The European Tributary States of the Ottoman Empire in the Sixteenth and Seventeenth Centuries the Ottoman Empire and Its Heritage

Home , Cossack Hetmanate, Giray dynasty, Khan Sahib, Mircea the Elder, Republic of Ragusa

The European Tributary States of the Ottoman Empire in the Sixteenth and Seventeenth Centuries The Ottoman Empire and its Heritage

Politics, Society and Economy

Edited by Suraiya Faroqhi, Halil İnalcık and Boğaç Ergene

Advisory Board fikret adanir – antonis anastasopoulos – idris bostan palmira brummett – amnon cohen – jane hathaway klaus kreiser – hans georg majer – ahmet yaşar ocak abdeljelil temimi – gilles veinstein†

VOLUME 53

The titles published in this series are listed at brill.com/oeh The European Tributary States of the Ottoman Empire in the Sixteenth and Seventeenth Centuries

Edited by Gábor Kármán and Lovro Kunčević

LEIDEN • BOSTON 2013 Cover illustration: Portrait of the Ragusan tribute ambassador and nobleman Marojica Caboga, late 17th century, by an anonymous painter. Courtesy of Dubrovački muzeji (Dubrovnik Museums).

Library of Congress Cataloging-in-Publication Data

The European tributary states of the Ottoman Empire in the sixteenth and seventeenth centuries / edited by Gábor Kármán and Lovro Kunčević. pages cm. — (The Ottoman Empire and its heritage, ISSN 1380-6076 ; volume 53) Includes bibliographical references and index. ISBN 978-90-04-24606-5 (hardback : acid-free paper)—ISBN 978-90-04-25440-4 (e-book) 1. Turkey—History—Ottoman Empire, 1288–1918. 2. Europe, Eastern—Relations—Turkey. 3. Turkey—Relations—Europe, Eastern. 4. Europe, Eastern—Politics and government— 16th century. 5. Europe, Eastern—Politics and government—17th century. 6. Turkey—Politics and government—16th century. 7. Turkey—Politics and government—17th century. I. Kármán, Gábor. II. Kunčević, Lovro.

DR511.E96 2013 947.0009’031—dc23 2013016824

This publication has been typeset in the multilingual “Brill” typeface. With over 5,100 characters covering Latin, IPA, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities. For more information, please see www.brill.com/brill-typeface.

ISSN 1380-6076 ISBN 978-90-04-24606-5 (hardback) ISBN 978-90-04-25440-4 (e-book)

Copyright 2013 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Global Oriental, Hotei Publishing, IDC Publishers and Martinus Nijhoff Publishers.

All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher.

Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change.

This book is printed on acid-free paper. CONTENTS

Acknowledgments ...... ix

Introduction ...... 1

Section One The Legal Status of the Ottoman Tributaries

The Legal and Political Status of Wallachia and Moldavia in Relation to the Ottoman Porte ...... 9 Viorel Panaite

Sovereignty and Subordination in Crimean-Ottoman Relations (Sixteenth–Eighteenth Centuries) ...... 43 Natalia Królikowska

Between Vienna and Constantinople: Notes on the Legal Status of the Principality of Transylvania ...... 67 Teréz Oborni

Janus-faced Sovereignty: The International Status of the Ragusan Republic in the Early Modern Period ...... 91 Lovro Kunčević

Cossack Ukraine In and Out of Ottoman Orbit, 1648–1681 ...... 123 Victor Ostapchuk

Section Two The Diplomacy of the Tributary States in the Ottoman System

Sovereignty and Representation: Tributary States in the Seventeenth-century Diplomatic System of the Ottoman Empire ...... 155 Gábor Kármán vi contents

Diplomatic Relations between the Ottoman Empire and the Republic of Dubrovnik ...... 187 Vesna Miović

Enemies Within: Networks of Influence and the Military Revolts against the Ottoman Power (Moldavia and Wallachia, Sixteenth–Seventeenth Centuries) ...... 209 Radu G. Păun

Section Three Military Cooperation between the Ottoman Empire and Its Tributaries

The Friend of My Friend and the Enemy of My Enemy: Romanian Participation in Ottoman Campaigns ...... 253 Ovidiu Cristea

The Military Co-operation of the Crimean Khanate with the Ottoman Empire in the Sixteenth and Seventeenth Centuries ..... 275 Mária Ivanics

‘Splendid Isolation’? The Military Cooperation of the Principality of Transylvania with the Ottoman Empire (1571–1688) in the Mirror of the Hungarian Historiography’s Dilemmas ...... 301 János B. Szabó

The Defensive System of the Ragusan Republic (c. 1580–1620) ...... 341 Domagoj Madunić

Section Four Instead of a Conclusion: on the “Compositeness” of the Empire

The System of Autonomous Muslim and Christian Communities, Churches, and States in the Ottoman Empire ...... 375 Sándor Papp contents vii

What is Inside and What is Outside? Tributary States in Ottoman Politics ...... 421 Dariusz Kołodziejczyk

Notes on Contributors ...... 433

Personal Names ...... 439 Place Names ...... 446

Acknowledgments

In the making of this book we have incurred many debts and it is a great pleasure to renew our thanks here. In the first place, we are deeply grateful to Robert Born and Nenad Vekarić, who provided unwavering support from the two institutions under whose auspices the conference was organized, the project group “Ottoman Orient and East Central Europe” at Geisteswissenschaftliches Zentrum Geschichte und Kultur Ostmittel europas an der Universität Leipzig and Zavod za povijesne znanosti HAZU u Dubrovniku. We also owe our warmest thanks to Judith Rasson, Arnold Ross, Katalin Stráner and Valerie Joy Turner for their painstaking work on the linguistic proof-reading of this sizeable volume. Moreover, we are indebted to Gáspár Katkó and Nedim Zahirović whose expertise in Orien- tal languages helped to resolve a number of philological dilemmas. Timo Stingl, who made the maps for the volume, also has our gratitude for his fine work. Finally, we owe our warmest thanks to the series editors and the volume’s reviewers, from whom we received invaluable comments which helped improve the book.

The editors Dubrovnik / Leipzig 15 December 2012

Introduction

It is a well-known fact that the expansion of the Ottoman Empire proceeded through various phases of power relations between the states and peoples of the invaded lands and the empire itself. Tributary states, such as the Bulgarian tsardoms of Trnovo and Vidin, Serbia, or many of the Aegean Islands were, in the course of renewed attempts to enlarge the territory under the sultan’s rule, incorporated into the empire. By the second half of the sixteenth century, however, a more or less stable system of tributary states was formed, which, despite the ever-changing scale of imperial influence, remained largely constant. The acknowledgment of the Ottoman overlords was a fundamental element of the policies of the rulers of Crimea, Moldavia, Ragusa, Transylvania, and Wallachia during the early modern period. In other entities, such as Cossack Ukraine, or Northern Hungary, the Ottoman tributary position was a short-lived experience, which is, nevertheless, worthy of attention. The status of Ottoman tributary states has, until recently, all but escaped the attention of researchers of the Ottoman Empire and has mainly been treated in the context of national historiographies. Croatian, Italian, Hungarian, Romanian, and Transylvanian German historians have dedicated numerous studies to the contacts between the Ottoman Empire and the country to which they declared allegiance. It is hardly surprising that in spite of the growing number of recent publications on the Crimea, it remains the European tributary that attracts the least scholarly interest, as no nineteenth- or twentieth-century nation state has claimed the legacy of the khanate as its own. National historiographies reveal many important details in the history of these states; however, their focus on single states as closed units tends to distort the interpretative process. Seeing the single case as unique—or applying only a rather coarse definition of the other states being in a “better” or “worse” situation—often results in misinterpretations of Ottoman attitudes and local responses, the nature and causes of the connection. In recent decades, the most important development of historical scholarship that had its impact on the histories of the Ottoman tributary states was the re-assessment of the history of the empire itself, with an emphasis on its composite state character, a re-evaluation of its activities, and the identification of its inner logic of empire-building. The shift of perspective 2 introduction from one focused on the “Ottoman yoke” to the “Pax Ottomanica”— or, rather, an emphasis on the negotiation between imperial and local perspectives—has resulted in a more nuanced understanding of the history of Ottoman-dominated Europe. Moreover, a change in the assessment of the Ottoman tributary status took place, with more understanding for the imperial perspectives. There have even been very promising initiatives for systematic comparative studies in this field; however, they have so far focused mainly on the legal framework of the tributary status, based on Ottoman sources, usually with a special focus on one tributary state. Given the wide geographic area covered and the linguistic variety of the sources, it is clear that only an international network of scholarly exchange would render it possible to create a nuanced understanding of the experience of tributary states that would “give voice” to both imperial and domestic sides, and at the same time take into account general trends and regional differences. From 22 to 23 May 2009, a conference in this vein was convened in Dubrovnik (Croatia) as a cooperation between the project “Ottoman Orient and East Central Europe: Comparative Studies in the Perceptions and Interactions in the Border Zones” at the Geisteswissenschaftliches Zentrum Geschichte und Kultur Ostmitteleuropa an der Universität Leipzig (GWZO) and the Institute for Historical Sciences in Dubrovnik of the Croatian Academy of Sciences and Arts (HAZU). It was the first event aimed at creating an international scholarly network, such as described above, by bringing together well-known experts of the various tributary states to discuss the state of the art and further possibilities of research. The topics of the papers presented at the conference were selected to cover the broadest possible field, maintaining a balance topically as well as regionally. The goal of the convenors was to provide a current overview of the historiographies of the individual states and thereby open up new perspectives for further research. The program thus covered the most important topics in the history of the Ottoman tributary states, offered the widest possible regional scope (Ragusa, Wallachia, Moldavia, Tran- sylvania, Crimea, and even the Cossack hetmanate), and even provided a conceptual panel that addressed, in detail, the meaning of tributary status from the Ottoman perspective. This volume, based on the material presented at the conference in Dubrovnik, but completed with additional studies by experts in various fields, was compiled with a similar aim. The articles are grouped into three sections which cover the crucial aspects of the political relationship between the empire and its tributary states. The first section is dedicated introduction 3 to the legal status of various Ottoman satellites, the second to the diplomatic communication between these polities and the Sublime Porte, and the third to the military cooperation between them and the imperial army. A fourth section concludes by looking at the empire as a whole, demonstrating the fragmentary nature of both its legal structure and its factual control, thus providing the broader context for the studies in the volume. Opening the section dedicated to the legal status of the Ottoman satellites, Viorel Panaite analyzes the gradual changes in the legal status of Moldavia and Wallachia which reflected the growing control of the Porte over these principalities during the fifteenth and sixteenth centuries. Besides consulting the rich legal material, Panaite tackles this controversial issue from a broader perspective, taking into account the narrative and diplomatic sources as well as the legal self-understanding of the Christian elites of these lands. In the following study, Natalia Kró- likowska investigates the relationship between the Crimean Khanate and the Ottoman Empire, analyzing both the legal interpretations of the khanate’s status and its factual autonomy from the Porte. Królikowska reconstructs this relationship by analyzing a number of politically revealing practices and symbols: the enjoyment of royal prerogatives such as the collection of taxes or sharing in the spoils of war, the procedures regarding the inheritance of the throne, the peculiar attire of the khan and the insignia given to him by the sultan. Teréz Oborni provides a fresh approach to the legal status of Transylvania by considering not only its relationship toward the sultan, but also the drastically different legal position which the principality continually renegotiated toward the Habsburgs. What emerges from her study is an intriguing picture of a small polity struggling to achieve the best possible position between the two great powers which interpreted Transylvania’s legal position in contradictory ways. Similar questions concerning multiple parallel commitments and even polyvassalage are discussed by Victor Ostapchuk. His narrative account of the seventeenth-century attempts for a close cooperation between Cos- sack Ukraine and the Ottoman Empire also presents an analysis of the most important legal documents regulating these contacts, as well as a broad context to render the sometimes surprisingly flexible attitudes of the players understandable. Finally, in his contribution Lovro Kunčević reconstructs the profoundly different understandings of the legal status of the Ragusan republic, that held by the Ottomans on the one hand and the Christian states on the other. Using this example to stress that legal status was a changing situational construct, the article argues for the possibility 4 introduction of multiple legal statuses of early modern polities, especially those which participated in different “international societies.” The second section, dedicated to diplomatic communication between the tributary states and the empire, opens with the article of Gábor Kármán, who presents the place of the tributary states in the diplomatic system of the Sublime Porte. Contrasting the history of the Transylvanian embassy in Constantinople to a wide comparative material, his analysis of four segments of diplomatic representation suggests that the dividing line between the treatment of the Romanian principalities and the other tributary states was much stronger than that between the latter and states independent from the Ottoman Empire. In her study of Ragusan diplomacy toward the Sublime Porte, Vesna Miović analyzes various ways in which the small republic interacted with its powerful protector, ranging from diplomatic ritual all the way to delivering confidential information. This summary of her extensive research touches on issues such as the structure of the republic’s diplomatic corps, its rhetoric in front of the Ottomans, the functioning of its espionage network, and its methods of transporting confidential information (e.g., ciphers and couriers). The study of Radu G. Păun investigates the relationship between the Porte and the Moldavian and Wallachian elites by focusing on the revealing situations of the revolts against the Ottomans. Offering a broad contextualization of these critical events, Păun analyzes the Porte’s mechanisms of control over the Danu- bian principalities, the biographies of the individual princes, the networks of influence they tried to control, as well as the impact of European crusading plans and eschatological expectations on these uprisings. The section on the military cooperation between the tributary states and the Ottomans opens with Ovidiu Cristea’s article, which deals with the little-researched topic of military aid provided by the Moldavian and Wallachian princes to the imperial army. Cristea addresses the origins of this arrangement, the numbers, tactical role, and types of Romanian troops, the battlefields where they fought, and the intervals for which they served. Along similar lines, Mária Ivanics deals with the military cooperation between the Ottomans and the Crimean Khanate, reconstructing its major episodes and crucial characteristics. Her study provides an overview of the military issues, such as the structure and size of the Tartar armies and their methods of supply and warfare, but also investigates the broader economic and political effects of such Tartar engagements. The Transylvanian military cooperation with the Ottoman Empire is presented in János B. Szabó’s study, which provides an analytical overview of its various forms. From the participation of the neighboring pashas’ armies in introduction 5 the princes’ campaigns to sending Transylvanian auxiliaries to the various battlefields, it presents a vision of the diverse mutual benefits gained by each party, as well as the drawbacks of the cooperation, when both sides had good reasons to regard the other as unreliable. Dealing with a very different kind of polity, the small Republic of Ragusa, Domagoj Madunić analyzes the organization of its defensive system, in particular the effect that Ottoman protection—guaranteed by the tributary status—had on its formation. On a more general level, his text sheds new light on one important Ottoman borderland, revealing that even the militarily weak states such as Ragusa, if located strategically, could play a significant role in the strategic planning of the great powers. The fourth and concluding section assumes a somewhat different perspective than the rest of the volume: going beyond the individual satellites, it looks at the entirety of the empire. It does that, however, not by adopting the traditional unifying perspective from the center, but by assessing the Ottoman commonwealth from its peripheries and from the viewpoint of the numerous autonomous communities within it. Thus Sán- dor Papp offers a comprehensive overview of various kinds of religious, communal, and state autonomies within the empire, encompassing both Muslim and Christian communities. Besides offering such a remarkable catalogue, Papp also tackles the history of Islamic legal doctrines on autonomy, regarding Muslim and Christian self-government. The thus revealed “compositeness” and fragmentary nature of the empire is further accentuated in the study of Dariusz Kołodziejczyk, which demonstrates a rather surprising point: how difficult it is to categorically determine what was “inside” and what was “outside” of the Ottoman state. Blurring the empire’s borders, Kołodziejczyk shows the constantly changing Ottoman interpretations of legal status of neighboring polities, and also stresses that many apparently integral parts of the empire were in reality beyond the reach of Istanbul. Though most contributions focus on the history of single tributaries and use the comparative method only tangentially, they most certainly enter into conversation with each other and thus provide a more comprehensive view of the tributary states’ history than that previously available. Crossing the borders of several national historiographies and languages, the volume can thus serve as the first stage of a more ambitious project to address the hitherto neglected comparative history of Ottoman satellites, and provide valuable new insights into the functioning of the empire as a whole, and its frontier with the Christian world in particular.

Section One

The Legal Status of the Ottoman Tributaries

The Legal and Political Status of Wallachia and Moldavia in Relation to the Ottoman Porte

Viorel Panaite

Studying tributary status is largely a matter of understanding the character of one’s sources and asking the right questions. Islamic juridical and religious sources—such as the Qur’an, the traditions (hadis), Islamic legal treatises (kitab as-siyar, kitab al-cihad), juridical opinions (fetva), and others—offer a theoretical model for defining the legal positions of non-Muslim subjects and foreigners. However, in order to decipher the official Ottoman view on the political position of tributary states, provinces, rulers, and people, it is necessary to apply a wider range of documentation. This also requires a deeper analysis of administrative and diplomatic chancellery documents that reflect the political, military, and diplomatic practice more directly and therefore render a better understanding possible. Several of the above mentioned types of documents are relevant to the subject of this study: imperial charters, diplomas, and letters (‘ahdname-i hümayun, berat-ı hümayun, name-yi hümayun, hatt-ı şerif, etc.), orders of the sultan (hüküm), petitions (‘arz), reports (telhis), and others. At the same time, given the insufficiency of official documents, the Ottoman chronicles (generally called Tevarih-i Al-i ‘Osman) are also primary sources for the study of the Ottoman policy in Southeastern Europe during the fourteenth and fifteenth centuries. Considering Süleyman the Magnificent’s epoch as a turning point in the political and juridical status of Wallachia and Moldavia vis-à-vis the Porte, in this article I mainly respond to the following three questions: Did the Ottomans conquer the principalities of Wallachia and Moldavia? Were Wallachia and Moldavia in the Ottoman Empire? And, did long- term “treaties” exist between the Ottoman Empire and the two Danubian principalities?

From Acknowledgment of Allegiance to Conquest

The Shakespearean dilemma “to be or not to be” does not work for Roma- nian historians in respect to the Ottoman conquest. Following the nationalistic ideal of “neither conquerors, nor conquered,” some scholars attempt 10 viorel panaite to solve the problem by inverting the question: “Why did the Ottomans never conquer the Romanian countries?”1 In my view, it is necessary to clearly affirm that—after a long period of paying homage—the Ottomans succeeded in conquering Wallachia and Moldavia during the reign of Süleyman the Magnificent. During the fifteenth and sixteenth centuries, the Ottomans used the offensive cihad ideology for the legal justification of their military expeditions against Christian states in Southeastern and Central Europe. The term küffar was used abundantly in Ottoman juridical, official, and narrative sources for the designation of all non-Muslims. As Muslim juris- prudence regularly ignored the notion of ethnic groups, Wallachians and Moldavians, in their quality as Christians, were included in the category of “infidels” (harbiler, sing. harbi) or “enemy infidels” (harbi küffar) and, more specifically, in that of the “people of the book” (ehl-i kitab). In the Ottoman juridical view (which was Hanafi), the world was divided into two parts, i.e., the “house of Islam” (dar ül-Islam) and “house of war” (dar ül-harb). The former included all territories under the sultan’s authority: not only those governed by şeri‘at and ruled by Muslim governors, but also the tributary provinces and states. The latter, also called the “house of infidels” (dar ül-kefere), referred to all foreign territories and states, and it was in this realm that—for a longer or shorter period—the countries of Southeastern Europe were included in the fifteenth and sixteenth centuries.2

1 Most modern Romanian historians strongly reject the idea of the Romanian principalities’ conquest (see Petru P. Panaitescu, “De ce n-au cucerit turcii Ţările Române” [Why did the Turks not conquer the Romanian countries?], in Interpretări româneşti: Studii de istorie economică şi socială [Romanian interpretations: Studies in economic and social history], ed. Ştefan S. Gorovei and Maria M. Székely (Bucharest, 1994), 111–119 (first published in Revista Fundaţiilor Regale 11, no. 5 (1944)); Mihai Maxim, Ţările Române şi Inalta Poartă: Cadrul juridic al relaţiilor româno-otomane în evul mediu [The Romanian countries and the Sublime Porte: Legal frames of Romanian–Ottoman relations in the Middle Ages] (Bucharest, 1993), 111–142 (Chapter “De ce n-au cucerit Otomanii Ţările Române”). Other historians use the term “conquest,” yet emphasize the difference in the Ottoman methods of conquering. Accordingly, from the late fourteenth century on, Southeastern European states, such as Bulgaria, Serbia, the Byzantine Empire, and Bosnia were crushed and dis- mantled by the invading Ottoman Turks, and were then directly incorporated into the rising Ottoman state. In return, Wallachia and Moldavia “were preserved as puppet states in name only” for nearly five hundred years (Dennis P. Hupchick, Culture and History in Eastern Europe (New York, 1994), 59). 2 For details, see Viorel Panaite, The Ottoman Law of War and Peace: The Ottoman Empire and Tribute Payers (Boulder, CO, 2000), 77–126. the legal and political status of wallachia and moldavia 11

From the late fourteenth to the mid-sixteenth century, the principalities of Wallachia and Moldavia were also considered to be a part of the “house of infidels”3 or “house of war” (in Ottoman, harbi vilayetleri, dar ül-harb, dar-i harb, diyar-i harb). In this context, the fifteenth-century chroniclers considered the Danube as a line of demarcation between the “house of Islam” and the “house of war.” Kıvami mentions that the army led by Bayezid II in 1484 “crossed the waters to the other bank of the Danube, in the house of war.”4 Theoretically, there was a permanent state of war between the “house of war” and the “house of Islam.” The ultimate aim of the Ottoman leaders was to bring these territories under their authority, to incorporate them into the “house of Islam,” and in this manner build a pax ottomanica. The best instrument for achieving this goal, from a juridical point of view, was to solicit Allah’s path, which meant, practically, waging a military campaign (gaza) against the Christian leaders of Southeastern and Central Europe. During that period, the two Danubian principalities were thus objects of an Ottoman conquering policy. In this vein, akıncı expeditions were frequently launched toward the north of the Danube to support, keep under control or punish certain local princes. Periodically, even great military campaigns (gaza, pl. gazavat) were organized under the direct command of sultans, grand viziers or beylerbeyis of Rumeli. Let us mention in particular the military campaigns led against Wallachia by the sultans Bayezid I in 1395, Mehmed I in 1417, and Mehmed II in 1462, and those against Moldavia by Mehmed II in 1476, Bayezid II in 1484, and Süleyman the Magnificent in 1538.5 For about one hundred and fifty years, holy war,

3 Diyar-i küffar, diyar-i küfr, kefere vilayetleri, kafiristan. Cf. Friedrich Giese, Die altos- manischen anonymen Chroniken (Tevârih-i Al-i Osman), vol. 1 (Breslau, 1922); idem, Die altosmanische Chronik des Aşıkpaşazade (Leipzig, 1929); Hüseyin Nihal Atsız, ed., Oruç Beğ Tarihi (Istanbul, 1972); Tursun Beg, The History of Mehmed the Conqueror, ed. Halil İnalcık and Rhoads Murphey, Biblioteca Islamica (Minneapolis and Chicago, 1978); Ibn Kemâl, Tevârih-i Âl-i Osman [The chronicle of the Ottoman dynasty], vol. 7, Tenkidli transkripsiyon [Critical transcription], ed. Şeraffetin Turan (Ankara, 1991); İbn Kemâl, Tevârih-i Âl-i Osman, vol. 8, Transkripsiyon [Transcription], ed. Ahmet Uğur (Ankara, 1997); Çiftcioğlu N. Atsız, Osmanlı Tarihleri [Ottoman chronicles], vol. 1 (Istanbul, 1949). 4 “(. . .) sulere Tunanın öte kenarına dar ül-harb’e geçdiler” Kıvami, Fetihnâme-i Sultan Mehmed [The book about the conquests of Sultan Mehmed], ed. Franz Babinger (Istanbul, 1955), 304–306. 5 For instance, Mehmet Neşri, when relating the history of the Ottoman dynasty in the natural order of the sultans’ succession, also wrote “stories” (hikayet) about the “sacred campaigns” commanded directly by the padishahs against Wallachia and Moldavia: “The Story of the Sacred Expedition in Wallachia” (Hikâyet-i Gazâ-yı Eflâk), referring to 1462; “The Story of the Second Sacred Expedition in Moldavia” (Hikâyet-i Gazâ-yı Kara-Boğdân 12 viorel panaite gazis and the Danube river had a common history, as the revealing phrase by the fifteenth-century Ottoman annalist Kıvami testifies: “Danube as a gazi river.”6 To justify their peaceful contacts with non-Muslims, the Ottomans adopted rules and notions specific to classical Islamic law; yet, they applied them in direct relation to practical circumstances. The Moldavian scholar Prince Dimitrie Cantemir (1673–1723) grasped very well the meaning of the Ottoman law of peace: Muslims were bound to fight against Chris- tians to spread Islam, “yet that law determines the manner in which peace may and ought to be made.” He therefore made a distinction between two standard situations. If the infidels had not accepted conversion to Islam or poll-tax paying, the Ottomans were allowed to interrupt the holy war and make truces with the Christian rulers. If the infidels had, however, accepted conversion to Islam or poll-tax paying, it was an obligation for the Muslims to stop the holy war and make a “firm and lasting peace” with non-Muslims.7 For the purposes of the present investigation, the acceptance of tribute paying by Christian rulers is of the highest relevance. There are several obvious cases for this from the fourteenth and fifteenth centuries in the southeastern European tributary states of the Porte, with longer or shorter intermissions: Byzantium between 1372 and 1453, Serbia from 1372 to 1459, the Bulgarian tsardoms in the 1380s–1390s, Bosnia between 1389 and 1463, Albanian lordships from 1385 to 1478, the Morean despotate, and the North Aegean Cycladic Islands (e.g., Lesbos, Lemnos, Imbros) in the fifteenth century. These cases should be taken as points of comparison to the fifteenth-century legal and political position of the principalities of Wallachia and Moldavia. Most of the above-mentioned states were eventually transformed into territories of direct Ottoman administration. The case of certain south Aegean Cycladic Islands (the Duchy of Naxos together with its dependencies, Paros and Andros), Ragusa, and the principality of Transylvania between 1541 and 1699 deserve to be mentioned specifically,

saniyen), referring to 1476. See Mehmed Neşri, Kitâb-ı Cihan-nüma: Neşri Tarihi [The mirror of the world: The chronicle of Neşri], vol. 2, ed. Faik Reşit Unat and Mehmed A. Köymen (Ankara, 1987; 1947), 755–759, 829–833. 6 Kıvami, Fetihnâme, 306–307; Viorel Panaite, Pace, război şi comerţ în Islam: Ţările române şi dreptul otoman al popoarelor: Secolele XV–XVII [Peace, war and commerce in Islam: The Romanian countries and the Ottoman Law of Nations: fifteenth to seventeenth centuries] (Bucharest, 1997), 278–292. 7 Demetrius Cantemir, The History of the Growth and Decay of the Othman Empire, ed. Nicholas Tindal (London, 1734), 276n1. the legal and political status of wallachia and moldavia 13 as they enjoyed a long-term status of tributary protection, similar to Wal- lachia and Moldavia from the fifteenth to the eighteenth century. They shared a common feature: their position as autonomous principalities or provinces was preserved in spite of their altered juridical and political forms. Finally, the juridical status of the Crimean Khanate—as a client state with a large internal autonomy in the Ottoman Empire—should also be compared to that of the Wallachian and Moldavian principalities, even though the khans did not pay tribute to the sultans.8 The political moments when princes of Wallachia and Moldavia accepted to pay tribute and submitted themselves to the Porte were called închinare in Romanian sources (Latin inclinare, Greek proskynesis), which can be translated as “acknowledgment of allegiance” or “homage-paying.” In Ottoman chronicles and documents, the following simple or phrasal verbs were used to indicate “submission and obeisance”: boyun eğmek, boyun vermek, tapmak, ita‘at ve inkıyad etmek, yüz sürümek, yüz sürmek. The tradition of “homage-paying” indicates a long process of regulating the relations between the Porte and Wallachia, later also Moldavia, which was marked by a series of conflicts and the conclusion of temporary pacts between sultans and hospodars. The fifteenth- and sixteenth-century sources that report the Wallachian and Moldavian princes’ acknowledgment of allegiance toward the Porte implied at least one of the following political and diplomatic actions: the presentation of the voievod himself, or that of his representative at the Porte, and sometimes even an envoy of the great boyars of the country; submission to the sultan willingly, not merely as a result of defeat in an open military confrontation; conclusion of a peace agreement between the two parts; payment of a sum of money, invariably considered as a tribute by the Ottomans (called either cizye or harac); the granting of protection by the sultan; acknowledgment of enthronement, and the granting of some distinguishing marks of investiture to the voievod or to his delegate. Limiting the moment of “homage-paying” to a single year and a specific voievod—as chroniclers of the seventeenth and eighteenth centuries did—seems to be a mistake. The tradition of “homage-paying,” by its ambiguous chronology, indicates rather a genuine process of the regulation of relations between the Porte and Wallachia, later also Moldavia, marked by a series of events

8 For details, see Panaite, The Ottoman Law of War and Peace, 147–156. 14 viorel panaite traditionally regarded as conclusions of peace agreements between sultans and voievods.9 In the case of Wallachia (Eflak), this process began during the reign of Mircea the Elder (Mircea cel Bătrân, 1386–1418),10 and was completed after Mehmed II’s campaign of 1462, by the enthronement of Radu the Handsome (Radu cel Frumos, 1462–1475, with interruptions).11 During this period, the most important developments took place in the years 1391–1395, 1417 and 1462. Yet, a complete and long-term submission was only secured in the third decade of the sixteenth century.12 Relating the astounding developments from that period, the Wallachian chronicler Radu Popescu registered an attempt to make peace with the Porte, initiated by Radu of Afumaţi (Radu de la Afumaţi, 1522–1529, with interruptions) and Wallachian boyars, probably in 1524: But putting their heads together, they found this: [as] they were few in a small country, they could not fight an emperor who conquered and overran so many countries and has a great many arms. They all decided that the voievod [should] go to the Porte and bow his head to the emperor’s lap. And so he did. He went with many boyars.13

9 For details, see ibid., 196–214. 10 Both Ottoman and non-Ottoman sources referred to Mehmed I’s expedition against Wallachia, and the peace agreement concluded with its prince between 1416 and 1420; however, the exact dates of these episodes are uncertain, and the protagonists’ identity is doubtful as well. For the seventeenth-century English historian Richard Knolles, this moment meant that Wallachia became “tributary to the Turk.” For details, see Richard Knolles, The Turkish History, from the Original of that Nation, to the Growth of the Ottoman Empire: with the Lives and Conquests of their Princes and Emperors (London, 1687), vol. 1, 170; Nicolae Stoicescu, Vlad Ţepeş (Bucharest, 1976), 85–136; Maxim, Ţările Române şi Inalta Poartă, 230–240. 11 In Ottoman terms, Mehmed II’s campaign of 1462 signified “complete submission,” passing “under the grand sultan’s domination,” and even the “conquest” of Wallachia. “Pay- ing homage” in 1417 was a personal decision; “paying homage en masse” took place only after about fifty years, in 1462. “The padishah of Islam (. . .), crossed the Danube, entered Wallachia, all Wallachian people came [and] paid homage to him” (“Padişah-ı İslam (. . .) Tuna’yı geçüb Eflak vilayetine girüb cemi’ Eflak’ın vilayeti halkı gelüb tapdılar”) Neşri Tarihi, vol. 2, 756, 757. 12 “Cronica lui Macarie” [The chronicle of Macarie], in Cronicile slavo-române din sec. XV–XVI [Slavic-Romanian chronicles from the fifteenth to the sixteenth centuries], ed. Ioan Bogdan and Petre P. Panaitescu (Bucharest, 1959), 93; Michael Bocignoli’s report in Călători străini despre Ţările Române [Foreign visitors in the Romanian countries], vol. 1, ed. Maria Holban (Bucharest, 1968), 178; Nicolae Stoicescu, Radu de la Afumaţi (Bucharest, 1983), 81–88. 13 Istoriile domnilor Ţărîi Româneşti de Radu Popescu [The histories of the Romanian lands by Radu Popescu], in Cronicari munteni [Muntenian chroniclers], ed. Mihail Grego- rian, vol. 1 (Bucharest, 1961), 274. the legal and political status of wallachia and moldavia 15

Taking these considerations into account, we can better understand the unqualified satisfaction that the Wallachian voievod expressed in his letter to the inhabitants of Sibiu from 1 February 1525: I went to the Turkish Porte and I stayed in the presence of the Turkish emperor and all the high officials of the Turkish country. The Wallachian reign was given to me by God, firstly, and then by the Turkish emperor, and I came back again alive and healthy (. . .), and I took the Wallachian throne and reign.14 In Moldavia, the process of homage-paying began in 1455–1456 and ended in 1538,15 but the peace agreements concluded by Stephen the Great (Ştefan cel Mare, 1457–1504) with Mehmed II Fatih (in 1480–1481) and Bayezid II (in 1486) are also relevant episodes. The Ottoman chronicles mostly emphasize Süleyman the Magnificent’s expedition in 1538 and its consequences on the legal and political status of Moldavia. Then, very much like events in Wallachia in 1462, a “homage paying en masse” took place which proved to have more important consequences than the previous peace agreements concluded between sultans and Moldavian voievods: “Young princes and boyars (boyarları) of the Moldavian vilayet, kowtow- ing in all possible ways and taking deep bows,” wrote Mehmed-Paşa Küçük Nişanci (d. 1571), “prayed for a new hospodar (hakim) to be appointed to rule over them. Then, on the 21st Rebi ül-ahır, in the year 945 [16th Sep- tember 1538], when [Ştefan Lăcustă, 1538–1540], one of the young princes of their descent, was appointed, they pledged themselves to send their harac yearly.”16 In the third and fourth decades of the sixteenth century, during the rule of Süleyman the Magnificent, important changes took place in the

14 Grigore George Tocilescu, 534 documente istorice slavo-române din Ţara Românească şi Moldova privitoare la legăturile cu Ardealul 1346–1603 [534 historical Slavic-Romanian documents from Wallachia and Moldavia concerning their contacts with Transylvania 1346–1603] (Bucharest, 1931), 426 (no. 308). 15 In the Ottoman chronicles, the homage of Voievod Petru Aron (1451–1457, with interruptions) in 1455–1456 is not registered, although the first Ottoman official documents that used the term “peace” (barış) between Mehmed II and Moldavia were issued in 1455–1456. See Mustafa A. Mehmet, Documente turceşti privind istoria României [Turkish documents concerning the history of Romania], vol. 1, 1455–1774 (Bucharest, 1976), doc. 1, 2. 16 Mehmed Paşa, “Tarih-i Nişancı” [The chronicle of Nişancı], in Cronici turceşti privind Ţările române: Extrase [Turkish chronicles concerning the Romanian countries], vol. 1, Sec. XV–mijlocul sec. XVII [Fifteenth to mid-seventeenth century], ed. Mihail Guboglu and Mus- tafa Mehmet (Bucharest, 1966), 295. See also Nasuh Matrakçi, Fetihname-i Kara-Boğdan [Book about the conquests of Kara-Boğdan] (Istanbul, R. 1284). 16 viorel panaite power relationships of Southeastern and Central Europe. The Ottomans conquered the Hungarian-held city of Belgrade in 1521, incapacitated the Hungarian army at the battle of Mohács in 1526, and occupied their capital, Buda in 1541. All these episodes led to the disintegration of the medieval Kingdom of Hungary.17 Moreover, as a response to Polish claims and Habsburg attacks, Süleyman the Magnificent established a definitive Otto- man control over the territories north of the Danube.18 The campaign of 1538 against the Moldavian prince Petru Rareş (1527–1546, with interruption), the second conquest of Buda, and the transformation of Transylva- nia into a tributary principality in 1541 also affected the political status of Wallachia. According to the Ottoman legal view, the territories falling under Mus- lim control were considered “conquered,” regardless of whether they had been taken “by force” or had submitted themselves “willingly.” In spite of the resistance and peace agreements from the fifteenth to the mid- sixteenth century, the principalities of Wallachia and Moldavia were eventually conquered by the sultans, who consequently invoked the “right of sword” (kılıç hakkı) over them. In the case of Wallachia, the idea that the country was conquered became more common after the third and fourth decades of the sixteenth century. Later, in an “imperial diploma” (berat-ı hümayun) of 1586, Mehmed III stated that “the above-mentioned country was conquered by power of our swords.”19 In his narrative about the surrender of Suceava, the capital of Moldavia in 1538, İbrahim Peçevi affirms that “in those days, the country of Moldavia came under the rule

17 For details, see Gábor Barta, La route qui mene à Istanbul 1526–1528 (Budapest, 1994). 18 For details, see Mihnea Berindei and Gilles Veinstein, L’Empire Ottoman et les Pays Roumains 1544–1545: Étude et documents (Paris and Cambridge, 1987), 17–46; Pál Fodor, “Ottoman Policy towards Hungary, 1520–1541,” Acta Orientalia Academiae Scientiarum Hun- garicae 45, nos. 2–3 (1991): 271–345; Cristina Feneşan, Constituirea principatului autonom al Transilvaniei [The formation of the autonomous Principality of Transylvania] (Bucha- rest, 1997); Radu Constantinescu, Moldova şi Transilvania în vremea lui Petru Rareş: Relaţii politice şi militare (1527–1546) [Moldavia and Transsylvania in the time of Petru Rareş: Political and military contacts] (Bucharest, 1978); Ştefana Simionescu, “Ţările române şi începutul politicii răsăritene antiotomane a Imperiului habsburgic (1526–594)” [The Romanian countries and the beginning of eastern anti-Ottoman politics in the Habsburg empire], Revistă de Istorie 28 (1975): 1197–1214; eadem, “Les relations de la Moldavie avec les Habsbourg pendant le règne de Petru Rareş (1527–1538; 1541–1546),” Revue Roumaine d’Histoire 16 (1977): 463–475. 19 Enacted by Murad III on 6 September 1586, to confirm Mihnea II Turcitul (Mihnea II the Turned-Turk) as voievod of Wallachia: Mihai Maxim, Culegere de texte otomane [Col- lection of Ottoman texts], vol. 1, Izvoare documentare şi juridice (sec. XV–XX) [Documen- tary and legal sources, fifteenth to twentieth centuries] (Bucharest, 1974), doc. 14. the legal and political status of wallachia and moldavia 17 of the Ottoman countries.”20 In Ottoman documents of the seventeenth to eighteenth centuries, the idea of the conquest of these countries was frequently connected to Süleyman the Magnificent’s rule, invested with the meaning of a turning point in the relations of the Porte with the tributary principalities of Wallachia and Moldavia. It is important to emphasize that as a rule, neither the Ottoman authorities, nor the autochthonous princes and nobles invoked the conquest of their countries in a negative sense. In this regard, it is worth having a closer look at the phrase “being separated at chancellery, spared of viola- tions, and free in all respects” (“mefruzü’l-kalem ve maktu’l-kadem min- küll-il-vücuh serbest olub”), frequently invoked in Ottoman documents.21 Romanian historians and jurists correlated this expression only with the autonomous status of Wallachia and Moldavia. As a matter of fact, this statement was incompletely quoted in order to avoid emphasizing that this juridical position had been applied toward the tributary principalities “since the imperial conquest” (“feth-i hakaniden berü”).22

Tribute-payers and Protected Peoples

The juridical nature of the relations of Wallachia and Moldavia with the Porte is a main topic of Romanian historiography on Ottoman affairs. Certain researchers accept only a de facto relation, which they see as constantly affected by abuses.23 This interpretation stands, however, in conflict with a large spectrum of legal, political, and administrative sources that prove the existence of a de jure status of Wallachia and Moldavia within the system of pax ottomanica. Having in mind another formula Romanian historians hold dear, according to which the Romanian principalities were at the “border of three empires” (i.e., Ottoman, Habsburg, and Russian),24 the second question we address in these pages is, Were Wallachia and Moldavia within the Ottoman Empire?

20 İbrahim Peçevi, “Tarih,” in Cronici turceşti, vol. 1, 481. 21 Valeriu Veliman, Relaţiile româno-otomane (1711–1821): Documente turceşti [Romanian- Ottoman contacts 1711–1821: Turkish documents] (Bucharest, 1984), doc. 93, 104, 107, 112, 140, 117, 118, 121, 127, 130, 141, 210, etc. 22 For details, see Panaite, The Ottoman Law of War and Peace, 214–222. 23 For instance I.P. Marcu, “Idei despre stat si drept în opera lui Dimitrie Cantemir” [The idea of state and law in the works of Dimitrie Cantemir], Studii si cercetari juridice 3 (1973): 497. 24 Veniamin Ciobanu, La graniţa a trei imperii [At the border of three empires] (Iaşi, 1985); Leonid Boicu, Principatele Române în raporturile politice internaţionale (secolul al 18 viorel panaite

In his fundamental work, published forty years ago, Nicolaas H. Bieg- man formulated a parallel question as a dilemma about Ragusa. He emphasizes that while there were many Ottoman sources treating Dubrovnik and the empire as separate entities, it is more correct to say “Dubrovnik and the rest of Turkey” than “Dubrovnik and Turkey.” For this reason, he gave the following title to the chapter analyzing this dilemma: “Dubrovnik’s Posi- tion within the Ottoman State.”25 Romanian contributors to this topic are fixated on proving that the principalities of Moldavia and Wallachia enjoyed a superior status in comparison to the Balkan territories. In order to achieve this aim, several concepts were invented—many of them improper or equivocal—to describe the status of the principalities, including concepts such as “submitted states,” “dependent states,” “vassal states,” “autonomous states,” “effective dependence,” “Ottoman domination,” “autonomy,” and so on.26 The concepts of “vassalage” and “autonomy” deserve special treatment. Most historians and jurists have defined the relations between tribute- paying princes and sultans from a European legal and political point of view, considering them to be in a vassal–suzerain relationship. Concern- ing this, two observations must be emphasized. On one hand, the concepts of “vassal” and “suzerain” were used to define the relationship between the Ottoman Empire and the Romanian principalities in a later period, especially in eighteenth- and nineteenth- century official European documents, narrative writings, and translations of Ottoman documents.27 Western historians and jurists used the “vassal

XVIII-lea) [The Romanian principalities in international political reports, eighteenth century] (Iaşi, 1986). 25 Nicolaas H. Biegman, The Turco-Ragusan Relationship according to the Firmans of Murad III (1575–1595) Extant in the State Archives of Dubrovnik (The Hague and Paris, 1967), 29–45. 26 A few titles include G. Duma, Situaţia juridică internaţională a Ţărilor Române faţă de Imperiul otoman [The international law position of the Romanian countries toward the Ottoman Empire] (Bucharest, 1947); Mihai Maxim, “L’autonomie de la Moldovie et de la Valachie dans les actes officiels de la Porte au cours de la seconde moitié du XVIe siècle,” Revue des Études Sud-Est Européennes 15, no. 2 (1977): 207–230; idem, “Le statut des Pays Roumains envers la Porte Ottomane aux XVI–XVIIIe siècles,” Revue Roumaine d’Histoire 24, nos. 1–2 (1985): 29–50; idem, Ţările Române şi Inalta Poartă. 27 Moreover, this medieval notion would not have been used in nineteenth-century diplomatic terminology in western Europe, with the exception of definitions of the relations between the Ottoman Empire and Bulgaria in the years 1878–1909 and between Great Britain and South Africa. See J.L. Brierly, The Law of Nations: An Introduction to the Inter- national Law of Peace (Oxford, 1963; 1928), 133–136. Several manuscripts from the archives of the French minister of foreign affairs were concerned with the vassal provinces and countries of the Ottoman Empire: Mémoires et documents divers sur les provinces et sur les the legal and political status of wallachia and moldavia 19 states” idiom to characterize the political status of the Danubian principalities toward the Porte in the nineteenth century. It was a complementary formula to the notion of suzerainty that defined a relation of dependency between a stronger state and a weaker one.28 According to Arthur de Claparède, secretary of the Swiss legation in Vienna and author of a work on diplomatic law, the “nominal vassalage status” applied to the Danubian principalities was defined by the fact that the state was an Otto- man subject, but de facto sovereign; possessing the rights to wage war and to conclude peace, to negotiate, and to be represented in foreign countries by legation.29 It was the next step after their usage for the contemporary situation that these terms began to be used to depict the power relations of the preceding centuries. The Romanian historians adopted, in their turn, the concepts of “vassalage” and “suzerainty” to define the relationship between sultans and voievods, without uniformity concerning chronology or even the content of the term.30 It must be noted that these concepts were inappropriate to the Ottoman world. As a matter of fact, before the Treaty of Adrianople in 1829, an equivalent to the Western term “suzerainty” cannot be found in Ottoman documents. It was only after 1829 that this term was used

pays vassaux de l’Empire ottoman: originaux, copies et imprimés des XVIIIe et XIXe siècles sur la période de 1700 à 1839 (Paris, Archives de Ministère des Affaires Étrangères [henceforth AMAE]), Mémoires et documents, Tome 13; Mémoires et documents sur la Moldavie et la Valachie: minutes, originaux, copies et imprimés du XIXe siècles sur la période de 1825 à 1855, AMAE, Mémoires et documents, Tome 48. 28 In this respect, Henry Wheaton spoke about “the Principalities of Moldavia, Wal- lachia and Serbia, under the suzerainty of the Ottoman Porte and the protectorate of Russia.” See Henry Wheaton, Commentaire sur les éléments du droit international et sur l’histoire des progrès du droit des gens (Leipzig, 1873), vol. 3, 36. 29 Arthur de Claparède, Essai sur le droit de représentation diplomatique d’après le droit international moderne (Geneva, 1875), 113. 30 For instance, some of them came to “Westernize” the political status of the Prin- cipalities of Wallachia and Moldavia by adopting the notion of vassalage for the period after 1538—rather paradoxically, as this was actually the period during which Ottoman habits began to penetrate the Danubian territories on a massive scale. Thus, an ambiguous image was created about the stages of the relationship between the Porte and the two tributary principalities through their definition; first according to Ottoman criteria and terms (for instance, tributaries) and then according to medieval Western ones (vassals). See, among others, Gheorghe Zagorit, “Stabilirea suzeranităţii turceşti în Moldova: Cu argumente că prima capitulaţie atribută lui Bogdan III a fost facută de Ştefan cel Mare la 1497” [The stabilization of the Turkish suzerainty in Moldavia: With arguments that the first capitulation attributed to Bogdan III was made with Stephen the Great], Convor- biri Literare 48 (1914): 710–728; N.A. Constantinescu, Începuturile şi stabilirea suzeranităţii turceşti în Moldova [The beginnings and stabilization of the Turkish suzerainty in Molda- via] (Bucharest, 1914). 20 viorel panaite abundantly in the translations of official Ottoman documents to Euro- pean languages.31 Numerous tributary or non-tributary regions in the Ottoman Empire enjoyed more or less limited self-government, the extent of this self- government was dependent on historical, geographical, political, diplomatic, and military circumstances. Nonetheless, the notion of “autonomy” was a creation of modern historians and jurists to define this reality. No serious specialist would contest the autonomy of Wallachia and Moldavia; at the same time, it is necessary to stress that this autonomy could exist only within another state—in this case, the Ottoman Empire—and not outside of it. It is also important to emphasize that there was no contradiction between the terms “tributary,” “protection,” “vassalage,” and “autonomy.” In a sense, they were complementary, characterizing the same reality from different points of view. The notion of “tributary” defines legal and political status from the perspective of the voievods’ duties, while “protectorate” characterizes the same status from the point of view of the Porte’s responsibilities. “Vassalage” and “autonomy” describe the two aspects as a whole; in fact, the latter term means the protection the sultan provided in return for the infidels’ acceptance to pay a tribute. In my opinion, a more suitable method to find a proper terminology for the legal and political status of Wallachia and Moldavia within the pax ottomanica would be to return to the sources rather than invent notions beyond them. Furthermore, it is necessary to approach the subject not from a nationalistic political perspective, but from the vantage point of Islamic law as practiced and amended by Ottoman legislators and rulers. The Ottoman system of government was not uniform throughout the empire. Realizing this, early European observers and then modern historians frequently tried to decipher, classify, and analyze it, while emphasizing the pragmatism of Ottoman statesmen. Some historians tried to

31 For instance, in the French translation of the ferman of 24 October 1866 that acknowledged Prince Charles of Hohenzollern and imposed on him the duty to observe Ottoman suzerainty over the Unified Principalities as part of the Ottoman Empire, the concept of “suzeraineté” is ubiquitous: “À respecter dans leur intégrité mes droits de suzeraineté sur les Principautés-Unies qui font partie intégrante de mon Empire, dans les limites fixées par les stipulations des anciennes Conventions et par le Traité de Paris de 1856.” This ferman was granted after the Grand Vizier Ali Pasha’s notice of 19 October 1866 to Prince Charles, which included the conditions of his sanction as the hereditary prince of the Unified Principalities, and after Prince Charles’ affirmative answer of 10 October 1866 (see Gabriel Noradounghian, Recueil d’actes internationaux de l’Empire Ottoman (Paris, 1897), vol. 3, 258). the legal and political status of wallachia and moldavia 21 offer us a global picture of the Ottoman system of government. In Albert Howe Lybyer’s opinion, the Ottoman Empire consisted, first, of a great body of lands which were directly administered according to a system that was exceedingly intricate but approximately uniform; second, of a number of regions less directly administered under special regulations; third, of numerous tributary provinces; and fourth, of certain protected or vassal states.32 In my opinion, despite the differences of administration or government, and the degree of subordination, autonomy or suzerainty, all these areas—from the lands under direct administration to the tributary “states”—constituted the Ottoman Empire. Ottoman pretensions to universal sovereignty—be it by secular or divine right—reached their height during Süleyman the Magnificent’s rule (1520–1566).33 It was during this time that the Ottoman view on the legal and political position of the princes, inhabitants, and territories north of the Danube was fully drawn. Süleyman was the first sultan who firmly stated his rights as a landlord over Wallachia and Moldavia, emphasizing in letters addressed to Christian sovereigns that their territories belonged to the “Well-protected dominions” (Memalik-i mahruse), their inhabitants were the sultan’s subjects (re‘aya), and their princes his tribute-payers, who were appointed to rule Ottoman provinces (vilayet) as governors (bey). As the strongest argument, let us quote from a letter dispatched by Süleyman Kanuni to King Sigismund I of Poland in the spring of 1531— that is, five years after the battle of Mohács (1526) and seven years before the military campaign toward Moldavia (1538): The voievod of Moldavia, as well as that of Wallachia, is my slave and tribute-payer (kulum and haracgüzarımdır); their possessions are included among our other Well-Protected Dominions (Memalik-i mahrusemiz), like Bosnia and Semendria, which are my estates (. . .) We have been informed that the above-mentioned voievod [Petru Rareş of Moldavia] dispatched an

32 William Eton, A Survey of the Turkish Empire (London, 1798), 287–288, 297; Thomas Thornton, The Present State of Turkey (London, 1809), vol. 2, 298–307; Albert Howe Lybyer, The Government of the Ottoman Empire in the Time of Suleiman the Magnificent (Cam- bridge and Oxford, 1913), 29; D.E. Pitcher, An Historical Geography of the Ottoman Empire from Earliest Times to the End of the Sixteenth Century (Leiden, 1972), 124–134; Metin Kunt, “Sixteenth-Century Ottoman Policies and Problems: Introduction,” in Süleyman the Mag- nificent and His Age: The Ottoman Empire in the Early Modern World, ed. Metin Kunt and Christine Woodhead (London and New York, 1995), 33–38. 33 On all these juridical and political changes, see Colin Imber, Ebu’s-su‘ud: The Islamic Juridical Tradition (Stanford, 1997), 65–79. According to him, around 1500, the Ottomans still limited their pretensions “to a claim to rightful sovereignty in Anatolia and the Balkan peninsula” (ibid., 74). 22 viorel panaite

envoy to you. Who is he to dare to have the audacity to send an emissary to you? He and the voievod of Wallachia are my tribute-payers and slaves. Consequently, we issued severe orders so that they should not dare to dispatch envoys, no matter to whom; no emissary is allowed to be sent to them either; if someone has a question to them, he must appeal to our powerful Porte, which is open at all times, and state what he needs.34 As a matter of fact, this letter is only one of a series of documents from the third and fourth decades of the sixteenth century; documents that reflect the changes in the power relations in Central Europe and also a new Ottoman legal and political view on the status of the Danubian principalities.35 One can conclude that Moldavia and Wallachia, from that moment, stopped being a subject and became merely an object of international law. The new power relations were, increasingly, legitimated through concepts and rules from classical Islamic law, which enjoyed an increasing recovery in the Ottoman world after the conquest of the Arab lands. When Ibrahim al-Halebî (d. 956/1549)36 came to Istanbul from Egypt after its conquest by Selim I in 1517, the writings on Islamic law multiplied and their quality improved.37 Also, the same period was marked by the

34 Letter of 18 April–17 May 1531, Grigore George Tocilescu and Alexandru I. Odobescu, eds., Documente privitoare la istoria românilor culese de Eudoxiu de Hurmuzaki [Docu- ments of the history of the Romanians collected by Eudoxiu de Hurmuzaki], Supple- ment, vol. 2, part 1, 1451–1575 (Bucharest, 1891) [henceforth Hurmuzaki, Supl. II/1], doc. IX. The same phrases can be found in a letter of evvel Cemâzi’levvel 938 / 10–19 January 1531: “İmdi sizinle sabıkda makrur olan mu‘ahede-yi hümayunum. Mezbur Boğdan voyvodası kulum ve harac-güzarımdır” (Archiwum Glówne Akt Dawnych, Warsaw [henceforth AGAD], AKW-Tureckie, K. 67, t. 33, no. 75. Likewise, in a letter from the fall of evasıt Rebi’levvel 938 / 23 October–1 November 1531, the Grand Vizier İbrahim Pasha reinforced this official image, writing to Sigismund I in similar terms (AGAD, AKW-Tureckie, K. 67, t. 32, no. 72). A different date is indicated in Hurmuzaki, Supl. II/1, doc. XIV: 12 October– 10 November 1531. 35 See Zygmunt Abrahamowicz, Katalog dokumentów tureckich : Dokumenty do dziejów Polski i krajów osciennych w latach 1455–1672 / Catalogue des documents turcs: Documents concernant la Pologne et les Pays Voisins de 1455 à 1672 (Warsaw, 1959). 36 Joseph Schacht, “al-Halebî,” in Encyclopaedia of Islam, ed. Peri J. Bearman et al. (henceforth EI2) (Leiden, 1979), vol. 3, 90; Ahmed Özel, Hanefi Fıkıh Âlimleri (Ankara, 1990), 114–115. 37 Ibrahim al-Halebî’s work was translated several times from Arabic to Turkish, first by Mehmed Tahir b. Mehmed Râhîmî (d. 1065/1654–1655) in the first half of the seventeenth century (Mehmed Tahir b. Mehmed Râhîmî, Tercüme-i Mültekâ’l-ebhür (Topkapı Sarayi Müzesi Kütüphanesi (henceforth TKSMK), E.H. 788; ms. of 637 folios, copied in 1221/1806–1807). the legal and political status of wallachia and moldavia 23 famous şeyh ül-Islams Zenbilli Ali Efendi (1503–1526),38 Kemalpaşazade (1526–1534),39 and Ebussuud Efendi (1545–1574),40 whose juridical opinions (fetvas) took a more rigorous approach to the relationship between Muslims and non-Muslims than ever before, and included notions such as sovereignty (hükumet), conduct of state (siyer), holy war (cihad), protected peoples (zimmi), tribute (harac), poll-tax (cizye), and foreigners (müste‘min). Certain juridical texts on the pacts with tributary princes are relevant for the changes in the political and juridical status of Wallachia and Moldavia. Let us quote here the famous Hanafi jurist Muhammad ash-Shaybani (d. 189/805), who stated that it would be illicit to conclude an agreement with a non-Muslim prince and permit the latter to maintain unlimited powers over the people of his realm. These kinds of agreements had nevertheless been in practice in the frontier zone bordering the Christian world; the dilemma was what the Muslim attitude toward these agreements should be. Was it legal to violate this kind of “peace,” or should only some of its stipulations have been abrogated? Let us quote once again the Siyar of ash-Shaybani:

38 Zenbilli Ali Efendi had been in office as şeyh ül-Islam between 1503 and 1526, the year he died. I have consulted the manuscript Fetava-yi Ali Efendi (Süleymaniye Kütüphanesi, Fatih 2390). 39 Kemalpaşazade (1468–1535), in the capacity of kadıasker, attended Selim I’s campaign to Egypt in 1516–1517, and accompanied Süleyman the Magnificent to Hungary in 1526. İbn Kemal studied Islamic law with Molla Lutfi (d. 1494), was şeyh ül-Islam between 1526 and 1534, and wrote many juridical and religious works, including some legitimating the jihad against the Safavids of Iran. See İsmail Hâmi Danişmend, Izahlı Osmanlı Tarihi Kronolojisi [A detailed Ottoman chronology] (Istanbul, 1947–1948), vol. 2, 430–431; Halil İnalcık, The Ottoman Empire: The Classical Age 1300–1600 (New York and Washington, DC, 1973), 173–174, 178; Cronici turceşti, vol. 1, 191–192; Richard C. Repp, The Müfti of Istanbul: A Study in the Development of the Ottoman Learned Hierarchy (Oxford, 1986), 224–240. 40 Ebussuud Efendi (898–989/1492–1574) was şeyh ül-Islam for thirty years, between 1545 and 1574, and is considered one of the greatest Ottoman holders of this office. His fetvas were taken as examples during the seventeenth and eighteenth centuries (Repp, The Müfti of Istanbul, 272–304; Hasan Basri Erk, Meşhur Türk Hukukçuları (Célébres juristes turcs) (Istanbul, s. a.), 117–139; Colin Imber, Ebu’s-su‘ud. Some of his fetvas are published: M.E. Düzdağ, Şeyhülslam Ebussuud Efendi fetvalar ışıgında 16. asır türk hayatı [Sixteenth- century Turkish life in the light of Şeyhülslam Ebussuud Efendi’s fetvas] (Istanbul, 1983); Friedrich Selle, Prozessrecht des 16. Jahrhunderts im Osmanischen Reich: Auf Grund von Fet- was des Scheichülislame Ebüssuud und anderer unter des Regierung des Sultans Süleiman des Prächtingen (Wiesbaden, 1962); Ahmed Akgündüz, Osmanlı Kanunnâmeleri [Ottoman legal books], vol. 4, Kanunî Sultan Süleyman Devri Kanunnâmeleri [Ottoman legal books from the age of Sultan Süleyman], part 1, Merkezi ve Umumî Kanunnâmeler [Central and general law books] (Istanbul, 1992), 40–100. 24 viorel panaite

I asked, [the Muslims] did so and entered into such an agreement with him [the non-Muslim prince] on such terms and he became a protected person of theirs, [what would the ruling be]? He replied: [The Muslims] would look into those terms of the agreements that were illegal and improper; they would abrogate them and would observe only those terms that were proper.41 In my opinion, the above-mentioned Islamic rules provided legal justification for the Ottomans to modify the juridical and political status of the Principalities of Wallachia and Moldavia during the reign of Süleyman the Magnificent and its development in the following centuries.42 Two notions were commonly used in Ottoman sources to depict the position of voievods, inhabitants, and territories of Wallachia and Mol- davia: tribute (cizye or harac) and protection (himayet or siyanet). Since the Wallachian and Moldavian voievods acknowledged their allegiance to the Porte in the fifteenth century, they became “tribute-payers” (sing. haracgüzar). Until the nineteenth century this concept was most frequently used in Ottoman and non-Ottoman sources to define the legal and political status of voievods, inhabitants, and territories of Wallachia and Moldavia.43 In a complementary manner, tribute-paying implied Ottoman protection, which ensured that local princes could go on ruling over their territories and the internal customs of their countries would be

41 Another quote is also enlightening: “I asked: If he [the ruler of the infidels] did not become a Muslim nor was given the benefit of a peace agreement, nor did he become a zimmi, but he proposed to the Muslims [a peace agreement] on condition that he be a protected person and pay the Muslims a tribute (harac), on condition that he be allowed to exercise over the people of his realm such powers as he wished, such as those of behead- ing or crucifying, or the like, which are not proper that he should exercise in the territory of Islam, [would such an agreement be made]? He replied: It would not be right for the Muslims to make any peace agreement with him on such conditions.” Both quotes (supplemented with further explanatory additions) are from The Islamic Law of Nations: Shaybani’s Siyar, trans. Majid Khadduri (Baltimore, 1966), 152–153. 42 Panaite, The Ottoman Law of War and Peace, 335–338. 43 Even Western sources utilized these terms, as in the reports of French ambassadors in Istanbul. For instance, Michel de Codignac called the Moldavian and Wallachian voievods “tributaries of the sultan” (tributaires du Grand Seigneur) in 1554; in 1612, Achille de Harlay wrote that “all three provinces are from ancient times tributaries of the sultan” (“toutes ces troys provinces sont depuis un long temps tributaires du Grand Seigneur”); and in 1702 M. de Ferriol noted that the “word tributaries is for Wallachians and Molda- vians” (“le mot de tributaires est pour les Valaques et pour les Moldaves”) Grigore George Tocilescu and Alexandru I. Odobescu, eds. Documente privitóre la Istoria Românilor [Docu- ments for the history of the Romanians], Supplement, vol. 1, part 1, 1518–1870 (Bucharest, 1886) (henceforth Hurmuzaki, Supl. I/1), doc. XVII (letter of 3 April 1554 to Henry II); doc. CCXXVII (report of 21 April 1612 from Pera); doc. DXXVII (report of 2 October 1702 to the King of France). the legal and political status of wallachia and moldavia 25 preserved. “The country is under our due protection, and its subjects pay me tribute,” states an “imperial diploma” (berat-i hümayun) of 1586.44 Let us emphasize that the concepts “tribute” and “protection” were also used in relation to Wallachian and Moldavian princes during the fifteenth century, but—according to the Ottoman law of peace—the attribute “temporary” had to be applied to them. In my opinion, it would be wrong to adopt, uncritically, the juridical terminology of sixteenth- to eighteenth-century sources, and use it retroactively to depict fifteenth- century power relations toward Wallachian and Moldavian princes, as certain Ottoman chroniclers such as İdris Bitlisi, Kemalpaşazade and others were inclined to do. From the time of Süleyman’s reign, “tribute” and “protection” should be defined as “permanent,” characterizing a “firm and lasting peace” (according to the formulation of Dimitrie Cantemir). To show what was behind the notions of “tribute-payer” (haracgüzar) and “protection” (himayet), I draw a synoptic picture of the place of voievods, inhabitants, and territories of Wallachia and Moldavia within the pax ottomanica system.45 Throughout the fifteenth to eighteenth centuries, various Christian and Ottoman titles were used to show the rank and status of the rulers of Wallachia and Moldavia, including tekfür, hakim, bey, voyvoda, hospodar, domn, kıral, ban, prince, duke, etc.46 Originally, some of these terms were specific to Ottoman or non-Ottoman sources; later, however, they borrowed from each other and the concepts appeared in both kinds of sources. From the beginning of the sixteenth century, the prevalent term used for the rulers of Wallachia and Moldavia in Ottoman sources was that of voyvoda. This title was adopted by Ottomans from Slavic and used, over time, with the meaning of “governor”; in some contexts, the position was seen as being synonymous with the subaşı, a provincial official in charge of collecting revenues.47

44 Granted by Murad III to the Voievod Mihnea II the Turkified of Wallachia, in 994/1586 (Maxim, Culegere, doc. 14). 45 See also Panaite, The Ottoman Law of War and Peace, 338–339. 46 Gustav Bayerle, Pashas, Begs, and Effendis: A Historical Dictionary of Titles and Terms in the Ottoman Empire (Istanbul, 1997). 47 According to Dimitrie Cantemir, voyvoda was a word “signifying the General of an army, but by the Poles given to Governors of Provinces. The Turks are wont to call by this name the Christian Princes of Moldavia, Walachia and Transylvania, who are tributary to them, especially in the imperial Mandates and Letters of the Prime Vizir, as Boğdân vayvodası, Iflâk or Erdel vayvodası” (Cantemir, Othman Empire, 193n38). 26 viorel panaite

Beginning with the rule of Süleyman, the nomination of a new tributary prince implied transferring to him—as would happen also in the case of Ottoman governors—a part of the sultan’s prerogatives, which were, according to official documents, the following: “to rule and govern the country,” “the protection and safeguarding of the subjects,” and “the defense of the country.” Voievods were included in the hierarchy of Otto- man dignitaries, and had the privilege of at least two horsetails (tuğ). They were invested by imperial diplomas (berat-i hümayun) and were ordered by hüküms, like any Ottoman beylerbeyi, sancakbeyi, or kadi.48 The status of tribute-payers (haracgüzar) implied that they observed imperial orders and fulfilled all political, military, and financial duties (hizmet) commanded of them by the Porte. Relevant to the status of a tributary prince were, first, the payment of the harac, which was also seen as “collecting poll-taxes” (“cizyelerinin tahsili”) and, second, the political and military duties, which were generally expressed by the phrase “be friend of our friends and enemy of our enemies” (“dosta dostumuza ve düşman düşmanımıza olub”). The latter implied, as a rule, that they adopt the same foreign policy as the Porte; participate with an army in imperial campaigns; support the Ottoman governors from the neighboring provinces; and inform the central and local Ottoman authorities about any hostile intentions or actions of the neighboring Christian states against the Porte.49 As noted, Wallachians and Moldavians, as inhabitants of the “house of war” and infidels, were called “enemies” (sing. harbi) or “enemy infidels” (harbi küffar) until the fifteenth century. At the same time, the term mu‘ahid—used for non-Muslims who had concluded a “covenant” with Muslims (“people of the covenant,” otherwise also called as ehl ül-‘ahd)— was only applied to Wallachian and Moldavian rulers. Later, during the sixteenth and seventeenth centuries, it was extended to entire populations, just like the notion “tribute-payer” (haracgüzar). This concept survived also into the eighteenth century, as represented by the fetva of 1717, according to which the Wallachians were defined as “infidels

48 For details, see Panaite, The Ottoman Law of War and Peace, 340–376; idem, “Power Relationships in the Ottoman Empire: The Sultans and the Tribute-paying Princes of Wal- lachia and Moldavia from the Sixteenth to the Eighteenth Century,” International Journal of Turkish Studies 7 (2001), 26–53. 49 For details, see Panaite, The Ottoman Law of War and Peace, 377–405; idem, “The Voyvodes of the Danubian Principalities—as Harâcgüzarlar of the Ottoman Sultans,” in Ottoman Borderlands: Issues, Personalities and Political Changes, ed. Kemal H. Karpat and Robert W. Zens (Madison, WI, 2003), 59–104. the legal and political status of wallachia and moldavia 27 who concluded an agreement with Muslims” (“keferesi ehl-i İslam ile mu‘ahidin olan”).50 In the Ottoman legal view, if an enemy (harbi) accepted to pay tribute and to obey the sultan, he became a zimmi, i.e., a non-Muslim protected subject. In Ottoman sources, he was also called a “tribute-payer” (harac- güzar) or an “infidel who is not from the house of war” (pl. “kefere ki dar ül-harb’dan olmıya”). After the rule of Süleyman, sultans referred to Moldo-Wallachians as their protected people (zimmi), who constituted the non-Muslim part of “tax-paying subjects” (re‘aya). In the terminology of Ottoman official sources, the inhabitants of the tributary principalities were successively assimilated to the sultan’s subjects on the right side of the Danube. The phrase in a document dating from 1017/1608, “the subjects of Moldavia are on the whole the same as the subjects of my Well- Protected Dominions” (“bi’l-cümle Boğdan re‘ayası sa’ir gibi Memalik-i mahrusem re‘ayası gibi olub”), became standard and was extended to Wallachians, Transylvanians, and Ragusans as well. Thus, it was affirmed that “the inhabitants of Dubrovnik are my tax-payer subjects just like my other subjects” (“Dubrovniklü sa’ir re‘aya gibi haracgüzar re‘ayalarımdır”). Consequently, their position was defined as similar to that of the Ottoman subjects on the right side of the Danube and distinguished from that of the people in the neighboring Christian states (considered harbi in the Hanafi view).51 The official Ottoman position was clearly outlined in a post-1672 order of Mehmed IV (1648–1687) dispatched to the Ottoman authorities along the western shore of the Black Sea: “there is no imperial permission from me that [the Wallachian population] should be treated like the Polish; the Wallachian is not an enemy population; it is the same as the population of my other Well-Protected Dominions” (“Eflak re‘aysı harbi değildir sa’ir Memalik-i mahrusemde olan re‘aya gibidir”).52 Even stronger evidence is offered by some unpublished juridical opinions ( fetvas) issued by the mufti Ali Efendi Akkermanlı (d. 1621), who defined Moldavians as

50 This fetva was included in a hüküm of evail Cemazi ül-evvel 1129 /13–22 April 1717 (Veliman, Relaţiile româno-otomane, doc. 24). 51 Tahsin Gemil, Relaţiile Ţărilor române cu Poarta otomană în documente turceşti, 1601– 1712 [The contacts of the Romanian countries with the Ottoman Porte in Turkish documents, 1601–1712] (Bucharest, 1984), doc. 14. See also, doc. 32: a chancery record of 1012/1604 referring to the Wallachian subjects; Nicolaas H. Biegman, “Ragusan Spying for the Otto- man Empire: Some 16th Century Documents from the State Archive at Dubrovnik,” Belleten 27 (1963), 106, 238. 52 Gemil, Relaţiile Ţărilor române cu Poarta otomană, doc. 155. 28 viorel panaite re‘aya and “protected peoples” (ehl üz-zimmet).53 As a result of their zimmi position, Wallachians and Moldavians enjoyed basically the same rights regarding their life, liberty, possessions, and religion as other subjects of the empire, unlike those with an enemy (harbi) status. The Ottoman state’s policy toward the re‘ayas was guided by the principle of “tax-payer protection”: not an abstract principle, but a concrete necessity in Ottoman society. As this policy was extended to the inhabitants of tributary principalities, imperial orders frequently stressed the voievods’ duty to work for the “peace and security of the subjects,” the “welfare and contentment of the re‘ayas,” and the “defense and protection of the re‘ayas.” Imperial orders to local Ottoman authorities emphasized three main aspects of zimmi protection, extended also to Wallachians and Moldavians: the prohibition of enslavement, the prohibition of plunder, and the protection of religion. To cause any damage to the tribute-payer subjects (haracgüzar olan re‘ayası) would be contrary to the şeri‘at and the kanun.54 Both Ottoman and non-Ottoman sources used a diverse legal, political, and administrative terminology in relation to the status of Wallachia and Moldavia in the Ottoman Empire. In Ottoman chancellery terminology, two main trends can be found. On one hand, they had a tendency for standardization; accordingly, the tributary principalities had the same status as other Ottoman provinces. Beginning with Süleyman’s reign, Wallachia and Moldavia were legally and politically considered to be the “patrimony of the sultan” (mülk-i mevrus), territories from the “house of Islam” (dar ül-İslam), and parts of the Ottoman Empire (Devlet-i Aliyye / “The Exalted State,” or Memalik-i mahruse / the “Well-protected dominions”).55 These formulas were invoked by the Porte in order to reject the political claims of neighboring Christian rulers. On the other hand, Ottoman chancelleries also used a specific terminology that stressed the autonomous position of Wallachia and Moldavia. In this vein, the terms most frequently applied to these countries emphasized their distinct status: “house of tributary protection” (dar ül-zimmet), “freedom” (serbestiyet), and “privileged provinces” (eyalât-i mümtaze)—all of these concepts were generally used to

53 He worked in the Ottoman territories northwest of the Black Sea and his works survived in Fetava-yi Akkirmani Maa Üskubi (Süleymaniye Kütüphanesi, Hkm. 405, f. 90b) and Fetava-yi ‘Ala Âkkermani (TKSMK, A 842). 54 For details, see Panaite, The Ottoman Law of War and Peace, 406–445; idem, “The Re’ayas of Tributary-Protected Principalities (the 16th–18th Century),” International Jour- nal of Turkish Studies 9 (2003), 79–104. 55 According to certain historians, all these formulas were used to express Ottoman suzerainty only: İsmail Hakki Uzunçarsılı, Osmanlı Tarihi [Ottoman history], vol. 4/1 (Ankara, 1956), 6; Veliman, Relaţiile româno-otomane, 13. the legal and political status of wallachia and moldavia 29 designate various financial, administrative, religious, and legal autonomies in the Ottoman Empire. These formulas were instrumental for the Porte and also for Moldo-Wallachians to invoke the protected status of these territories and inhabitants against the interference of Ottoman officials from neighboring sancaks. One of the most controversial questions relates to the kind of Islamic juridical realm into which Wallachia and Moldavia were included. The Shafi‘i concept of “house of peace”/“house of pact” (dar ül-‘ahd) was frequently adopted by modern historians and jurists in order to define tributary status, as it was considered useful to cover an intermediary reality, full of contradictions, situated between dar ül-İslam and dar ül-harb.56 I avoid using it to characterize the legal status of Wallachia and Moldavia for two reasons. First, the concept of dar ül-‘ahd did not belong to the Hanafi juridical school adopted by the Ottomans and consequently it was not theorized in their legal sources. The famous Shafi‘i scholar al-Maverdi (d. 450/1058) theorized about the concept of dar ül-‘ahd in a chapter dedicated to the classification of the territories coming under Muslim rule. So, under the term dar ül-‘ahd (in E. Fagnan’s French translation, Territoire du traité), al-Maverdi understands those territories that, based on a contractual pact, remained in non-Muslim possession, but paid in exchange a tribute (harac) equivalent to a capitation (cizye). But al-Maverdi himself admits that he did not find such a concept in the Hanafi works and emphasizes that Abu Hanifa included the above-mentioned territories in the “house of Islam”: “Abu Hanifa says their territory becomes dar ül-Islam by the treaty (bi’l sulh dar ül-İslam’a), they themselves become zimmis, and cizye is taken from them.”57 Second, the single concept of the “house of peace” (dar ül-‘ahd) is not acceptable to describe the legal status in

56 For example, it was used to define the legal status of the tributary states in South- eastern Europe, including Wallachia, Moldavia, Transylvania, Ragusa, and certain Aegean islands. See Biegman, The Turco–Ragusan Relationship, 30–32; Halil İnalcık, “Dâr al-‘ahd,” in EI2, 118–119; Aurel Decei, “Tratatul de pace—sulhnâme—încheiat între sultanul Mehmed al II-lea şi Ştefan cel Mare la 1479” [Peace treaty—sulhname—concluded between Sultan Mehmed II and Ştefan cel Mare in 1479], Revista Istorică Română (Bucharest) 15, no. 4 (1945): 465–494; Ion Matei, “Quelques problemes concernant le regime de la domination ottomane dans les Pays roumains (concernant particulièrement la Valachie),” Revue des Études Sud-Est Européennes 10 (1972), 66–81, and 11 (1973), 81–95; Maxim, Ţările Române şi Înalta Poartă, 143–168. 57 Mawerdi (Aboul-Hasan Ali), Al-Ahkâm al-Sultâniyah (Les Statuts Gouvernementaux ou règles de droit public et administratif ), trans. and ed. Edmond Fagnan (Algiers and Paris, 1915), 290–291. See also the English edition: al-Mawardi, Al-Ahkam as-Sultaniyyah: The Laws of Islamic Governance, trans. Asadullah Yate (London, 1996), 201–202, and Decei, “Tratatul de pace,” 137. 30 viorel panaite regard to the Porte because of the many historical changes that occurred throughout the long period of Ottoman–Romanian relations. By the criterion of sovereignty, the Hanafi jurists ash-Shaybani (d. 805) and as-Sarakhsi (d. 1090) defined two categories of territories: dar ül- muvada’a (“house of reconciliation,” or “house of truce”) and dar ül- zimmet (“house of submission,” or “house of tributary protection”). These concepts did not designate territories outside the “house of Islam” or the “house of war,” but rather distinct areas included in both basic spheres. In official Ottoman documents, the use of the terms muvada‘a, ehl-i zimmet (zimmi), cizye, and harac imply that these theoretical concepts of dar-i muvada‘a and dar-i zimmet were also applied in praxis. It is our task to find the correspondence between legal formulas and the political situation of the two tributary principalities. In my opinion, considering their incomplete and temporary submission until the reign of Süleyman the Magnificent, the principalities of Wallachia and Moldavia should be considered part of the “house of reconciliation” (dar ül-muvada‘a), because their rulers concluded temporary arrangements with the stipulation that they would pay the capitation (called cizye or harac) to the Islamic ruler, and, in return, the plundering raids against their countries would cease. From Süleyman’s reign on, Wallachia and Moldavia should be considered—according to the Ottoman legal view— part of the “house of tributary protection” (dar-i zimmet), a section of the “house of Islam,” because their inhabitants accepted to pay tribute and obey the Muslim power; they also obtained the privilege of being ruled by a native leader, chosen and appointed by the sultan. In order to complete the picture of the position of the tributary principalities within the pax ottomanica system, it is necessary to take into consideration the autochthonous and the Western European perspectives as well. From the age of Süleyman, the Wallachian and Moldavian princes and nobles had increasingly come to recognize that their supreme ruler resided in Istanbul and it was him to whom their countries belonged. On 24 April 1537, with a Polish expedition impending against Moldavia, Petru Rareş clearly affirmed in a letter to the Polish king that “This country belongs to the all-high and victorious Turkish emperor, my all-generous lord, and he has given it to us to rule over it: long live His Imperial Majesty (. . .) And we hope that the Turkish army will come to defend the Emperor’s country!”58

58 “Ţara aceasta e a prea-înălţatului şi nebiruitului Împărat turcesc, domnul mieu cel prea-milostiv, şi ni-a dat-o nouă s’o stăpânim: să trăiască Măria Sa Împărătească (. . .) şi the legal and political status of wallachia and moldavia 31

Similarly, the European powers considered and treated Wallachia and Moldavia as provinces of the Ottoman Empire. In this respect, the diplomatic and consular reports are significant. In 1612, Achille de Harlay, the French ambassador to the Porte, noted that “all of these three provinces are long time tributaries of the Grand Seigneur.”59 In fact, Western observers used similar terms to characterize the tributary status of Ragusa within the Ottoman Empire. In a 1766 report to the French king, Louis XV, the consul André Alexandre le Maire wrote about the “total and formal subjection of the Ragusans to the Ottoman Porte” and described the juridical status of the inhabitants of Dubrovnik as “a subject people to their sovereign” and the sultan’s re‘aya (“this term does not imply a tributary but a subject as well”).60 By the late eighteenth and early nineteenth century, European observers made fewer distinctions among the Ottoman provinces. In his work on the Turkish Empire published in 1798, William Eton spoke about “the pashaliks or governments most immediately connected with the seat of the empire” and “the more distant provinces,” in which category he also added Egypt, Wallachia, and Moldavia, noting that “they may be considered connected with the Porte rather by treaty than as integral parts of the empire.”61 A few years later, in The Present State of Turkey published in 1809, Thomas Thornton presented the general System of Turkish Government toward the Tributary Subjects, explaining the Turkish government institutions put into place when they abolished native institutions after their conquests as a consequence to their new conquests: While the Turkish power was in a state of progressive aggrandizement, it was the constant policy of the government to expel the nobles and great landed proprietors from those countries that they have incorporated with their empire, and to make a new division of the lands according to the arrangements of their peculiar civil and military system.62

mai nădăjduim că pentru apărarea ţerii Împăratului va veni oaste turcească!” (Nicolae Iorga, Scrisori de boieri, scrisori de domni [Writings of the boyars, writings of the princes] (Vălenii-de-Munte, 1925), doc. XXIII). 59 Report of 21 April 1612, from Pera (Hurmuzaki, Supl. I/1, doc. CCXXVII (toutes ces troys provinces sont depuis un long temps tributaires du Grand Seigneur). 60 Zdenko Zlatar, Between the Double Eagle and the Crescent: The Republic of Dubrovnik and the Origins of the Eastern Question (Boulder, CO, and New York, 1992), 1. 61 Eton, Turkish Empire, 287–288, 297. 62 Thornton, The Present State of Turkey, vol. 2, 298–307. 32 viorel panaite

According to Thornton’s view, this was the rule. In some cases, the Otto- mans applied “exceptions to the usual mode of Turkish government,” and this is how the tributary provinces of the empire, such as Egypt, Walla- chia, and Moldavia were created. With the Islamic Ottoman terminology in mind, and in order to avoid the confusion caused by using Western concepts in mélange with Islamic notions, I prefer to group the diverse historical and juridical terminology under a single idiom, i.e., tributary-protected provinces (principalities), to define the legal and political status of Wallachia and Moldavia. It is necessary to point out, however, that regardless of which concept is used— tributary states, tributary provinces, autonomous states, autonomous provinces, tributary principalities or buffer-protectorates—all of them mirror the organizational peculiarities of the pax ottomanica system.63

Customs and Covenants

The overriding issue for Romanian historians has long been to find the so-called “treaties” which were supposed to have regulated the legal status of Wallachia and Moldavia vis-à-vis the Porte over the long term; thus these historians have almost completely ignored the problem of customary law in the relations between the Ottoman state and the two Danubian principalities. In fact, given the huge amount of documentary information and primarily the Ottoman sources of the sixteenth to eighteenth century, custom—and not the supposed long-term “covenants”—should be seen as the main legal source for the status of Wallachia and Moldavia. There are three main cases in which custom (adet, kadim, mu‘tad, ayin, etc.) was invoked in historical sources: 1) Ottoman ceremonial usages or payments established in connection with the voievods’ enthronement; 2) confirma- tions of political, military, and financial duties of the tribute-payer voievods; 3) reinforcements of specific rules of self-government in Wallachia and Moldavia. Some sources register the mechanisms of the formation of customs. These refer primarily to setting up new obligations and increasing the sum of old payment duties—a practice that began in the second part of

63 For details, see Panaite, The Ottoman Law of War and Peace, 446–484; idem, “Wal- lachia and Moldavia according to the Ottoman Juridical and Political View, 1774–1829,” in Ottoman Rule and the Balkans, 1760–1858: Conflict, Transformation, Adaptation, ed. Antonis Anastasopoulos and Elias Kolovos (Rethymno, 2007), 21–44. the legal and political status of wallachia and moldavia 33 the sixteenth century and reached a climax in the eighteenth century. On one hand, while competing for the throne, the rivals extended old obligations, adopted some Ottoman customs, and even assumed new duties. On the other hand, the sultans’ orders (hüküm) could impose Ottoman traditions on the tributary principalities, or establish new practices in their relationship toward the empire.64 After 1774, local boyars put forward numerous claims in order to strengthen the distinctiveness of the Romanian principalities as political entities. As a consequence, the sultans confirmed the above-mentioned customary practices by imperial orders and laws (hüküm, ferman, hatt-i şerif, sened, and even kanun-name, etc.), that is, by documents that were used to regulate the internal affairs of the empire—thus making it clear that they considered Wallachia and Moldavia to be Ottoman tributary provinces.65 By contrast, the Wallachian and Moldavian nobility realized that the prolonged customary practices regularly observed in the relations

64 For details, see Panaite, The Ottoman Law of War and Peace, 313–334; idem, “Custom in the 16th–18th Centuries Ottoman–Romanian Relationships (Starting Points for a Histo- riographical Debate),” Revue des Études Sud-Est Européennes 31 (1993), 171–185. 65 In Ottoman documents, the miscellaneous rules that related to the status of a state’s self-government were generally called “regulation of the country” (nizam-i memleket). These rules concerned the institution of the voyvodalık; the collection of the poll-tax; the protection of Wallachian and Moldavian territories and subjects against neighboring Otto- man authorities; the legal status of Turkish merchants in Moldavia and Wallachia (Maria Matilda Alexandrescu-Dersca Bulgaru, “L’origine des khatt-i sherifs de privilèges des Principautés roumaines,” Nouvelles Études d’Histoire (Bucharest) 6, no. 1 (1980): 259–262). From several fermans one can deduce that the Ottoman subjects (especially merchants and soldiers) from the neighboring territories and fortresses (such as Hotin, Brăila, Vozia, Tighina, and Trnovo) were not allowed to enter Wallachia and Moldavia without imperial authorization. See Gemil, Relaţiile Ţărilor române cu Poarta otomană, doc. 190 (from 1691); Veliman, Relaţiile româno-otomane, doc. 41, 143, 148, 158 174, 184 (from 1720, 1760–1761, 1766, 1775, 1782). Ahmed Reşid, the kadi of Trnovo wrote in a report from 1691 that “the inter- diction of permitting somebody from the vicinity to enter without a [specific] order the countries of Moldavia and Wallachia was part of the regulation of the country” (“Eflâk ve Boğdân memleketlerine êtrafdan kimesnenin bila-ferman duhûluna ruhsat verilmesi egerçi nizam-ı memleketden olub”). This regulation was acknowledged by the Muslims of Trnovo by reading the sultan’s order and the grand vizier’s letter, which were then copied in the șeri‘at registers; see the report of 6 Receb 1102 / 5 April 1691 (Gemil, Relaţiile Ţărilor române cu Poarta otomană, doc. 190). See also Veliman, Relaţiile româno-otomane, doc. 143, 148 (“memleket-i mezbûrenin bu def ’a virilen nizamın istikrar ve istimrarı içün”), 158 (memleket-i Boğdân’in nizam-kadîmi). The decision, in the eighteenth century, to limit the rule of the voievods to a duration of three years was also considered a “rule and regulation” (ka‘ide ve nizam); see a document from 1749 in Veliman, Relaţiile româno-otomane, doc. 120. Finally, documents from 1776 and 1812 speak about the regulation (nizam) of collecting the tribute (cizye) from the principalities of Wallachia and Moldavia (Veliman, Relaţiile româno-otomane, doc. 176, 241). 34 viorel panaite between the Porte and its tributary principalities could provide a basis for recuperating a real internal autonomy. Through the codification of these customary practices, “treaties” (“capitulations”) were concluded between the Porte and Wallachia, as well as Moldavia: a fact that simultaneously implied the elimination of the “new practices” that had been established in an abusive manner during the eighteenth century.66 This also gave birth to the exaggerated myth of “old and long-term privileges” granted to certain Wallachian and Moldavian princes in the fourteenth to the sixteenth century on the occasion that they acknowledged their allegiance to the Porte. According to the eighteenth-century annalists, these acknowledgments of allegiance in Wallachia would have been conditioned by “agreements of obedience” (in Romanian, învoială or tocmeală) between sultans and princes. They were traditionally assigned to Mircea the Elder and Basarab Laiotă, who would have concluded them in 1391/1393 and 1460 (!?) respectively. The stipulations (legături) of these two “capitulations,” which defined a wide-ranging self-government status for Wallachia, were listed by the boyars in memorials addressed to Count Orloff in 1772.67 Concern- ing Moldavia, the tradition of old privileges was established primarily through the writings of Nicolae Costin and Dimitrie Cantemir, who registered the “instruments of the sultans, by which they ratified the conditions” with the sons of Stephen the Great (i.e., Bogdan the Blind and Petru Rareş) in 1511–1512 and 1529 as historical reality.68 This tradition was

66 They were characterized as “vile practices” (in Romanian, obiceiuri spurcati) by the Moldavian chronicler Ion Neculce, see his Opere: Letopiseţul Ţării Moldovei şi O samă de cuvinte [Works: The chronicle of Moldavia and a collection of tales], ed. Gabriel Ștrempel (Bucharest, 1982), 736. 67 Mihail Cantacuzino banul, Genealogia Cantacuzinilor [The genealogy of the Canta- cuzinos], ed. Nicolae Iorga (Bucharest, 1902), 67–70; Dionisie Fotino, Istoria generală a Daciei sau a Transilvaniei, Ţării Munteneşti şi a Moldovei [The general history of Dacia, that is Transylvania, Muntenia and Moldavia], trans. George Sion (Bucharest, 1859), 216–217; Istoria Ţerei Romănesci de fraţii Tunusli [The history of the Romanian lands by the Tunusli brothers], trans. George Sion (Bucharest, 1863), 66–70. 68 According to Dimitrie Cantemir, the main “articles” of the old Moldavian “treaties” are the following: “Among other numberless privileges granted them according to the times, the chief was that wherein it was expressly said, that Moldavia voluntarily and without compulsion offered her obedience to the Turkish Empire, and therefore it is the Sultan’s will that all her Churches, religious Rites, and Laws be untouched, and nothing more [be] required of the Prince, but that he send every year by faithful Boyars to the resplendent Port four thousand gold Crowns, forty bred Mares, twenty-four Falcons, in the name of Pishkiesh, a present of gift.” Also, Dimitrie Cantemir wrote that “The instruments of the Sultans, by which they ratified the conditions, were formerly kept among the public records of Moldavia, but in my time by the command or permission of John the legal and political status of wallachia and moldavia 35 then continued by Alexandru Beldiman, from whom we have a manuscript based on excerpts from “the sources of Neculaiu Costin” under the title The Compacts by which the Country was Submitted by Bogdan Voivode, Ruler of Moldavia, during the Reign of Bayezid the Second.69 In the period between 1774 and 1829, Wallachian and Moldavian nobles—enjoying the support of Russia—tried to present the “old privileges” of Wallachia and Moldavia, which had supposedly been established in writing by the so-called “capitulations,” as an “integral part of European public law.” Based on these apocryphal texts, the boyars tried to build a new “political identity” and to obtain international recognition. They were indeed considered by Ottoman and Russian delegates at the negotiations leading to the Peace Treaty of Adrianople (1829) as diplomatic sources for the political status of the Romanian principalities. The “capitulations” also received European recognition, as, besides the imperial orders (hatt-i şerifs) and Ottoman-Russian treaties, they were also invoked as the legal basis of the Ottoman sovereignty and the privileges enjoyed by the principalities.70 The tradition of “old privileges” fabricated by Wallachian and Molda- vian noblemen was also accepted by Western scholars in the eighteenth and nineteenth centuries.71 William Eton, for instance, in his book A Sur- vey of the Turkish Empire (printed in 1798), suggests that the legal status of Wallachia and Moldavia, as well as that of Egypt, might have been a result of peace agreements: “They may be considered connected with the Porte rather by treaty than as integral parts of the empire.”72

Sobiesky King of Poland, in his invasion of Moldavia in the year 1686, they were taken, and I know not whether very wisely, burnt publicly at Jassy, with this Declaration to the crowds of people, ‘Behold! his royal Majesty thus frees you from the Turkish yoke.’ ” (Cantemir, Othman Empire, 188–189n32). 69 Tractaturile prin care s-au închinat ţara, de către Bogdan voevod, domnu al Moldoviei, împărăţind Baizet al 2-lea (Biblioteca Academiei Române, ms. 566, f. 126v–136). This text was published by Mihail Kogălniceanu in Arhivă Românească 2 (1845): 347–364 and then in his Cronicele României seu Letopiseţele Moldaviei şi Valahiei [Chronicles of Romania, or Moldavia and Wallachia], vol. 3 (Bucharest, 1874), 450–459. 70 Dimitrie A. Sturdza and Cezar Colescu-Vartic, ed. Acte şi documente relative la istoria Renascerei României [Papers and documents concerning the history of Romania’s rebirth], vol. 1, 1391–1841 (Bucharest, 1900), 321. 71 Actes diplomatiques constant l’Autonomie Politique de la Roumanie, ed. J.-A. Vaillant (Paris, 1857); “Deux premiers traités de protection que la Valachie et la Moldavie ont conclu avec la Porte Ottomane,” Revue d’Orient 1 (1861), 89–90. 72 Eton, A Survey of the Turkish Empire, 287–288, 297. In the first part of the nineteenth century, the tradition of “old privileges” was present in French writings (where I have seen the term). In his “Report on the Romanian principalities” of 10 May 1834, Bois-le-Comte emphasizes that the legal status of Wallachia and Moldavia was based on the same “terms” 36 viorel panaite

The authenticity of the texts of the “old privileges” was first questioned at the beginning of the twentieth century by the Romanian historian Con- stantin Giurescu, who proved that those that were supposed to have been granted by Ottomans to Moldavian hospodars in 1511–1513 and 1529 are actually apocryphal writings from the eighteenth century.73 His conclusions are also valid for the so-called Wallachian “capitulations,” assigned to the prince Mircea the Elder in 1391/1393 and, later, to Laiotă Basarab (1473–1477, with interruptions) in 1460. Yet, Romanian historians have not given up the idea that “special treaties” were concluded between the Porte and the Romanian principalities; treaties that would have regulated the legal and political status of Wallachia and Moldavia on a long-term basis.74 The myth became much stronger in Romanian historiography during the Communist regime and these apocryphal writings are frequently quoted even today in nationalistic works and very recent textbooks for students as basic documents of the Ottoman–Romanian relations of the late fourteenth to the nineteenth century. In my view, it is necessary to state clearly that the Ottomans did not grant any “special treaty” (“capitulations”) between the fourteenth and the mid-sixteenth century that regulated the legal and political status of Wal- lachia and Moldavia on a long-term basis. The main reason being that this kind of “treaty” was contrary to Ottoman diplomatic practice.75 In fact, the term “capitulations” should not be used by historians and jurists at all

(conditions) since 1460 (?), respectively 1536 (?): “Les conditions qui furent accordées dans ce premier moment, aux Valaques et aux Moldaves, ont servis de base jusqu’à ce jour à leur constitution politique,” Hurmuzaki XVII, doc. DXI, 328–334. 73 Constantin Giurescu, Capitulaţiile Moldovei cu Poarta Otomană [The capitulations of Moldavia with the Ottoman Porte] (Bucharest, 1908). 74 Ştefan Ştefănescu, Ţara Românească de la Basarab I “Intemeietorul” până la Mihai Viteazul [Wallachia from Basarab I “the Founder” to Michael the Brave] (Bucharest, 1970), 101–140; Alexandrescu-Dersca Bulgaru, “L’origine” 251–263; Şerban Papacostea, “Tratatele Ţării Româneşti şi Moldovei cu Imperiul otoman în secolele XIV–XV: ficţiune politică şi realitate istorică” [The treaties of Wallachia and Moldavia with the Ottoman Empire in the fourteenth and fifteenth centuries: Political fiction and historical reality], in Stat, societate, naţiune: Interpretări istorice [State, society, nation: Historical interpretations] (Cluj-Napoca, 1982), 93–106; Mihai Maxim, “Din istoria relaţiilor româno-otomane—‘Capitulaţiile’ ” [From the history of the Romanian–Ottoman contacts: “Capitulations”], Anale de Istorie 28, no. 6 (1982), 34–68; idem, “Din nou pe urmele vechilor tratate ale Moldovei şi Ţării Româneşti cu Poarta Otomană” [Again on the track of the old treaties of Moldavia and Wallachia with the Ottoman Porte], Revistă de Istorie 40, no. 2 (1987), 157–172; idem, Ţările Române şi Inalta Poartă, 25–111, 197–261. 75 For details, see Panaite, The Ottoman Law of War and Peace, 232–282; idem, “Peace Agreements in Ottoman Juridical and Diplomatic View (15th–17th century),” in Pax Oto- mana: Studies in Memoriam of Prof. Dr. Nejat Göyünç, ed. Kemal Çiçek (Ankara and Haar- lem, 2001), 277–308. the legal and political status of wallachia and moldavia 37 in connection to the early relations between the Ottoman state and the principalities. Medieval chancelleries issued many kinds of documents for internal or external use (capitularium, capitulatio), which had a common feature, i.e., they were divided into chapters. The term capitulations originated in the words caput or capitulum (pl. capitula), Latin for chapter, paragraph.76 It was only due to their division into articles, and not for other reasons, that peace agreements were also named capitulations in Western Europe.77 Considering the ‘ahdnames granted by sultans as treaties, Europeans applied an autochthonous concept, i.e., capitulations, to them. This is also documented by the sixteenth- and seventeenth-century Western translations of ‘ahdnames, in which the two words were considered similar. The Latin capitulationes, the French capitulations, the Italian capitulatione, the English capitulations, and the German Capitulation were the Western terms used in translations of ‘ahdnames granted to France,78

76 Naturally, the Italian capitolo (pl. capitoli), capitolare, capitolato and the French chapitre, chapitrer have the same origin. 77 The other name in use was pactum (in Italian patto, in French pacte). For details, see François-Alphonse Belin, Des Capitulations et des traités de la France en Orient (Paris, 1870), 7–17 (Chapitre Ier: Sens et signification du terme capitulations; son correspondant dans les chancelleries orientales). For example, the Byzantine–Genoese treaty of 1304 and the one concluded between Venice and Cyprus in 1328 were both called capitula. These terms can also be found in Latin, French, and Italian versions of eleventh- to fourteenth-century peace and commerce treaties between Christian states or cities and North African Muslim princes. In 1186, “capitoli della pace” were negotiated and established between Pisa and Tunis. All these treaties were divided into “chapters”: primo capitolo, secondo capitolo, and so on (Marie Louis de Mas Latrie, Traités de paix et de commerce et documents divers concernant les relations des Chrétiens avec les Arabes de l’Afrique septentrionale au Moyen Âge (Paris, 1866), vol. 1, 28–30, 66–69, 283–284 etc.). 78 In France, they were included in seventeenth- and eighteenth-century miscellanies, such as the Recueil de pièces officiells folio 2 recto, comprenant les capitulations (‘ahdname- i şerif ), see Edgar Blochet, Catalogue des Manuscrits Turcs de la Bibliothèque Nationale, vol. 1 (Paris, 1932), 53. Also, “les hautes et heureuses Capitulations,” “des anciennes Capitu- lations” in Henry III’s letter to the Grand Vizier Sinan Paşa of 7 July 1582. (Hurmuzaki, Supl. I/1, doc. CXXIV). Concerning England, by comparing the Ottoman text of commercial privileges granted to the English merchants in 1580 with the Italian translation of 5 April 1583, we can see that the formula ‘ahdname-i hümayun was translated imperiale capitulazione, and the whole document was entitled Capitoli dati alla Regina d’Inghiltera. See Susan Skilliter, William Harborne and the Trade with Turkey 1578–1582: A Documentary Study of the First Anglo–Ottoman Relations (London, 1977), 86–89, 232–239. For Poland, see the following examples: for the 1568 ‘ahdname, “Capitulatione tra sultan Selim Imperator de Turchi et Sigismondo Augusto Re di Polonia,” Ioan Slavici, ed. Documente privitoare la istoria românilor culese de Eudoxiu de Hurmuzaki [Documents for the history of the Romanians collected by Eudoxiu de Hurmuzaki], vol. 8, 1376–1650 (Bucharest, 1894) [henceforth Hurmuzaki VIII], doc. CXCI); “Capitulations du Turc avec les Polonnois,” “les Capitulations de Pologne” French letters of 1617 and 1623 (Hurmuzaki, Supl. I, vol. 1, doc. CCLXVI, CCCVI; “nostra eccelsa capitulatione,” in an Italian translation of the 1564 ‘ahdname (Hurmuzaki, VIII, doc. CXXXIII). For the Habsburg Empire: “die fridens capitulation,” on the 1606 treaty, in a 38 viorel panaite

England, Poland, the Habsburg Empire, or Russia. When translating the Ottoman texts, the Europeans organized their contents by articles or chapters.79 As any document divided into chapters could be called capitulation in Western chancelleries, the same label was applied to the Otto- man ‘ahdnames as well.80 Other meanings of the concept capitulation(s) were also adopted into Ottoman Turkish (Kapitülasyonlar) from the early eighteenth century; it was also used to refer to the system of privileges granted to foreign merchants in the Ottoman Empire, or for a special treaty that regulated the position of foreigners inside the empire.81 The “peace agreements” concluded between the Ottoman Empire and the Christian rulers belonged to the general category of “contract” (‘akd) or “covenant” (‘ahd). According to Muslim legal texts (kitab as-siyar, fetvas), there were two kinds of “covenant” (‘ahd): a) “contract of tributary protection” (‘ahd üz-zimmet) and “truce” (muvada‘a). Until the nineteenth century, the term used by the Ottoman chancellery for peace agreements by which the Porte regulated its peace relations, alliances, international trade or

German letter of 16 March 1614 (Ioan Slavici, ed. Documente privitoare la istoria românilor culese de Eudoxiu de Hurmuzaki [Documents for the history of the Romanians collected by Eudoxiu de Hurmuzaki], vol. 4, part 1, 1600–1649 (Bucharest, 1882), doc. CCCCLXIV). For Russia, see “capitulationes” in the Latin translation of a ferman of Ahmed III, in the Ottoman–Russian treaty of 1711, Ioan Slavici, ed. Documente privitoare la istoria românilor culese de Eudoxiu de Hurmuzaki [Documents for the history of the Romanians collected by Eudoxiu de Hurmuzaki], vol. 6, 1700–1750 (Bucharest, 1878), doc. LVII). 79 The Ottoman texts of imperial charters (‘ahdnames) include a succession of peace and trade “conditions” (sing. şart; pl. şartlar, şurut, şerayt) or “problems” (hususlar). The beginnings of the articles were marked by the conjunction “and” (ve) until the end of the seventeenth century, when it started to be signaled with numbers, as in the Western chancelleries. 80 Few authors proposed other options. According to Ignace de Testa, the notion capitulations was used in European diplomacy for the designation of proper treaties (“bilateral documents”), and could therefore not be applied to ‘ahdnames (seen as “unilateral documents”). In his opinion, the proposed translation for ‘ahdname should be “lettres-patentes” (see his Recueil des traités de la Porte Ottomane avec les puissances étrangères (Paris, 1864), vol. 1, 6). Susan Skilliter called the document of 1580, by which Murad III had granted commercial privileges to English merchants, a “diploma (berât) incorporating the privileges” or “Imperial treaty letter” (Skilliter, William Harborne, 86–89, 232–239). 81 By “capitulation” one can also understand an agreement between two enemies that prescribed the surrender of one to the other. In Italian, capitolare means “to submit” and capitulazione “surrender.” With this in mind, certain scholars defined the Capitulations as documents which confirmed that the sultans were forced by Western powers to give up certain parts of their sovereignty. For the Turkish usage, see Nasim Sousa, The Capitula- tory Regime of Turkey: Its History, Origin and Nature (Baltimore, 1933), 2–3; Halil İnalcık, “Imtiyâzât,” in EI-2, vol 3, 1217; Gérard Pélissié du Rausas, Le régime des capitulations dans l’Empire Ottoman (Paris, 1910), vol. 2, 1, 28; Reşat Ekrem, Osmanlı Muahedeleri ve Kapitül- asyonlar (1300–1920) ve Lozan Muahedesi (24 Temmuz 1923) (Istanbul, 1934), 402. the legal and political status of wallachia and moldavia 39 the condition of foreigners was imperial charter (‘ahdname-i hümayun) (this is what today’s historians and jurists label as peace treaties, treaties of alliance, treaties of friendship, treaties of commerce, treaties of vassalage, and so on). The term ‘ahdname is composed of two words, the Arabic ‘ahd (pl. ‘uhud), which means oath, compact, or covenant; and the Persian nâme, i.e., charter.82 This document was used to confirm in writing the granting of “protection” (aman), and the existence of a “contractual pact” (‘ahd, ‘akd, ‘akd-i ‘ahd, sulh) between sultans and an individual, a community, or a state.83 Even though ‘ahdnames had the form of an unilateral document, their contents implied “contract and alliance” (‘akd ü-ittifak) or “pact and agreement” (‘ahd ü-misak).84 Apart from the usual ‘ahdname,85 the terms mu‛ahede (mu‛ahedename), sulh (sulhname),86 and musalaha (musalahaname)87 began to appear in eighteenth-century Ottoman manuscripts—alone or in combinations (sulh mu‛ahedesi)—to refer to treaties of peace concluded with European powers, such as the Habsburg Empire and Russia; but also to refer to entire manuscript compilations of these peace agreements.88 The Moldo-Wallachian sources, when they referred

82 Joseph Schacht, “‘Ahd,” in EI2, vol. 1, 263. Dimitrie Cantemir defined the term ‘ahdname in the following words: “an Arabian Word compounded of Ahd, a Covenant or Condi- tion, and Namé, a Letter. By this Name the Letters are signified which foreign Ambassadors, after a Peace obtained, procure from the Sultan, to their respective Princes, containing the Terms of the Peace (which the Arabians call Mevad) and ratified with the Dura, or Char- acter of the Imperial Name” (Cantemir, Othman History, 88n35). 83 İnalcık, “Imtiyâzât,” 1208. There were ‘ahdnames issued to appoint dignitaries or successors (Schacht, “‘Ahd,” 263). 84 See the examples of the 1604 ‘ahdname to France in Ahmed Feridun Bey, Mecmu‘a-ı Münşe’at es-Selatin (Istanbul, 1264–1265/1848–1849), vol. 2, 404; or the 1612 ‘ahdname to Holland in Alexander H. de Groot, The Ottoman Empire and the Dutch Republic: A History of the Earliest Diplomatic Relations 1610–1630 (Leiden and Istanbul, 1978), 246. 85 The manuscript ‘Ahd-nâme sûretleri is a compilation of the peace agreements between 1084 and 1146/1673 and 1733 (TKSMK, H. 1636). 86 From the fifteenth century, the word sulh continually and generally designated a peace concluded with any foreign state (TKSMK, R. 1325: risâle on the conclusion of peace (sulh) with the Habsburg Empire and Russia in 1739; TKSMK, E.H. 1438: Rusya ile sulh). 87 TKSMK, E.H. 1438, 165b–214a (on the treaties with Russia of 1791–1792). 88 The manucript Sulh-name-i ‘Amuca-zade Hüseyin Paşa relates the negotiations leading to the peace of Karlowitz (TKSMK, R. 1311). The manuscripts Suver-i mekatib-i musalaha Nemçe and Esnayi musalahada tevarüd eden mekatib also includes the treaty of Passarowitz of 1131/1719 (TKSMK, R. 1946, R. 1953). The manuscript Mu‘ahedat-ı hümayun 975–1200 Osmanlı devleti ile Nemçe ve sair devletler arasında ‘akd olunan bazimuahedeler is a collection of the treaties with the Habsburg Empire, Russia, and other states concluded between the years 975 and 1200 / 1567 and 1796 (Belediye Kütüphanesi, Muallim Cevdet, K. 4; partially in Arhivele Naționale ale României, Bucharest (henceforth ANR), mf. Turcia, r. 53, fr. 894–1004). The manuscript Risâle-i mu‘âhedât including the 1798 treaty with 40 viorel panaite to Ottoman peace agreements, alluded to the practice of taking oaths,89 or directly borrowed the Ottoman concept of ‘ahdname (in Romanian transcription, ahtinâme or actenamede).90 The ‘ahdnames that confirmed the conclusion of a peace with Euro- pean rulers were valid only for a limited period of time: either for the number of years indicated, or for the length of the two ruler’s reigns. The former practice followed Islamic law, which interdicted the conclusion of perpetual agreements with non-Muslim states, and recommended that a fixed and specified number of years should be registered in the text of the agreement.91 The latter followed a medieval custom, in which there did not exist an idea that official documents, including treaties, would have perpetual validity after the death of the ruler that issued them. Chroni- cles concerning the Balkan tributary princes of the fifteenth century offer evidence that because of their connection to the persons who concluded them, Ottoman sultans, as well as Christian princes, were aware of the provisional character of covenants.92 From the late fourteenth to the mid-sixteenth century, Ottomans concluded temporary peace agreements with the Wallachian and Moldavian princes, either in verbal or in written form. In the narrative sources, these were most frequently called “oaths” and “pledges” (in Ottoman ‘ahd, in Greek horkos, in Romanian jurământ), because the oath had long been

England and the secret treaty with Russia of 27 Receb 1213/1 January 1799 (TKSMK, fd. Bağdad 237; ANR, mf. Turcia, r. 51, fr. 647–663). 89 The ‘ahdname granted to John Zápolya in 1528 by Süleyman the Magnificent was called “carte cu giuramânt” in Romanian chronicles (Miron Costin, Opere, ed. Petre P. Panaitescu (Bucharest, 1958), 290), whereas the treaty between John VI Kantakuzenos and Orhan or Murad I was referred to as “juramânt mare” (Mihail Moxa, Cronica universală [Universal chronicle], ed. Gheorghe Mihăilă (Bucharest, 1989), 187). 90 The Romanian term for the documents that granted a lifelong rule to Constantin Brâncoveanu, voievod of Wallachia between 1688 and 1714, were “atişirif ” and “hatişerifuri” (from the Turkish hatt-ı şerif ); see Radu logofătul Greceanu, Istoria domniei lui Constantin Basarab Brâncoveanu Voievod (1688–1714) [The history of the rule of Voievod Constantin Basarab Brâncoveanu], ed. Aurora Ilieş (Bucharest, 1970), 126, 144–145. 91 Pélissié du Rausas, Capitulations, vol. 1, 23–24; Herbert J. Liebesny, “The Development of Western Judicial Privileges,” in Law in the Middle East, ed. Majid Khadduri and Herbert J. Liebesny (Washington, DC, 1955), 309–333. 92 According to Şükrüllah, the compact between Mehmed I Çelebi and the Wallachian prince in 1417 was to be valid only “during the sultan’s life” (Şükrüllah, Tevârih, in Cronici turceşti, vol. 1, 32). A similar limitation, that is, the validity of the contract “until the end of the [sultan’s] life,” was indicated by Ducas referring to fifteenth-century Ottoman– Byzantine relations, and specifically to the peace agreements of 1416 and 1451; see Ducas, Istoria turco-bizantină (1341–1462) [Turkish–Byzantine history 1341–1462] ed. Vasile Grecu (Bucharest, 1958), 158, 288. See also Laonic Chalcocondil, Expuneri istorice [Historical expo- sitions], ed. Vasile Grecu (Bucharest, 1958), 219. the legal and political status of wallachia and moldavia 41 the most important warranty of fidelity practiced in diplomatic relations. Let us note that very few of the “oaths” concluded by the Ottomans with Southeastern European rulers in the fifteenth century have been preserved in a chancellery form (not a single one with Byzantium or with the Wallachian princes, and only one with Hospodar Stephen the Great of Moldavia; in contrast, there are, fortunately, more imperial charters preserved from those granted to Ragusa).93 In the ‘ahdname granted to Stephen the Great in 1480–1481, Mehmed II Fatih promised protection only in his own name, without placing obligations on any of his successors to respect the covenant: “neither his wealth nor his country will be attacked by me.”94 Moreover, the sultan limited the validity of the documents with the following words: “As long as from his part one shall see right actions, he will enjoy the same imperial benevolence (şahanemle) and will be protected by the same imperial magnanim- ity (husrevanemle) on my part.”95 Thus, the sultan related the validity of the pact to the behavior of the other part, invoking as an argument the Qur’anic verse 9:17: “As long as they stand true to you, stand true to them: for Allah does love the righteous.” Sultans obliged themselves by oath not to violate the peace stipulations as long as (madam ki) the Christian princes did not break them first. The result was, of course, that the validity had no practical time limitation, and depended only on the Ottoman authorities’ will. According to the şeri‘at, the aman and ‘ahd lost their validity when the term expired. In the practice of the Ottoman chancellery, the ‘ahdnames lost validity when the prescribed number of years ran out or the reign of the ruler who had granted the document ended. If there was an agreement of interests by both parties, the ‘ahdnames could be renewed. The renewal of peace agreements when a new sovereign came to the throne was a medieval customary practice, also used in Southeastern Europe, and included the relations between Ottomans and tributary princes.96

93 Panaite, The Ottoman Law of War and Peace, 264–282; idem, “Câteva observaţii privind reglementările de pace româno-otomane în secolul XV” [Some remarks on the Romanian–Ottoman peace settlements of the fifteenth century], in Românii în istoria universală, vol. 3/1 (Iaşi, 1988), 469–492; idem, Pace, război şi comerţ în Islam, 314–328. 94 Mehmet, Documente turceşti, vol. 1, doc. 5. Actually, the standard formula of the ‘ahdnames would have been “me, too” (ben dahi, benden dahi); see the ‘ahdname of 1607 granted to Poland in Feridun, Münşe’at, vol. 2, 417. 95 Mehmet, Documente turceşti, vol. 1, doc. 5. 96 This was the practice in the relations between Ragusa and Hungary, Serbia, Bosnia; see Biegman, The Turco–Ragusan Relationship, 16n38. It was also used in the relations of Moldavia and Wallachia with Poland and Hungary; see Tocilescu, 534 documente; Ioan 42 viorel panaite

In addition, the renewal of peace agreements also took place after the occurrence of military conflicts when the Porte had not destroyed the enemy state by debellatio. The relations with the principalities of Walla- chia and Moldavia from the late fourteenth century to the mid-sixteenth century offer us a picture in which war and peace were everyday topics.97

In conclusion, I prefer to follow the manner of Ottoman fetvas and offer short answers to the questions asked at the beginning of these pages: To the question, did the Ottomans conquer Wallachia and Moldavia? I reply, yes, under the rule of Süleyman the Magnificent. To the question, were Wallachia and Moldavia within the Ottoman Empire? I reply, yes, from the rule of Süleyman. And to the question, were there long-term “treaties” between the Ottoman Empire and the two Danubian principalities? I reply, no. With the adoption of these answers, we are led to a logical and hopefully realistic description of Wallachia and Moldavia’s status in relation to the Porte.

Bogdan, Documente privitoare la relaţiile Ţării Romaneşti cu Braşovul în secolele XV si XVI [Documents concerning the relations of Wallachia with Braşov in the fifteenth and sixteenth centuries] (Bucharest, 1905), vol. 1. 97 Ducas, Istoria, 296; Ştefanescu, Ţara Românească, 37–100. The same picture was characteristic for the relations between the Porte and Byzantium; see Georges Ostrogorsky, Histoire de l’État byzantine, trans. Jean Gouillard (Paris, 1977), 555–595. Sovereignty and Subordination in Crimean-Ottoman Relations (Sixteenth–Eighteenth Centuries)

Natalia Królikowska

During the last two centuries, Crimean-Ottoman relations attracted mod- erate attention from European scholars; until World War II, the khanate was treated as a vassal state of the Ottoman Empire.1 Many scholars believe that Ottoman suzerainty over the Tatars was stipulated in a treaty concluded between Khan Mengli I Giray and Sultan Mehmed II in 1478;2 however, the existence of this treaty was questioned by Halil İnalcık in the 1940s3 and effectively disproved by later scholars. One of them, Alan Fisher, points out that the degree of the khanate’s dependency on the

1 The titles of two well-known nineteenth-century studies epitomize this attitude: see Joseph Hammer-Purgstall, Geschichte der Chane der Krim unter osmanischer Herrschaft vom 15. Jahrhundert bis zum Ende 18. Jahrhundert (Vienna, 1856); Vasilii Smirnov, Kryms- koe khanstvo pod verkhovenstvem Otomanskoi Porty do nachala XVIII veka [The Crimean Khanate under the suzerainty of the Ottoman Porte] (St. Petersburg, 1887) and the second part entitled “Krymskoe khanstvo pod verkhovenstvom Otomaskoi Porty v XVIII v. do prisoedinieniia k Rossii” [The Crimean Khanate under the suzerainty of the Ottoman Porte in the eighteenth century to its incorporation into Russia], Zapiski Odesskogo Obsh- chestva Istorii i Drevnostii 15 (1889): 152–403. The claim is also well illustrated by the passages quoted below: “Mengli Giray (. . .) was sent back to the Crimea as khan; nevertheless, from this time onward, the Turks started to control the Perekop Tatars and also positioned their troops on the peninsula. In return, the khans gained assurances of eternal alliance and brotherhood [with the Ottomans].” (“Menglikerey (. . .) nazad do Tauryki odesłany z tytułem kana; wszelako od tej daty Turcy powzięli mieć najwyższą zwierzchność nad Tatarami Perekopskiemi, stawiać na Tauryce swoje garnizony, a kanom tytuł braterstwa i ligi, wiecznej zostawuiąc.”) See Adam Naruszewicz, Tauryka czyli wiadomości starożytne o stanie i mieszkańcach Krymu do naszych czasów [Tauryka, that is, archaic information about the state and population of the Crimea until our time] (Warsaw, 1805), 90; “Still we cannot but feel a conviction that where the elements of civilized life, of great ability, of rule, of farsighted policy, of taste for the acts of peace, are found in monarchs such as those here instanced (and others might be named), there must be at least the possibility of a civilized community in their subjects. Yet, at the same time, we cannot but ask, what could a country do, whose rulers were subjects to such a tyrannous caprice as these khans suffered at the hands of the degraded Sultans?”; see Anthony Grant, An Historical Sketch of the Crimea (London, 1855), 101. 2 Vasilii Smirnov, Krymskoe khanstvo, 294. 3 Halil İnalcık, “Yeni Vesikalara Göre Kırım Hanlığının Osmanlı Tabiliğine Girmesi ve Ahidnamesi Meselesi” [The subordination of the Crimean Khanate to the Ottomans and 44 natalia królikowska

Ottomans was in continual flux between the late fifteenth century and the end of the eighteenth century. He questions the view that the Tatars were servile vassals of the Ottoman Empire during this period. Fisher also argues that they conducted a separate foreign policy, which sometimes differed from that of the Ottomans.4 The study of Carl Max Kortepeter provides a fascinating example of such an independent Crimean ruler, Gazi II Giray (1588–1596; 1597–1608). The author demonstrates that the different aims of the Ottomans and the Crimean clans made it possible for the khan to seize enough power to control his vassals from one side, and to conduct more independent policies toward the Ottomans from another.5 The studies of Akmes Nimet Kurat on the early eighteenth century demonstrate that in this period the khans could not act independently of Ottoman influence. To illustrate this, the author draws a moving picture of Khan Devlet II Giray (1699–1702, 1708–1713), yet another outstanding Crimean ruler, who failed in his attempts to conduct an anti-Russian policy independently of the Ottomans. In 1711, during the war with Russia in Moldavia, he was unable to convince the Ottomans to pursue their advantage over Tsar Peter I: going against the khan’s advice, Grand Vizier Baltacı Mehmed Pasha concluded the Prut armistice on the terms proposed by the tsar.6 As Alan Fisher suggests, “at some undetermined time between 1600 and 1750, the entire configuration of Ottoman-Crimean Tatar political relations changed completely.”7 The shifts in Crimean-Ottoman relations can also be seen in the Crimean chronicles. Their authors conceptualized mutual relations, including the status of the khanate versus the Ottoman Empire, in diverse ways. The first chronicle written by Remmal Hoca in the mid-sixteenth century reflects these ambiguities, which are clearly present in his description of

the question of the treaty in the light of previously unknown documents], Belleten 8 (1944): 185–229. 4 Alan Fisher elaborates on this subject in numerous articles, collected in two volumes; see idem, Between Russians, Ottomans and Turks: Crimea and the Crimean Tatars (Istanbul, 1998); idem, A Precarious Balance: Conflict, Trade and Diplomacy on the Russian-Ottoman Frontier (Istanbul, 1999). 5 Carl Max Kortepeter, Ottoman Imperialism during the Reformation: Europe and the Caucasus (London, 1972). 6 Akmes Nimet Kurat, İsveç Kıralı Karl’ın Türkiye’de Kalışı ve Bu Sırada Osmanlı Imperatorluğu [The stay of the Swedish King Charles in Turkey and the Ottoman Empire in this period] (Istanbul, 1943), 399–458; idem, Prut Seferi ve Barışı [The Prut campaign and the peace] (Ankara, 1953), vol. 2, 501–502. 7 Alan Fisher, Between Russians, Ottomans and Turks, 67. sovereignty and subordination in crimean-ottoman relations 45 the circumstances in which Sahib Giray (1532–1551) ascended to and then lost the Crimean throne. In his description of Sahib Giray’s ascension to the throne, the chronicler depicts the audience of the Crimean noblemen before Sultan Süleyman the Magnificent in 1532. According to Remmal Hoca, as soon as the Tatars requested the change of their ruler, Sa‘adet I Giray, the sultan answered: “It is not possible to depose the Crimean khans, because for a long time they were the padishahs with the right to sovereignty, hutbe [the sermon following the Friday prayer in which the Islamic ruler’s name is mentioned] and [the right to mint] coins.”8 Nevertheless, when Saadet I Giray, the unwanted khan, left his state, the sultan nominated Sahib Giray in his place. The new ruler was sent to the Crimea as the Ottoman appoin- tee with the insignia of power given to him by Süleyman the Magnificent. Sahib Giray obtained from the sultan a standard and a horsetail, as well as janissaries and money. Remmal Hoca suggests that Sahib Giray’s loss of power and life was closely connected to the deterioration of his relations with the Sublime Porte. The chronicler states that during his reign, Sahib Giray took part in numerous Ottoman military campaigns; but in 1546, when he fabricated an excuse not to send auxiliary troops to support the sultan’s campaign in Iran, the pashas at the Sublime Porte alerted the sultan. In the words of the chronicle: Our Lord, thanks to you the khan has ascended to the throne, you have augmented his treasury and strengthened his army. [But now] he shows disrespect to [You,] Our Lord. The proof is that he now dares to oppose your orders and make excuses for not sending auxiliary troops. He thinks he will gain possession of all the territories [in the Crimea] the moment you pass away. (. . .) If he unites his forces with the Nogays, no one will be a match for him and [be able to] resist. Now is the time to take precautions.9 Sahib Giray lost the support of the Ottomans and a few years later was killed by his Nogay vassals; whereupon the sultan appointed another Crimean khan in his place. The story raises important issues about Tatar-Ottoman relations and provokes various questions concerning the degree of sovereignty of the Crimean Khanate. Was it an Ottoman vassal state? What did it mean to

8 “Ḳırım Ḫānları ma’zul olmaz eba’an ced ṣaḥibi ṣaltanat ve ṣaḥibi ḫutbe ve ṣaḥibi sikke pādişāhlardır.” See Remmal Hoca, Tārīh-i Ṣāḥib Giray Ḫān [The history of Sahib Giray Han], ed. Özalp Gökbilgin (Ankara, 1973), 20. 9 Remmal Hoca, Tārīh-i Ṣāḥib Giray Ḫān, 114–115. 46 natalia królikowska be a vassal or a sovereign in the Muslim world? What were the attributes of sovereignty and subordination in Tatar-Ottoman relations? To limit this impossibly broad field, this study, based on information drawn from chronicles of the Crimea, and Crimean court registers, as well as European and Ottoman accounts,10 addresses the following questions:

– What were the limits of Ottoman control over the khanate? Did the claim to the Genghisid heritage raised by the khanate affect Tatar- Ottoman relations?11 – Could the Tatars force the Sublime Porte to change its policy toward the khanate or toward other polities? – Did Tatar-Ottoman relations change during the early modern period? If so, what were the turning points? – Were there symbols of Ottoman suzerainty over the khanate?

For the purposes of this study, we must also define what it meant to be a vassal and a sovereign state in the Muslim world. As Colin Imber states,

10 Crimean judicial records, 121 volumes altogether, are preserved in the Russian National Library in St. Petersburg (Otdel Rukopisev Rossiiskoi Natsionalnoi Biblioteki, St. Petersburg [hereafter, ORRNB]), Fond 917). Before 1799, the collection was handed over to the Russian authorities by Mehmed Ağa, the last Crimean kadiasker; it was kept in the archives of the governor of Simferopol till 1905, when it was sent to the public library in St. Petersburg; see Olga Vasileva, “Krymsko-Tatarskie Rukopisnye Materialy v otdele ruko- pisii” [Crimean Tatar manuscript materials in the department of manuscripts], Rossiiskaia National’naia Biblioteka: Vostochnyi Sbornik 5 (1993): 37–45. The collection contains judicial records of the kadiaskers and kadis of the main Crimean cities (Bahçesaray, Kara Su, Gözleve), along with a few registers whose territorial affiliation remains unidentified. Two registers of real estate left by Christians, who emigrated from the Crimean peninsula in 1778, are also included in this collection. 11 The Girays’ descent from Genghis Khan was recognized by the Ottomans and gave the khans a unique prestige in the Ottoman world. Although the sultans dissociated themselves from the Mongol tradition (see Leslie Peirce, The Imperial Harem: Women and Sovereignty in the Ottoman Empire (New York and Oxford, 1993), 159), they competed with the khans over the political inheritance of the Golden Horde. The history of the unsuccessful Astrakhan campaign in 1569 clearly demonstrates the Tatar-Ottoman competition for supremacy over the steppe. After the Russian conquest of Kazan and Astrakhan, the Crimean khan made an attempt to regain these territories without Ottoman support. He tried to convince the tsar to return this area to him, “because from the old days Astra- khan and Kazan were part of the Muslim world and the yurt of the khans of our dynasty”; quoted from Mikhail Khodarkovsky, Russia’s Steppe Frontier: The Making of Colonial Empire, 1500–1800 (Bloomington, IN, 2004), 116. Only after the tsar denied this demand, did the khan give reluctant support to the Ottoman campaign heading to Astrakhan in 1569. Nonetheless, the Tatar troops abandoned the Ottomans at their earliest convenience, near the river Volga. Apparently, the Ottoman presence in the steppe posed such a serious threat to the khan that he preferred to leave Astrakhan in Russian hands. sovereignty and subordination in crimean-ottoman relations 47 the Hanafi theory of rulership offered a simple and realistic definition of a legal sovereign “as a person who seizes and effectively exercises power.”12 Although this was the real basis of Ottoman rule, it was not something which the sultans could admit. Instead, they made elaborate claims to rightful sovereignty, culminating in the sixteenth-century idea of universal rule. Ottoman sultans, according to Colin Imber, adopted the Hanafi jurists’ idea that the following four acts express exclusive royal authority:

1. performing the Friday prayer in the sultan’s name; 2. collecting the poll-tax and tribute from non-Muslims; 3. obtaining one-fifth of all plunder taken in a holy war; and 4. executing the fixed penalties.13

These areas played a symbolic role in the exercise of the sultan’s power. In the following few pages I address the question, who exercised power in these four areas in the Crimean Khanate: the khan or the sultan?

Friday Prayer

Remmal Hoca and Abdullah, son of Rizvan Pasha, in reference to Sahib Giray and Mengli I Giray respectively, state that Friday prayers were performed in the khan’s name.14 Evliya Çelebi, however, observed during his travel through the Crimea in the 1660s that the khan’s name was mentioned after that of the sultan.15 Çelebi states that the sultan’s name appeared in the Crimean hutbe as early as the reign of Selim I (1512–1520).16 Yet another version is found in the chronicle of Seyyid Muhammed Riza, who attributes the change of customs to İslam II Giray. The khan, according to the chronicler, introduced the sultan’s name into the Friday prayers in 1584; by doing so, he wanted to compensate the sultan for the damages

12 Colin Imber, Ebu’s-suud: The Islamic Legal Tradition (Edinburgh, 1997), 73. 13 Ibid., 79–94. 14 Remmal Hoca, Tārīh-i Ṣāḥib Giray Ḫān, 20; Abdullah, son of Rizvan Paşa, La chronique des steppes kipchak. Tevārīh-i Dešt-i Qipčak du XVIIe siècle, ed. Ananiasz Zajączkowski (Warsaw, 1966), 34. 15 Evliya Çelebi, Księga podróży Ewliji Czelebiego [The book of travels by Evliya Çelebi], ed. Zygmunt Abrahamowicz, Aleksander Dubiński, and Stanisława Płaskowicka- Rymkiewicz (Warsaw, 1969), 174, 283. 16 Ilya Zaitsev, Krymskaia istoriograficheskaia traditsiia XV–XIX vekov [Crimean historiographic tradition, fifteenth-nineteenth centuries] (Moscow, 2009), 150. 48 natalia królikowska caused by previous Tatar raids in Moldavia.17 The motive suggested by Riza calls for further scrutiny, as the Tatars made no savage raids on Moldavia in the 1580s: it seems, therefore, that İslam II Giray did not have to beg for the sultan’s forgiveness. It is also noteworthy that Abdullah, son of Rizvan Pasha, who lived in the Ottoman Crimea only a half century later did not mention this fact in his description of the reign of İslam II Giray.18 Yet another Crimean chronicler, Abdulgaffar el-Kırımi, also did not include any passages about eventual changes in the hutbe in his description of the reign of İslam II Giray.19 From the silence of Abdullah, son of Rizvan Pasha and Abdulgaffar el-Kırımi we can at least conclude that the version found in Seyyid Muhammed Riza’s chronicle was not universally acknowledged in the khanate.20 In light of the aforementioned examples it becomes clear that the circumstances in which the sultan’s name was introduced into the Friday prayer in the Crimea were seen differently in a variety of early modern narrative accounts, and it is hard to reconcile the divergent descriptions on the subject. Therefore, the exact date and the reasons for changing the Crimean hutbe remain unclear and will require further research.

Collecting the Poll-tax and Tribute

Tax registers from the Ottoman Crimea dating back as early as the 1520s provide evidence for the fiscal autonomy of the Crimean Khanate.21 The orders of the Crimean rulers recorded in the judicial records demonstrate

17 Seyyid Muhammed Riza, Es-sebü’s-seyyar ili sem’ planet soderzhavshchii istoriu krymskikh khanow ot Mengli Girej Khana pierwogo do Mengli Girej Khana vtorogo t. e. s 871/1466 po 1150/1737 [The seven planets or the history of the Crimean khans from Khan Mengli I Giray to Khan Mengli II Giray], ed. Mirza Kazembeg (Kazan, 1832), 107. This story is repeated by Halim Giray, whose descriptions of events prior to the 1740s mirror the chronicle of Seyyid Muhammed Riza, see Halim Giray, Gülbünü Hanan yahud Kırım Tarihi/Rozo- vyi kust khanov ili istoria Kryma [The rosebush of the khans or the history of Crimea], ed. Ablakim Ilmi (Simferopol, 2004), 193. 18 Abdullah, son of Rizvan Paşa, La chronique des steppes kipchak, 38. 19 Abdulgaffar el-Kırımi, Ümdetü’t-tavarih [The essence of history], ed. N. Asim Bey (Istanbul, 1343/1927), 114. 20 Natalia Królikowska, “Law and Division of Power in the Crimean Khanate: A Study on the Reign of Murad Giray (1678–1683),” (PhD diss., University of Warsaw, 2010), 64–73. 21 About the exact dating of the oldest tapu tahrir defter from Ottoman Crimea, see Yücel Öztürk, Osmanlı Hakimiyetinde Kefe (1475–1600) [Caffa under the Ottoman Rule] (Ankara, 2000), xxi. sovereignty and subordination in crimean-ottoman relations 49 that their officials collected the poll-tax payable by non-Muslim subjects.22 Sources of this kind also offer data on other taxes paid to the khan by his Muslim and non-Muslim subjects.23 There is no evidence of any taxes from the khanate being paid into the Ottoman treasury. The tribute paid by Muscovy and the Polish-Lithuanian Commonwealth to the Crimea constituted an important part of khan’s income. Different terms with various meanings were applied to these periodic payments, but it is unfortunately beyond the scope of this study to discuss in detail all of the meanings attributed to them. For instance, in Ottoman Turkish documents they were referred to as uluğ hazine, tiyiş, vergi, kesim or pişkeş,24 terms which have various meanings from “great treasure” (uluğ hazine) to “tax” or “tribute” (vergi or pişkeş). Not surprisingly, Polish and Russian diplomats preferred to use terms with less humiliating connotations, like “gifts” (pominki in Russian25 and upominki in Polish)26 or “treasure” (kazna

22 ORRNB, Fond 317, defter 25/115a/4 (this identifies the case as recorded in judicial record no. 25, on page 115a, as the fourth entry); Natalia Królikowska, “Law and Division of Power in the Crimean Khanate,” 193–199. 23 ORRNB, Fond 917, defter 2/55a/1; 25/18a/2; 25/114b/1; 25/120a/2. 24 Halil İnalcık, “Power Relationships between Russia, the Crimea and the Ottoman Empire as Reflected in Titulature,” in Passé turco-tatar, présent soviétique: Études offertes à Alexandre Bennigsen, ed. Chantal Lemercier-Quelquejay, et al. (Louvain and Paris, 1986), 208–211. 25 See, for example, reports of Russian diplomats in 1683: Rossiiskii gosudarstvennyi arkhiv drevnikh aktov (Moscow) Fond 123, no 72, f. 491–532. 26 See, for instance, the Polish version of the Treaty of Zborów concluded in 1649 (Archiwum Główne Akt Dawnych, Warsaw [hereafter, AGAD], Archiwum Koronne War- szawskie, Dz. tatarskie, k. 62, t. 3, no. 334). It is of interest to note that while Polish diplomats denied Tatars the right to collect tribute from the Polish-Lithuanian Commonwealth, they willingly acknowledged their authority to do the same in Muscovy. In 1623, for example, Mehmed III Giray made the following demand in presence of a Polish envoy: “God wants you [the Poles] to pay us tribute, because it is our right to be well-off, to have food, drink and clothing thanks to the giaurs as well as to have our crops harvested by your slaves settled down in our villages.” The khan then added that since the Russians pay the tribute, the Poles should do the same. Although the envoy was not convinced by the last argument, he agreed that the Russians were obliged to pay as “they are from among your [Tatar] hordes.” The Polish-Lithuanian Commonwealth, the envoy continued, used to send the annual payment in the hope of military help and immunity from Tatar raids. It is noteworthy that twenty-five years earlier, the Polish king already had to deal with the problem of the former subordination of Lithuania to the Tatars. Sigismund III Vasa instructed his envoy that he, as the ruler of the newly created Polish-Lithuanian Commonwealth, would not accept a treaty that included the passage copied from earlier Tatar-Lithuanian treaties granting “Kiev and other castles as far as Sokal” by the Tatars. The king was determined not to include this formula, which would have placed him among the khan’s tributaries; see Dariusz Skorupa, Stosunki polsko-tatarskie, 1595–1623 [Polish-Tatar relations, 1595–1623] (Warsaw, 2004), 79, 252. 50 natalia królikowska in Russian).27 The Tatar interpretation was based on Mongol and Islamic traditions: according to the chronicler Seyyid Muhammed Riza, the khan had the right to this tribute because he was the heir of the Golden Horde. He refers to the annual payment delivered by Muscovy as cizye, i.e., the poll-tax payable by non-Muslims, thus he explains the practice of collecting payment from Muscovy in purely Islamic terms.28 It is clear that in the course of time the meaning of the periodic Rus- sian payments changed: the khans seem to have accepted its Muscovite interpretation, because of the growing power of Russia in the second half of the sixteenth century. For the Russians, it meant a gift or a payment that secured peaceful relations; such an interpretation can be found in a letter sent by İslam Bey Cegielski to Polish Crown Chancellor Mikołaj Prażmowski in the 1660s. Its author, the khan’s secretary and a Polish renegade, mentions a report sent by a Tatar spy in Muscovy, who states that the tsar planned to give a treasure (kazna)—which normally was sent to the Crimean khan—to the Kalmyks in return for their military support.29 The usage of the term kazna, therefore, by no means suggests Tatar sovereignty over Muscovy in the mid-seventeenth century. The Rus- sians ceased to deliver a treasure to the khan in 1700. The question arises, how did the Ottomans perceive Muscovy: as their tributary state or as the tributary state of the khanate? Diplomatic relations between these three states, some aspects of which have been discussed by Halil İnalcık and Alan Fisher, indicate that the Ottomans acknowledged the special rights of the khanate over Muscovy.30 As a sign of Ottoman recognition of Tatar sovereignty over Russia, the khan received their tribute for himself and mediated Ottoman diplomatic relations with Muscovy until 1700. Nonetheless, when needed, Tatars themselves used Ottoman power and authority to intimidate the Russians. This is clear, for example, from a letter sent by the khan’s vizier Sefer Gazi Ağa to Tsar Alexei Mikhailovich in 1662. The khan’s vizier reminds the Russian ruler to obey the will of “his Padishah [i.e., the Ottoman sultan], who at the same time is also Padishah of the tsar.”31

27 See note 25. 28 Seyyid Muhammed Riza, Es-sebü’s-seyyar, 99, 166. 29 AGAD, Archiwum Koronne Warszawskie, Dz. tatarski, k. 61, t. 140, no. 283. 30 İnalcık, “Power Relationships,” 175–208; Fisher, “Crimean Separatism in the Ottoman Empire,” 86–87. 31 The letter was quoted by İnalcık, “Power Relationships,” 196. sovereignty and subordination in crimean-ottoman relations 51

In summary, the Crimean khan enjoyed the royal right to collect Islamic taxes in the khanate; our sources indicate that his authority in this area was not limited by the sultan. Moreover, the latter recognized the khan’s right to collect annual payments from Muscovy and the Polish-Lithuanian Commonwealth. Although different meanings were attributed to the payments, it is clear that they represented the recognition of Tatar rights to the heritage of the Golden Horde. Because of this, the Tatars were inclined to label them as “tribute,” whereas those who paid them called them “gifts.” The Sublime Porte left these payments as well as the branch of its international policy concerning Muscovy in the khan’s hands as a way of recognizing his prestigious Genghisid descent.

Spoils of War

Spoils of war constituted an important source of income for the Crimean khan and his subjects. The Tatars took plunder and the khan claimed its percentage; yet, the khan’s share and the legitimate basis of its collection remain somewhat unclear. It seems that the rights of the khan to a share of war booty have their roots in both Islamic and Mongol traditions. According to Hanafi jurists, a Muslim sovereign had rights to one-fifth (hums) of all plunder seized during a holy war—although the interpretation varied as to the meaning of the term “holy war.”32 Mongol customs

32 The Koran Interpreted, trans. Arthur J. Arberry (New York, 1955), 201. The definition of holy war as understood by the Tatars remains unclear. Did they declare every war with non-Muslims as a holy war, or did some types of conflicts fall beyond this notion? If so, did they have the same doubts as the Ottomans did concerning wars against the Shiites and non-Muslims protected by treaty? Eventually, the Ottoman ‘ulemas justified warfare against both types of enemies and declared them holy wars. However, different laws applied to infidels and to the Shiites; the latter were considered apostates, and could not be enslaved, but rather gained the status of prisoner. Therefore, they could be killed or ransomed, but could not be owned; see Imber, Ebu’s-suud, 86–89. Although much about the Tatar understanding of holy war remains unknown, it seems that they applied this term not only to campaigns against non-Muslims, but also to those against the Shiites. Moreover, they took slaves during the campaigns against the Safavids, even against the will of the Ottomans. In 1578, for example, the Ottoman commander-in-chief, Osman Pasha, mentioned to the Tatars that “even in the war zone it had been Ottoman policy to protect the lives and property of the Muslims inhabitants. The Tatars replied that ‘for the Tatars, raiding was as necessary as worldly goods for the repose of ordinary people’ and they let it be known that they could not accept his restriction,” see Kortepeter, Ottoman Imperialism, 58; Remzi Kılıç, XVI. ve XVII. Yüzyıllarda OsmanlıIran Siyasi Antlaşmaları [Ottoman-Persian political treaties in the sixteenth and seventeenth centuries] (Istanbul, 2001), 101–103. 52 natalia królikowska also secured the ruler’s right to a share of war spoils: it was called savga and its amount varied by time and place.33 From Crimean narrative histories and documents we learn that the khan’s share of spoils was referred to by both terms, i.e., as Mongol savga and as Islamic hums. The chroniclers provide ambiguous figures of its percentage. In the 1650s, Mehmed Senai stated that the collection of savga was justified by ancient law (kanuni kadim). According to the chronicler, its amount equaled a one-fifth share (hums) prescribed by Islamic law (şeri‘at).34 Yet, in the mid-eighteenth century, Abdulgaffar el-Kırımi defined savga as a tithe of booty.35 From external sources we learn that the khan’s share was indeed limited to a tithe.36 It seems that while the khans justified their rights referring to both traditions, they were satisfied with a modest share that originated from Mongol customs. Such a decision might have reconciled the need to confirm royal authority with the economic interests of the khan’s subjects. The Ottomans did not interfere in the division of booty, which had already been taken by the Tatars. They tried, however, to set limitations on the territories and groups of people who could be plundered. The Tatars were supposed to respect the Ottoman Empire’s policy toward its neighboring and tributary states, thus Ottoman diplomacy also limited the possibilities for the Tatars to conduct holy wars. For instance, in February 1699, the sultan reminded the Tatars that they were not supposed to make incursions into Muscovite and Pol- ish territories because of the recently concluded treaty.37 Moreover, the Sublime Porte could restrict the types of booty taken by the Tatars. Sül- eyman the Magnificent, for example, forbade Khan Sahib Giray and his troops from enslaving inhabitants of Moldavia in 1538.38 It is noteworthy that the Tatars did not always submit to the Ottoman restrictions, as can be seen in the case of the Ottoman failure to convince the Tatars

33 Mehmed Senai z Krymu, Historia Chana Islama Gereja III [The history of Khan İslam Giray III], ed. Zygmunt Abrahamowicz (Warsaw, 1971), 176n329. 34 Mehmed Senai z Krymu, Historia, 111/30 (the page numbers refer to the Polish translation of the chronicle and its text in Ottoman Turkish). 35 Adet-i Tatar olan savga alınmak hususı mukaleme olındı ki nisfi humsi şer’idir hasıli ‘öşr olsa gerekdir; see Abdulgaffar el-Kırımi, Ümdetü’t-tavarih, 157. 36 Fiodor Lashkov, “Istoricheskii ocherk krymsko-tatarskogo zemlevladeniia” [A historical study on Crimean Tatar land ownership], Izvestiia Tavrıcheskoi Uchenoi Arkhivnoi Kommissii [hereafter, ITUAK] 23 (1895): 88; idem, ed., “Sbornik dokumentov po istorii krymsko-tatarskogo zemlevladeniia” [Collection of sources on the history of Crimean Tatar land ownership], Izvestiia Tavrıcheskoi Uchenoi Arkhivnoi Kommissii 23 (1895): 127. 37 ORRNB, Fond 317, defter 34/19b/1. 38 Remmal Hoca, Tārīh-i Ṣāḥib Giray Ḫān, 28/167. sovereignty and subordination in crimean-ottoman relations 53 not to plunder Muslims during their campaigns against the Safavids in the sixteenth century.39 Moreover, it seems that the Sublime Porte had problems controlling the Tatars even during their crossings through Otto- man territories. Viorel Panaite quotes an order of Selim II, in which the sultan reminds the Tatars—without success—of “the well-known rule of şeri‘at that stipulated obedient tributaries and protected peoples be released and of a hüküm by which the sultan had already ordered that the rebels be ransomed only by their Moldavian relatives and only at a price of 1,000 akçes each.”40 Referring to the war in Hungary (1593–1606), Caroline Finkel notes that “the Tatars often raided the Ottomans’ re‘aya, stealing their animals and possessions.”41 To protect their subjects, the sultans tried to restrict the Tatar route through Ottoman territories: Sul- tan Ahmed I, for example, ordered the Tatars not to cross the Danube on their way to the campaign against the Habsburgs in 1605.42 Ottoman military campaigns that the Tatars considered unimportant from a political perspective and which caused them financial losses constituted another problem in mutual relations. In such circumstances, the khan could rise against the sultan, or could face a rebellion by his unsatisfied Crimean and Nogay vassals. Vasilii Aitemiriev, a Muscovite envoy to the Crimea in 1695, described such a conflict: December 8 [1695]: On the request of the karaçı beys,43 the Şirin and Mansur Beys,44 the kaymakam Apkerim Efendi along with the mufti, the kadiasker and the well-known mirzas and ağas, who lived in Bahçesaray, set out to [a place] in the vicinity of Kara Su.45 There, they held a council from which the

39 Remzi Kılıç, XVI. ve XVII. Yüzyıllarda, 102. 40 Viorel Panaite, “The Re‘aya of the Tributary-protected Principalities: The Sixteenth through Eighteenth Century,” in Ottoman Borderlands: Issues, Personalities and Political Changes, ed. Kemal H. Karpat and Robert W. Zens (Madison, WI, 2003), 88. 41 Caroline Finkel, The Administration of Warfare: The Ottoman Military Campaigns in Hungary in 1593–1606 (Vienna, 1988), 204. 42 Le Khanate de Crimée dans les Archives du Musée du Palais de Topkapı, ed. Alexandre Bennigsen et al. (Paris, 1978), 143. For more examples of illegal Tatar raids against Ottoman territories, see Panaite, “The Re‘aya of the Tributary-protected Principalities,” 86–89. 43 Karaçı families formed the highest circle of the Crimean nobility. They traced their origins back to the Golden Horde and they owed their elevated status to their economic and military power. They were headed by beks, who extended a great influence over the khanate’s policy and held substantial power within its borders. 44 In the seventeenth century, the Şirins and the Mansurs were the most powerful karaçı families in the Crimean Khanate. 45 “White Rock” (Ak Kaya), which is situated in vicinity of Kara Su, was a traditional place of periodic assemblies (kurultay) of the Tatar nobility. 54 natalia królikowska

kaymakam and those others [who had arrived] from Bahçesaray returned on December 16. And the news spread from this council that it had been said that the khan and the Crimean army should return to the Crimea and defend their homes. For the last ten years, the Crimean warriors have helped the Turkish army every year. Not only have [the Tatars] received no benefit [from these campaigns], but they lost many people and horses. They kept dying, weakening and losing their horses during such long expeditions. Because of [these campaigns,] all the Tatars are going to perish. Only last summer, when the Crimean and Bucak armies were absent from home, the Russian army and the Don and Zaporozhian Cossacks raided the territories of Crimea and Bucak. They ravaged villages and settlements. They wiped out [their inhabitants]. Thus [the council] decided that [the Tatars] should not go to Hungary or distant Wallachia, but should defend the Crimea against the Cossacks’ pillage. It is written by the unbelievers in the Qur’an: If somebody has his own close enemy and instead of fighting that enemy helps to fight a faraway enemy, it is not right. And after they [i.e., the participants of the council] had written down their decisions, they sent [them] to the khan in Adrianople by a servant of the khan’s court, who was named Hacı Receb. And [they notified Khan Selim I Giray] that if he wished to remain their khan he should immediately return to the Crimea. If not, they would find another khan in the Crimea.46 It is clear that the Crimean nobles did not want to take part in the Otto- man holy war when they felt that it threatened their vital interests. More- over, they denied the khan the right to jeopardize their security in order to help the sultan in a distant campaign against infidels. In conclusion, collecting booty caused extensive trouble in Tatar- Ottoman relations. It seems clear that the Ottomans recognized the Tatars’ right to take spoils of war and the Crimean khan’s right to a share of one-fifth of all plunder. Yet, the Sublime Porte and the Tatars defined the lawful mode of acquisition of booty in a different manner. As noted above, the Ottomans tried to limit Tatar raiding activities in order to protect their non-Muslim vassals, subjects, and allies, and also considered pillaging Muslim enemies as an offense. The history of the Persian campaigns in the 1570–1580s indicates that the Tatars did not share this point

46 Vasilii Aitemiriev received this information from his translator, who was present in front of the kaymakam just before the departure of Hacı Receb, and saw the letter written to the khan; see A.I. Markievich, ed., “Spisok s stateinogo spiska poddiachavo Vasiliia Aitemirieva, posylannovo v Krym s predlozheniiem mirnykh dogovorov (7200/1692– 7203/1695)” [An excerpt from the diplomatic report of Poddiachi Vasilii Aitemiriev, sent to the Crimea with a proposal for peace negotiations], Zapiski Odesskogo Obshchestva Istorii i Drevnostei 19 (1896): 48–49. sovereignty and subordination in crimean-ottoman relations 55 of view: they considered the military conflicts with the Safavids a legitimate source of slaves and goods.

Execution of Fixed Penalties

The execution of fixed penalties for the five Qur’anic offenses—that is, fornication, false accusation of fornication, wine-drinking, theft, and highway robbery—constituted another area subject to the exclusive authority of the ruler. The Qur’anic requirements to prosecute, however, were so strict that they were seldom imposed in Islamic states. In their place, Ottoman jurists rather inflicted discretionary punishments such as flogging, fines or imprisonment, which in the Ottoman Empire were then executed by the sultan’s officials.47 The Crimean court registers reveal little about the execution of fixed penalties in the khanate. What is known suggests some analogies to the Ottoman practice. The khan assigned the task to the Crimean kadis, who were the most important officials in the province. They administered justice in the judicial districts (kazas), which covered the entire territory of the khanate.48 The kadis had a variety of minor officials at their disposal (naibs, katibs, muhzırs, kassams, delals), and could additionally rely on the help of other provincial officials, such as law enforcement officers called zabits. The latter, recruited from among the janissaries, reported to the kadis as well.49 From the court registers we learn that individuals accused of committing the Qur’anic offenses—for example, robbery on a highway50

47 Rudolph Peters, Crime and Punishment in Islamic Law: Theory and Practice from the Sixteenth to the Twenty-First Century (Cambridge, 2005), 75–102. 48 In general, the kadis were the khan’s appointees; this is clear in the opening formula of the Crimean judicial records, as well as in the khan’s extant orders recorded there. Yet, the kalga or the karaçı families seemed to influence the choice of the more important kadis in some districts. Still, one can argue that the latter, at least formally, remained the khan’s appointees, see Barbara Kellner-Heinkele, Aus den Aufzeichnungen des Said Giray Sultan: Eine zeitgenössische Quelle zur Geschichte des Chanats der Krim um die Mitte des 18. Jahrhunderts (Freiburg am Breigau, 1975), 212–246, 317; Charles Peysonnel, Traité sur le commerce de la mer Noire (Paris, 1787), 282; ORRNB, Fond 917 defter 25/3b/5; 23b/69a/1; 25/25a/4; 23a/146b/7. 49 For more detail about the Crimean judiciary, see Natalia Królikowska, “Law and Divi- sion of Power,” 87–158; 168–176; 189–228. 50 For instance, Ali Çelebi, a scribe of the beytülmal emini, accused a cotton worker, Aslan, of wounding a fellow worker. The official met Aslan and his suffering companion near the road linking Bahçesaray and Kara Su and brought both of them to the court. The defendant stated that his friend had wounded himself by accident while they were having a good time together. The injured worker confirmed this story: “It happened after we got 56 natalia królikowska or illicit sex51—were brought to the judicial court.52 By investigating the cases recorded in the Crimean court registers and narrative accounts, one can conclude that the Crimean kadis adjudicated the cases concerning the five Qur’anic offenses. Yet, they did not impose the fixed penalties prescribed by the şeri‘at. As in the Ottoman Empire, the kadis inflicted discretionary punishments, for example, flogging for illicit sex. Moreover, the Crimean kadis did not rely exclusively on classical standard proofs, such as confession or accounts of eyewitnesses, but also used torture to extract confessions from the accused in well-defined circumstances. It seems, therefore, that Crimean officials applied elements of administrative justice (siyaset), which, presumably, had the effect of extending the ruler’s authority, similar to the practice in the Ottoman Empire. It was the khan, and not the Ottoman sultan, who was responsible for delivering justice to the khanate’s inhabitants; therefore, the Crimean ruler exercised the privilege of an Islamic sovereign to dispense justice in

drunk. Because of that I wounded myself.” The scribe was satisfied with the explanation and withdrew his charges (ORRNB, Fond 917, defter 25/116a/5). The immediate reason for the official’s intervention was most probably the fact that the assault took place near the public road. 51 For instance, a Kara Su zabit, Bölükbaşı Cafer, brought to the local court a married woman named Saime and a certain İbrahim, son of Hasan. The official found them together at night in the woman’s house while her husband was absent. Saime defended herself by saying that she had found İbrahim already in her house when she had returned home in the evening. She stated that she had not invited him. On his part, perhaps to avoid more serious charges of robbery or rape, İbrahim maintained that he had come upon her invitation. He produced witnesses who confirmed his version. Consequently, both defen- dants were found guilty of having illicit sex and the judge sentenced them in all likelihood to flogging (ta‘zir ile hükm şod), see ORRNB, Fond 917, defter 25/31a/3; 25/48a/4. 52 It is noteworthy that yet another Qur’anic offense, theft (sarika), was rarely reported to the Crimean kadis. Presumably, it was difficult to successfully condemn someone for theft, because it was so narrowly defined in Islamic law (Peters, Crime and Punishement, 56). Perhaps for this reason, the aggrieved parties used to sue their opponents for usurpation, which had a broader definition. In case of usurpation, the original owner had to demand the return of goods himself. The story of Nicolas Kleemann, a merchant from the Habsburg Empire, who was robbed in his house by his Armenian translator, indicates that the aggrieved party could rely on the help of the judge and the zabits: he gave a detailed description of how the subordinates of the judge caught the accused Armenian and brought him before the court. His story also indicates that elements of administrative justice (siyaset) were applied in cases of this type in the Crimean Khanate; see Nicolas Kleemann, Voyage de Vienne à Belgrade et à Kilianova, Dans le pays des Tartares Budziacs et Nogais dans la Crimée, et Kaffa à Constantinople; au travers de la mer Noire; avec le retour à Vienne; par Trieste. Fait dans les années 1768, 1769 et 1770, par Nicolas-Ernest Kleemann: On y a joint description des choses les plus remarquables concernant la Crimée (Neuchatel, 1780), 90–107. sovereignty and subordination in crimean-ottoman relations 57 the territory of his state. It is worth noting that the khan’s officials used to apply elements of administrative justice, which brought some aspects of the judicial procedure and punishment under the authority of the ruler.

Symbols of Subordination and Sovereignty in Tatar-Ottoman Relations

From our investigation we can conclude that the Crimean khan exercised power in the “four areas” of exclusive royal authority. Yet it is clear that in all these areas his power was limited in some aspects by the Ottoman sultan. As noted, our sources do not reveal the exact circumstances in which the khans resigned their regal monopolies and accepted the protection of the Sublime Porte. The degree of the khan’s subordination to the Ottomans also remains unknown. It can be argued that the procedures of succession to the Crimean throne reflect these ambiguities. Crimean Tatar tradition provided some rules concerning succession. First, as the khanate was a continuation of the Genghisid Empire, the khan had to be a male member of the Giray dynasty, which traced its origins to Genghis Khan. Second, the eldest male was expected to come to the throne. Yet, the new khan was supposed to be selected by the kurultay. This assembly of Crimean nobles was dominated by the karaçı beys, the most powerful among the khan’s vassals. One should not disregard the fact that their diverging political goals, as well as the personal ambition of various members of the Giray dynasty, were of the utmost importance in the process of selecting a new khan. Thus, any member of the Giray dynasty could in principle and practice come to power as the candidate of a successful faction. It comes as no great surprise that the rival groups sought the Sublime Porte’s support to turn the scales in their favor. Consequently, the conflicts allowed the Sublime Porte to interfere more and more frequently in matters of Crimean succession. It should also be noted that the Ottomans and the Crimean nobles defined the sultan’s role in the selection of the khan differently: the former claimed to have the right to appoint the new ruler, while the latter limited the sultan’s role to the confirmation of the kurultay’s decision. In time, the Ottoman role grew more important. This was already evident during the struggle between Tohtamış Giray and Selamet Giray in 1608. Thanks to the noblemen’s support, Tohtamış Giray came to the throne after the death of his father, Gazi II Giray. As soon as the traditional 58 natalia królikowska

Tatar ceremony of accession had taken place,53 Crimean envoys were sent to the sultan to ask him to confirm this choice. The Ottomans, however, feared that Tohtamış Giray would continue his father’s policy of seeking independence; therefore, they supported Selamet Giray, the only living brother of Gazi II Giray. It is noteworthy that the prince was literally in the sultan’s hands at the time, as he was imprisoned in Istanbul for taking part in the rebellion of Deli Hasan (1602).54 It was probably this very fact that made the Crimean nobles decide in favor of Tohtamış Giray, a younger member of the dynasty. After the nomination, Selamet Giray left for the Crimea. At the same time, Tohtamış Giray departed to Istanbul to seek the sultan’s support personally, but he was murdered by clients of the kalga of Selamet Giray in the vicinity of Akkerman.55 Afterward, the Ottoman nominee seized the Crimean throne. In the first half of the seventeenth century the criteria used to choose the khans varied. Being the eldest male member of the dynasty did not secure the throne. This period was also marked by a growing Ottoman interference in Crimean succession: each and every successful candidate was supported by the sultan. Yet, quite often as soon as a Giray prince came to the throne, he tried to oppose Ottoman policy. A prince such as Mehmed III Giray managed to keep the Crimean throne, thanks to domestic support and the backing of the Cossacks, thus the Ottomans could not remove him for four years (1624–1628). His reign came to an end when he lost the support of the Nogays and the majority of his Crimean vassals.56 Beginning at the end of the sixteenth century, when the balance of power turned against the khanate, the position of the Tatars weakened to the extent that they were no longer allowed to choose their own rulers. The new khan was more and more frequently appointed by the sultan in the Ottoman capital and sent to the Crimea only afterwards, to be accepted by the local nobles. In the eighteenth century, the ceremony of

53 The traditional Tatar ceremony of accession emphasized the vital role of the Crimean karaçı beks, who carried the new khan seated on a felt cushion into the council’s room, where they literally put the ruler on the throne; see Seyyid Muhammed Riza, Es-sebü’s- seyyar, 126; Smirnov, Krymskoe khanstvo, 467. 54 According to Caroline Finkel, Selamet Giray supported the rebellion of Deli Hasan hoping to find allies to win the Crimean throne for himself; see her Osman’s Dream: The Story of the Ottoman Empire 1300–1923 (London, 2005), 182. 55 Seyyid Muhammed Riza, Es-sebü’s-seyyar, 126; Smirnov, Krymskoe khanstvo, 467–472. 56 For more about Crimean-Ottoman relations in the 1620s, see Sándor Papp, “Die Inau- gurationen der Krimkhane durch die Hohe Pforte (16.–18. Jahrhundert),” in The Crimean Khanate between East and West (15th–18th Century), ed. Denise Klein (Wiesbaden, 2012), 75–90. sovereignty and subordination in crimean-ottoman relations 59 kurultay election became just a formality. Nonetheless, the nobles could still force the Sublime Porte to remove an unpopular ruler.57 As we can see, there were two strands to the khan’s claims to legitimacy. First, he was justified to rule as a Genghisid prince, selected according to the Tatar rules of succession. Second, his right to rule was legitimized by Ottoman consent. In time, the latter justification grew increasingly important. These two legitimizing paths were also reflected on a symbolic level. The khans had some insignia of authority underlining their Genghisid heritage. The trident-shaped Giray family mark (tamga) was one of the most important symbols of steppe sovereignty, and was displayed not only at the entrance to the khan’s council room,58 but also on some items of vital symbolic meaning, such as diplomatic treaties issued by the khans59 or Crimean silver coins.60 The fact that the khans retained the right to maintain individual diplomatic relations and mint their own coins points to their status as sovereign rulers. Khanate ownership of a network of post stations perhaps constituted yet another symbol of the khans’ sovereignty inherited from the steppe. From Crimean sicils61 and European narrative accounts62 we learn that the khans made arrangements for the upkeep and expenses of the post stations, which were designed to serve the khan’s administration. It seems that the ruler authorized only a limited number of individuals to benefit from the system. It remains unclear whether the Tatars inherited the network of post stations from the Golden Horde, or whether it was modeled on the Ottoman postal system (ulak).63 According to Charles Peysonnel, the Crimean post stations were better organized than those of the Otto- mans, which might suggest that the Crimean network was inherited from the Golden Horde, rather than being an improved version of the Ottoman

57 See note 47. 58 Nicole Kançal-Ferrari, Kırım’da Kalan Miras: Hansaray [A treasure preserved in the Crimea: The palace of the khan] (Istanbul, 2005), 179–186. 59 Sagit Faizov, Tugra i vselennaia: Mokhabbat-name i shert-name krymskikh khanov i printsev v ornamental’nom, sakral’nom i diplomaticheskom kontekstakh [Tuğra and the world: Ornamental, sacral and diplomatic aspects of the mohabbetnames and şertnames of the Crimean khans] (Moscow and Bakhchesaray, 2002), 20. 60 Şevket Pamuk, A Monetary History of the Ottoman Empire (Cambridge, 2005), 105. 61 See the yarlığ of Selim II Giray on the organization of post stations (menzilhane) in 1744/1745: ORRNB, Fond 917, defter 67/1a/2. 62 Peysonnel, Traité sur le commerce, 299–300. 63 The Ottoman system of post stations seems to be indebted to the Mongol yam as well; Lütfi Paşa ve Tevārihi Āli Osman [Lütfi Pasha and the chronicles of the House of Osman], ed. Kayhan Atik (Ankara, 2001), 284. 60 natalia królikowska institution.64 Yet the origins of the Crimean post system remain unclear and merit further research. Many sovereign rulers emphasized their unique position by wearing special clothing with a symbolic design. The question of whether some parts of the costumes of the Crimean khans symbolized their claims to the heritage of the Golden Horde is of great interest. If such items existed at all, it was likely a cap with sable trim. The chronicle of Abdulgaffar el-Kırımi implies that this symbolized the khan’s authority.65 Moreover, Ottoman miniatures make clear that headgear constituted the most important difference between Ottoman and Tatar clothing.66 Sixteenth- century Ottoman artists depicted the khans wearing caps of different col- ors with fur trim. They seem to be of traditional Tatar design, as members of the khan’s retinue were painted in similar headgear.67 It is of the utmost importance that the Tatar headgear on the Ottoman miniatures resembles the so-called Monomakh’s cap in shape and design. This gold cap with sable trim was most probably given by Khan Özbek to the Muscovite grand prince Ivan Kalita (1325–1340); thus the khan of the Golden Horde must have sent it to his vassal as an insignia of authority. Until the mid-fifteenth century, it signified the subordination of Musco- vite rulers, but it continued to play the role of the crown long after the fall of the Golden Horde. In that period, the cap, which was co-opted to the Byzantine tradition, became a symbol of Muscovite sovereignty.68 In a similar vein, the Crimean rulers’ custom of wearing a sable cap might date back to the Golden Horde. Originally it may have been designed as a symbol of the Crimean governors’ subordination, but when they established an independent khanate, the sable cap must have become a symbol of their descent from Genghis Khan, and thus legitimized their right

64 For the Mongol postal system (yam), see Adam J. Silverstein, Postal Systems in the Pre-Modern Islamic World (Cambridge, 2007), 141–164. 65 Abdulgaffar el-Kırımi’s chronicle also includes a fascinating story about the kalga sultan of Gazi II Giray and the daughter of a Polish hetman. The girl was enslaved by the kalga during one of the Tatar raids into the Polish-Lithuanian Commonwealth. When she was finally returned home, she was pregnant. Her family claimed that the kalga was the father of the child. The chronicler depicted the dilemma of the Girays, who learnt that one of them, who might one day be entitled to “the sultan’s sable cap” (sultanın samur kalpağı) would remain in Polish hands; see Abdulgaffar el-Kırımi, Ümdetü’t-tavarih, 116. 66 Ilya Zaitsev, “Krymskie khany: portrety i suzhety” [The Crimean khans: Portraits and motives], Vostochnaia kollektsia 1 (2003): 86–93. 67 Cf. pictures of Mengli Giray and members of his retinue in the Hüner-name of Seyyid Lokman; Zaitsev, “Krymskie khany,” 87. 68 Donald Ostrowski, Muscovy and the Mongols: Cross-Cultural Influences on the Steppe Frontier 1304–1589 (Cambridge, 2002), 174–175. sovereignty and subordination in crimean-ottoman relations 61

Table 1. Insignia given to Khan Mengli I Giray by Sultan Mehmed II in various chronicles Chronicler Insignia Abdullah, son of Rizvan Pasha (1630–40s);69 drum and standard (“tabl ve ‘alem”) İbrahim Kefeli (1730–40s)70 drum and standard (“tabl ve ‘alem”) Seyyid Muhammed Riza (1750s)71 horsetail, standard, and money (“tuğ, ‘alem ve teşrifat hani”) Halim Giray (1820s)72 white felt cap and cap with gold embroidery (“ak börk, altınlı üsküf ”) to rule. As to the question of whether the khans passed down this special sable cap to the next ruler, the sources available reveal nothing about such a crown insignia; therefore, it is suggested that the Crimean khan had the right to wear a sable cap, but it was not necessarily inherited from his predecessors. In the Ottoman tradition, numerous items linked a sovereign and a vassal. The insignia of authority (hükümet alametleri) sent by the sultans emphasized Ottoman suzerainty over the subordinated polities. The types and the quality of items sent from the Sublime Porte varied depending on the rank of the addressee. According to some seventeenth- and eighteenth- century Crimean chroniclers, Mengli I Giray had already been given various symbolic attributes when he was appointed khan after the conquest of Kefe. It is worth noting that descriptions of the insignia given to Mengli I Giray made by the seventeenth- and eighteenth-century chronicles show differences in terms of their type and number. This list suggests that there was no unanimous tradition on the possible insignia of authority given to Mengli I Giray: their exact types and number remains unclear. It is even uncertain that this khan received them at all; the descriptions might have been influenced by later customs, as the above list consists of items given to Ottoman vassals throughout the early modern period.73 Each of the above mentioned hükümet alametleri had

69 Abdullah, son of Rizvan Paşa La chronique des steppes kipchak, 34. 70 İbrahim Kefeli, Tevarihi Tatar Han ve Dagistan Moskov ve Deşti Kıpçak Ülkeleri nindir [Chronicle about the lands of the Crimean Khan, about Dagistan, Muscovy and the Quipchak steppe], ed. İbrahim Otar (Eskişehir, 2005), 79. 71 Seyyid Muhammed Riza, Es-sebü’s-seyyar, 74. 72 Halim Giray, Gülbünü Hanan, 167. 73 See, for example, the insignia given to Khan Sahib Giray in 1532 or Khan Murad Giray in 1678; Remmal Hoca, Tārīh-i Ṣāḥib Giray Ḫān, 21; Franciszek Kluczycki, ed., Źródła do poselstwa Jana Gnińskiego wojewody chełmińskiego do Turcyi w latach 1677–1678 [Sources 62 natalia królikowska deep symbolic meanings in the Turkish-Ottoman tradition. The custom of giving a horsetail came from Turkish and Mongol shamanic traditions;74 standards and drums, mentioned by three out of four chroniclers, were among the most typical items given by the sultan to his vassals: according to Maria Pia Pedani, “in the ancient Turkish world the ruler used to give a banner to his new vassal, such as that given by the Selgiukid sultan to Osman.”75 Various types of headgear constituted the last type of insignia attributed to Mengli Giray. According to the Turkish-Ottoman tradition, bestowing a headgear created a bond between a sovereign and a vassal. Therefore, when the Crimean khan put on a headgear received from the Ottoman Empire instead of his own sable cap, he was acknowledging the sultan’s suzerainty.76 The above mentioned items were insignia typically given to both Muslim and non-Muslim vassals of the Ottoman Empire. From other sources we learn that the Crimean khans, in contrast to the Christian vassals, such as the Danubian hospodars, were given various types of weapons.77 Accord- ing to Maria Pia Pedani, it “was a symbol of peace relation and meant that the sultan trusted so much a person that he could send him a weapon.”78 This type of insignia emphasized the elevated position of the Crimean khans, and also symbolized their subordination to the Ottomans.

Conclusions

Crimean-Ottoman relations changed significantly during the early modern period. In the mid-sixteenth century accusations that Sahib Giray wanted to challenge Ottoman power may have found credibility by the chronicler’s readers.79 In 1569, the Tatars succeeded in foiling the Ottoman project of recapturing Astrakhan from Muscovite hands. Apparently, the Ottoman presence on the northern shore of the Caspian Sea threatened

for the embassy of the voievod of Chełmno, Jan Gniński to Turkey in the years 1677–1678] (Warsaw, 1907), 64–80. 74 For more details on the subject, see Bahaeddin Ögel, “Tuğ,” in Islam Ansiklopedisi (Istanbul, 1988), vol. 13, 1–5. 75 Maria Pia Pedani, “Sultans and Voivodas in the 16th c.: Gifts and Insignia,” Journal of International Social Research 1 (2007): 201. 76 Ibid., 200. 77 Remmal Hoca, Tārīh-i Ṣāḥib Giray Ḫān, 21; Zaitsev, “Krymskie khany,” 88. 78 Pedani, “Sultans and Voivodas,” 202. 79 Remmal Hoca, Tārīh-i Ṣāḥib Giray Ḫān, 114–115. sovereignty and subordination in crimean-ottoman relations 63

Tatar claims to the Genghisid heritage.80 Two years later, the khan attacked and burned Moscow. In the 1620s, Mehmed III Giray ruled the khanate without Ottoman consent. In the eighteenth century, however, the khans no longer dared to oppose the Ottoman sultans: for example, Devlet II Giray gave up the throne in 1702 despite the strong support of his vassals.81 When the turning point in Ottoman-Crimean relations occurred remains a question: it seems likely that it should be dated to the second half of the seventeenth century, and may be connected with changes in power relations in Central Europe and the Black Sea region. A report by the Polish emissary Stanisław Karwowski dates from this period, from which it is known that he was secretly received, at the end of 1671, by the Crimean khan Selim I Giray. The ruler had gathered his troops in a military camp near Bahçesaray for the war with the Polish- Lithuanian Commonwealth and then the khan made the following apolo- gies to the emissary for the uncommon circumstances of their meeting: “Do not mind that I receive you in this way, but I must keep our meeting a secret before the [Ottoman] kapıcı başı. After all, the Turks have already announced you as an enemy in all their lands.” When Karwowski stated that the Polish-Lithuanian Commonwealth trusted the agreement based on mutual friendship, as stipulated in the Polish-Tatar treaty, Selim I Giray answered: With all my soul I regret the loss of your friendship, because the former king showed me a lot of kindness. I know all your dignitaries very well, especially the Marshall with whom I spent such a long time during the war in Ukraine. But what shall I do? It is not my will, but the will of My Lord, My Emperor, with whose grace I ascended to this throne of mine. And I will tell you something in secret, but do not repeat it. If I saw you as strong as before, I would support you and I would say: “Hold on [My] Emperor, the Poles are not guilty, they are our friends, who stand by us whether in good or in bad.” The whole Crimea would repeat these words, because many of their rights are also violated by the Porte. But when I see you so impoverished, weak and friendless, what shall I do? I would ruin myself, while I would not save you (. . .) For a long time the Porte prepared against you, but either we advised against [the war with you] or we frightened them, [saying] that you are powerful and rich. But now when they have had a closer look at you, they would disdain of you and say that within a year they would take your land as far as Warsaw or Cracow. Please understand, therefore, that neither would I send my envoy with you to the King nor could I dismiss

80 Smirnov, Krymskoe khanstvo, 325. 81 Abddulgaffar el-Kırımi, Ümdetü’t-tavarih, 137. 64 natalia królikowska

you without conferring with the [Ottoman] kapıcı başı. Otherwise I would be suspected.82 The issues discussed enable us to draw some general conclusions about Tatar-Ottoman relations. First, the Ottomans successfully interfered in the appointments and depositions of the Crimean khans. Yet, the choice of the khanate’s new ruler was limited to the members of the Giray dynasty, and the Tatars could force the Sublime Porte to remove an unpopular ruler.83 The sultan gave the khan insignia of authority, which underlined the khan’s subordination to the Ottomans. In addition, the khan received military and financial support that strengthened his position vis-à-vis his vassals and neighboring rulers.84 Second, the khan raised a claim to be an independent ruler; this was symbolized by the insignia of authority that referred to his Genghisid descent. From Karwowski’s report we learn that the khan might have willingly acted independently, but to do so, he needed a strong ally. In the first half of the seventeenth century, for example, Mehmed III Giray managed to maintain an independence in policy-making for four years, but was eventually forced to leave the Crimea. Apparently, the Crimean Khanate was not powerful enough to oppose the Sublime Porte neither alone, nor with only modest external help: in time, the khans were more frequently forced to follow Ottoman policies, and this in turn led to tension between the Ottomans and the Tatars, who felt that their rights and vital interests had been violated by the former. Moreover, it should be noted that power in the khanate was divided between the khan and the noblemen described in Karwowski’s report as “the whole Crimea.” The noblemen, as stated by Aitemiriev, claimed to be able to depose and choose the khan by themselves, even against the Ottoman will. Whoever wanted to utilize the Crimean army apparently had to consider gaining not only the khan’s, but also the noblemen’s support. Finally, the khan exercised power in all four areas distinguished by the Hanafi jurists as symbols of sovereignty. It should be stated that in the

82 Acta Historica res gestas Poloniae illustrantia, vol. 2, Pisma do wieku i spraw Jana Sobieskiego [Papers concerning the age and activities of Jan Sobieski], ed. Franciszek Kluczycki (Cracow, 1880), vol. 1, part 1, 748–749. 83 For instance, Tatar nobles convinced the sultan to remove Kara Devlet Giray in 1717 and Saadet III Giray in 1724; Abdulgaffar el-Kirımi, Ümdetü’t-tavarih, 161, 164. 84 On financial support, see Alan Fisher, “Les rapports entre l’Empire Ottoman et la Crimée: L’aspect financier,” Cahiers du Monde Russe et Soviétique 13, no 3 (1972): 368–381; Mária Ivanics, “Entstehung und Quellenwert der krimtatarischen Tiyiş Defters,” Acta Ori- entalia Academiae Scientiarum Hungaricae 47 (1994): 115–122. sovereignty and subordination in crimean-ottoman relations 65 case of Friday prayers, tribute, and even spoils of war, the khan’s position was to some extent subordinated to and limited by the Ottoman sultan, who nonetheless did not interfere in the khanate’s tax and judicial policy. It is noteworthy that in these two areas the khan enjoyed similar prerogatives in the Crimea as the sultan in his own lands. It is clear that there is still much to learn about Tatar-Ottoman relations, and it is safe to say that a better understanding of this subject will follow from the deepening of our knowledge on regional diversities in the Ottoman world. The comparison of the khanate’s status with other non- core lands of the Ottoman Empire could provide answers to the fascinating question: to what extent was it unique?

Between Vienna and Constantinople: Notes on the Legal Status of the Principality of Transylvania*

Teréz Oborni

In the sixteenth and seventeenth centuries, the Principality of Transyl- vania, a vassal of the Ottoman Empire, functioned as a state with limited sovereignty. While the documents relating to Transylvania’s vassal status have been well-studied, little attention has been devoted to the documents defining the legal relationship between Transylvania and the Kingdom of Hungary under Habsburg rule. In fact, there are a significant number of treaties and agreements that attempt to define and re-define the complex relationship between the two states.1 What is more, research on these treaties and agreements shows that in many cases the two parties had contrasting interpretations of the relationship between Transyl- vania and its prince on one hand, and the Kingdom of Hungary and its monarch on the other. This article does not address the vassal status of the Principality of Transylvania because several studies have examined it already.2 Instead,

* For his suggestions and help in the translation of this paper’s final version, I thank Gábor Kármán; also Endre Sashalmi and Gábor Ágoston for our very stimulating discussions on the topic. 1 These documents are published in Roderich Gooss, ed., Österreichische Staatsver- träge: Fürstentum Siebenbürgen (1526–1690) (Vienna, 1911); Sándor Szilágyi, ed., Erdélyi országgyűlési emlékek [Documents of the Transylvanian Diets], 21 vols. (Budapest, 1874– 1898) (hereafter: EOE). 2 Pál Fodor, “Ottoman Policy towards Hungary,” Acta Orientalia Academiae Scientiarum Hungaricae 45, no. 2–3 (1991): 271–345; idem, Magyarország és a török hódítás [Hungary and the Ottoman conquest] (Budapest, 1991); Gábor Kármán, “Transylvania between the Ottoman and Habsburg Empires,” in Statehood Before and Beyond Ethnicity: Minor States in Northern and Eastern Europe 1600–2000, ed. Linas Eriksonas and Leos Müller (Brussels, 2005), 151–158; Gábor Barta, “The First Period of the Principality of Transylvania (1526– 1606),” in History of Transylvania, vol. 1, From the Beginning to 1606, ed. László Makkai and András Mócsy (Boulder, CO, 2001), 593–770; Ioan-Aurel Pop, “Tra gli Asburgo e gli Ottomani: la Transilvania alla metà del XVI secolo,” Annuario: Istituto Romeno di Cultura e Ricerca Umanistica 6–7 (2004–2005): 185–204; Călin Felezeu, “The Legal Status of Tran- sylvania in its Relations with the Ottoman Porte,” in History of Transylvania, vol. 2, From 1541–to 1711, ed. Ioan-Aurel Pop, Thomas Nägler, and András Magyari (Cluj-Napoca, 2009), 49–74. 68 teréz oborni

I focus on the complex relationship through which the principality was linked to the Kingdom of Hungary. The present study is only a brief summary of my ongoing research and covers only parts of the principality’s century-and-a-half-long history. Still, the study approaches the legal status of the Principality of Transylvania in the early modern period from a new point of view. From the perspective of the Kingdom of Hungary, the discussion of the legal status of the principality during its 150-year-long existence can be pursued along three main lines. The first revolves around the name and title of the ruler of Transylvania (voievod or prince). The second is the conceptualization and definition of Transylvania as a province (provincia) or a country (regnum) as laid down in the relevant treaties concluded between the representatives of the Kingdom and Transylvania. The third concerns the recurring debates over the free election of the prince (libera electio). The definition of these legal categories is crucial in order to understand the establishment and the legal nature of the principality, thus the treaties and agreements concluded between the Kingdom of Hungary and the Principality of Transylvania, as well as the laws passed at the diet of Transylvania constitute an important body of sources for our topic. As the topics and perspectives above are closely interrelated and appear together in most of the legal documents, the three aforementioned questions continue to reoccur throughout the analysis.

The Formation of the Transylvanian State and the Title of Its Ruler

In the Middle Ages, the territory called Transylvania functioned as an integral part of the Kingdom of Hungary. The legal terminology of the age—surviving well into the sixteenth and seventeenth centuries— conceptualized this relationship by calling Transylvania a member (membrum) of the Hungarian Holy Crown.3 At the same time, this theoretical

3 According to the organic theory of the Hungarian state, codified for the first time by István Werbőczy (1458?–1541), the country (regnum) constituted a body, and the concept of the Holy Crown included the king, the nation, and the body of the state; see Stephen Werbőczy, The Customary Law of the Renowned Kingdom of Hungary: A Work in Three Parts (1517), ed. and trans. Janos M. Bak, Peter Banyo, and Martyn Rady, with an introductory study by Laszlo Peter (Budapest, 2006); István Csekey, “Werbőczy és a magyar alkotmány- jog” [Werbőczy and Hungarian constitutional law], in Werbőczy István: Acta Juridico- Politica, vol. 2 (Kolozsvár, 1942), 71–90; Ferenc Eckhart, A szentkorona-eszme története [The history of the doctrine of the Holy Crown] (Budapest, 1941); József Kardos, A Szent Korona és a Szentkorona-eszme története [The history of the Holy Crown and the doctrine of the notes on the legal status of the principality of transylvania 69 framework implied that Transylvania was also a part of the body of the Hungarian state (regnum).4 In sixteenth-century usage, the concepts of regnum and that of the Holy Crown were interchangeable from a territorial point of view, as the parts constituting the state were also constitu- ents of the crown.5 In the Middle Ages, the term most frequently used for Transylvania was “partes Transylvanae,” and it was understood as “the Transylvanian parts of Hungary.” However, from the mid-fifteenth century on, Transylvania was referred to as a separate part of the Holy Crown (Regnum Transilvanense, Erdélyország). The deputies of the three estates of Transylvania participated in the diets of Hungary, the troops from Transylvania were part of the Hungarian army, and the regalian incomes from the territory were administered by the Hungarian royal treasury.6 Although Transylvania was a separate province of the Kingdom of Hungary—due partly to its geographical position, partly to its peculiar societal development—in the sense of political law, it existed as an integ ral part of the Hungarian state; this is also clear from the structure of its government. In the development of the provincial government, the huge geographical distance of Transylvania from the center (or, when it existed, the capital) of the Kingdom played an important role. By the first half of the thirteenth century, it was already customary for the king to appoint a high official called vajda (voievod, that is, military leader) to the top of the

Holy Crown] (Budapest, 1992); Gábor Hamza, ed., Studien über István Werbőczy (Buda- pest, 2001); Barna Mezey, ed., Magyar alkotmánytörténet [Hungarian constitutional history] (Budapest, 2003), 74–82, 117–119, 178–179; Kees Teszelszky, “The Holy Crown for a Nation: The Symbolic Meaning of the Holy Crown of Hungary and the Construction of the Idea of a Nation,” in Building the Past: Konstuktion der eigenen Vergangenheit, ed. Rudolf Suntrup and Jan. R. Veenstra (Frankfurt am Main, 2006), 247–259; Endre Nagy and Lajos Rácz, Magyar alkotmány- és közigazgatástörténet [Hungarian constitutional and institutional history], ed. Gábor Máthé (Budapest, 2007); Kees Teszelszky, Az ismeretlen korona: Jelentések, szimbólumok és a nemzeti identitás [The unknown crown: Meanings, symbols and the national identity] (Budapest, 2009). 4 In contemporary Hungarian political terminology, apart from the territory of the state, regnum also had a meaning close to res publica, that is, the political community living in the territory, see Eckhart, A szentkorona-eszme, 240. 5 Eckhart, A szentkorona-eszme, 246. 6 Elemér Mályusz, Az erdélyi magyar társadalom a középkorban [Hungarian society in medieval Transylvania] (Budapest, 1988); András Kubinyi, “Erdély a Mohács előtti évtizedekben” [Transylvania in the decades before Mohács], in Tanulmányok Erdély történetéről, ed. István Rácz (Debrecen, 1988), 65–76; Zsigmond Jakó, “Mátyás király erdélyi társadalompolitikájáról” [On the social policies of King Matthias in Transylvania], Korunk, ser. 3, 2, no. 4 (1990): 426–433; Pál Engel, Gyula Kristó, and András Kubinyi, Magyarország története 1301–1526 [A history of Hungary 1301–1526] (Budapest, 1998), 238–240. 70 teréz oborni

provincial administration. The voievod was one of the three major ranks of the Kingdom of Hungary and as such, he functioned as the locum tenens of the king of Hungary in Transylvania. He had the supreme right over administration, jurisdiction, and military leadership, bestowed upon him by the king. A parallel to his office can be found in the similar functions of another rank, the ban, who stood at the highest level of the administration of the Banate of Croatia, Slavonia, and Dalmatia, which territories were also seen as membrum of the Kingdom of Hungary. The yearly salary of the voievod was paid by the Hungarian royal treasury.7 The juridical praxis of the voievod followed the one established in the Hungarian royal court, and the court of appeal against his decisions was that of the Lord High Steward (országbíró) of Hungary.8 In the second half of the fifteenth century, aside from regulating local administrative issues, the most important task of the voievods became the military defense of the province. Otto- man attacks became more frequent, especially in the first quarter of the sixteenth century, and reached their peak during the mandate of János Szapolyai, who was a voievod from 1510 to 1526. Due to the attacks, the administration and the military of the province had to be re-organized and strengthened; thereby considerably increasing the power concentrated in the hands of the provincial governor. The task of the voievods to organize an efficient local administration was supported by the establishment and the progressive development of their chancelleries beginning with the first third of the fifteenth century.9 The history of Transylvania after the tragic battle of Mohács in 1526 is closely intertwined with that of the kingdom of János Szapolyai and the part of the country ruled by him. As is well known, the estates of Hungary elected two kings in 1526: János ( John) I Szapolyai and Ferdinand I of Habsburg. As both elections and coronations took place according to the

7 In the year of 1504, it was 9,000 florins per year, completed with salt worth approximately 3,000 florins. See Martinus Georgius Kovachich, Supplementum ad vestigia comitio- rum apud Hungaros ab ex ordio regni in Pannonia, usque ad hodiernum diem celebratorum (Buda, 1800), vol. 2, 305. 8 Iván Janits (Borsa), Az erdélyi vajdák igazságszolgáltató és oklevéladó működése 1526-ig [The juridical and charter-giving activities of the voievods of Transylvania until 1526] (Budapest, 1940), 56–57; Zsigmond Jakó, Az erdélyi vajda kancelláriájának szervezete a XVI. század elején [The structure of the chancellery of the voievod of Transylvania in the early sixteenth century] (Kolozsvár, 1947); Zsigmond Jakó, “Az erdélyi vajdák kinevezéséről” [On the appointment of the voievods of Transylvania], Levéltári Közlemények 63 (1992): 71–82. 9 Anna Pécsi, Az erdélyi fejedelmi kancellária kialakulása és okleveles gyakorlata 1571-ig [The establishment of the chancellery of the voievod of Transylvania, and its charters until 1571] (Budapest, 1938); Jakó, Az erdélyi vajda kancelláriájának szervezete. notes on the legal status of the principality of transylvania 71 regulations, both ruled as legitimate kings in the same country, but with its territory divided between the two: while John ruled the lands east from the river Tisza, together with Transylvania, Ferdinand held the western territories. The territorial division of the country was canonized with the Peace of Várad (1538), after a long period of civil war (1527–1538): this, however, was planned as a temporary solution until the death of King John, at which point both parts of the country were expected to be united under the rule of Ferdinand. Although John died in 1540, only two weeks after the birth of his son, the unification of the country did not take place. The most important reason for this was the Ottoman attack against the country in 1541 and the conquest of its capital, Buda. It should nonetheless be noted that eo ipso the birth of a successor changed John’s earlier plans altogether. Before his death he requested that the Hungarian estates not allow the country to fall under Habsburg rule, but rather should elect his son, János Zsigmond Szapolyai ( John Sigismund) as their new king—and this was exactly what happened in the autumn of 1540. John II Sigismund was, however, never crowned; therefore he could only use the title electus Rex (king-elect) until the end of his life, as it is clear from the analysis below. The relationship of Transylvania (and thus the later Principality of Transylvania) and the Ottoman Empire was also formed in peculiar circumstances. I believe that an important precedent to the formulation of Transylvania’s “special status” and its acknowledgment by the Porte was King John’s alliance with Sultan Süleyman in 1528.10 The stipulations granted in this alliance were then extended by the sultans to John Sigismund and later to his successor, István Báthory, who specifically asked for it; it also remained a point of reference for the future princes of Transylvania. Because of this connection, the conditions for the vassal status of the Principality of Transylvania turned out to be better than

10 Két tárgyalás Sztambulban: Hieronymus Łaski tárgyalása a töröknél János király nevé- ben / Habardanecz János jelentése 1528. nyári sztambuli tágyalásáról [Two negotiations in Istanbul: Hieronymus Łaski’s talks at the Porte on behalf of King John / Report of Johannes Habardanecz about his talks in Istanbul during the summer 1528], ed. Gábor Barta and Pál Fodor (Budapest, 1996); Gábor Barta, La route qui mène à Istanbul 1526–1528 (Budapest, 1994); idem, “The Forgotten Theater of War 1526–1528 (Historical Events Preceding the Ottoman–Hungarian Alliance of 1528),” in Hungarian–Ottoman Military and Diplomatic Relations in the Age of Süleyman the Magnificent, ed. Géza Dávid and Pál Fodor (Budapest, 1994), 93–130. 72 teréz oborni those of Moldavia and Wallachia, the other two vassal states in its closest proximity.11 According to the law promulgated by the estates of Transylvania in 1567, a new ruler would be freely elected from among their ranks after the death of their current lord, János Zsigmond.12 They claimed that the right of free election was given by the sultan to their ruler, who acquired a diploma to that effect at great expense. From 1571 onward, the Transylva- nian princes were elected by the Transylvanian estates based on the laws of 1567. Throughout the history of the principality, the estates insisted on their right to choose the prince freely or at least to maintain this illusion.13 This was also true in those cases when a preliminary agreement with the Porte was made about the person of the next prince, or even when the latter ordered Ottoman troops to bring a new prince to the principality, as in the cases of Gábor Bethlen (1613) or Mihály Apafi I (1661). As is clear from these cases, the demands of the Porte and loyalty to the Ottomans played a defining role in most elections, which caused some historians to think that the right of the Transylvanian estates for free election was never put fully into practice.14 However, I believe that the situation should be assessed in a more nuanced manner: in several cases, the power relations among the estates and the skillful politics of the candidate in winning

11 There were important differences between Transylvania and the two Romanian principalities. Unlike Moldavia and Wallachia, there were no Ottoman troops stationed in Transylvania. Whereas Istanbul often appointed the Moldavian and Wallachian voievods, the Transylvanian diet elected its rulers, whom the Porte usually accepted. Also, the practice by which the Romanian voievods had to send their sons as hostages to Istanbul was not introduced in Transylvania. For more detail, see Klára Hegyi, Török berendezkedés Magyarországon [Ottoman rule in Hungary] (Budapest, 1995), 20–23; Géza Dávid, “Admin- istration in Ottoman Europe,” in Suleyman the Magnificent and His Age: The Ottoman Empire in the Early Modern World, ed. İ. Metin Kunt and Christine Woodhead (London, 1995), 88–89. 12 Diet of Gyulafehérvár, 8 September 1567: “if His Majesty were to be taken from this world by the Lord Almighty, we should not be divided among us, which could lead us into utter destruction, but elect a prince for ourselves with common understanding; as His Majesty won this privilege from the sultan and gained a letter about this election at great expenses”; EOE, vol. 2, 1556–1576 (Budapest 1876), 335. 13 See Graeme Murdock, “ ‘Freely Elected in Fear’: Princely Elections and Political Power in Early Modern Transylvania,” Journal of Early Modern History 7, nos. 3–4 (2003): 214–244; László Makkai, “The Crown and the Diets of Hungary and Transylvania in the Sixteenth Century,” in Crown, Church and Estates: Central European Politics in the Sixteenth and Seventeenth Centuries, ed. Robert J. W. Evans and Trevor V. Thomas (Basingstoke, 1991), 80–91. 14 Zsolt Trócsányi, Az erdélyi fejedelemség korának országgyűlései (Adalék az erdélyi rendiség történetéhez) [Diets in the age of the principality of Transylvania: Contributions to the history of the Transylvanian system of estates] (Budapest, 1976). notes on the legal status of the principality of transylvania 73 them over had the most important impact on the outcome of the election. The choice of the Transylvanian estates had to be confirmed with an ‘ahdname by the sultan in all cases: the prince’s position became legal only through this document. The Porte was often dissatisfied with only having the right of confirmation and demanded preliminary information about the would-be prince. The inauguration ceremony to follow the election was meant to represent the sultan’s authority over Transylvania together with the symbols used during the ceremony: the decorated horse, the scepter, the fur cap, the flag, and the kaftan sent for the prince along with the ‘ahdname—the latter two symbolizing his fealty to the sultan.15 The territorial frames of the new country were defined by the decision of the sultan after the conquest of Buda (1541). Süleyman assigned Transyl- vania and a southern territory connected to it, the so-called Temesköz, to the widow and son of King John. The formation of the new state apparatus started in the legislative assemblies of the estates of eastern Hungary, held between 1542 and 1544, which followed the Ottoman occupation of Buda and central Hungary in 1541.16 At these meetings, the estates of Transylva- nia and the adjacent eastern territories of the Hungarian Kingdom created a new constitution for the nascent state, based on the constitution of the medieval Kingdom of Hungary, and also defined the operation of the leg- islature, government, and jurisdiction of the country.17 Thus, these general diets also served as constitutional assemblies. Antal Verancsics (secretary of King John I and of the dowager Isabella after his death) summarized the deliberations of the diet of December 1542 with the following words: “We have decided to accept the son of the deceased as our prince, to pay

15 See János B. Szabó, “Insignia of the Transylvanian Princes,” Majestas 4 (1996): 85–105; János B. Szabó and Péter Erdősi, “Ceremonies Marking the Transfer of Power in the Princi- pality of Transylvania in East European Context,” Majestas 11 (2003): 111–160. 16 Cristina Feneşan, Constituirea principatului autonom al Transilvaniei [The making of the autonomous Principality of Transylvania] (Bucharest, 1997); Teréz Oborni, “From Province to Principality: Continuity and Change in Transylvania in the First Half of the Sixteenth Century,” in Fight Against the Turk in Central Europe in the First Half of the 16th Century, ed. István Zombori (Budapest, 2004), 165–180; eadem, “A gyalui szerződés” [The treaty of Gyalu], in A magyar államiság első ezer éve, ed. Márta Font and István Kajtár (Pécs, 2000), 133–146; Gábor Barta, Az Erdélyi Fejedelemség születése [The making of the principality of Transylvania] (Budapest, 1979). 17 Katalin Péter, “Az erdélyi országgyűlés a kora újkori magyar fejlődésben” [The diet of Transylvania in early modern Hungarian history], in Kálmán Benda and Katalin Péter, Az országgyűlések a kora újkori magyar történelemben [Diets in early modern Hungarian history] (Budapest, 1987), 21; Gábor Barta, Az Erdélyi Fejedelemség, 72–91. 74 teréz oborni tribute to the sultan, and ask King Ferdinand not to take us into danger, and let us look after ourselves if he cannot help us.”18 In 1543, the estates decided that they would send the tribute offered to the sultan and thank the Sublime Porte for appointing the son of King John as the new ruler of the country. Simultaneously, they requested that the sultan allow them to elect another prince (princeps) from their own ranks if the child king died.19 The diets held between 1542 and 1544 had two important consequences. First, by recognizing Sultan Süleyman’s appointees, Queen Isabella and her son John Sigismund, as their rulers, the Transylvanian estates acknowledged the suzerainty of the Ottoman sultan. Paying an annual tax to the sultan also showed the country’s vassal status. During the assemblies, the estates repeatedly declared their loyalty to the sultan and expressed that they regarded themselves as subjects of the Porte. By the middle of the century, the terminology referring to these general diets of the principality had become standard and remained unchanged until the 1690s: “the estates and orders of the three nations of Transylvania and of the adjacent parts of Hungary” (status et ordines trium nationum regni Transilvaniae Partiumque Hungariae eidem annexarum).20 In the mid-sixteenth century when the Transylvanian estates repeatedly attempted, on the initiative of Vienna, to join the remaining parts of the Hungarian Kingdom under Habsburg rule, the Ottoman government intervened. Constantinople made it clear that the two parts of Hungary could under no circumstances be held by the same ruler. Yet, despite its obvious limited sovereignty and dependence on the Ottoman Empire, the Principality of Transylvania established a special relationship with Habsburg Hungary. King Ferdinand I never renounced his claim to Transylvania.21 More- over, he made considerable efforts to annex it to his kingdom; however, he was only able to retain Transylvania for six years. After this interim period

18 “(. . .) statutum est, ut filium defuncti principem habeamus, ut cesari Turcarum tributum mittamus, ut rogemus Ferdinandum Regem, ne in periculum nos adducat si tueri nequit et sinat nosmet ipsos nostri curam gerere” EOE, vol. 1 (1540–1556) (Budapest, 1875), 172. 19 EOE, vol. 1, 176. 20 EOE, vol. 1, 189. 21 See Teréz Oborni, “Die Plane des Wiener Hofes zur Rückeroberung Siebenbürgens 1557–1563,” in Kaiser Ferdinand I.: Ein mitteleuropäischer Herrscher, ed. Martina Fuchs, Teréz Oborni, and Gábor Ujváry (Munster, 2005), 277–298; idem, “Die Herrschaft Ferdi- nands I. in Ungarn,” in Kaiser Ferdinand I.: Aspekte eines Herrscherlebens, ed. Alfred Kohler and Martina Fuchs (Munster, 2003), 147–166. notes on the legal status of the principality of transylvania 75 of Habsburg rule, between 1551 and 1556, the next stage in the evolution of the new state was the formation of its governmental and institutional structure. These were organized by Queen Isabella, following her return from Poland (1556). The Transylvanian estates enacted their right to the free election of their ruler in 1567, but they asked for (and received) the approval of the Porte and not that of the king of Hungary. It was obvious from the beginning of the formation of the new state that the consent of the sultan was essential for the election of the Transylvanian ruler. While the principality was dependent on the Ottoman Porte during the reign of John Sigismund, in the 1560s it carried out ongoing negotiations with King Maximilian of Hungary (r. 1564–1576; Maximilian II as Holy Roman Emperor) on the principality’s legal status and the title of its ruler. The two issues were closely linked. The special relationship between Transylvania and the Kingdom of Hungary was acknowledged for the first time in the Treaty of Speyer, concluded in 1570 by John Sigismund and King Maximilian of Habsburg. In this treaty, the Transylvanian ruler renounced his title of “elected king of Hungary” (electus rex Hungariae) and was content with the title of prince of Transylvania and of the Adjacent Parts of Hungary (Princeps Transyl- vaniae et partium regni Hungariae eidem annexarum). 22 He also acknow ledged that Transylvania was part (membrum) of the Hungarian Crown and that the king of Hungary was his sovereign.23 The agreement created a peculiar situation, at least in theory: the sovereignty of the principality was limited by both the Ottoman Empire and the Kingdom of Hungary, ruled by the Habsburgs. Nevertheless, neither party acknowledged the other’s authority over the principality. It is noteworthy that the text of the treaty does not mention the Ottoman Empire, neither does it refer to John Sigismund as a ruler connected to the sultan in any way. The parties codified the borders according to the status quo, and defined their relationship toward each other according to their agreement, reached after years of exhaustive negotiations.

22 Gooss, Österreichische Staatsverträge, 182–204. See Teréz Oborni, “Erdély közjogi helyzete a speyeri szerződés után (1571–1575)” [The legal status of Transylvania after the treaty of Speyer], in Tanulmányok Szakály Ferenc emlékére, ed. Pál Fodor, Géza Pálffy, and István György Tóth (Budapest, 2002), 291–306. 23 “(. . .) serenissimus princeps unacum suis descendentibus posteris habebit et recog noscet Sacram Caesaream Regiamque Maiestatem pro capite totius Christianitatis, pro rege Hungariae et maiore et superiore suo, ipsamque Transylvaniam ac Partes Hungariae, quas possidet, pro membro regni Hungariae” Gooss, Österreichische Staatsverträge, 191. 76 teréz oborni

However, the Treaty of Speyer did not solve the problem altogether. John Sigismund died in the very year of its conclusion (1571) and a legal crisis arose in Transylvania. On one hand, the treaty declared that after the extinction of the Szapolyai dynasty, the territories under their rule would revert to the king of Hungary, so it seemed likely that Maximilian would launch a military campaign against Transylvania, justified by the legal principle that declared Transylvania to be part (membrum) of the Hungarian Holy Crown. On the other hand, the estates of Transylvania were afraid that after the death of the last Szapolyai, the Porte would want to change the status of the principality and would intensify its pressure, rather modest till that moment. The estates, while afraid of both dangers, insisted on their right to free election, and chose a new ruler, István (Stephen) Báthory, who turned out to be extremely talented. He solved the crisis by not taking the princely title; instead, he called himself a voievod (vajda), thereby displaying loyalty toward both emperors. Báthory made a secret oath of loyalty to Maximilian, in which he declared that he regarded himself as voievod of the king of Hungary in Transylvania.24 This oath was kept as a closely guarded secret from the Porte; only a few Transylvanian aristocrats knew about it. The title voievod was also acceptable to the Porte, for it showed Báthory’s subjugated status: it was also the title of the rulers of Istanbul’s two Romanian vassals, Wallachia and Moldavia.25 With the consent of his counselors, Báthory sent a copy of the ‘ahdname of John Sigismund to the Porte as early as

24 The secret oath, taken on 25 May 1571, read as follows: “Ego Stephanus Báthory de Somlyó, Vaywoda Transsylvanus et Comes Siculorum juro per Deum vivum etc. Quod Serenissimo et potentissimo Principi et Domino, Domino Maximiliano, Electo Romano- rum Imperatori semper augusto, ac Hungariae, Bohemiae, Dalmatiae, Croatiae, Sclavo- niae etc Regi, Archiduci Austriae etc. Domino meo semper clementissimo eiusdemque heredibus et successoribus, legitimis scilicet Hungariae regibus a modo imposterum, fidelis, obediens ac subditus atque in hoc vaivodatus mei officio, hostibus suae Maiestatis hostis, amicis amicus ero. Arces et bona suae Maiestatis, que manibus meis fuerint tradita, pro sua Maiestate et eiusdem successoribus legitimis videlicet Hungariae Regibus fideli- ter conservabo et conservari curabo, omnes status et ordines Regni Transylvaniae in suis juribus et libertatibus tenebo, et tam cntra internos, quam externos hostes, pro viribus defendam. Omnibus soram me causantibus rectum et verum iudicium et iustitiam admin- istrabo, prece, premio, odio, favore, amore et complacentia semotis ac quandocunque per Eandem Caesaream et Regiam Maiestatem praefatum officium Vaivodatus, unacum arci- bus et bonis fidei meae concreditis a me repetetur, idem libenter sine omni difficultate et subterfugio eidem suae Maiestati vel cui eadem jusserit, restituam et assignabo. Ita me Deus adiuvet.” Published in Endre Veress, Báthory István erdélyi fejedelem és lengyel király levelezése [The correspondence of István Báthory prince of Transylvania and king of Poland], vol. 1, 1556–1575 (Kolozsvár, 1944), 115–116. 25 Sandor Papp, Die Verleihungs-, Bekraftigungs- und Vertragsurkunden der Osmanen fur Ungarn und Siebenbürgen: eine quellenkritische Untersuchung (Vienna, 2003), 11–140. notes on the legal status of the principality of transylvania 77

1572, and asked that the same document be assigned to him as well. This was important in order to have the same powers and warranties that the deceased prince had possessed, and to secure the position of Transylvania from the side of the Ottoman Empire. In the same year he also managed to obtain the consent of the Porte regarding his new title.26 While the new ruler of Transylvania was trying to save his country from the military and political interventions of both powers, he formed his own political view regarding the status of his country: in his opinion, Transylvania lay between the two most powerful rulers of the world, in their pivots (in arcto), thus the country could only be preserved by seeking the goodwill of both emperors. One of them had to be won by declaration of obedience; the other by presents and subservience.27 István Báthory began to use the title of prince officially only after he became king of Poland in 1576. As king, he tried to maintain a relationship with the Porte that was neither too hostile, nor too friendly. Holding a much more important position, he still wanted to maintain the balance that he aimed for as the ruler of Transylvania.28 Báthory’s election as Pol- ish king brought no change to the legal position of Transylvania, nor did its inhabitants think that their situation would be altered. Rudolph I, the new king of Hungary, was nevertheless forced to accept Báthory as prince, thus his Polish royal title was a strong political asset. By the mid-1570s, a delicate situation had arisen: Transylvania was “officially” the vassal of the Porte and feared Istanbul’s power; at the same time, the principality secretly recognized the suzerainty of the king of Hungary, and repeatedly ensured the latter and the Viennese government of its loyalty in secret letters and agreements.

26 Ferman of Selim II to István Báthory, at the beginning of March 1572. EOE, vol. 2, 462; see also the letter of Báthory to Maximilian II, 18 November 1572, in Veress, Báthory István, 222 (Nr. 221). 27 “(. . .) considerantes Provinciam istam inter duos potentissimos orbis terrarum Monar- chas, ceu in arcto positam, humanisque viribus et auxilio prorsus destitutam, non aliter, quam utriusque Imperatoris favore nobis conciliato, . . . conservari et gubernari posse, omni tempore, totisque viribus iuxta officium nostrum id dedimus operam, omnesque huius provinciae reditus et facultatis eo convertimus, ut altero Imperatorum obsequio, altero vero muneribus et subiectione nobis conciliato, provincia ista quam maxime tuta et quieta ab externis hostibus omnique periculo conservata maneret.” Wolffgangus de Bethlen, Historia de rebus Transylvanicis (Cibinii, 1783), vol. 2, 333. 28 About the relationship of Báthory as king of Poland and the Ottoman Empire, see Felicia Roşu, “Contractual Majesty: Electoral Politics in Transylvania and Poland–Lithuania (1571–1586)” (PhD diss., Georgetown University, Washington, DC, 2009), 146–150; Kemal Beydilli, Die polnischen Königswahlen und Interregnen von 1572 und 1576 im Lichte osmani scher Archivalien (Munich, 1976). 78 teréz oborni

Treaties and Agreements between Transylvania and the Kingdom of Hungary at the Beginning of the Seventeenth Century

The Treaty of Speyer was followed by other agreements, which were adjusted to the actual political situation and regulated the relationship between Transylvania and the Kingdom according to contemporary power relations. It is, however, important to emphasize that these treaties did not codify the actual political and legal situation, but rather signaled the pretensions of the kings of Hungary with regard to Transylvania. The kings used various means in order to regain Transylvania and the eastern parts of Hungary: they repeatedly besieged the castles situated on the northeastern border, attempting to extend their own territories toward Transylvania; and tried to obtain the support of influential Transylva- nian aristocrats by way of political and diplomatic maneuvers, and hence divert the country from the Porte and force the prince to declare loyalty to the Habsburgs. Whenever possible, they also launched military campaigns to remove the princes supported by the Porte; and finally, through diplomatic negotiations, Vienna intended to restore Transylvania’s former status as a province of Hungary. Apart from the short period under Ferdinand I, the other attempt to re-incorporate Transylvania into the Kingdom of Hungary took place in the first years of the seventeenth century, during the Long Turkish War. Prince Zsigmond Báthory (r. 1588–1602, with interruptions) resigned from his throne for the benefit of Rudolph in 1602, and thus Transylvania was united with the western half of the medieval Kingdom of Hungary between 1602 and 1604, under the rule of the Habsburgs. At this time, the king of Hungary built his claim on former agreements, partly on the Treaty of Speyer, and partly on the Treaty of Prague (1595). In the Treaty of Prague, Zsigmond Báthory and Rudolph I entered into an alliance to undertake a joint struggle against the Ottomans. The two parties also agreed that in the event Zsigmond Báthory died without an heir, Transylvania would revert to the king of Hungary and the Hungarian Crown. It was declared that in this case the king would appoint a gover- nor (gubernator) or a voievod as the leader of the province.29 The Long

29 “(. . .) ut decedente sine haeredibus masculinis ipso principe vel illis superstitibus, ac in hac eius linea legitima deficientibus ipsa Transylvania eique annexae ditiones non in alterius quam in Suae Maiestatis eiusque successorum Hungariae Regum ditionem et potestatem tanquam verum et inseparabile membrum absque ulla controversia pervenire debeat. (. . .) Tenebitur praeterea rex Ungariae a tempore quo Transylvania ad coronam notes on the legal status of the principality of transylvania 79

Turkish War and this phase of Zsigmond Báthory’s rule were the most critical years of the history of Transylvania, not only because of the hardships of the war, but also because of the sudden, almost yearly changes in the political status of the principality.30 The final resignation of Zsigmond Báthory in 1602, when he ceded power to King Rudolph, was in fact his fourth resignation in only a few years. In this period of turmoil, two other members of the Báthory family, Boldizsár and András (the latter a former cardinal), were also princes of Transylvania for short periods. In several other instances during the same period, the country was governed by a locum tenens: for a time, Zsigmond Báthory entitled his wife, Maria Chris- tierna of Habsburg, to be his deputy; she held the same position later, in the name of King Rudolph of Hungary. The voievod of Wallachia, Michael the Brave, was also a locum tenens of Rudolph and thus the governor of the province for almost a year from November 1599 on. Directly after the last resignation of Zsigmond Báthory, between 1602 and 1604, General Giorgio Basta led the country’s administration as a military governor; following this, royal commissioners came to Transylvania and immediately suspended the fundamental laws of the country as well as the activities of the diet. A more detailed discussion of developments in this period would, however, require another study. Although Zsigmond Báthory left the principality for good in 1602, Tran- sylvania could not be placed under the authority of the king of Hungary permanently. It is known that the peace treaties of Zsitvatorok and Vienna, which ended the “Long War” and the Bocskai insurrection of 1604–1606 respectively, jointly determined Transylvania’s legal and political relations toward the Ottoman Empire and the Kingdom of Hungary. As a consequence of István Bocskai’s successful insurrection and the Ottoman support given to him, the principality survived as a separate political entity, and was subordinated to the Ottoman Empire under the terms set in the mid-sixteenth century.31 Due to their unsuccessful attempt to annex

devolvetur, unum pro suo placito ex ordinibus Transylvaniae gubernatorem seu Waivo- dam illorum provinciarum creare.” Gooss, Österreichische Staatsverträge, 228. 30 See Meinolf Arens, Habsburg und Siebenbürgen 1600–1605: Gewaltsame Einglie- derungsversuche eines ostmitteleuropäischen Fürstentums in einen frühabsolutistischen Reichsverbund (Cologne, Weimar, and Vienna, 2001). 31 Andrea Molnár, Fürst Stefan Bocskay als Staatsmann und Persönlichkeit in Spiegel seiner Briefe 1598–1606 (Munich, 1983); Teréz Oborni, “Der Siebenbürgische Staat von Bocskai,” in “Einigkeit und Frieden sollen auf Seiten jeder Partei sein”: Die Friedensschlüsse von Wien (23.06.1606) und Zsitvatorok (15.11.1606), ed. János Barta, Manfred Jatzlauk, and Klára Papp (Debrecen, 2007), 63–73. 80 teréz oborni

Transylvania, the Habsburgs also had to accept Transylvania’s status. However, they did not give in, and carried out a decade-long political maneuvering to achieve their goal. The election of the princes of Transylvania was a recurrent issue in the negotiations and in both secret and public agreements concluded between the princes of Transylvania and the kings of Hungary in the sixteenth and seventeenth centuries. The Hungarian kings repeatedly expressed their opinion that the appointment of the princes of Transylvania, which was in their view a province of the kingdom, was their prerogative. I have already noted how the title of the ruler of Transylvania was defined in negotiations during the 1560s and 1570s; this question also had an outstanding significance in the agreements in the first half of the next century. During the negotiations leading to the Treaty of Vienna in 1606, the head of the Habsburg delegation, Archduke Matthias, summarized his views regarding the legal status of the Principality of Transylvania and its prince as follows: István Bocskai, the elected prince of Transylvania, should be allowed to keep his title, but with the clear understanding that the province itself belonged to the Kingdom of Hungary, and Bocskai and his legitimate heirs should acknowledge the sovereignty of the Hungarian king. To prove their loyalty, they should pay an annual tax, and a Tran- sylvanian delegation should be present at the Hungarian diet. If Bocskai were to die, Transylvania would revert to the Hungarian king, who could then give the province to whomever he wished.32 One of the Hungarian diets of the early seventeenth century also ruled that Transylvania and its adjacent territories should belong to the Hungarian Holy Crown.33 Nego- tiations regarding the handing over of Transylvania were continually

32 “(. . .) placet ut is Transylvaniae principatum una cum titulo et indigniis provinciae et suis propriis gentilitiis ad modum quo Sigismundus Bathori habuit, teneat; ita tamen, ut a nomine principis Hungariae, eiusque indigniis penitus abstineat, singulariter vero ipsa provincia Regno Hungariae incorporata maneat et propterea tam ipse, quam sui successores ex legitimo thoro in recognitionem superioritatis quotannis munera quaedam suae Maiestati et eiusdem successoribus regibus Hungariae per certum hominem mittant; et ad omnes diaetas per nuntios suos compareant; et ut ipsis sine liberis masculis ex legitimo thoro natis decedentibus haec ipsa Transsylvania cum partibus ad ipsum spectantibus, prout tenuerint et possederint, rursum ad regnum devolvatur et liberum regi sit, eandem cui voluerit, conferre.” Árpád Károlyi, ed., Magyar Országgyűlési Emlékek / Monumenta Comitialia Regni Hungariae (Budapest, 1883), vol. 11, 427–428. 33 Korpona, 1605. “Regnum hoc ipso a se invicem non separabitur, sed et haec pars regni quam illi daret, una cum Transylvania a Corona dependeret, qua inparte multas expensas et curas Maiestatis Caesareae Celsitudo Sua, dominus noster Bochkay sublevaret.” Ibid., 494; Kálmán Benda, “Absolutismus und ständischer Widerstand in Ungarn am Anfang des 17. Jahrhunderts,” Südost-Forschungen 33 (1974): 85–124. notes on the legal status of the principality of transylvania 81 renewed with the kings of Hungary, Matthias II (1608–1619) and Ferdi- nand II (1619–1637) during Gábor Báthory’s reign (1608–1613) and in the first five years of the rule of Gábor Bethlen (1613–1629). This new round of negotiations began in 1608 between King Matthias II and the newly elected prince Gábor Báthory. Lasting several years, the goal of these discussions was to establish peace between the Kingdom of Hungary and Transylvania and to settle the disputes regarding the legal status of the latter. During the negotiations, the Viennese royal court reit- erated its standpoint that the prince of Transylvania was the subject of the king of Hungary, irrespective of the fact that he was appointed by the Ottoman sultan. In order to make this point, the ruler of Transylvania was consistently addressed by Vienna as a voievod and not as a prince, thereby reflecting his subordination to the Hungarian king. Gábor Báthory was willing to establish a closer relationship with the Kingdom and even to acknowledge the sovereignty of the king of Hun- gary, but he was forced to abdicate before he could do so.34 Báthory concluded two important treaties with the king: first in 1608 and then again in 1613. The first, the so-called Treaty of Kassa, proclaimed, among other things, that Transylvania and the adjoining counties could not be alienated from the Crown of Hungary by the prince in any way other than at that time.35 This meant that the principality was still interpreted as a part of the Crown, but it was also acknowledged that it existed in a somewhat alienated form, under the temporary rule of a prince. Matthias II ratified the treaty after his coronation as king of Hungary, on 4 December 1608.36 However, we find proof already in the narratio part of the ratification document that the two sides continued to differ in their interpretations about the status of the prince. In the document, the prince of Transylvania was called by the king “our faithful follower, the illustrious prince, voievod of Transylvania and comes of the Seklers.”37 This formulation suggests a clear claim that the king of Hungary held sovereignty over the prince. In the concluding clause following the articles of the agreement, however, the prince is only referred to as “wayvoda Transylvaniae.” So, even if the title “prince” does appear in the text, it is always completed with the title

34 Teréz Oborni, “Báthory Gábor megállapodásai a Magyar Királysággal” [The agreements between Gábor Báthory and the Kingdom of Hungary], in Báthory Gábor és kora, ed. Annamária Jeney-Tóth, Klára Papp, and Attila Ulrich (Debrecen, 2009), 111–122. 35 Gooss: Österreichische Staatsverträge, 376–377. 36 Ibid., 374–376. 37 “(. . .) cum fidelis nostri illustris principis domini Gabrielis Báthory de Somlio, way- wodae Transsylvaniae, Siculorum comitis,” ibid., 376. 82 teréz oborni

“voievod”; this, together with other documents, shows that the king preferred to call Gábor Báthory simply a voievod. Through this, they referred back to the medieval governmental system described in the introductory part of this article, according to which the voievod of Transylvania had always been a royal office holder in the province. The treaty of 1608 brought no satisfaction to either side; therefore, negotiations were continued between Gábor Báthory and Matthias II. In June 1612, the prince sent a memorandum to the Transylvanian estates. In it, he declared that while he was working on establishing an alliance with the king of Hungary against the Ottomans, he was also trying not to antagonize the sultan, because it was well known that a country where the Turks appointed the ruler would find itself in a miserable situation: it would be better for all of us to die rather than for a prince to come to our beautiful country from the Porte of the Turkish emperor against our free election; we would reach the pitiful condition of Wallachia and Moldavia, and we would shrink just like them, as a result of such frequent changes.38 The Hungarian royal councilors summarized their views regarding the status of Transylvania in a document dated 7 January 1613. They emphasized that Transylvania belonged to the Kingdom of Hungary according to all their existing laws, thus Ottoman occupation had to be avoided.39 The agreement that was concluded between Gábor Báthory and Matthias II in Pozsony in 1613 repeated the articles agreed upon in Kassa: both sides promised mutual aid; stated that Gábor Báthory should remain prince of Transylvania and ruler of the adjacent territories of Hungary, as well as comes of the Seklers; and that his successors would inherit the same titles. The accord also declared that the kings of Hungary acknowledged—and would acknowledge in the future—the right of the Transylvanian estates to freely elect their princes.40 The Transylvanian estates knew very well

38 “(. . .) mindnyájunknak jobb volna meghalni, hogysem mint szabad választásunk ellen a török császár portájáról jönne fejedelem szép hazánkban, Havasalföld és Molduva keserves állapotjára jutnánk, és gyakor változásokban, mint szintén ők, mi is szintén úgy fogynánk.” Memorandum of Prince Gábor Báthory to the estates, 26 June 1612, EOE, vol. 6, 1608–1614 (Budapest, 1880), 238. 39 “(. . .) ne per simultates inquitorum hominum vel etiam ulla ratione ipsa provincia— qua omni jure sacrae Regni Hungariae Coronae iurisque ipsius propria habetur—hostibus Turcis usurpanda et adimenda relinquatur.” ibid., 265. 40 “(. . .) ipse et successores eius legitimi principes, et Transilvania in libera electione per suam Maiestatem Caesaream et Regiam eiusque successores legitimos reges Hungariae et regnum conservabuntur (. . .) tam ipse, quam futuri principes et sui successores similiter etiam Transylvaniae et partes Hungariae eidem subiectae obstricti sint ad observationem huius transactionis.” ibid., 269. notes on the legal status of the principality of transylvania 83 that they had already acquired the right to freely elect their princes from the Ottoman Porte, and had never consulted Vienna regarding their choice (what is more, in several instances their election was against Vienna’s will). Thus, this point was clearly only a manifestation of the claims from the side of the king of Hungary that they, and not a ruler of another empire, should be the ones to give permission to Transylva- nians to free election. In an annexed secret document, Gábor Báthory also promised the king of Hungary to take his side in the event that he started a war against the Ottomans.41 Gábor Bethlen, the prince with the greatest diplomatic talent in seventeenth-century Transylvania, was also aware that his title and position had to be accepted by both emperors; this had been a tradition and constraint for the princes of Transylvania since the reign of István Báthory. In 1613, right after his ascension to the throne, Bethlen entered into negotiations with the Viennese government about the status of Transylvania and his own position. In the spring of 1614 at the assembly of the Habsburg countries and lands in Linz, Melchior Khlesl, bishop of Vienna and head of the Habsburg government, proclaimed the Viennese court’s opinion regarding Bethlen’s election and Transylvania’s legal status. According to him, Bethlen’s rule in Transylvania was illegal, because he had become prince with the direct support of the Porte; he also used his title illegitimately, because Transylvania was a province of Hungary. Thus, he said, the country should be handed over to the king of Hungary and if this could not be achieved by peaceful means, he proposed to use arms.42 The letters of the Viennese government and the king’s councilors from this period clearly suggest that the Habsburg government simply did not acknowledge the fact that Transylvania was Istanbul’s vassal. His correspondence with György Thurzó, palatine of Hungary (1609– 1616) and thus the highest representative of the country’s estates, offers a closer look at Khlesl’s conceptions about Transylvania and Gábor Bethlen, and clarifies the differences he saw between the situation of the present ruler of the principality and the earlier “voievods.” The bishop made a

41 Gooss, Österreichische Staatsverträge, 420. 42 About the assembly in Linz, see Bálint Ila, “Az 1614-iki linzi egyetemes gyűlés” [The general assembly in Linz in 1614], in A gróf Klebelsberg Kuno Magyar Történetkutató Intézet Évkönyve, vol. 4, ed. Dávid Angyal (Budapest, 1934), 231–253; Benda, “Absolutismus”; Joachim Bahlcke, “Durch ‘Starke Konföderation wohl stabiliert’: Ständische Defension und politisches Denken in der habsburgischen Ländergruppe am anfang des 17. Jahrhunderts,” in Kontakte und Konflikte: Böhmen, Mähren und Österreich: Aspekte eines Jahrtausends gemeinsamer Geschichte, ed. Thomas Winkelbauer (Horn-Waishofen, 1993), 173–186. 84 teréz oborni clear distinction: the Ottomans had never tried to take Transylvania the way they wanted at that point, and they had never laid any claim over the parts of Hungary; and they had never interfered with the agreements concluded with the earlier voievods, “John, Christoph, Stephen, Sigismund, and Gabriel” in any way either—in marked contrast to the present situation.43 Khlesl suggested that if parts of Hungary were also to be ruled by the Turkish governor (that is, Bethlen), the Ottomans would achieve their goal, that is, to have Transylvania belong to them. He proposed to start negotiations with the Transylvanians, so that they would receive everything they wanted (and what they would otherwise get from the Turks) from the emperor—thus, they would acknowledge the rule of the emperor over themselves. If they could agree upon this with Thurzó, Khlesl concluded, they could, together with others, also convince Matthias II with good arguments, and this would lead to the removal of Bethlen, the recu- peration of Transylvania, and even the skillful avoidance of war.44 On 23 August 1614, Matthias II received the Transylvanian envoys at an audience and proclaimed that his greatest wish was to return the country and its neighbors to a peaceful condition.45 In his written resolution, he emphasized the same arguments that he used throughout the parleys: he asked that the Transylvanians take care that the Turks do not get involved in the negotiations, because that would cause great reluctance among the Hungarian estates. He declared that for him the greatest problem was that the Turks treated Transylvania as their own property and Bethlen behaved as if he were their governor. What is more, he complained, the Turks had asked him to abandon his rights to Transylvania while they treated the country as if this had already happened; and they asked him for the right

43 “Etenim ad hunc modum nunquam Turca Transylvaniam petiit: partes autem Hun- gariae neque ad hunc modum, neque ad alium unquam alias postulavit; compactatis quoque nos inter nec non Joannem, Christophorum, Stephanum, Sigismundum, Gabri- elem aliosque vajvodas erectis, nunquam ille sese immiscuit, ita ut in praesentiarum facit,” Khlesl to Palatine Thurzó, Linz, 16 August 1614; Joseph Hammer-Purgstall, Khlesl’s des Cardinals, Directors des geheimen Cabinetes Kaisers Mathias, Leben (Vienna, 1850), vol. 3, 110. The “voievods” mentioned by Khlesl—who confused the chronological order—are János (II) Zsigmond Szapolyai (1559–1571), István Báthory (1571–1576), Kristóf Báthory (who ruled as voievod (deputy) of his brother, István Báthory, between 1581 and 1583), Zsigmond Báthory (1588–1602), and Gábor Báthory (1608–1613). 44 “Si igitur in isto Illustrissima Dominatio Vestra et ego concordamus, nihilum dubito, quin id ipsum etiam aliis, medio solidarum rationum persuadeamus, Bethlehemum redu- camus, Transylvaniam retineamus, bellum dextre evitamus.” Ibid., 110. 45 EOE, vol. 7, 1614–1621 (Budapest, 1881), 154. notes on the legal status of the principality of transylvania 85 to send a pasha to the country when—and wherever—they wanted to, canceling the old treaties between Transylvania and Hungary.46 These are just two small examples from the huge bulk of documents that have been preserved from the negotiations between 1613 and 1615, but they are characteristic of the opinion of the king and his court about Transylvania: as if Gábor Bethlen’s seizure of power were the first time that the principality was threatened by coming under Ottoman sovereignty, and it had not been the case since more than a half century already. The conclusion to the crisis came with the the Treaty of Nagyszombat, signed in 1615, and confirmed two years later. By accepting it, Bethlen clearly yielded to the Viennese government. The agreements included a public and a secret part. In the public treaty, the king accepted that the inhabitants of Transylvania would elect their own ruler in the future.47 Accordingly, Transylvania and the adjacent parts of Hungary, together with their fortifications and castles, were never supposed to be alienated from the Crown of Hungary; what is more, the Transylvanians agreed to do everything in their power to maintain them in this position. Bethlen was allowed to rule the parts of Hungary as their lord (dominus) with the same privileges that his predecessors had been granted from the kings of Hun- gary. The king and his successors also obliged themselves to keep these articles and defend Transylvania and its elected ruler (electus) against any hostile menace; and maintain Transylvania’s old privileges and not oppose the country in any way. Bethlen and his legitimately elected heirs were obligated to help the king of Hungary and his successors against any enemy (with the exception of the Turks), and, if necessary, let the royal armies into the territories of Transylvania and the parts of Hungary.48 Two phenomena concerning the text of this part of the agreement are worth highlighting: Bethlen was mentioned in the treaty in two different ways—as electus, but also as illustrissimus dominus, that is someone who is the dominus of the part of Hungary belonging to Transylvania. His quality as a prince was left unnoted and Transylvania was in every case

46 “Turca ipse supervenit, qui Transylvaniam pro sua asserere, sibique vendicare niti- tur, ac partes Hungariae domino Gabrieli Bethlen tanquam gubernatori suo restitui atque universae Transylvaniae a Sacra Maiestate renunciari postulet (. . .) Turcae liberum permis- sum sit bassam ibi quandocunque libuerit, collocare.” Ibid., 155. 47 “(. . .) Transylvaniae status et ordines maneant in libera electione eorum domini et idem electus et provincia as observationem Tirnaviensium articulorum adstringatur.” Gooss, Österreichische Staatsverträge, 442. 48 The copy ratified by Bethlen is published, ibid., 440–447. 86 teréz oborni referred to as provincia. Bethlen himself called Transylvania a country (Regnum nostrum Transylvaniae) and himself a prince in the document of ratification—and it was this quality to which he obliged himself and his country to keep the stipulations of the treaty.49 A more detailed definition of the legal connection between Transylva- nia and the Kingdom of Hungary was provided in the secret points that completed the treaty; from our perspective, the first and twelfth points are the most important. The first proclaims that the principality should continue to enjoy the right of free election as it did, granted by the king, until Buda and Eger were free from Ottoman rule. After this, Transylvania and the parts of Hungary would return to the earlier position that they had held in the time of the old kings of Hungary, when they were governed by royal officials. Until then, however, the king permitted that the free princely position and rule (liber principatus et dominum) would remain intact for Bethlen and his elected successors.50 In the other article discussed here, the prince acknowledged Matthias II and his successors as the heads of the entire Christian world, kings of Hungary, his principals, and superiors; whereas he declared Transylvania and the parts of Hungary as an inseparable part of the Crown of Hungary and subject to the king. He also proclaimed that he would never interfere with the rights of the Crown.51 The treaty was complete with its public and secret parts, as they together regulated the relationship of Transylvania and its prince with the Kingdom of Hungary and its ruler, and resolved—albeit only temporarily—the severe political crisis after the election of Bethlen. The main goal of the public document was to disentangle the political–military

49 “(. . .) ad observationem eorum omnium etiam successores nostros legitimos Transyl- vaniae principes et regnum obligatos esse volumus in perpetuum.” Ibid., 447. 50 “Quantum ad statum Transylvaniae Partiumque ei annexarum attinet. In hac libera electione principatus quem nunc habent, per nos et successores nostros legitimos Hunga- riae reges conservabuntur, donec Dei beneficio Buda et Agria a potestate Turcarum (nisi antea voluerint) eliberabuntur. Qua eliberata Transsylvania cum partibus sibi annexis in pristinum statum, uti tempore antiquorum regum Hungariae per officiales regios guber- nata fuerat, redeat. Quos si autem nobis viventibus divino auxilio, Sua Maiestas Caesa- rea regiaque vel eiusdem successors consequi poterunt, quousque nos supervixerimus in libero nostro principatu Transylvaniae partiumque ei annexarum, permaneamus. Interim autem tam pro nobis, quam successoribus nostris legitime electis Transylvaniae principi- bus liber principatus et dominium salva et intacta permaneant.” Ibid., 449–450. 51 “Quod Sacratissimam Caesaream Regiamque Maiestatem eiusque legitimos successores pro capite totius Christianitatis et rege Hungariae, majoribus et superioribus suis agnoscant. Et Transylvaniam partesque ei subiectas pro inseparabili membro coronae Regni Hungariae recolunt et recognoscunt, neque iuri coronae praeiudicabunt.” Ibid., 452. notes on the legal status of the principality of transylvania 87 conflict, and thus its most important results were the acknowledgement of the territorial integrity of Transylvania, the restitution of the parts of Hungary, and the king’s approval of the free election of the prince. This latter point is all the more noteworthy, because it meant the continuation of a tradition: although this right was gained for the first time by John Sigismund from the Porte and by the grace of the sultan, the kings of Hungary confirmed it in their treaties with Transylvania. In the same vein, the Treaty of Nagyszombat also acknowledged this right, albeit with the condition that the king was supposed to rule the parts only as a dominus. The public points of the treaty not only failed to touch upon the relations of Transylvania to the Sublime Porte, but they also did not mention the overlord of the principality. Even so, the secret points of the treaty emphatically suggest that both parties were fully aware of the power of the mighty Ottoman Empire. The points of the secret treaty affirm the theoretical basis of the king’s approach during the negotiations, as well as in its final wording: namely, the legal fiction that Transylvania was still a part of the Crown of Hungary, therefore Bethlen could rule over the territory only with his approval. The temporary separation of the moment was acknowledged, but Tran- sylvania was supposed to be reincorporated when Buda and Eger were reconquered. The twelfth point of the secret treaty, summarized above, follows almost word by word a parallel article in the Treaty of Speyer. Although the parties renewed the 1615 treaty in 1617, in the relevant documents Bethlen was again addressed either by the title of dominus or by his name. This is a clear sign that Vienna did not recognize him as prince of Transylvania. I end the examples here, partly because the nature of the relationship between Vienna and Transylvania changed significantly when Beth- len joined the Thirty Years War in 1619. Also, with the above examples I only wish to demonstrate the rather complex nature of the relationship between Transylvania and Vienna and that of the Transylvanian estates’ right to freely elect their rulers. For six years, Bethlen bore the attacks and his own belittlement with endless patience, and attempted to assert his country’s status to the Habs burgs. When he finally joined the Thirty Years War in 1619 against the Habsburg forces, a new state of affairs developed. Bethlen’s military and diplomatic successes strengthened his position and that of Transylvania as a separate state, which the king of Hungary also had to accept. The Peace Treaty of Nikolsburg (1622) was thus conceived in an entirely different political situation—and also, in a very different manner. In this, 88 teréz oborni

Bethlen was called a prince who concluded the treaty as coequal to the emperor. No mention was made about whether and how Transylvania would belong to Hungary; the assent to free election was also missing; the parties concluded the peace as rulers of two sovereign countries, which nevertheless are neighbors with intertwined issues.52 This remained the state of affairs under the Rákóczi princes as well. We must also mention how the negotiations with the Viennese government were communicated to the Sublime Porte. On the whole, the Tran- sylvanians attempted to keep the negotiations secret, or when this was not possible, they tried to present the relationship as insignificant as they could, and prove their unswerving loyalty to the sultan. Transylvania’s resident envoys in Istanbul did outstanding service in this regard.53 Secrecy about Transylvanian-Habsburg negotiations was critical, as any attempt at transferring loyalty to the Habsburgs could and did provoke punitive Otto- man campaigns. Several times during the history of the principality—for instance in the above mentioned cases of 1571 and during Bethlen’s seizure of power; but also later, during the Ottoman campaign of 1660—Transyl- vanian politicians feared several times that the Porte planned to change Transylvania’s status to an Ottoman province or vilayet. * * * In conclusion, it can be argued that the princes of Transylvania, in their delicate position between the two empires, were forced to carry out a balanced policy. While Transylvania was an Ottoman vassal state, its princes repeatedly acknowledged that the principality existed as a province of the Kingdom of Hungary. Thus, the legal status of the principality was determined by its double dependency on its neighboring empires and its sovereignty was limited by both Vienna and Istanbul. Throughout the sixteenth and in the beginning of the seventeenth century, the kings of Hungary from the Habsburg dynasty regarded Transylvania as part and member of the Hungarian Crown and succeeded in maintaining their legal claim in this regard. Although the principality came under brief, direct Habsburg rule on two occasions (1551–1556

52 Ibid., 545–554; Gerald Volkmer, Das Fürstentum Siebenbürgen: Aussenpolitik und Völkerrechtliche Stellung (Heidelberg and Braşov, 2002), 139–142. 53 Vencel Bíró, Erdély követei a Portán [The envoys of Transylvania at the Porte] (Cluj, 1921). On the organizational questions concerning the embassy, see Gábor Kármán, “A 17th Century Odyssey in East Central Europe: A Biography of Jakab Harsányi Nagy” (PhD diss., Central European University, 2010), chapter 2. notes on the legal status of the principality of transylvania 89 and 1602–1604), these proved to be only temporary. Furthermore, in Hun- garian political thinking, the medieval and organic view of the state first described in the Tripartitum (1517) survived throughout the sixteenth and seventeenth centuries. According to this view, the theoretical unity of the members of the medieval Kingdom of Hungary (membra Regni) was maintained despite the tripartite division of the kingdom, and in spite of the fact that the weakened Hungarian Kingdom had only minimal influence over the Principality of Transylvania, which was a vassal state of the Ottoman Empire.

Janus-faced Sovereignty: The International Status of the Ragusan Republic in the Early Modern Period

Lovro Kunčević

Do we wish, o most esteemed gentlemen, to give account of our affairs to Ragusans? . . . To miserable persons, to a people for whom I do not know whether it is free or enslaved? To Ragusans which the Porte in its documents names and treats as subjects? Venetian patrician Bernardo Nani in 17531 This essay is an attempt to answer Nani’s dilemma regarding the liberty of Ragusans. More precisely, it is an attempt to investigate the consequences of being an Ottoman tributary state for the legal status and claims to independence of the early modern Ragusan republic. Over the last five hundred years answers to this vexing question have changed quite drastically, depending on who was asked and when. Nevertheless, they always boiled down to one of two elementary choices. In the legal sense, Ragusa was either characterized as a dependent polity under the supreme rule of the sultan or as an independent state paying the tribute only in order to ensure peace and commercial privileges in the empire. The goal here is to offer a new interpretation of the Ragusan legal status, one which surpasses this traditional dichotomy. More precisely, by rethinking the functioning of international law in the early modern period, I seek to demonstrate that the usual opposition between legal dependence and independence is anachronistic. In the first part of the essay I address the obligations and privileges of Ragusa according to Ottoman documents

1 “Volemio, volemio, eccellentissimi signori, render conto ai Ragusei dei nostri processi? (. . .) A miserabili persone, a un popolo, che non so, se sia libero o schiavo? A Ragusei, che la Porta nelle soe carte tratta e denomina per sudditi?” This fragment from Nani’s speech regarding a diplomatic dispute between Ragusa and Venice in the mid-eighteenth century is taken from Simeon Ljubić, “O odnošajih medju republikom mletačkom i dubrovačkom od početka XVI stoljeća do njihove propasti” [On the relations between the Venetian and Ragusan Republic from the beginning of the sixteenth century until their fall], Rad JAZU 54 (1880): 146. I must use this opportunity to warmly thank my colleague and friend Vesna Miović, who has helped me immensely in finding my way in the Ottoman maze. 92 lovro kunčević and the degree of self-governance that the city enjoyed as a consequence. In the second part I reconstruct the peculiar interpretation of tributary status promoted by the Ragusan government and contrast it with the very different Ottoman understanding. In the third part I consider the claims to “objectivity” of these two irreconcilable interpretations within their proper historical context—the interaction of two mutually alien legal cultures, Ottoman and Ragusan—which results in a somewhat heterodox interpretation of Ragusan legal status. The essay ends with several general reflections on the functioning of international law and inter-state treaties over the civilizational border between Islam and Christianity in the early modern period.

The Legal Status and Self-governance of the Ragusan Republic

The fundamental legal documents which codified the relationship between the Ottoman Empire and the Ragusan republic were the ‘ahdname (capitulations) confirmed and reissued to Ragusa by each new sultan. After long negotiations and two short-term arrangements (1430 and 1442), with the ‘ahdname of 1458 Ragusa became, definitively, an Otto- man tributary state and held that status until Napoleon conquered it in 1806.2 Despite a few minor later modifications, the basic obligations and privileges of the republic were arranged in 1458 and only reconfirmed by the sultans in subsequent centuries. According to the capitulations, the obligations of the city-state were relatively minor: it owed a somewhat vaguely defined “fidelity,” “truthfulness,” and “submission” to the sultan and had to pay an annual tribute (harac), stabilized at 12,500 gold coins in 1481. In return, the sultan promised to protect the city, guaranteed that his subjects would not harm Ragusans, and granted them extensive

2 The ‘ahdname of 1458, preserved only in a Slavic version, has been published in Bran- islav M. Nedeljković, “Dubrovačko-turski ugovor od 23. oktobra 1458. godine” [Ragusan- Ottoman treaty of 23 October 1458] Zbornik Filozofskog Fakulteta 9, no. 1 (1970): 390–393; Boško I. Bojović, Raguse (Dubrovnik) et l’Empire ottoman (1430–1520) (Paris, 1998), 21–30. Basic overviews of Ottoman-Ragusan relationship can be found in Ivan Božić, Dubrovnik i Turska u XIV i XV veku [Ragusa and Turkey in the fourteenth and fifteenth centuries] (Belgrade, 1952); Toma Popović, Turska i Dubrovnik u XVI veku [Turkey and Ragusa in the sixteenth century] (Belgrade, 1973); Vuk Vinaver, Dubrovnik i Turska u osamnaestom veku [Ragusa and Turkey in the eighteenth century] (Belgrade, 1960); Vesna Miović-Perić, Na razmeđu: osmansko-dubrovačka granica (1667.–1806.) [‘On the Frontier: Ottoman-Ragusan border 1667–1806] (Dubrovnik, 1997); Mladen Glavina, “An Overview of the Formation and Functioning of the Institute of Capitulations in the Ottoman Empire and the 1604 Dubrovnik Capitulation,” Prilozi za orientalnu filologiju 58 (2008): 139–166. status of the ragusan republic in the early modern period 93 trading privileges in the empire. In fact, by far the greatest part of any Ragusan ‘ahdname reads like a detailed trade agreement. It regulates the status of Ragusan merchants in the sultan’s lands, allows traders from any country—including those currently at war with the Porte—to come to the city, and, most importantly, it grants Ragusans the unprecedented customs rate of only two percent on all of their wares in the empire.3 As N. H. Biegman stresses, when the Ragusan ‘ahdname is compared with those of other tributary states, one is “struck by the amount of independence” it conferred upon the small republic.4 Interestingly enough, this independence was much more a consequence of what the capitulations did not say than what was explicitly stated in them. Namely, the Ragusan ‘ahdname failed to mention, and thereby regulate, a number of crucial issues strictly regulated in the cases of other tributary states. For example, there is not a word on the relationship of Ragusa with other countries, military cooperation with the empire, religion, the election of ruling officials or administration in general. All of these issues were left to the self-governance of the local patriciate, resulting in what Biegman characterizes as the “absolute autonomy” of the republic.5 As a matter of fact, what the ‘ahdnames did say explicitly was also not particularly obligatory or politically restrictive. Not only was the part dealing with politics short, but, even more importantly, it was formulated in a vocabulary much vaguer and less repressive than those in the capitulations of other tributary states. Thus, for example, Transylvanian ‘ahdnames explicitly stated that the province was a part of sultan’s “divinely

3 With the exception of Istanbul, where they paid five percent, and Adrianople and Bursa where they paid three. The Slavic versions of the fifteenth-century ‘ahdname were published in Ćiro Truhelka, Tursko-slovjenski spomenici dubrovačke arhive [Turkish-Slavic monuments in the Ragusan archives] (Sarajevo, 1911); Ljuba Stojanović, Stare srpske povelje i pisma [Old Serbian charters and letters], book 1, part 2 (Belgrade and Sremski Karlovci, 1934); Bojović, Raguse. For the texts and interpretation of capitulations from the later centuries, especially the sixteenth, see Nicolaas H. Biegman, The Turco-Ragusan Relationship According to the Firmāns of Murād III (1575–1595) Extant in the State Archives of Dubrovnik (The Hague and Paris, 1967); Vesna Miović, Dubrovačka Republika u spisima osmanskih sultana [The Ragusan republic in the documents of the Ottoman sultans] (Dubrovnik, 2005); Gliša Elezović, Turski spomenici [Turkish monuments], vol. 1, part 1 (Belgrade, 1940); Vesna Miović, Dubrovačka diplomacija u Istambulu [Ragusan diplomacy in Istanbul] (Zagreb and Dubrovnik, 2003), 15–20. On the ‘ahdname as a diplomatic form in general, see Hans P. A. Theunissen, “Ottoman–Venetian Diplomatics: The ‘Ahd-names: The Historical Background and the Development of a Category of Political-Commercial Instruments Together with an Annotated Edition of a Corpus of Relevant Documents,” Electronic Journal of Oriental Studies 1, no. 2 (1998): 185–190, 218–263, 306–309. 4 Biegman, The Turco-Ragusan Relationship, 51–52. 5 Ibid., 51–53. 94 lovro kunčević protected possessions,” that it was “subjected to our empire” or that the land was “given by our mercy” to the newly elected prince.6 Such explicit statements affirming the sultan’s supreme rule over the land or postu- lating him as the source of the ruler’s power never appear in Ragusan capitulations. The only formulations suggesting the sultan’s sovereignty were occasional mentions of the “fidelity,” “obedience” or “submission” of the Ragusan patriciate and statements that the city would be “under the mighty protection of my empire” or “in a state of safety and protection.”7 Of course, even these formulations could have far-reaching implications if interpreted in the tradition of Islamic law. However, the point is that those implications were never spelled out and written down in the capitulations. As a result, the legal position of Ragusa toward the empire always remained somewhat undefined—in a kind of conceptual sfumato—and this, as is shown below, made it an easy object of (mis)interpretation. It seems that the relative vagueness of the ‘ahdname and, even more, the broad privileges which Ragusa enjoyed, caused a certain amount of confusion in the Ottoman administration. Problems were most visible when it came to defining the legal status of Ragusan merchants in the empire. According to Islamic law, Ragusans were zimmis, non-Muslim subjects of the sultan: they were routinely mentioned as such, and also as re’aya (subjects), in official Ottoman documents.8 They enjoyed a number of privileges, however, which traditionally did not pertain to the zimmi population and which distinguished them sharply from the Christian subjects of the sultan. The most important was the customs rate guaranteed by the capitulations. According to Islamic law, zimmis had to pay a four percent customs’ tax on their wares, while Ragusans paid only two, even less than the Muslim subjects, who paid three percent. Other important privileges which distinguished the Ragusans from zimmis were their rights, repeatedly stated by fermans, to ride horses, wear Islamic clothing, and even carry arms in the empire, albeit only in “dangerous places.”9 Such a

6 Sándor Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden der Osmanen für Ungarn und Siebenbürgen (Vienna, 2003), 193, 221, 256. The English translations are mine. 7 These examples are taken from two typical ‘ahdnames, that of Selim I (1513) and Murat III (1575); Miović, Dubrovačka Republika, 13; Biegman, The Turco-Ragusan Relation- ship, 56–59. 8 For zimmi status in general, see Majid Khadduri, War and Peace in the Law of Islam (Clark, NJ, 2006), 175–201, especially 195–201. For mention of Ragusans as re’aya and zimmis, see Biegman, The Turco-Ragusan Relationship, 33–34. 9 See examples in Miović, Dubrovačka Republika, 169, 183. On the whole issue, see also Biegman, The Turco-Ragusan Relationship, 84–85. status of the ragusan republic in the early modern period 95 peculiar Ragusan status, as Biegman remarked, occasionally required a kind of “doublethink” on the part of the Ottoman administration: in theory Ragusans were zimmis, but, in actuality, due to their specific privileges, they were not. Speaking of these privileges, on some occasions Ottoman bureaucrats were even led to explicitly distinguish between Ragusans and the inhabitants of dar ül-İslam or to number Ragusans among foreign “Franks,” thereby implying that they were not under the rule of the sultan at all.10 Probably the most revealing example of the insecurities surrounding the status of Ragusans was the recurring problem with the customs rate they were supposed to pay. It seems that Ottoman officials could not decide whether Ragusans were zimmis or foreign merchants [!] and therefore whether they should pay a four or five percent customs’ duty on their wares. There are several sixteenth-century fermans in which the sultan addressed such problems and pointed out that the Ragusans paid a specific and unique percentage of their own, different from any other traditional legal category of the population.11 The difficulty of reconciling the status of the Ragusans with the traditional framework of Islamic law is also revealed by several seventeenth-century fermans in which the sultan stated quite a surprising thing: that Ragusans were not zimmis but, somewhat tautologically, “Ragusan people” or, even, “true Ragusan merchants.”12 Similar confusion was apparent when it came to defining the political status of Ragusan territory. While Ottoman documents sometimes spoke of the republic as an integral part of the Ottoman state, Biegman warns that “many fermans seem to distinguish between Dubrovnik and the Empire,” for example, when discussing Ragusan merchants coming to the “divinely- protected Possessions.”13 Of course, all of this does not mean that on the purely theoretical or doctrinal level there was any doubt on the Ottoman side concerning the status of Ragusa. Regardless of which legal school one followed, whether

10 Biegman, The Turco-Ragusan Relationship, 34, 48. 11 Of course, some of this was surely a deliberate means of extorting money—yet it was no less symptomatic since it was clearly believable that one could be unsure of Ragusan status. For examples of such cases, see Miović, Dubrovačka Republika, 147, 148, 149; Bieg- man, The Turco-Ragusan Relationship, 34; Popović, Dubrovnik, 79–82. 12 Mladen Glavina, 17. Yüzyıl Başında Osmanlı İmparatorluğu ile Dubrovnik Cumhuri- yeti İlişkileri [The relationship between the Ottoman Empire and Ragusan republic at the beginning of the seventeenth century], (MA thesis, University of Ankara, 2009), 117, 137. I am profoundly grateful to Mr. Glavina for having alerted me to this wonderful example. 13 Biegman, The Turco-Ragusan Relationship, 35. 96 lovro kunčević situating tributary states within the “House of Islam” or the “House of Treaty,” the basic result was the same. From the Ottoman perspective Ragusa was under the rule of the sultan, even in his absolute power.14 However, the specific mission of mediating with the Christian Mediterra- nean which the small republic performed made it necessary for the Porte to grant it extensive privileges that did not fit well with tributary status conceived in a traditional way. Simply put, mediation—primarily of trade goods, but also information and prisoners of war—required neutrality, but in order to make Ragusa a truly neutral state, the Ottomans had to renounce a number of traditional requirements of a tributary position, such as the contribution of military aid or adherence to the same foreign policy. In other words, the inconsistencies which appeared in the Otto- man documents in determining the legal status of Ragusa were largely the result of a discrepancy between the traditional legal norms and the unprecedented privileges which were granted to the small republic for pragmatic reasons. These privileges enabled Ragusa to enjoy remarkable self-governance, the extent of which becomes fully apparent only in comparison with other tributary states. As Géza Dávid points out, there were already significant differences between the autonomy of Transylvania and that of Walla- chia and Moldavia. The Transylvanian position was better in at least five aspects: its prince was elected by the local nobles and only confirmed by the sultan; it gave no hostages to the Porte; there were no Ottoman garrisons on its territory; its harac was relatively small compared to its economic strength; and, finally, it did not have the obligation to send annual food gifts to Istanbul.15 All of these points also held true for Ragusa, plus a few additional advantages. Unlike Transylvania, it was not obliged to provide military aid; there were absolutely no restraints on its foreign policy; and its ruler was not confirmed by the sultan (that being practically impossible since the rector changed every month). In fact, besides the somewhat abstract “obedience” and “submission,” the only permanent and concrete Ragusan obligation was to pay the harac. On top of this, the Porte would occasionally request other forms of assistance, such

14 For more details of the Ottoman legal doctrine regarding tribute-paying states, see the important work of Viorel Panaite, The Ottoman Law of War and Peace: The Ottoman Empire and Tribute Payers (Boulder, CO, 2000), passim, especially 77–84, 127–136, 461–466; Biegman, The Turco-Ragusan Relationship, 29–36. 15 Géza Dávid, “Administration in Ottoman Europe,” in Süleyman the Magnificent and His Age, ed. Metin Kunt and Christine Woodhead (London and New York, 1995), 71–90. status of the ragusan republic in the early modern period 97 as gathering information on Christian armies, transporting the sultan’s envoys to the West or sending a skilled labor force to neighboring sancaks. However, these requests did not significantly impede Ragusan self- governance. Not only were they quite sporadic, but, more importantly, whenever it was deemed necessary the Ragusan government—humbly yet unambiguously—refused them.16 All in all, the privileged legal position combined with the absence of concrete mechanisms of Ottoman control (e.g., garrisons and hostages) resulted in the Ragusan patriciate effectively exercising supreme political power in both domestic and foreign affairs. In other words, the councils of the aristocratic republic were at the top of the decision-making chain in all matters normally pertaining to the jurisdiction of an early modern government. For example, they independently elected and removed all state officials; created, changed, and enforced all law on Ragusan territory; sent out official diplomatic missions and negotiated with foreign powers, even those currently at war with the Sublime Porte. Although careful not to antagonize the Ottomans, for centuries Ragusan patricians ruled the city-state without interference from the Porte and without asking for its approval of their political decisions. This has led the majority of specialists in Ragusan history to characterize the republic’s position in the following way: while de iure Ragusa was an Ottoman “vassal,” de facto it was an independent state.17 Nonetheless, when speaking of the factual independence of Ragusa, two important caveats must be made. First, this independence was quite fragile. The small republic was clearly at the mercy of the overwhelmingly superior Ottoman Empire, which ruled the whole of its Balkan hinterland. Importantly, it was precisely Ragusa’s manifest weakness that probably induced the Ottomans to tolerate its independence. The Porte needed

16 On the various Ragusan services for the Ottomans, see Popović, Dubrovnik, 22–24, 54, 76, 83, 105–106, 138–139, 143; Nicolaas H. Biegman, “Ragusan Spying for the Ottoman Empire: Some 16th-Century Documents From the State Archive at Dubrovnik,” Belleten 26, no. 106 (1963): 237–255; Miović, Dubrovačka republika, 33–63. 17 For a few typical examples, see Robin Harris, Dubrovnik: A History (London, 2003), 82, 85, 425; Bernard Stulli, “Pregled državnopravne historije dubrovačke republike” [An overview of the history of the international status of the Ragusan republic], in Bernard Stulli, Studije iz povijesti Dubrovnika (Zagreb, 2001), 273–274; Ilija Mitić, Dubrovačka država u međunarodnoj zajednici (od 1358. do 1815) [The Ragusan republic within the international community: From 1358 to 1815] (Zagreb, 2004), 101–102; Božić, Dubrovnik i Turska, 340–344. A similar, albeit more reserved, interpretation has recently been offered by the influential S. Faroqhi, who writes about Ragusan “autonomy or even independence of a sort.” Suraiya Faroqhi, The Ottoman Empire and the World Around It (London and New York, 2007), 89. 98 lovro kunčević a neutral and therefore independent state as a mediator with Christian Europe and Ragusa was an ideal candidate for this delicate role. It had an extensive trade network and ample privileges all over the Mediterra- nean; at the same time it represented no military threat and was dependent upon the goodwill of the empire. The second caveat to note is that speaking of the factual independence of Ragusa does not mean that the republic could ignore the interests of the Sublime Porte. Quite the contrary, Ragusa went to great lengths to avoid provoking the neighboring superpower, especially in its foreign policy. However, a somewhat narrow space for maneuver due to considerations of the great powers was a common problem of all the small states of the epoch—other examples being Lucca and Genoa, whose independence is normally not questioned. The crucial point is the absence of coercion; despite giving constant consideration to the Ottoman view, political decisions were made in the Ragusan republic for centuries without any interference and even without asking the approval of the Sublime Porte.

The Ragusan Interpretation of Its Tributary Status

As noted, most specialists in Ragusan history would agree with the claim that the republic was factually independent, but in the legal sense a “vassal” state under the supreme rule of the sultan. Indeed, Ragusa paid the harac and its status was regulated by ‘ahdnames which, despite their comparatively non-obligatory vocabulary, nevertheless clearly referred to “submission” and “obedience” owed to the sultan. Although based on convincing reasons, this simple characterization of the republic’s position nevertheless disregards, even distorts, one important aspect of the Ottoman-Ragusan relationship. It creates a false impression that the “vassal” position of Ragusa was an undisputed and generally accepted truth, when in fact it was a highly contested issue. This becomes clear when one examines the ways in which it was interpreted by contemporaries, especially by one group whose voice should not be ignored—the Ragusans themselves. As this section shows, Ragusans drastically reinterpreted the meaning of tributary status and claimed something very different from modern scholars. Simply put, they emphatically insisted that their republic was a fully independent state. Before reconstructing the peculiar Ragusan interpretation of tributary status one important question must addressed: What made it possible for Ragusans to interpret their relationship with the Sublime Porte status of the ragusan republic in the early modern period 99 so differently from the Ottomans and so far from the original meaning of the capitulations? The most obvious reason was the aforementioned brevity and vagueness of Ottoman documents when it came to defining the legal status of Ragusa. The republic’s position remained in a conceptual sfumato since its ‘ahdnames addressed the political issues very laconically, while its fermans were characterized by frequent inconsistencies, even contradictions. However, an even more important reason that enabled Ragusans to reinterpret their tributary position was the fact that the relationship between the Ottomans and the republic was established across a civilizational border, between two sides that had totally different legal and political cultures. This deep conceptual chasm was the main area of Ragusan manipulation. Since the two partners spoke different political and legal languages, Ragusans were able to ensure that many embarrass- ing meanings conveniently got “lost in translation.” Although there are numerous examples of such ideological maneuvers later in the text—especially regarding tribute—a few illustrations at this point will add clarity. Telling examples are the politically compromising concepts usually mentioned in Ottoman documents, such as “fidelity” (sadakat), “servitude” (ubudiyyet), “obedience” (itaat), and “protection” (emn ve eman or himayet).18 These concepts, which in Islamic law had clear connotations of submission, were far less repressive when interpreted in the context of Western political traditions, especially in humble Ragusan diplomatic rhetoric. Namely, in every letter directed to an important Christian ruler the Ragusan government insisted on its “fidelity” (fedeltà), “obedience” (obedienza), and “servitude” (servitù), and also invoked his “protection” (prottetione). When the Ottoman documents were read against this background most of their political repressiveness disappeared, since the compromising concepts appeared as nothing but empty and polite phrases, typical of Ragusan diplomatic rhetoric. Even more subversively, interpreted from this perspective the Ottoman-Ragusan relationship did not seem substantially different from the relationship of Ragusa with any other powerful state, since they were described in strikingly similar terms.19

18 These concepts were normally used by the Ottoman chancellery to define the relationship of tributary princes to the Porte (Panaite, The Ottoman Law, 197). They also sometimes appear in the ‘ahdnames issued to foreign principalities such as Venice or Poland due to the Ottoman legal fiction of the sultan’s universal rule. 19 When it came to such “convenient” translations the most problematic concept was clearly that of “subjection/submission,” which also appears occasionally in Ottoman 100 lovro kunčević

There is no doubt that the Ragusan government was aware of the different meanings of these key concepts in its own usage and in that of the Ottomans. However, it cultivated this ambivalence by persistently using the same vocabulary toward the Porte as it did toward any other foreign power, thus preserving the cherished independence of the republic. The Ottomans, on their side, tolerated Ragusan (mis)interpretations, since, as noted, the city’s role as a mediator required it to be seen as an independent state by the Christian international community. Supported by strong pragmatic reasons, such a chronic “misunderstanding” between Ragusa and the Ottomans lasted for centuries. The basic mechanism was as simple as it was effective: by re-contextualizing the repressive concepts, translating them from one political vocabulary to another, Ragusans turned what were originally assertions of Ottoman sovereignty into mere diplomatic phrases. All of this made it possible for the Ragusans to proclaim their independence in quite unambiguous terms, at least to Christian princes. Several far-reaching claims characteristic of sixteenth- and seventeenth-century Ragusan diplomacy toward Western states include the following: the republic is an “absolute prince” (principe assoluto) with supreme jurisdiction over its subjects and territory; God is the direct and only source of its “liberty” (libertas/libertà); this “liberty” has unquestionable historical foundations since Ragusa had (allegedly) never acknowledged the rule of a foreign lord[!].20 Of course, such forceful proclamations of independence made explaining the republic’s tributary status even more urgent. However, when it came to justifying their close ties with the “infidel” to Christian Europe, the crucial issue was not the (in)dependence of Ragusa, but its Christian loyalties. The persistent claim that the annual tribute to the Ottomans did not mean subjection, but was simply the price of non-aggression and enormous trading privileges, was normally not challenged. What was documents. Yet even here there are examples of it being used as a non-obligatory phrase of flattery toward Western rulers. For example, in 1570, the Ragusan ambassador stated to the pope that his city had never wavered in its fidelity toward the Holy See, but “was always most obedient and subjected to it” (“anzi che sempre g’è stata obedientissima et soggetta”) (Državni Arhiv u Dubrovniku (State Archives of Dubrovnik, henceforth DAD), Lettere di Ponente 1 (1566–1570), fol. 224r). 20 For more on the ideology of Ragusan liberty see Lovro Kunčević, “Discourses on liberty in early modern Ragusa,” in Freedom and the construction of Europe, vol. I, Quentin Skinner and Martin van Gelderen, eds. (Cambridge, 2013), 195–214; and “O dubrovačkoj libertas u kasnom srednjem vijeku” [On Ragusan libertas in the Late Middle Ages], Anali zavoda za povijesne znanosti HAZU u Dubrovniku 46 (2008): 9–64. status of the ragusan republic in the early modern period 101 challenged was the morality of the Ragusan policies toward the Ottomans. It was known that the city-state paid substantial tribute and lived in a highly profitable peace with the “infidel”, occasionally even assisting him against fellow Christians (albeit this could rarely be proven). These facts were used extensively by the enemies of the city-state, primarily the Vene- tians, who repeatedly accused the Ragusans of being “false Christians,” selfish profiteers who gained immense profits by helping the “Turk” in his assault on Respublica Christiana. In order to respond to these accusations, Ragusan diplomats began to represent the republic’s tributary status in quite a surprising way—as a great favor to Christianity. The basic claim was that the close ties with the “infidel” enabled Ragusa to perform the crucial task of protecting the remaining Christian communities under Ottoman rule. Ragusan diplomats stressed the unique privileges of their city which enabled it to maintain churches in “infidel” lands, send mis- sionaries there, publicly celebrate the Christian religion, and in general provide legal protection to the Christian populations in the Ottoman Bal- kans. All in all, when it came to explaining its tributary status to Western Europe, the central issue was not whether Ragusa was an independent state or not, but whether it was a traitor or a valiant defender of the Chris- tian republic.21 If Ragusa could represent itself as a fully independent state to Western audiences with relative ease due to their vague knowledge of Ottoman politics, a completely different ideological and diplomatic challenge was to legitimize this claim before the Sublime Porte. As a matter of fact, most of the time the goal of the Ragusan government was far more modest—to simply avoid tackling the topic at all. The basic strategy of Ragusan diplomacy in Istanbul was to avoid opening general questions about the city’s status, and instead obfuscate the whole issue with non-obligatory humble phrases. For example, in endless letters and diplomatic instructions the Ragusans represented themselves as “the oldest and most devoted harac-payers” or “the most faithful tributaries and servants of the Gate of Happiness,” who for centuries had “served the Ottoman dynasty” praying “God to grant long life and victories to his Highness.” Although this rhetoric

21 This Ragusan rhetoric was in fact a peculiar adaptation of the widespread European topos of antemurale/propugnaculum Christianitatis. On the accusations against Ragusa and the responses of its apologists, see Lovro Kunčević, “Retorika granice kršćanstva u diplomaciji renesansnog Dubrovnika” [Rhetoric of the frontier of Christendom in the diplomacy of Renaissance Ragusa], Anali zavoda za povijesne znanosti HAZU u Dubrovniku 48 (2010): 179–211. 102 lovro kunčević was characterized by an extremely humble tone, on closer inspection it becomes clear that it relied heavily on the tendentious re-contextualization mentioned above. It contained nothing really prejudicial to Ragusa’s liberty, at least from the Western perspective. Ragusan diplomats never went beyond general phrases about “fidelity” and “service” to the sultan, clearly avoiding more compromising words such as “submission.” Simi- larly, the relationship between the sultan and the city-state was usually characterized by the term “protection” (protettione), which had comparatively neutral connotations in the Western tradition and was regularly applied to important Christian rulers as well. Finally, the Ragusans were normally described by relatively non-obligatory words such as “tributaries” or “servants,” while the words “vassals” or “subjects” were carefully avoided. Admittedly, there were a few instances during the early modern period when Ragusans flirted with the claim that they were the subjects of the sultan. Typically for Ragusan diplomacy, however, such statements were strongly ambivalent and could be understood in at least two ways. A telling example comes from 1520, when the senate instructed its ambassadors to Istanbul to complain about the Catalan consul in Alexandria, who claimed jurisdiction over Ragusan merchants. The ambassadors were supposed to ask for a berat which would state that Ragusan merchants should be handled by the local kadi and should be treated “as his [the sultan’s] vassals since we are tributaries and his faithful servants, paying the customs of these lands as vassals.”22 A hundred years later, in 1621, the Ragusan government again tried to procure privileged treatment by blurring the distinction between its own subjects and those of the Porte. The ambassadors to Istanbul were supposed to complain to the grand vizier about a certain tax, “which the Venetians do not request from the subjects of the Grand Signore, but leave them free with their wares; according to the same reasoning, we should be treated and favored in the same way, as the ancient and most faithful of his tributaries.”23 Although definitely accentuating their ties with the empire, strictly speaking what the Ragusan government claimed in these statements was not that Ragusans were the

22 The sentence is even more ambivalent in Italian: “como soi vassalli attento che siamo charazari, et soy fideli servitori pagando li dretti de detti paesi como vassalli” (DAD, Lettere di Levante 19 (1504–1526), fol. 175r). 23 “(. . .) che li Signori Venetiani non pretendino questo datio delli suditi del Gran Signore, ma li lassino liberi con loro mercantie, cosi per conseguenza con istessa ragione debbiamo esser trattati e fauoriti noi come antichi e fedelissimi suoi tributari” (DAD, Let- tere di Levante 43 (1615–1622), fol. 240v). status of the ragusan republic in the early modern period 103 subjects of the sultan. In both instances it insisted only that they should be treated as the sultan’s subjects or vassals since they were his “tributaries.” There is, clearly, a great difference between claiming one is a subject and claiming one should be treated as a subject.24 To the best of my knowledge, during the sixteenth and seventeenth centuries there was only one situation in which Ragusans went beyond the carefully guarded boundaries of their diplomatic rhetoric and clearly admitted Ottoman sovereignty. It was during the immensely traumatic crisis of Lastovo, 1602 to 1606, when this important island rebelled against the republic, acknowledged Venetian sovereignty and was occupied for several years by the Serenissima. This led to a genuine diplomatic war between the two Adriatic republics in which Ragusa demanded immediate and unconditional restitution of its possession and mobilized Euro- pean courts in order to put pressure on the Venetians. In such a situation, which seemed to many merely the first act of the Venetian occupation of the whole republic, the support of the Porte was desperately needed. Consequently, Ragusan diplomats in Istanbul went much further than usual. They insisted that Venetians were trying to “take us away, with our obedience and tribute, from the Grand Signore and to submit us to them.” They proclaimed that the patricians would sooner “lose their state, sons and lives than submit to another empire than that of the Grand Signore” or “accept another patron than our legitimate one, the sultan Ahmed, the invincible emperor.”25 Finally, they even admitted that their own

24 Ambivalent statements such as those quoted above were the basis of a claim, sometimes also found in the historiography, that the Ragusans themselves occasionally stated that they were the subjects of the sultan (for example, Vinko Foretić, Povijest Dubrovnika do 1808 [History of Ragusa until 1808], vol. 1 (Zagreb, 1980), 211–212; Miović, Dubrovačka diplomacija, 17). However, besides one exceptional example analyzed below, during the sixteenth and seventeenth centuries there is no documentary evidence for this. There are, of course, examples of Ottomans describing Ragusans in that way, while on the Ragu- san side there is only the aforementioned ambivalent rhetoric. On the other hand, in the eighteenth and early nineteenth centuries Ragusans seem to have been less circumspect and indeed stated on several occasions that they were subjects of the sultan. For such late examples, see Ljubić, “O odnošajih,” 144–146; Lujo Vojnović, Pad Dubrovnika [The fall of Ragusa] (Zagreb, 1908), book 2, 54–55. 25 These quotations are all from a codex containing copies of letters from Ragusan diplomats in Istanbul to their government during the critical period of 1604–1606 (DAD, Arhiv Bassegli, box 7, B 3/1 Corrispondenza delli Signori Senatori di Ragusa Marco Bassegli, Jacomo Bobali Sorgo, Valentino Giorgi, Proculi Relativa a una loro ambasciata a Constan- tinopoli dal 21 Giugno 1604 fino al Luglio 1606). The quotations translated above are from letters from late 1604 and early 1605. Their originals are as follows: fol. 48r: “sotraerci, et con l’obedienza, e con il tributo, da G. S-re e someterci a loro”; fol. 19r: “noi prima pat- iremo di perder il stato e fioli e vite istesse che sogiacere ad altro imperio che a questo 104 lovro kunčević subjects owed allegiance to the sultan, claiming that the Venetians in fact demanded “that we lift ourselves from the protection of the Grand Signore and submit ourselves to them, and that our subjects begin to recognize only them as their superiors and no longer His majesty nor our lords.”26 In all likelihood, these exceptional statements were the most explicit acknowledgments of fealty that the Porte received from Ragusa during the sixteenth and seventeenth centuries. Yet even they remained somewhat abstract, since what was acknowledged was only the sovereignty of the sultan in the most general sense, with in fact quite a few caveats. Namely, even in this dramatic situation Ragusan diplomats were expected to remain circumspect regarding the vocabulary used to characterize the city’s relationship to the sultan. This is nicely illustrated by a letter from 1605, in which the senate warned the ambassadors that the word slaves (schiavi) with which they were addressed by the Grand Vizier should be “avoided at any cost” and that the senators were “greatly offended and surprised since one could say tributaries (tributari) or, at most, servants (servi).”27 Moreover, even in this period of strikingly submissive rhetoric in diplomats’ conversations with Ottoman dignitaries there emerged glimpses of the typical subversive interpretation of Ragusa’s tributary relationship, which is extensively analyzed below. For instance, the ambassadors pointed out that the tribute was paid “in order for our liberty to be preserved and conserved,” stressed that it was Ragusans themselves who “of our own purest will submitted with the tribute,” and even stated “our signori are deemed free by God [!] and maintained by the Grand Signore.”28

felicissimo del G. S-re”; fol. 21v: “mai ne acceteremo altro padrone che il nostro legitimo Sultan Achmet Invictissimus Imperator.” 26 “. . . la domanda loro e non in altro che noi ci leviamo dalla protetione dal G. S-re e si sotometiamo a essi, e che i nostri suditi riconosiciano loro soli per superiori e patroni e non piu ne S. M. ne i nostri S-ri.” (Ibid., f. 103v, the letter of 9 December 1605). 27 The senate wrote: “Ci pare avvertirvi che quella parola schiavi, o sia robovi, con la quale veniamo esser chiamati, schifiate omninamente, et in spetial nelle scritture, come voi Cerva [one of the ambassadors] sapete benissimo, che si é fatto sempre mai, perche replicandola tante volte in quella nostra suplica presentata al Sig-r Bassa, c’ha offeso molto, e na’ha recato gran maraviglia, potendo dire tributarij, et ad piu servi.” (DAD, Lettere di Levante 41 (1604–1608), fol. 107r–107v, letter of 30 December 1605). 28 “(. . .) voglino havere di censurare le acioni di Nostri Sri iudicati liberi dal Dio, e mantenuti dal G. S-re perche mentre le acioni loro fosero soggete alla corecione di Vene- ziani non sariano liberi ma vasali di essi Venetiani in modo che vanamente saria pagato il tributo per eser conservata e difesa la liberta nostra” (DAD, Arhiv Bassegli, box 7, B 3/1 Corrispondenza delli Signori Senatori di Ragusa, fol. 64v, letter of 23 April 1605; “. . . al hora status of the ragusan republic in the early modern period 105

However, besides the exceptional rhetoric of the Lastovo crisis and the vague rhetoric usually applied by Ragusan diplomats, there was yet a third way of speaking of tributary status in front of the Ottomans, one which is the most relevant for this study. Namely, Ragusan diplomats occasionally abandoned the customary polite phrases and explicated an interpretation of the tributary relationship which must have seemed highly subversive from the Ottoman perspective. As is apparent from the examples below, such breaches of the usually cherished conceptual sfumato happened in situations of profound crisis—in fact, when it was estimated that contra- dicting the Ottomans was better than letting them act based on their own understanding of the city’s status. A good summary of this subversive interpretation of tributary status can be found in an angry letter of the senate to its ambassadors in Istan- bul in 1677. The senate reacted to a major blunder of its diplomats at the Porte. Faced with Ottoman demands for an enormous sum of money, the ambassadors composed and distributed a memorandum expound- ing the Ragusan defense. The problem was that in this text they not only interpreted Ragusan tributary status in a way unacceptable to the senate, but also used some of the usually carefully avoided words to describe it. After remarking that the memorandum is “incredibly prejudicial to the honor and liberty of this republic,” the senate explained its grievances: At the very beginning of your memorandum you assert that long ago we submitted spontaneously to Sultan Orhan, who gave us and left us the free possession of our city. This assertion is more than prejudicial to our liberty because we never submitted to the Grand Signore, but simply recommended ourselves to his protection with the offer of an annual tribute [“raccoman- datisi alla sua prottetione con offerta d’un annuo tributo”]. Equally, the abovementioned Sultan Orhan did not give us the free possession of our city, whose liberty was left to us by our ancestors and not given to us by any prince whatsoever . . . In the next chapter you state that all of this [the treaty with Orhan] was confirmed by all of his successors. They, however, have never sought more nor confirmed to us anything but the privileges of the ‘ahdname in which there is not a word about us being subject to the sultan, nor of him leaving us free our city. These words presume that the city was in his possession, which is not true. It is very different to say—as the ‘ahdname does say—that we can freely govern and to say that he left us free our city . . .

che ci sotometimo di pura volonta nostra col tributo al G. S-re” (Ibid., fol. 54v, letter of 2 March 1605). 106 lovro kunčević

Finally, you conclude the document by saying that we are the most sincere tributaries and subjects. We are still amazed at you having inserted this word subjects (sudditi), a word so inimical to our liberty which we enjoy by the mercy of God before the whole world. There is a great difference between being a subject and being a tributary. Many and even great princes are tributaries of other princes, but most definitely not their subjects.29 Clearly, this is an interpretation of tributary status drastically different from that of the Ottomans. The fundamental difference was the unambiguous claim that Ragusa was not under the rule of the sultan. As the senate insisted, Ragusans “never submitted to Grand Signore,” they were not his “subjects,” the republic’s “liberty” was inherited from their ancestors and “not given to us by any prince whatsoever.” The senate based such claims on an important historical myth about Sultan Orhan, a narrative which represented the beginnings of tributary relationship in a way wonderfully suited to the republic’s interests. For centuries, the city’s diplomats at the Porte claimed that the ancestors of the patricians, due to their political prudence or even divine revelation, understood the future grandeur of the empire as early as the fourteenth century, when it was still a mediocre power in Asia Minor. In order to ensure the sultan’s defense from their enemies and a bright future for the republic in general, they sent an embassy to Orhan and arranged for his “protection” in return for an annual tribute. The ideological agenda behind this story is clear: the Ragusans entered the tributary relationship of their own will and freely negotiated the privileges and obligations it entailed. Not only did they themselves ask for Ottoman protection, but, even more importantly, they did so without any coercion since Orhan was a distant ruler who represented no threat whatsoever.30 Indeed, speaking of the beginnings of the tributary relationship in the 1677 letter, the senate obviously picked its words carefully, stating that

29 Jovan Radonić, ed., Dubrovačka akta i povelje [Ragusan acts and charters], III/2 (Belgrade, 1939), 890–891. 30 Some of the examples of this story are DAD, Lettere di Levante 37 (1590–1592), fol. 9r; Ante Liepopili, “Dopisi Marojice Kaboge vladi dubrovačkoj” [Dispatches of Marojica Kaboga to the Ragusan government] Glasnik dubrovačkog učenog društva “Sveti Vlaho” 1 (1929): 127. According to another, less frequent, version of the story, Ragusan ambassadors originally brought only “gifts” and tribute was a later addition (see, for instance, the instruction to the ambassadors to Selim II in 1566: Radonić, Dubrovačka akta i povelje, II/2, 141). This story also appears in Ragusan historiography, descriptions of the city, and even literature (Natko Nodilo, ed., Chronica Ragusina Junii Restii item Joannis Gundulae (Zagreb, 1893), 153; Fran- cesco M. Appendini, Notizie istorico-critiche sulla antichità storia e letteratura de’ Ragusei, vol. 1 (Ragusa, 1802), 295; Simeon Ljubić, ed., Commissiones et relationes venetae. tomus 3 (annorum 1553–1571) (Zagreb, 1880), 77; Jaketa Palmotić, Dubrovnik ponovljen i Didone [Ragusa restored and Didone], ed. Stijepo Skurla (Ragusa, 1874–1875), 60, 250). status of the ragusan republic in the early modern period 107

Ragusans had simply “recommended themselves” to the sultan, offering an annual tribute. The delicate question of the tribute’s meaning was in turn solved by stressing one distinction which was crucial for the Ragusan political tradition—a “great difference” between the status of “subjects” and that of “tributaries.” This fundamental distinction rested on a blatant example of what has been labeled “tendentious re-contextualization”: that is, on the maxi- mally benevolent Ragusan interpretation of the meaning of the tribute paid to the Ottomans. While in Islamic law the payment of harac implied acknowledging the supreme rule of its recipient, in the Western political tradition, the giving of tributum/tributo does not necessarily have that con- notation. The tribute could have been paid for possession of certain territories, as the price of economic privileges, military alliance or peace itself, none of which conferred inferior legal status on the one paying (albeit it usually hurt his prestige). Not surprisingly, it was this latter understanding that Ragusans adopted when interpreting the meaning of the tribute they paid to the Ottomans, thereby denying the sultan’s sovereignty over their republic.31 Thus, in their blunter moments Ragusan diplomats at the Ottoman court claimed that the tribute was paid exclusively for military protection by the Porte. This understanding emerged clearly in 1605, during the Lastovo crisis when the submissive rhetoric was temporarily replaced by the subversive discourse. After one vizier told the Ragusans they should wait a bit for the sultan’s protection against the Venetians, the ambassador responded with uncharacteristic bluntness: If you want us to wait in this ruin and dissolution of our state, let his High- ness of Grand Signore also wait for the tribute from our signori, which we pay not because he gave us or sold us our city and its territory, nor because he conquered us with arms, but exclusively in order that he defends us with his invincible hand in situations like this.32

31 For the connotations of harac in Islamic law and the possibility that the Christian side interpreted it differently, see Panaite, The Ottoman Law, 205–207; Biegman, The Turco-Ragusan Relationship, 32. Ragusans themselves traditionally paid a number of such “harmless” tributes for possession of parts of their territory and various privileges abroad. Therefore, it was only natural that they attempted to represent harac as something similar. For other Ragusan tribute payments, see Mihajlo Dinić, “Dubrovački tributi: Mogoriš, Svetodmitarski i Konavoski dohodak, Provižun braće Vlatkovića” [Ragusan tribute: The Mogoriš, the tribute for St. Demetrius and Konavle, the Provižun of brothers Vlatković], Glas SAN 168 (1935): 203–257. 32 “Se la vole che noi aspetiamo in questa rovina e disolutione del stato, aspeti ancora la M. del G. S-re il tributo da nri SS-ri: pagatole non perche c’habbia dato o venduto la Cità o il territorio ne sogiogatoci a quelo con l’arme, ma solamente perche siamo difesi dal 108 lovro kunčević

Far from appearing as the sovereign of Ragusa, the Ottoman Empire here seems like a gigantic mercenary hired to protect the small republic. Yet explicating the specific purpose of tribute was not the most provocative thing Ragusan diplomats did in this passage. Even more subversive was their accentuation of its conditional nature: if the sultan did not fulfill his duty of protection, Ragusans did not have to fulfill their duty of tribute. Such an interpretation of the Ottoman-Ragusan relationship, insisting on strict reciprocity and mutual obligations, is clearly incompatible with Islamic law, according to which tributary status was a result of the unilateral act of the sultan, a privilege granted by his mercy. If Ragusans could legitimately refuse to give harac if the sultan did not protect them, then their status was based on a mutually binding agreement between the two sides—upon a contract. This is yet another example of tendentious re-contextualization, of how the Ragusans used the gap between legal cultures in order to soften compromising concepts via a convenient “translation”: if for the Ottomans the ‘ahdname was a privilege, for the Ragusans it was a contract (patto). That Ragusans interpreted tributary status in the spirit of the Western contractual tradition is confirmed by the 1590 instruction to ambassadors in Istanbul in which the senate touched upon a delicate question: What is the final legitimate response if the sultan fails to fulfill his obligations? The circumstances provoked such radical thinking, since the diplomats had to resist the initiative of one courtly clique to turn a part of Ragusan territory into a sancak. Claiming that such a move was illegal, they were to warn the grand vizier that alienating part of republic’s territory would mean that the sultan was breaking his vow in the ‘ahdname, in which he guaranteed the unhindered possession of the whole Ragusan district. Then they were supposed to make a point which, from the Ottoman perspective, must have amounted to treason: it is not a task nor a custom of great rulers to abuse their own laws and disregard the contracts (patti) which were made and confirmed with the most solemn oaths; because, if they were to do so they would first offend God, and their own souls, and also they would give legitimate cause to their allies [conferederati] to retreat from their protection . . .33

invitissimo suo bracio nelle simile ocorentie.” (DAD, Arhiv Bassegli, box 7, B 3/1 Corrispon- denza delli Signori Senatori di Ragusa, fol. 103r; letter of 9 December 1605). 33 “(. . .) non essendo uffitio nè costume delli Gran Signori preuaricare le leggi loro, et disprezzar i patti fatti et confermati con sollennissimi giuramenti perche quando facessero contra li patti, et giuramenti prima offenderebbeno Iddio, et l’anime loro, e poi darebbeno status of the ragusan republic in the early modern period 109

Although formulated as a general point, the message was clear. If the sultan broke his oath and forfeited his obligation of protection, the Ragusans— his “allies”[!]—could legitimately withdraw from their tributary relationship to the Sublime Porte. The tributary status of the city is here quite unambiguously represented not as a result of the sultan’s privilege, but of a contract between Ragusa and the empire. Consequently, it was a reciprocal relationship that could be revoked by either side if the other did not respect its contractual obligation. The idea that Ragusa was an independent state enjoying contractu- ally defined protection by the sultan naturally led to other propositions which were at odds with the Ottoman understanding of the city’s status. For instance, it made Ragusans refute the idea that their territory in any way belonged to the sultan. Thus, the two ambassadors, writing to the senate in 1677, stated: “if we are asked [by the Ottomans] to whom our city belongs, we will not say it belongs to the sultan, but to the Ragusan patriciate under the protection of His Highness.”34 Even clearer was the Ragusan diplomat who, resisting the above mentioned Ottoman attempt to turn a part of the Ragusan district into a sancak, in 1590 stated that the republic’s territory was inherited from “our predecessors” and concluded quite bluntly: “we [Ragusans] do not possess a single palm of soil which belongs to the Grand Signore.”35 Along the same lines, the patricians never subscribed to the idea that their power was in any way derived from the sultan. The claim that power was enjoyed “by God’s will and the sultan’s decision” and other similar statements which appear in this period in the Wallachian and Moldavian context were absolutely inconceivable in the Ragusan political discourse. When speaking of their “liberty” Ragusans always stated that it was granted only by God and inherited from their ancestors. Even during the Lastovo crisis the most they agreed to was the somewhat unclear claim mentioned above that while “deemed” free by God, they were only “maintained” by the sultan.36

legitima causa alli loro confederati di ritirarsi dalla loro protettione (. . .)” DAD, Lettere di Levante 37 (1590–1592), fol. 89r; letter to the ambassadors Michael Gozze and Lucian Caboga, 12 July 1590. 34 Liepopili, “Dopisi Marojice Kaboge,” 127. 35 “(. . .) noi non possediamo pur un palmo di terreno, che sia del Gran Signore . . .” DAD, Lettere di Levante 37 (1590–1592), fol. 34v; instruction to the ambassadors Andrea Bonda and Paolo Stefano Gozze, 13 June 1590. A similar example is in ibid., fol. 43v–44r. 36 For several examples of the Wallachian and Moldavian understanding, see Panaite, The Ottoman Law, 163, 346, 468. For more detail on the drastically different Ragusan viewpoint: Kunčević, “O dubrovačkoj libertas,” 58–60. 110 lovro kunčević

To summarize, in the rare instances when the usual vague rhetoric did not suffice, Ragusan diplomats at the Ottoman court presented a more precise but also far more subversive interpretation of republic’s tributary status. Ragusa had never “submitted” to the sultan, but “recommended” itself of its own free will to his protection and therefore Ragusans were not Ottoman “subjects” or “vassals.” The harac the city paid was not an acknowledgment of Ottoman sovereignty but merely a price for (military) protection. Neither the possession of Ragusan territory nor the power of the ruling patriciate derived from the sultan. Finally, if the sultan did not fulfill his obligations, the republic could legitimately withdraw from the tributary relationship. In sum, the Ragusans represented their relationship with the Porte as a conditional and revocable contract between two partners who were surely unequal in factual power and prestige, but were both—on the bottom line—independent states. All of this begs the question: How could the Ragusans afford such discourse in front of Ottoman dignitaries? Why was it tolerated? The answer is simply to put things in context. First, one should not forget that these provocative statements are fairly atypical, being hand-picked from the usual Ragusan diplomatic discourse in Istanbul which was servile and non-provocative—although always carefully vague regarding the exact nature of the Ottoman-Ragusan relationship. Moreover, these statements themselves were usually couched in phrases about undying loyalty and were frequently pronounced “with tears in the eyes and a crying voice,” as the diplomatic instructions put it. Second, one should not forget that such statements were regularly followed by a bribe, which surely made them more acceptable. Finally, as has been stressed repeatedly, one should keep in mind the pragmatic Ottoman attitude toward the small republic, the fact that the Porte tolerated Ragusan trivializations of tributary status because they were seen as necessary in their role as mediator with the Christian Mediterranean.

The Multiple Legal Statuses of Early Modern Ragusa

The last question this essay seeks to address is: What should one do with the peculiar Ragusan interpretation of its tributary status? Should it be taken into account or should it be dismissed completely as the boasting of a small republic? Put more generally, which of the two drastically different interpretations—Ragusan or Ottoman—should be taken as objective when it comes to defining the legal status of the republic? Was Ragusa a status of the ragusan republic in the early modern period 111 fully independent state or a dependent polity under the supreme rule of the sultan? As noted, modern historiography has mostly opted for the latter solution—the majority of specialists agree that Ragusa was de iure an Ottoman “vassal.”37 Although far from being wrong, this solution is nonetheless only a partial and therefore unsatisfactory description of the international status of Ragusa. It is problematic because, on the plane of international law, it absolutely favors the Ottoman understanding of the republic’s tributary status, dismissing altogether the very different Ragu- san interpretation. Such an easy dismissal seems unjustified for a simple reason: this interpretation was far from being exclusively Ragusan; it had quite a formidable following. The central and fundamental claim of Ragu- san self-representation—that the city was de iure an independent state regardless of its tributary status—was accepted not only by Ragusans but also by the vast majority of political thinkers and governments in early modern Europe. This is clearly visible when one examines the relevant works of sixteenth- and seventeenth-century European political thinkers, historians, and travel writers, in which Ragusa is treated like any other independent Christian state. Since in the Western political tradition tribute did not necessarily imply submission, Christian commentators saw nothing problematic in stating that the city was “free,” “sovereign” or sui iuris, even as they mentioned its annual tribute to the Ottomans.38 If these Western authors offered any explanation of the tribute’s meaning at all, they

37 For a few typical examples, see footnote 17. Here one should also mention a notable exception to this general trend—the influential historian Vinko Foretić, who persistently refused to acknowledge Ottoman sovereignty over Ragusa. Echoing the traditional Ragu- san interpretation of tributary status, on several occasions Foretić insisted that the Repub- lic had never been under the rule of the sultan, but simply “acknowledged the Ottoman Empire as a dominant power,” and paid harac, “which in fact was the tribute for the price of peace.” Vinko Foretić, “Dubrovnik u doba Marina Držića” [Ragusa in the age of Marin Držić], in Vinko Foretić, Studije i rasprave iz hrvatske povijesti (Split and Dubrovnik, 2001), 302, 304, 305; similarly, in Foretić, Povijest Dubrovnika, vol. 1, 315–325. 38 For instance, the Venetian nobleman Giovanni Battista Giustiniano, in his mid- sixteenth century description of the city, mentions that Ragusans paid an annual tribute to the Ottomans but immediately afterward continues: “They govern themselves in the way of the Republic, and they live free and not subjected to anyone [sic] choosing for themselves every month a ruler who is called rector . . .” (“Si regono a republica, et vivono liberi non sottoposti ad alcuno et fanno ogni mese un capo, ch’ha nome di rettor (. . .)”) Simeon Ljubić, ed., Commissiones et relationes venetae, tomus 2 (annorum 1525–1553) (Zagreb, 1877), 249. A similar observation, comfortably joining liberty and paying tribute, is found in the description of Ragusa by a Friulan geographer, Giuseppe Rosaccio, published in 1598: “It [Ragusa] maintains itself in liberty by paying 14 thousand ducats to the Turk,” Giuseppe 112 lovro kunčević closely echoed the explanations traditionally given by Ragusans themselves. Tribute was represented as either the price of “peace” and Ottoman “protection,” which actually enabled the city to preserve its “liberty,” or, following the explanation typical of Ragusan Renaissance historiography, as nothing more than the price of the abundant economic privileges that the city enjoyed in the empire.39 All of these commonplaces were summarized by Melchior Seydlitz, a sixteenth-century Silesian travel writer, who apparently saw no difference between the international status of Ragusa and that of the Venetian republic: The city of Ragusa is a lordship for itself [“eine Herrschafft vor sich selbst”], like Venice; it has some land and islands subjected to itself; as I am told, it pays to the Turks 14,000 ducats annually so that its territory is left in peace and in order to safely trade and travel on the sea and in their [Turkish] lands.40 That Ragusa was perceived as a fully sovereign state by Christian observers is also confirmed by probably the greatest authority imaginable— Jean Bodin, the very inventor of the modern concept of sovereignty. In his immensely influential Six Books of the Republic, Bodin points out that what constitutes a commonwealth (respublica) is the presence of sovereign power, regardless of its population or the size of its territory. He continues by making an immensely interesting point: Just as the mouse is as much numbered among animals as is the elephant, so the rightly ordered government of only three households, provided they are subject to a sovereign authority, is just as much a commonwealth as a great empire. The principality of Ragusa, which is one of the smallest in Europe, is no less a commonwealth than the empires of the Turks and the Tartars which are among the greatest in the world . . .41 This passage clearly shows that one of the most influential political thinkers of the early modern period not only perceived Ragusa as a sovereign

Rosaccio, Viaggio da Venetia, a Costantinopoli, Per Mare e per Terra, insieme a quello di Terra Santa (Venice, 1598), 21–23. 39 Good examples of such trivializing explanations of the tribute can be found among the numerous descriptions of Ragusa collected by Jorjo Tadić, Promet putnika u starom Dubrovniku [The traffic of travelers in old Ragusa] (Dubrovnik, 1939), 185, 186, 188, 191, 192, 202. See also Radovan Samardžić, “Nekoliko francuskih putopisaca XVII veka o Dalmaciji i Dubrovniku” [Several seventeenth-century French travel writers on Ragusa], Zbornik Filo- zofskog fakulteta knjiga 7, no. 1 (1963): 364–365, 368. 40 Melchior Seydlitz, Gründtliche Beschriebung der Wallfart nach dem heiligen Lande (Görlitz, 1591), no pagination [148]. 41 Jean Bodin, Six Books of the Commonwealth, trans. M.J. Tooley (New York, 1967), 7–8. status of the ragusan republic in the early modern period 113 state, but even as being “no less” a sovereign state than the Ottoman Empire itself! Importantly, Bodin was far from being the only Western political writer of the sixteenth and seventeenth centuries to think along these lines. Although most of them mentioned the Ottoman tribute, numerous other influential authors—such as Francesco Sansovino, Louis Moréri, and Johannes Gothofredus—referred to Ragusa in their works as a free republic (it was frequently evoked together with other republican regimes such as Venice, the Swiss cantons or Lucca).42 Some authors went much further, uncritically reproducing even the most megalomaniac claims of the Ragusan patriciate. Thus, in his mid-seventeenth century description of the world, Luca de Linda not only states that the republic was an independent state at the present, but also claimed that it “never [sic!] bore the servile yoke of another’s dominion.”43 The first serious doubts about the independence of the Ragusan republic seem to have appeared only in the second half of the seventeenth century, when several Western commentators began to speak about the “apparent,” “imagined” or “fantastic” liberty of Ragusa. However, it is important to stress that these doubts were not due to the tributary status, but due to the small size and the military impotence of the republic, which resulted in its permanent dependence on the goodwill of more powerful states. In other words, in the late Baroque Europe of territorially vast and powerful monarchies, claims to full sovereignty by a microscopic and helpless city-state were treated with increasing skepticism.44 It was only in the eighteenth century, with its stricter notions of sovereignty and its better knowledge of the Ottoman commonwealth, that some—but far from all—European commentators raised the question of whether Ragusa was truly an independent state due to its relationship with the Sublime

42 For references to Ragusa as an independent state in these influential works, see Fran- cesco Sansovino, Del governo dei Regni e delle Republiche cosi antiche come moderne. Libri XVIII (Venice, 1561), 123–129; Joannes Ludovicus Gothofredus, Archontologia cosmica, sive imperiorum, regnorum, principatuum rerumque publicarum omnium per totum Terrarum Orbem (Frankfurt, 1628), 623–627; Louis Moréri, Le Grand dictionnaire historique, ou le Mélange curieux de l’histoire sacrée et profane (Paris, 1759), vol. 9, 24. 43 Luca di Linda, Le relationi de descrittioni vniversali et particolari del mondo (Venice, 1664), 640–641. 44 Several examples of seventeenth-century authors doubting Ragusan sovereignty on these grounds are Vjekoslav Jelavić, “Doživljaji Francuza Poullet-a na putu kroz Dubrovnik i Bosnu (godine 1658)” [The experiences of the French traveler Poullet on his voyage through Ragusa and Bosnia (in the year 1658)], Glasnik Zemaljskog muzeja u BiH 20 (1908): 24–27; Paul Rycaut, The Present State of the Ottoman Empire, in Three Books Containing the Maxims of the Turkish Politic, Their Religion, and Military Discipline (London, 1668), 66; Samardžić, “Nekoliko francuskih putopisaca,” 376–377. 114 lovro kunčević

Porte.45 However, the majority of political and geographical works of the period reproduced a different image of the city-state: Ragusa was seen as an irrelevant but independent polity that preserved its “liberty”—treated with more or less sarcasm—by relying on a number of stronger states as its protectors. Importantly, the Ottomans were usually mentioned as simply one among many protectors of the republic, not substantially different from the Habsburgs, Spain, Venice, the Papacy, France or Naples.46 Although much harder to interpret than the explicit reflections of political thinkers, the diplomatic practice of Christian states yields similar conclusions concerning the international status of Ragusa. European governments treated the republic in the same way as any other state, regardless of its tributary status.47 Stated forcefully (and somewhat anachronistically), Ragusa was perceived as a legitimate independent subject of international relations, with whom diplomacy was to be conducted directly and without expecting any interference from the Ottoman Empire. Thus, for example, official Ragusan envoys were treated according to the customary diplomatic protocol and granted the usual immunities and privileges of representatives of foreign powers. Equally so, Christian rulers corresponded with the Ragusan government following the normal diplomatic routines, acknowledged its official documents, and issued

45 A good example is the Scottish geographer Thomas Salmon, who seems unable to make up his mind, writing: “Although they rule themselves in a certain way independently from any other sovereign, at the same time they have the habit of including themselves among the subjects of the Ottoman Porte to which they are paying an annual tribute in cash; the price, they say, of the protection provided to them by the Turks”. Thomas Salmon, Lo stato presente di tutti i paesi, e popoli del mondo, naturale, politico e morale (Venice, 1753), vol. 20, part 1, 381. 46 Several typical examples are Johann Hübner, Allgemeine Geographie aller vier Welt-Theile, durch und durch verb., vielfältig verm. und bis auf gegenwärtige Zeiten fortge- setzt von G.F.K, part 1 (Dresden and Leipzig, 1761): 780–781; “Raguse,” in Encyclopédie, ou dictionnaire raisonné des sciences, des arts et des métiers, ed. Denis Diderot and Jean le Rond D’Alembert (available online: http://artflx.uchicago.edu/cgibin/philologic/getobject .pl?c.12:1953.encyclopedie0311.7712745); D. Anton Friderich Büschings Erdbeschreibung. Zweyter Theil, welcher Ost- und West-Preußen, Polen und Litauen, Galizien und Lodomerien, Ungarn, die denselben einverleibten Reiche und Siebenbürgen, die Republik Ragusa und das osmanische Reich, enthält (Hamburg, 1788), 600–601. 47 The only European state that refused to acknowledge Ragusa as an independent republic in this period was its powerful rival, Venice. However, this was not due to the Ragusan tributary position, since such Venetian reservations were clearly present even before the first Ragusan ‘ahdname. The reason can probably be found in the fact that before its emancipation in 1358, Ragusa was a Venetian possession and always remained a highly desired acquisition by its powerful neighbor. See Mitić, Dubrovačka država, 138–139; idem, “Za Serenissimu—Dubrovnik samo komuna, nikad Republika” [For the Serenissima— Ragusa is only a commune, never a republic], Dubrovnik 5 (1976): 65–69. status of the ragusan republic in the early modern period 115 their own to Respublica Ragusina. Moreover, the republic made standard international treaties—signed and ratified by both sides—directly with major European powers without any interference from the Ottomans.48 Finally, Ragusan neutrality was formally acknowledged and respected during the numerous wars of the period, including those between the Ottoman Empire and Christian powers. This shows that Ragusa was not only perceived as a legitimate international subject, but also as being a polity clearly distinct from the Ottoman Empire. All in all, during the sixteenth and seventeenth centuries the Ottoman Empire did not intervene in a single instance as the de iure overlord in Ragusa’s diplomatic commu- nications with Western powers, nor was there any instance of Christian powers refusing to conduct diplomacy with Ragusa on the grounds of it being perceived as a dependent tributary state.49 By now, two important facts have been established. First, the Ragu- san republic was perceived as an independent state by the vast majority of early modern commentators. Second, in the diplomatic practice of European governments it was treated in the same way as any other independent state. When all of this is taken into account, it becomes clear that the widely accepted perception of the legal status of Ragusa—that it was a “vassal” state under the rule of the sultan—is less obvious than it appears at first. The problem is that this clearly was not the way in which the majority of contemporaries understood Ragusan international status. In other words, such an understanding of the Ragusan legal position is problematic because it rests on one dubious assumption: that the only

48 The international treaties of Ragusa are relatively rare because of the fact that the city did not participate in any European war after the mid-fifteenth century. However, two examples that spring to mind are the treaty of friendship and trade with France (1776) and the never-enacted treaty of 1684, in which Ragusa negotiated to come under the “protection” of the Austrian Habsburgs once they had expelled the Turks from its hinterland. For the treaty with France, see Lujo Vojnović, La monarchie française dans l’Adriatique; histoire des relations de la France avec la république de Raguse (1667–1789) (Paris and Barce- lona, 1917), 202–230; for the treaty with the Emperor Leopold I, see Zdenko Zlatar, Between the Double Eagle and the Crescent: The Republic of Dubrovnik and the Origins of the Eastern Question (Boulder, CO, and New York, 1992). 49 To the best of my knowledge, even in the eighteenth century there is only one example of a European power denying Ragusa the status of independent state because of its tributary status. The Ragusan ambassador to St. Petersburg in the early 1770s was denied an audience with the Empress Catherine and not recognized as a representative of a sovereign power. However, this episode is not only exceptional, it also took place very late and involved a relatively new European player, the Russia of Catherine the Great. See Foretić, Povijest Dubrovnika, vol. 2, 262–268; Jovan Radonić, ed., Dubrovačka akta i povelje (Belgrade, 1951), vol. V, 403–404, 412, 442–445, 472. 116 lovro kunčević correct interpretation of Ragusan international status was the Ottoman one, while that of the Christian side (including Ragusa itself ) should be dismissed altogether. The question is: Why should that be the case? Why should one, adopting the Porte’s official position, state that Ragusa was de iure a vassal state under the sultan’s rule and completely ignore a totally different understanding held by the overwhelming majority of international community? Why was Istanbul more qualified to define Ragusan status than Rome, Paris, Madrid or, for that matter, Ragusa itself? What exactly gave the Sublime Porte such a privileged perspective with regard to issues of international law and order? Somewhat cynically one could argue that what makes the Ottoman interpretation of Ragusan status the “objective” one is the fact that the Ottomans had overwhelming military superiority enabling them to, if necessary, force their view upon the small city-state. However, the Spanish—primarily the neighboring viceroys of Naples—possessed equally overwhelming military superiority and yet they treated Ragusa as an independent state. In other words, even if one insists on the factual relationship of power, again it is not clear why Istanbul’s perspective would be the one privileged since there were other powerful states whose interpretation of Ragusan status differed greatly from that of the Porte. Another problem with accepting the Ottoman interpretation of Ragu- san status as the sole “objective” one becomes apparent after a certain reductio ad absurdum, i.e., after an attempt to investigate the final consequences of this position. If one tries to be consistent in believing the Ottomans about the legal status of other states, a number of fantastic and telling results occur. For example, as tribute payers both Poland and Ven- ice should be considered under the rule of the sultan and enjoying essentially the same dependent legal status as Ragusa or Transylvania. Equally, the French king should in fact be granted the title of the “Emperor of the French,” while the sixteenth-century Holy Roman Emperor should be mentioned merely as the “King of Vienna.”50 Of course, today none of this is taken seriously; it is dismissed without a blink as the fantastic and megalomaniac discourse of the Porte. Yet this is the very same discourse—the official Ottoman interpretation of international relations resting on Islamic law—on the basis of which Ragusan international legal

50 Faroqhi, Ottoman Empire, 78–79. Intriguing examples are also to be found in Maria Pia Pedani, Dalla frontiera al confine (Rome, 2002), 9–10; Bernard Lewis, The Muslim Dis- covery of Europe (New York and London, 1982), 205–208. status of the ragusan republic in the early modern period 117 status is normally determined. If one is deeply skeptical, even totally dis- missive of this discourse when it comes to some states, why is it uncritically accepted when it comes to the others? Why precisely should one believe it regarding Ragusa and completely reject it in relation to Venice, Poland or France? Where and why does the utterly fantastic and routinely dismissed discourse turn into a reliable, even exclusive, vantage-point on early modern international relations? All of this does not mean that the Ottoman interpretation of Ragusan status should be rejected altogether—far from it. What should be rejected, however, is the idea that it is the only correct and objective answer regarding the de iure position of the Ragusan republic. Once one takes into account the nature of international law and relations during the sixteenth and seventeenth centuries, it becomes clear that there were at least two, if not more, equally valid views of the legal status of Ragusa—depending on who was asked to define it. Perhaps the best way of clarifying this point is to borrow the influential concept of “international society” developed in international relations theory. According to the classical formulation by Hedley Bull, an international society is “a group of states“ which “have established by dialogue and consent common rules and institutions for the conduct of their relations, and recognize their common interest in maintaining these arrangements.”51 These “common rules and institutions” shared by the members of the same international society include things such as fundamental political values, international law, the norms of diplomacy, tacit understandings about how to conduct war or trade, and sometimes also the institutional forums for settling inter-state disputes. While the contemporary world is integrated in a “universal international society,” the early modern epoch was characterized by the simultaneous existence of several of them, typical examples being China, Christian Europe, and the so-called “Arab-Islamic” (in fact Ottoman) international society, centered in Istanbul and stretching from Morocco to Afghanistan.52

51 Hedley Bull and Adam Watson, “Introduction,” in The Expansion of International Society, ed. Hedley Bull and Adam Watson (Oxford, 1984), 1. For the classical statement of the idea of international society, see also Kai Alderson and Andrew Hurrell, Hedley Bull on International Society (Basingstoke, 2000), especially, 3–73; Hedley Bull, The Anarchical Society: A Study of Order in World Politics (New York, 1977). 52 On the emergence of a universal international society, see Hedley Bull, “The Emer- gence of a Universal International Society,” in The Expansion of International Society, ed. Hedley Bull and Adam Watson (Oxford, 1984), 117–126; Adam Watson, The Evolution of 118 lovro kunčević

Clearly, different international societies will have completely different interpretations of international order in general and the legal status of individual states in particular. This elementary lack of consensus between international societies is what makes asking for a single and objective answer concerning the legal status of early modern Ragusa profoundly anachronistic. The Ragusan republic was strongly integrated in two separate international societies—Christian Europe and the Ottoman (“Arab-Islamic”) sphere—each of which had its own specific legal and political culture and its own peculiar conceptualization of international order. Therefore, if one does not want to anachronistically disregard the major civilizational gap between the two international societies of early modern Europe, one must admit the obvious—Ragusa had multiple legal statuses.53 In other words, there were at least two equally valid answers to the question of the legal status of the Ragusan republic, depending on which part of the international community was asked to define it. From the vantage point of Ottoman international society it was a dependent polity under the sultan’s rule, a part of either dar ül-‘ahd or dar ül-İslam. From the vantage point of European international society—and itself—it was perhaps not a very prestigious state, yet nevertheless it was fully independent. One could say that from the legal perspective Ragusan sovereignty was “Janus-like”: the status of the republic was strikingly different depending on whether it was perceived from the East or from the West. Again, there is absolutely no reason to assume that any of these interpretations is true and the others false, since none of their proponents had a special claim to “objectivity.” If there is one objective answer regarding Ragusan independence it does not concern its legal dimension but the factual exercise of power. De iure Ragusa was both a dependent polity under Otto- man rule and an independent republic; de facto it was only one thing—an independent state.54

International Society (New York, 1992). On the multiple international societies of pre modern Europe and on Arab-Islamic international society: Bull, “Introduction,” 3–4. 53 On the difference between the two societies, see Thomas Naff, “The Ottoman Empire and the European States System,” in The Expansion of International Society, ed. Bull and Watson, 143–170; Bull, “The Emergence.” More specifically on the profound differences between Christian and Islamic international law in this period, see Khadduri, War and Peace, 42–43; idem, The Islamic Law of Nations: Shaybānī’s Siyar (Baltimore, 1966), 64–67. 54 Even today, in the more or less uniform “universal international society” there are polities whose status is interpreted in drastically different ways by various members of the international community, such as Kosovo, Taiwan, Abkhazia, and South Ossetia. It is only natural that in the early modern period, lacking the universally accepted norms status of the ragusan republic in the early modern period 119

Conclusion

To sum up, the broadly accepted idea that Ragusa was a vassal state under the supreme rule of the Sublime Porte is only a partial and therefore unsatisfactory answer regarding its legal status. It ignores the crucial fact that the overwhelming majority of the contemporary international community clearly did not perceive Ragusa in this way, but, quite the contrary, saw it as a de iure independent state. Such different perceptions of the republic were a natural consequence of it being integrated in two different international societies with their own specific systems of international law. Once this profound gap between the legal cultures of early modern Europe is taken into account, it becomes clear that insisting on a single “objective” answer regarding the legal status of Ragusa is necessarily anachronistic. Simply put, the republic’s legal status was relational and multiple, depending on which international society defined it. Although Ragusa was probably the most salient example, there are traces of such a double legal status among other tributary states as well. Probably the best example is Transylvania, another Ottoman satellite with broad autonomy. Depending on who was asked, it was either a tributary state under the rule of the sultan or an integral part of the Kingdom of Hungary under Habsburg sovereignty. These two contradictory interpretations were, importantly, also accepted by the Transylvanians themselves, who, depending on the situation, agreed to one or the other. Again, as in the Ragusan case, there is no reason to assume that one interpretation should be proclaimed as the only true and objective one.55 In fact, Transylvanian legal status had a further element of complexity, since its interpretation by the Habsburgs changed over time, even quite drastically, as the contribution of Teréz Oborni in this volume shows. While in the treaties of Speyer, Prague, and Nagyszombat the Habsburgs affirmed their usual position about Transylvania being a province of Hungary, in Kassa

of international law, such divergences of interpretation were even more frequent and profound. 55 Clearly, the Ottoman interpretation was more frequently enforced, but on two occasions the Habsburgs also had control of the country, enforcing their understanding of its status. It is generally a mistake to reject one interpretation of legal status in favor of another simply because one was put into practice more frequently. Would we be more prone to say the Habsburg understanding was “objective” had they controlled Transylvania four or five times in the sixteenth and seventeenth centuries or had they controlled it for several decades? Where, then, exactly is the limit? How much factual control over a territory is necessary to make the controller’s understanding of its legal status “objective”? 120 lovro kunčević and Pozsony they admitted that at the moment it existed in an alienated form under the rule of a separate prince. Finally and most interestingly, in the Peace Treaty of Nikolsburg the emperor acknowledged the ruler of Transylvania as his equal with no mention of Habsburg sovereignty whatsoever. It seems, therefore, that Transylvania not only had dual legal status, but that—differently from Ragusa—its Western variant changed drastically over time. Moreover, as in the Ragusan case there are striking—but, admittedly, less frequent—examples of Western commentators mentioning Transylvania as a fully independent state despite its tributary status. As if echoing the Ragusan distinction between its status as a tributary and that of a subject, Sir Thomas Roe, English ambassador to the Porte stated in 1625: “Gabor [Bethlen] is a free prince to us, in our consideration of him, and in his owne estimation of himselfe; and though hee pay tribute to the grand signor, hee is not therefore a subiect.”56 Last but not least, similar to Ragusa and even more salient because of its military relevance, Transylvania made fully-fledged international treaties directly with West- ern powers without any legal interference from the Porte. However tricky it might be to reconstruct a polity’s status from the highly pragmatic and inconsistent diplomacy of the early modern period, this fact nonetheless suggests that from the perspective of Christian princes Transylvania was an independent international subject, not a mere Ottoman possession.57 One could expand the general point of this essay even further by remarking that multiple and relational legal status was not only characteristic of tributary polities, but of all states that were in contact with more than one international society. The difference, however, was that the more distant or more powerful European states could easily dismiss the Ottoman interpretation of their standing as an irrelevant abstraction, while for tributary states the dual legal status was a concrete political fact that profoundly affected their quotidian functioning. The extensive trading privileges that Ragusa enjoyed are an excellent case in point. On one hand, the city had immense privileges in the Ottoman Empire, since it

56 Sir Thomas Roe to George Abbot (Constantinople, 9/19 June 1625) The Negotiations of Sir Thomas Roe, in His Embassy to the Ottoman Porte, from the Year 1621 to 1628, ed. Samuel Richardson (London, 1740), 408. For this reference—and generous help regarding Transyl- vanian matters in general—I am profoundly grateful to Gábor Kármán. 57 For an overview of the international status of Transylvania, see Gerald Volkmer, Das Fürstentum Siebenbürgen 1541–1691: Aussenpolitik und völkerrechtliche Stellung (Heidel- berg, 2002); Gábor Kármán, “Transylvania between the Ottoman and Habsburg Empires,” in Statehood Before and Beyond Ethnicity: Minor States in Northern and Eastern Europe, 1600–2000, ed. Linas Eriksonas and Leos Müller (Brussels, 2005), 151–158. status of the ragusan republic in the early modern period 121 was seen as a vassal. On the other, it had equally significant privileges in Western Europe, especially Habsburg domains, because it was perceived as an independent state (no Habsburg would have granted such commercial advantages to the subjects of the sultan). In sum, “the most faithful tributaries of the Grand Signore” were entitled to a totally different set of privileges and obligations than the independent Respublica Ragusina. In large part Ragusan prosperity stemmed from the fact that the republic managed to keep the benefits of both statuses at the same time. In conclusion, one more caveat should be made. The legal status of early modern polities was relational and multiple not only becaues of the changing perspectives of different international societies. International societies themselves should not be imagined as monolithic. Within the uniting framework of shared political and legal traditions there was still more than enough space for states to disagree on each other’s legal status (the most famous early modern example surely being the United Provinces until 1648). Importantly, one could make yet another shift of perspective, going “below” the level of individual states, and analyze different interpretations of the legal status of polities from the level of individual authors: political thinkers, historians, legal scholars, etc. Whatever the perspective one might take, however, the basic point remains the same. It is mislead- ing to insist on determining a single objective answer regarding the legal status of an early modern polity. What historians should do instead is reconstruct the ever-changing and frequently competing interpretations, the purposes they served, the audiences they targeted, and consequences they had. This study is a modest attempt in that direction.

Cossack Ukraine In and Out of Ottoman Orbit, 1648–1681

Victor Ostapchuk

In the second half of the seventeenth century a great upheaval occurred in the Ukrainian territories of the Polish–Lithuanian Commonwealth that led to the unraveling and eventual transformation of the international order in Eastern Europe. For more than a generation the revolt against the Polish-Lithuanian Commonwealth sparked in 1648 by Bohdan Khmel- nytsky, hetman of the Zaporozhian Cossacks, and the ensuing wars and social upheavals, to greater or lesser extents drew in most near and distant neighbors—in particular the Ottoman Empire, the Crimean Khanate, Moldavia, Transylvania, Muscovy, and Sweden. This whirlwind of events eventually brought the Ottoman Empire and Muscovy into their first major military conflict (the 1569 Ottoman Don-Volga-Astrakhan expedition and later proxy encounters in the North Caucasus notwithstanding) and contributed to the demise of the Commonwealth. By the late 1660s the Ottomans felt compelled to reverse their centuries-old policy of avoiding expansion beyond the northern Black Sea coastal region and engage in an active northern policy that led to a struggle for the steppes between the Dniester and Dnieper Rivers and beyond. Between the Ukrainian revolt of 1648 and the Treaty of Bahçesaray of 1681, when the Porte effectively abandoned its active northern Black Sea policy, though it still held on to Podolia, a major subplot emerged: the search by Cossack hetmans and Ottoman sultans and viziers and their respective envoys for mutually agreeable terms by which Cossack Ukraine,1 once a fierce foe of the Turks and Tatars, could become a subject

1 By “Cossack Ukraine” we refer to those areas of Ukraine dominated by the Ukrainian (as opposed to the Russian Don) Cossacks—originally these were in the lower Dnieper region, known as Zaporozhia (south of the modern city of Zaporizhe). After 1648, we use the term “Cossack Ukraine” to indicate in addition those territories no longer under the control of the Commonwealth, that is, the provinces of Kiev, Bratslav, and Chernihiv, but not the predominantly ethnic Ukrainian provinces of the Polish Crown Podolia, Volhynia, and eastern Galicia (region of Lviv, so-called Red Ruthenia). Although the boundaries of Cossack Ukraine waxed and waned during the upheavals of our period the conventional name Hetmanate has been used to refer to this polity because of the presence of 124 victor ostapchuk of the Porte. This essay seeks to provide an interpretation and better understanding of the Ukrainian-Ottoman encounter during this turbulent and pivotal period. While it brings into play key primary sources as well as essential secondary literature, because of space limitations, it relies on generally accepted knowledge to provide an outline of events and their relevant contexts.2 To understand the swings in the Porte’s stance toward the Black Sea and the expanses to its north, it is necessary to place this portion of the northern frontiers in the longue durée.3 After the fall of Constantinople,

state-like institutions (e.g., internal administration, courts, army, foreign relations). From the late 1660s through the end of our period Cossack Ukraine was de facto divided in two along the Dnieper—in the west the Right-Bank Hetmanate and in the East the Left- Bank Hetmanate. In summary, although in this essay we use both “Cossack Ukraine” and “Hetmanate,” we prefer the former during the periods of war and internal anarchy discussed here, namely, the Khmelnytsky years (1648–1657) and thereafter the so-called “Ruin” a name that Ukrainian historiography has applied to the devastating period that ended in the early 1680s. 2 The most detailed and still authoritative coverage of the Khmelnytsky era is that of Mykhailo Hrushevsky’s multi-volume Istoriia Ukrainy-Rusy, now available in an Eng- lish translation that thanks to careful editing supersedes the original Ukrainian version: Mykhailo Hrushevsky, History of Ukraine-Rus’, vol. 8, The Cossack Age, 1626–1650, ed. Frank E. Sysyn, trans. Marta Olynyk; vol. 9, book 1, The Cossack Age, 1650–1653, ed. Serhii Plokhy and Frank E. Sysyn, trans. Bohdan Strumiński; vol. 9, book 2, part. 1, The Cossack Age, 1654–1657, ed. Serhii Plokhy and Frank E. Sysyn, trans. Marta Olynyk; vol. 9, book 2, part 2, The Cossack Age, 1654–1657, ed. Yaroslav Fedoruk and Frank E. Sysyn, trans. Marta Olynyk (Edmonton and Toronto, 2002–2010). Some major gaps in evidence on the Ottoman side were filled by the Czech Iranist and Ottomanist Jan Rypka: “Z korespondence Výsoké Porty s Bohdanem Chmelnickým” [From the correspondence of the Sublime Porte with Bohdan Khmelnytsky], in Z dějin Východní Evropy a slovenstva: Sborník vénovaný Jaro- slavu Bidlovi, profesoru Karlovy University k šedesátým narozeninám, ed. Miloš Weingart, et al. (Prague, 1928): 346–350, 482–498; “Weitere Beiträge zur Korrespondenz der Hohen Pforte mit Bohdan Chmel’nyćkyj,” Archiv Orientální 2 (1930): 262–283; “Dalši příspěvek ke korespondenci Vysoké Porty s Bohdanem Chmelnickým” [Another contribution on the correspondence of the Sublime Porte with Bohdan Khmelnytsky], Časopis Národniho Musea 105 (1931): 209–231. The fundamental work on the Doroshenko period remains the magnum opus of a descendent of the same family: Dmytro Doroshenko, Het’man Petro Doroshenko (New York, 1985). Important also is a joint study by the last two authors, which includes some Ottoman source evidence: Dmytro Dorošenko and Jan Rypka, “Hejt- man Petr Dorošenko a jeho turecká politika” [Hetman Petro Doroshenko and his Turk- ish policy], Časopis Národniho Musea 107 (1933): 1–55. Finally it is worth mentioning an erudite and original investigation of realities and counterfactuals of the Cossack-Ottoman relationship—to what extent was an Ottoman alignment a viable and desirable alternative for the Hetmanate? Dariusz Kołodziejczyk, “Tertium non datur? Turets’ka al’ternatyva v zovnishii politytsi Kozats’koi derzhavy” [No third way? The Turkish alternative in the foreign policy of the Cossack state], in Hadiats’ka Uniia 1658 roku, ed. Pavlo S. Sokhan’, Viktor Brekhunenko et al. (Kiev, 2008), 67–80. 3 For a more elaborate version of the following presentation of the Black Sea as an Ottoman mare nostrum see Victor Ostapchuk, “The Human Landscape of the Ottoman cossack ukraine in and out of ottoman orbit, 1648–1681 125

Mehmed the Conqueror (1451–1481) moved to turn the Black Sea into an “Ottoman lake” by taking control of key ports and fortresses on various shores of the sea and most significantly, by gaining control of the southern shore of the Crimea and establishing suzerainty over the Crimean Khan- ate in 1475. Control of the Black Sea meant, above all, control of trade so that the immense food, raw material, and human resources of the Black Sea basin could be channeled toward the interests of empire-building, including the provision of regular and affordable supplies for the growing capital city of Istanbul. Halil İnalcık has indicated the significance of control of the sea for the strength and well-being of the Ottoman Empire.4 As Gheorghe I. Brătianu demonstrated, the move to control the Black Sea occurred almost as a corollary of control of the Straits and is analogous to earlier Byzantine and Venetian control of the Straits, which amounted to control of Black Sea trade (the former directing it to Constantinople and the latter to the Mediterranean). And while this control was established relatively easily along most stretches of coast by simply taking over Italian emporia and fortresses, there were two achievements in which Ottoman success was no trivial matter: the establishment of suzerainty over the Crimean Khanate—a Chinggisid successor state of the Golden Horde— and Bayezid II’s (1481–1512) conquest of Kili and Akkerman at the mouths of the Danube and Dniester in 1484. Rather than attempting to establish themselves in the steppe zone to the north by subduing the Tatars, the Ottomans seemed to understand the nature of the northern Black Sea steppes, and realized that attempting to conquer this somewhat arid and sparsely populated zone would be futile and pointless. Thus they did not attempt to take direct control, instead they established Ottoman provincial rule in strategic coastal areas, such as Kefe, Azak, Akkerman, Kili, and, in 1528, Özi. In these locales Ottoman institutions were not fully established (e.g., there was no timar system in the province of Kefe). Instead of attempting to eliminate the region’s largely nomadic order, they established a mutually benefi- cial economic relationship with the Crimean Khanate. In this new situation, the Tatars mounted ever larger raids into the southeastern regions

Black Sea in the Face of the Cossack Naval Raids,” Oriente Moderno 20 n.s. (2001): 23–95, esp. 27–37. 4 On the Ottoman closing of the Black Sea, see Halil İnalcık, “The Question of the Clos- ing of the Black Sea under the Ottomans,” Αρχείον Πόντου 35 (1979): 74–110; idem, Sources and Studies on the Ottoman Black Sea, vol. 1: The Customs Register of Caffa, 1487–1490, ed. Victor Ostapchuk (Cambridge, MA, 1996). 126 victor ostapchuk of Poland–Lithuania (mainly Ukraine) and southern Muscovy to obtain captives for the vast Ottoman slave market. The Ottoman–Crimean relationship, although not without its periodic conflicts, afforded the Porte sufficient influence among the Tatars in the Crimea and beyond to manip- ulate the steppe region in its favor. This important development set the stage for the ensuing centuries. Being the first power not only to take control of the Black Sea but also to establish a strong relationship with the inhabitants of the steppes, the Ottomans in effect locked their northern neighbors Poland–Lithuania and Muscovy out of the Black Sea region for several hundred years. In the steppe zone controlled by a militarily formidable Crimean Khanate that directed its raiding activity against its northern neighbors, the Porte had a very effective “active” buffer zone protecting its Black Sea dominion. Thus from the late fifteenth to the late seventeenth century the Ottomans displayed no interest in expanding past the northern seaboard of the Black Sea and were satisfied with a passive, defensive stance in the region, pre- ferring to concentrate their efforts on expansion in central Europe, the Mediterranean, and in the East. Exceptions to this policy, such as the Don- Volga-Astrakhan campaign of 1569 or the Hotin War of 1621, only served to underscore the difficulty and futility of expansion to the north. At the same time, as Professor İnalcık has pointed out, while maintaining a passive Black Sea policy the Ottomans were, apparently, keenly aware of the relative strengths of the two major powers to the north, Poland–Lithuania and Muscovy, and concerned that neither of these powers became powerful enough to challenge Ottoman dominion over the Black Sea.5 What about Ukraine? It is one of the great ironies of the history of this region that the power and sway of the steppe, which in the Mongol era destroyed the East Slavic empire of Kievan Rus’, was in the post-Mongol period a prime stimulant for the rebirth of East Slavic power and sway in that same steppe. The dangerous conditions that a successor state of the Golden Horde (and therefore also of the Mongol Empire), namely the Crimean Khanate, maintained in the Pontic steppes eventually led to a response by the sedentary peoples. This response was partly modeled on the post-Mongol tradition of kazaklık, a state of vagabondage in the steppe wilderness that involved raiding, often with the goal of attracting a

5 Halil İnalcık, “The Origin of the Ottoman–Russian Rivalry and the Don–Volga Canal (1569),” Annales de l’Université d’Ankara 1 (1947): 47–110; idem, “Osmanlı-Rus Rekabetinin Menşei ve Don-Volga Kanalı Teşebbüsü (1569),” Belleten 12 (1948): 349–402. cossack ukraine in and out of ottoman orbit, 1648–1681 127 retinue of followers in order to gain political power.6 Desperate or adventurous inhabitants of the lands to the north of the steppes and even further afield, men from of all walks of life—runaway serfs, hunters, trap- pers, traders, townsmen, and even nobles—learned, often by example from the Tatars and Nogays, how to survive in the “Wild Field,” as the Pontic steppes were referred to then. With the help of gunpowder weapons and tactics, they went on the offensive against the Tatars and became a formidable force that by the end of the sixteenth century challenged the hinterland whence they originally came, that is, the military forces of the joint Lithuanian and Polish state. These were, of course, the Cossacks. Relevant for us is their Ukrainian incarnation, whose refuge was in the hard-to-access waterways, islands, and marshes of the middle Dnieper “beyond the rapids” (zaporozhe), the so-called Zaporozhian Sich (sich, “stockade”). In this environment, came another perhaps unexpected transformation—land dwellers became mariners and in the latter part of the sixteenth century the Zaporozhian Cossacks embarked on a spectacular career raiding first the northern and western coasts of the Black Sea and eventually all of its shores and shipping lanes. By the 1610s the Zaporozhians (and also their Russian counterparts, the Don Cossacks), became a primary problem for the Ottoman Empire: the Black Sea was no longer a peaceful and prosperous “Ottoman lake.” After military adventures in Ottoman, Muscovite, Danubian, and Polish– Lithuanian lands, and possibly even further west as mercenaries in the Thirty Years War, by mid-century, the Ukrainian Cossacks became increasingly confident actors on the international scene. This was in spite of the fact that they remained nominal subjects of the Polish–Lithuanian Com- monwealth and even experienced severe repression in response to several serious uprisings in the 1630s. Mostly thanks to the Cossack phenomenon, which managed to “cossackify” the aspirations and lifestyles of much of the peasant and town populations of the hinterland, the former heartlands of Kievan Rus’ gradually took on a new identity. Aside from social and economic factors, crucial was a revival of the endangered Orthodox heritage and culture that was in a life-death struggle (and dialogue) with Polish, in particular Roman Catholic, culture. While the old East Slavic

6 On Turkic kazaklık as well as Slavic Cossackdom now see Joo-Yup Lee, “The Forma- tion of the Qazaqs and the Socio-Political Phenomenon of Qazaqlïq” (PhD diss., University of Toronto, 2012). 128 victor ostapchuk word okraina, that originally meant “borderland,” in the sixteenth century was a place name referring to the middle Dnieper region, it eventually came to denote the southern and eastern borderlands of the Common- wealth. However, with the new period inaugurated by the events of 1648, Ukraina7 as a place name increasingly came to denote wherever the Cos- sack movement was active, even including all the provinces inhabited by Ruthenians.8 Thanks to the military prowess and steppe survival skills of both the Tatars (those connected with the Crimea as well as the independent, fully- nomadic Tatars of the steppe, known as Nogays) and the Zaporozhian Cossacks, for centuries the Ottoman, Polish–Lithuanian, and Muscovite empires could not subdue the borderlands between them. Only in the late eighteenth century, when the Russian Empire, having eliminated the khanate and hetmanate, completed the incorporation of the territory that would become modern Ukraine. Moreover, just as practically speaking the perennial threat from the south for the two northern powers were Tatar raiders rather than the Ottomans, so for the Ottomans the first challenge to their Black Sea realm and the greatest threat from the north were Cossack interlopers rather than the northern powers. And in each case the challenges were often directed at the heartlands and not merely the borderlands. By the time of Khmelnytsky the survival of Ukrainian Cossackdom seemed to hinge on its ability to become a legitimate player on the international scene rather than merely a destabilizing frontier phenomenon.

Bohdan Khmelnytsky’s Explorations in Multi-vectorism and the Porte

The dean of modern Ukrainian historiography, Mykhailo Hrushevsky, has provided an insightful explanation of Khmelnytsky’s foreign policy machinations. In his efforts to find a place in the international community for the nascent Ukrainian polity, the hetman navigated between and

7 On the evolution of the okraina/Ukraina as words and concepts, see Omeljan Pritsak and John S. Reshetar, Jr., “The Ukraine and the Dialectics of Nation-Building,” Slavic Review 22 (1963): 224–255 and now most recently Natalia Iakovenko, “Choice of Name versus Choice of Path: The Names of Ukrainian Territories from the Late Sixteenth to the Late Seventeenth Century,” in A Laboratory of Transnational History: Ukraine and Recent Ukrainian Historiography, ed. Georgiy Kasianov and Philipp Ther (Budapest and New York, 2009), 117–148. 8 Ruthenian is a conventional term for East Slavs in Poland–Lithuania, in opposition to those in Muscovy, mostly commonly applied to Ukrainians, but also to Belorussians. cossack ukraine in and out of ottoman orbit, 1648–1681 129 within several major systems of states and powers. One system was an anti-Catholic block of Orthodox and Protestant states: Muscovy, Ukraine, Moldavia, Wallachia, Transylvania, and Sweden. Operating within this system, Khmelnytsky politicked and connived to bring about an anti- Commonwealth alliance of at least some of these significant powers. Such an alignment could even accommodate membership for a part of the Commonwealth itself, namely, the Grand Duchy of Lithuania, whose ruling elite was heavily Protestant. Another completely different system was an anti-Ottoman coalition involving Moldavia, Wallachia, and Tran- sylvania—vassal states of the Ottomans—Muscovy and even the Com- monwealth itself. The elimination of the Crimean Khanate and truncation if not destruction of the Ottoman Empire offered tempting prospects to all these players, not least Ukraine. A variation of this system was a Euro- pean, largely Catholic alliance, involving Venice, then bitterly at war with the Ottomans, and Poland–Lithuania, including Cossack Ukraine. Indeed Khmelnytsky participated in the discussions for setting up such a crusade, hosting envoys of the Venetians and lending hope to the Poles that he would abandon his war on them and re-channel the energies of the Cos- sacks to the south. Yet another system involved Ukraine and the Crimea with possibly the Commonwealth against Muscovy, the Don Cossacks, and even some of the Circassians. Finally, there was the Ottoman system, in which Ukraine would find a place in an orbit of the Porte, similar to the Crimean Khanate, Moldavia, Wallachia, and Transylvania.9 The international relations of this era were made inscrutable by Khmelnytsky’s ability to operate within several of these systems practically simultaneously. For his contemporaries he was a most slippery ally or opponent. He is no less slippery as an object of research to the historian, who—unaware of these various systems—can easily fall into error and take at face value words and actions of this consummate politician to mean that the hetman was certainly an ardent and sincere follower of, for example, the Muscovite tsar or the Ottoman sultan, when at any given time he may have been so only partially or even not at all. The initial spark that supposedly set off the great Ukrainian revolt was almost a private affair—a conflict in early 1648 between Khmelnytsky and a Polish noble over property and a woman. It should be noted that in the years prior a perfect storm was gathering, thanks to huge dissatisfaction

9 Hrushevsky, History of Ukraine-Rus’, vol. 8, 523–524; Serhii Plokhy, The Cossacks and Religion in Early Modern Ukraine (Oxford, 2001), 274–278. 130 victor ostapchuk with the socioeconomic and political state of affairs among various sectors of Ukrainian society, be they Cossacks, peasants, nobles, or churchmen. In any event, feeling deeply wronged at the hands of the Polish administration, which refused to give him redress, Khmelnytsky went to the Zaporo- zhian Sich and roused its garrison to attack the Polish authorities, who had severely curtailed their rights and activities after the last major Cos- sack uprising ten years earlier. With each success on the battlefield more Cossacks joined Khmelnytsky, including the so-called Registered Cossacks who were on the payroll of the Polish Crown and considered more loyal to it. By the end of the 1648 campaigning season the forces of Khmelnytsky had spectacularly defeated the armies sent to stop him and rolled through Ukraine to the edge of Poland proper. A critical defining occurrence that determined whether the movement unleashed by Khmelnytsky would be another essentially social and economic revolt rather than what can be called an early modern national movement was the hetman’s stay in Kiev during the winter of 1648–1649. Prior to this his demands were primarily directed at redressing Cossack grievances. But in Kiev Khmelnytsky was swept away by the tumultuous reception that he was given by Ukrainian townspeople, church leaders, and intellectuals. He realized that what he had started was something much larger than he had ever imagined. During that winter Khmelnytsky incorporated into his agenda the strivings of the disaffected elements of Ukrai- nian society and prepared for an all-out war with the Commonwealth. It was during the winter of 1648–1649 that the idea of forming some sort of an independent or semi-independent Ukrainian state began to circulate. Already the seeds for the internationalization of the conflict were planted and the winter months inaugurated frequent diplomatic traffic between Khmelnytsky and neighbors of the Commonwealth.10 Khmelnytsky would never even have made his triumphant entry to Kiev at the end of 1648 and his revolt would have been no more than just another Cossack uprising had the hetman not initially made a bold and decisive move: he sent messengers to the Crimea and obtained the military support of the Crimean Khanate. In the following weeks and months and years it was demonstrated again and again that Tatar cavalry operating with Cossack infantry was a most formidable combination capable

10 Hrushevsky, History of Ukraine-Rus’, vol. 8, 515–529. See also Plokhy, Cossacks and Religion, 221, 227–228. cossack ukraine in and out of ottoman orbit, 1648–1681 131 of defeating just about any force the Commonwealth could muster. The Crimea, however, entered into the Ukrainian–Polish conflict out of self- interest. On one hand, the Cossack onslaught on the Commonwealth was evidence that a major northern power, which in the first half of the seventeenth century was the stronger of the two (as compared to Muscovy) was weakening. Disorder in the north was always to the khanate’s advantage and the endless campaigning brought by the Khmelnytsky war with Poland yielded enormous opportunities for booty and the acquisition of captives for the slave market, a mainstay of the Crimean economy. On the other hand, in the long term the emergence of a strong Cossack entity geographically closer to the Crimea than the Commonwealth or Muscovy was not a favorable prospect for the khanate. The Cossacks, as a potential rival, threatened the very existence of the khanate while as allies they eliminated Ukraine as a source of captives, leaving the Crimea with the prospect of relying on lands further removed to the north, in the Com- monwealth and Muscovy. Throughout his career, Khan İslam Giray III (1644–1654) made sure that Khmelnytsky persevered in his struggle with the Poles, yet he prevented the hetman from attaining total victory. Thus in several great and potentially decisive battles, at Zboriv in 1649, at Berestechko in 1651, and at Zhvanec’ in 1653, İslam Giray’s actions prevented Cossack victories and forced Khmelnytsky to negotiate a compromise with the Commonwealth. Without İslam Giray, the Khmelnytsky movement would surely have met with defeat or been forced to come to terms with the Poles, yet with him as sole ally, total victory could never be. Khmelnytsky realized this early in the struggle. After the campaign of 1648, the hetman sent envoys near and far in order to connect with all states that were present or potential enemies of the Commonwealth. Aside from the Crimea, his major hopes were Transylvania, whose ruler György Rákóczi I (1630–1648) had ambitions for the Polish throne, and Muscovy, the old rival of the Commonwealth that had lost significant territories to Poland–Lithuania in the wars of the first half of the century. The hetman was able to reach an understanding in principle for military support from the Transylva- nian prince’s son and successor, György Rákóczi II (1648–1660, with interruptions), though it would be several years before his armies moved against Poland. Almost from the beginning of his struggle with the Com- monwealth, Khmelnytsky offered to enter into the suzerainty of the tsar, hoping to thereby draw Muscovy into the war with its western neighbor. However, for almost six years Moscow steadfastly rebuffed Khmelnytsky’s 132 victor ostapchuk offers, repeatedly maintaining that it could not actively aid him since it had an “eternal peace” with the Commonwealth established in 1634 by the Treaty of Polianovka. There is evidence suggesting that envoys of Bohdan Khmelnytsky began traveling to Istanbul in 1648.11 However, it was only after the setback in Zboriv in August 1649, when İslam Giray forced Khmelnytsky to come to terms with the Commonwealth by withdrawing his support from battle just as the Cossacks were on the verge of defeating the Polish army, that Khmelnytsky set out in earnest to establish closer relations with the Otto- mans. His immediate goal was to appeal to the sultan to order the Crimean khan to support the hetman without hesitation. Ultimately it was hoped that the Ottomans would take an active role, at least by ordering military support from the vassal states of Moldavia, Wallachia, and Transylvania, if not by putting their own forces into the struggle with the Commonwealth, thereby assuring his chances for victory. For Khmelnytsky, however, there were even greater attractions in the Ottoman system. Once it became clear that reconciliation with the Poles was impossible, the hetman and his colleagues began to think in terms of establishing a Cossack or Ukrainian state, either independent or allied with some other state. Of course the lack of a ruling dynasty was a serious weakness as far as gaining internal and external legitimacy was concerned. The model of Moldavia and Wallachia, Orthodox and culturally akin societies with a degree of autonomy and relatively unhampered ruling elites under the protection of the Porte, was an alluring prospect for the Ukrai- nian elites. The idea of seizing or being appointed by the sultan to the

11 This has been a point of controversy. Pritsak insists that not only had Cossack envoys already visited Istanbul by 1648, but that a short-lived trade pact was then concluded with the Porte. Abrahamowicz and Hösch argue that there is insufficient evidence for such a treaty, and do not recognize any diplomatic relations in that year. Pritsak has replied to his critics and there are other, Polish sources referring to such contacts that have been pointed to (see Stepankov reference below). For us here whether relations began in 1648 or 1649 is of no consequence. See Omeljan Pritsak, “Das erste turkisch-ukrainische Bundnis (1648),” Oriens 6 (1953): 266–298, esp. 280–285. Edgar Hösch, “Der türkisch–kosakische Ver- trag von 1648,” Forschungen zur Osteuropäische Geschichte 27 (1980): 233–248; Zygmunt Abrahamowicz, “Comments on Three Letters by Khan Islam Gerey III to the Porte (1651),” Harvard Ukrainian Studies 14 (1990): 132–143, esp. 137–138; Omeljan Pritsak, “Shche raz pro soiuz Bohdana Khmel’nyts’koho z Turechchynoiu” [Once again on the union of Bohdan Khmelnytsky with Turkey], Ukrains’kyi arkheohrafichnyi shchorichnyk 2 n.s. (1993): 177–192; Valeryi Stepankov, “Mizh Moskvoiu i Stambulom: Chy isnuvala problema vyboru protek- tsiii u 1648–1654 rr.” [Between Moscow and Istanbul: Did a problem of choosing protection exist in 1648–1654?], Ukraina v Tsentral’no-Skhidnii Ievropi 4 (2004): 223–236, esp. 225–226. cossack ukraine in and out of ottoman orbit, 1648–1681 133 voievodship of Moldavia must have occurred to Khmelnytsky very early on. When the current voievod (or hospodar) of Moldavia, Vasile Lupu, got wind of Khmelnytsky’s plan, he tried to appease him by assuring him that someone of his greatness deserved to become the ruler (kniaz, “prince”) of Rus’ (Ruthenia).12 For that matter the idea of a separate Ukrainian entity gained some currency among the Tatars and Ottomans; thus the Ottoman chronicle of Na‘ima has İslam Giray in 1648 telling the Ottomans, “if God is willing, my intention is to have a Ruthenian king (Rus kıralı) appointed by the sultan, just as the Moldavian [voyvoda] is.”13 When by 1650 it became obvious that the Porte was unwilling to allow Ukraine to extend its direct influence into Moldavia, Khmelnytsky pursued a plan of marrying his eldest and most able son Tymish to the daughter of Vasile Lupu, thereby establishing a degree of dynastic legitimacy and political access to Moldavia. It was on this card that Khmelnytsky put much of his hope; he pursued it for the next two years, finally succeeding in 1652 in forcing Lupu to give the hand of his daughter Roksanda (also known as Helen, perhaps as in “Helen of Troy”) to Tymish, only to have his dynastic dreams shattered when his son fell in battle during one of his military interventions there in the following year. In any event, from fall 1648 until spring 1651 there was a frequent exchange of envoys between Chyhyryn, Khmelnytsky’s capital, and Istanbul, involving not only envoys from the sultan, but from other high officials, such as viziers and the aga of the janissaries.14 From the very beginning the hetman’s representatives received a seemingly warm welcome. As noted, the Ottomans were most pleased to have the Cossacks as a friendly rather than hostile neighboring power: throughout the correspondence the Porte reminds Khmelnytsky to keep the Cossacks away from the Black Sea. In return the Ottomans pressured the Crimean khan to remain loyal to Khmelnytsky and made vague promises of sending to Khmelnytsky’s aid “whatever number of troops” he needed. Khmelnytsky’s letters to the Porte characteristically include statements that he wishes to be, requests to be considered, or even is a subject, or more literally, slave (Turkish kul, Arabic ‘abd) of the sultan. In his letters to the hetman,

12 Hrushevsky, History of Ukraine-Rus’, vol. 8, 525. 13 “İnshallah ‘azimetüm budur ki Rus kıralını Bogdan gibi taraf-i saltanatdan nasb etdürerim.” Mustafa Na‘ima, Tarih-i Na‘ima, vol. 4 (Istanbul, 1281–1283/1864–1866), 287. 14 Pritsak, “Das erste turkisch-ukrainische Bundnis”; Rypka works cited in note 2; András Riedlmayer and Victor Ostapchuk, “Bohdan Xmel’nyc’kyj and the Porte: A Document from the Ottoman Archives,” Harvard Ukrainian Studies 8 (1984): 453–473. 134 victor ostapchuk the sultan acknowledges this but does not explicitly and unambiguously refer to him as his subject. There are some indications that though very pleased with this situation, the Ottomans were careful about or even wary of accepting the hetman and his Cossack army. In one letter dated 10 August 1650 (12 Şa‘ban 1060) there is phrase to the effect that “we are close to accepting your request for slavery.”15 We know that the Ottomans were nervous about Cossack military and political interventions in affairs of its vassal states and were opposed to the marriage of Tymish Khmelny- tsky to Lupu’s daughter. For that matter, it is difficult to judge to what degree Khmelnytsky was sincere in his declaration of being an Ottoman subject. We know that at the same time he was working to convince the Muscovite tsar to accept him into his suzerainty. As we shall see below, the hetman not only kept his options open by appealing to several sovereigns at once; he would reveal the fact of his applications to one sovereign in order to blackmail the other into acting in his favor. In any event, by March 1651, a letter from the Porte explicitly refers to the hetman as being a subject of the sultan (the document uses both the Turkish kulluk and Arabic ‘ubudiyyet, “state of slavery, servitude”), and though it comes closer to granting unambiguous subject and protected status, it stops just short of it. It states that if the hetman continues to be faithful, remains in good relations with the khan, keeps the center informed of events with frequent envoys, then he will surely be placed “under the shadow of protection” and be granted an ‘ahdname-i hümayun (“imperial letter of oath”).16 The ‘ahdname had several related usages in Ottoman diplomatics and diplomacy: a unilateral granting of peace and protection to a polity of lesser stature, and acceptance of a bilateral peace treaty to a state of essentially equal stature and status.17 Of course, given the hetmanate’s parvenu status any ‘ahdname to the hetman could only be a unilateral grant of protection by the sultan to

15 “Ricanuz kabula karin olub kulluga kabul etmişüzdür.” Rypka, “Z korespondence Výsoké Porty s Bohdanem Chmelnickým,” 346–350, 482–498, esp. 487ff. 16 Dated first decade Ramadan 1061/22 February–3 March 1651. A full-size facsimile and a translation by Joseph von Hammer-Purgstall, albeit with a mistaken conversion of the original’s Hicri date, was published in N.I. Kostomarov, “Gramota Sultana Tureckogo Moxammeda IV, Bogdanu Khmel’nitskomu i vsem voisku Zaporozskomu: V dekabre 1650” [A document of the Turkish Sultan Mohammed IV for Bohdan Khmelnytsky and the entire Zaporozhian Host. December 1650], in Pamiatniki izdannye Vremennoiu Kommisieiu dlia razroba drevnikh aktov 3 (1852): 436–440; Rypka, “Korrespondenz der Hohen Pforte mit Bohdan Chmel’nyćkyj,” 268ff. 17 Dariusz Kołodziejczyk, Ottoman–Polish Diplomatic Relations (15th–18th Century): An Annotated Edition of ‘Ahdnames and Other Documents (Leiden, 2000), 3–46, 68ff. cossack ukraine in and out of ottoman orbit, 1648–1681 135 an inferior and thus contain a clear formalization of the suzerain-vassal relationship. In late July another Ottoman envoy was sent to Chyhyryn with a sultanic letter promising much: aid from forces in Dobruca; orders to the rulers of Transylvania, Wallachia, and Moldavia to be ready for action; and the eventual issuance of an ‘ahdname.18 Despite the prospects that an Ottoman orientation would be offered to Khmelnytsky and Ukraine, by late 1653 it was clear to the hetman that the promise was not about to be realized. No military aid was forthcoming from the Ottomans and the alliance with the Tatars was unraveling, as once again the khan reached an understanding with the Polish king, leaving Khmelnytsky vulnerable. In general, it seems that the possibilities for plunder and captive-taking in the lands under the Polish Crown were nearly exhausted. With his son Tymish dead, the Danubian principalities opposed to him because of his military interventions in Moldavia in 1652, and the Porte displeased therewith, Khmelnytsky’s Ottoman system lay in shambles. As for the Ottomans, we must remember that not only was the main Ottoman army bogged down in a long war in Crete; during these years the Ottoman government was beset by some of the worst internal difficulties in the history of the empire: intrigues from various factions within the harem, as well as infighting between the janissary establishment and the court sipahis. In 1648 Sultan İbrahim was deposed and executed, and there were eight different grand viziers over six years. Given this situation, it is not surprising that the Ottomans did not engage in an active northern policy, even though the Khmelnytsky movement presented probably the best opportunity hitherto for them to move into Ukraine and expand the borders of the empire well beyond the Black Sea coast. Though a historian should avoid speculation, one can surmise that if the Ottomans really wanted to, they could have found a way to alter their commitments, either by coming to an agreement with Venice and ending the war over Crete, or by temporarily easing this commitment, and taking advantage of the weakness of the Commonwealth, the timidity of Mus- covy, and the plea from Ukraine. For example, when it was decided that Prince Rákóczi had become too independent in his dealings with Euro- pean states, in 1658 Grand Vizier Mehmed Köprulu set aside the war with Venice to march on Transylvania.19 In the early 1660s further troubles with

18 Rypka, “Dalši příspěvek,” 215–219. 19 Metin Kunt, et al., Türkiye tarihi, vol. 3: Osmanlı devleti, 1600–1908 [History of Turkey, vol. 3: Ottoman State, 1600–1908] (Istanbul, 1990), 28. 136 victor ostapchuk

Austria compelled the Ottomans again to leave the Venetian campaign on the back burner and campaign in Central Europe. But then Köprülü was a different kind of leader, and had he been in power earlier who knows what might have happened in Eastern Europe? Despite the difficulty of the moment, the Ottoman attitude vis-à-vis the northern Black Sea frontier during the Khmelnytsky period bears a striking similarity to their attitude in earlier periods: if no power became too strong, that is, strong enough to threaten their control of the Black Sea region, the Ottomans were content to stay out of the northern steppes, leaving the Crimean Khanate to tend to the Porte’s (as well as its own) interests in the region.

Ukraine between Muscovy and the Ottomans

Finally, in late 1653 Muscovy agreed to enter into a war with the Common- wealth and accept Ukraine under its protection. After years of appeals by Ukrainian politicians and churchmen to the religious and cultural affini- ties between Ukrainians and Russians, as well as threats to go over to the Turks and Tatars and go to war against Muscovy, Moscow finally agreed to a major change of course. Although appeals to protect its coreligionists against the Catholic Poles were recognized, it seems that the main factor was Moscow’s fear that if it did not join Khmelnytsky, the hetman and Ukraine would become unequivocal subjects of the Ottoman Empire.20 An equivalent danger was that the hetman would yield to the current urgings of the khan to resubmit to the Polish king and join a Crimean– Commonwealth war effort against Muscovy. The 1654 Treaty of Pereiaslav by which Bohdan Khmelnytsky accepted the protection of the Muscovite tsar greatly altered the nature and course of the struggle over Ukraine. However, because Russia eventually pre- vailed vis-à-vis Poland–Lithuania and the Ottoman Empire and became the sole ruler of Ukraine, Pereiaslav is too often seen as more of a turning point than it actually was; in other words, it is often assumed that when, in 1654, Russia entered solidly into the conflict, the tide had turned against the Poles and Ottomans. In fact, Pereiaslav was only the starting

20 Cf. a reason given in the decision of the Zemskii Sobor to accept Ukraine under the suzerainty of the tsar, 1 October 1653: “And so that not to let them go into the subjection of the Turkish sultan or Crimean khan . . .,” Vossoedinenie Ukrainy s Rossiei: Dokumenty i materialy v trekh tomakh. vol. 3: 1651–1654 gody [The reunification of Ukraine with Russia: Documents and materials in three volumes], ed. P.P. Hudzenko, et al. (Moscow, 1953), vol. 3, 414, no. 197. cossack ukraine in and out of ottoman orbit, 1648–1681 137 point for Muscovy/the Russian Empire—the road to Russian supremacy in Ukraine and in the northern Black Sea region was still be long and hard. The Ottoman reaction to the treaty is revealing. The sources on this are not abundant, but from reports of Polish and Cossack diplomatic missions to the Porte, it seems that at first the Ottomans did not believe that Khmelnytsky had really sided with the tsar and when the existence of the treaty was confirmed they were clearly displeased; at one point in late 1654 an envoy arrived in Chyhyryn ordering Khmelnytsky to end his treaty with Moscow; if he failed to do so, the sultan threatened to send all his armies against the Cossacks.21 However, if there was a strong reaction, it did not last long; it should be noted that in the Ottoman chronicles the Treaty of Pereiaslav is passed over without mention. Already by the beginning of 1655 the Ottomans adopted a much more conciliatory tone toward the Cossack envoys. Khmel- nytsky’s envoys maintained that the khan had abandoned them and he was forced to conclude a military alliance with Moscow against Ukraine’s enemies. The envoys professed the hetman’s continued allegiance to the sultan, entreating him to accept Ukraine under the sultan’s hand, while ordering the Crimeans and other Ottoman vassals not to harm Ukraine.22 From Istanbul’s point of view, as at first from Bahçesaray’s point of view too, Pereiaslav was but one of a series of Khmelnytsky’s demarches in his struggle with the Commonwealth and it saw no fundamental change in the erratic and anarchic international scene in Eastern Europe. Thus at the end of June/beginning of July 1655 the sultan issued another letter to Khmelnytsky. It is a rather extraordinary document: the Ottoman text—surviving only in copy form in the so-called Göttingen Codex—is perhaps the elusive ‘ahdname that the Porte had long refrained from issuing to the hetman. Though it does not refer to itself as an ‘ahdname, it invokes the hetman to “send according to custom an ‘ahdname” to the Porte.23 Accepting the hetman’s argument that he had been forced to

21 Hrushevsky, History of Ukraine-Rus’, vol. 9, book 2, part 1, 472. 22 E.g., G.A. Sanin, Otnosheniia Rossii i Ukrainy s Krymskim Khanstvom v seredine XVII veka [The relations of Russia and Ukraine with the Crimean Khanate in the middle of the seventeenth century] (Moscow, 1987), 104. 23 “Sen dahi ayinüñ üzere ‘ahdname gönderesin.” Rypka, “Dalši příspěvek,” 227. Note the bilateral nature of the transaction—in response to the sultan’s issuance of this (implicit?) ‘ahdname, the hetman is requested to issue his confirmatory ‘ahdname. Cf. Kołodziejczyk, Ottoman–Polish Diplomatic Relations, 4–5. Until recently there was no evidence of such a reciprocal mission as requested by the Porte. However, in 2004 Vera Chentsova discovered evidence that there was a Cossack mission to Istanbul in November or December 1655, which Jaroslav Fedoruk views as indirect evidence that Khmelnytsky attained Ottoman 138 victor ostapchuk turn to Moscow for military aid, the letter expresses joy that the hetman returned to the sultan’s protection and promises to force the Crimea to align itself again with the Cossacks. Russian historians have dismissed Bohdan Khmelnytsky’s reaffirmation of vassalage to the Porte as merely a tactic on the part of the hetman to obtain support or neutrality from the Crimea, since after all the hetman’s aides kept Moscow informed of his continued ties to the Porte. It is true that by resubmitting to the Porte in 1655 Khmelnytsky did not consciously and deliberately intend to leave the suzerainty of Moscow. However, it seems that he was, indeed, playing the two against each other, telling each side that he had allied with the other only for tactical reasons—in fact, Khmelnytsky was maintaining the Otto- man system in which he, inter alia, had operated for so many years. If indeed the Ottomans perceived no grave danger from the new Ukrai- nian–Muscovite relationship, the Crimean Khanate reacted strongly to it, and already in 1654 was closely cooperating with the Commonwealth to break up this new alignment and force the Ukrainian Cossacks to join the Tatars and Poles in an attack on Muscovy, which it considered the stronger northern neighbor once Ukraine was aligned with it. Following Khmelnytsky’s death in 1657, instability in Ukraine increased, as dissatisfaction with Muscovite policies and tensions between the upper and lower strata of Cossack society led to rebellions and struggles between pro-Commonwealth and pro-Muscovite groupings. Throughout most of the next ten years Moscow and Warsaw were at war over Belorussia and Ukraine. For brief periods the Cossacks reached rapprochements with the Tatars, but mostly the Crimea remained aligned with the Commonwealth. In 1658 the Union of Hadiach negotiated by Khmelnytsky’s successor, Hetman Ivan Vyhovsky set up a new, tripartite Commonwealth, wherein the Grand Duchy of Rus’/Ruthenia would be on equal footing with the Kingdom of Poland and Grand Duchy of Lithuania. This seemed a reasonable resolution of the decade-long conflict and an end to the multiple systems that drove the tangled international relations of the nascent hetmanate (and for that matter would have meant the end of the hetmanate as such).24 However the inertia of social, political, and military conflicts

vassalage in 1655. See his “Overcoming Stereotypes and Analyzing Ukrainian Foreign Pol- icy,” in Hrushevsky, History of Ukraine-Rus’, vol. 9, book 2, part 2, xxvii–lix, esp. xlv–xlvi. 24 See Andrzej Kamiński, “The Cossack Experiment in Szlachta Democracy in the Polish-Lithuanian Commonwealth: The Hadiach (Hadziacz) Union,” Harvard Ukrainian Studies 1 (1977): 178–197; Hadiats’ka Uniia 1658 roku, ed. P. Sokhan’, Viktor Brekhunenko et al. (Kiev, 2008). cossack ukraine in and out of ottoman orbit, 1648–1681 139 led to the fall of Vyhovsky leaving the tripartite Commonwealth stillborn. By 1660 Ukraine was effectively divided along the River Dnieper between the Commonwealth controlled west and Muscovite controlled east. In the years immediately following the death of Khmelnytsky we have no evidence of diplomatic relations between the Cossacks and the Otto- mans. This could simply be for lack of sources, but it seems as if in the early years of the Köprülü era the Porte was satisfied to leave the northern policy to the Crimea given the anarchic and indecisive course of the struggle between Moscow and Warsaw. Instead energies were poured into dealing with outstanding problems in Transylvania and the Danubian principalities, other internal problems, and above all bringing the Cretan War to a successful conclusion.

Petro Doroshenko: A Willing and Devoted Vassal of the Porte?

The de facto partition of Ukraine between Commonwealth and Muscovite halves did not last long. The Commonwealth was first to lose effective control of its half. In 1663 a rebellion broke out against its rule and by 1665 the strongly pro-Commonwealth hetman, Pavlo Tetera, was forced from the scene. In the meantime the Tatars broke with the Commonwealth and again supported the Cossacks. In late 1665, with their help Petro Dorosh- enko captured the hetmanship over Right-Bank Ukraine. Of old Cossack lineage, Hetman Doroshenko was a strong and gifted leader, intent on taking Ukraine out from under both the Commonwealth and Muscovy. Very early in his hetmanship we can see Doroshenko turning to the Otto- mans for support. By 1666 we see a renewal of the Cossack–Ottoman correspondence.25 Developments in the relations between the Common- wealth and Muscovy forced the Porte to pay more and more attention to the north. At this time the two northern powers, exhausted by perennial and inconclusive warfare, were engaged in peace negotiations that culminated in the Truce of Andrusovo in 1667. Completely ignoring Ukrainian preferences and interests, Warsaw and Moscow made the de facto partition of Ukraine along the Dnieper (with the exception of Kiev going to Muscovy) official for the next thirteen and a half years. There were also provisions in the treaty for cooperation against the Tatars and Ottomans.

25 Letter of the grand vizier from Shevval 1076/April 1666 in the Göttingen Codex, see Dorošenko and Rypka, “Dorošenko a jeho turecká politika,” 9–10. 140 victor ostapchuk

The Andrusovo Treaty was met with a very negative reaction in Ukraine, especially in the Right-Bank. At the same time not only the Tatars, but also the Ottomans reacted strongly to the arrangement. In contrast to the Treaty of Pereiaslav, Andrusovo was regarded by the Porte as a dangerous change in the international situation. With the two northern powers reaching an agreement and thereby having good prospects for establishing strong presences in their respective halves of Ukraine, the Ottomans feared they could soon fall victim to the alliance. The Ottoman chronicle of Raşid cites a letter from the Porte to the King of Poland soon after the treaty warning him to abandon what it calls “mutual assistance and union” (ta‘azud ve ittihad) with the Muscovites.26 As Andrusovo made no provisions to accommodate Doroshenko’s hetmanate, it seemed only a matter of time before he would be eliminated. Given the plight that the Commonwealth–Muscovite alliance brought to Doroshenko and the threat that it implied to Ottoman control of the Black Sea, the interests of Doroshenko and the Ottomans converged. It was the Treaty of Andrusovo that stimulated a more active Ottoman policy in this region. In 1668 the Left-Bank hetman Ivan Brukhovetsky rebelled, sending out proclamations in his realm explaining his break with Moscow: “[Mos- cow together with Poland] want to raze and plunder Ukraine, our dear homeland, and destroying all the great and lesser inhabitants, turn it into nothing.”27 Besides contacting Doroshenko, Brukhovetsky opened direct negotiations with the Crimean Khanate and the Porte to submit Left- Bank Ukraine under the Porte. However, the Ottomans, still tied up in the war with Venice over Crete, were in no position to enter into active involvement. Later that year Doroshenko and his Cossack army set out across the Dnieper with Tatar support and on 8 June 1668, after eliminating Brukhovetsky, became the sole hetman of all Ukraine, both Left- and Right-Bank (respectively east and west of the Dnieper). This achievement brought Doroshenko to the height of his power. The Ottoman attitude to this is not clear, but without any concrete military support from the Porte, despite his high degree of popularity at the time, Doroshenko was unable to hold on to the Left Bank and it soon lapsed back under Muscovite

26 Mehmed Raşid, Tarih-i Raşid [The chronicle of Raşid] (Istanbul, 1282/1865–1866), vol. 1, 138. 27 Akty, otnosiashchiesia k istorii iuzhnoi i zapadnoi Rossii [Documents concerning the history of southern and western Russia] [henceforth Akty IuZR] (St. Petersburg, 1867), vol. 7, 47. cossack ukraine in and out of ottoman orbit, 1648–1681 141

control. Certainly the Crimean Khanate seems to have been ill-disposed to his enhanced stature and power and at the time of Doroshenko’s triumph it supported a rival hetman from the Zaporozhian Sich, Petro Sukhovy (Sukhovienko). Doroshenko understood that he would not be able to survive as the sole hetman and so he returned to his earlier idea of putting Ukraine under the protectorate of the Ottomans. Apparently the Ottomans were eager to do whatever they could at the time to take advantage of such an opportunity. On 10 August 1668 envoys from the sultan arrived at Chyhyryn. Although we do not have the letter that they delivered, there were reports that the sultan offered to have Doroshenko become his subject without requiring that Ukraine pay any taxes (harac) and he promised to give him the same status as the Crimea. The only stipulation was that he allow the garrisoning of 1,000 janissaries in Chyhyryn and the same number in Kodak, the latter being a strategic fortress on the Dnieper ( just south of today’s Dnipropetrovsk and at the northern end of the nine cataracts after which the Zaporozhe region began). Doroshenko reportedly consented to these terms, with the exception that 1,000 janissaries be garrisoned only in Kodak. Doroshenko then called together his officers for a counsel in which it was decided to accept Ottoman protection on the basis of seventeen articles, the most important of which are as follows: they would always be ready to go to war against the enemies of the sultan; they would request that all Crimean, Nogay, Circassian, and Bucak (Southern Bessarabia) forces that come to the hetman’s aid be placed under his and his successors’ operational command; as a symbol of this, the sultan would grant him a mace, standard, and horse tail banner (tug, horse tail attached to a flag staff ); receipt of such symbols would not mean that the Cossacks were to be considered simple subjects or tax paying tributaries nor would their hetman be changed at will by the sultan; that Turkish and Tatar armies would cause no harm to his people, nor enslave and send them to Istanbul; Moldavian and Wallachian armies sent to aid the hetman would bring no harm and the Ukrainian clergy would receive its ordination from the patriarch of Constantinople; the hetman would continue to sit until the end of his life and not be deposed by anyone from the Porte or by any Cossacks who do not agree with him and who want to install a new hetman; that the hetman would reunite the entire Ruthenian Orthodox nation, from Peremyshl’ and Sambor in the West, to Kiev and beyond, to the Vistula and Niemen Rivers, and to Sevsk and Putivl’ in the East and liberate them from the the Poles and Moscow; in court cases between a Cossack and Turk each would have recourse to his own courts; and that 142 victor ostapchuk the sultan and khan make no agreements with other states, especially the Polish king and Muscovite tsar, without notifying the hetman.28 There is also a report that the Ottoman çavuş went to each of Dorosh- enko’s colonels one by one and closely questioned them to determine if they really wanted to be under the sultan or if Doroshenko had forced them to say this. They replied that they were not forced by the hetman to say this and that they wanted to be subjects of the sultan like the Molda- vians and Wallachians. On 24 December 1668 envoys of Doroshenko presented a letter of submission to the sultan with a request for a horse tail pole and standard.29 By June 1669 the Porte issued a patent (berat, nişan) granting Doroshenko all of Cossack Ukraine as an Ottoman sancak or province. The original is preserved in Moscow and there is a chancery copy in Istanbul. The following passages are relevant here: The possessor of this celebrated royal mandate and exhibitor of this eloquent and justly eminent imperial document (. . .) the dedicated and devoted hetman of the three Cossack peoples—the Sarı Kamış (Zaporozhians), the Barabaş (Left-Bank Cossacks) and the Potkal (Right-Bank Cossacks)30— Petro Doroshenko (. . .) sent envoys and emissaries to my prosperous court (. . .) and he offered service (hidmet) and submission (‘ubudiyyet) to our Sub- lime Porte (. . .) and he was numbered among the totality of subjects of my customary grace—Wallachia, Moldavia and others (. . .) [and] he requested to be given the horse tail pole (tug) and standard (‘alem) and banner (sancak), in order to become ruler of the regions and I gave my imperial assent to his request with the stipulation that he remain constant on the path of obedience (. . .) and I showed my favor by confirming his appointment, gave him jurisdiction over the three regions as a sancak (. . .) he should guard and defend his country, and preserve and protect the populace; he should take proper steps and measures in the provinces, and keep order and discipline among the three Cossack peoples and on occasion of campaign (. . .) he should arrive at the appointed place with his army in order and in formation (. . .) all of the populace should have recourse to his government in all necessary matters, major and minor, within his jurisdiction (. . .)31

28 Akty IuZR, vol. 8, no. 73, 218–220; see also Doroshenko, Doroshenko, 213–216. 29 “Ve yigirminci [Receb 1079] (24 December 1668) Salı gün kazak hatmanı olan Doroşenkonun dahi devlet-i ‘aliyyeye qulluq etmek üzere tug ve ‘alem ricasiyle gelen elçisi paye-i serir-i a‘laya yüz sürüb ‘arz eyledügi ‘ubudiyyet namesine müsa‘ide olındı,” Silahdar Fındıklılı Mehmed Aga, Silahdar Tarihi (Istanbul, 1928), vol. 1, 509. 30 The identification of these three names is in Pritsak, “Das erste turkisch-ukrainische Bundnis,” 293–295. 31 Original: Rossiiskii gosudarstvennyi arkhiv drevnikh aktov [Moscow], fond 89, opis 2, no. 36 (I thank Dariusz Kołodziejczyk for providing me with a copy of this document); Ottoman chancery copy: Başbakanlık Osmanlı Arşivi [Istanbul], İbülemin Hariciye 52. cossack ukraine in and out of ottoman orbit, 1648–1681 143

Important points in the berat include the broad degree of autonomy granted, the lack of a requirement to pay the harac, and the Ottoman government’s recognition of Doroshenko’s authority as applying to all of Ukraine: Right-Bank, Left-Bank, and Zaporozhia. Despite the high tone and the impression of good relations that the Ottoman chronicles and this document suggest, even in the beginning there was some tension between the desiderata of Doroshenko and that of the Porte. Aside from the fact that at the time the Crimeans were supporting a different candidate to the hetmanship, Sukhovy (against whom Doroshenko spent almost a year fighting, before finally eliminating him), there is evidence that the Porte wanted to install in the place of Doroshenko Iurii Khmelnytsky, son of Bohdan Khmelnytsky, who himself had already been hetman twice before and had proven to be weak and untalented, that is a potentially useful non-entity. The Ukrainian historian, Dmytro Doroshenko, has suggested that like the Poles and Muscovites, the Ottomans did not see it in their interest to have a man of strong will and competence in the seat of the hetman.32 The contemporary Ukrainian chronicle known as Litopys sam- ovydtsia [Eyewitness chronicle] relates that the Ottomans were not overly eager to take in the Cossacks because they had allegedly betrayed every other sovereign of theirs, and the Turks supposedly reminded the Cos- sacks that it was not they who needed the Cossacks but the Cossacks who needed them and so they had better be obedient.33 Whether or not the Ottomans really intended to install the young Khmelnytsky in place of Doroshenko, they kept him in reserve for possible installation later in a manner reminiscent of the way they traditionally kept the Crimean Khan- ate and Danubian principalities in line by maintaining a reserve of potential rival princes. The Ottomans were unable or unwilling to intervene until two years later when Jan Sobieski, together with a rival candidate for the hetmanate, Mykhailo Khanenko, carried out an offensive against Doroshenko. A correspondence recorded in the Ottoman chronicles between Istanbul and Warsaw on the subject of the Commonwealth’s attempts to unseat Dor- oshenko and take back Right-Bank Ukraine reveals some of the differing assumptions and attitudes of the two sides. Take, for example, the letter of

32 Doroshenko and Rypka, “Hejtman Petr Dorošenko,” 24. 33 Letopis’ Samovidtsa po novootkrytym spiskam [The eyewitness chronicle according to newly discovered copies], ed. Orest Levitskii (Kiev, 1878), 104. 144 victor ostapchuk

Grand Vizier Ahmed Köprülü to the Polish deputy chancellor from early 1672: (. . .) you have written saying, “The country of Ukraine (Ukraniya memleketi) is our hereditary possession and its people are our subjects (re‘aya).” But in truth when God (. . .) wills that the people of a country who are continuously in debility and rebellion be granted security and mercy, they seek the protection of a padishah of the rank of Alexander (. . .). Although the people of the Cossack dominion (kazak vilayeti) have for a long time themselves [been a separate] people (kavim), by way of accord they entered into obedience under you with guaranties and oaths and however many conditions and terms. They remained such a situation for a considerable length of time. Finally, being unable to bear the oppression and encroachment and the tyranny and torment that you have rendered them, which is contrary to the treaty and compact between you and them, the Cossack people with their possessions and their souls and so as to defend their country have in totality withdrawn their obedience to you and have taken to the sword to do battle and make war with you (. . .) and then they turned to the Crimean khan for refuge while you continued your encroachment and aggression against them—it has now been more than twenty years that they have been fighting and battling protecting their possessions and souls from you (. . .) and when they let their request for refuge and for the horse tail pole and the banner to be known to the padishah (. . .) they were accepted into the servitude of the padishah and (. . .) the horse tail pole and banner were granted to them. After they passed a period of time in such a situation you said “We plan to appoint another governor and judge to that country and the country of Ukraine is our hereditary possession.” How can this be? And to say of a people who have renounced obedience to you for so much time and have endured being in the middle of battle and war with you, “[They] are our subjects” can only be “correct” by a forced interpretation.34 This letter was a kind of declaration of war: in early summer 1672 the main Ottoman army with Sultan Mehmed IV (1648–1687) at its head marched on the strategic fortress of Kamianets’ (Kamaniçe in Turkish; Kamien- iec in Polish; today Kamianets’-Podil’s’kyi) located north of the Dniester River, in Podolia. By the beginning of September, after a nine-day bom- bardment, the fortress fell and Sultan Mehmed IV entered the city with Doroshenko at his side. While Doroshenko’s Cossacks together with the Crimean Tatars played an important role in vanguard battles ahead of the Ottoman army, the actual capture of the fortress itself was achieved by the latter. Thereafter the Ottomans, Tatars, and Cossacks went north

34 Silahdar, Tarih, vol. 1, 570–572; cf. Ukrayna memleketi in Raşid, Tarih, vol. 1, 263– 266. Both forms—Ukraniya and Ukrayna—occur in the Ottoman sources of the time. cossack ukraine in and out of ottoman orbit, 1648–1681 145 to besiege Lviv, but the siege was lifted after a ransom was promised. In October the Ottomans and Commonwealth forces reached an agreement at Buchach, where a treaty known by the same name was signed. Accord- ing to it, the province of Podolia was ceded to the Ottoman Empire, the Commonwealth agreed to pay a yearly tribute, and Right-Bank Cossack Ukraine maintained autonomy under Ottoman protection. Thereafter the Ottomans set about turning Podolia into a regular Ottoman eyalet or province known as the eyalet of Kamaniçe: a beylerbeyi and other provincial officials were appointed, and a survey of the population for tax purposes (defter-i tahrir or mufassal) was compiled. For Doroshenko this was a severe disappointment since he had hoped that the rather prosperous province of Podolia, whose peasantry was completely Ukrainian, would fall under his jurisdiction. Instead he was left with only the war-torn and depopulated Right-Bank Cossack territories. To maintain decent relations, the Ottomans relented and also granted him the southeastern Podolian city of Mohyliv as a life tenure. A scrutiny of the actions that followed the conquest of Polodia makes Ottoman motivations for moving into Ukraine clearer. While before the Poles they claimed that they were going to war to protect their new Cossack subjects of the “country of Ukraine” (Ukrayna memleketi), this was a secondary consideration. First, it cannot be denied that at this time the Commonwealth was weak and an easy, attractive prey to a sultan who had never before had the opportunity to lead a holy war, or gaza. Rather than occupying and defending the Cossack land of Doroshenko, the Otto- mans directed their thrust further west against Kamianets’, the magnificent fortress that guarded some important routes into both Ukraine and Poland. Second, control of Kamianets’ and indeed of Podolia was of great strategic importance in more than one respect. Its importance was not only due to the fact that it guarded passages into Poland and Ukraine. Metin Kunt has proposed that during this period, the Ottomans were erecting new regular provinces on their European borders so as to gain better control of older autonomous entities traditionally under rather loose control of the center. By forming the two new eyalets of Uyvar and Yanova, the principality of Transylvania would no longer be a borderland and eventually the Ottomans could eliminate its ruling establishment and turn it into a regular province. Similarly, Kunt has suggested that with the eyalet of Kamaniçe the Ottomans would be in a position to eliminate the autonomous status of Moldavia. Certainly an immediate benefit that could be expected from the eyalet of Kamaniçe was enhanced control over both 146 victor ostapchuk

Moldavia and the Crimean Khanate. Moreover, a strong presence in Pod- olia put the Ottomans into a powerful position over their subjects to the east, the Right-Bank Cossacks.35 As it turned out, war and conflict plagued most of Podolia’s existence as an Ottoman province until 1699 when it reverted to the Commonwealth in the Treaty of Karlowitz. Nevertheless, some stability was attained and during the years of a lull in conflict and with the Ottoman presence in Podolia a considerable volume of Ottoman documentation on the area was generated, the most important of which is a large mufassal tahrir defter, a full edition of which has been published by Dariusz Kołodziejczyk.36 The decision to enter Ukraine compelled the Ottomans to engage in campaigns there for seven straight years. The critical year for Ottoman rule in Ukraine with Doroshenko as their vassal was 1674. In that year Muscovy came very close to a direct military confrontation with the Ottoman Empire. Moscow was hoping that Doroshenko would betray the Turks and pledge his allegiance to Muscovy, but the Left-Bank hetman Samoilovych, fearing that he would be replaced as hetman by Doroshenko, moved against the latter, crossing the Dnieper into Right- Bank Ukraine. Muscovite forces had to follow and very soon they, along with Samoilovych, occupied much of Doroshenko’s territory. Doroshenko himself was besieged in his capital, Chyhyryn. The Ottoman army headed by the sultan himself moved into Ukraine to save Doroshenko. But before an Ottoman–Muscovite confrontation could occur, the Muscovite forces withdrew from the Right-Bank in the face of the Crimean army headed by Khan Selim Giray I (1671–1704, with interruptions). Nevertheless, the Ottoman forces moved across Doroshenko’s realm, besieging and destroying fortresses, towns, and villages that had gone over to the Muscovites. The chronicle of Silahdar details this harsh campaign in which many settlements were razed and plundered and their inhabitants killed or enslaved. The chronicle refers to these settlements as rebellious (‘isyan eden)—a harsh reaction is allowed by Islamic law

35 İ. Metin Kunt, “17. Yüzyılda Osmanlı Kuzey Politikası Üzerine Bir Yorum” [An interpretation of seventeenth-century Ottoman northern policy], Boğaziçi Üniversitesi Dergisi: Hümaniter Bilimler 4–5 (1976–1977): 111–116. 36 Dariusz Kołodziejczyk, The Ottoman Survey Register of Podolia (ca. 1681): Defter-i Mufassal-i Eyalet-i Kamaniçe (Cambridge MA, 2004). See also the same author’s mono- graphic treatment of Ottoman Podolia: Podole pod panowaniem tureckim: Ejalet Kamien- iecki 1672–1699 [Podolia under Ottoman rule: The eyalet of Kamianets’ 1672–1699] (Warsaw, 1994). cossack ukraine in and out of ottoman orbit, 1648–1681 147 against subjects (zimmis) who rise in rebellion against the Darü’l-İslam.37 Many of the harsh reprisals were perhaps due to Kara Mustafa Pasha’s well-known brutal leanings—he had led bloody seizures, for example, the capture of the towns of Ladyzhyn and Uman. In all fairness, it must be pointed out that Doroshenko himself harshly suppressed towns that had sided with the Muscovites. In any event, the harsh reaction of the Ottomans and Doroshenko meant for all practical purposes the end to a viable Cossack Ukraine as a subject of the Porte. Now the population was firmly against Doroshenko and the Turks, whereas before the people were rather well-disposed to Ottoman protection.38 The violence of 1674 caused a large migration of the population of Right-Bank Ukraine to the Left-Bank. Doroshenko himself held on for another two years, but finally, in 1676, he surrendered Chyhyryn to the Muscovites, who treated him relatively well, allowing him to live out his remaining years in far off Muscovite provinces. From the Ottoman decision to take the best part (Podolia) for themselves, one gets the impression that they were skeptical about the viability of Cossack Ukraine as a polity, even as a vassal. Its instability made them treat it as, at best, a frontier buffer zone. In reading the Ottoman chronicle accounts of their systemic destruction of strategic fortresses and palankas in Ukraine, it seems that they were deliberately trying to ensure that the region would remain a thinly populated frontier or borderland which a foe, be it Muscovy or the Commonwealth, could not re-occupy and easily establish a strong presence in. Given the geopolitical situation and internal realities in Doroshenko’s realm, the Ottomans apparently decided not to risk too much and not to place too much faith in a vassal Cossack state. Rather, they acted conservatively to assure the best defense of their nearby realms.

37 In this connection an interesting question: did the Porte consider Ottoman Ukraine as dar ül-İslam? It should be noted that no harac was imposed on this “tributary” (better vassal) territory; Silahdar at one point mentions returning back to the dar ül-İslam from Ukraine, but this refers to 1674 after the rebellions that according to Islamic law would have meant the subject territory would have reverted to dar ül-harb status. Perhaps we have here a certain inertia in labeling or reflection of the unsettled conditions in Ottoman Ukraine, Silahdar, Tarih, vol. 1, 642–643. 38 In fact there was a strong trend of philoturkism with Ottoman entry into Ukraine as the Turks were seen as able to establish order and justice, in contrast to the vicious anarchy that characterized the Ruin period until this point. 148 victor ostapchuk

Iurii Khmelnytsky and Ottoman Triumph in and Disappearance from Ukraine

The Ottomans continued to recognize some use for the Cossack vassal state and appointed Iurii Khmelnytsky as hetman in place of Doroshenko. The fall of Doroshenko brought the Ottomans into their first war with Muscovy as the presence of Muscovite forces in the Right-Bank could not be tolerated by the Porte. There is no space to go into these events in any detail here. In 1677 an Ottoman force was unable to dislodge the Musco- vites and the Left-Bank Cossacks from Chyhyryn and so in 1678 Sultan Mehmed IV led a major expedition that, thanks to Ottoman expertise in engineering, demolitions, and other aspects of siege warfare, successfully captured Chyhyryn. What was left of the fortress itself was razed to the ground. Interestingly, from an Ottoman gazaname on the Chyhyryn campaign we learn that Iurii Khmelnytsky himself asked the Ottomans to destroy Chyhyryn and requested and was allowed to move his capital to the town of Nemyriv (approximately 300 kilometers to the west). Moreover the same gazaname relates that before withdrawing at the end of the campaign season, the Ottomans made sure that the young Khmel- nytsky gained control of his Ukraniya memleketi.39 Thus here we see that the Ottomans were still thinking that some sort of Ukrainian vassal state on the Right-Bank might continue. After 1678 there were no more significant Ottoman campaigns in Ukraine and in 1681 peace was reached with Muscovy in Bahçesaray (and in 1682 confirmed in Istanbul). It is clear that the Ottomans had no interest in extending their rule beyond the Dnieper by going to war against Muscovy over the Left-Bank hetmanate, even though there had been opportunities and on several occasions Doroshenko had invited the Ottomans to undertake a joint campaign to oust his rivals in the east and unify Ukraine. In the treaty with Muscovy it was agreed that the Dnieper River would be the border although Moscow would continue to hold Kiev and the surrounding area as it had since Andrusovo. Ottoman suzerainty over the Right-Bank was recognized although it was to be a kind of demilitarized buffer zone: frontier fortresses were not to be built or rebuilt. The Zaporo- zhia, however, was to belong to neither side. Neither side had much of an appetite to go to war with the other and it was no accident that the Ottomans, even prior to the treaty of 1681, did not attempt to consolidate

39 Lubomyr Hajda, “Two Ottoman Gazanames concerning the Chyhyryn Campaign of 1678” (PhD dissertation, Harvard University, 1984), 204, 207, 246, 249. cossack ukraine in and out of ottoman orbit, 1648–1681 149 their position in Ukraine by building up a strong military presence there. A frontier buffer zone seemed to suit their interests. All in all, what we have called an Ottoman “active northern policy” in the 1670s was actually more defensive than expansionist in its goals. Developments in the north between the Commonwealth and Muscovy and Doroshenko’s invitation had left them no choice but to intervene. Aside from taking strategic Podolia, the Ottomans had no interest in conquest to the north or east. We submit that this was in line with their traditional, centuries-old Black Sea policy to do what is necessary for the security of their Black Sea dominion. In the 1680s, rather than engage in wars of conquest on the frontiers of the Commonwealth or Muscovy, they preferred a no less difficult conflict in central Europe which was an attractive object of conquest. That the Ottomans were not inclined to develop their control in Ukraine and instead left it as it was and went on to fight a probably even more dangerous war for Vienna tells us much about Otto- man goals and policy in the northern Black Sea region.

Tributaries, Vassals, Polyvassalage

For the outlined period, Ottoman–Ukrainian rapprochement was an uneven process that took place in a climate of great instability and flu- idity, involving interests that were often as contradictory as they were overlapping. Even in the Ottoman camp, both the Crimea and Moldavia preferred to have specific relationships with the hetmanate rather than accommodate Ottoman interests.40 Certainly the vagaries of the situation in the north, as well as problems elsewhere, both at home and abroad, explain why it took the Porte so many years to agree to become fully involved in Ukraine. Hence, there was so much back and forth, especially in the Khmelnytsky period, with repeated requests to be accepted into the Ottoman fold and repeated prevarications by Istanbul. So to what degree was Cossack Ukraine an Ottoman entity in this period? Since Islamic-style tribute (harac) was never imposed and scarcely discussed,41 technically

40 For lack of space we have not delved into the complexities of Crimean–Cossack relations. Suffice it to say that the khanate was inclined to view the Ukrainian Cossacks as being subject to it and to act as an intermediary between Chyhyryn and the Porte (cf. the Crimean Khanate vis-à-vis certain polities in the North Caucasus in the mid-sixteenth century). 41 E.g., according to an indirect source—a letter written in October 1648 by a Pol- ish official in Kamianets’ to one in Lviv—Khmelnytsky’s envoys promised inter alia to pay harac in the manner of Moldavia and Wallachia and supply military support to the 150 victor ostapchuk speaking, we cannot call the hetmanate an Ottoman tributary.42 This is, of course, why we have preferred the term “vassal,” of course not in the original Western medieval sense, but in the sense of the relationship between a subject state and a suzerain, a state in which there are mutual obligations—mainly non-aggression and protection of the subject by the suzerain in exchange for, when needed, military service by the subject on behalf of the suzerain, and possibly rendering tribute. The question of the degree to which Khmelnytsky was ever a formal vassal of the sultan is obscured by repeated references of his entering into or being in a state of servitude/slavery to the sultan. While in the case of Doroshenko we have both an Ottoman original and an Ottoman copy of a patent of investiture—the nişan/berat cited above—that leaves no doubt as to his formal relationship to the Porte, for Khmelnytsky the most solid document attesting formal vassal status is the sultan’s letter to him from mid-1655, also cited above. However, Na‘ima’s chronicle under the year 1063/1653 maintains that in this year Khmelnytsky was granted some of the standard symbols of investiture—the banner (‘alem) and drum (tabl), along with a patent (berat) granting him his territories as an Ottoman province (eyalet).43 The “multiple ambiguity” of Ukraine’s situation in these years was crucial in deciding the nature of relations with the Porte and, for that matter, with all the surrounding powers. The multiple systems in which Khmelnytsky operated dictated that his relations with the surrounding powers remained as fluid, murky, and as undefined as possible. Similarly, Doroshenko, even while a subject of the Ottomans, maintained contacts with Poland and Muscovy, and continually explored possibilities of rapprochement with either or both of them in the event the vicissitudes of the international situation required him to abandon his Ottoman

tune of 200,000 Cossack troops only if the sultan ordered the Tatars to support the Cos- sacks against the Poles, see Jakub Michałowski, Księga pamiętnicza [Memoirs], ed. Antoni Zygmunt Helcel (Cracow, 1864), 211. 42 While it is known that Islamic law requires a subject polity to pay tribute to an Islamic state if it is to avoid jihad being waged against it, we have not been able to confirm this as a specific stricture of the Hanafi school of law or any madhhab for that matter, and we suspect that in early stages of subjugation, for example, in the Ottoman conquest of the Balkans, principalities became Ottoman vassals without tribute and only in exchange for military support. Of course we have no way of knowing whether Cossack Ukraine would have proceeded along a similar path and eventually become subject to harac had its relationship with the Porte lasted. 43 Na‘ima, Tarih, vol. 5, 278–279. cossack ukraine in and out of ottoman orbit, 1648–1681 151 orientation.44 Recently a leading Ukrainian historian of this era, Taras Chukhlib, has characterized the relations of the Cossack Hetmanate as being one of “polyvassalage”—the various bilateral relations in which it operated allowed the hetmanate to be a vassal of several suzerain states simultaneously.45 Of course simultaneous vassalage is nothing new to this part of the world. Thus, in the sixteeenth century some of the polities of the North Caucasus managed not only to be vassals of the Crimean Khanate and/or the Ottoman Empire and Muscovy, but members of their ruling class had no problem professing Christianity and Islam before their various suzer- ains at virtually the same time.46 In the seventeenth century, Moldavia managed, in certain periods, to act as a subject of the Polish–Lithuanian Commonwealth without relinquishing Ottoman subjecthood. While the term polyvassalage may accurately describe the actual situation in all of these cases, it is problematic, as proposed, because it pertains more to a de facto situation or even a tactic of international relations rather than to a legitimate system recognized by all parties.47 Certainly there is evidence that the Ottomans, Poles, and Muscovites were often aware of Khmel- nystky’s and Doroshenko’s rapprochements with opposing parties, but usually they chose to pretend they were not aware of any dual loyalty if they could not depose them, or the hetmans were able to convince their suzerain that they were engaging with an opposite party only for tactical or military reasons. The failure of the Ukrainian-Ottoman venture was a true turning point in the history of the northern Black Sea basin. The lapse of active Otto- man involvement there meant that at this stage a more independent Ukrainian state would not come to be. For the Ottomans it eventually meant repeated direct confrontations with the Russian Empire. That the

44 Taras Chukhlib, “Hetmanuvannia Petra Doroshenka: Prychyny ‘virnosti’ ta ‘zrady’ koroliu, sultanu i tsariu (1665–1676 rr.)” [The hetmanship of Petro Doroshenko: Causes for the ‘loyalty’ and ‘treason’ toward the king, sultan, and tsar (1665–1676)], Ukrains’kyi istorychnyi zhurnal, no. 1 (2007): 39–61. 45 See Taras Chukhlib, Kozaky i monarkhy: Mizhnarodni vidnosyny rann’omodernoii derzhavy, 1648–1721 [Cossacks and monarchs: International relations of the early modern state, 1648–1721] (Kyiv, 2009). 46 See Murat Yasar, “The North Caucasus in the Second Half of the Sixteenth Century: Imperial Entanglements and Shifting Loyalties” (PhD diss., University of Toronto, 2011). 47 The term “condominium” is applied to situations in which two or more powers shared sovereignty over a territory by common agreement, as was the case in certain sectors of the Ottoman–Hungarian frontier in the sixteenth and seventeenth centuries. This term cannot be applied here given a lack of any common agreement by the suzerain powers. 152 victor ostapchuk

Ukrainian Cossacks, who were so adept in fighting in this region ended up on the Russian rather than Ottoman side in the Russo-Ottoman wars of the eighteenth century was of no mean significance. Reliance on the old defensive Black Sea policy in the second half of the seventeenth century, a policy that previously had been so appropriate given the geopolitical rea- lia of the region, meant that in the end Ottoman presence in the northern Black Sea would not last. Section two

The Diplomacy of the Tributary States in the Ottoman System

Sovereignty and Representation: Tributary States in the Seventeenth-century Diplomatic System of the Ottoman Empire*

Gábor Kármán

The question of sovereignty has haunted the Hungarian historiography on the Principality of Transylvania since the beginning of its institutionaliza- tion in the nineteenth century. On one hand, the well-known fact that the confirmation of the sultan was necessary to become the ruler of the country, that Transylvanians paid tribute to the Sublime Porte, and that their foreign policy had to be negotiated with the court in Constantinople all indicate the way the principality as a state under the “control of the Turk” was seen. On the other hand, the Ottomans’ indifference toward the domestic matters of the principality, the skillful means the princes sometimes used to counter Ottoman claims, and their relatively wide opportunities, especially in the first half of the seventeenth century, to create their own power base in the region and engage in European armed conflicts, made another interpretation possible. Several historians at the turn of the twentieth century held the view that Transylvania should be regarded as practically an independent country, with some minor obligations toward the Ottoman Empire, to which they were also connected as allies.1 Their motivation—as Vencel Bíró, one of the early analysts of the question put it—was, in several cases, a kind of national self-esteem (entirely understandable and acceptable to him), an “elevating feeling” they had when they came across data that suggested that Transylvania was not a slave to the sultan.2 The question of Transylvania’s sovereignty,

* In the course of my research I have relied on the linguistic expertise of my colleagues Heiner Grunert, Sabine Jagodzinski, and Nedim Zahirović, whom I wish to thank here. 1 A monograph in this vein was written by Gyula Vajda, Erdély viszonya a Portához és a római császárhoz mint magyar királyhoz a nemzeti fejedelemség korszakában [The relationship of Transylvania to the Porte and the emperor as king of Hungary in the epoch of the national principality] (Kolozsvár, 1891). 2 Vencel Bíró, “Erdély és a porta” [Transylvania and the Porte], Századok 57 (1923): 77. On similar motivations in other historiographies related to the history of the tributary states, see the contribution of Viorel Panaite in this volume. 156 gábor kármán however, was also important for more recent generations of historians who developed a more distanced attitude toward the past. The assessment of the anti-Habsburg wars in Hungary, led by princes of Transylvania in the first half of the seventeenth century, depends decisively upon how one sees the legal status of the principality before the Sublime Porte. Treat- ing them as wars for independence which aimed to establish a national kingdom in the place of the Habsburg rule—which was a dominant trend in the historiography of the 1960s to the 1980s—has recently been challenged with the argument (among others) that those who were supposed to establish it were themselves subjects of the sultan.3 Probably the most promising possibility of reaching a nuanced understanding of questions related to Transylvania’s sovereignty is that suggested by this volume: to see the relationship of the principality in a comparative perspective. This method was suggested in the 1920s by Vencel Bíró; he also provided valuable data for differentiating the status of Transylvania from that of Wallachia and Moldavia. However, he intentionally limited his attention to these three tributary states, excluding all others with the claim that their status was obviously different. He treated Ragusa as practically an independent country, similar in its payment of tribute to Venice, Poland or the Habsburgs; and he dismissed the Crimean Khanate as a Muslim state that proved to “have more luck” than its Christian counterparts.4 However, these assumed differences between the states that historiography calls tributaries and those regarded as independent were not necessarily strictly differentiated by the Sublime Porte. As I show in this paper, the dividing line was drawn with varying criteria in the Ottoman system of interstate relations. Moreover, I also find it profitable to broaden the circle of the entities involved in the inquiry to include non-state actors. Recently there has been an increase in studies on the Ottoman borderlands which show that the control of the center was rather more limited on the peripheries than earlier research took into account.5 The histories of different tributary states record a variety of actions taken by the pashas of Bosnia, Silistria, Temesvár, and Buda which they initiated without any prior directives from

3 See, for instance, the rather heated debate between Géza Pálffy and Sándor Őze in Zita Horváth, ed., Tanulmányok Bocskai Istvánról [Studies on István Bocskai], Publicationes universitatis Miskolcinensis, Sectio Philosophica, tom. 13, fasc. 2 (Miskolc, 2008). 4 Bíró, “Erdély és a porta,” 78–79. 5 Gábor Ágoston, “A Flexible Empire: Authority and Its Limits on the Ottoman Fron- tiers,” International Journal of Turkish Studies 9 (2003): 15–29. tributary states in the 17th-c. diplomatic system 157

Istanbul.6 These regional governors seem to have challenged the monopoly of the Ottoman state in the field of foreign policy, an unimaginable situation for anyone using the nineteenth-century concept of sovereignty. In order to grasp the mechanisms of the political system at the Sublime Porte in the seventeenth century—and the place of Transylvania in it— one has to make clear how the differences among the territories which were inside and outside of the empire were represented. Can we identify distinct boundaries among them, or should we imagine them as a con- tinuum of political entities flowing into each other without clear borders? The comparative approach has already been successful in providing important results for the study of the tributary states of the Ottoman Empire. The ‘ahdnames, fermans, and berats issued by the sultan as inauguration documents for the rulers of these political entities described—or through the absence of the topic, alluded to—the rights and obligations of the territories toward the Sublime Porte. These legal documents, with their relatively standardized formulas and scribal language, promise to be a fruitful field for comparative study, and a handful of Turcologists has recently taken up this challenge.7 Here I introduce another sphere with comparably standardized elements which make it possible to provide a systematic survey of Transylvania’s place in the political system of the Sublime Porte: the peculiarities of diplomatic representation in Constanti- nople. For my purposes, four major questions deserve particular attention. First, how were the embassies structured? What different forms and ranks

6 See, for instance, the case when support was given to a Transylvanian pretender, István Bethlen, by the pashas of Buda and Temesvár in 1636: Imre Lukinich, “Bethlen Ist- ván támadása 1636-ban” [The attack of István Bethlen in 1636], Századok 44 (1910): 98–112, 212–227. About the role of the pasha of Silistria during the Wallachian crisis of 1655, see Lidia A. Demény, Lajos Demény, and Nicolae Stoicescu, Răscoală seimenilor seu răscoală populară? 1655, Ţară Românească [A revolt of the seimens, or a popular revolt? 1655, Wallachia] (Bucharest, 1968); on the significance of the pasha of Bosnia for Ragusa, see Vesna Miović, “Beylerbey of Bosnia and Sancakbey of Herzegovina in the Diplomacy of the Dubrovnik Republic,” Dubrovnik Annals 9 (2005): 37–69. 7 Here I primarily mean the studies of Sándor Papp; in his edition of the official documents given to Transylvania, he also dedicated one chapter to the relevance of his conclusions for the question of the inauguration documents given to the two Romanian voievodates, see his Die Verleihungs-, Bekräftigungs- und Vertragsurkunden der Osmanen für Ungarn und Siebenbürgen (Vienna, 2003), 137–140; idem, “Christian Vassals on the Northwestern Border of the Ottoman Empire,” in The Turks, vol. 3. Ottomans, ed. Hasan Celal Güzel, C. Cem Oguz, and Osman Karatay (Ankara, 2002), 719–730. See also his contribution to this volume. Viorel Panaite uses comparative material in his discussion of the legal status of Moldavia and Wallachia: The Ottoman Law of War and Peace: The Ottoman Empire and Tribute Payers (Boulder, CO, 2000). See also the study of Dariusz Kołodziejczyk in this volume. 158 gábor kármán of deputies did various political entities use to represent their interests in the Ottoman capital? Second, can the embassy’s location be taken as an indicator of the prestige of the representatives housed there? Third, what was the ceremonial aspect of their presence, the formal acts connected to the entry of the diplomats to Constantinople and their audiences with the grand vizier and the sultan? Last, what conclusions can be drawn from the occasional mistreatment of diplomats? By contrasting the characteristic practices of the principality to other tributary states, to powers not dependent on the Ottoman Empire, and, where applicable, to Ottoman regional governors, my aim is to provide results relevant for a wider circle of reference, beyond the specific case of Transylvania.

Sovereignty and the Right to Send Representatives

“(. . .) There is not a more illustrious Mark of Sovereignty than the Right of sending and receiving Ambassadors.”8 As the comment of Abraham de Wicquefort, an early theoretician of diplomacy, shows, the seventeenth- century Western European international system used the diplomatic representation of its actors as an indicator of their rank in the hierarchy. Here, the emphasis is on the use of the word “ambassador,” as the Dutch diplomat made a clear distinction between two categories of diplomats and also dedicated long passages to those of the “second rank”: the agents, secre- taries, and commissioners.9 Thus, he concludes that while minor powers of limited sovereignty could send envoys, there was an elevated category of diplomats—those invested with the characteristic of repraesentatio, the claim to be treated the same way as would be due to their ruler if he were present in person—which was limited to fully sovereign rulers. The distinctions between various categories of diplomats appeared as early as in sixteenth-century writings and became a commonplace in later treatises. The practice was, however, even less clear than the rather versatile terminology of theoretical works, especially from the late seventeenth century on, when new terms were introduced in order to indicate the claims of rulers to the acknowledgment of their elevated rank; the confusion lasted until the international regulation of this field at the Congress of Vienna.10 For the

8 Abraham de Wicquefort, The Embassador and His Functions (London, 1716), 6; the original French text was published in 1681. 9 Wicquefort, The Embassador, 33–43. 10 Erich H. Markel, Die Entwicklung der diplomatischen Rangstufen (Erlangen, 1951); Charles Howard Carter, “The Ambassadors of Early Modern Europe: Patterns of Diplo- matic Representation in the Early Seventeenth Century,” in From the Renaissance to the tributary states in the 17th-c. diplomatic system 159 seventeenth century, it is impossible to establish a precise hierarchy of ranks that would have been accepted by every state—but it is certainly justified to accept Wicquefort’s idea that contemporary practice had at least two categories of diplomats and that the rank of the sender determined the category of representative he was allowed to commission to foreign places.11 At first glance, the case of the Transylvanian embassy in Constantino- ple seems to suggest that this conclusion was also valid for the Ottoman system of diplomacy.12 Two distinct groups can be identified among the diplomats of the principality: chief envoy ( főkövet) and orator (kapitiha). The task of the former was to deliver the tribute of the principality to the Sublime Porte each year, but the prince also sent them to negotiate matters of great importance. These matters could be new, unprecedented demands in the tributary relationship (such as raising the tribute of the country or asking for its military cooperation with Ottoman armies), problems concerning the legitimacy of the Transylvanian ruler (the princes’ attempts to gain recognition for their appointed successors or against a serious pretender) or even major atrocities committed by neighboring pashas. Apart from extraordinary situations—some were not allowed to leave Constantinople for years—the chief envoys usually stayed in the Ottoman capital only as long as their negotiations lasted, which, in most cases meant no more than a few weeks. Most of the chief envoys came from high-ranking aristocratic Transyl- vanian families or at least from the higher administration of the principality. If it was seen as necessary, the three estates of the Transylvanian diet could send their three chief envoys together with those of the prince, in

Counter-Reformation: Essays in Honour of Garrett Mattingly, ed. Charles H. Carter (New York, 1966), 269–295; Miloš Vec, “ ‘Technische’ gegen ‘symbolische’ Verfahrensformen? Die Normierung und Ausdifferenzierung der Gesandtenränge nach der juristischen und politischen Literatur des 18. und 19. Jahrhundert,” in Vormoderne politische Verfahren, ed. Barbara Stollberg-Rilinger (Berlin, 2001), 559–590. 11 André Krischer, “Souverenität als sozialer Status: Zur Funktion des diplomatischen Zeremoniells in der Frühen Neuzeit,” in Diplomatisches Zeremoniell in Europa und im Mitt leren Osten in der frühen Neuzeit, ed. Ralph Kauz, Giorgio Rota, and Jan Paul Niederkorn (Vienna, 2009), 1–32, especially 10–22; Barbara Stollberg-Rilinger, “Die Wissenschaft der feinen Unterschiede: Das Präzedenzrecht und die europäische Monarchien vom 16. bis zum 18. Jh.,” Majestas 10 (2002): 1–26. 12 The most important overview of the principality’s diplomacy at the Porte was written by Vencel Bíró, Erdély követei a Portán [The envoys of Transylvania at the Porte] (Kolozsvár, 1921). On the latter part of the seventeenth century, see István Molnár’s informative Rozs nyai Dávid török deák [The Turkish scribe Dávid Rozsnyai] (Budapest, 1909), 39–128. For a summary in German, see Georg Müller, Türkenherrschaft in Siebenbürgen: Verfassungs rechtliches Verhältnis Siebenbürgens zur Pforte 1541–1688 (Hermannstadt-Sibiu, 1923), 70–142. 160 gábor kármán order to show the unity of the country on a specific question.13 The higher prestige of this post was clear to both the Transylvanian and Ottoman sides. In one case, Grand Vizier Tabanıyassı Mehmed Pasha warned the recently appointed prince, György Rákóczi I (1630–1648), that his conduct was causing a serious drawback to dealing with his issues. Contrary to his predecessor, Gábor Bethlen (1613–1629), who frequently sent chief envoys, Rákóczi did not even send official letters to the Sublime Porte, but tried instead to negotiate everything through his orators, which, according to the grand vizier, was a mistake. In these circumstances he was not able to stand up for the prince’s claims in front of the sultan because he had no official proof to show the padishah concerning Rákóczi’s intentions.14 The most important task of the orators, as resident diplomats of their prince at the Sublime Porte, was to inform their lords about developments in the Ottoman capital. As is clear from the remark of Tabanıyassı Mehmed Pasha, they could not initiate negotiations at the highest level as there was no possibility of asking for audiences with the sultan. However, they were in regular contact with the grand vizier and other high dignitaries and thus were able to raise issues of minor importance or prepare the ground for the missions of the chief envoys. The orators were also the first medium through which the head of the Ottoman Empire addressed the prince; the grand vizier transmitted any new demand or question to Transylvania through the orators. As I discuss in more detail below, they also fulfilled some of the functions of a hostage, through whose person the Porte could exert pressure upon the prince. The orators also had a key role in the principality’s foreign policy, as they helped the princely court in making decisions, by providing—sometimes remarkably detailed—information about the Ottoman Empire and its environments that they collected in Constantinople (or, in most of the 1660s–1670s, from Adrianople, where the sultan’s court had moved). They were also useful in maintaining contacts with other Christian embassies at the Sublime Porte.15 The orator’s office was not popular among Transylvanians, because of its lower prestige and the high risk of the post—and also because of the sometimes rather miserable conditions they suffered in Constantinople. Their enthusiasm was also dampened by the fact that, although it was supposed to last no more than a year—the new orator always went to Constantinople

13 Bíró, Erdély követei, 1–78; Molnár, Rozsnyai, 44–49. 14 István Szalánczy to György Rákóczi I (Constantinople, 14 October 1632), in Sándor Szilágyi, ed., Levelek és okiratok I. Rákóczy György keleti összeköttetései történetéhez [Letters and documents of the history of the eastern contacts of György Rákóczi I] (Budapest, 1883), 58. 15 Bíró, Erdély követei, 79–94; Molnár, Rozsnyai, 112–122. tributary states in the 17th-c. diplomatic system 161 with the chief envoy who brought the tribute—political tensions between the principality and the Ottoman Empire often meant that diplomats were forced to stay much longer. Therefore, it often happened that the orators begged the prince in their letters to let them return home from “this hate- ful land of pagans without God, where there are no relatives, no Chris- tianity, and no mercy.”16 It should be no surprise that the persons who filled the ranks of this diplomatic corps did not come from the highest strata of Transylvanian society; most of them were of petty noble origin. The closest parallel to this duplex system of Transylvanian representation can be found in the structure of the Habsburg embassy after 1606. During the sixteenth century the rulers of the Holy Roman Empire were theoretically not allowed to station a diplomat in Constantinople permanently, although the method which they used for dealing with this prohibition actually resulted in establishing a more or less standing representation. They often sent ad hoc ambassadors who tried to stay in the Ottoman capital as long as possible. It was only with the Peace of Zsitvatorok (1606) that the possibility of establishing a permanent embassy opened up. It also prescribed that every three years each side should send a ceremonial delegation to the other; the system thus created, which included both a standing representation and ambassadors coming at regular intervals, shows the most similar characteristics with the Transylvanian practice.17 The element of extraordinary ambassadors was entirely missing from the structure of the other Western European embassies, probably because of the long distances involved; none of the embassies of France, England or the Netherlands (established in 1559, 1583, and 1612) followed the same

16 László Kubinyi to Mihály Apafi I, Constantinople, 2 October 1677, Áron Szilády and Sándor Szilágyi, eds., Török-magyarkori állam-okmánytár [A collection of state papers from the Turkish-Hungarian age] (Pest, 1871), vol. 5, 452. In any case, it was not unusual for an orator to return to Constantinople after having spent some time at home after his first year in the Ottoman capital. Occasionally, one can even find “diplomat families,” as in the case of the Szalánczys, several members of which had roles in the Transylvanian foreign policy toward the Porte. See Klára Jakó, “A Szaláncziak (Egy fejezet az erdélyi fejedelem- ség keleti diplomáciájának történetéből)” [The Szalánczis: A chapter from the history of the eastern diplomacy of the principality of Transylvania], in Emlékkönyv Imreh István születésének nyolcvanadik évfordulójára, eds. András Kiss, Gyöngy Kovács Kiss, and Ferenc Pozsony (Kolozsvár, 1999), 199–210. On the situation and life conditions of the Transylva- nian orators in Constantinople, see my article “ ‘Verdammtes Konstantinopel’: Türkenbild in der siebenbürgischen Gesandtschaft an der Hohen Pforte,” in Imperiale Koexistenz im Donauraum, ed. Norbert Spannenberger (Stuttgart, 2013, forthcoming). 17 Peter Meienberger, Johann Rudolf Schmid zum Schwarzenhorn als kaiserlicher Resident in Konstantinopel in den Jahren 1629–1643: Ein Beitrag zur Geschichte der diplomatischen Beziehungen zwischen Österreich und der Türkei in der ersten Hälfte des 17. Jahrhunderts (Bern and Frankfurt am Main, 1973), 11–14, 34–35. 162 gábor kármán pattern. Communication between the rulers of these countries and the sultans was conducted exclusively through their resident ambassadors in Constantinople.18 The motivation is less evident in the case of Venice; nevertheless, the bailo—the head of the first resident embassy at the Sub- lime Porte, whom the sultans inherited from the Byzantine Empire—also had to share his responsibilities with ad hoc ambassadors sent from the Serenissima only in extraordinary situations.19 The model used by Moldavia and Wallachia—the political entities whose status was probably closest to Transylvania—shows a somewhat different structure. There are some traces of the duplex system of a standing embassy and ad hoc envoys sent to negotiate matters of primary importance, but there is scant data about the latter; in contrast to Tran- sylvania, this method was not used regularly. The vast majority of issues were taken care of by the resident envoys (sing. capuchehaia), of whom there were at least two, but sometimes even four at the same time at the Sublime Porte—a method unknown in Transylvania.20 According to some

18 See the overview on the French and English embassies: George Frederick Abbott, Turkey, Greece and the Great Powers: A Study of Friendship and Hate (London, 1916), 17–166; and the overviews of the particular embassies of France: Gérard Tongas, Les relations de la France avec l’Empire Ottoman durant la première moitié du XVIIe siècle, et l’Ambassade à Constantinople de Philippe Harlay, Comte de Césy, 1619–1640 (Toulouse, 1942), 5–42; Michael Hochedlinger, “Die französisch–osmanische ‘Freundschaft’ 1525–1792: Element antihabs burgischer Politik, Gleichgewichtsinstrument, Prestigeunternehmung—Aufriß eines Problems,” Mitteilungen des Instituts für Österreichische Geschichtsforschung 102 (1994): 108–164; Géraud Poumarède, “La querelle du sofa: Étude sur les rapports entre gloire et diplomatie,” Histoire, économie et societé 20, no. 2 (2001): 185–197; on England, see Alfred C. Wood, A History of the Levant Company (London, 1935); and on the Netherlands, see Alex- ander H. de Groot, Ottoman Empire and the Dutch Republic: A History of the Earliest Diplo- matic Relations 1610–1630 (Leiden and Istanbul, 1978). 19 On the functioning of the Venetian embassy, see the recent publications of Eric R. Dursteler, “The Bailo in Constantinople: Crisis and Career in Venice’s Early Modern Dip- lomatic Corps,” Mediterranean Historical Review 16, no. 2 (2001): 1–30; idem, Venetians in Constantinople: Nation, Identity, and Coexistence in the Early Modern Mediterranean (Bal- timore, 2006), 23–40. 20 For overviews on the Moldavian and Wallachian representatives in Constantinople, see Ioan D. Condurachi, Soli şi agenţi ai domnilor Moldovei la Poartă in secolul al XVII- lea [Envoys and agents of the Moldavian rulers at the Porte in the seventeenth century] (Bucharest, 1920); Aurel H. Golimas, Despre capuchehăile Moldovei şi Poruncile Porţii către Moldova până la 1829: Contribuţii la cunoaşterea raporturilor de drept dintre Moldova şi Turci [About the capuchehaias of Moldavia and the orders to Moldavia before 1829: Contri- butions to the knowledge about the legal contacts between Moldavia and the Turks] (Iaşi, 1943); Viorel Panaite, “Reprezentanţa diplomatică a Ţării Româneşti la Poarta Otomană în epoca lui Constantin Brâncoveanu” [The diplomatic representation of Wallachia at the Sublime Porte in the period of Constantin Brâncoveanu], Revista de Istorie 42 (1988): 877–894; Ion Matei, Reprezentanţii diplomatici (capuchehăi) ai Ţării Româneşti la Poarta tributary states in the 17th-c. diplomatic system 163 opinions, the resident representatives of the Romanian voievodates should not be regarded as diplomatic staff, but rather as administrators whose only task was to facilitate the communication between the center of the empire and the voievods—as the latter were interpreted by the sultans not as independent rulers, but as governors of imperial provinces.21 Such provincial governors, however, had an arrangement which in practice also created a duplex system of permanent and ad hoc representatives. According to Paul Rycaut, who spent several years at the English embassy in Constantinople, the delegates of regional governors were always present at the Porte; seventeenth-century Transylvanian reports also testify that the pashas of Buda or Temesvár always sent one of their local officials to Constantinople when they wanted to raise any significant issue.22 Hypothetically, we can assume that the multiplication of Romanian capuchehaias was necessary in order to overcome the problem of the missing duplex system. If some of them commuted between Constantinople and their respective capitals they could fulfil the role of the resident representatives as well as that of the ad hoc legates—an arrangement made necessary by the close dependence of the voievodate’s political elites upon the developments at the Sublime Porte. This Moldo-Wallachian system of—at least nominally—using only resident envoys for communication is all the more noteworthy because Ragusa followed exactly the opposite method; rather surprisingly, the city- state did not have a resident embassy in Constantinople until the late seventeenth century. The only official diplomatic contact with the sultan was maintained by the two envoys sent with the tribute each year and ad hoc embassies when the need arose. It seems that the Sublime Porte did not expect Ragusa to station a resident envoy in Constantinople and the information-gathering needs of the city-state at the Ottoman capital were sufficiently fulfilled by the service of her merchants and confidants.23

Otomană [The diplomatic representatives (capuchehaias) of Wallachia at the Sublime Porte], ed. Nagy Pienaru and Tudor Teoteoi (Bucharest, 2008). 21 Condurachi, Soli, 29. 22 See, for example, the letter of István Váradi and Jakab Harsányi Nagy to György Rákóczi II (Constantinople, 22 September 1655) in Sándor Szilágyi, ed., Erdély és az északkeleti háború [Transylvania and the northeastern war] (Budapest, 1890), vol. 1, 556; István Rácz’s letter to György Rákóczi I (Consantinople, 19 May 1642) in Szilágyi, ed., Levelek és okiratok, 672. Cf. Paul Rycaut, The History of the Present State of the Ottoman Empire (London, 1686), 166. 23 Nicolaas H. Biegman, The Turco–Ragusan Relationship According to the Firmāns of Murād III (1575–1595) Extant in the State Archives of Dubrovnik (The Hague and Paris, 1967), 36–38; Vesna Miović-Perić, “Methods Used by the Ragusan Diplomacy in Its Contacts with the Ottoman Empire,” in International Symposium: Diplomacy of the Republic of Dubrovnik, 164 gábor kármán

This was an unusual solution for a Christian tributary state of the Otto- man Empire; maintaining diplomatic contacts only through non-resident ambassadors was characteristic for powers politically independent from the sultan, such as Poland–Lithuania, Muscovy or Sweden. In all these cases, resident embassies were founded only in the late seventeenth, or eighteenth century.2425

Table 1. The diplomatic representation of various political entities at the Sublime Porte, mid-seventeenth century Political entity Permanent residents Ad hoc envoys Venice √ France √ The Holy Roman Empire √ √ England √ The Netherlands √ Poland √ Muscovy √ Transylvania √ √ Moldavia √ Wallachia √ Ragusa √ Crimean Khanate24 (√) √ Cossack Hetmanate25 √ Pashas of Buda, Silistria, etc. √ √

Diplomatic Academy Year-Book, vol. 3, ed. Svjetlan Berković (Zagreb, 1998), 257–266; Vesna Miović, Dubrovačka diplomacija u Istambulu [Ragusan diplomacy in Istanbul] (Zagreb and Dubrovnik, 2003), 21–29. See also her contribution in this volume. 24 There is no overview of the Tatar khan’s diplomatic representation in Constantino- ple; however, it seems that he did not have a resident for the entire seventeenth century. For the late 1650s there are various data about an agent of his stationed in Constantinople; see Claes Rålamb, “A Relation of a Journey to Constantinople,” in Collection of Voyages and Travels, ed. John Churchill (London, 1732), vol. 5, 688; letter of Jakab Harsányi Nagy to György Rákóczi II (Constantinople, 7 July 1656), in Sándor Szilágyi, ed., Okmánytár II. Rákóczy György diplomáciai összeköttetéseihez [Documents on the diplomatic connections of György Rákóczi II], Monumenta Hungariae Historica. Ser. I. Diplomataria, no. 23 (Buda- pest, 1874), 405. In the 1630–40s, however, he must have used only ad hoc envoys; in his final relation, the Habsburg resident, Johann Rudolf Schmid, enumerated every diplomatic representative that he had seen at the Porte during his stay, including envoys from the Georgian tributaries of the Ottoman Empire and Indian ambassadors, but he does not mention a Tatar resident. Cf. Meienberger, Johann Rudolf Schmid, 223–271. 25 Although Omeljan Pritsak published an Ottoman–Cossack treaty from 1648 that ordered the establishment of a resident embassy, the authenticity of this document was later seriously questioned and no data shows that the arrangement was ever made, cf. Omeljan Pritsak, “Das erste türkisch–ukrainische Bündnis (1648),” Oriens 6 (1953): 266– 298; and Edgar Hösch, “Der türkisch–kosakische Vertrag von 1648,” Forschungen zur Osteu- ropäische Geschichte 27 (1980): 233–248. tributary states in the 17th-c. diplomatic system 165

When this information is put in a table, it becomes clear that a pattern is hard to find; there was no clear differentiation based on the prestige or relative independence of the political entities. If one inquires further and apart from the structural specificities of the embassy, also takes into account the terms used for the envoys, the picture becomes even more blurred. Rycaut declared in a rather straightforward way that: the Turks make no difference in the name between an Ambassador, Resi- dent, Agent, or any petty Messenger sent or Residing upon a publick Affair; the name Elchi, serves them to express all; though they have the name of Kapikihaya, which signifies an Agent at Court; and is commonly attributed to those who reside at the Port, for the Princes of Transilvania, Moldavia and Walachia, and also for the agents of Pasha’s.26 At first glance, his interpretation of a prestige-based distinction between the envoys of the tributaries and the independent powers seems to be valid; the titles of the Transylvanian, Moldavian, and Wallachian resident envoys in their local languages all derive from the Turkish term kapı kehayası, which was also used for the representatives of provincial governors to the Sublime Porte.27 However, it would be hard to interpret along these lines the fact that the Habsburg ambassador residing in Constanti- nople after 1606 was referred to in Ottoman documents with a synonymous version of the same term—kapı kethüdası—which was also used for the Transylvanian orators.28 This, however, did not mean that he would have enjoyed less honor than the English, French or Dutch residents, who were called elçi.29 In spite of coming from a tributary state, the Ragusan

26 Rycaut, The History, 166. 27 For the nineteenth-century use of the term, see Joseph von Hammer, Des osmani schen Reichs Staatsverfassung und Staatsverwaltung, dargestellt aus den Quellen seiner Grundgesetze (Vienna, 1815), vol. 2, 134. On the earlier period (in a more restrictive manner), see Bülent Arı, “Early Ottoman Diplomacy: Ad Hoc Period,” in Ottoman Diplomacy: Conventional or Unconventional?, ed. A. Nuri Yurdusev (Houndmills, 2004), 47. 28 See the examples from 1606, 1615, and the 1630s in Lajos Fekete, ed., Türkische Schriften aus dem Archive des Palatins Nikolaus Esterházy 1606–1645 (Budapest, 1932), 219– 220, 223, 281, 295, 317, 436. For an example from the latter part of the century, see Nedim Zahirović, “Podaci o Bosanskom ejaletu u orijentalim rukopisima Konzularne akademije u Beču” [Data about the eyalet of Bosnia in the oriental manuscripts at the consular academy of Vienna], Prilozi (Institut za Istoriju u Sarajevu) 34 (2005): 226. For the Transylvanian kapı kethüda, see Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 268; for kapı kehayası there, see Zdenka Veselá-Přenosilová, “Contribution aux rapports de la Porte Sub- lime avec la Transylvanie d’aprés les documents turcs,” Archiv Orientální 33 (1965): 571. 29 For several examples of preferences for the Habsburg resident envoys at the Porte over their French and English colleagues, see Bertold Spuler, “Die europäische Diplomatie in Konstantinopel bis zum Frieden von Belgrad (1739),” Jahrbücher für Kultur und Geschichte der Slaven 11 (1935): 202–203; Meienberger, Johann Rudolf Schmid, 61–62. On the use of elçi for the Dutch resident, see de Groot, Ottoman Empire, 121. On the English and French, see 166 gábor kármán envoys were always called elçis and never kapı kehayası, a fact that also speaks against seeing this differentiation as one based on the prestige of the office.30 The latter example points to a functional divide between the two roles: kapı kehayası/kethüdası would mean a resident envoy or ambassador, whereas an elçi would be an ad hoc delegate who returned to his ruler after having accomplished his specific mission. The fact that during the seventeenth century, the Transylvanian chief envoys—in contrast to the resident orators—were called elçi in Turkish documents also supports this conclusion.31 In a similar vein, the ‘ahdname given to Imre Thököly, the founder of a short-lived tributary state in Northern Hungary, also speaks of his kapı kehayası at the Porte as well as his elçi to be sent there.32 How- ever, the custom of using the term elçi for Western European envoys contradicts the functional interpretation of the word. All in all, we can conclude that even if the Ottoman terminology referring to the various categories of diplomats was far from the simple system described by Rycaut, it still did not come close to the chaotic Western European situation. The primarily functional divide between the two main groups was not observed consistently, mainly for reasons of prestige.33

Susan A. Skilliter: William Harborne and the Trade with Turkey 1578–1582: A Documentary Study of the First Anglo–Ottoman Relations (London and Oxford, 1977), 240, 249. 30 Mladen Glavina, “An Overview of the Formation and Functioning of the Institute of Capitulations in the Ottoman Empire and the 1604 Dubrovnik Capitulation,” Prilozi za Orijentalnu Filologiju 58 (2008): 155. 31 From 1605, see Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 268. From the 1650s, see Franz Babinger, “Zwei türkische Schutzbriefe für Georg II. Rákóczi, Fürsten von Siebenbürgen, aus den Jahre 1649,” Le Monde Oriental 14 (1920): 121; Jan Rypka, “Die türkischen Schutzbriefe für Georg II Rákóczi, Fürsten von Siebenbürgen, aus dem Jahre 1649,” Der Islam 18 (1929): 216. From the 1670–80s, see Zdenka Veselá-Přenosilová, “Zur Korrespondenz der Hohen Pforte mit Siebenbürgen (1676–1679),” Archiv Orientální 26 (1958): 594–596; eadem, “Contribution,” 563–571. The data refer both to chief envoys arriving at the Porte with the tribute, and those delivering important letters from the prince. 32 Sándor Papp, “Thököly Imrének és Közép-Magyarország népének kiállított szerződéslevél” [The treaty given to Imre Thököly and the people of Middle Hungary], in István Seres, Thököly Imre és Törökország / İmre Thököly ve Türkiye (Budapest, 2006), 278– 289. Another copy of the document was edited by Zdenka Veselá-Přenosilová, “Quelques chartes turques concernant la correspondence de la Porte Sublime avec Imre Thököly,” Archiv Orientální 29 (1961): 556. It seems, nevertheless, that during the short existence of the state of Orta Maçar (1682–1685), no standing embassy was established at the Porte, cf. Zsolt Trócsányi, Teleki Mihály (Erdély és a kurucmozgalom 1690-ig) [Mihály Teleki: Transyl- vania and the Kuruc movement until 1690] (Budapest, 1972), 276, 286. 33 This double character of the division between elçi and kapı kehayası is also clear from the seventeenth-century vocabulary of François Mesgnien-Meninski, who translated “kapu ketchuda vulg. kapy kihaja” as “Residens, Agens Principis alicujus apud alterum,” tributary states in the 17th-c. diplomatic system 167

This is also clear from the use of alternative terms; whereas some had a functional meaning, such as the balyos reserved only for the Venetian envoy and vekil for plenipotentiary delegates for negotiations, others had a clear evaluative character.34 Before the seventeenth century, Ottoman documents referred to the chief envoys of Transylvania simply with the term adam (man, that is, “a man of the prince”). The first appearance of the word elçi can be connected to the re-establishment of their control over the principality after the Long Turkish War at the end of the sixteenth century—which surely mirrors an upgrade of Transylvania’s prestige, a concession by the Ottomans.35 Nevertheless, due to the fragmented character of the typology and terminology of the representation of foreign powers at the Sublime Porte, it is impossible to specifically define the scale of Transylvanian sovereignty based solely on these qualities.

Foreign Representatives in the Constantinopolitan Space

Urban space is always heterogeneous and specific meanings and values are attributed to specific locations. Thus, mapping the topography of the diplomats’ living quarters is a promising way to measure the degree

thereby giving both the functional and the prestige-based interpretation, see his Thesaurus linguarum orientalium turcicæ, arabicæ, persicæ. . . (Vienna, 1680), col. 3621. 34 Bertold Spuler, “Bālyōs,” in Encyclopaedia of Islam, second edition, ed. H.A.R. Gibb, et al. (Leiden, 1986), vol. 1, 1008; see also in form of baylos / bailos in several ‘ahdnames given to Venice: Hans Peter Alexander Theunissen, “Ottoman–Venetian Diplomat- ics: The ‘Ahd-names: The Historical Background and the Development of a Category of Political-Commercial Instruments Together with an Annotated Edition of a Corpus of Relevant Documents,” Electronic Journal of Oriental Studies 1, no. 2 (1998), 478–479, 481– 483, 620–624, and so on. Vekil was used in this vein in the ‘ahdname given to the Polish king, Michael, in 1672; see Dariusz Kołodziejczyk, Ottoman–Polish Diplomatic Relations (15th–18th Century): An Annotated Edition of ‘Ahdnames and Other Documents (Leiden, 2000), 502. In the Transylvanian context, the term was used for the envoys sent by the three nations of the country to the Sublime Porte for negotiations; see Veselá-Přenosilová, “Contribution,” 569–571. 35 Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 268. On the use of adam, see ibid., 178, 215, 228. In a similar vein, the envoys of Transylvania, Moldavia, and Wallachia are called elçis in a document from 1599, in spite of the fact that the envoys of the two Romanian voievodates were usually referred to as merdüman (men). The change is most probably due to their mission’s importance and the attempts of the Porte to make these countries return to the faith of the sultan. See Mihai Maxim, “New Turkish Docu- ments Concerning Michael the Brave and His Time,” in idem, L’Empire Ottoman au nord du Danube et l’autonomie des Principautés Roumaines au XVIe siècle (Istanbul, 1999), 151; and for the interpretation, idem, “Nouveaux documents turcs sur les cadeaux protocolaires (peşkeş) et les insignes du pouvoir (hükümet alâmetleri),” in idem, Romano-Ottomanica: Essays and Documents from the Turkish Archives (Istanbul, 2001), 84. 168 gábor kármán of prestige they enjoyed. Unfortunately, the location of diplomats’ residences as an analytical attribute show many similarities to their typology and the related terminological questions; at first glance, the method offers clear-cut results, but these become less distinct upon a closer examination of the data. The residence of the Transylvanian orators was in Balat, a quarter with a mainly Jewish population, situated at the northwestern corner of Con- stantinople between the walls of the city and Fener, the Greek quarter of the city. Its precise location has not been simple to identify. In contrast to the Moldavian and Wallachian embassies, whose churches maintained the names of Vlah Sarayı and Boğdan Sarayı even after diplomatic representatives abandoned the building, the name Erdel Sarayı was forgotten after the turn of the eighteenth century, when the principality was incorporated into the Habsburg Empire and ceased to have a foreign policy of its own. It was the pioneering Hungarian Turcologist, Imre Karácson, who found the location on a small street called Macarlar Yokuşu Sokak; this he based on the recollections of local people and by comparing the site to early modern descriptions.36 It was here that the rather simple two-story building must have stood which provided housing for the Tran- sylvanian orators and chief envoys.37 According to descriptions in diplomatic reports, it had stalls on the lower story and rooms on the upper one. Although in the 1650s, on the occasion of repair work, an attempt was made to enhance the representative character of the rooms (at least those belonging to the chief envoy), the house generally must not have made a captivating impression. It was not only the consequence of the lack of funds for renovation—which was a frequent complaint in the orators’ reports—but also of the negligence of some diplomats, who did not expend much effort to maintain the condition of the premises. A report from the 1630s noted that so much horse-dung had accumulated in the courtyard during the service of the previous orator that there was no longer room for any horse other than the chief envoy’s.38 Ironically, not long

36 Imre Karácson, “Az erdélyi ház Konstantinápolyban” [The Transylvanian house in Constantinople], Budapesti Hírlap 27, no. 125 (1907), 31. On the churches of Blach Sarayı and Bogdan Sarayı, see Wolfgang Müller-Wiener, Bildlexikon zur Topographie Istanbuls: Byzantion, Konstantinopolis, Istanbul bis zum Beginn des 17. Jh. (Tübingen, 1977), 102, 108. 37 The scattered data about the Erdel Sarayı were collected by Sándor Mika, “Az erdélyi ház Konstantinápolyban” [The Transylvanian house in Constantinople], Budapesti Szemle 130 (1907): 1–21. See also Bíró, Erdély követei, 106–112. 38 See the letter from Boldizsár Sebessy to György Rákóczi I (Constantinople, 21 Febru- ary 1636) Szilágyi, ed., Levelek és okiratok, 345. tributary states in the 17th-c. diplomatic system 169 after the major renovation in the 1650s, the Erdel Sarayı remained empty for an extended period because the orators followed the sultan’s court when it moved to Adrianople. During the 1670s an exiled Transylvanian pretender, Miklós Zólyomi, took shelter in the abandoned house.39 The Vlah Sarayı and the Boğdan Sarayı were also located in the same area, Fener and Balat, respectively. The former, the house of the Walla- chian kapı kehayasıs, is mentioned for the first time in a document from the late seventeenth century, but the source suggests that it had existed as early as the sixteenth century. The existence of the Moldavian house is documented from the mid-sixteenth century, from approximately the period when the embassy would have been established.40 In contrast, the still-functioning western European embassy buildings in Galata and Pera, on the northern shores of the Golden Horn, are evidence that most of the foreign ambassadors established their residences there. In the mid- sixteenth century the Venetian embassy was the first to move to Vigne di Pera and place the activities of the bailo in the vineyards outside the walls of the former Genoese colony of Galata. This provided much more comfortable circumstances than the busy and crowded merchant settlement and more liberty than their former residence in Constantinople proper.41 The British embassy moved there in the late sixteenth century and was joined before the third decade of the seventeenth century by the French and Dutch representatives.42 These data suggest that there was a geographical separation between the representatives of the tributaries and those independent of the Ottoman Empire.43 If one takes into consideration that the Erdel Sarayı was given to the Transylvanians by the sultan, the situation can even be seen as a result of a conscious Ottoman decision.44

39 Mika, “Az erdélyi ház,” 17; Molnár, Rozsnyai, 123–125. 40 Golimas, Despre capuchehăile Moldovei, 99–107; Matei, Reprezentanţii diplomatici, 200–204. 41 Tommaso Bertelè, Il palazzo degli ambasciatori di Venezia a Constantinopoli e le sue antiche memorie (Bologna, 1932), 81–87; Dursteler, Venetians, 25–26. 42 On the English, see Wood, A History, 238; on the French, see Spuler, “Die europäische Diplomatie,” 198; on the Dutch, see de Groot, Ottoman Empire, 191, 323n4; Marlies Hoen- kamp-Mazgon, Palais de Hollande in İstanbul: Het ambassadegebouw en zĳn bewoners sinds 1612 [The embassy building and its inhabitants since 1612] (Amsterdam, 2002), 5–40. 43 This model is suggested, among others, by Dóra Kerekes, Diplomaták és kémek Kon- stantinápolyban [Diplomats and spies in Constantinople] (Budapest, 2010), 30–31. 44 On the sultan’s donation of the Erdel Sarayı to Transylvania, see Bíró, Erdély követei, 106; Mika, “Az erdélyi ház,” 2. 170 gábor kármán C onstantinople M ap 1. E mbassy buildings in seventeenth-century tributary states in the 17th-c. diplomatic system 171

Among the tributaries, it is again Ragusa that contradicts this general model. The magistrates of this city-state made a point of not buying a house for their representatives in Constantinople, rather they instructed their diplomats to find shelter in cheap accommodations and live in poor conditions during their missions, in order to support the discourse of poverty Ragusa so readily used toward its Ottoman overlord. They chose lodgings either in Fener or in the center of the city, sometimes in Pera; their house was only built there in the eighteenth century, when a Ragusan consul resided permanently in Constantinople.45 Placing the Habsburg embassy into this theoretical system of segregation between the tributaries and independent powers is even more problematic. During the sixteenth century, ambassadors of the emperor were usually placed in the Elçi Han, by the so-called Tavuk Pazarı (chicken market) in central Constantinople. The building, as its name suggests, was dedicated to housing the ambassadors who came to Constantinople— according to some reports, in the sixteenth century it also provided lodgings for envoys from Transylvania, Moldavia, and Wallachia. It seems that although the building—which was eventually named Nemçe Hanı, the han of the Germans—was large, it was also uncomfortable, and its central location did not provide the privacy necessary for fulfilling the tasks of a standing representative.46 Therefore, after 1606, the Habsburg residents looked for another residence, and for most of the seventeenth century, they chose houses in Fener.47 A variety of ambassadors from other powers, such as Poland and Sweden, are also reported to have found shelter on the southern shores of the Golden Horn. Many were actually housed in the Boğdan Sarayı, provoking a debate among later historians on whether the Porte had confiscated the building and taken it away from the

45 See the article of Vesna Miović in this volume and Miović, Dubrovačka diplomacija, 214–215. In the late sixteenth century, Fener was preferred (see Biegman, The Turco–Ragusan Relationship, 38). 46 On the Elçi Han, see Karl Teply, Kaiserliche Gesandtschaften ans Goldene Horn (Stutt‑ gart, 1969), 208–217; Ernst D. Petritsch, “Zeremoniell bei Empfängen habsburgischer Gesandtschaften in Konstantinopel,” in Diplomatisches Zeremoniell in Europa und im Mitt leren Osten in der Frühen Neuzeit, ed. Ralph Knauz, Giorgio Rota, and Jan Paul Niederkorn (Vienna, 2009), 309; Ferenc Luttor, “Adalékok az eldsi-hanhoz” [Contribution about the Elçi Han], Túrán 3 (1918): 286–288. 47 Meienberger, Johann Rudolf Schmid, 53–55; Spuler, “Die europäische Diplomatie,” 197–198. 172 gábor kármán

Moldavian capuchehaias, or whether the name actually referred to a complex consisting of two buildings.48 Thus, contrary to the case of the terminology concerning the diplomats, in the case of the topography of their residences it is clear that the differences were not based on prestige, but reflected a functional separation. Rather than searching for a hierarchy among the various places, it is more convincing to argue that the decisive element was whether or not the state’s interests in maintaining contacts with the Ottomans were primarily of an economic nature. The diplomats who represented more or less numerous merchant communities preferred to settle closer to the merchants than to the center of the Ottoman government. The latter was more vital for the representatives of political entities whose relations with the Ottoman Empire focused almost entirely on settling political questions.49 Thus, the location of their lodgings was not an indicator that the Otto- man authorities disadvantaged the diplomats of tributary states. There was, however, an element related to their housing which shows their inferior position toward the envoys of independent powers. Even if the embassy residences in the vineyards of Pera were far from the elaborate palaces built in the eighteenth century, they were suitable to house recep- tions and banquets given by the ambassadors. Even the more humble lodgings of the Habsburg resident in Fener was home to a number of festivities.50 There is no trace that Transylvanian envoys would have ever participated in any of these or organized one themselves. Of the tributary states, only Ragusa might have allowed itself the cost of such an event,

48 Golimas, Despre capuchehăile Moldovei,104; Matei, Reprezentanţii diplomatici, 201–202. Unfortunately, it is not clear where the residence of the Polish resident embassy was after its establishment in 1678, but they did not move to Pera before the mid-eighteenth century, see Jan Reychman, “Rzekoma siedziba ambasady dawnej Rzeczypospolitej w Stam- bule” [The alleged headquarters of the old embassy of the Rzeczpospolita in Istanbul], Rocznik Orientalistyczny 17 (1951–52): 399–414. 49 This is probably also how to understand Peter Meienberger’s differentiation between “the neighboring states” and the other group (without further specification), who, according to him, differed in “their manner of interstate relations,” see Meienberger, Johann Rudolf Schmid, 52. 50 Dursteler, Venetians, 26; de Groot, Ottoman Empire, 223; Meienberger, Johann Rudolf Schmid, 63; Laura Coulter, “An Examination of the Status and Activities of the English Ambassadors to the Ottoman Porte in the Late Sixteenth and Early Seventeenth Centu- ries,” Revue des Études Sud-Est Européennes 28 (1990), 85. A seventeenth-century Turkish album gives an impression of the exterior of the bailo’s residence, see Franz Taeschner, Alt-Stambuler Hof- und Volksleben: Ein türkisches Miniaturenalbum aus dem 17. Jahrhundert (Hannover, 1925), no. 48. tributary states in the 17th-c. diplomatic system 173 but the magistrates of the city-state, as noted earlier, were motivated not to acknowledge their wealth in front of the sultan. Thus, a prestige-based segregation was indeed created among the representatives at the Sublime Porte, but it was not geographical, nor was it created by the Ottomans, rather it was based on the hierarchy of the Christian states as seen by their diplomats in Constantinople.

Ceremonies at the Sublime Porte

Diplomatic representation as a sphere of symbolic politics enjoys a renewed interest in historiography today. Due to the influence of anthropology on the discipline, historians have tended to abandon the old attitude of treating ceremonies as a quite unnecessary and sometimes ridiculous by-product of Realpolitik. Diplomats parading through the streets of friendly, or even hostile, powers, debates about the correct order of seating or signing a treaty are considered of interest in their own right, as important battlefields in the power struggle between states. After the fall of the Respublica Christiana, in the process of the formation of the early modern international system, a new hierarchy of states and rulers was formed. Ceremo- nies had to mirror the actual place of a single entity in this hierarchy, but they were also a means of challenging it.51 The situation was no different at the Sublime Porte; the diplomats in Constantinople could be sure that their reception was indicative of the Ottoman support they enjoyed—or lack of it. It is clear from the instructions of the princes of Transylvania to their ambassadors that they were aware of the implications of diplomatic ceremony. To quote Prince György Rákóczi I: As our honor as a ruler partly consists of the way the entry of Your Lord- ship to Constantinople is received, and the Christian orators there can judge

51 William J. Roosen, “Early Modern Diplomatic Ceremonial: A Systems Approach,” Journal of Modern History 52 (1980): 452–476; Barbara Stollberg-Rilinger, “Zeremoniell als politisches Verfahren: Rangordnung und Rangstreit als Strukturmerkmale des früh- neuzeitlichen Reichstags,” in Neue Studien zur frühneuzeitlichen Reichsgeschichte, ed. Johannes Kunisch (Berlin, 1997), 91–132; eadem, “Höfische Öffentlichkeit: Zur zeremoni- ellen Selbstdarstellung des brandenburgischen Hofes vor dem europäischen Publikum,” Forschungen zur Brandenburgischen und Preußischen Geschichte n.s. 7 (1997), 144–176; Axel Gotthard, “Die Inszenierung der kurfürstlichen Präeminenz: Eine Analyse unter Erprobung systemtheoretischer Kategorien,” and Helmut Neuhaus, “Der Streit um den richtigen Platz: Ein Beitrag zu reichsständischen Verfahrensformen in der Frühen Neuzeit,” both in Vor- moderne politische Verfahren, ed. Barbara Stollberg-Rilinger (Berlin, 2001), 303–332, and 281–302. 174 gábor kármán

the attitude of the Porte toward us accordingly, so Your Lordship has to take special care that your reception and welcome would be solemn and honorable.52 As a rule, the entry of the chief ambassador of Transylvania into Constan- tinople was staged in a ceremonial form. The orator residing at the Porte met the chief envoy and his entourage at Küçükçekmece, four hours from the imperial city, where, after having prepared themselves, they started their entry. They were met halfway, at Davutpaşa, by an Ottoman retinue. In the first half of the seventeenth century, a high-ranking Ottoman dignitary, the çavuş başı, came to join this procession, but this custom was dropped after 1639.53 Even later, however, a group of çavuşes came to accompany the Transylvanian diplomat on the way to his lodgings. Their number varied, in most cases between twelve and eighteen, and the chief envoys were usually convinced that it served as an indicator of the Porte’s attitude toward their mission. Even in 1657, when the grand vizier greatly resented the policies of Prince György Rákóczi II (1648–60) (especially his Polish campaign), his ambassador was accorded fourteen çavuşes for his entry, thereby giving satisfaction and—as it turned out, false—relief to the diplomat.54 Some days after the ceremonial entry to the city, a series of audiences took place. First, a ceremonial audience was held with the grand vizier, which was limited to conveying the greetings of the prince. This was followed by a private audience, sometimes in the divan, where actual negotiations could take place. At this latter event, the date of the chief envoy’s

52 Instructions of György Rákóczi I to Chief Envoy István Serédy (Gyulafehérvár, 4 October 1641) Áron Szilády and Sándor Szilágyi, eds., Török-magyarkori állam-okmánytár [A collection of state papers from the Turkish-Hungarian age], vol. 3 (Pest, 1870) (henceforth TMÁO III), 82. 53 After his return to Transylvania in 1638, István Szalánczy could still report to György Rákóczi I (Gyulafehérvár, 16 May 1638) that “the çavuş başı personally came to meet me with numerous chief çavuşes to the grove under the gardens of Davutpaşa, where the envoy of any emperor and king is usually welcomed, they would not come further for anyone, and would not even come this far for everyone.” Antal Beke and Samu Barabás, eds., I. Rákóczy György és a Porta: Levelek és okiratok [György Rákóczi I and the Porte: Letters and documents] (Budapest, 1888), 463. Eight years later, he had to let the prince know that since the 1639 embassy the çavuş başı would not come to meet the Transylva- nian chief envoys (Constantinople, 2 January 1646), ibid., 786. On the participation of the çavuş başı at the ceremonial entries of the Habsburg ambassadors, see Teply, Kaiserliche Gesandtschaften, 175–178. 54 Letter of Ferenc Thorday to Ákos Barcsai (Constantinople, 4 June 1657) Sándor Szilá- gyi, ed., Erdély és az északkeleti háború [Transylvania and the northeastern war] (Budapest, 1891), vol. 2, 323. On the entry of the chief envoy into Constantinople in general, see Bíró, Erdély követei, 24–26. tributary states in the 17th-c. diplomatic system 175 audience with the sultan was also fixed, which was the ceremonial peak of his mission. Chief envoys from Transylvania were often taken to the Saray on the day when the janissary troops were paid in order to witness this presentation of the Ottoman army’s might. Entering through the First Court of the Saray, they were led to the Second Court, which housed the main administration of the empire. After courtesies at the divan, the envoys participated in a festive meal with the grand vizier. The audience with the ruler himself took place in the Chamber of Petitions, at the entrance of the Third Court, the part of the Saray secluded as the private space of the sultans. Before entering the room, the chief envoys, the orators, and their retinues received honorary robes (kaftans); then they were led to greet the sultan and to hand over their gifts as well as the tribute of Transylvania, if that was the purpose of the mission. As with the first audience with the grand vizier, this was not an occasion designed for negotiations; the chief envoy was only allowed to give a greeting oration and deliver the letter of his prince to the grand vizier to be placed in front of the sultan. In many cases, the sultan did not say a single word; but even if he did, his communication was almost always limited to short questions addressed to the diplomats. After this ceremony, the chief envoys could return to their lodgings.55 In general terms, the audiences granted to the chief envoys of Tran- sylvania did not differ much from those staged for other foreign ambassadors. Even some of the characteristic points which seem to suggest the low status of the Transylvanian envoys—such as the silence of the sultan—turn out to have been generally the same. As a matter of fact, Transylvanian sources report that the ruler talked to the diplomats more frequently than was customary in the seventeenth century.56 None of the

55 On the progress of the audiences, see Bíró, Erdély követei, 28–52. On the ceremonial function of the various buildings in the Saray, see Gülru Necipoğlu, Architecture, Ceremo- nial and Power: The Topkapı Palace in the Fifteenth and Sixteenth Centuries (New York and Cambridge, MA, 1991). 56 Maria Pia Pedani, “The Sultan and the Venetian Bailo: Ceremonial Diplomatic Pro- tocol in Istanbul,” in Diplomatisches Zeremoniell in Europa und im Mittleren Osten in der Frühen Neuzeit, ed. Ralph Knauz, Giorgio Rota, and Jan Paul Niederkorn (Vienna, 2009), 296–297; Necipoğlu, Architecture, 101–103. On the roots of the sultan’s silence in the sixteenth century, see Konrad Dilger, Untersuchungen zur Geschichte des Osmanischen Hof zeremoniells im 15. und 16. Jahrhundert (Munich, 1967), 82–87; Pál Fodor, “Sultan, Imperial Council, Grand Vizier: Changes in the Ottoman Ruling Elite and the Formation of the Grand Vizieral Telhis,” Acta Orientalia Academiae Scientiarum Hungaricae 47 (1994): 80. The credibility of the reports of some seventeenth-century Polish ambassadors about having long conversations with the sultan have recently been questioned by Dariusz Kołodziejczyk, “Semiotics of Behavior in Early Modern Diplomacy: Polish Embassies in 176 gábor kármán important details of the ceremony for foreign ambassadors were missing in the case of the chief envoys of Transylvania. This does not mean, however, that there were no differences. Some of these were of quantity. The number of çavuşes sent for the ceremonial entry and also for the procession—even in the cases when the Transylva- nian envoys were pleased by it—was much lower than for the ambassadors coming from independent and more powerful rulers. Some ambassadors themselves helped create the spectacle offered to the population of Con- stantinople; the diplomats of Poland–Lithuania, for instance, generally came with a retinue of several hundred men. In other cases, however, such as that of the Dutch ambassador, Cornelis Haga, the Ottoman government made sure that they would make an impact, as he was given a hundred çavuşes—in contrast to the much smaller scale of the Transyl- vanian processions.57 Also, the number of kaftans given to the members of the embassy differed in the various cases, and the Transylvanians, as a rule, were given fewer than their Western colleagues. In 1666, Dávid Rozsnyai noted in his diary that they had an audience immediately after the French delegation, whom they could see were given fifteen or sixteen kaftans; the Transylvanians only received six. Rozsnyai notes bitterly that this was very few, but he does not compare the number to the Western envoys—it was obvious that they should get more—rather, he compares it to the twelve kaftans usually given to Transylvanian diplomats. Thus, even the number which would have made the Transylvanians content was only three-quarters of the number given to the French embassy.58

Istanbul and Bahçesaray,” Journal of Early Modern History 7 (2003): 245–252, and Tetiana Grygorieva, “Symbols and Perceptions of Diplomatic Ceremony: Ambassadors of the Polish–Lithuanian Commonwealth in Istanbul,” in Kommunikation durch symbolische Akte: Religiöse Heterogenität und politische Herrschaft in Polen–Litauen, ed. Yvonne Kleinmann (Stuttgart, 2010), 129–130. 57 “A True Declaration of the Arrival of Cornelius Haga (with Others that Accompanied him) Ambassador for the General States of the United Netherlands at the Great City of Constantinople,” in The Harleian Miscellany; Or, a Collection of Scarce, Curious, and Enter- taining Pamphlets and Tracts, as well in Manuscript as in Print, Found in the Late Earl of Oxford’s Library, Interspersed with Historical, Political, and Critical Notes (London, 1809), vol. 3, 214. On the size of the Polish embassies, see Dariusz Kołodziejczyk, “Polish Embas- sies in Istanbul or How to Sponge on Your Host Without Losing Your Self-Esteem,” in The Illuminated Table, the Prosperous House: Food and Shelter in Ottoman Material Culture, ed. Suraiya Faroqhi and Christoph K. Neumann (Würzburg, 2003), 51–52. 58 Dávid Rozsnyai’ diary, entry 2 February 1666, in Sándor Szilágyi, ed., Rozsnyay Dávid, az utolsó török deák történeti maradványai [The historical monuments of the last Turkish scribe, Dávid Rozsnyai] (Budapest, 1867), 298. tributary states in the 17th-c. diplomatic system 177

Transylvanian diplomats nevertheless did everything in their power to avoid the introduction of further quantitative restrictions on their audiences. An instruction of Prince Gábor Bethlen from 1628 states that in the previous year his envoy had reported an outrageous phenomenon; apart from the chief envoy and the two orators—that is, the one from the previous year and the one for the coming year—no one from the embassy was allowed to enter the Chamber of Petitions.59 This was certainly contrary to the custom, as there were usually many more people in attendance at even this last phase of the ceremonies. In the retinue of eleven people joining the above-mentioned Cornelis Haga in 1621, there were even some Dutch merchants who had little to do with the embassy itself and only happened to be in Constantinople at that time. The insult was so obvious that Bethlen ordered his chief envoy to protest, and even to threaten the grand vizier that if the old custom were not reinstated, Transylvania would not pay tribute the next year.60 Such threats might have been effective on some occasions—later data shows that the subsequent audience with the sultan was not restricted to the three most important members of the embassy—but in other cases Transylvanian diplomats had to accept even qualitative changes the Otto- man administration instituted in ceremonies. After 1635, the Sublime Porte gave up the practice of granting the chief envoy a second audience with the sultan before his departure.61 In contrast to Ragusa, where the Senate continued to fight against a similar arrangement until the early eighteenth century, the chief envoys of Transylvania had to be content with only one audience from that time on—which in most cases meant that after submitting the tribute they only had to wait until the grand vizier met them once more and gave them official answers for the prince before they set out on the journey home. This custom created a serious gap between the treatment of the Transylvanian diplomats and that of other foreign embassies who always had a last audience before departing and considered it a serious insult if it was not granted to them.62

59 Instruction of Gábor Bethlen to Mihály Tholdalagi (Fogaras, 21 October 1628), Áron Szilády and Sándor Szilágyi, eds., Török-magyarkori állam-okmánytár [A collection of state papers from the Turkish-Hungarian age] (Pest, 1869) (henceforth TMÁO II), vol. 2, 83. 60 Ibid. For Haga’s audience, see “A True Declaration,” 214–217. 61 István Szalánczy’s report to György Rákóczy I about his embassy to the Porte (Gyulafehérvár, 16 May 1638) Beke and Barabás, eds., I. Rákóczy György, 466. 62 Such as happened in the case of the Swedish embassy of 1657–58, see Gábor Kármán, “Svéd diplomácia a Portán 1657–58: Claes Rålamb és Gotthard Welling konstantinápolyi követsége” [Swedish diplomacy at the Porte: The embassy of Claes Rålamb and Gotthard 178 gábor kármán

In the same year that the second audience was abandoned, another new element was introduced to the diplomatic ceremonies staged for Transylvania, leading the dignitaries of the principality to even more dire criticism. Chief Envoy Mihály Tholdalagi and the other members of the embassy had to kiss the carpets interwoven with gold at the feet of the sultan. This novelty, which was interpreted, with some right, as a humiliation, caused a small uproar in Constantinople and the Habsburg ambassador reported it to Vienna with considerable indignation.63 It again constituted a difference toward other foreign ambassadors, who were, like Tholdalagi, taken by the arms and walked in front of the sultan, where they only had to bow deeply and kiss the sleeves of the ruler’s garment— a custom which was, in any case, too much for some diplomats—most of the Polish ambassadors tried not to mention this in their reports.64 In spite of the protests, the new arrangement for Transylvanian envoys continued; it was only abandoned at the end of a long and troubled phase of the relationship between the Transylvanian prince and his overlord, as part of a compromise in which the new prince, György Rákóczi II (among other arrangements), agreed to the higher tribute that had long been requested by the Porte.65

Welling to Constantinople], Sic Itur ad Astra 13, nos. 1–2 (2001): 76. On Ragusa’s struggle for the second audience, see Miović, Dubrovačka diplomacija, 75. 63 Mihály Tholdalagi reported to Prince György Rákóczi I in 1637 that it was already the third year when they had to follow this new custom, the first being an audience not in Constantinople, but in a military camp close to Üsküdar (Gyulafehérvár, 19 July 1637), Szilágyi, ed., Levelek és okiratok, 485–486. This is probably the reason why the Habsburg resident, Schmid, only reacted to this novelty in the next year; see Meienberger, Johann Rudolf Schmid, 28. Maria Pia Pedani suggests that it was introduced as a general custom in 1636 (Pedani, “The Sultan,” 296), but this does not follow from the source of her data, which only narrates the Transylvanian case, cf. Joseph von Hammer-Purgstall, Geschichte des Osmanischen Reiches, Vom Regierungsantritte Murad des Vierten bis zum Frieden von Carlowitz 1623–1699 (Pest, 1835), vol. 3, 162. 64 See Kołodziejczyk, “Semiotics,” 245–252; Tetiana Grygorieva, “Practical Perfor- mance and the Ceremonial Rhetoric of Peacemaking: Peacemaking Procedure between the Polish–Lithuanian Commonwealth and the Ottoman Empire after the Khotyn War, 1621–1624,” in Osmanischer Orient und Ostmitteleuropa, eds. Robert Born and Andreas Puth (Stuttgart, 2013, forthcoming). On the various interpretations of the custom of leading the ambassador in front of the sultan, see Dilger, Untersuchungen, 52–60; Petritsch, “Zeremon- iell,” 314–315. For English examples of kissing the sleeves of the sultan, see Coulter, “An Examination,” 76; Susan Skilliter, “‘The Stately Porch of the Grand Signor at Constantino- pel’: The Ottoman Court as Seen by Elizabethan Travellers: Comparisons and Contrasts,” in Europäische Hofkultur im 16. und 17. Jahrhundert, ed. August Buck, et al. (Hamburg, 1981), vol. 1, 85. 65 István Serédy’s letter to György Rákóczi II (Constantinople, 5 June 1649) Szilágyi, ed., Erdély, vol. 1, 76. On the political context, see Sándor Papp, “II. Rákóczi György és a Porta” [György Rákóczi II and the Porte], in Szerencsének elegyes forgása: II. Rákóczi György és tributary states in the 17th-c. diplomatic system 179

This difference was thus only in practice for a relatively short period. Of those which lasted longer, the most important was not related to meeting the sultan, but rather to the earlier phases of the ceremony. Whereas some ambassadors were embarrassed to kiss the sleeves of the sultan’s robe, those coming from tributary states had to do the same for the grand vizier. This custom was not known from the diplomatic practice of any state independent of the Ottoman Empire, but is documented for Transyl- vania and Ragusa.66 Ragusan envoys also showed this kind of veneration toward other dignitaries of the empire, such as the pashas of Bosnia. It is not known whether or not the Transylvanians kissed the sleeves at their audiences with the pasha of Buda, who played a similar role for the principality as the Bosnian pasha did for the Ragusans; their reports do not go into detail and only use the standard formulation: “I saluted him decently in the name of Your Grace.”67 At the Porte, however, the Transylvanians were sometimes even warned against an exaggerated enthusiasm for the custom. In 1642 the grand vizier sent word that when visiting the divan it was only his sleeves that the chief envoy had to kiss, not those of all present.68 We can thus conclude that there was a distinction between the diplomatic ceremonial for tributary states and that for independent powers; even if the number of elements in which it was manifested varied, some of them were always observed. Even so, the most important distinction expressed through the diplomatic ceremonial was not the one that disadvantaged Transylvania and Ragusa, but the one which gave them precedence over Moldavia, Wallachia, and the representatives of provincial governors. It is not surprising that the occasional emissaries coming from the pashas of Buda, Bosnia or Silistria did not have a parade entry into town, but in light of the above discussion, it is noteworthy that the extant accounts about the submission of the Moldavian and Wallachian tribute do not mention any similar ceremony. The early eighteenth-century Descriptio Moldaviae by the former voievod, Dimitrie Cantemir, states that kora, ed. Gábor Kármán and András Péter Szabó (Budapest, 2009), 130–134; Gábor Kármán, Erdélyi külpolitika a vesztfáliai béke után [Transylvanian foreign policy after the peace of Westphalia] (Budapest, 2011), 141–147. 66 Bíró, Erdély követei, 32; Miović, Dubrovačka diplomacija, 53. 67 See, for instance, Boldizsár Sebessy’s memoir of his mission (24 May 1638) Szilágyi, ed., Levelek és okiratok, 579. Cf. Miović, “Beylerbey,” 9. 68 Letter of István Rácz to György Rákóczi I (Constantinople, 30 December 1642) TMÁO III, 179. In 1637, Mihály Tholdalagi and the other envoys kissed the sleeves of the robes of at least four dignitaries, see his report from Gyulafehérvár, 19 July 1637 in Szilágyi, ed., Levelek és okiratok, 482. 180 gábor kármán as a rule, an ağa was sent to the country, where he counted the tribute collected before his arrival. Then the money was transferred to Constanti- nople by the voievod’s envoy and submitted to the defterdar, who counted it again and acknowledged the delivery. Although Cantemir does mention that the transfer of the tribute ended with an audience with the grand vizier, where the Moldavian diplomats were given kaftans and an official letter to the voievod, it is clear that they did not see the sultan during the process.69 This method of paying tribute may have been the end of a longer process—a Venetian source from 1609 suggests, for instance, that the Wallachian envoys transferred their tribute to the sultan—but it must have been in practice for the better part of the seventeenth century; records from the 1640s mention the ağa sent to the Romanian countries to collect the tribute.70 Transylvanian sources concerning the diplomacy at the Porte are preserved in larger numbers from the 1620s on and there are many reports that the chief envoys were invited to the audience with the sultan on the same day as the Polish, French or Ragusan diplomats. There is no mention, however, that they met an envoy from the two Romanian voievodates there. They had, however, an opportunity to meet the voievods in person in Constantinople; in 1668 Transylvanian diplomats had to give precedence to Radu Leon of Wallachia at their audience with the sultan.71 This instance points to another important ceremonial difference between Transylva- nia and the two Romanian voievodates; the rulers of the latter had to be invested with their titles in Constantinople, and were also often requested to visit the Ottoman capital in person. This never happened to the Tran- sylvanian princes, most of whom never saw the city.72 Moldavian and

69 Dimitrie Cantemir, Beschreibung der Moldau: Faksimiledruck der Originalausgabe von 1771 (Bucharest, 1973), 253–255. See also Condurachi, Soli, 52. 70 In a letter to György Rákóczi I, prince of Transylvania, his orator, István Rácz, noted that the voievod of Moldavia was humiliated by the Porte, as it was only the “főpuskás” (probably referring to the silahdar ağa) sent for their tribute, whereas the sultan sent the kapıcı başı in 1636 (Constantinople, 14 October 1642) TMÁO III, 141. See, for another description with the same content from the 1670s, Matei, Reprezentanţii diplomatici, 131–135; as a counter-example, he refers to the report of Simeon Contarini to the doge (Pera, 27 March 1609) Documente privitoare la Istoria Românilor culese de Eudoxiu de Hurmuzaki [Documents concerning the history of the Romanians collected by Eudoxiu de Hurmu- zaki], vol. 4, part 1, 1600–1650, ed. Ioan Slavici (Bucharest, 1884), 303. 71 Entry 5 December 1668 in Ernő Tóth, “Hidvégi id. Nemes János naplója az 1651–1686. évekről. II. közlemény” [The diary of János Nemes Sr. of Hídvég from the years 1651–1686. Part 2], Történelmi Tár 25 (1902): 398. 72 On the investiture and confirmation ceremonies of the voievods, see Viorel Panaite, The Ottoman Law, 351–359. tributary states in the 17th-c. diplomatic system 181

Wallachian diplomats were also discriminated against in other aspects of symbolic politics; as re‛ayas, inhabitants of the tributary states were theoretically not allowed to ride horses in the empire. The privilege was, however, granted to Transylvanian and Ragusan diplomats—but not to those of the two voievodates, at least according to a Transylvanian source describing the 1620s.73

A Negative Indicator: Mistreatment of Diplomats

The quantitative ceremonial indicators of Transylvania’s prestige show decreasing numbers after 1657; fewer çavuşes were sent for the ceremonial entry of the chief envoy into Constantinople, fewer kaftans were granted to the diplomats before entering the Chamber of Petitions, and so on.74 This devaluation of the principality’s prominence can also be observed in another sphere of the diplomatic milieu in Constantinople: questions related to the diplomatic immunity of ambassadors. The literature about the Western embassies at the Sublime Porte reports a fair number of cases of mistreatment of foreign diplomats by Ottoman dignitaries. The extraterritoriality and immunity of envoys—a concept just being formed in seventeenth-century Western Europe— enjoyed a rather limited validity in early modern Constantinople. Even if it was not the general custom to use violence against them, we know of many examples in which diplomats were threatened and incarcerated or their lodgings were searched by the Ottoman authorities.75 What is more, some Ottoman documents suggest that these instances cannot be regarded as deviations from a general custom, rather they were based on the Ottoman theory of diplomatic representation, in which the diplomats

73 János Kemény, “Önéletírása” [Autobiography], in Kemény János és Bethlen Miklós művei, ed. Éva V. Windisch (Budapest, 1980), 90. On the privilege granted to Ragusans for horse-riding, see Biegman, The Turco–Ragusan Relationship, 85, especially n3. 74 Prince Mihály Apafi I complained in a supplication to the sultan in 1686 that the respect for his chief envoys and orators had declined in the last years; see his instruction to Boldizsár Macskási (Nagyszeben, 22 January 1686) Sándor Szilágyi, ed., Diplomatarium Alvinczianum/Alvinczi Péter okmánytára, vol. 1, 1685–1686. augustus 27-dikéig, Monumenta Hungariae Historica Diplomataria, no. 14 (Pest, 1870), 74–75. 75 Meienberger, Johann Rudolf Schmid, 47, 49–52; Bertelè, Il palazzo, 176–208; Tongas, Les relations, 40–42. 182 gábor kármán had the status of protected non-Muslims (müste‘min), but were also officially regarded as hostages (rehn) and bailsmen (kefil).76 Considering the recurrent character of this phenomenon, it is remarkable that there are no data from before 1657 about the serious mistreatment of Transylvanian diplomats at the Sublime Porte. This could be the result of the destruction of most sources on the principality’s Ottoman diplomacy in the sixteenth century, but the report written to Grand Vizier Köprülü Mehmed by the Transylvanian diplomats in that year states that what happened to them had been unheard of earlier in the history of Tran- sylvania; namely, they were arrested and sent to the prison of the Seven Towers.77 Even earlier the diplomats of Transylvania had something of the character of hostages. The Sublime Porte repeatedly refused to grant permission for the chief envoy to leave Constantinople; this was a used as a method of putting pressure on the prince. The longest stay was more than three years, between 1645 and 1648, instead of the usual few weeks; the victim of this diplomatic tug-of-war, Chief Envoy István Szalánczy, died in the Ottoman capital.78 In any case, Transylvanian diplomats protested fiercely in this earlier period against even the first signs of infringement on their personal freedom, such as happened in 1637, when the grand vizier stationed two janissaries in front of the Erdel Sarayı, in order to vouch for the envoys’ safety—according to the official explanation.79 In this period, their complaints were generally successful. The situation changed after 1657. The failure of György Rákóczi II’s Pol- ish campaign proved fatal; the prince could not accept that he had fallen into disfavor with Grand Vizier Köprülü Mehmed and tried repeatedly to return to the Transylvanian throne; this brought several Ottoman penal expeditions and civil war to the country. During these years of turmoil,

76 Anton C. Schaendlinger, “Der Status der Habsburgischen Gesandten an der Hohen Pforte zur Zeit Süleymans des Prächtigen,” in XXIII. Deutscher Orientalistentag vom 16. bis 20. September 1985 in Würzburg: Ausgewählte Vorträge, Zeitschrift der Deutschen Morgen- ländischen Gesellschaft: Supplement, no. 7, ed. Einar von Schuler (Stuttgart, 1989), 229– 234. This theoretical stance was also clear to contemporary Western diplomats, see Rycaut, The History, 160–165. 77 Österreichisches Staatsarchiv, Haus-, Hof- und Staatsarchiv Türkei I. Kt. 128. Fasc. 63/b. Conv. D. fol. 68–70. 78 His arrival in Constantinople was noted in his letter to György Rákóczi I from the 2 January 1645 (Beke and Barabás, ed., I. Rákóczy György, 786), and his death was reported by the Orator Ferenc Gyárfás to the same prince (Constantinople, 14 January 1648; Szilágyi, ed., Levelek és okiratok, 895–898). On his career, see also Jakó, “A Szaláncziak.” 79 See the memoirs of Boldizsár Sebessy (sine dato, Beke and Barabás, ed., I. Rákóczy György, 289), and the letter of Mihály Tholdalagi and his co-diplomats to György Rákóczi I (Constantinople, 25 January 1637; Szilágyi, ed., Levelek és okiratok, 462–463). tributary states in the 17th-c. diplomatic system 183

Transylvania endured severe losses; one third of its territory was lost to the Ottoman Empire (where the eyalet of Várad was founded in 1660), and the tribute was raised from 15,000 guldens to 40,000. The decline of the principality’s prestige was also marked by a change in the procedure for appointing princes. Instead of granting them an ‘ahdname, an offical ceremonial document also used for interstate treaties, from 1659 on the princes were confirmed with an imperial order in the form of a berat.80 While there was no example of Transylvanian diplomats being placed in custody during the earlier centuries, from this time on it became a regular occurrence.81 Hardly had the embassy of 1657 been released from the Seven Towers when, during the autumn of 1658, another group of diplomats was placed first under house arrest at the Erdel Sarayı, then later transported to the dreaded fortress.82 Even after the pacification of Transylvania it remained clear that the personal freedom of diplomats was not respected by the Sublime Porte. After several cases of threats and intimidation, in 1667 the Orator János Dávid was arrested and dragged by soldiers before the kadi for a minor offense.83 In December 1679, the Transylvanian diplomats were questioned by the grand vizier about some Ottoman soldiers killed at the borders of the principality, and when they could not present a satisfactory answer, Orator János Sárosi was arrested and put in chains. The chief envoy was helpless unless he wanted to risk his own safety, as he expressed in a sarcastic remark: “I started to call the Seven Towers the Hungarian saray.”84 Only three years later, another orator, Bálint Szilvássy, was taken into custody for at least three weeks on the orders of the grand vizier because some prisoners of the kızlar ağası were

80 Papp, “II. Rákóczi György,” 162–169. 81 In general, see Molnár, Rozsnyai, 118–119. 82 See Johannes Lutsch—Jurnal de captivitate la Istanbul (1658–1661) / Johannes Lutsch— Tagebuch seiner Gefangenschaft in Istanbul (1658–1661), ed. Cristina Feneşan and Costin Feneşan (Timişoara, 2006); Bíró, Erdély követei, 54–55. 83 Letter of Dávid Rozsnyai to Mihály Apafi I (Adrianople, 25 May 1667) Áron Szilády and Sándor Szilágyi, eds., Török-magyarkori állam-okmánytár [A collection of state papers from the Turkish-Hungarian age] (Pest, 1870), vol. 4, 356. 84 Letter of László Székely to Mihály Teleki (Constantinople, 15 December 1679), Sámuel Gergely, ed., Teleki Mihály levelezése [The correspondence of Mihály Teleki], vol. 8, 1678–1679 (Budapest, 1926) [henceforth; TML VIII], 558. See also the letter of Székely and András Keczeli to Mihály Apafi I (Constantinople, 10 December 1679) Sándor Szilágyi, ed., Erdélyi országgyűlési emlékek történeti bevezetésekkel [Documents of the diets of Transyl- vania, with a historical introduction], vol. 17, 1679–1682 (Budapest, 1894) [henceforth: EOE XVII], 68; and the letter of Székely to Mihály Teleki (Constantinople, 9 December 1679) TML VIII, 556. 184 gábor kármán suspected of hiding in the Transylvanian embassy.85 The frequency and gravity of this mistreatment clearly separated the principality’s embassy from those of the independent powers and shows that after 1657 their prestige fell and became closer to that of the Moldavian and Wallachian capuchehaias, who were often reported, throughout the entire period, to have been treated harshly by Ottoman dignitaries.86 * * * As promised in the introduction of this chapter, the survey of various elements concerning the diplomatic representation of Transylvania at the Sublime Porte provides a more nuanced understanding of the political status of the principality. Certain analytical tools, such as the typology of the diplomats and the terminology connected to them, which we anticipated would show the distinctions between the treatment of tributaries and independent states proved ill-suited to this task. Divisions also turned out to run along different lines in the case of the topography of the residents’ lodgings. Using other topics of analysis it is, however, possible to point out that distinct boundaries did exist between the peripheries of the empire and its tributaries, respectively, those states which—even though they sometimes paid tribute to the sultan—should be considered independent. Surprisingly, however, both the quantitative and qualitative differences in the diplomatic ceremonies in Constantinople proved fewer between Transylvania and the independent states than among the principality and the two Romanian voievodates. This conclusion can be further contextualized by discussing cases of mistreatment, which also point to an important chronological divide in the prestige Transylvania enjoyed at the Sublime Porte. Transylvanian diplomats were certainly aware of the stakes in the games of diplomacy they played (which were also played with them) and they pulled every string to avoid any signs that would indicate the loss of prestige for their country. The place of the principality in the hierarchy is probably best shown by identifying their most important opponent in the ceremonial contest: Transylvanian and Ragusan diplomats did their best to avoid each other at audiences with the sultan because of the risk that

85 Letter of Bálint Szilvássy to Mihály Apafi I (Constantinople, 5 July 1682), EOE XVII, 285–286. 86 Condurachi, Soli, 55–58; Matei, Reprezentanţii diplomatici, 176–178. tributary states in the 17th-c. diplomatic system 185 the other’s ceremonial precedence would be demonstrated.87 In light of the ongoing discussions, also represented by this volume, this conclusion still does not answer the question posed at the beginning of the paper, whether Transylvania should be regarded as an “independent country” or a “slave of the sultan.” The answer only shows that the question itself is much too simplistic and should be dismissed, as it does not encompass the subtly refined character of early modern political realities.

87 Bíró, Erdély követei, 46; Miović, Dubrovačka diplomacija, 74. Mihály Tholdalagi wrote to György Rákóczi I (Constantinople, 27 July 1636) that he chose to postpone his audience with the sultan in order to avoid having to give precedence to the Ragusans, TMÁO II, 385–386.

Diplomatic Relations between the Ottoman Empire and the Republic of Dubrovnik

Vesna Miović

Introduction: The Beginnings of a Tributary Relationship

By the end of the fourteenth century, the Ottomans had conquered much of the Balkans, thus making their first contacts with the Ragusan merchants who were engaged in trade activities throughout the peninsula. At that time, Dubrovnik authorities still hoped that the alliance of Christian powers would prevent the further advance of Ottoman armies. Conse- quently, Ragusans were reluctant to establish more regular relations with the Ottomans, the Porte in particular, but dealt with the matter only on an episodic basis, mainly for commercial purposes. Attempts at improving the relations between the Republic of Dubrovnik and the Porte became more concrete in 1430 and 1442.1 However, the two states remained fairly distant until the reign of Sultan Mehmed II and his systematic conquest of Serbia, Bosnia, and the bulk of Herzegovina in Dubrovnik’s immediate neighborhood.

1 The first Ragusan tribute ambassadors were dispatched to the Porte in 1430 at the prompting of Murad II. Although the main motive behind their mission was Dubrovnik’s conflict with the Ottoman vassal and Bosnian lord Radosav Pavlović, they also discussed the issue of Ragusan merchants, after which Murad II issued a ferman confirming the freedom and safe conduct of Ragusan trade in Ottoman territories. After the first fall of Serbia (1439), Murad II—imprisoning Ragusan merchants and threatening to attack Dubrovnik— demanded a tribute of 10,000 ducats from Ragusa. Finally, a compromise was reached in 1442: the Ottomans agreed to an annual gift of silver dishes worth 1,000 ducats. It should be noted, however, that the two above mentioned documents of Sultan Murad II regulated and legally defined the framework of Ragusan trade in the Balkans as it had been conducted before the Ottoman arrival to the Balkans; issuing such documents was, in a broader sense, an Ottoman practice elsewhere as well. Following the restoration of Serbia and its despot in 1446, Ragusans were no longer required to pay a regular tribute; thus, the earlier legal status (that is, before the Ottoman occupation of Serbia) was re-established. In pursuit of peace after a succession of military defeats, Sultan Murad II officially renounced the Ragusan “gift.” See Ivan Božić, Dubrovnik i Turska u XIV i XV veku [Ragusa and Tur- key in the fourteenth and fifteenth centuries] (Belgrade, 1952), 1–22, 49–60, 104–111; Vinko Foretić, Povijest Dubrovnika do 1808 [History of Ragusa until 1808] (Zagreb, 1980), vol. 1, 12, 117–120, 203, 206, 207, 211, 212, 264, 265. 188 vesna miović

The reign of Mehmed II the Conqueror brought significant changes to the relations of the two states. Given that at this time Ottoman troops were marching toward the republic, Ragusans could no longer hope to receive help from Christendom. In 1458, they resumed negotiations with the sultan, agreed to pay an annual tribute of 1,500 gold coins and pledged loyalty to the Ottoman Empire.2 In exchange, the sultan gave the Ragu- sans the capitulation (‘ahdname), in which he guaranteed to protect their republic. The sultan also confirmed a number of privileges to Ragusan merchants throughout the Ottoman territories.3 In spite of this agreement, Ragusans spent most of the middle decades of the fifteenth century fortifying their city in preparation for an anticipated attack. It became clear over time, however, that Mehmed was hesitant about conquering the republic: he continued to increase the tribute until it had grown tenfold4 and tried to persuade Ragusans to entirely and voluntarily submit to Ottoman rule. His plans of conquest were aimed at the Italian Peninsula, for which a conquered Dubrovnik would have served as a suitable foothold.5 The Ottoman danger abated with Sultan Mehmed’s death in 1481. Yet, it was during his reign that the legal foundations of the relations between the Ottoman Empire and the Republic of Dubrovnik had been fully defined. Mehmed’s successor, Bayezid II, settled Dubrovnik’s annual payment at 12,500 gold coins. Following the conquest of Herzegovina in 1482, the Ottomans became the republic’s sole neighbors on land.

2 In order to lift the blockade (Tur. yasak) from the Ragusan border, the ambassadors bringing the tribute in 1651 tried to stress the fact that Ragusans voluntarily accepted the paying of the tribute; this latter circumstance cannot be stressed enough in this context. The ambassadors were aware that the Ottomans differentiated between tributaries that voluntarily accepted paying the tribute and those subjected to it by military means. See Državni Arhiv u Dubrovniku [State Archives of Dubrovnik; henceforth DAD] Diplomata et Acta (Acta Turcarum), 7/2.1, vol. 24, nos. 1087–1089. 3 Branislav Nedeljković, “Dubrovačko-turski ugovor od 23. oktobra 1458. godine” [Ragusan-Turkish treaty of 23 October 1458], Zbornik Filozofskog fakulteta 11, no. 1 (1970): 363–390; Foretić, Povijest, vol. 1, 227–228. 4 Mehmet II increased the amount of the Ragusan tribute concomitant with his military successes, threatening to conquer the Republic and incarcerate Ragusan merchants: in 1468 he increased it to 5,000 gold coins, in 1472 to 9,000, and in 1473 to 10,000. In 1476, the sultan decided that Ragusan merchants had to pay an import and export customs duty of 5 percent. A year later he acknowledged the complaints of the Ragusan government and appointed it lease-holder of customs duty. The lease amounted to 2,500 gold coins per year. In 1480, he once again increased the tribute by 2,500 gold coins, which then, in addition to the duty lease, climbed to 15,000 gold coins. 5 Foretić, Povijest, vol. 1, 235; Božić, Dubrovnik i Turska, 188–205. diplomatic relations between the ottomans and dubrovnik 189

In any case, the Republic of Dubrovnik managed to retain its integrity during and following the first Ottoman conquests. It was, however, not by chance that such a fortunate fate had befallen Dubrovnik: in fact, the republic seemed to have exactly what the Ottomans were in need of. The Ottomans lacked the following:

1) merchants to trade and conduct business throughout the Balkans— and Ragusans were well-known as skillful merchants, businessmen, and bankers; 2) a neutral port through which they could trade with Venice, Spain, the Papal State, and so on, even if they happened to be at war with them; 3) neutral territory for the exchange of the prisoners of war; 4) a mediator between the Ottomans and the Christian states of the West; and 5) a “window” to the Mediterranean through which information about the Christian states in this basin would be gathered.

Dubrovnik was able to meet all these demands, but only on the condition that it remain neutral. In other words, both the East and West had to perceive it as neutral. This meant that the Ottomans could not burden Dubrovnik with the customarily employed vassal restrictions; they had to leave it free. This fact is confirmed by Dubrovnik’s unusually brief capitulation, which contained only roughly drawn political and commercial articles. One of them clearly testifies to the special status enjoyed by the republic: “The territory of Dubrovnik is open to all whether arriving by land or sea, whether a friend or enemy of the sultan.”6 This regulation evidently departs from the customary vassal terms, presupposing that an enemy of the sultan is by default the enemy of the sultan’s vassal. Addi- tionally, the Dubrovnik capitulation fails to mention the regular obligation of providing military and logistic support to the sultan.7

6 Vesna Miović, Dubrovačka Republika u spisima osmanskih sultana [The Ragusan republic in the documents of the Ottoman sultans] (Dubrovnik, 2005), 13. 7 Viorel Panaite, The Ottoman Law of War and Peace: The Ottoman Empire and Tribute Payers (Boulder, CO, 2000), 160, 164. For the text of Dubrovnik’s capitulation, see Nikolaas Hendrik Biegman, The Turco–Ragusan Relationship according to the Firmans of Murad III (1575–1595) Extant in the State Archives of Dubrovnik (The Hague and Paris, 1967), 56–59. 190 vesna miović

Dubrovnik between East and West

Given the circumstances, the Ottomans did not endanger the existence of the republic and its commercial development. Also, it must be repeatedly stressed that they gave a fresh impetus to the formation of a politically neutral Republic of Dubrovnik, presenting it as such to Christendom, which in turn accepted this role of Ragusa, since it was also precisely what it needed. The overall protection of the republic against its greatest enemies, the Venetians, may serve as an illustration of this peculiar international arrangement in which Ragusan neutrality was supported by both East and West. The Ottomans were Dubrovnik’s main protectors, but not the only ones: Spain also protected Dubrovnik from the Venetians. For instance, during the war of the First Holy League against the Ottomans in 1539, Dubrovnik was threatened by Venice. On this occasion, a Spanish military expert arrived in Ragusa to assist the inhabitants in fortifying their city; this was done because Dubrovnik had an impressive commercial fleet sailing under neutral flag and its vessels were frequently chartered by the Spaniards.8 Ottoman–Ragusan relations were under considerable strain at the time, with the Ottomans accusing the Ragusans of supporting the Holy League. This, however, had no serious long-term effects on their relations.9 The presence of the Spanish military expert in Dubrovnik during the war provoked no reaction from the Porte, which may seem curious, but only at first sight; the situation is not so surprising in light of the foundations of Ottoman–Ragusan relations. Besides the Porte and Spain, Dubrovnik also enjoyed the protection of the Popes. During the wars of both Holy Leagues against the Ottomans (1538–1540 and 1570–1573), Venice tried to involve Dubrovnik in the League, which would have destroyed its neutrality and led it straight into ruin. Upon the prompting of the Ragusans, the Popes opposed this idea on both occasions. The reason is simple: the Curia depended upon Ragusan services to spy on the Ottomans and protect Catholics inhabiting the territory of Ottoman Herzegovina.10 Up until the 1520s, the republic had two protectors to whom it paid tribute—the Porte and the kings of Hungary and Croatia. To the latter,

8 Jorjo Tadić, Španija i Dubrovnik u XVI veku [Spain and Ragusa in the sixteenth century] (Belgrade, 1932), 68–69, 80. 9 Foretić, Povijest, vol. 2, 40–41. 10 Toma Popović, Turska i Dubrovnik u XVI. veku [Turkey and Dubrovnik in the sixteenth century] (Belgrade, 1973), 264–266. diplomatic relations between the ottomans and dubrovnik 191

Dubrovnik paid an annual tribute of 500 ducats, based on an agreement made with King Louis the Great in 1358. The relationship between the Republic of Dubrovnik and the Hungarian Kingdom lasted until the battle of Mohács in 1526. After the death of King Louis II, the new claimants of the Hungarian throne also demanded tribute from the republic, but after lengthy diplomatic maneuvers the Ragusans managed to retreat from such close relations with a state that was the main target of the Ottoman expansion.11 Clearly, Dubrovnik enjoyed an extremely favorable position in international affairs. In order to preserve this position, however, Ragusan authorities and the diplomatic corps expended great effort. The so-called “tribute ambassadors” had an extremely important role in maintaining relations with the Porte. These diplomats—always two noblemen—delivered the annual harac to the sultan and were supposed to resolve current issues with the empire, such as the importation of Ottoman grains to Dubrovnik; the protection of Ragusan merchants in the Ottoman territories; the maintenance of the monopoly of transit trade going through Ragusan ports; the maintenance of the Ragusan monopoly of selling salt to the Ottoman subjects in Herzegovina; and protection against North African corsairs and bandits from Montenegro and Herzegovina, and so on. This was, nevertheless, only a part of the Ragusan diplomats’ tasks; in order to preserve the good international position of the republic, far more was needed. Ragusans had to be fully and accurately informed about everything that concerned the Ottomans, this for two main reasons: first, to protect the republic’s interests and, second, to deliver information on the Ottomans to the Christian powers. For this they needed a carefully constructed and maintained diplomatic network. The most important part of the network was naturally in Istanbul. Apart from the tribute ambassadors, the Istan- bul network consisted of numerous other individuals: consuls, dragomans, charge d’affaires, as well as unofficial persons—Dubrovnik’s confidants.

The Ragusan Diplomatic Network in Istanbul

The government of Dubrovnik frequently collaborated with confidants, mostly Ragusan merchants or physicians who worked for the government unofficially, under the guise of their profession. Adept in customs

11 Foretić, Povijest, vol. 1, 133–134; vol. 2, 8–35. 192 vesna miović and relations with the Ottomans, they assisted the ambassadors in every situation. Their role was particularly important in the period before the first Ragusan resident consul to Istanbul was appointed (1688). It was in these times that the Ragusan government relied on Bernard Giorgi, physician Daniel Abeatar, Solomon Abeatar, Gabriel Battitore, Mattatia Ben Castiel, Augustin Tudisi, Trojan Sorgo, Vicko Querini, and physician Frano Crasso.12 Among these, Castiel and Crasso stood out particularly. Due to his reports and other services, the republic became so indebted to Ben Castiel that he was provided with a state house, free of charge, when he returned to Dubrovnik in 1640.13 Crasso was the grand vizier’s physician and as such he could accomplish a great deal. He acted as a mediator in nearly every communication between the Ragusan ambassadors and the grand vizier. It was solely a result of his influence that in 1641 the grand vizier dismissed Şahin Pasha, a Herzegovinian sancakbeyi who was thoroughly antagonistic toward Dubrovnik. He was replaced with a sancakbeyi Crasso was very familiar with.14 Like others, Crasso also reported about events at the Porte. For example, he gave a humorous account of Sultan İbrahim I from a medical perspective: “[d]ue to the captivity he had endured, His Excellency the sultan suffered from depression which rendered him numb and unable to have intercourse with women for some time after ascending to power.”15 In the “mosaic” of Dubrovnik’s diplomatic corps in Istanbul, interpret- ers of Ottoman Turkish, the so–called dragomans, played a very important part. Like other European countries, Dubrovnik was also aware of the risk of assigning such an important role to Ottoman subjects; ultimately, educating Ragusans proved to be the only possible solution. The so–called giovani di lingua (language youths), or students of foreign language, as we would call them today, began to appear in Dubrovnik around the middle of the sixteenth century.16 However, the languages which were difficult to master (besides Ottoman Turkish, also, for instance, Arabic and Persian), and the fear of “a different world” took their toll: there were very few language youths and educated dragomans. It was not until the eighteenth

12 DAD Let. Lev. vol. 47, fols. 91–92, 106, 120v, 124v, 134–135v, 144–145, 148v, 178–179v, 185v–188, 189v, 190, 229v, 234. 13 DAD Let. Lev. vol. 44, fols. 48, 49a; vol. 47, fols. 229, 234–235v. 14 DAD Acta Consilii Rogatorum [henceforth cited as Cons. Rog.], vol. 97, fol. 92v, 103; DAD Let. Lev. vol. 48, fols. 21v–22v, 49v, 90v–91v, 98–99v, 102, 108–109v, 139; DAD Acta et Diplomata 17 [henceforth AD 17], series 76, vol. 1838, no. 10, 11; vol. 1861, no. 1, 2, 4, 9. 15 DAD AD 17, vol. 1861, no. 11. 16 DAD Let. Lev. vol. 27, fol. 12v. diplomatic relations between the ottomans and dubrovnik 193 century that they became more common.17 From the beginning of the eighteenth century until 1804, the famous Ragusan dragomans who traveled to Istanbul with the tribute ambassadors were Luka Lučić, Andrija Magrini, Miho Zarini, Miho Božović, Nikola Radelja, and Antun Vernazza.18 In the meantime, the government in Dubrovnik came to a logical solution: from the sixteenth century, it began to rely on its merchants, who had become fluent in the language and customs during their stay in the empire.19 In the seventeenth century, Dubrovnik’s dragomans consisted almost exclusively of Ragusan merchants from Ruschuk, Belgrade, Sofia, and Plovdiv. Most European countries found ways to make up for the lack of dragomans. Even though Venice had been trying to adopt a variety of methods throughout the centuries to educate its own youth as successfully as possible, they were not able to raise much interest, and the difficulties young Venetians—as well as their Ragusan counterparts—encountered proved too great a hurdle. Venice, therefore, began to rely on the so-called Levan- tines, Ottoman subjects of European descent. The quarters of Galata and Pera in Istanbul housed dragoman dynasties intertwined through marital ties; the job of a dragoman came to be passed on from father to son in each family. They formed a caste of sorts, similar to unofficial nobility. During the seventeenth and eighteenth centuries, the following families were known: Alleon, Borisi, Brutti, Carli, Chabert, Dane, Dantan, Donà, Draperis, Fonton,20 Fortis, Grillo, Hübsch von Grosstahl, Navoni, Negri,

17 For an extensive account on educating Ragusan language youths see Bogdan Kriz- man, O dubrovačkoj diplomaciji [On Ragusan diplomacy] (Zagreb, 1951), 120–126; Vesna Miović-Perić, “Dragomans of the Dubrovnik Republic: Their Training and Career,” Dubrovnik Annals 5 (2001): 81–94. 18 DAD Let. Lev. vol. 67, fol. 207; vol. 68, fols. 81, 81v; vol. 69, fol. 71; vol. 70, fols. 145v–151v; vol. 71, fol. 137v; vol. 72, fol. 51; vol. 73, fols. 104, 104v; vol. 74, fols. 114–126; vol. 76, fols. 1–11v; vol. 77a, fols. 1–13, 188–208; vol. 77b, fol. 237; vol. 79, fols. 224–240v; vol. 82, fol. 166v; vol. 84, fols. 209–229v; vol. 86, fol. 96v; vol. 88, fols. 204–230v: item 25; vol. 91, fol. 173: item 21; vol. 94, fols. 123–141 item 27; vol. 96, fols. 118–133: item 23; vol. 98, fols. 43–58: item 20; vol. 101, fols. 15–28v: item 36, 83; vol. 103, fol. 112v; vol. 104, fol. 117; vol. 106, fols. 33–36, 72–73: item 48; vol. 107, fols. 53, 110–133v: item 48; vol. 108, fols. 7–8v; vol. 109, fols. 46–67v: item 48; vol. 110, fols. 17–18v, 204v, 243. DAD Cons. Rog. vol. 138, fol. 185v; vol. 146, fol. 99; vol. 150, fol. 45; vol. 154, fols. 227v, 235; vol. 157, fols. 82, 100v. DAD Officiales Rationum, vol. 18, c. 106. 19 DAD Let. Lev. vol. 19, fols. 163v, 176; vol. 20, fols. 15, 193v; vol. 21, fols. 43–44; vol. 25, fols. 117v–119v, 199v–200v; vol. 26, fols. 69v, 192v, 193; vol. 27, fols. 12, 136, 136v; vol. 28, fols. 27v–28v; vol. 29, fols. 92, 92v, 118, 206, 206v; vol. 30, fols. 12v, 139v, 218v; vol. 32, fol. 244v. 20 Karlo Fonton, a Russian consul in Dubrovnik, was a member of the same clan. The first Russian consul to Dubrovnik, Antun Gika (1788–1800), was the descendant of those dragomans of the Porte who, after resigning from their post, were assigned to Moldavian rulers. 194 vesna miović

Olivieri, Perone, Pini, Pisani, Salvago, Tarsia, Testa, Timoni, and others. The most distinguished among them belonged to the supreme authority, i.e., the Serenissima Comunita di Pera e Galata council, better known as Magnifica Comunita.21 The council was headed by a principal (most frequently from the Testa clan, one of the oldest dragoman dynasties), a deputy principal and twelve councilmen.22 To put it simply, members of these dragoman families had a kind of a dual citizenship, since they were officially protected by the Ottomans and the country they worked for. For instance, the members of the Testa family worked for diplomats from Genoa, Venice, Sweden, Russia, Holland, Austria, England, Prussia, and France.23 Dragoman family members’ national affiliation was in itself an imaginary concept and therefore both the Ottomans and especially the Europeans eyed them with suspicion. The Ottomans considered them people of a different religion, while Europeans found them overly familiar with Ottoman public and political life and suspiciously at ease with everyday life and traditions in the empire. This probably explains the mock poem that circulated around the Mediterranean at that time, part of which said, “A Pera ci sono tre malanni: peste, fuoco e dragomanni” (“Pera

See Nora Şeni, “Dynasties de drogmans et levantinisme a Istanbul,” in Istanbul et les langues orientales, Varia turcica, no. 31, ed. Frédéric Hitzel (Paris and Montréal, 1997), 173; Maren Frejdenberg, “Izvještaji Antuna Gike, prvog ruskog konzula u Dubrovniku” [The reports of Antun Gika, the first Russian consul in Ragusa] Anali Zavoda za povijesne znanosti HAZU u Dubrovniku 37 (1999): 219. 21 In 1453, when Sultan Mehmed II conquered Istanbul, he granted religious autonomy and the right to free trade to Genoese merchants residing in the Galata quarter, on the condition that they pay him capitation. With the sultan’s permission, Genoese merchants would in the future elect one from their ranks to be their representative before Ottoman authorities and to act with the help of the council, i.e., the aforementioned Magnifica Comunita. Initially, members of the community were heads of Veneto-Genoese families residing in Galata. See Jozef Matuz, Osmansko Carstvo [The Ottoman empire] (Zagreb, 1992), 43, 44; Livio Missir de Lusignan, “Une aristocratie “inclasable”: les drogmans,” in Istanbul et les langues orientales, Varia turcica, no. 31, ed. Frédéric Hitzel (Paris and Mon- tréal, 1997), 154; Şeni, “Dynasties de drogmans,” 162, 163. 22 Additionally, distinguished Greeks from the Fener quarter had been members of the Council for some time, and together with the Levantines, they shared influential duties with the dragomans of the Porte and with those of the Ottoman fleet. The Fanariots, however, managed to seize a monopoly of the above mentioned duties and therefore separated from the rest in the mid-seventeenth century. This was the first reason why Levantine supreme authority weakened and finally diminished toward the end of the century. The Testa clan continued to run the Levantine community. See Andrei Pippidi, “Drogmans et enfants de langue: La France de Constantinople au XVIIe siècle,” in Istanbul et les langues orientales, Varia turcica, no. 31, ed. Frédéric Hitzel (Paris and Montréal, 1997), 147; Maria de Testa and Antoine Gautier, “Deux grandes dynasties de drogmans: les Fonton et les Testa,” ibid., 182, 185, 193–196; Şeni, “Dynasties de drogmans,” 163, 170–174. 23 See de Testa and Gautier, “Deux grandes dynasties,” 177. diplomatic relations between the ottomans and dubrovnik 195 holds three troubles: the plague, the fire, and the dragomans”).24 Drago- mans also stood out with their special attire, an integral part of which was the Ragusan cap.25 Venetian diplomats and consuls kept sending home warnings that the Levantine dragomans were neither sufficiently reliable nor trustworthy: Since they reside in Istanbul with their families, where they own real estate, Ottoman benevolence is far more important to them than that of Venice or any other country they are working for. Therefore, they will not always tell the Turks what they were ordered to, nor will they convey to us everything that was said to them by the Turks. Besides, they constantly insist that we should bear lavish gifts to the Ottomans, for their own personal reasons. Each ruler should, therefore, appoint solely his own subjects as dragomans, who could serve him loyally and with dedication.26 Venetians took measures by beginning to control—that is, limit—the number of Levantine youths among their foreign language students. In order to maintain the loyalty of those dragomans who were natives of Venice, the bailos proposed that Venetian dragomans who married Levan- tine women would lose their jobs—this practice was also adopted by the French and the Austrians.27 Clearly, the Levantine dragomans possessed significant political power and influence due to their specialized skills and their strong connections at the Porte, both usually bequeathed to them by their families. Foreign consuls and diplomats were openly frustrated over the fact that they could not function without dragomans, even in their private lives. Bailo Ber- nardi, for instance, bitterly remarked: “What else is a bailo to do but speak with the tongue of another, listen with the ear of another, and lastly, act with the help of the brain of another.” Saint-Priest, a French consul, drew a similar conclusion: “A consul cannot, nor will ever be something more than his dragoman’s clerk.”28

24 Isabella Palumbo Fossati Casa, “L’École vénetienne des ‘giovani di lingua,’ ” in Istan- bul et les langues orientales, Varia turcica, no. 31, ed. Frédéric Hitzel (Paris and Montréal, 1997), 113; Alexander H. de Groot, “Protection and Nationality: The Decline of the Drago- mans,” ibid., 235, 236. 25 Şeni, “Dynasties de drogmans,” 163. 26 Gligor Stanojević, Vijesti o Turskoj [News concerning Turkey] (Pietro Busenello: Noti- zie Turchesche) (Sarajevo, 1960), 146–147; Francesca Lucchetta, “Un progetto per una scuola di lingue orientali a Venezia nel settecento,” Quaderni di studi Arabi 1 (1983): 1. 27 Francesca Lucchetta, “Una scuola di lingue Orientali a Venezia nel settecento: il secondo tentativo,” Quaderni di studi Arabi 2 (1984): 24, 26, 39. 28 Carla Coco and Flora Manzonetto, Baili Veneziani alla Sublime Porta (Venezia, 1985), 107; de Groot, “Protection and Nationality,” 241. 196 vesna miović

Ragusans were quite pragmatic in dealing with the same problem. All Ragusan resident consuls in Istanbul—except for the first one, Luka Barca (1688–1709)—were professionally trained dragomans: Luka Chirico (1709–1749), Đuro Chirico (1749–1767), Đuro Curić (1768–1787), and Fred- erik Chirico (1787–1807). They were all married to members of Levantine families such as the Testa, Pini, Pisani, Timoni, Fonton, and Dantan— each of them a source of valuable information. Ragusan merchant Petar Čingrija (Dubrovnik’s charge d’affaires in Istanbul 1755–1768), as well as the Ragusan physicians Mato Bratis and Ivo Beneveni (Dubrovnik’s confidants during the second half of the eighteenth century) were also married to members of Levantine families.29 Thus, the Republic of Dubrovnik established strong ties with the Levantines in a more “secure” way than employing them. The following example shows the manner in which Ragusan authorities benefited from these family ties. According to unreliable data, Luka Chirico, Ragusan consul to Istanbul, married a certain Miss Blummer, the sister of an Eng- lish dragoman, whose name is unknown. They had three children: two daughters, Katarina and Antonija, and a son named Đuro. Katarina married the English dragoman Angelo Timoni. Antonija married a member of the Pini dragoman family, and after his death, she married again, this time the Ragusan dragoman Andrija Magrini.30 In 1749, after his father’s death, Đuro received the post of Ragusan consul. Since he was young and inexperienced, his brother-in-law Timoni was appointed temporary consulate supervisor.31 In 1760, Đuro married Elizabeta, the daughter of Frederich Hübsch, an Austrian merchant, who had close ties to Levantine dragomans.32 Hübsch was one of the few who, like Angelo Timoni, used his ties to the Chirico family to become a Ragusan confidant of foreign

29 For an extensive account on Ragusan ties with the Levantine families, see Vesna Miović, Dubrovačka diplomacija u Istambulu [Ragusan diplomacy in Istanbul] (Dubrovnik, 2003), 115–122. 30 Alcuni brevi ceni circa il rinnomato Luca di Barca nativo di Ragusa, già Primo Console Generale della Repubblica di Ragusa, e di varie altre Potenze Europee appresso la Sublime Porta di Constantinopoli scritto da Marco B. Marinovich, Cittadio Raguseo, DAD Rukopisna ostavština Ernesta Katića, fol. 197; Vuk Vinaver, Dubrovnik i Turska u XVIII veku [Dubrovnik and Turkey in the seventeenth century] (Belgrade, 1960), 11; DAD Acta et Diplomata 18 [henceforth cited as AD 18], series 76, vol. 3164, no. 82; vol. 3168, no. 109. 31 DAD Let. Lev. vol. 77a, fols. 126v, 127, 129–131; Ilija Mitić, Konzulati i kozularna služba starog Dubrovnika [Consuls and consular service of ancient Ragusa] (Dubrovnik, 1973), 69, 122–123. 32 Maria, Frederich Hübsch’s second daughter, was married to an English dragoman, Stefano Pisani (Şeni, “Dynasties de drogmans,” 162, 170). diplomatic relations between the ottomans and dubrovnik 197 descent. The Ragusan government entrusted him with appeasing the Eng- lish consul, who accused the Ragusans of collaborating with the French during the Seven Years’ War (1756–1763).33 Duval, the brother-in-law of Pero Čingrija, the Ragusan chargé d’affaires in Istanbul, was the dragoman at the French consulate.34 So, while Đuro Chirico and Frederich Hübsch tried to neutralize the English consul’s animosity toward the republic, the tribute ambassadors could surreptitiously communicate with the French consul through Čingrija and Duval.35 Moreover, the Ragusan authorities made use of Čingrija and Duval when they wanted to turn the French consul in Istanbul against his colleagues stationed in Dubrovnik, Le Maire (1758–1764) and Le Prévost (1764–1771).36 In doing so, the authorities were not at all concerned that these French officials kept close ties with each other. The consul in Istanbul had been ordered by the king of France to keep track of Ragusan activities and keep them under control; however, he was rather controlled by Ragusans. For instance, it was Čingrija’s duty to secure the consul’s help in dismissing the infamous Le Maire, and for this purpose, he regularly received long letters from Dubrovnik about every movement of the French diplomat.37 Čingrija was so persuasive with the French consul in Istanbul that the latter finally sided with the Ragusans. When in 1764 the new French consul, Le Prévost, arrived in Dubrovnik, Čingrija managed to learn everything he had reported to the French consul in Istanbul about his reception in

33 The reason for the English consul’s accusations was a Frenchman who had begun building a warship in Dubrovnik at the beginning of the war. In retaliation, English corsairs started confiscating Ragusan ships. Hübsch worked on behalf of the Republic of Dubrovnik for about thirty years (DAD Let. Lev. vol. 79, fols. 204–205v, 243–246v; vol. 80, fols. 90–98, 102–103, 106–107v; vol. 82, fols. 58, 58v, 138–160v: item 16; vol. 83, fols. 62v, 69v; vol. 84, fols. 40, 40v, 103, 103v, 161, 161v, 209–229v: item 24; vol. 85, fol. 13; vol. 86, fols. 89–101v: item 16; vol. 88, fol. 153; vol. 89, fol. 28; vol. 92, fols. 28, 34v, 103v; vol. 94, fols. 123–141: item 20, fol. 173; vol. 104, fols. 112, 117v, 121v, 195–196v). 34 Marija Čingrija, most likely Petar’s daughter, married the Dutch dragoman Giuseppe Testa (de Testa and Gautier, “Deux grandes dynasties,” 193; DAD AD 18, vol. 3165, no. 25, 32). 35 DAD Let. Lev. vol. 79, fols. 204–205v, 243–246v; vol. 80, fols. 90–98, 102–103; vol. 82, fols. 58v, 85, 138–160v (item 16); vol. 83, fol. 62v; vol. 84, fols. 103, 103v, 209–229v (item 17); vol. 86, fols. 89–101v (item 16); vol. 88, fols. 204–230v (item 18). 36 During the 1740s Ragusan maritime affairs expanded rapidly and the French sought to control them together with the city’s economic development in general. This is the reason Le Maire was the first French person to be given the post of the French consul in Dubrovnik (before Le Maire, a native Ragusan served in this capacity). Naturally, the Ragu- san authorities did not welcome Le Maire, not only on account of his task, but also because he was known as a man with an irritating and short-tempered nature. Le Prévost also received a cold welcome for the same reasons (Foretić, Povijest, vol. 2, 235–238, 248–250). 37 The “job” of dismissing Le Maire was completed by Frano Sorgo-Bobali, Ragusan ambassador to Paris. 198 vesna miović

Dubrovnik and about the republic in general. Thus, with the help of the French consul in Istanbul, Ragusans learned about Le Prévost’s thoughts, feelings, and intentions.38

Ragusan Espionage Activities in the Ottoman Empire

All the aforementioned Ragusan representatives were, however, merely part of a far broader espionage network. Between 1500 and 1550 the republic established 44 consulates in the western Mediterranean, and from 1750 to 1808 their number rose to 57. In the eastern Mediterranean, the numbers for those periods were 6 and 27 respectively. Whenever Ragu- san maritime activities experienced a period of stability, Ragusan ships frequented every significant port in the Mediterranean and even the Atlantic.39 Not only consuls and diplomats, but all Ragusans, including merchants and seamen, made it their duty to gather information; in other words, the authorities received news from anyone who, residing on foreign ground for any purpose, may have estimated that what he had heard or seen could serve the republic’s interests.40 Every Ragusan was aware of the fact that the Ragusan authorities were to be promptly informed about the current plans of both the Ottoman Empire and other European countries. Furthermore, Ragusan authorities gave the following instructions to tribute ambassadors: they should stress to the Ottomans at every step that the republic had been and would be loyal to them from the beginning of time to the end of the world; they should give special attention to anyone possibly plotting against the Republic of Dubrovnik, its trade, salt trade, maritime affairs, etc.; they should report on all issues deemed worth knowing, with complete accuracy and trustworthiness; especially important news should be sent by emergency couriers; matters conducted

38 DAD Let. Lev. vol. 79, fols. 244v–246v, 247–249; vol. 80, fols. 155, 155v; vol. 81, fols. 146–147v; vol. 83, fols. 10–15v, 20–23, 35–36, 96, 96v, 108; vol. 85, fols. 218–219, 258; vol. 86, fols. 1–12v, 17, 19–35, 55–56; Foretić, Povijest, vol. 2, 235–247, 269–273. 39 Mitić, Konzulati, 202–206; Antonio Di Vittorio, “Un grande nodo postale tra Ori- ente e Occidente in éta moderna: La Repubblica di Ragusa,” in Ragusa (Dubrovnik) una Repubblica Adriatica: Saggi di storia economica e finanziaria, ed. Antonio di Vittorio, Sergio Anselmi, and Paola Pierucci (Bologna, 1994), 67–69. 40 See such reports from Ragusans with no specific diplomatic mission in DAD AD 17, vol. 1810; DAD AD 18, vol. 3175, 3176. diplomatic relations between the ottomans and dubrovnik 199 should not be discussed with anyone privately, not even upon their return to Dubrovnik.41 Dragomans had an obligation to continually visit all possible sources of information.42 It was a well-known fact that they were capable spies; moreover, no information could be deemed reliable without them, as expressed by the Ragusan ambassador of 1676: “As for news, we have nothing trustworthy to report to Your Excellencies since we are devoid of a dragoman acquainted with reliable people to obtain the news from.”43 On several occasions, Ragusan ambassadors had to pay for their soldiers, couriers, and servants to be released from the dungeon, some of whom must have been imprisoned for careless eavesdropping.44 The spy network sometimes included the services of locals from Istan- bul, who came across useful information and traded it like merchandise. Ambassadors had a special budget for espionage services and information.45 In 1765, an unidentified local person, who was used as a valuable source of information, required three barrels of the best wine in exchange for just one favor; the Ragusans sent even more, in anticipation of his future services.46 People of Ragusan descent, mainly Ottoman subjects, were also employed: for instance, an illegitimate son of the Ragusan nobleman Pavao Caboga, who converted to Islam and lived in Istanbul, proved to be much inclined toward the republic.47 Tribute ambassadors and other representatives were committed to reporting on a variety of details from Istanbul. They wrote about anything and everything: Ottoman dignitaries and foreign envoys; dismissals from the Porte; the current favorites of the sultan and the most distinguished of dignitaries; marriages; personalities; tendencies; the sultan and his ministers’ plans, including those of the newly appointed Bosnian governors and sancakbeyis of Herzegovina; dignitaries of Bosnian descent; events in Ottoman provinces, even the farthest ones in Asia, followed by those in

41 DAD Let. Lev. vol. 69, fol. 10 items 6, 8, 12, 19, 26. 42 These reports are filed in archival documents under the name Denunziones. DAD AD 17, vol. 1776, no. 3. 43 DAD AD 17, vol. 1849, no. 7. 44 DAD Cons. Rog. vol. 102, fol. 53; vol. 173, fol. 146v; vol. 174, fol. 194. 45 The special expense reports of the 1677 tribute ambassadors list 50 thalers intended for spies (DAD AD 17, vol. 1839, no. 86). Also see DAD Let. Lev. vol. 41, fols. 114, 157, 160v; AD 17, vol. 1839, no. 19. 46 DAD Let. Lev. vol. 86, fol. 42. 47 DAD Let. Lev. vol. 42, fol. 50. Pavao Caboga’s son was not the only descendant of a Ragusan nobleman to convert to Islam. Leonard Niko Giorgi’s son Miho converted during the first half of the sixteenth century (DAD Testamenta Notariae, vol. 39, fol. 134). 200 vesna miović

Wallachia, Moldavia and Transylvania; trade and exchange rates; crops and grain prices; plague epidemics; illnesses and deaths; fires; courses of the Ottoman fleet and army; wars and rebellions, and, in addition, the heads of killed rebels exposed in Istanbul.48 In 1730, Ragusan authorities demanded ambassadors to write only to us, in detail, pay special attention to describing the new sultan’s spirit and temperament, followed by those of the ministers—particularly of the grand vizier and his deputy (kethüda), and others—that our ambassadors usually have to come into contact with. What can be expected of the grand vizier? What can we hope to get from others? Which ones are guaranteed to keep their positions? Which are the ones that seem to be losing theirs? Name the new and the old ones among dignitaries. Which ones are in the sultan’s favor? What are we to expect from the new government?49 The Republic of Dubrovnik used a variety of codes in order to send and receive spy information.50 One of them, the so called Cifra di Constanti- nopoli, is considered to have been the most frequently used, clearly illus- trating the importance of Istanbul as a source of information.51 Moreover, in the correspondence between Ragusan authorities and a Ragusan confidant, the sender’s signature and the recipient’s address usually belonged to a private person, sometimes one who did not even exist.52 Confidential mail was delivered by trustworthy couriers.53 Depending on the importance of the mail they carried, couriers were classified either as regular (in the archival sources: corriere ordinario, a piedi) or emergency (corriere espresso, a cavallo, apposta).54 The most confidential news was “carried” to Dubrovnik in a special way: “do not take a risk in writing

48 DAD AD 17, vol. 1782, no. 18; vol. 1814, no. 4; vol. 1818, no. 3a; vol. 1827, no. 5; vol. 1850, no. 5. 49 DAD Let. Lev. vol. 72, fols. 219–229. 50 Bogdan Krizman, “Diplomatska šifra Dubrovačke Republike,” Arhivist 2 (1951): 49, 50. 51 DAD AD 18, vol. 3277/2, no. 108. For more on the Cifra di Constantinopoli as an advanced numerical code, see Bogdan Krizman, Diplomati i konzuli u starom Dubrovniku [Diplomats and consuls in the old Ragusa] (Zagreb, 1957), 110–121; Krizman, “Diplomatska šifra,” 48–50; Krizman, O dubrovačkoj diplomaciji, 61, 62. 52 Krizman, O dubrovačkoj diplomaciji, 62. For instance, Ragusan authorities addressed their letters to “Beniamin Saidini in Saida,” “Ortensio Torabbi in Pera,” “Israel Bonivades Quardi Dios in Galata,” and so on. Ragusan confidants wrote back to the authorities using addresses like “Ferdinand Nicoboli in Pesaro,” “Josef Garzeros in Macerata,” “Ottavio Luminelli in Ancona,” see DAD AD 18, vol. 3277/2, no. 143, 192, 209. 53 Žarko Muljačić, “Pomorske i kopneno-pomorske poštanske veze starog Dubrovnika” [Postal connections of old Ragusa over the sea and the sea and land] Naše more 5–6 (1962): 185; DAD Let. Lev. vol. 44, fols. 170v, 251–253v; vol. 45, fols. 21v–23; vol. 46, fols. 155, 234; vol. 48, fols. 107v, 108; vol. 49, fol. 75; vol. 50, fols. 105v, 106. 54 DAD Let. Lev. vol. 46, fols. 15v–18; vol. 55, fol. 30v; vol. 59, fol. 87v; vol. 70, fol. 194. diplomatic relations between the ottomans and dubrovnik 201 to us, but inform us by word of mouth, using your servants or other reliable individuals, capable of conveying your exact words to us.”55 It is well-known that the Ragusans regularly forwarded news on the Ottomans to the Papal Curia, Spain, Austria, and other European courts. News to Austria was sent directly by Ragusan representatives in Istan- bul to those in Vienna. In the eighteenth century, Ragusans made use of the Viennese Schuller bank, a loyal Ragusan bank partner with business connections all around the world, as their mediator in delivering news and diplomatic mail. It is no wonder that the manager of this bank was appointed as Ragusan chargé d’affaires in Vienna in 1804.56 Throughout the seventeenth century, particularly during the Cretan War, letters to the Papal Curia were sent from Dubrovnik and signed by “Lucio Pisone,” “Martino de Turra,” “Fabritio de Tersis.” The sender would always address the recipient with Beatissimo Padre, and continue that “a friend from Istanbul” had notified him of certain news through a letter with a certain date. Had it not been for the secretary of the Vatican Chan- cellery who marked them with “The Republic of Dubrovnik,” the letters themselves could not in any way have revealed who the actual sender was.57 “Friends from Istanbul” most often referred to tribute ambassadors. A comparison of the reports sent to Rome and the correspondence between the Ragusan authorities and tribute ambassadors makes it clear that Ragusans were committed to informing the Pope. In cases when they were unable to do so, they used honest reasons as excuses, and continued to inform him as soon as possible. In a letter dated 22 November 1647, “Martino de Turra” informed the Roman Curia that “Ottomans in Istanbul were doing everything possible in order to prevent letters from being sent; to everyone’s surprise, they even halted couriers of Ragusan ambassadors on two occasions, and seized their letters, both official and private ones.”58 Several days earlier, the Ragusan authorities wrote to the ambassadors ordering them to stop sending letters in cipher, be extremely cautious

55 Ragusan authorities’ letters to the 1664 ambassadors Andrija Pozza and Miho Resti (DAD Let. Lev. vol. 59, fol. 88). See also DAD Let. Lev. vol. 60, fols. 13, 13v; vol. 62, fols. 104v, 239, 239v; vol. 63, fol. 84; vol. 64, fols. 99, 99v. 56 Krizman, O dubrovačkoj diplomaciji, 139, 140; Krizman, “Diplomatska šifra,” 47; Vinko Ivančević, “Veze Joakima Stullija i Dubrovačke Republike s bečkim bankarima Schuller” [The connections of Joakim Stulli and the Ragusan republic with the Viennese bankers Schuller], Filologija 12 (1984): 497. 57 Ivan Dujčev, Avvisi di Ragusa, documenti sull’Impero Turco nel sec. XVIIe sulla guerra di Candia (Roma, 1935), XII–XV; Paolo Preto, I Servizi Segreti di Venezia (Milano, 1994), 235–239. 58 Dujčev, Avvisi di Ragusa, 118. 202 vesna miović when writing ordinary ones, and “stay quiet” until the Porte issued them a permit to leave for Dubrovnik. The diplomats, of course, continued to write, but only about irrelevant issues; this resulted in an interruption in the transfer of information to the Roman Curia that ended only in the summer of 1648, with the arrival of the new tribute ambassadors to Istanbul. “Martino de Turra” filled the void with news he continued receiving from a trustworthy person in Belgrade, most probably Diodono Bosdari, who had been ordered by the Ragusan authorities at the end of 1647 to report on everything worth knowing.59 The new pair of tribute ambassadors sent extensive reports on the dethronement and execution of Sultan İbrahim, and other events at the Porte. These reports also found their way to the Roman Curia through “Martino de Turra.”60 The Ottomans were very much aware of Ragusan spying activities; moreover, they claimed the republic itself was filled with spies working for the enemies of the empire. Indeed, a Spanish informant concluded that Dubrovnik was “the best place on Earth to obtain information on the sultan.”61 Apart from that, Dubrovnik also served as a place spies used to cross over into Ottoman territory: in 1591, the Bosnian governor was issued a ferman ordering loyal çavuş wardens to follow Ragusan ambassadors on their journey through Bosnia, since they were always accompanied by infidel spies.62 In spite of everything, the Ottomans were quite confident in receiving Ragusan news about activities in the Western world. Ragusans often remonstrated about their dedication to the Ottomans, going as far as claiming that “[w]e pay not one, but two tributes, since we always have to spend great amounts of money on our people around the world in order to gather information on events, and then forward them to the blessed Porte.”63 Owing to the scrolls of Ragusan reports, the Bosnian governor was superbly informed on all events in the world.64 The sultan issued

59 DAD Let. Lev. vol. 49, fols. 222v–224v. 60 Dujčev, Avvisi di Ragusa, 119–143; DAD AD 17, vol. 1830. Cf. Dujčev, Avvisi di Ragusa, 140–143 and DAD AD 17, vol. 1830, no. 16. 61 Mirjana Polić Bobić, Među križom i polumjesecom: Dubrovačke dojave španjolskom dvoru o Turcima u XVI. stoljeću [Between the cross and the crescent: Ragusan reports to the Spanish court regarding the Ottomans in the sixteenth century] (Zagreb, 2000), 13. 62 DAD AD 17, vol. 1826, no. 1; Nicolas H. Biegman, “Ragusan Spying for the Ottoman Empire,” Belleten 27 (1963): 239. 63 Biegman, The Turco–Ragusan Relationship, 129. 64 During the Austrian–Ottoman war 1737–1739, Ragusans were informing the governor of the Bosnian eyalet on the activities of the Austrians, the French, the English, and the reverberations of the war in Europe. During 1762 and 1763, they sent him reports diplomatic relations between the ottomans and dubrovnik 203

special fermans to ambassadors, confirming that he had received the news and was expecting new ones.65 The credential letters given by the sultan to the ambassadors contained his request to the Ragusan government to keep him informed on all activities occurring in areas close to Dubrovnik, and anything else worth knowing.66 His requests were rather specific at times, for example in 1587, when he asked for information on the Span- ish fleet.67 The Porte’s dignitaries had equally specific requests.68 Besides, they always expected tribute ambassadors to share all of the latest news on Western countries. In 1619, for instance, the grand vizier greeted them with the following questions: how strong was the Spanish fleet; what about the Italian fleet; could they possibly unite; were France and Flanders preparing their fleets; how many ships could Flanders arm, etc.69 Since the ambassadors always arrived at the Porte bearing the latest news, the silence that occurred whenever it was not in their interest to speak was considered suspicious. They then had to “reveal” news of no importance just to get by. In 1771, when the Russian fleet arrived in the Mediterranean, it was estimated that informing the Porte any further would pose too big a risk. Ragusan authorities ordered tribute ambassadors to leave Istanbul as soon as possible. In order to cover up for the silence, they “confided” to the Porte’s ministers that the Russian fleet had arrived in the Mediterranean—a fact that did not involve any danger since it was something everyone already knew.70 Ragusans frequently used news from the West as a diplomatic tool to buy benevolence and services at the Porte. From 1615 to 1620, during

on the developments of the Seven Years’ War (1756–1763). Generally speaking, they kept him informed on news of the world and routes of Christian fleets: DAD Copia Lettere Diverse, vol. 2, fols. 53v, 56v–58v, 59, 71, 71v, 109–110, 112, 112v, 115v, 116, 118v, 119, 122v, 123, 128, 128v, 130v, 131, 135v, 136, 139–140, 145v–146v, 158, 158v, 161–162, 259–260v; vol. 3, fols. 47, 56; vol. 4, fols. 380, 381, 393. Let. Lev. vol. 69, fols. 156v–159, 193–197v; vol. 70, fols. 135v–140v; vol. 75, fols. 122–124. DAD Acta Turcarum, vol. B 10, no. 10; vol. B IV, no. 168, 178, 181, 185; vol. B 15, no. 97. 65 Such fermans were issued to Ragusans by Sultans Beyazit II in 1500, Süleyman I in 1541, 1547, and 1551, and Selim II in 1570 and 1571. See Biegman, “Ragusan Spying,” 237–255; Ante Vučetić, “Spomenici dubrovački” [Ragusan documents], Srđ 5 (1906): 572–574. 66 Miović, Dubrovačka Republika, 301 (no. 1168), 337 (no. 1433), 338 (no. 1439), 389 (no. 1834), 399 (no. 1915). 67 Biegman, The Turco–Ragusan Relationship, 129. 68 In 1706 Ragusans sent news on the Venetian fleet at the defterdar’s request (DAD Let. Lev. vol. 67, fols. 220–224v). In 1712, they sent an emergency courier with news for the grand vizier (DAD Let. Lev. vol. 69, fols. 43v, 44). 69 DAD AD 17, vol. 1819, no. 24. 70 DAD Let. Lev. vol. 92, fols. 34v, 35. 204 vesna miović the conflict with Spain, Venetians continually plotted against Ragusans, accusing them of having entered an alliance with Spaniards. The Porte’s ministers were quite irritated, and the ambassadors patiently tried to mol- lify them with news on the Spanish fleet that they had obtained through Dubrovnik from Ragusan confidants in Italy. They finally succeeded in 1619, the exact moment they managed to win over the grand vizier with their ongoing reports. He no longer paid any attention to Venetians; moreover, he ordered them to stop plotting against Ragusans.71 At the same time, Ragusan authorities granted an expense of 300 thalers to a pair of ambassadors so they could acquire Oriental items: souvenirs that they planned on giving to Ossuna, the viceroy of Naples, the initiator of the current conflict with Venetians.72 In another case in 1655, the ambassadors could not obtain a permit to export grain from the empire, but ultimately they managed to break the grand vizier’s resistance by providing him with details about the activities of the French fleet against Naples. The Ragusan authorities proceeded to instruct the ambassadors: in case the French consul discovered what they had done, he should be told in no uncertain terms that they had no other choice, they needed Ottoman grain as they could not get Neapolitan grain because Naples had been attacked by the French.73 Finally, it is important to stress that not all the information accumulated in Dubrovnik may be characterized as intelligence in the narrow sense. Part of it consisted of news that Ragusans managed to gather and distribute rather fast. This way Ragusans efficaciously filled much of the media gap in this part of the world. That also answers the question why Ragusans, due to their double espionage, i.e., offering information to opposing sides, sometimes faced the fury of both East and West, but never with serious consequences: they gained more than they lost from Ragu- san double intelligence. Ultimately, Ragusans made sure that other states understood that the Republic was necessary for them the way it was.

71 Vuk Vinaver, “Turska i Dubrovnik u doba španske invazije Jadranskog mora (1617– 1619 godine)” [The Ottoman empire and Ragusa in the period of the Spanish invasion of the Adriatic (1617–1619)], Istoriski glasnik 1–4 (1952), 21–60; Foretić, Povijest, vol. 2, 85–89; Stjepan Ćosić and Nenad Vekarić, “Raskol dubrovačkog patricijata” [The split of the Ragu- san patriciate], Anali Zavoda za povijesne znanosti u Dubrovniku 39 (2001): 317–318. 72 DAD Let. Lev. vol. 43, fol. 124v. 73 DAD Let. Lev. vol. 56, fols. 59v–61v. diplomatic relations between the ottomans and dubrovnik 205

Diplomatic Methods: “Poverty” and Tears

During the sixteenth and seventeenth centuries, only Ragusans were permitted to sell salt to Ottoman subjects in Herzegovina. In the territory ranging from Split to Durrës, the Ragusan harbor was the main port for East–West transit trade for a long time; thus, Ragusan merchants enjoyed a number of extensive privileges in the Ottoman lands. These facts, together with the Ragusan double intelligence, shaped the power of Dubrovnik, which must have been obvious to all and must have led them to the conclusion that the republic was immensely rich. Indeed, Dubrovnik’s wealth was a fact, but Ragusans diligently tried to hide it. Istanbul was the main stage where, for specific reasons, they tried to represent themselves as completely powerless and poor, yet the oldest and the most faithful Otto- man tribute-payers. The tribute itself was a strong object of the republic’s diplomatic manipulation: Ragusans paid tribute and in exchange they expected to be protected not only against the Venetians, but also against hunger and their alleged poverty. On account of the tribute, Ragusan diplomats at the Porte regularly gained the right to buy great amounts of Ottoman grains and livestock. Each time the republic’s salt or transit trade monopoly was threatened, Ragusans demanded Ottoman protection, as otherwise “they would perish.”74 Essentially, the alleged Ragusan poverty was meant to provoke protective inclinations toward the Republic of Dubrovnik in the Ottomans. The main actors in this “poverty drama” at the Porte were the Ragusan tribute ambassadors: the delivery of tribute was regularly used to emphasize that “12,500 ducats were drained from the Ragusans’ blood.” Overall, the ambassadors were expected to skillfully outwit their opponents with all kinds of dissimulation in order to achieve a diplomatic goal. The Ragu- san authorities issued the following warning to one of them: We find it extremely difficult that you are accomplishing absolutely nothing. If all you are doing is standing there and doing nothing but watching and waiting, our efforts and expenses were all in vain. Negotiations with Turks require skill. This is why we sent you there, not simply to watch and wait.75 The ambassadors were even supposed to cry if necessary in order to fulfill their diplomatic aims. It seems that Ragusan ambassadors could literally cry rivers whenever they had to. They were trained to cry on cue,

74 Miović, Dubrovačka Republika, 100–110. 75 DAD Let. Lev. vol. 96, fol. 55. 206 vesna miović

according to instructions explaining in front of whom, where, and on what words exactly they should start crying. One could suspect that the tears mentioned in letters written by Ragusan authorities and ambassadors were simply a metaphor suggesting that they should perform convincingly and dramatically. Any doubts, however, are cast aside by the dispatch of Sekundo Gozze, the tribute ambassador of 1678, who claimed that he had shed so many tears in front of the Grand Vizier Kara Mustapha that they drenched his suit. Other ambassadors also cried, but Sekundo Gozze used his crying skill to grotesque proportions. According to his own letters, he shed tears always and everywhere. He cried so much that merely the mention of his name evoked nausea among Ottoman dignitaries. When he wanted to talk to one of them yet again, the dignitary replied: “I will see him, but if he plans on crying, he better not come.” Faced with the kethüda of Kara Mustapha together with three other Ragusan ambassadors, the kethüda pointed in his direction and asked: “Is this the crying one?”76 Whenever humbleness and tears turned out to be ineffective, the ambassadors were to change tactics immediately, and were supposed to act as stubborn and daring diplomats, prepared to risk their own lives in order to achieve their goal. If necessary, they would also drop a subtle hint that, in the event that the Ottomans failed to be benevolent, all Ragusans would be forced to flee Dubrovnik (leaving it to the Venetians)—nevertheless, this threat to leave the republic was reserved for critical situations as the most dramatic of diplomatic methods. On other occasions, Ragusan diplomats resorted to misinforming the Porte. For example, at the beginning of the eighteenth century, they tried to persuade the Ottomans to decrease the tribute the republic paid. Ragu- san diplomats of the time did their best to convince the sultan that the city could not bear to submit the yearly tribute any longer. The Porte delegated the governor of the Bosnian eyalet, Dubrovnik’s closest neighbor, to inspect these claims. Bribed by the Ragusan diplomats, the Bosnian governor sent a report in which he confirmed the poor situation of the city. On the basis of this report, the Porte agreed to decrease the republic’s tribute and determined that Ragusans would have to pay an amount of 12,500 gold coins every three years until their situation improved. The tribute would then have to be delivered annually again. Several times during the eighteenth century, the Porte’s ministers investigated Dubrovnik’s real

76 DAD AD 17, vol. 1834, no. 7; vol. 1849, no. 11, 13, 18. diplomatic relations between the ottomans and dubrovnik 207 financial situation, but they made the same mistake every time, request- ing the Bosnian governor to make an estimate, unaware that Bosnian governors were ready to offer political services to the republic in Bosnia as well as at the Porte. In exchange, Ragusans would offer the governors what they needed, mostly ready money and various commodities such as clothes and food. With this ongoing exchange of favors, the Ragusan authorities and Bosnian governors were loyal neighbors.77 Nonetheless, the “poverty drama” remained the most important Ragu- san diplomatic method, one used on every possible occasion. Ragusans showed their alleged poverty by offering official gifts which were not at all precious: silver platters and fabrics given to the Ottoman diginitaries by the tribute ambassadors did not stand in comparison with the abundant gifts given, for example, by Venetians or Austrians. Moreover, Ragusan tribute ambassadors were obliged to prove their poverty not only at the Porte, but also at every step through the Ottoman territories. While Euro- pean diplomatic and consular representatives resided in luxurious palaces in Pera, a respectable quarter in Istanbul, the Ragusan government never even planned to build or purchase a palace; rather their ambassadors stayed in pitiful rented hovels instead. The expression of their alleged poverty was regulated by law, and neither the ambassadors nor members of their entourage were allowed to keep or buy a horse worth more than 100 ducats, an offense punishable by a fine of 500 ducats. In a letter to the ambassadors, the Senators explained the decree in detail: Our forefathers always took special care not to show luxury with their valuable horses and horse-adornments, to the Turks, here or elsewhere, especially at the Porte. They were always presented with misery and we constantly claimed that the tribute had been barely collected from our scarce sources. Do not, therefore, keep or buy any expensive horses. On your way around Istanbul or to the Porte, do not make use of the noticeable and valuable horses or horse-adornments, but be modest, as you should, so as to avoid rumors and unpleasantness.78 Indeed, in 1563 the ambassador Šimun Benessa was accused of “parading around Istanbul with two horses and three servants.”79 The duty to hide

77 Miović, Dubrovačka diplomacija, 189–194; Vesna Miović, Dubrovačka Republika u spisima namjesnika Bosanskog ejaleta i Hercegovačkog sandžaka [The Ragusan republic in the documents issued by the governors of the Bosnian eyalet and the sancak of Herzegovina] (Dubrovnik, 2008), 51–57; eadem, “Beylerbey of Bosnia and Sancakbey of Herzegovina in the Diplomacy of the Dubrovnik Republic,” Dubrovnik Annals 9 (2005): 37–69. 78 DAD Let. Lev. vol. 58, fols. 134v, 135. 79 Popović, Turska i Dubrovnik, 207–208. 208 vesna miović

Ragusan richness on pain of punishment is even described in Ragusan folk songs: according to one of these, the Ragusan authorities had one of their subjects hanged because he kept boasting about Ragusan wealth while in Istanbul.80

Conclusion

In the fifteenth century, the Ottomans estimated that the Republic of Dubrovnik could offer them a concrete benefit. Since Ragusan neutrality was the basis of that benefit, Ottomans left the republic to survive as a free state; thus, Ragusans found themselves in a position that was highly profitable, but also very hard to sustain. Ragusan authorities and the diplomatic corps went to great effort to show the importance of Dubrovnik to both the East and West and to reap profit in such complicated circumstances. Hence, Ragusan diplomacy was bold, daring, hazardous, and often followed the patterns of classical warfare in the true meaning of word. This kind of diplomatic battle was often fought under the disguise of humbleness—a truly challenging task. The centuries-long political turmoil, along with the specific geographical position of Dubrovnik, largely determined the nature of its diplomacy as an essential device of its political existence. For over three hundred years, Ragusans made every effort to hide their wealth by presenting themselves as the most loyal, oldest, poor- est, and, above all, utterly harmless tributary of the Ottoman Empire. Such a diplomatic strategy relied heavily on the talent of the Ragusan tribute ambassadors to act out in tears the role of impoverished tributaries living in strained circumstances.

80 Maja Bošković-Stulli, Pjesme, priče, fantastika [Poems, stories, fantasy] (Zagreb, 1991), 14. The manuscript of the poem, entitled Dubrovčanin Vid–Đivo Bričilo [The Ragusan Vid–Đivo Bričilo], can be found in Arhiv Instituta za etnologiju i folkloristiku (Archive for the Institute for Ethnology and Folklore), Zagreb, manuscript no. 267. Enemies Within: Networks of Influence and the Military Revolts against the Ottoman Power (Moldavia and Wallachia, Sixteenth–Seventeenth Centuries)*

Radu G. Păun

In 1590, the Venetian bailo Giovanni Moro wrote about the Ottoman tributary principalities of Moldavia, Wallachia, and Transylvania in these terms: These provinces are extremely rich in all sort of goods and they are well peopled, and their princes, who are Christians, would maybe make one day some useful service to Christendom, in order to free themselves from the tyranny of the Turks, since they have, besides their own forces—which could be considerable if unified—those of the Roman Emperor, of the king of Poland, and of the Muscovite ruler, who are, all of them, very powerful and enemies of the Ottomans.1 The most substantial problem, in Moro’s view, seemed to be the “rude nature” of the Orthodox Christians; but even so, he did not rule out the possibility that a general revolt against the Porte would finally happen, if some exceptional leader (capo) were to appear, one who was able to

* This work was supported by a CNCS grant—UEFISCDI, IDEI, PCE-2011-3-0309. I want to express my gratitude to my colleagues and friends Prof. Olga Katsiardi-Hering (Athens), Vera Tchentzova (Moscow), Gábor Kármán (Leipzig), and Ovidiu Olar (Bucharest), who kindly provided me with invaluable information and advice during the preparation of this paper. 1 “Sono quelle province abbondantissime di ogni sorte di viveri e ben popolate, e i loro principi, che sono cristiani, per liberarsi dalle continue tirannie de’turchi, potriano un giorno apportar qualche notabil servizio alla cristianità, avendo, oltre le proprie forze, che unite sariano molto considerabile, la forza dell’Imperatore, del Polono e del Moscovita, principi molto potenti e nemici della grandezza ottomana.” In Moro’s view, Moldavia “può mettere insieme 9,000 archibusieri a piedi e più di 25,000 uomini a cavallo armati secondo l’uso della cavalleria turchesca,” while Wallachia: “può valersi di mille buoni archibusieri a piedi e di dieci mila uomini a cavallo,” in such a way that, united with Transylvanian forces, whose number would reach “fino a 50,000 uomini a cavallo e 10,000 a piè,” it would result in a considerable army of “85,000 cavalli e 20,000 pedoni.” Eugenio Alberi, Le relazioni degli ambasciatori veneti al Senato durante il secolo decimosesto, series 3, vol. 3 (Flor- ence, 1855) [henceforth, Alberi III], 346. 210 radu g. păun engage his people in some noble and generous action against the Otto- man tyranny.2 Moro, like every Venetian ambassador in Constantinople, had enough knowledge of politics to understand the role of the capi in anti-Ottoman Christian uprisings, and it is precisely these capi and their networks that this chapter is devoted to, with special reference to the Moldavian and Wallachian revolts against the Porte.

The Facts

The first episode examined here occurred in 1574, when the Moldavian Prince Ioan the Terrible (Ioan cel Cumplit, 1572–1574) refused to pay the ordinary tribute and rebelled against the sultan. The Ottoman reaction was immediate: after a short campaign, Ioan was defeated and executed and the Ottomans put Petru the Lame (Petru Şchiopul, 1574–1591, with interruptions) on the throne—the first Wallachian prince to rule Moldavia. Twenty years later, this scenario repeated itself: Mihai the Brave of Wallachia (Mihai Viteazul, 1593–1601) and Aron the Tyrant of Molda- via (Aron Tiranul, 1591–1595, with interruption) ordered the massacre of the Turkish creditors and troops in Bucharest and Iaşi and declared war against the Porte. Aron soon lost his throne, while Mihai continued the war with Transylvanian and Austrian support. In 1599–1601, he succeeded to take Wallachia, Moldavia, and Transylvania under his rule, put himself under the Habsburgs’ suzerainty and fought the Porte until 1601, when he was killed by Giorgio Basta’s mercenaries. Revolts remained more or less cyclical later on as well: in 1620, the Croatian-born prince of Moldavia, Gaspar Graziani (1619–1620), a former Ottoman dragoman, rebelled against the sultan with Polish aid. Again, the Ottomans managed to restore order, and Graziani was assassinated by his own noblemen. Peace lasted for some forty years, but in 1659 a new revolt broke out in Wallachia, led by Mihnea III (1658–1659), in alliance with György Rákóczi II, prince of Transylvania (1648–1660, with interruptions), and Constantin Şerban, prince of Moldavia (1659–1661, with interruption). After some initial success, Mihnea fled to Transylvania, where he died

2 “Ma forse che, mossi dalla disperazione che qualche volta suol mettere ardire anche negli animi bassi, potrebbero un giorno, piacendo a Dio, darsi un capo che li guidasse a qualche azione nobile e generosa, con notabil pregiudizio di quell’imperio per levarsi il giogo dal collo e rimettersi in libertà,” ibid., 362. networks of influence and the military revolts 211 in obscure circumstances. The last episode of the story occurred in 1711, when the Moldavian Prince Dimitrie Cantemir (1710–1711) joined Peter the Great’s army against the sultan. On the Prut borders, the Ottomans crushed the Moldavian–Russian forces, Cantemir left his country and Rus- sia abandoned anti-Ottoman plans for some fifty years.3 Thus, between 1566 and 1711, the Moldavian and Wallachian princes, vassals of the Porte, rebelled at least five times against their overlord, which means, grosso modo, one uprising each thirty years. Revolt thus appears to be quite a common life experience transmitted by social actors directly to their offspring, since almost each generation of that time lived through at least one. In addition, as this retrospective overview indicates, revolt represented an effective political solution always within reach. At this point, the question is, are such phenomena only the result of specific political and military circumstances (abusive tribute increases, European military coalitions against the Porte, etc.) or, are they recurrent episodes of a more complex process reflecting a more general attitude toward Otto- man rule. Should we analyze these revolts one by one, as individual cases, or consider them in a serial perspective?4 I suggest the use of a serial and comparative approach of the revolts. In order to accomplish this, I take four factors into account. The first relates to the dynamics of the relationship between the Porte and the vassal principalities of Moldavia and Wallachia from the mid-sixteenth century onward. The second refers to the general European policy toward the Ottomans, whose main feature was, at that time, the quasi-permanent oscillation between the idea of “Crusade” and the “capitulations’ regime.” The third—and very important—point concerns the personal careers of the social actors who orchestrated the revolts, namely the capi Moro spoke about. Where did they come from? What did they think about Ottoman power? What did they hope for in these seemingly hopeless wars? Were

3 For a general survey of these events, see Nicolae Iorga, Istoria Românilor [The history of the Romanians], vol. 5, Vitejii [The braves], ed. Constantin Rezachievici (Bucharest, 1998), and vol. 6, Monarhii [The monarchs], ed. Ştefan Andreescu (Bucharest, 2000). 4 The revolts of the tributaries are rarely addressed in works concerning Ottoman history; see, for instance, The Ottomans and the Balkans: A Discussion of Historiography, ed. Fikret Adanir and Suraiya Faroqhi (Leiden, 2002); The Cambridge History of Turkey, vol. 3, The Later Ottoman Empire, 1603–1839, ed. Suraiya N. Faroqhi (Cambridge, 2006). An important study related to this topic has now been published by Olga Katsiardi-Hering, “Von den Aufständen zu den Revolutionen christlichen Untertanen des Osmanischen Reiches in Südosteuropa (ca. 1530–1821): Ein Typologisierungsversuch,” Südost-Forschungen 68 (2009): 96–137. I am grateful to Prof. Katsiardi for kindly giving me the opportunity to read the manuscript of her paper before it was published. 212 radu g. păun they completely unaware of the risk they took? Finally, yet importantly, we must also refer to the “spiritual atmosphere” of that time and to ideas on the destiny of the Ottoman Empire.

The Overlord and His Tributaries

Within a century following the conquest of Constantinople, Ottoman military expansion had completely changed the balance of power in Southeastern and Central Europe and established the Sublime Porte as the major power in the region. For Moldavia and Wallachia, the immediate consequence of this process was a considerable increase in Ottoman political control over them. As Viorel Panaite points out, after 1535–1538 the twin principalities started to be seen by the Porte as “conquered territories,” being therefore included within the “well-protected dominions” of the empire.5 One of the most relevant indicators of this situation is the generalization of the custom to appoint the rulers of the two principalities in Constantinople, by the sultan himself. Indeed, of the thirteen princes who reigned in Wallachia between 1545 and 1594, no less than eleven gained the throne in that way, while in Moldavia election by local nobility and nomination by the sultan alternated during the same period. It is also noteworthy that almost all the reigning princes of Wallachia were dismissed at least once in their careers, but only two of them regained their throne, while others died either under the executioner’s sword or in exile, or they converted to Islam. Other nominations, although decided and officially pronounced, never became effective, despite the significant amount of money pretenders poured into the coffers of the Porte. The “rotation principle,” as Karen Barkey calls it,6 began to apply to the twin principalities as well.

5 Viorel Panaite, Pace, război şi comerţ în Islam: Ţările Române şi dreptul otoman al popoarelor (secolele XV–XVII) [Peace, war, and commerce in Islam: The Romanian principalities and the Ottoman law of nations, sixteenth–seventeenth centuries] (Bucharest, 1997); idem, The Ottoman Law of War and Peace: The Ottoman Empire and Tribute Payers (Boulder, CO, 2000). See also his contribution in this volume, as well as Tahsin Gemil, Românii şi otomanii în secolele XIV–XVI [Romanians and Ottomans, sixteenth–seventeenth centuries] (Bucharest, 1991); Mihai Maxim, Ţările Române şi Înalta Poartă: Cadrul juridic al relaţiilor româno-otomane în evul mediu [The Romanian principalities and the Sublime Porte: The juridical framework of the Romano–Ottoman relations in the Middle Ages] (Bucharest, 1993). 6 Karen Barkey, “In Different Times: Scheduling and Social Control in the Ottoman Empire, 1550 to 1650,” Comparative Studies in Society and History 38, no. 3 (1996): 460–483; networks of influence and the military revolts 213

In this context, it seems fair to assume that the main scene of the decision-making process moved gradually to Constantinople, where candidates to the princely thrones and/or former princes dismissed by the sultan spent their time and resources trying and hoping to gain, or regain, the master’s favor. A pool of pretenders formed, employed by the sultan and his circle as instruments of pressure and sources of continual provocation against one another.7 They all needed money in order to buy information and power, and information in order to better know how to use money and power. As many foreign observers stressed, very few things could be achieved in Constantinople without the intervention and influence of Ottoman high officials. Everything had a price, including the support of the sultan’s favorites, who were able to open the doors of the Saray and impose their opinion even on the Imperial Council.8 Non-Muslim elites in Constantinople played the role of cultural mediators in these affairs, for they already possessed a good perspective on the Ottoman system, and gradually became a kind of secondary “power elite” of the Porte.9 Thus, pretenders to the Moldavian and Wallachian thrones had to maintain strong connections within this milieu in order to (re)gain power.

eadem, Bandits and Bureaucrats: The Ottoman Route to State Centralization (Ithaca, NY and London, 1994). 7 This situation was also made possible by the relative vagueness of the rules regulating the succession to power in both Moldavia and Wallachia. The argument “of blood” (“os de domn”) allowed this right to practically anyone among the heirs of the late reigning princes, whether legitimate or not, real or pretend; see Radu G. Păun, “La circulation des pouvoirs dans les Pays Roumans au XVIIe siècle: Repères pour un modèle théorique,” New Europe College Yearbook 1998–1999 (Bucharest, 2001): 265–310. 8 Nicolae Iorga, Studii şi documente cu privire la istoria Românilor [Studies and documents concerning Romanian history], vol. 4, Legăturile Principatelor române cu Ardealul de la 1601 la 1699 [The relations of the Romanian principalities to Transylvania 1601–1699] (Bucharest, 1902), 190 (Dutch report, 13 October 1629); Documente privitoare la Istoria Românilor culese de Eudoxiu de Hurmuzaki [Documents concerning the history of the Romanians collected by Eudoxiu de Hurmuzaki], vol. 9, part 1, 1650–1747, ed. Ioan Slavici (Bucharest, 1897), 56 (Venetian report, 22 August 1654). 9 I am using Wolfgang Reinhard’s definition of “power elite,” see his “Power Elites, States Servants, Ruling Classes and the Growth of State Power,” in Power Elites and State Building, ed. Wolfgang Reinhard (Oxford, 1996), 5–7. I added the attribute “secondary” here because non-Muslim elites in Constantinople had no direct access to the decision-making process, although they could influence it. On this elite in Constantinople, see Radu G. Păun, “Well- born of the Polis: Ottoman Conquest and the Reconstruction of the Greek Orthodox Elites under Ottoman rule (15th–17th centuries),” in Türkenkriege und Adelskultur in Ostmittel europa vom 16. bis zum 18. Jahrhundert, ed. Robert Born and Sabine Jagodzinski (Ostfildern, 2013), forthcoming. 214 radu g. păun

In addition, Ottomans tried to envision a “security solution,” the ultimate aim of which was to integrate the future rulers of the twin principalities into Ottoman culture. One of the instruments to achieve this purpose was the institution of the hostages (sons or brothers of the reigning princes) and the custom of enrolling them in the müteferrika corps.10 However, this solution was not without faults. In fact, most of the aforementioned princes involved in revolts fell into the category of Constantinople-educated pretenders who had spent a great deal of their lives in the empire and had a good understanding of the Turkish language and customs. In short, they acted as “men of the system,” in the full sense of the term. What is more, their masters employed them as a kind of “crisis solution.” Ioan replaced Bogdan Lăpuşneanu (1568–1572), whose Polish connections were considered dangerous by the Ottomans, as they were engaged in a war against the Holy League. Aron was appointed instead of Petru the Lame, who had abdicated and left the country, and Mihai was seen as the most appropriate person to ease the internal conflicts in Wallachia in the context of a new war between the Porte and the Habsburgs. The same applies to Graziani, whose experience as a political negotiator recommended him to the Moldavian throne in the tense circumstances of Polish–Ottoman relations. As for Mihnea, he was sent to Wallachia to fight and capture the disobedient local prince Constantin Şerban (1654–1658) and to secure a bridgehead for a future campaign against György Rákóczi II. In a similar context, Cantemir had to replace Nicolaos Mavrocordatos (1709–1716, with interruption) in Moldavia with the task of keeping watch on Russian movements in the direction of the Ottoman territories. Why did they choose to revolt against their overlord? Why did they do so, and not the princes elected by the country, who had lived there all their life? Let us say first that the above list of anti-Ottoman revolts does not necessarily reflect the general attitude of Moldavian and Walla- chian society as a whole, as proven by the fact that almost all the revolting

10 As Constantino Garzoni noted in 1573: “Ci è in oltre la compagnia dei muteferica, che è di quattrocento, per la maggior parte figli d’uomini principali turchi et di cristiani ancora, e al presente vi sono due fratelli del Valacco e del Moldavo cristiani, quali servono il Gran Signore, aspirando a quel governo dopo la morte dei loro fratelli; e stanno in quella corte ostaggi della sincerità dell’animo loro. Tutti questi muteferica stanno sempre appresso la persona del Gran-Signore (. . .) e (. . .) sono veramente i gentiluomini della persona del Gran-Signore,” Eugenio Alberi, Le relazioni degli ambasciatori veneti al Senato durante il secolo decimosesto, series 3, vol. 1 (Florence, 1840) [henceforth Alberi I], 413–414. See also The Encyclopaedia of Islam, ed. by C.E. Bosworth et al. (Leiden and New York, 1993), vol. 7, 794. networks of influence and the military revolts 215 princes (the exception seems to be Mihai) fought against strong internal opposition. Ioan was betrayed by his nobles in 1574, while Aron never gained the support of the Moldavian elite, who granted him later the nickname “the Tyrant.”11 Graziani had an even sadder destiny; he finished his life assassinated by his own boyars. In 1659, the Wallachian magnates informed the Ottomans about Mihnea’s plans, and the prince eliminated a number of them in order to avoid massive betrayal, which happened nonetheless.12 Even in 1711, when the news announcing the arrival of the Russian emperor came, the Moldavian elite did not show much enthusiasm to fight the Ottomans. Given this, one could logically wonder what the relevance of these revolts is for our discussion. The answer lies outside the two principalities, namely in Constantinople and in the Ottoman Empire itself, for it was there that the idea of reconquest persisted among Christians. For many of them, this was an ongoing belief that became stronger with each Christian attempt to fight the “infidel.” On the other hand, “marciare verso Con- stantinopoli” (a term of Peter Bartl) appears to be not only a dream, but a permanent concern and a major orientation of European politics in the sixteenth–seventeenth centuries, blessed by the Holy See and grounded on the old Crusading tradition.13 Regardless of whether or not this orientation produced concrete results, it shaped a state of mind in which hope instigated action and action inspired hope, one in which final victory was always seen as possible, by both Christians under Ottoman rule and by the “new Crusaders” who thought themselves close to freeing the “Second Rome.” Hope and action (whether “real” or “imagined”) not only grew together as they fed one another, but continuously produced information concerning

11 A Polish source states that when Aron was about to start his revolt, some of his boyars fell on their knees in front of him with crosses in their hands (!) and begged him to abandon this idea, see Ilie Corfus, Documente privitoare la istoria României culese din arhivele polone: Secolele al XVI-lea şi al XVII-lea [Documents concerning Romanian history collected from Polish archives, sixteenth–seventeenth centuries] (Bucharest, 2001), 80 (1 December 1594). 12 Alexandru Ciorănescu, “Domnia lui Mihnea III (Mihail Radu), 1658–1659” [The reign of Mihnea III (Mihail Radu), 1658–1659], Buletinul Comisiei Istorice a României 14 (1935): 172–177. 13 See, among others, James Hankins, “Renaissance Crusaders: Humanist Crusade Literature in the Age of Mehmed II,” Dumbarton Oaks Papers 49 (1995): 111–207; Éva Bóka, “Crusading Tradition in the Seventeenth Century European Political Thought,” Südost- Forschungen 32 (1994): 39–59. 216 radu g. păun

“the real state of Turkish affairs.”14 An important mass of news came to and left Constantinople every day through multiple channels, ultimately reaching the important cities and courts of Europe. Logically enough, the reliability of this news was often doubtful and frequently one piece of information proved to be exaggerated or even biased, while another came to announce exactly the opposite; but it was precisely this alter- nation that fed a certain state of tension and maintained people’s interest in the matter. The same was true about what one could call “news of wishful thinking,” that is, news announcing events people hoped for, which did not necessarily have to be true to be considered as such and to fire the imagination.15 Information of various kinds, whether “true” or “false,” whether specifically conceived as a part of political propaganda or not, clearly functioned to persuade the public it reached, in the sense that it inspired them to action and guided their perception by fashioning a particular image of the Ottomans and of the attitudes to adopt toward them. It is in this precise context that one should consider the historical phenomena we are dealing with here, as we try to go beyond the borders of the positivist sense of history grounded on the logic of “real facts,” which counts victories and defeats, successes and failures. Something more complex and certainly more subtle than the “pure” and cold “strategic reason” seems to be at stake in these cases: something referring directly to the

14 Carl Göllner provides a list of sixteenth-century writings concerning the Ottomans in his Turcica: Die europäischen Türkendrücke des XVI. Jahrhunderts, vol. 1, 1501–1550 (Bucha- rest and Berlin, 1961), and vol. 2, 1550–1600 (Bucharest and Baden-Baden, 1968). Their number is, however, much larger, see Gregory James Miller, Holy War and Holy Terror: Views of Islam in German Pamphlet Literature (PhD diss., Boston University, 1994). See also Ivan Dujčev, Avvisi di Ragusa: Documenti sull’Imperio Turco nel sec. XVII e sulla Guerra di Can- dia (Rome, 1935); Zdenĕk Šimeček, “Ottoman Expansion in Czech Reports of the 16th and the Beginning of the 17th Century,” in Ottoman Rule in Middle Europe and Balkan in the 16th and 17th Centuries (Prague, 1978), 252–287; Daniel C. Waugh, “Diplomatic Channels as a Source for Foreign News in Muscovy” (paper delivered at the Conference of Mus- covite History, 1–4 September 1975, http://faculty.washington.edu/dwaugh/publications/ diplsourcesfornewsoxfordpaper1975). 15 In March 1572, news from Naples informed “que se dezía en Constantinopoli que los Reyes de Roscia, de Polachia de Carabordam (Moldavia) havían juntado 15 mill cavallos y 30 mill infantes, conjurandose con el Emperador contra el Turco.” Consequently, “toda la Romelia y provincias de Griegos y Arnautes esperan al Señor Don Juan con grandissimo desseo para levantarse contra el Turco,” Alexandru Ciorănescu, Documente privitoare la istoria românilor culese din arhivele din Simancas [Documents concerning the history of the Romanians collected from the Simancas Archives] (Bucharest, 1940), 56. Similarly, a Viennese newspaper of 5 March 1594 reported that an anti-Ottoman league including Mus- covy and Persia was already concluded; see Šimeček, “Ottoman Expansion,” 264. networks of influence and the military revolts 217 personality of the capi Moro spoke about, to their individual life experience in a world that proved to be much larger than the way in which we are used to understanding it today. When looking at their individual biographies, it appears that all the “revolting princes” shared contemporary ideas that animated this world, for they lived closely connected to the major information centers of their time and communicated with the networks feeding these centers. In most cases, their promotion to princes of Wallachia and Moldavia was precisely a result of these connections and networks that acted as lobbies and interest groups on the Constanti- nopolitan stage.

The Capi’s Personal Stories

Three of our actors came from the within the Ottoman system and followed the “classic” pattern of ascension in Ottoman bureaucracy. Gaspar Graziani owed his position to the powerful Iskender Pasha,16 who promoted him to Duke of Paros and Naxos and later to prince of Moldavia.17 From the Ottoman point of view, Graziani fully deserved this honor, for

16 Iorga, Studii şi documente, vol. 4, CXXXVI and CXXIXn1. Graziani’s career was presumably facilitated by the fact that Grand Vizier Murad Pasha (1606–1611) was himself of Croa- tian origin, as were some other Ottoman grandees, see Manfred Stoy, “Das Wirken Gaspar Gracianis (Graţianis) bis zu seiner Ernennung zum Fürsten der Moldau am 4 Februar 1619,” Südost-Forschungen 43 (1984): 59; idem, “Gaspar Graţiani, Fürst der Moldau 1619–1620: Seine marginale Rolle in den Anfängen des Dreißigjährigen Krieges,” Mitteilungen des Insti- tuts für Österreichische Geschichtsforschung 112 (2004): 306–315. See also Castilia Manea- Grgin, “Neobičan knez na moldavskom prijestolju: Hrvat Gašpar Graziani (1619–1620.)” [An unusual prince on the Moldavian throne: The Croat Gašpar Graziani, 1619–1620], Povijesni prilozi 30 (2006): 51–77. On Iskender Pasha, see Balázs Sudár, “Iskender and Gábor Bethlen: The Pasha and the Prince,” in Europe and the “Ottoman world”: Exchanges and Conflicts (16th–17th centuries), ed. Gábor Kármán and Radu G. Păun (Istanbul, 2013), forthcoming. 17 The translator of Na‘ima’s chronicle (1620) considers that Graziani’s various services “pleurent (sic!) à Skender Pacha, qui, le trouvant d’ailleurs intelligent, l’envoya deux ou trois fois en Allemagne, pour y traiter des affaires,” and then promoted him to Duke of Naxos. Later on, “il (Graziani) desira de se voir Vaivode de Bogdanie, ce qu’il obtint à force d’argent, et sous la caution de Skender Pacha, qui, l’ayant reconnu fidèle dans toutes les occasions, n’hezita pas de luy accorder cette grace,” Nicolae Iorga, Acte şi Fragmente cu privire la istoria Românilor adunate din depozitele de manuscrise ale Apusului [Documents and fragments concerning Romanian history collected from the Occidental archives], vol. 1 (Bucharest, 1895), 56. Graziani continually claimed his right to the throne as a reward of his services, see Iorga, Studii şi documente, vol. 4, CXXVI (Habsburg report, October 1615); Documente privitoare la Istoria Românilor culese de Eudoxiu de Hurmuzaki [Documents concerning the history of the Romanians collected by Eudoxiu de Hurmuzaki], vol. 4, part 1, 1600–1650, ed. Ioan Slavici (Bucharest, 1884), 365 and 376 (Venetian reports, 3 Sep- tember 1616, and 16 February 1619) [henceforth Hurmuzaki IV/2]. 218 radu g. păun he had played a key role in many diplomatic missions on behalf of the Porte, to Florence (1613), Belgrade (1613), Linz, and Vienna to negotiate a new peace arrangement between the Porte and the Holy Roman Empire (1614–1615). As to Mihnea, he appears in the eyes of foreign observers as a Greek gentleman, son of the late prince Radu Mihnea (died in 1626), and brought up in the house of the old Kenan Pasha. On a different occasion, the same source reports that he was brought up within the Saray, where he had the favor of the “sultana” (presumably the sultan’s mother) and of the sister of the sultan.18 It was through these connections, and especially thanks to the mediation of the Saray’s women, that Mihnea acquired the money to buy the throne.19 In Cantemir’s case, it was the milieu of amateurs of letters and arts that opened his way to high Ottoman political spheres,20 as arts and politics

18 “(. . .) allevato dentro il Seraglio, dove ha potuto tenere il favore della Sultana e della sorella del Gran Signore,” Venetian reports of 16 March 1658, and August 1658, Documente privitoare la Istoria Românilor culese de Eudoxiu de Hurmuzaki [Documents concerning the history of the Romanians collected by Eudoxiu de Hurmuzaki], vol. 5, part 2, 1650–1699, ed. Ioan Slavici (Bucharest, 1886), 42 [henceforth Hurmuzaki V/2], and Alexandru Ciorănescu, “Documente privitoare la domnia lui Mihail Radu (1658–1659) culese mai cu seamă din archivele Veneției” [Documents concerning the reign of Mihail Radu (1658–1659) collected mainly from the Venetian archives], Buletinul Comisiei Istorice a României 13 (1934): 28. If we are to trust Venetian agents, even after launching the revolt against the sultan, Mihnea “teneva pur anco corrispondenza con la Regina Madre del Gran Signore,” who “vorebbe veder rovinato il Primo Visir, e non tien bon animo verso il presente Gran Signore [Meh- met IV, 1648–1687], amando il secondogenito e bramandolo all’imperio esaltato,” ibid., 127 (6 December 1659). The Venetians were, however, wrong: Köprülü owed his position to the Queen Mother Turhan Sultan and to the influential faction gravitating around her, which supported him during the entire period of his office (1656–1661), see Caroline Finkel, Osman’s Dream: The Story of the Ottoman Empire, 1300–1923 (New York, 2005), 254. 19 This was 50,000 thalers with no interest from Atiké, and some 30 purses from two Jewish bankers of hers, see Ciorănescu, “Domnia lui Mihnea,” 97. New data about the Con- stantinopolitan connections of Mihnea are provided by Gábor Kármán, “The Networks of a Wallachian Pretender in Constantinople: The Contacts of the Future Voivode Mihail Radu 1654–1657,” in Europe and the “Ottoman World” (forthcoming). 20 This is why contemporaries state that Cantemir owed his throne to his knowledge of Ottoman customs and letters and to his musical skills. He arrived in Constantinople at the age of fifteen as a hostage on behalf of his father (1688). He stayed for three years and then went back to Iaşi, but for two years only; in 1693, after a reign of three weeks, he returned in Constantinople in order to live there for seventeen more years, first as an exiled prince, then as kapı kethüdası of his brother Antiochus (1695–1700). Unfortunately, no data are available about him between 1705 and 1710; see Paul Cernovodeanu, “Dimitrie Cantemir la Constantinopol (1700–1710): Activitatea politică şi culturală” [Dimitrie Cantemir in Con- stantinople (1700–1710): Political and cultural activity], in Omagiu Virgil Cândea la 75 de ani, ed. Paul H. Stahl (Bucharest, 2002), 143–152; Ştefan Lemny, Les Cantemir: l’aventure européenne d’une famille princière au XVIIIe siècle (Paris, 2009). networks of influence and the military revolts 219 proved inseparable during the Ottoman “Tulip Era.” The astrologer Nefioğlu, for instance, whom Cantemir knew very well, was on one hand a person initiated into the Palace’s secrets, and on the other, a close client of the Grand Vizier Rami Mehmed, who also appears to have been Cantemir’s patron for a time. The grand treasurers of the Palace, Davul İsmail Effendi and Latif Çelebi were also great amateur musicians, and it was on their demand that Cantemir composed his treatise on Otto- man music, in order to dedicate it to Sultan Ahmed III.21 When Davul became the Tatar khan’s kapı kethüdası in Constantinople, Cantemir saw his position considerably strengthened; soon after he was granted with the Moldavian throne, thanks to the khan’s intervention, as was the case of his brother Antiochus (1695–1707, with interruption).22 It seems likely that Grand Vizier Baltacı Mehmed, who shared an interest in music together with some of Cantemir’s friends, also had a say in this affair. Baltacı Mehmed needed a trustworthy client on the Moldavian throne, one who would be able to provide him with effective support in the war against Russia he was lobbying for with the ongoing help of the khan and the exiled Swedish King Charles XII.23 Cantemir’s connections within the milieu of Constantinopolitan diplomacy must have also contributed to Baltacı Mehmed’s choice.24

21 Iorga, Istoria, vol. 6, 417; Lemny, Cantemir, 69–71. Cantemir’s friendship with the court painter Abdülcelil Çelebi Levni certainly contributed to the growing sphere of his political connections, some of which Dimitrie had inherited from his father. 22 French diplomats in Constantinople saw Antiochus as a “créature du Khan des Tartares,” considering that Dimitrie himself “avait obtenu cette Principauté (. . .) à la rec- ommendation du Roy de Suède et du Khan des Tartares,” Documente privitoare la Istoria Românilor: Urmare la colecţiunea lui Eudoxiu de Hurmuzaki [Documents concerning the history of the Romanians. Continuation of Eudoxiu de Hurmuzaki’s collection], supplement 1, vol. 1, 1518–1780, ed. Grigore George Tocilescu and Alexandru I. Odobescu (Bucha- rest, 1883) [henceforth Hurmuzaki Suppl. I/1], 352 (23 February 1700) and 395; Iorga, Acte şi Fragmente, vol. 1, 309 (Italian and English reports, 8 and 19 December 1710). 23 Nicolae Iorga, “Carol al XII-lea, Petru cel Mare şi ţerile noastre (1709–1714)” [Charles XII, Peter the Great, and our countries, 1709–1714], Analele Academiei Române: Memoriile Secţiei Istorice, series 2, 33 (1910): 57–127; Lemny, Cantemir, 91–92; Ion Eremia, “Politica rusofilă a lui Dimitrie Cantemir—mit şi realitate” [The Russophile politics of Dimitrie Cantemir—myth and reality], in Dimitrie Cantemir: Fürst der Moldau, Gelehr ter, Akteur der europäischen Kulturgeschichte, ed. Klaus Bochmann and Vasile Dumbravă (Leipzig, 2008), 61–63. See also Veniamin Ciobanu, Carol al XII-lea şi românii / Charles XII et les Roumains (Bucharest, 1999). Demetrius’ father Constantin had a good reputation as a soldier, and this strongly recommended him to the Porte in 1685. 24 Dimitrie asserts that he had good relations with French ambassadors Châteauneuf (1689–1699) and Ferriol (1699–1710), but neither of them confirms his account. However, it is quite likely that he had some contacts with foreign diplomatic representatives in Con- stantinople; see Lemny, Cantemir, 65–67. 220 radu g. păun

All this indicates that alternative resources played a crucial role in one’s career in Constantinople. In the three cases presented so far, the positions they succeeded at acquiring within the Ottoman system made our protagonists highly interesting for “external” groups looking for firsthand data about Ottoman power and for competent mediators to use in political negotiations with the Porte. It was in this capacity that the future revolting princes succeeded in setting up their own networks of influence in the diplomatic milieu in Constantinople, which in turn increased their importance in the eyes of the Ottomans, who also needed skillful and reliable mediators in international affairs. Again, Graziani proved to be a relevant case, since he stood for a long time in the very medias res of political affairs, holding an official position in the Ottoman apparatus. His skills had already been “tested,” for he was recommended to the Ottoman dignitaries by his previous career as secretary and dragoman of the English embassies in Venice and Constantinople.25 Once he became an Ottoman diplomat, Graziani preserved his contacts with old protectors and friends, for whom he often served as a source of information.26 As proven by his entire career, he was such a great master in networking and trafficking information that it is difficult to say with whom he did not entertain relations: his correspondents included Pol- ish magnates, Emperor Ferdinand II, Cosimo II de’ Medici, grand duke of Tuscany, Don Pedro Téllez-Gíron de Osuna, vice-king of Naples and Sicily,

25 A later source saw him as “a man for special uses entertained amongst the English, whose brother and sister were both taken and admitted amongst the Turks, being formerly Christians of Transilvania, or Austria,” Thomas Gainsford, The Glory of England or a True Description of Many Excellent Prerogatives and Remarkable Blessings, Whereby Shee Triumpheth over All the Nations in the World (London, 1620; 1618), 192. Regardless of the author’s vague knowledge of Graziani’s family, the idea that he had been there “for special uses” is not wrong, since the Croatian dragoman knew—as the Habsburg agent Starzer noted—German, Croatian, Italian, English, Turkish, and Greek; see Stoy, “Die Wirken,” 58. It is his English connections that put him in contact with Moldavian issues, for he had also served the pretender Ştefan Bogdan, backed by the English Crown; see Laura Coulter, “An Examination of the Status and Activities of the English Ambassadors to the Ottoman Porte in the Late Sixteenth and Early Seventeenth Centuries,” Revue des Etudes Sud-Est Européennes 28, nos. 1–4 (1990): 69. 26 He presumably had the support of the English embassy for getting the throne; see Coulter, “An Examination,” 57 and 70. Baron von Molard recommended him to Emperor Mathias II for the same purpose: Andrei Veress, Documente privitoare la istoria Ardealului, Moldovei şi Ţării Româneşti [Documents concerning the history of Transylvania, Moldavia and Wallachia], vol. 9, Acte şi scrisori (1614–1636) [Charters and letters, 1614–1636] (Bucha- rest, 1937), 78 (16 April 1616). It is likely that he acted as a double or even multiple agent— one of the reasons why he was not too much appreciated by some of his contemporaries, see Stoy, “Die Wirken,” 116. networks of influence and the military revolts 221 and Philip III (1598–1621), king of Spain. In his desire to strengthen his position, he also intended to establish some ties with the Venetian milieu in Constantinople, by asking the bailo and the doge for the hand of the daughter of the Venetian Grand Dragoman Marc’ Antonio Borisi. It seems this marriage was never concluded, but we nevertheless find Bernardo Borisi, Marc’ Antonio’s brother, as Graziani’s proxy in Moldavia.27 When meeting the leaders of the Militia Christiana in December 1618, Graziani had already been introduced to this milieu and was knowledge- able of the projects fomented by the Duke of Nevers and his followers; it is more than likely that he had already met at least some of them during his diplomatic missions in the service of the Porte.28 New documents from Italian, Ragusan, and Viennese archives will probably shed more light on this topic and on Graziani’s connections throughout the Euro- pean diplomatic milieu. It is, however, fair to assume that his career was not accidental, and the revolt he launched against his overlord was far from being a neutral and isolated fact. On the contrary, Graziani’s actions should be regarded in the context of the various contemporary attempts to incite Balkan peoples to revolt against the Porte, orchestrated mainly by certain Italian rulers (Savoy, Mantua, Tuscany, Parme) and Spain, with

27 Andrei Pippidi, “Quelques drogmans de Constantinople au XVIIe siècle,” in idem, Hommes et idées du sud-est européen à l’aube de l’âge moderne (Bucharest and Paris, 1980), 133–159; Cristian Luca, Ţările Române şi Veneţia în secolul al XVII-lea [The Romanian principalities and Venice during the seventeenth century] (Bucharest, 2007), 71–83. 28 The military-religious order Militia Christiana was founded in 1618 on the initiative of Charles Gonzague, Duke of Nevers, and brought together warrior nobles from France, Ital- ian states, and the Holy Roman Empire, with the openly declared goal to deliver the Chris- tians from the Turkish “tyranny.” One of the leaders was Count Adolf of Althan, whom Graziani met during the 1614–1615 Habsburg–Ottoman negotiations. Marconnet, who was responsible for the east European region of the Militia, also belonged to Graziani’s contacts; see his letter to the Duke of Nevers (5 October 1618): Elvira Georgescu, “Trois princes roumains et le projet de croisade du Duc de Nevers,” Revue Historique du Sud-Est Européen 9, nos. 10–12 (1934): 340–341; Stephanos I. Papadopoulos, Η κίνηση τοῦ Δούκα τοῦ Νεβὲρ Καρόλου Γονζάγα γιὰ τὴν απελευθέρωση τῶν βαλκανικῶν λαῶν (1603–1625) [The action of Charles Gonzague, Duke of Nevers, to liberate the Balkan peoples, 1603–1625] (Thessalo- niki, 1966), 258–259. Graziani had contacts with former Wallachian Prince Radu Şerban (1602–1611, with interruption), starting with 1616, see Veress, Documente, vol. 9, 74–76 (28 March 1616). Some of Graziani’s old diplomatic connections (Charles Emanuel of Savoy, the Great Duke of Tuscany, the Duke of Mantua, and others) were among those who welcomed Nevers’ project from the very beginning; see Carol Göllner, “La Milice Chrétienne, un instrument de croisade au XVIIe siècle,” Mélanges de l’Ecole Roumaine en France 13 (1935–1936): 84–85; idem, “Beziehungen der Rumänischen Wojewoden Radu Şerban, Nico- lae Pătraşcu und Gaspar Gratiani zur Milice Chrétienne,” Revue des Etudes Sud-Est Euro- péennes 6, no. 1 (1968): 71–83; Ştefan Andreescu, ‘Restitutio Daciae,’ vol. 2, Relaţiile politice dintre Ţara Românească, Moldova şi Transilvania în răstimpul 1601–1659 [Political relations between Wallachia, Moldavia, and Transylvania, 1601–1659] (Bucharest, 1989), 9–11. 222 radu g. păun the blessing of the Holy See.29 Considering his position, it is certain that Graziani had good information about these attempts, as well as about the numerous uprisings they inspired against the Porte. Nevers’ Balkan policy,30 upheld by Rome and by the famous Père Joseph,31 provided him with an excellent opportunity for action. The fact that Graziani helped György Hommonai Drugeth in his 1619 attempt to overthrow Gábor Beth- len, prince of Transylvania (1613–1629), and the presence of the Polish noble Samuel Korecki alongside him during the final battle against the Porte indicate a common plan drawn up for a general anti-Ottoman war.

29 Angelo Tamborra, Gli stati italiani e il problema turco dopo Lepanto (Florence, 1961); Zdenko Zlatar, Our Kingdom Come: The Counter-Reformation, the Republic of Dubrovnik, and the Liberation of the Balkan Slavs (Boulder, 1992), 261–297. Two of Graziani’s prox- ies in Moldavia, Ragusan nobles Marino de Resti (Restić) and Marino de Gradi (Gradić) had played a role in the “Great Conspiracy” in Ragusa and spent some years afterward as mercenary commanders at Italian courts (Mantua, Savoy). For more detail, see Zlatar, Our Kingdom, 280–332; Stjepan Ćosić and Nenad Vekarić, “The Factions within the Ragusan Patriciate (17th–18th c.),” Dubrovnik Annals 7 (2003): 7–81, especially 12ff; idem, Dubrovačka vlastela između roda i države: Salamankezi i sorbonezi [Ragusan patriciate between kin- ship and state: The Salamancanists and the Sorbonnists] (Zagreb and Dubrovnik, 2005), 25–50. See also Peter Bartl, Der Westbalkan zwischen Spanischer Monarchie und Osmani schem Reich: Zur Türkenproblematik an der Wende vom 16. zum 17. Jahrhundert (Wiesbaden, 1970), 170ff. This aspect of Graziani’s politics has been so far ignored by historians; I thank Lovro Kunčević, who has drawn my attention to it. Restić had served in Bethlen’s army, but “recusando di gire con Bethlem Gabor, Prencipe di Transilvania, contro l’Imperatore, s’era ritirato in Moldavia,” Giovanni Battista M[ont]alb[an]i, Vera relatione e aviso de Moldavia . . . (Vienna, 1620), in Nicolae Iorga, “Manuscripte din biblioteci străine relative la istoria românilor: Al doilea memoriu” [Manuscripts from foreign libraries concerning Romanian history: Second account], Analele Academiei Române: Memoriile Secţiei Istorice, series 2, 21 (1899): 45. It was perhaps him that Graziani would have intended to grant with the command of the Uskoks, whom he wanted to bring to Moldavia as a personal guard; Hurmuzaki IV/2, 378 (Venetian report, 9 June 1619). 30 At the very moment Graziani ascended the throne, Achille de Harlay wrote from Constantinople to the king of France: “je le (Graziani) tiens pour homme qui facilement serviroit au dessein dont il se bruit tant de Monsieur de Nevers,” Hurmuzaki Suppl. I/1, 184 (10 February 1619). On Nevers’ Balkan actions, see Bernard de Xivrey, “Mémoire sur une tentative d’insurrection organisée dans le Magne de 1612 à 1619 au nom du Duc de Nevers,” Bibliothèque de l’Ecole des Chartes 2 (1840–1841): 532–553; Jacques Humbert, “Charles de Nevers et la milice chrétienne, 1598–1625,” Revue Internationale d’Histoire Militaire 68 (1987): 85–109; Papadopoulos, Η κίνηση. The fact that Patriarch Timothy was forced to excommunicate the participants of anti-Ottoman conspiracies is telling of the importance Ottomans accorded to these events; see Ioannis K. Hassiotis, “Spanish Policy Towards the Greek Insurrectionary Movements of the Early Seventeenth Century,” in Actes du IIe con- grès international des études du Sud-Est Européen, vol. 2, Histoire (Athens, 1970), 325. 31 Gustave Fagniez, “Le Père Joseph et Richelieu: le projet de croisade (1616–1625),” Revue des Questions Historiques 24, no. 46 (1889): 5–59; Papadopoulos, Η κίνηση, 97–147; Benoist Pierre, “Le père Joseph, l’Empire Ottoman et la Méditerranée au début du XVIIe siècle,” Cahiers de la Méditerranée 71 (2005) (http://cdlm.revues.org/index968.html, accessed on 1 April 2011). networks of influence and the military revolts 223

In this adventure, the former dragoman represented a precious strategic resource for the Militia’s leaders, since he never hesitated to use the networks he was supposed to build up and/or to maintain as an Ottoman diplomat for his own personal purposes.32 Mihnea’s case confirms this assertion, for he also tried to mobilize old personal contacts (especially with Venice and Transylvania), in order to shape a military coalition directed against the Ottomans. While in Con- stantinople, he had many intelligence assets within the Saray (“molte con- fidenze in serraglio”), thus he was not only interesting to, but also feared by both potential friends and competitors.33 If Venice chose to play very cautiously with him, Rákóczi’s agents in Constantinople constantly looked for his friendship and cooperation, and even advocated his claims to the Wallachian throne, which strategy proved to be efficient for Rákóczi, who gained a faithful and determined ally.34 This was not the case of Venice, whose caution continued even after Mihnea’s promotion to prince. However, available evidence indicates that Mihnea possessed accurate knowledge about the political events of the time and provided the Venetian diplomats with reliable information.35

32 Iorga, Studii şi Documente, vol. 4, CXXXVII; CXL–CXLI; idem, “Manuscripte,” 36n7; Manea-Grgin, “Neobičan knez,” 67–69. After ascending the throne, Graziani “manteneva continovamente spie alla Porta et alla Corte di Skender-Passà [his protector!] che lo raguagliavano di mano a mano d’ogni risolutione e conseglio di Turchi,” Giovanni Battista M[ont]alb[an]i, Vera relatione e aviso de Moldavia . . . (Vienna, 1620), in Iorga, “Manu- scripte,” 45. It seems that the Porte knew about Graziani’s participation in Militia Christi- ana for a long time, but preferred to postpone his dismissal until an appropriate moment came; see Iorga, Studii şi Documente, vol. 4, CXLIIIn3 (Dutch report, 24 August 1620). The main person responsible for this was Gábor Bethlen. 33 When he had to end his 1655–1656 secret negotiations with Mihnea, the Venetian secretary Ballarino was careful to do it in a very “smooth” manner, considering the influence this person enjoyed within the Imperial Palace; see Hurmuzaki V/2, 27 (25 February 1656). Evliya Çelebi also testifies that Mihnea had open access to the Imperial Palace; see Ciorănescu, “Domnia lui Mihnea,” 92–93. 34 These relations, in which the nodal point was Jakab Harsányi Nagy, go back at least to 1654, when Mihnea apparently swore loyalty to Rákóczi. It was through Harsányi that Mihnea established contacts with the Swedish diplomats sent to the Porte to support Rákóczi, and it seems that he played an important role in the Constantinopolitan affairs and intrigues; see Gábor Kármán, “A 17th Century Odyssey in East Central Europe: A Biog- raphy of Jakab Harsányi Nagy” (PhD diss., Central European University, Budapest, 2010), especially chapter 2. It is certain that Mihnea had contacts with the Habsburg ambassador Simon Reniger and maybe with Spanish and French agents as well; see Ciorănescu, “Dom- nia lui Mihnea,” 94. See also Kármán, “The Networks of a Wallachian Pretender.” 35 Mihnea continually claimed his own “intiera cognitione de’ fini e de’ pensieri de’ Turchi,” which came from the fact he “ha uscito della casa dell’Imperatore et (. . .) adesso ha buona conrispondenza con detto (. . .) et sa quello che il Turco pensa,” Ciorănescu, “Documente,” 9–10 (27 July 1658), and 19 (1658). It should be added that the prince knew 224 radu g. păun

During the secret negotiations with Ballarino in 1655–1656, he alluded to Cossack proposals for an anti-Ottoman alliance, but confessed that he had reservations about the matter.36 In his view, the preferable path was entirely different and much more secure, namely to support a revolt in Mani, where he maintained reliable contacts among the local leaders.37 If the Cossack alliance was a cliché at that moment,38 the idea of a revolt in Mani eventually proved to be well-founded: a violent uprising broke out there in 1659, that is, at the very moment Mihnea himself raised his sword against the sultan.39 Moreover, if we are to believe Montecuccoli’s testimony, a great plan for an offensive project (a combined attack on sea and ground) was drawn at the end of 1658, which would have opened the way for the Venetian navy to besiege Constantinople.40 Whether true or not, this fits perfectly with Mihnea’s proposals to organize a diversionary action in the Black Sea regions (“una buona diversione . . . per la parte di

Arabic, Persian, Turkish, Romanian, and Greek, and probably Italian and Latin, as he was known as a poet and erudite, idem, “Domnia lui Mihnea,” 89 (Evliya Çelebi’s testimony). 36 Hurmuzaki V/2, 22–23 (9 October 1655). The Venetian Senate remained cautious: ibid., 24 (2 December 1655); 27 (28 February 1656). See also Luca, Ţările Române, 177–202. 37 In his discussions with Ballarino, Mihnea pretended he was many times solicited by the “popoli di Brazzo di Maina con espresse speditioni fatte a lui di diversi Metropoliti, Arcivescovi, et del loro Patriarca per farlo risolver di condursi al loro governo,” Hurmuzaki V/2, 22 (9 October 1655). 38 The idea of an anti-Ottoman alliance between Moldavia, Wallachia, and the Cossacks appears many times in diplomatic correspondence. In 1654, the Polish envoy to Constanti- nople spoke about the Cossacks’ participation in a Russian-led campaign against the Porte, in order that the tsar could deliver the Greeks from the Ottoman yoke and restore the Byzantine Empire. He added that the tsar had sent spies (priests and monks) to instigate the Greek people to revolt; see Corfus, Documente, 281–282; Andreescu, Restitutio, vol. 2, 242ff; 272n92. This information can be now corroborated to that provided by Harsányi Nagy; see Kármán, “The Networks of a Wallachian Pretender.” 39 On this episode, see Athanasios E. Karathanassis, “Επαναστατικές κινήσεις στὴν Πελοπόννησο στὰ 1659” [Insurrectional actions in Peloponnesus in 1659], Πελοποννησιακά 8 (1971): 239–260; Νikolaos B. Tomadakis, “Ο Οικουμενικός Πατριάρχης Ιωαννίκιος Β΄ από Ηρακλείας” [The ecumenical Patriarch Ioannikios II of Heraclea], Λακωνικαί Σπουδαί 2 (1975): 127–161. 40 Ştefan Andreescu, “The Relations between Venice and the Romanian Principalities during the War of Candia,” in Italia e Romania, due popoli e due storie a confonto (secc. XIV–XVIII), ed. Sante Graciotti (Florence, 1998), 159–171, here 168–169. The idea was to use the Cossacks and the celali revolt in Asia. Luca expressed doubts about the authenticity of this project; however, in December 1658, the Bulgarian Catholic missionary Petar Parčević exposed a similar plan to the Viennese Court; see Luca, Ţările Române, 187. networks of influence and the military revolts 225

Mar Negro”)41 together with the Cossacks in order to facilitate the movement of the Venetian fleet toward the Dardanelles.42 All the immediate political circumstances indicated here must not be neglected when analyzing the revolts, which must be interpreted in relation to the concrete means the social actors were informed about and to their capability to use these circumstances for personal purposes. Gra- ziani, Mihnea, and Cantemir appear to be in a privileged position in this respect, precisely because of the status they held in the Ottoman system. However, the general framework in which revolts broke out appears to be much more complex than suggested by the sum of the “global” and “regional” events that happened to occur at the same moment in time, as it also encloses the structural elements of the system in which revolts took place and the tensions provoked by the dynamics of these structures. Mih- nea’s story unveils some of these elements and tensions, namely the hid- den mechanisms of the struggle for power between competing networks on the Constantinopolitan scene.43 According to the Venetian diplomats, the rebellion of the Wallachian prince was provoked by the intrigues of Panagiotis Nikousios, Mihnea’s personal enemy (“nemico suo”) and none other than Köprülü’s favorite client and a powerful patron in Greek Ortho- dox circles in Constantinople.44 The reason for these intrigues seems to be, among others, the rivalry between the two men for the Wallachian throne.45

41 Ciorănescu, “Documente,” 25 (Thomassy’s second account to the Senate, 23 August 1658); 28 (Venetian report from Rome, August 1658). 42 Ibid., 35–36 (Venetian report from Rome on Thomassy’s account in front of the Pope, 14 September 1658). The defensive weakness of the Dardanelles was well-known; see Fin- kel, Osmans’s Dream, 263. Venetian naval victories in spring 1659 perhaps played a role in Mihnea’s decision to revolt, see Ciorănescu, “Domnia lui Mihnea,” 146. 43 On this harsh competition for power, see Radu G. Păun, “Pouvoirs, offices et patronage dans la Principauté de Moldavie au XVIIe siècle: L’aristocratie roumaine et la péné- tration gréco-levantine” (PhD diss., École des Hautes Études en Sciences Sociales, Paris, 2003). 44 “Questa ribellione del Valacco fosse natta da timore di amari sentimenti del Primo Visir contro di lui fomentati da Panagiotti, onde stasse attendendo ben da vicino non pur e la depositione dalla carica, ma la perdita della vita,” Hurmuzaki V/2, 63 (30 September 1659). The words Ballarino assigned to Nikousios (for whom Mihnea’s act was a “stravaganza” provoked by a “pazzo e precipitoso furore”) are nearly the same that Patriarch Dositheus Notaras and the Cantacuzenoi employed. In fact, all of them belonged to the same interest group; see Andrei Pippidi, Tradiţia politică bizantină în ţările române în secolele XVI–XVIII [The Byzantine political tradition in the Romanian principalities, sixteenth-eighteenth centuries] (Bucharest, 1983), 211–215. Ballarino remains cautious; for him, the revolt “puo esser ben derivata da occulti uffitij e concerti,” which was at least partially true. 45 As Ballarino noted, “Panagioti aspira a quel posto e lo procura con le vie possibili,” but “si crede però che il Visir non vogli privarsi di lui,” Ciorănescu, “Documente,” 122 226 radu g. păun

Thus it becomes easier to understand why a “single-card player” had no chance within the Ottoman system, as competitors who paid more and/or had more direct and efficient access to the fountain of favor could appear at any moment and eliminate their weaker rivals. In that case, alternative resources proved to be of crucial importance for any social actor involved in the play, who sometimes even chose to use them against the establishment, as shown by Mihnea’s case.46 The same applies to the revolting princes who came from outside the Ottoman system, namely those who spent only the final part of their pretenders’ career in Constantinople (Ioan, Aron, and Mihai). In their cases, too, alternative networks played an influential role; not only in their promotion, but also in the preparation for the revolts. However, they followed an inversely oriented logic of political ascension, as they received access to the Ottoman decision-making milieu through secondary channels. In other words, their primary resources came mainly from outside and/or from the lower levels of the system. If the “insiders” (Graziani, Mihnea, Cantemir) played on their close proximity to the Palace, the “outsiders” compensated for this lack by mobility. When they traveled across Europe and/or throughout the Ottoman Balkans, they were harbored at European courts and/or noble houses, and therefore acquired a good knowledge of the international political milieu and succeeded in establishing valuable connections therein. Mobility also allowed them to look at the Ottoman system from afar and learn about the plans and strategies European powers had drawn up in respect to the Porte. Additionally, traveling throughout the Ottoman Balkans meant that they acquired direct knowledge of everyday life in the empire, as well as of the attitudes and hopes of local people.

(23 October 1659). This could explain why, when Greek notables met in January 1656 in order to prepare the replacement of the two reigning princes of Moldavia (Gheorghe Ştefan) and Wallachia (Constantin Şerban), the choice of the “conspirators” did not point to Mihnea (although he was present at the meeting) but to Radu Leon (issued from a Mol- davian Graecized family, prince of Wallachia, 1664–1669). Whatever the case, it is a fact that Nikousios finally managed to get the throne for his own friend and client, Gheorghe Ghica (1658–1659). 46 In the same vein, there is information indicating that Ioan’s rebellion was partially provoked by the exactions of Mihail Cantacuzenus “Shaytanoglu” over the Moldavian prince, who probably had ascended to the throne against Shaytanoglu’s will. Gerlach reports that Ioan refused to pay the 50,000 ducats Shaytanoglu had requested from him as a kind of protection tax; in response, the latter accused the prince of wanting to “hand over Moldavia to the Polish king,” which would have provoked Ottoman military intervention; see Stefan Gerlachs deß Aeltern Tage-Buch (Frankfurt am Main, 1674), 463–465; Ştefan S. Gorovei, Muşatinii [The Muşat dynasty] (Bucharest, 1976), 115ff. networks of influence and the military revolts 227

It was on the basis of these events, which troubled the whole of Europe in the 1560s and 1570s, and also by taking into account his well-established connections in various diplomatic circles, and his previous experience as a soldier and a direct witness of these events that Ioan broke his relations with his overlord in 1574.47 As documentary evidence shows, as soon as he ascended to the throne, he tried to activate his connections in Poland,48 Vienna,49 and maybe Moscow,50 in order to survey possibilities for political

47 I refer here to the period Ioan spent in Rhodes, which coincides with the events related to the battle of Lepanto. On the political context, see Keith M. Setton, The Papacy and the Levant (1204–1571), vol. 4, The Sixteenth Century from Julius III to Pius V (Phila- delphia, 1984); Ιoannis Κ. Hassiotis, Οι Έλληνες στις παραμονές της ναυμαχίας της Ναυπάκτου [The Greeks at the eve of the Lepanto battle] (Thessaloniki 1970), 48–76; Manoussos Manoussacas, “Lepanto e i Greci,” in Il Mediterraneo nella seconda metà del ‘500 alla luce di Lepanto, ed. Gino Benzoni (Florence, 1974), 215–243; Andrei Pippidi, “Les Pays Roumains et Lépante,” in idem, Hommes et idées, 25–53; José Manuel Floristán Imizcoz, Fuentes para la politica oriental de los Austrias: La Documentación Griega del Archivo de Simancas (1571– 1621) (Léon, 1988); idem, “Felipe II y la empresa de Grecia tras Lepanto (1571–78),” Erytheia 15 (1994): 155–190. It seems impossible that someone like Ioan, who had a direct interest in Ottoman affairs, had not learned about these events, nor about the Venetian attempts to mobilize his former protector Ivan the Terrible in a common crusade against the “infidel,” since news of this kind reached even the far distant England; see Calendar of State Papers: Foreign Series of the Reign of Elisabeth, vol. 10, 1572–1574, ed. Allan James Crosby (London 1876), 28, no. 83. It was in Rhodes that Ioan started to trade jewels for a while and to make acquaintances in Greek-Levantine merchant circles. It is possible that it was precisely these connections that helped him gain the throne. 48 Fearing a diplomatic conflict with the Ottomans, the Polish king Henri of Valois refused Ioan’s demand for assistance; see Nicolae Iorga, “La France dans le Sud-Est de l’Europe (I),” Revue Historique du Sud-Est Européen 12, nos. 1–3 (1936): 21–68, here 56. See also, Theodor Holban, “Henri de Valois en Pologne et les Roumains,” Bulletin de la Sec- tion historique de l’Académie Roumaine 23 (1942): 81–124. It should be noted that between 1558–1559 and in 1563, Ioan lived in Poland, under the protection of Jan Firley, voyvode of Lublin. 49 As voiced by the letters he sent to Hans Rueber, the chief commander of Upper Hungary; see Documente privitoare la Istoria Românilor culese de Eudoxiu de Hurmuzaki [Documents concerning the history of the Romanians collected by Eudoxiu de Hurmu- zaki], vol. 2, part 1, 1451–1575, ed. Ioan Slavici (Bucharest, 1891) [henceforth Hurmuzaki II/1], 630–631 (17 June 1572). See also Rueber’s report, ibid., 650–654, and Eugen Denize, “Românii, Liga Sfântă şi otomanii în epoca bătăliei de la Lepanto (1570–1574)” [The Roma- nians, the Holy League and the Ottomans in the age of the Lepanto battle, 1570–1574], Studii şi Materiale de Istorie Medie 21 (2003): 159–179. In 1563, Ioan joined the Habsburg armies hoping for Maximilian II’s help in obtaining the Moldavian throne. 50 After a failed attempt at the throne (1551), Ioan went to Ivan the Terrible’s court, where he married a daughter of the Knez Semion of Rostov. Years later, Ivan proposed him a new Russian marriage (Ioan’s first wife and their child had died), with the daughter of the knez Msistlavski, but this marriage remained unconcluded; see Gorovei, Muşatinii, 120ff. Given these facts, the hypothesis of a Moldo–Russian anti-Ottoman project of alliance appears to be quite plausible. The Ottomans feared this continually, Hurmuzaki II/1, 640–641 (Karel Rym’s report, 4 September 1572), and 692 (Rym’s and David Ungnad’s joint report, 20 June 1574); Dinu C. Giurescu, Ioan Vodă cel Viteaz [Prince Ioan the Brave] 228 radu g. păun and military alliances, to secure himself a place to escape if necessary, and to give himself a security option in order to avoid any attacks from behind. These networks proved to be efficient in at least one case, since the allies Ioan counted on were not the Cossacks, as most historians have thought, but private troops of Polish nobles.51 The experiences of Aron and Mihai provide more detail about the dynamics of influence networks, as well as about the role they played in the preparation of anti-Ottoman revolts. Before ascending the throne, Aron spent a good part of his life in Lemberg, one of the key centers of information traffic at the northern frontier of the Ottoman Empire, and in Prague, where he had the opportunity to meet, among others, Archduke Ferdinand, and the imperial chancellor Pezzen (who between 1578–1581 functioned as a secretary of the Habsburg embassy and later as Habsburg ambassador in Constantinople in 1587–1592, respectively).52 It was precisely at that point (1589) that Rudolph II conceived a new Holy League project, which was to include, besides Spain and the empire, Muscovy, some Georgian princes, and Persia.53 When the Habsburg–Ottoman hostilities broke out, Aron was already in Constantinople to initiate his first attempts to gain the throne. His peregrinations proved useful: during his

(Bucharest, 19662), 131; Matei Cazacu, “Familles de la noblesse roumaine au service de la Russie, XVe–XIXe siècles,” Cahiers du Monde Russe 34, no. 1 (1993): 211–226, here 213. Spanish diplomats took it for a fact, Ciorănescu, Documente privitoare la istoria românilor, 67–68 (letter from Naples, June 1574). 51 Mihnea Berindei, “Le problème des ‘Cosaques’ dans la seconde moitié du XVIe siècle: A propos de la révolte de Ion Vodă, voïévode de Moldavie,” Cahiers du Monde Russe et Soviétique 13, no. 3 (1972): 338–367. Ioan was not the first one to think about an anti- Ottoman coalition, see Ştefan Andreescu, Restitutio Daciae, vol. 1, Relaţiile politice dintre Ţara Românească, Moldova şi Transilvania în răstimpul 1526–1593 [Political relations between Wallachia, Moldavia and Transylvania, 1526–1593] (Bucharest, 1980), 46. 52 Ilie Minea, “Aron Vodă şi vremea sa” [Prince Aron and his time], Cercetări istorice 8–9, no. 1 (1932–1933): 149–150; Gorovei, Muşatinii, 128–129. 53 Josef Polišenský, “The Turk Question in European Policy of the 16th and 17th Centu- ries,” in Ottoman Rule in Central Europe, 222. After Lepanto, a strategic change took place, as military actions against the Porte were to be carried out mainly through the Balkan regions, see Angelo Tamborra, “Dopo Lepanto: lo spostamento della lotta antiturca sul fronte terrestre,” in Il Mediterraneo, 371–393. In Gregory XIII’s view it was the Polish– Habsburg union that had to be the core of the new league. Stefan Bathory’s exceptional position as prince of Transylvania and king of Poland (1575–1586) was supposed to allow new attempts to shape a large anti-Ottoman front, see Ludwik Boratyński, “Stefan Batory i plan ligi przeciw Turkom (1576–1584)” [Stefan Bathory and the plan of a league against the Turks, 1576–1584], Rozprawy Akademii Umiejętności: Wydział historiczno-filozoficzny 19 (1903): 197–347; Rodica Ciocan, “Etienne Bathory et l’idée de Croisade,” Balcania 8 (1945): 154–178. All this means that negotiations and crusade projects started to involve Vienna, Prague, Warsaw, and even Moscow, much more than Venice or Madrid, although the idea to coalesce Spanish and Austrian Habsburgs was much cherished by the Pope. networks of influence and the military revolts 229 stay in Transylvania, he met Prince Zsigmond, who became his friend and patron, at Báthorys’ court; some time later, Boldizsár Báthory recommended him to the very influential Edward Barton, British ambassador in Constantinople.54 Aron’s sojourn in Alba Iulia occurred at the same time as Mihai the Brave’s stay at the same court, and it is likely that the two met in Báthory’s entourage. While Aron acquired court experience in Poland, Prague, and Transylvania, Mihai spent his youth in Constantinople, and traveling through the Balkan area as a sheep merchant (celep) working for his influential Greek-born uncle Yani, before he made a brilliant political career in Wallachia under the reign of several princes and the protection of the same Yani.55 Once his princely origins were revealed, he had to leave the country, since every other pretender regarded him as an enemy. Like Aron, Mihai enjoyed Zsigmond Báthory’s favor in Transylvania, and he was also recommended to Barton by the same Boldizsár Báthory as Aron had been.56

54 Barton’s influence was so high that the Papal envoy in Poland thought that “quello ambasciatore si facea Turco, et (. . .) si dubitava che dovesse diventare un gran Bassà,” Documente privitoare la Istoria Românilor culese de Eudoxiu de Hurmuzaki [Documents concerning the history of the Romanians collected by Eudoxiu de Hurmuzaki], vol. 11, Acte din secolul al XVI-lea relative mai ales la domnia şi viaţa lui Petru Vodă Şchiopul, 1517–1612 [Sixteenth-century documents concerning mainly the reign and life of Petru the Lame, 1517–1612], ed. Nicolae Iorga (Bucharest, 1900) [henceforth Hurmuzaki XI], 309 (26 Janu- ary 1593); see also 316–317. Barton’s network deserves further study; see Ioan I. Podea, “A Contribution to the Study of Queen Elisabeth’s Eastern Policy (1590–1593),” in Mélanges d’Histoire Générale, ed. Constantin Marinescu (Bucharest, 1938), 423–477; Ludovic Demény and Paul Cernovodeanu, Relaţiile politice ale Angliei cu Moldova, Ţara Românească şi Tran- silvania în secolele XVI–XVIII [England’s political relations with Moldavia, Wallachia, and Transylvania, sixteenth-eighteenth centuries] (Bucharest, 1974), 27ff. Aron was also supported by Barton’s proxy, Paolo Mariani, see Podea, “A Contribution,” 451–452; Eric R. Dursteler, Venetians in Constantinople: Nation, Identity, and Coexistence in the Early Modern Mediterranean (Baltimore, 2006), 139–141. 55 Being a celep was a very lucrative business, see Bistra Cvetkova, “Le service des celep et le ravitaillement en bétail dans l’Empire ottoman (XVe–XVIIIe siècles),” Etudes Historiques 3 (1966): 145–172; Elena Grozdanova, “Bergleute (Madenci), Salzgewinner (Tuzcu) und Celeps als Bevölkerungsgruppen mit Sonderpflichten und Sonderstatus im Osmanischen Reich: Versuch einer vergleichenden Analyse,” Südost-Forschungen 56 (1997): 105–121. 56 Podea, “A Contribution,” 464–465; George D. Florescu and Dan Pleşia, “Mihai Viteazul—urmaş al împăraţilor bizantini” [Mihai the Brave, successor of the Byzantine emperors], in Scripta Valachica (Târgovişte, 1972), 161; Ştefan Andreescu, “Familia lui Mihai Viteazul” [Mihai the Brave’s family], in Mihai Viteazul: Culegere de studii, ed. Paul Cerno- vodeanu and Constantin Rezachievici (Bucharest, 1975), 225–241; idem, “Mihai Viteazul, Cantacuzinii şi marea bănie de Craiova” [Mihai the Brave, the Cantacuzenoi and the “great bănie” of Craiova], in idem, Restitutio Daciae, vol. 3, Studii cu privire la Mihai Viteazul [Studies about Michael the Brave] (Bucharest, 1997), 13–37. 230 radu g. păun

Barton acted as political protector and provider of funds for Aron based on the solid relations he had already established within the Constantino- politan milieu.57 His task was facilitated by Zsigmond Báthory’s intervention and by the relations Aron had already established with some Polish noble families58 and with high ecclesiastic figures like Patriarch Jeremias II Tranos and his close counselor Nikephorus “the Deacon,”59 that is, two people later found among Mihai’s partisans, too. In addition, there are some interesting connections between Aron’s and Mihai’s movements, the key to which would be, in my opinion, in the Cantacuzenoi’s network in Constantinople, to which both Aron and Mihai were closely linked, even if not as relatives, as some historians believed,60 at least as political clients, like a number of princes before them. A tradition in this respect was already created by Mihai Cantacuzenus “Shay- tanoglu,” and was continued by his relatives.61 After Shaytanoglu’ violent death in 1578, the affairs of the family were handled by Yani “the Ban,” who some historians believe was the brother of Shaytanoglu, while others consider him a relative through marriage. Whether a true Cantacuzenus or not, Yani inherited Shaytanoglu’s political and business connections

57 The reactions provoked by Aron’s dissmisal in 1592 show who his real supporters were and how strong they turned out to be. Indeed, a real uprising broke out when Aron’s creditors (and first of all the janissaries) saw the money they invested in Aron vanishing. Barton was not completely innocent in this regard and eventually succeeded to regain the throne for his client; see Minea, “Aron,” 172, 180n4, and 222. The amount of money at stake was between 1 and 1.5 millions gold coins; see Ciorănescu, Documente privitoare la istoria românilor, 97 (21 September, and 16 November 1591); Hurmuzaki IV/2, 157 (Venetian report, 12 September 1591). Barton vouched for his protégé in various transactions and even presented himself as a pledge of the prince in some trials: ibid., 177 and 179–180 (4 May and 20 June 1594). Aron was also backed by the French ambassador, the sultan’s favorite wife, two Jewish bankers, and some Greek merchants; see Podea, “A Contribution,” 452; Eric D. Tappe, Documents Concerning Romanian History (1427–1601) Collected from the British Archives (London, The Hague, and Paris, 1964), 61, and 68–69. 58 His sister Marica married into the Orzechowski family, while the daughter born from this union became the wife of the noble Vasilij Czołchanski, Gorovei, Muşatinii, 131. 59 Hurmuzaki IV/2, 159 (Venetian report, 30 May 1592); Podea, “A Contribution,” 451–452. 60 Historians presumed that Aron married a woman from the Cantacuzenus family. There is no direct evidence of this, although it seems plausible that some relation of kin- ship existed. With regard to Mihai, some historians believe he was a Cantacuzenus himself through his mother; see Florescu and Pleşia, “Mihai Viteazul”; Andreescu, “Familia”; idem, “Mihai Viteazul, Cantacuzinii”; Matei Cazacu, “Stratégies matrimoniales et politiques des Cantacuzène de la Turcocratie (XVe–XVIe siècles),” Revue des Etudes Roumaines 19–20 (1995–1996): 157–181. On the Cantacuzenus family, see Jean M. Cantacuzène, Mille ans dans les Balkans: Chronique des Cantacuzène dans la tourmente des siècles (Paris, 1992). 61 See Cazacu, “Stratégies”; Păun, “Well-born of the Polis.” networks of influence and the military revolts 231 and employed them in the same manner, which allowed the Ragusan Pascal de’ Marini Poli to note: In Constantinople the Grand Ban stayed on behalf of the prince, a clever man who is the prince’s friend and who has helped him very much to regain his throne with his wit and his money as well, because he is extremely rich; so, when he is in Constantinople there is no need for the prince to go there.62 Marini Poli paints here the portrait of the perfect broker, a kind of prince’s “double,” and so was indeed this “Greek,” brother of Mihai’s mother:63 a clever and “prudent” rich man, enjoying great prestige among his peers and the trust of his superiors, always able to handle difficult affairs and to act as a provider of funds when the situation called for it.64 He represented the perfect choice for any pretender to the throne, and it was not by accident that he held the office of chargé d’affaires (kapı kethüdası) in Constantinople for years, on behalf of many Wallachian and Molda- vian princes. He also did this for Aron, which shows that Aron and Mihai belonged to the same circle gravitating around Yani’s powerful figure. We understand now Patriarch Jeremias’ and Nikephorus’ proximity to both Aron and Mihai, as they belong in fact to the Cantacuzenoi’s network: the first one as an old client of the late Shaytanoglu, and the second as a nephew of Shaytanoglu’s sister.65

62 “A Constantinopoli per il principe sta Ban Grande, homo savio e amico di principe, qual in sua absentia l’a aiutato molto a tornar nel stato, si con sua prudentia, come anco con danari, ch’e richisimo, di modo che, stantte lui a Constantinopoli, domnia (i.e., the prince) non ge andara là,” Nicolae Iorga, “Contribuţiuni la istoria Munteniei în a doua jumătate a secolului al XVI-lea” [Contributions to Wallachian history in the second half of the sixteenth century], Analele Academiei Române: Memoriile Secţiei Istorice, series 2, 17 (1895–1896): 24. Reference is made here to the second reign of Mihnea “the Turk” (1585– 1591). Marini Poli was close to the Wallachian dynasty, as his brother married the prince’s cousin. The ban was the highest officer in Wallachia, the governor of the western part of the country. 63 “(. . .) avunculus, vel matris frater, gente Graecus, qui et ipse antea dicto Bahni (Ban) officio functus erat,” Iorga, “Contribuţiuni,” 82–83. 64 He defended his prince (at that time Petru the Lame) before the imperial council; see Nicolae Iorga, “Două texte greceşti privitoare la ţările noastre” [Two Greek texts concerning our countries], Revista Istorică 19, nos. 1–3 (1933): 6–9; Nicolae Bănescu, Un poème grec vulgaire relatif à Pierre le Boiteux de Valachie (Bucharest, 1912). Within less than two years Yani managed to borrow, on behalf of Petru the Lame, around 1,200,000 aspers from Jews, Greeks, and Ottoman officers; see Păun, “Pouvoirs, offices,” 306. His credit was so high that even when he had a debt of around 500,000 gold coins, he could still borrow more money; see Iorga, “Contribuţiuni,” 82; Hurmuzaki IV/2, 180. See also Hurmuzaki XI, 309–310 (after 29 January 1593). Yani also acted as a mediator during the 1590–1591 Ottoman–Polish negotiations, see Florescu and Pleşia, “Mihai Viteazul,” 145ff. 65 Constantine N. Tsirpanlis, “A Prosopography of Jeremias Tranos (1536–1595) and His Place in the History of the Eastern Church,” The Patristic and Byzantine Review 4, no. 3 232 radu g. păun

Yani’s successor in the leading position of the Cantacuzenoi’s network was the son of Shaytanoglu, Andronicus, who became Mihai’s first counselor soon after 1593.66 It was he who collected the large amount of money needed in order to buy the throne for Mihai and it is quite certain that the 1595 anti-Ottoman revolt was planned in his house rather than in Bucharest or Iaşi. The large extent to which the Cantacuzenus family and its relatives were engaged in Mihai’s revolt is telling. Besides Andronicus, who was one of the main actors of the play, at least three other members of the family (Andronicus’ brothers Theodor, Dumitru, and Ion) held key positions in Mihai’s political apparatus, while some relatives (Nikepho- rus and Dionysius Rhalle, Metropolitan of Trnovo, whose mother was a Cantacuzenus) and followers actively supported Mihai’s political actions.67 While it is highly doubtful that Mihai really dreamed of the reconquest of

(1985): 155–173; Nicolae Iorga, “Nichifor Dascălul exarh patriarhal şi legăturile lui cu ţările noastre (1580–1599)” [The patriarchal exarchus Nikephorus the Deacon and his relations with our countries, 1580–1599], Analele Academiei Române: Memoriile Secţiei Istorice, series 2, 27 (1904): 183–200; Ştefan Andreescu, “Exarhul patriarhal Nichifor Dascălul, cneazul Constantin de Ostrog şi Mihai Viteazul” [The patriarchal exarchus Nikephorus the Deacon, Prince Constantine of Ostrog and Mihai the Brave], in idem, Restitutio, vol. 3, 81–115. 66 A letter Patriarch Jeremias sent to the former Moldavian Prince Petru the Lame reveals the great favor Andronicus enjoyed at the Ottoman court, to such an extent that almost nothing was impossible for him, as proved by the “great things” he already accomplished; see Hurmuzaki XI, 370–371 (November 1593). When addressing the same Petru, Andronicus himself openly declared that it was him who “made the ban Mihai prince of Wallachia,” together with Nikephorus the deacon, who shared his indignation, seeing that the country had been ruled for long years by “the servants of his servants,” who all disap- pointed him in the end; Hurmuzaki XI, 373–374; new edition in Mihai Viteazul în conştiinţa europeană [Mihai the Brave in the European consciousness] (Bucharest, 1982), vol. 1, 66–67. See also Andreescu, “Mihai Viteazul, Cantacuzinii”; Cazacu, “Stratégies,” 175ff. 67 Andreescu, “Mihai Viteazul, Cantacuzinii,” 25–26, and 32; Nicolae Iorga, “Sfătuitorul bizantin al lui Mihai Viteazul: mitropolitul Dionisie Rali Paleologul” [Mihai the Brave’s Byzantine counselor: Metropolitan Dionysius Rhalle Palaiologus], Revista istorică 5, nos. 1–2 (1919): 26–35. Recent research seems to indicate, however, that Dionysius Rhalle (of Trnovo) and Dionysius Palaiologus (of Kyzikos) were two different persons; see Ioannes K. Hassiotis, “Ο Μητροπολίτης Κυζίκου Διονύσιος Παλαιολόγος καὶ η δράση του στη Δυτική καὶ την Ανατολική Ευρώπη κατά τον 16ο αιώνα” [The Metropolitan of Kyzikos Dionysius Palaiologus and his activity in western and eastern Europe, sixteenth century], Ηπειρωτικά Χρονικά 44 (2010): 575–680. At present I cannot establish which of the two (maybe both of them) was Mihai’s follower. Whichever is the case, it is certain that a good number of Mihai’s counselors, captains, and agents were of Balkan origin; see Emil Turdeanu, “Quelques ragusains auprès de Michel le Brave,” Etudes Roumaines 1 (1938): 41–74; Ştefan Andreescu, “Boierii lui Mihai Viteazul” [Mihai the Brave’s boyars] in idem, Restitutio, vol. 3, 339–419; Andrei Pippidi, “Notes et documents sur la politique balkanique de Michel le Brave,” Revue Roumaine d’Histoire 23, no. 4 (1981): 41–62; idem, “Résurrection de Byz- ance ou unité politique roumaine? L’option de Michel le Brave,” Revue des Etudes Sud-Est Européennes 13, no. 3 (1975): 367–378. networks of influence and the military revolts 233

Byzantium,68 as Orthodox people under Ottoman rule hoped, and as is suggested by the memory he left in Balkan folklore,69 it seems true that the principalities of Moldavia and Wallachia were dominated by the Byz- antine imperial family of the Cantacuzenoi, whose members still hoped that the Ottoman Empire would soon come to an end.70 It is clear, in my view, that Aron’s and Mihai’s promotion to princes of the twin principalities was the result of the combined action of at least three networks: the Cantacuzenoi and their followers; Báthory’s nexus of political connections, which would have driven him to restore the ancient Dacia, as the humanists of his court used to claim;71 and Barton’s network, whose main target was to preempt Catholic advancement in the region and to maintain the peace necessary for English trade.72 While the initial circumstances satisfied everyone’s interests, the consequences turned out to be entirely different. If Barton himself and his queen clearly disliked a war against the Porte, the Báthorys, supported by Rudolph II and the Holy See, had a completely different view on the matter (as well as the Cantacuzenoi, albeit for different reasons). Aron and Mihai chose the war option on their advice, with their help, and, at least for a while, under their authority. Zsigmond’s abdication (1598) changed the situation completely and pushed Mihai toward a new kind of politics in respect to Transylvania, Moldavia, and the empire, one that eventually allowed him to take Zsigmond’s place as the overlord of the three principalities.73 From

68 Ibid.; Ştefan Andreescu, “Michele il Bravo e l’idea di riedificazione dell’Impero bizantino: le testimonianze degli ambasciatori veneziani a Costantinopoli,” in Dall’Adriatico al Mar Nero: Veneziani e Romeni, tracciati di storia comune, ed. Grigore Arbore Popescu (Rome, 2003), 51–66. 69 Iulian Ştefănescu, “Epopeia lui Mihai Viteazul în lumea greco-rusească în secolii XVII şi XVIII” [Mihai the Brave’s epopée within the Greco–Russian world, seventeenth- eighteenth centuries], in idem, Opere istorice (Bucharest, 1942), 79–109; Alexandru Iordan, “Mihai Viteazul în folklorul balcanic” [Mihai the Brave in the Balkan folklore], Revista Istorică Română 6 (1936): 361–381; Alfred Vincent, “From Life to Legend: The Chronicles of Stavrinos and Palamidis on Michael the Brave,” Thesaurismata 25 (1995): 165–238. 70 Nicolae Iorga, “Les grandes familles byzantines et l’idée byzantine en Roumanie,” Bulletin de la Section historique de l’Académie roumaine 18, no. 1 (1931): 14–15. 71 See Levente Nagy, “Dalla Dacia alla ‘Monarchia Hungarica’: Il progetto di Luigi Ferdinando Marsili riguardante la descrizione dell’Europa Sud-Orientale,” in La storia di un ri-conoscimento: I rapporti tra l’Europa Centro-Orientale e la Penisola italiana dal Rina- scimento all’Età dei Lumi, ed. Cristian Luca and Gianluca Masi (Brăila and Udine, 2012), 353–393, here 366–369. 72 Báthory’s good relations with England and Barton facilitated these political alliances; see Podea, “A Contribution”; Demény and Cernovodeanu, Relaţiile, 27–31, and 67–74. 73 The Alba Iulia treaty (1595) confirmed Zsigmond Báthory’s position as Mihai’s and Aron’s overlord and set the grounds of their common military action against the Porte. On this occasion, Andronicus was one of the leaders of the Wallachian delegation. Soon 234 radu g. păun this point onward, of the three networks that supported him in 1593, only the “Greek” one remained on his side until the very end.74 In most of the cases presented above, revolts appear to be practically intrinsic to the Porte’s relationship with the tributary states. When the arena of power moved to Constantinople, pretenders to the Wallachian and Moldavian thrones had to follow the general logic of the Ottoman system, whose main features were, in my view, the rotation principle and the rivalry between factions and networks of power and influence. As time went by and competition for power in Constantinople became harsher, the Ottoman custom to “rotate” princes and consequently dismiss them without any serious reason, confiscate their wealth, put them to death or send them to exile, generated a great deal of insecurity and general unpre- dictability, as princes never knew for how long they were to rule and what their destiny was going to be. In other words, the more our protagonists became involved in the Ottoman system, the more pressure the system put on them. Logically enough, this pressure (both economic and political) increased dramatically in times of war. In the tributaries’ case war was almost completely unpredictable, for they found themselves in a state of war each time their overlord decided so, and this because they had no control over the decision to declare war and to conclude peace. In such a context, the “loyalty option” often revealed its limits as a political solution,75 while “overmuch loyalty,” that is, anticipating Ottoman demands or even offering more than the overlord asked for, proved to be highly contradictory in the medium and long run, for such a practice gradually increased the pressure on the vassal principalities and consequently made the Porte’s requests even harder to fulfill.76 Similarly, the “voice option” meant inter- nalizing and following the current Ottoman political practice of spending considerable resources to build up and maintain strong influence

after the revolts broke out, Zsigmond replaced Aron with Ştefan Răzvan. In 1598, Zsig- mond abdicated, and Mihai used the new circumstances to denounce the Alba Iulia commitments and take possession of Transylvania (1599). In the next year, he also occupied Moldavia; see Petre P. Panaitescu, Mihai Viteazul [Mihai the Brave] (Bucharest, 1936); Alexander Randa, Pro Respublica Christiana: Die Walachei im ‘Langen’ Türkenkrieg der katholischen Universalmächte (1593–1606) (Munich, 1964). 74 Andronicus was executed by Mihai’s enemies in October 1601: Ciorănescu, Docu- mente privitoare la istoria românilor, 204 (Spanish report). 75 I follow Albert O. Hirschman’s paradigm here; Exit, Voice and Loyalty: Responses to Decline in Firms, Organizations and States (Cambridge, MA, 1970). 76 I refer to the tribute and other financial obligations. Even so, it occurred quite rarely that “overly loyal” vassal princes were successful in holding their throne for a long time. networks of influence and the military revolts 235

networks in Constantinople. This did not mean the complete avoidance of risks, since quite often these networks chose to follow their own policy, which sometimes meant turning against their formal patrons, as the Cantacuzenoi’s case indicates.77 All these tensions and conflicts generated by the dynamics of the Ottoman system played an important role in the vassals’ decision to choose the “exit” option, especially when favorable circumstances were provided by Christian coalitions directed against the Porte. Seen from this angle, revolt might be considered a means of coun- terattack, but one must also keep in mind the supplementary pressure put on the vassal princes by the divergent interests of the competing powers in the area, such as Poland, the Habsburgs or Russia. Caught between these and their “rightful” overlord, Wallachian and Moldavian princes had to make a delicate choice: if in a context of war they chose “loyalty” and the Christians had won, they would have placed themselves on the side of evil by fighting their coreligionists; if the Ottomans won and they had chosen “exit,” they would have lost their position anyway.78

The Spiritual Atmosphere

Despite all these quite dramatic experiences, belief in the imminent fall of the Ottoman power constantly fed a crusade-like idea in both the Eastern and Western parts of Europe. Since Jerusalem turned out to be too far off an ideal for the “new crusaders,” it was the Second Rome which gradually took on Jerusalem’s features and stood for centuries as an object (and a fountain at the same time) of a mass of “political prophecies.” Their impact on the immediate political and military actions should not be neglected, since people living in the “City of the Cities”—whether they were Christians or Muslims—always kept in mind that their capital’s destiny was fatally

77 I refer here especially to the decisions of Yani and Andronicus to abandon their current protégés and orient themselves toward the rival branch of the Wallachian dynasty, from which they chose Mihai. 78 This dilemma was shared by the Greek-Levantine notables (the archontes) in the sultan’s service; see Păun, “Pouvoirs, Offices.” The answer they chose during the seventeenth century illustrates a different attitude toward the matter, as they proved to be more interested in the preservation of the Ottoman system, which consequently meant the preservation of their own position therein. The anti-Ottoman attitude is to be found mainly among Church leaders, as proven by the Greek-Russian “crusade” negotiations in the 1650s; see below. 236 radu g. păun related to the world’s destiny itself.79 At this point, Christian eschatological hopes/fears and those of the Ottomans corresponded to one another and grew together, being constantly stimulated by real events. It seems difficult to believe that the revolting Moldavian and Wallachian princes would have lived in complete isolation from this general state of mind. While in Constantinople, they could learn about the Muslim ideas concerning the new millennium (which began in 1591–1592), and about the “eschatological panic” (Alphonse Dupront) that came along with it.80 The Venetian bailo Matteo Zane describes this fear, which gained great strength “because of a prognostic the Turks have,” namely that some thousand years after Muhammad’s birth great troubles would occur in the empire; and since the moment was now expected to come, the incidents in Hungary against the Porte lent further authority to the prognostic.81 Zane’s testimony dates from 1594, the very moment when Mihai and Aron decided to revolt against the sultan. Six years later, Girolamo Cappello notes that everyone regarded Sultan Mehmed III (1595–1603) with disdain because of, among other issues, some “inauspicious prophecies” that predicted the fall of the empire under his reign. The logic of these prophecies was simple: since both the first and the last emperor of Byzantium had the same name (Constantine) and they were both the sons of empresses called Helene, and since the conqueror of Constantinople was named Mehmed, son of Murad, so the Empire was supposed to be lost under a sultan having the same name of Mehmed, son of another Murad—which was indeed the case. As a reasonable diplomat, Cappello did not take seriously all these “superstitious and futile ideas”; yet, he had to acknowledge the impact they had upon the masses, in such a way that the sultan enjoyed not only less esteem, but was also

79 Stéphane Yérasimos, “Apocalypses Constantinopolitaines,” Critiques 543–544 (1992): 610. See also Andreas Külzer, “Konstantinopel in der apokalyptischen Literatur der Byz- antiner,” Jahrbücher der österreichischen Byzantinistik 50 (2000): 51–76; Wolfram Brandes, “Der Fall Konstantinopels als apokalyptisches Ereignis,” in Geschehenes und Geschriebenes: Studien zu Ehren von Günther S. Heinrich und Klaus-Peter Matschke, ed. Sebastian Kolditz and Ralf C. Müller (Leipzig, 2005), 453–469. 80 The idea of some Muslim prophecies announcing the end of the Empire appeared soon after 1453, and quickly became a topos; see Kenneth M. Setton, Western Hostility to Islam and Prophecies of Turkish Doom (Philadelphia, 1992), 16ff. Very useful data on this topic is provided by Kaya Şahin, “Constantinople and the End Time: The Ottoman Con- quest as a Portent of the Last Hour,” Journal of Early Modern History 14 (2010): 317–354. 81 “Il (. . .) timore prende assai di fomento per un pronostico che hanno essi turchi, che intorno alli mille anni dal nascimento di Maometto nasceran disturbi in quello stato; ed essendo al presente circa questi anni, li accidenti chi occorrono in Ungheria a danno de’Turchi accrescano al pronostico autorità,” Alberi III, 442–443. networks of influence and the military revolts 237 less obeyed by his subjects, a fact that provoked a great confusion in the Ottoman government.82 The ideas expressed by Venetian diplomats already had a long history and were destined for a brilliant career. Marginalia and interpolations on the Pseudo-Sphrantzès Chronikon Majus predicted the fall of the Ottoman Empire for 1655 by extrapolating the logic of the so called horoscope of the Arab nation (attributed to Stephanos of Alexandria) to the Ottomans; another text of the same kind assigned to Emperor Leo the Wise was used as a supplementary proof in this respect.83 The same year, Paisius Lig- arides, Mihnea’s intimate counselor, wrote about the Muslim prophecies concerning the fall of the Ottoman Empire he had found in the chronicles and the history books of the Ottomans themselves. Ligarides even established the precise moment this event should occur: 1678.84 Some years before the fatal deadline was reached, the Croatian scholar Juraj Križanić

82 “L’hanno fatto cadere in mal concetto di tutti, che con infausto pronostico dicono che nel suo tempo sia per mancar l’Imperio, confrontando insieme la corrispondenza dei due Costantini imperatori, figliuoli di due Elene, con i due Maometti, figliuoli d’Amurat il primo, ch’acquistò Costantinopoli, e questo che dicono dover esser l’ultimo possessore, di maniera che passando questi superstiziosi e vani concetti nella plebe, viene il re poco stimato e manco obedito, donde nasce poi la confusione, in che al presente si trova quel governo,” Maria Pia Pedani, Relazioni di ambasciatori veneti al Senato, vol. 14, Relazioni inedite, Costantinopoli (1508–1789) (Padova, 1996), 398; see also 422, and 443. 83 Marie-Hélène Congourdeau, “Byzance et la fin du monde: Courants de pensée apocalyptiques sous les Paléologues,” in Les traditions apocalyptiques au tournant de la chute de Constantinople, ed. Stéphane Yerasimos and Bernard Lellouch (Paris, 1999), 66–67ff. See also C.J.G. Turner, “An Oracular Interpretation Attributed to Gennadius Scholarios,” Ελληνικά 21 (1968): 40–47. On the pseudo-Sphrantzès’ work, see Vasile Grecu, “Georgios Sphrantzes: Leben und Werk—Makarios Melissenos und sein Werk—Die Ausgaben,” Byzantinoslavica 26 (1965): 62–73; Ricardo Maisano, “Il manoscritto Napoletano II E.25 e la storia della tradizione dello pseudo-Sfranze,” Ιταλοελληνικά: Rivista di cultura greco-moderna 2 (1989): 126–129. On the life and activity of Makarios, his brother and the son of this one, who were themselves involved in anti-Ottoman projects and revolts, see Ioannes K. Hassi- otis, Μακάριος, Θεόδωρος και Νικηφόρος οι Μελισσηνοί (Μελισσουργοί) (16ος–17ος αι.) [Makarios, Theodoros and Nikephoros Melissenoi (Melissourgoi), sixteenth-seventeenth c.] (Thessalo- niki, 1966). For similar prophetic ideas about the imminent fall of the Ottoman Empire in 1622, see Zdenko Zlatar, The Slavic Epic: Gundulić’s Osman (New York, 1995). 84 Astérios Argyriou, Les exégèses grecques de l’Apocalypse à l’époque turque (1453–1821) (Thessaloniki, 1982), 112. On Paisius’ Chresmologion, see Tudor Teoteoi, “L’Europe confes- sionnelle dans l’Oracle inédit de Païsios Ligaridis,” Nouvelles études d’histoire (Bucharest) 10 (2000): 91–96; idem, “La tradition byzantine de l’Oracle inédit de Païsios Ligaridis,” Revue des Etudes Sud-Est Européennes 39, nos. 1–4 (2001): 19–26; Ovidiu-Victor Olar, “Profeţie şi istorie: Note asupra câtorva manuscrise călătoare prin Ţările Române (Matei al Mirelor şi Paisie Ligaridi)” [Prophecy and history: On some traveling manuscripts through the Roma- nian lands, Mathaios of Myra and Paisius Ligarides], in Manuscrise bizantine în colecţii bucureştene / Byzantine Manuscripts in Bucharest Collections, ed. Ileana Stănculescu (Bucha- rest, 2009), 35–46; Iamilia Ramazanova, “Bukharestskii spisok Khrismologiona Paisiia Liga rida: Paleograficheskoe i kodikologicheskoie issledovanie” [The Bucharest manuscript of 238 radu g. păun met in Constantinople the great dragoman of the Porte, Panagiotis Nikou- sios, who showed him some “Arabian books” announcing the end of the Ottoman Empire under Mehmed IV (1648–1687).85 Westerners also shared these ideas,86 for they intensively circulated all over Europe thanks especially to Bartholomew Georgewitz influential work Prognoma sive praesagium Mehemetanorum (1545),87 and the Byzantine eschatological and apocalyptic writings, which were copied, translated into various European languages, and reinterpreted according to contemporary events.88 It was not by accident that the interest in this

Paisius Ligarides’ Chresmologion: paleographical and codicological approach], Vestnik Rossiiskii gosudarstvennyi gumanitarnyi universitet 7, no. 50 (2010): 178–192. 85 Argyriou, Exegèses, 111–112; Ivan Golub, “Contribution à l’histoire des relations de Križanić avec ses contemporains (1651–1658),” in Juraj Križanić (1618–1683) Russophile and Ecumenic Visionary, ed. Thomas Eekman and Ante Kadić (The Hague and Paris, 1976), 128. Nikousios owned at least one manuscript anthology of Byzantine and post-Byzantine prophecies that he gave as a present to the Wallachian chamberlain Constantin Can- tacuzino; see Nicolae Iorga, “Manuscripte din biblioteci străine relative la istoria româ- nilor: Întâiul memoriu” [Manuscripts from foreign libraries concerning Romanian history: First account], Analele Academiei Române: Memoriile Secţiunii Istorice, series 2, 20 (1898): 239–240. It is not impossible that precisely this manuscript inspired Ligarides’ work, for he spent some time at the Cantacuzinos’ court in Wallachia. New data about the eschatological ideas of Nikousios are provided by Georgios Koutzakiotis, Αναμένοντας το τέλος του κόσμου τον 17ο αιώνα: ο Εβραίος Μεσσίας και ο Μέγας Διερμηνέας [Awaiting the end of the world: The Hebraic messiah and the great dragoman] (Athens, 2011). Križanić himself is the author of the Tolkovanie istoricheskikh prorochestv [Explanation of historical prophecies] (1674); see Sobranie sochinenii Iuriia Krizhanicha [Complete works of Juraj Križanić], vol. 2, ed. M. Sokolov (Moscow, 1892). 86 In 1620, Père Thomas de Navarre informed Père Joseph of some “Arabian wise men” who predicted the fall of the Ottoman state precisely for that moment. Lazaro Soranzo also refers to the legend around the names: since Constantinople was founded by one Con- stantine and lost by another one, and the Roman Empire began with one Augustus and ended with another one, Ottoman Constantinople will also be lost by one Mehmed, since it was conquered by one; see Apostolos E. Vakalopoulos, Πρώιμοι απελευθεροτικοί αγῶνες των Ελλήνων κατὰ την Οθωμανοκρατία (1453–1669) [Early liberation struggles of the Greeks during the Ottoman rule, 1453–1669] (Athens, 2000), 372; Ottomano (Ferara and Milan, 1599); Setton, Western Hostility, 38. The same idea is expressed by the Sieur du Vignau, former secretary of the French ambassador in Constantinople in his Etat présent de la puissance ottomane, avec les causes de son accroissement et celles de sa décadence (Paris, 1687; English edition in 1688). 87 Georgewitz’s work was published almost eighty times during the sixteenth century and translated into seven languages; see Setton, Western Hostility, 30–31ff. The print editions are listed in Göllner, Turcica. 88 I refer here to “revelations” and “oracles” ascribed to Prophet Daniel, Methodius of Patara, the Sybil, Emperor Leo the Wise, and Saint Andrew Salos, and the various interpretations of the Revelation of John. From the mass of works devoted to this topic, I should mention Paul J. Alexander, The Byzantine Apocalyptic Tradition (Berkeley and Los Angeles, 1985); Agostino Pertusi, Fine di Bisanzio e fine del mondo: significato e ruolo storico delle profezie sulla caduta di Costantinopoli in Oriente e in Occidente, ed. Enrico Morini (Rome, networks of influence and the military revolts 239 literature increased precisely during the second half of the sixteenth century, when Mediterranean and Balkan areas experienced a long series of real and imagined wars against the Turks. Lepanto greatly stimulated this interest and left an enduring memory in both Italian and Greek cultures, a memory reactivated during each new anti-Ottoman war.89 The chronology of the manuscripts and print editions of Oracula Leonis corresponds to the moments of high tension in the relations between “Christian Europe” and the Ottomans.90 Zacharias Skordylis produced two of the so-called “Barozzi manuscripts” of the Oracula between 1572 and 1573, that is, just after the battle of Lepanto. In Skordylis’ view, the fall of the Ottoman Empire was imminent and was supposed to occur during the reign of Selim II (1566–1574).91 In 1596, in the very middle of the

1988); Katerina Kyriakou, Οἱ Ἱστορημένοι χρησμοί τοῦ Λεοντοῦ ς᾽ τοῦ Σοφοῦ: χειρόγραφη παράδοση καὶ ἐκδόσεις κατὰ τοὺς ΙΕ´-ΙΘ´ αἰῶνες [The historical oracles of Leo VI the wise: Manuscript and print tradition, fifteenth–nineteenth centuries] (Athens, 1995); Vasilka Tǎpkova- Zaimova and Anisava Miltenova, Historical and Apocalyptical Literature in Byzantium and Medieval Bulgaria (Sofia, 2011). The Romanian versions are listed by Andrei Timotin, “La littérature eschatologique byzantine et post-byzantine dans les manuscrits roumains,” Revue des Etudes Sud-Est Européennes 40, nos. 1–4 (2002): 151–167. For the role of these texts in the Western world, see Jean Dény, “Les pseudo-prophéties concernant les Turcs au XVIe siècle,” Revue des Etudes Islamiques 10, no. 2 (1936): 201–220; and, more recently, Yoko Miyamoto, “The Influence of Medieval Prophecies on Views of the Turks: Islam and Apocalypticism in the Sixteenth Century,” Journal of Turkish Studies 17 (1993): 125–145. 89 Achille Olivieri, “Il significato escatologico di Lepanto nella storia religiosa del Mediterraneo del Cinquecento,” in Il Mediterraneo nella seconda metà del ‘500 alla luce di Lepanto, ed. Gino Benzoni (Florence, 1974), 257–276; Letizia Pierozzi, “La vittoria di Lepanto nell’ escatologia e nella profezia,” Rinascimento 34 (1994): 317–363. In his Chro- nographia, Georgios Klontzas (1540–1608) brings together a mass of oracular texts in a way that frames Lepanto as the culminating event of universal history, an event which had been announced by the prophecies he mobilizes therein, and which announced by itself, according to the same prophecies, the ultimate victory over the “infidel.” Klontzas situates the fulfillment of Leo’s pronouncement in the years 1590–1592, a moment which, accidentally or not, corresponded to the Islamic millennium and to the beginning of a new Holy League. See Olga Gratziou, “History and Actuality in the Oracular Manuscripts Illuminated by Georgios Klontzas,” Proceedings of the 8th International Cretological Confer- ence (Heraklion, 2000), vol. B1, 151–162; Mary Williams, “After Lepanto: Visualizing Time, History, and Prophecy in the ‘Chronographia’ of Georgios Klontzas” (MA thesis, University of Notre Dame, 2007). 90 All twelve illuminated manuscripts of Oracula Leonis were made in the sixteenth century, and they were applied continually to the successive Ottoman sultans; see Anto- nio Rigo, “L’opera ‘profetica’ di Francesco Barozzi tra Creta e Venezia,” in Storia e figure del’Apocalisse fra ‘500 e ‘600, ed. Roberto Rusconi (Rome, 1996), 79. The majority were produced around 1590–1600, namely during the Thirteen Years War, see Cyril Mango, “The Legend of Leo the Wise,” Zbornik Radova Vizantološkog Instituta 6 (1960): 79–80. 91 The Venetian bailo Stefano Tiepolo narrated a discussion between Süleyman the Magnificent and an Arabian (or Armenian) man, who supposedly told the sultan that the Ottoman dynasty will end with him; Süleyman replied: “Not with me but with my 240 radu g. păun

Holy League’s war against the Porte, the Oracula were printed in Brescia along with Georgewitz’s writing, which foretold the end of the Ottoman power under Mehmed III (1595–1603). The first Greek edition of the text appeared in 1618, at the very moment when the Militia Christiana was created, while the Latin-Greek edition of Petrus Lambeccius was published in Paris in 1655, during the war of Candia and four years before Mihnea’s revolt.92 It is of course absolutely legitimate to consider all of the above as coin- cidences, since many other print editions of this kind were published, and their publishing dates do not necessarily coincide with any significant revolts against the Porte. Proof exists, however, that at least some of the social actors we are dealing with were familiar with these ideas, as they possessed, and presumably used to read, works of this kind.93 The private library of Antonius Cantacuzenus, for instance, contained (besides the famous Apology of the emperor John Kantakuzenos against the Muslims) the prophecies of Pseudo-Methodius and Emperor Leo, as well as some other similar writings.94 Antonius’ son Georgios (Shaytanoglu’s son-in-law

successor,” see Antonio Rigo, Oracula Leonis: Tre manoscritti Greco-veneziani degli ora- coli attribuiti all’imperatore bizantino Leone il Saggio (Bodl. Baroc. Marc. gr. VII. 22, Marc. gr. VII.2) (Venice, 1988), 13. 92 Rigo, “L’opera ‘profetica,’ ” 86. 93 Rigo doubted the impact of these texts in the Greek world under Ottoman rule: “L’opera ‘profetica,’ ” 82–83. However, Gerlach testified some belief of this kind, and he was very well informed, since he moved in the Cantacuzenoi’s circle in Constantinople, Patri- arch Jeremias included; see Vakalopoulos, Πρώιμοι, 371 (1576). On the private libraries in Constantinople, see Georges K. Papazoglou, Βιβλιοθῆκες στήν Κωνσταντινούπολη τοῦ ις΄ αἰώνα (Κωδ. Vind. Hist. Gr. 98) [Libraries in Constantinople during the sixteenth century. Codex Vind. Hist. Gr. 98] (Thessaloniki, 1983). The great number of manuscript copies is also telling. According to Francesco Barozzi, “si ritrovano molti essemplari in quest’isola (Crete) et in altri lochi della Grecia,” Rigo, Oracula Leonis, 17. Indeed, specialists count around fifty sixteenth-century manuscripts only of Leo’s Oracula: Lydie Hadermann-Misguich, “Les livres d’oracles illustrés par Georges Klontzas: Leur chronologie,” Zograf 26 (1997): 133–138. Ironically, when deploring the Romaioi’s fate and their interest in various “lying prophecies,” Matthew of Myra provides indirect evidence that people really believed in them: Alfred Vincent, “Byzantium Regained? The History, Advice and Lament by Matthew of Myra,” Thesaurismata 28 (1998): 275–347; Olar, “Profeţie şi istorie.” 94 Oraculum Methodii Episcopi Patharensis, praeterea ex visionibus Danielis Prophetae et oraculis Astronomiae, et oraculum Cosdroae regis Persarum ad Ducem Romanum explicatum a Laskare ad haec Leonis Philosophi imperatoris omnia qualia oracula a Magno Constantino usque ad Antichristum et finem mundi; see Johann Hartung, Bibiotheca sive Antiquitates Urbis Constantinopolitanae (Strasbourg: 1578). I used Du Verdier’s edition: Les bibliothéques françoises de la Croix du Maine et de Du Verdier, Sieur de Vauprivas: Nouvelle édition (Paris, 1773), vol. 6, 243ff, here 298. networks of influence and the military revolts 241 and Andronicus’ brother-in-law) also possessed the prophecies of Leo,95 as did Manuel Eugenikos, the owner of some illustrated (“cum figuris”) manuscripts containing the same texts.96 The list does not end here. In Patriarch Jeremias’ collection of manuscripts, Hartung found, besides Cantacuzenus’ Apology, Andrew of Caesarea’s commentary on the Proph- ecy of Daniel (Sanctissimi Metropolitae Caesareae Cappadociae Domini Andreae Explicatio in visiones Danielis Prophetae).97 Relevant data about the interest in and the use of old prophecies in new contexts are provided by an eighteenth-century writing briefly introduced by Alfonso Carile. The first part of the text foretells the restoration of the Byzantine Empire by the Palaiologus’ family (γένος τοῦ Παλαιολογοῦ), which seems to recall the actions of Charles de Nevers and the Militia Christiana. If Carile’s interpretation is correct—and I think it is—we have here one more indication that inasmuch as Ottoman affairs are concerned, prophecy was closely associated with political decisions and actions not only for reasons of propaganda, but also because this was part of the general view of politics.98 The second “noyau historique” Carile talks about referred, in his view, to the battle of Lepanto, but the alliance

95 Ibid., 300–301. Henri Omont, “Martin Crusius, Georges Dousa et Théodose Zygoma- las,” Revue des Etudes Grecques 10, no. 3 (1897): 70. 96 Oracula omnia, cum figuris, putà principium Methodii, Episcopi Patarensis, together with the Visiones Danielis Prophetae, & Cosdroae, Regis Persarum & Theophili, Presbyteri Romani, translata à Georgio Zegabeno praetereaque Leonis Philosophi & Imperatoris; see Hartung, Bibliotheca, 320. The manuscript cum figuris might be one of those that later served as a basis for the first Greek edition of the Oracula Leonis, for it is known that this one was based on a manuscript brought from Constantinople by Georgios Dousa, see Rigo, Oracula Leonis, 99n48. Dousa’s contacts with the Cantacuzenoi were mediated by Barton: Omont, “Martin Crusius,” 69–70. Leo’s Oracula were also included in the famous Chronography of pseudo-Dorotheus (Venice 1631), which circulated throughout the entire Orthodox world, and was translated into Russian by Gabriel of Nazareth and by Arsenius the Greek: Christos P. Laskarides, Η στάση της Ρωσίας στον Πόλεμο της Κρήτης (1645–1669) [Russia’s attitude during the Cretan War, 1645–1669] (Thessaloniki, 2002), 303–324 (Greek text), and 331–392 (Russian text). See also Nadezhda Chesnokova, Khristianskii Vostok i Rossiia: Politicheskoe i kul’turnoe vzaimodeistvie v seredine XVII veka [Christian East and Russia: Political and cultural relations at the middle of the seventeenth century] (Мoscow, 2011), 159–204. 97 Hartung, Bibliotheca, 287–288. 98 Alfonso Carile, “Une prophétie inédite en néo-grec et en vénitien sur la chute de l’Empire ottoman,” Byzantinische Forschungen 17 (1991): 32–33. Again, eastern and western ideas about the immediate fall of Ottoman power responded mutually. Père Joseph’s action to shape a large coalition against the Porte is a case in point. Related to this and to Militia Christiana, one could recall Marco Aurelio Scaglia’s attempts to find a Biblical prophecy to apply to the new order and to the holy mission associated with it: Göllner, “Militia,” 66. In parallel, Artus Tomas d’Embry published Tableaux prophétiques prédisant la ruine de la monarchie turque (Lyon and Paris, 1620) and dedicated it to the Duke of Nevers. 242 radu g. păun between Venice and the “blonde nation” (ξανθός γένος) the text mentions rather seems to point to the Venetian–Russian projected alliance during the war of Crete.99 If so, it is certain that Carile’s text is not unique: several years after having composed the Trionfo del illustrissimo (. . .) signor Alvise Mocenigo (1649), the Greek scholar Gerassimos Vlachos turned toward Moscow and wrote Θρίαμβος κατὰ τὴς τῶν Τουρκῶν βασιλείας (Tri- umph over the Turkish empire, 1656–1657), which he dedicated to Alek- sei Mikhailovich.100 Vlachos’ text echoed Leo’s prophecies and explicitly identified the “blonde nation” with Russians.101 At the same moment, two important events occurred in Moldavia and Wallachia: in 1656, Gheorghe Ştefan (1653–1658) concluded a treaty with the tsar in which Greek clergy- men played a crucial role,102 and Ligarides completed his Chresmologion,

99 In 1657, Venice incited the Greeks to call on the tsar to protect Orthodox Christians: Philip Longworth, “Russian–Venetian Relations in the Reign of Tsar Aleksey Mikhailovich,” Slavonic and East European Review 64, no. 3 (1986): 383ff, 398. In 1660, the ex-princes of Wallachia and Moldavia, Gheorghe Ştefan and Constantin Şerban, addressed the same call to the Russian Court: Ion Nistor, “Contribuţii la relaţiunile dintre Moldova şi Ucraina în veacul al XVII-lea” [Contributions to seventeenth-century Moldo–Ukrainian relations], Analele Academiei Române: Memoriile Secţiei Istorice, series 3, 13 (1933): 214 (no. L). 100 Argyriou, Exegèses, 93. The work was soon after translated into Russian: Ekkehard Kraft, Moskaus griechisches Jahrhundert: Russisch–griechische Beziehungen und metabyzan- tinischer Einfluss, 1619–1694 (Stuttgart, 1995), 80; Laskarides, Η στάση, 252–253ff. On him, see Vassilis Tatakēs, Γεράσιμος Βλάχος ὁ Κρὴς (1605/7–1685): Φιλόσοφος, θεολόγος, φιλόλογος [Ger- assimos Vlachos from Crete (1605/7–1685): Philosopher, theologian, philologist] (Venice, 1973). 101 In 1653, the ex-ecumenical patriarch Athanasios Patellaros alluded to exhorting the tsar to deliver Constantinople and “accomplish the old prophecies,” Boris L. Fonkić, Greaco–Russian Contacts from the Middle of the XVIth Century up to the Beginning of the XVIIIth Century: The Greek Documents in Moscow Archives. Catalogue of Exhibition (Mos- cow, 1991), 49–50 (22 March 1654), and 34–35 (21 October 1646); Olga Alexandropoulou, “Το ροσσικό ταξίδι του Αθανάσιου Πατελλάρου και ο Προτρεπτικός Λόγος του προς του τσάρο Αλέξιο (1653)” [The Russian travel of Athanasios Patellaros and his Exhortation addressed to Tsar Alexis, 1653], Μνημων 21 (1999): 9–35; Laskarides, Η στάση, 271. See also Vera G. Tchentzova, “Les documents grecs du XVIIe siècle: pièces authentiques et pièces fausses. 1. L’ex-patriarche œcuménique Athanase Patélare à Moscou en 1653,” Orientalia Christiana Periodica 68, no. 2 (2002): 335–374. 102 Vera G. Tchentzova, “Dionysos Iviritis et les pourparlers entre la Moldavie et la Russie en 1656,” in Închinare lui Petre Ş. Năsturel la 80 de ani, ed. Ionel Cândea, Paul Cer- novodeanu, and Gheorghe Lazăr (Brăila, 2003), 581–605. They acted as mediators between the Wallachian and Moldavian princes and the tsar, and between the tsar and the Cos- sacks. Three Orthodox patriarchs were involved in these negotiations: Parthenius of Con- stantinople (executed by the Ottomans in March 1657), Païsios of Jerusalem, and Macarius Za’im of Antioch. See also Andreescu, Restitutio, vol. 2, 241–242ff. Some leading figures of the Serbian Church participated as well. Patriarch Gavrilo Rajić of Peć was executed by the Ottomans for this reason, precisely in 1659: Istorija Srpskog naroda [The history of the Serbian people], vol. 3/2, Srbi pod tuđinskom vlašću 1536–1669. [The Serbs under Turkish rule, 1536–1699], ed. Radovan Samardzić (Belgrade, 1993), 86–93. networks of influence and the military revolts 243 a large collection of prophecies and oracles foretelling the ruin of the Ottoman Empire by the “blonde nation from the North.”103 In parallel, in Constantinople, Mihnea tried to survey Venice’s availability to fight the Porte and shared, as Claes Rålamb’s testimony clearly indicates, the prophetic ideas about the imminent fall of the Ottoman power.104 A common state of mind seems to have been shared by certain Greek, South-Slavic, Wallachian, and Moldavian circles, whose main features appear to be a strong belief in the imminent fall of the Ottoman power on one hand, and the restoration of the “Christian (i.e., Orthodox) Empire” by the Russians on the other. Put in this context, Mihnea’s highly ritualized crusade-like behavior bears a clear eschatological meaning, as he played not only as a kind of basileus tōn Romaiōn redivivus but also as an alle- gory of the Last Emperor to come.105 The popularity of these ideas in the Balkan world grew constantly, as they were stimulated by Russian military successes and by the conviction that the Ottoman state had entered a period of rapid and fatal decay.106 Peter the Great’s successful politics

103 Argyriou, Exegèses, 112. See also Teoteoi, “La tradition byzantine”; Olar, “Profeţie şi istorie.” The work was dedicated to the tsar and proved to be quite popular in Russia: Ernst Christoph Suttner, “Panteleimon (Paisios) Ligarides und Nicolae Milescu: Ein Beitrag zur Frage nach der Offenheit des wallachischen Fürstentums für das Bildungsgut der Zeit im zweiten Drittel des 17. Jahrhunderts,” Kirche im Osten 26 (1983): 73–94; Chesnokova, Hristianskij Vostok i Rossija, 173–176. On Paisius’ career in Russia, see Charalambos K. Papastathis, “Païsios Ligaridis et la formation des relations entre l’Église et l’État en Russie au XVIIe siècle,” Cyrillomethodianum 2 (1972–1973): 77–85; Ihor Ševčenko, “A New Greek Source for the Nikon Affair: Sixty-One Answers Given by Paisios Ligarides to Tsar Aleksej Mixailovič,” Palaeoslavica 7 (1999): 65–83. 104 See Kármán, “The Networks of a Wallachian Pretender.” 105 I argued for this in my “Pouvoir, Croisade et Jugement Dernier au XVIIe siècle: Le vécu et l’invisible,” in Ius et Ritus: Rechtshistorische Abhandlungen über Ritus, Macht und Recht, ed. Ivan Biliarsky (Sofia, 2006), 213–283. Ideas of the kind were also spread in Russia of the 1650s and 1660s, see Daniel C. Waugh, “Seventeenth-Century Muscovite Pamphlets with Turkish Themes: Toward a Study of Muscovite Literary Culture in its European Set- ting” (PhD diss., Harvard University, 1972), 42. 106 Western travelers continually testify to the popularity this idea enjoyed among Balkan people; see Athanasios E. Karathanasis, “Pierre le Grand et l’intelligentia grecque (1685–1740),” in Les relations gréco-russes pendant la domination turque et la guerre d’indépendance grecque: Premier colloque (Thessaloniki, 1983), 45 (Spon and Wheeler, Epirus, 1675–1676); Apostolos E. Vakalopoulos, Ἱστορία τοῦ νέου Ἑλληνισμοῦ [The history of Neohellenism], vol. 4, Τουρκοκρατία, 1669–1812 [The Turkocracy, 1669–1812] (Thessaloniki, 1973), 69 (La Guilletière, Lacedemony, 1676); Pitton de Tournefort, Relation d’un voyage du Levant fait par ordre du Roy (Paris, 1717), vol. 1, 71 (Crete, 1700). In 1688, the Wallachian Prince Şerban Cantacuzino (Cantemir’s father-in-law) called tsars Ivan and Peter “New Messiah,” whose mission was to destroy the tyranny of the “New Pharaoh” (the sultan): Lidia E. Semionova, “Stabilirea legăturilor diplomatice permanente între Ţara Românească şi Rusia la sfârşitul secolului al XVII-lea şi începutul secolului al XVIII-lea” [The establishment of permanent Wallachian-Russian relations, end of the seventeenth—beginning 244 radu g. păun increased the pro-Russian feelings of even more Greeks. The capture of Azov in 1696, for instance, inspired an anonymous author to dedicate to the tsar a Θρίαμβος (in the same vein as that mentioned above, by Vlachos) exhorting him and his “dignitaries and generals” to go to war against the Ottomans. Two years later, the German scholar Stanislaw Reinhard Axtelmeyer published his Prognostikon (Augsburg, 1698), which was quickly translated into Greek by Yoannis Komninos.107 In 1709, Anastasios Mihail of Naousa (Macedonia), who later became Cantemir’s secretary in Russia, glorified Peter’s victory at Poltava in his Βασιλικὸν Θέατρον (Impe- rial theater).108 No direct evidence is known about the influence these ideas exerted on Cantemir’s actions, but it seems likely that the Moldavian prince, whose relations with and knowledge of Greek culture were excellent, knew of their popularity. Yet his own leading idea of the “growth and decay of the Ottoman Empire,” although inspired by the Renaissance reflection on the historical life of the states (René de Lucinge, Pétis de la Croix, etc.), remains connected to the historical Biblical tradition, whose ultimate fulfillment would have been the Christian triumph over the “infidel.”109

of the eighteenth century], Romanoslavica 5 (1962): 31. A contemporary Venetian report describes the unrest provoked in Istanbul by the finding of a Bulgarian-written inscription “che nell’anno 1691 doverà Constantinopoli cader in mano de’Christiani, e del 1693 serà tenuto un concilio per l’union delle chiese,” which event “havesse aumentato fra Turchi la confusione,” Setton, Western Hostility, 19n20 (24 November 1689). 107 Corneliu Dima-Drăgan and Mihai Caratașu, “Un manuscrit grec inédit du docteur Jean Comnène,” Revue des Etudes Sud-Est Européennes 9, no. 1 (1971): 107–120. 108 Karathanasis, “Pierre,” 48n21, and 49–50; Vakalopoulos, Ἱστορία, vol. 4, 70. The 1711 campaign was accompanied by Russia’s attempts to incite Balkan Orthodox people to rise against the Porte: Apostolos E. Vakalopoulos, “ Ὁ Μέγας Πέτρος καὶ οἱ Ἕλληνες κατὰ τὰ τέλη τοῦ 17. καὶ τὶς ἀρχὲς τοῦ 18. αἰ.” [Peter the Great and the Greeks at the end of the seventeenth and the beginning of the eighteenth century], Επιστημονικὴ Επετηρὶς τῆς Φιλοσοφικῆς Σχολῆς τοῦ Αριστοτελείου Πανεπιστημίου Θεσσαλονίκης 11 (1969): 245–259. See also Nikolas Pissis, “Τροπές της ‘ρωσικής προσδοκίας’ στα χρόνια του Μεγάλου Πέτρου” [Turns of the “Russian expectation” in the age of Peter the Great], Μνημων 30 (2009): 37–60; idem, “Das verän- derte Russland und das griechische Gelehrtentum nach 1700,” in Griechische Dimension südosteuropäischer Kultur seit dem 18. Jahrhundert: Verortung, Bewegung, Grenzüberschrei- tung, ed. Maria Oikonomou, Maria A. Stassinopoulou, and Ioannis Zelepos (Frankfurt am Main, 2011), 155–168. On the role of prophecies in modern Balkan cultures, see John Nicolo- poulos, “From Agathangelos to the Megale Idea: Russia and the Emergence of Modern Greek Nationalism,” Balkan Studies 26, no. 1 (1985): 41–56; Marios Hatzopoulos, “Oracular Prophecy and the Politics of Toppling Ottoman Rule,” Historical Review/La Revue Histo- rique 8 (2011): 95–116; Traian Stoianovich, “Les structures millénaristes sud-slaves aux XVIIe et XVIIIe siècles,” in Actes du 1er Congrès international des études balkaniques et sud-est européennes, vol. 3, Histoire (Sofia, 1969), 809–819. 109 Pippidi, Tradiţia, 88; idem, “La décadence de l’Empire ottoman comme concept historique de la Renaissance aux Lumières,” Revue des Etudes Sud-Est Européennes 35, nos. 1–2 networks of influence and the military revolts 245

The Revolts of the Tributaries and the Ottoman System

Despite the lacunary and apparently incoherent character of the data presented so far, I think there is enough evidence to assert that hope in the future reconquest of the “City” remained strong among the Christian subjects of the Porte. This represents, I think, the biggest difference between the early modern view about Ottoman power and our present perspective: while we are used to looking at Ottoman rule as a reality whose beginning and end are clearly established, for sixteenth- and seventeenth-century actors this was but a temporary experience, the end of which could occur at any moment. This attitude appears to be more visible in the Balkan regions than in the tributary principalities of Moldavia and Wallachia, where political elites were concerned to preserve their position. This explains why they did not generally follow along with revolts, and even dissuaded and/ or blocked this kind of enterprise. Indeed, anti-Ottoman projects were numerous, but few of them became realities, as local princes were aware that they could not count on their nobles’ support in case of need, and they also feared betrayal.110

(1997): 5–21; Lemny, Cantemir, 140. This role was assigned by him to Peter the Great’s Russia: Eugen Lozovan, “D. Cantemir: Monarchiarum physica examinatio,” Romansk Insti- tut Københavns Universitet, 113 (1983): 1–21. In his work The System or the Structure of the Muslim Religion, written in Latin before 1719 and published in Russian in 1722, a large part (Book 3) was devoted to the Muslim prophecies on the conquest of Constantinople by the Russians; see Sistemul sau întocmirea religiei mahomedane, ed. Virgil Cândea and Anca Irina Ionescu (Bucharest, 1987); Virgil Cândea, “Présage et eschatologie dans l’œuvre de Démètre Cantemir,” Bulletin de l’Association Internationale d’Etudes du Sud-Est Européen, 19–23, nos. 1–2 (1993): 73–78. On the intellectual history of the idea of the rise and decay of states, see, among others, Peter Burke, “Tradition and Experience: The Idea of Decline from Bruni to Gibbon,” Daedalus, 105, no. 3 (1976): 137–152. 110 This explains the hesitant attitude of the Moldavian princes Gheorghe Ştefan and Ştefan Petriceicu when they had to choose between the Christians and the Porte. Although caught between the two, neither dared to openly declare war against the Ottomans: the first one moved to action when everything was lost anyway, while the second left the Ottoman camp on the very eve of the final confrontation with the Poles; see Iorga, Istoria, vol. 6, 228–233. See also Cronicile slavo-române din sec. XV–XVI publicate de Ioan Bogdan [The fifteenth to sixteenth-century Slavonic–Romanian annals published by Ioan Bogdan], ed. Petre P. Panaitescu (Bucharest, 1959), 150–151 (the hostile attitude of the chronicler Azarius toward Ioan the Terrible’s revolt); Istoria Ţării Româneşti, 1290–1690: Letopiseţul cantacuzinesc [The history of Wallachia, 1290–1690: The Cantacuzenoi’s annals], ed. Con- stantin Grecescu and Dan Simonescu (Bucharest, 1960), 132–133 (the Wallachian magnates’ reaction toward Mihnea’s plans), and Cronica lui Ion Neculce copiată de Ioasaf Luca: Man- uscrisul “Mihail” [The chronicle of Ion Neculce copied by Ioasaf Luca: The manuscript “Mihail”], ed. Zamfira Mihail and Paul Mihail (Bucharest, 1980), 122 (the opinion of the commander-in-chief of Cantemir’s army in 1711). 246 radu g. păun

“Foreign” princes had a completely different outlook on the issue, according to their particular ideas and personal biography, which allowed them to know (or to believe they did) the weaknesses of the Ottoman system and the way to exploit them in an efficient manner. Their “training” in Ottoman power circles and/or their European experience proved to be, however, a double-sided atout and generated some contradictory consequences. On one hand, this experience provided them with arguments and means to grasp power, and some idea about how to use it in accordance with Ottoman and European policies. On the other hand, this alienated them from the political cultures and traditions of their own countries and did not allow them to structure the local networks necessary to rule and gain the trust of the locals, who often regarded them as the Ottomans’ agents and considered their martial intentions as insti- gations inspired by the Turks themselves.111 When engaging in their war against the Porte, the revolting princes were thus forced to count much less on local nobles’ support than on external allies, whose force was supposed to persuade and mobilize the locals as well. This rarely happened, however, and Cantemir’s case is telling in this respect.112 Similarly, this made them suspect in the eyes of Christian powers interested in a war against the Porte.113 A long and often laborious work of persuasion then proved necessary, in order for the revolting princes to be accepted as reliable allies; this generally left sufficient time for the Ottomans to react. This reaction proved to be efficient—so efficient, in fact, that the Porte did not consider it necessary to significantly change the general principles of its policy toward its tributaries. In other words, despite the revolts, the Ottomans felt confident enough in their own power to apply the same “rotation principle” they had always employed and to appoint princes

111 Graziani’s and Mihnea’s cases are telling. On Cantemir, see Cronica lui Ion Neculce, 116; Eremia, “Politica rusofilă,” 61–66. 112 Iorga, “Carol XII”; Andrei Pippidi, “Politique et histoire dans la proclamation de Démétrius Cantemir en 1711,” in idem, Hommes et idées, 187–215; Eugen Lozovan, “Une fail- lite diplomatique: l’alliance de D. Cantemir avec Pierre le Grand (1711),” Buletinul Bibliotecii Române: Freiburg im Bresgau 7 (9) (1979): 1–36; Lemny, Cantemir, 94–100. 113 The Papal envoy Cardinal Comuleo (Komulović) considered it inappropriate to share any anti-Ottoman projects with Mihai, because the latter was much too dependent on the sultan: Ştefan Andreescu, “Mihai Viteazul şi Roma” [Mihai the Brave and Rome], in idem, Restitutio, vol. 3, 228–229. Habsburg diplomats did not consider Graziani’s offer for an anti-Ottoman alliance, as they thought he was a man of the Ottomans and even intended to “make himself a Turk”; see Anton Mesrobeanu, “Nuovi contributi sul vaivoda Gaspare Graziani e la guerra turco-polacca del 1621,” Diplomatarium Italicum 3 (1935): 132. The same reason explains the Venetian and Roman reluctance toward Mihnea’s proposals: Ciorănescu, “Domnia lui Mihnea.” networks of influence and the military revolts 247 recruited from among the same category of the “men of the system” to the Moldavian and Wallachian thrones. We should then revisit Dimitrie Cantemir’s idea that it was because of the revolts that the Porte decided— just after Ştefan Petriceicu’s treason in 1673—to deprive the local people of the right to rule Moldavia and grant it to the Constantinopolitan Greeks.114 As presented so far, at that precise moment revolts were already a century-old cyclical phenomenon that Ottomans had continually faced and solved with short-term measures. The major change occurred two decades later, when, as the same Cantemir suggests—and he is right this time—the Porte decided to appoint as princes individuals with families and houses in Constantino- ple, people who were not familiar with matters of war;115 that is, Greeks from Constantinople, whose resources depended completely on the Otto- man state’s power. In other words, the Ottomans simply restricted the s/election area to the Greek (or Greek-identified) milieu, which had long ago (since the Cantacuzenoi’s action) abandoned the idea of fighting the Ottomans, precisely because they functioned as a secondary power elite within the system to which they owed their social status.116 From 1711 onward, this strategy was applied systematically and without exception, allowing historians to speak about the “Phanariot regime.” Unlike the Wallachian and Moldavian elites and the Constantinopoli- tan Greek patriciate, Balkan Christians had everything to gain and nothing to lose, thus their support was consistently sought by the revolting princes. Practically every Wallachian and Moldavian revolt—except for Ioan’s action—had a Balkan dimension, which was also taken into account by European powers engaged in the war against the Porte. In this respect, military raids to the southern Danube were supposed to intimidate Ottoman forces but also—and more importantly—to instigate local Christians to revolt and feed a state of tension and a heart of struggle in the region.117 In many cases, it was the Church that served as a relay

114 Dimitrie Cantemir, Vita Constantini Cantemyrii, cognomeno Senis, Moldaviae Princi- pis, ed. Dan Sluşanschi, Ilieş Câmpeanu, and Andrei Pippidi (Bucharest, 1996), 86. 115 “Qui habeat domum familiamque suam Constantinopoli, praeterea imbellem et rerum bellicorum ignarum,” ibid. In Cantemir’s view, the first “Greek” prince would have been Demetrius Cantacuzenus (1674–1675; 1684–1685), who came from a Constantinopo- litan family indeed. 116 Păun, “Well-born of the Polis.” 117 Mihai’s exploits were celebrated by Balkan people; see Andrei Pippidi, “Au sujet des peuples de l’Europe du Sud-Est dans la politique internationale à la fin du XVIe et au début du XVIIe siècle,” East European Quarterly 10, no. 1 (1976): 117. On the Balkan connections of Graziani’s revolt, see Mesrobeanu, “Nuovi contributi,” 200–201. Mihnea’s envoys 248 radu g. păun between the Wallachian and Moldavian princes and the Balkan regions, and it is not insignificant that many of the “revolting vassals” were closely connected to ecclesiastic circles and had at least one clergyman as intimate counselor.118 On the regional level, the Balkan policy of the revolting princes was constantly redoubled by systematic attempts to shape a common Moldo- Wallachian anti-Ottoman front and even to engage therein—when circumstances permitted—also the Principality of Transylvania.119 Habs burg, Polish, Cossack and/or Russian alliances were also considered, in order to strengthen and secure this front. It seems fair to assume that the Wallachian and Moldavian revolts were not merely individual actions, but deliberate attempts to coalesce Christians against the “infidel” and/or to respond to European anti-Ottoman projects, from Lepanto to Peter the Great’s “crusade.” In this context, many of the revolting princes understood that Western anti-Ottoman projects were intimately linked to the Holy See’s missionary task in the Orthodox lands and therefore tried to

in Venice and Rome insisted that the prince “s’offerisce di passar il Danubio (. . .) con 30 mille soldati” for going into the Ottoman provinces, “dove lui ha li huomini giurati” and ready to raise arms against the Turks: Ciorănescu, “Documente,” 43. This step appeared to be crucial in Mihnea’s plans to go “alle mura di Constantinopoli”: ibid., 33–35; 95–96. At least three Bulgarian agents of Mihnea are known: the Franciscans Gabriel Thomassy and Grigorij of Kiprovac, and Peter Bakšić: ibid., 21 (Rákóczi’s letter to the Doge, 3 July 1658). On the 1711 episode: Andrei Pippidi, “Ecouri ale evenimentelor din 1711 în corespondenţa diplomatică” [Echoes of the 1711 events in diplomatic correspondence], in Românii în istoria universală, vol. 3/3, Izvoare străine pentru istoria românilor, ed. Ștefan S. Gorovei (Iaşi, 1988), 404–405. 118 Mihai the Brave and Aron cooperated with Dionysus Rhalle, archbishop of Trnovo, and with Nikephorus the Deacon, while Mihnea’s intimate counselor was Paisius Lig- arides, a former Catholic missionary in Constantinople who later became the Orthodox archbishop of Gaza. On him, see Émile Legrand, Bibliographie hellénique, ou description raisonnée des ouvrages publiés par des Grecs au dix-septième siècle, vol. 4 (Paris, 1894), 8–61; Francisc Pall, “Les relations de Vasile Lupu avec l’Orient orthodoxe et particulière- ment avec le Patriarcat de Constantinople, envisagées surtout d’après les lettres de Lig- aridis,” Balcania 8 (1945): 137–139; Harry T. Hionides, Paisius Ligarides (New York, 1972); Philip Longworth, “The Strange Career of Paisios Ligarides,” History Today 45, no. 6 (1995): 39–45. 119 Ioan’s first move was to replace the Wallachian pro-Ottoman Prince Alexandru Mircea with his own protégé Vintilă, while Mihai tried—and temporally succeeded—to take all three principalities under Habsburg suzerainty. For the same reason, in the Mili- tia Christiana’s and Graziani’s plans, Wallachia was committed to Radu Şerban, Mihai’s former captain, and Transylvania to György Hommonai, instead of the Protestant pro- Ottoman Gábor Bethlen. Similarly, Mihnea’s action must be regarded in the broader context of the alliance with György Rákóczi II and Constantin Şerban, while Peter the Great’s anti-Ottoman plans included, besides Cantemir’s Moldavia, the Wallachia of Constantin Brâncoveanu (1688–1714). networks of influence and the military revolts 249 adapt their politics to this reality. Aron was seen by Rome as a virtual Catholic and Cardinal Comuleo even stated that the prince converted to the Roman faith.120 Similarly, Graziani (who was himself a Roman Catho- lic until he had to convert to Orthodoxy as prince of Moldavia) was dedicated to improving the status of the Roman Church in Ottoman Hungary and maintained a correspondence with Pope Paul V and with the Jesuits in Constantinople.121 Mihnea’s movements indicate the same orientation. As recent research shows, he even pleaded to unify the Church, declaring himself ready to embrace the Roman faith.122 We can thus affirm that the idea of a united Christendom, whose first and most important purpose would have been to fight the “infidel,” did not die, despite the religious and political fractures dividing Christian Europe. In the Orthodox world under Ottoman rule, this idea continued to inspire some “Crusades after the Crusades,” to put it in Alphonse Dupront’s words.123 Could we speak then about a kind of misunderstanding between the Western view of a “divided Christendom,” as Dupront stated it, and the Eastern idea of an “united corps” of the Christian Church? This remains, of course, a question for debate.124

120 Călători străini despre Ţările Române [Foreign travelers on the Romanian countries], ed. Maria Holban, Paul Cernovodeanu, et al., vol. 3 (Bucharest, 1971), 376; see also Andreescu, “Mihai Viteazul şi Roma.” 121 Virginia Vasiliu, “Il principato moldavo e la curia papale fra il 1606–1620,” Diploma- tarium Italicum 2 (Rome, 1930): 20–21; Mesrobeanu, “Nuovi contributi,” 136–137, 152, 176 (14 August 1617), 183–184 (January 1618), 199 (12 December 1620); Manea-Grgin, “Neobičan knez,” 71. 122 Pope Alexander VII advised him to make a declaration of faith before the Catholic archbishop of Sofia, Hurmuzaki V/2, 56–57 (15 October 1658), see also 65 (letter of the Orthodox metropolitan of Wallachia to the Pope, 2 October 1659); Ciorănescu, “Docu- mente,” 26 (Venetian report, August 1658); Păun, “Pouvoir, Croisade,” 270–272; Ovidiu- Victor Olar, “Orthodoxie et politique I. Le Synode de Târgovişte (Janvier 1659),” Archævs: Etudes d’Histoire des Religions/Studies in the History of Religions 11–12 (2007–2008): 177–204. 123 Note the affinity between Ioan’s and Mihnea’s ideas of Christendom, as expressed by their common motto “Si Deus nobiscum, quis contra nos?,” Corfus, Documente, 72 (Ioan’s letter to the Polish Senate, fall 1572); Păun, “Pouvoir, Croisade,” 223. 124 Certain historians continue to be too influenced by Alphonse Dupront’s masterful work, Le mythe de Croisade, vols. 1–4 (Paris, 1997). Dupront’s perspective allowed Géraud Poumarède, for instance, to consider sixteenth–seventeenth century anti-Ottoman wars as a kind of “common” (that is, quite trivial) experience, and to seriously doubt the authenticity and the impact of the prophecies concerning the end of the Ottoman power: Pour en finir avec la Croisade: Mythes et réalités de la lutte contre les Turcs aux XVIe et XVIIe siècles (Paris, 2004). As Olivier Chaline rightly argues, it remains to be proven if these conclusions apply also to the East and Central European mind, “Les théâtres de batailles dans le sud-est de l’Europe,” XVIIe siècle 229 (2005): 581n8.

section three military cooperation between the ottoman empire and its tributaries

The Friend of My Friend and the Enemy of My Enemy: Romanian Participation in Ottoman Campaigns

Ovidiu Cristea

Political relations between Wallachia and Moldavia, on the one hand, and the Ottoman Empire, on the other, have long been a favorite subject for modern Romanian historiography. There is, however, an obvious imbal- ance here; although numerous studies highlight the role of these two countries as defenders of Christendom, most historians all but ignore the occasions in which they worked together with the Ottoman Empire.1 Virgil Ciocîltan has devoted one study to Vlad II Dracul’s (1436–1442, 1443–1447) participation in Murad II’s Transylvanian campaign of 1438;2 there have also been some references in general works of historical synthesis,3 and Mihai Maxim published some documents related to the subject, with commentaries.4 Other than these, though, there has never been any systematic attempt to analyze how Wallachian and Moldavian voievods participated in Ottoman campaigns.

1 The theme of Wallachia’s role as “the shield of Christendom” has been the object of numerous studies, especially after 1990 when the topic gained significant currency. See Vlad Georgescu, Istoria ideilor politice româneşti (1369–1878) [A history of Romanian political thought] (Munich, 1987), 77–78; Andrei Pippidi, “La Croisade au bas Danube: les Roumains comme rempart de la chrétienté,” in Histoire des idées politiques de l’Europe centrale, ed. Chantal Delsol and Michel Maslowski (Paris, 1998), 77–89; Liviu Pilat, “Con- ceptul de Poartă a Creştinătăţii în retorica voievozilor Moldovei, 1475–1538” [The concept of the gate of Christendom in the rhetoric of the Moldavian voievods], in Ideologii politice şi reprezentări ale puterii în Europa [Political ideologies and representations of power in Europe], ed. Alexandru-Florin Platon, Bogdan-Petru Maleon, and Liviu Pilat (Iaşi, 2009), 139–174. 2 Virgil Ciocîltan, “Între sultan şi împărat: Vlad Dracul în 1438” [Between sultan and emperor: Vlad Dracul in 1438], Revista de Istorie 29 (1976): 1767–1790. 3 See, for instance, Ştefan Andreescu, Restitutio Daciae: Relaţiile politice dintre Ţara Românească, Moldova şi Transilvania în răstimpul 1526–1593 [Restitutio Daciae: Political relations between Wallachia, Moldavia, and Transylvania in the period 1526–1593] (Bucha- rest, 1980), 53–54, 56, 58, 67, 69, 75, 79, 82. 4 Mihai Maxim, Noi documente turceşti privind Ţările Române şi Înalta Poartă 1526–1602 [New Turkish documents on the Romanian principalities and the Sublime Porte] (Brăila, 2008). 254 ovidiu cristea

The first difficulty comes in establishing when the two principalities began to take part in the sultans’ wars at all. We might suppose that this must post-date the first tribute payments, but there are still disagreements about when and how a Wallachian or Moldavian prince first joined an Ottoman military expedition. The cause of the controversy is primarily the sources, which are often contradictory and lacking in detail. Sometimes a Romanian presence in the campaigns can only be deduced from Otto- man sources, while Christian documents ignore the Romanian aspects of the sultans’ wars. Sometimes, however, the situation is reversed: Christian sources mention Wallachian or Moldavian involvement, while the Otto- man sources are silent. In addition to this aspect, research is also needed on whether military contributions were only sought in the sultans’ European wars, on whether there was a significant difference between Moldavia and Wallachia regarding the troops they sent, and on whether the Porte showed a particular preference for certain types of military units (cavalry, infantry or auxiliaries). At the same time, further study is needed on the extent to which the princes of the two realms actually responded to the Porte’s requests, on whether the type of troops varied over the course of time, and on whether their military services were limited to only a few months or could last for the duration of the campaign. Equally, we might ask where exactly the Wallachian and Moldavian troops saw action, how effective they were, and what part treaty obligations played, whether they were respected or shirked. The absence of any such analysis in Romanian historiography is all the more surprising when we consider that the vassal relationship to the Porte has never been denied, and that many older and more recent scholars have studied the Romanian principalities’ duties and obligations, looking especially at the amount of tribute paid over time.5 The present study aims to answer some of the questions set out above, without claiming to be a complete overview of the problem. In particular I am aware that research can only yield real results in this area when it is undertaken along with a study of how military participation in Ottoman campaigns was related

5 The most comprehensive treatment of the subject remains Mihai Maxim, Ţările Române şi Înalta Poartă: Cadrul juridic al relaţiilor româno–otomane în Evul Mediu [The Romanian principalities and the Sublime Porte: The legal context of Romanian–Ottoman relations in the Middle Ages] (Bucharest, 1993). romanian participation in ottoman campaigns 255 to other obligations toward the empire (paying tribute, giving gifts,6 supplying provisions and materiel,7 allowing troops to cross their territory, sending hostages). Thus, this article addresses only selected aspects of the topic, considering only military contributions and setting aside the fact that on many occasions Wallachia and Moldavia supplied the Ottoman Empire with oarsmen, food, and logistical support.8 Equally, I only consider those cases in which Romanian troops served in the sultan’s army; the inverse case, where the Porte sent forces to support one or another of the Romanian princes, deserves a study in its own right, although the treaty obligations, at least theoretically, provided that the signatories would be “the friend of our friend and the enemy of our enemy.” Finally I suggest that a nuanced understanding of the military relationship also requires an analysis of the political vocabulary used in diplomatic correspondence.9 Although the same terms were used repeatedly from the fifteenth to the seventeenth century, the Wallachian and Mol- davian voievods’ choice of certain discursive elements varied significantly depending on the ebb and flow of power relations at any given moment and according to their political objective in writing.

Termini Post and Ante Quem

Having suffered raids from Ottoman troops throughout the final decades of the fourteenth century, Wallachia became an Ottoman vassal state during the reign of Mircea the Elder (Mircea cel Bătrân, 1386–1418), even if there is no consensus as to exactly when the sultan’s sovereignty was accepted. One plausible view holds that tribute was first paid immediately after

6 Especially horses and falcons; see, e.g., Maxim, Noi documente, doc. 6, 72–73, and the editor’s comments, 73–80. 7 As was the case in 1572, when the sultan sought the aid of the Wallachian prince for the materials needed to rebuild the fleet; cf. Nicolae Iorga, “Românii şi lupta de la Lepanto” [Romanians and the battle of Lepanto], Revista Istorică 10, nos. 4–6 (1924): 106–107; idem, “Ordinul lui Selim al II-lea către Alexandru Vodă al Ţării Româneşti, 1572” [Selim II’s orders to Alexandru of Wallachia], Revista Istorică 11, nos. 7–9 (1925): 153. 8 In 1543 Wallachia supplied 45 oarsmen to the galleys on the Danube; cf. Maxim, Noi documente, doc. 25–28, 157–171. This contribution was exceptional, since the oarsmen were paid a sum totaling 6,750 aspers. 9 A first sketch on the significance of this topic is Ovidiu Cristea, “Datoria noastră creştinească: Preliminarii la o istorie a limbajului politic în epoca lui Ieremia Movilă” [Our Christian duty: Preliminaries to a history of political vocabulary in the time of Ieremia Movilă], in Movileştii: Istorie şi spiritualitate românească, vol. 2, Ieremia Movilă: Domnul. Familia. Epoca (Suceviţa, 2006), 107–122. 256 ovidiu cristea the crusade of Nicopolis (1396), when Sultan Bayezid I was at the peak of his power and the Wallachian prince was in a vulnerable position.10 It is, however, hard to prove that the Wallachian ruler was bound to furnish troops to the sultan’s army from this same point in time. The hypothesis that Mircea fought in Bayezid’s forces at the battle of Ankara has been conclusively disproven by the Orientalist Aurel Decei, who showed that this argument was based on a misreading. Decei shows that the use of the term V-l-k/Vulkoglu in an anonymous chronicle published by Friedrich Giese refers to the Serbian Prince Stefan Lazarević rather than to the Wal- lachian ruler.11 All things considered, it seems that we can date the first Wallachian– Ottoman military collaboration to Mircea’s reign. Ottoman12 and Byzantine13 sources agree that the Wallachian prince took part in the wars of Ottoman succession supporting the pretender, Musa Çelebi. After the Ottoman prince sought refuge in Wallachia, one of the voievod’s sons led a detachment to join Musa’s troops in fighting against Süleyman for the throne. Musa besieged the citadel of Mesembria in 140914 with Wallachian aid, and subsequently was successful in taking control of the empire’s Euro- pean territories.15 This joint venture probably explains, at least in part, the hostile attitude of Musa’s eventually successful rival, Mehmed I Çelebi, toward the Wallachian rulers in the first years of his reign. After

10 Discussed in detail in Maxim, Ţările Române şi Înalta Poartă, 208–217; see particularly 216–217 for arguments in favor of the year 1396. 11 Aurel Decei, “A participat Mircea cel Bătrân la lupta de la Ankara?” [Did Mircea the Elder fight at the battle of Ankara?], in idem, Relaţii româno–orientale [Romania’s eastern relations] (Bucharest, 1978), 9–14; the same opinion is also held by Nicoară Beldiceanu, “Les Roumains à la batialle d’Ankara: Quelques données sur leur organisation militaire dans la Péninsule Balkanique,” Südost-Forschungen 14 (1955), where he argues that the Vlachs mentioned in the anonymous chronicle were from the Balkan Peninsula. 12 Cronici turceşti privind Ţările Române [Turkish chronicles on the Romanian principalities], vol. 1, sec. XV–mijlocul sec. XVII [Fifteenth to mid-sixteenth century], ed. Mihail Guboglu and Mustafa Mehmet (Bucharest, 1966), 84. 13 Laonic Chalcocondil, Expuneri istorice [Proofs of histories], ed. Vasile Grecu (Bucha- rest, 1958), 113–116. Other Byzantine and Western sources are collected and presented by Dimitris J. Kastritsis, The Sons of Bayezid: Empire Building and Representation in the Otto- man Civil War of 1402–1413 (Leiden and Boston, 2007), 137–144. 14 Peter Schreiner, Die Byzantinischen Kleinchroniken, vol. 1, Einleitung und Text (Vienna, 1975), 215. 15 Şerban Papacostea, “La Valachie et la crise de structure de l’Empire Ottoman,” Revue Roumaine d’Histoire 25 (1986): 23–33 shows that Mircea’s choice to support Musa was rooted in the existing conflict between Wallachia and the Byzantine Empire, which supported Süleyman I. In contrast, Kastritsis, The Sons of Bayezid, 135, argues that Musa acted with the support of Emperor Manuel II Paleologos. romanian participation in ottoman campaigns 257 an extensive campaign along the Danube frontier in 1419–1420,16 Mircea once more became a vassal of the Porte and was bound, according to the chronicler Şükrullah, not just to pay tribute and send hostages, but also, “whenever there is war and the sultan shall ask for their support [the Wal- lachians], to send troops according to the sultan’s will.”17 References to Wallachian participation in Ottoman campaigns multiply from the end of Mircea the Elder’s reign, and the practice was maintained until the end of the seventeenth century. The last Ottoman military expedition in which a Wallachian prince participated took place in 1690, when Constantin Brâncoveanu joined Ottoman forces at Zărneşti in defeating a Habsburg army led by General Donat Heissler.18 It is even more difficult to establish the exact time when this obligation started in the case of Moldavia. Clearly this must have taken place some- time after 1456, the year when tribute payment to the Ottoman Empire began.19 Recently the thesis has been revived that the principality must be counted as having come under the Porte’s hegemony as early as the rule of Voievod Petru Muşat (1375–1391), based on a catalogue entry in a Polish archive that mentions a treaty, now lost, concluded with Sultan Bayezid I in 1392.20 Other sources, however, make no mention of diplomatic or military ties between Moldavia and the Ottoman polity before the fifteenth century, so unless further documents are discovered this hypothesis must be treated with some caution. It is certainly true that indirect testimony dates the sultan’s first orders to raise troops to the reign of Stephen II (Ştefan II, 1433–1447, with

16 Viorica Pervain, “Lupta anti-otomană a Ţărilor Române în anii 1419–1420” [Wallachia’s war against the Ottomans in 1419–1420], Anuarul Institutului de Istorie şi Arheologie Cluj 19 (1976): 55–79; Virgil Ciocîltan, “Competiţia pentru controlul Dunării inferioare, 1412–1420” [The struggle to control the lower Danube], Revista de Istorie 35 (1982): 1191–1203; Eliza- beth A. Zachariadou, “Ottoman Diplomacy and the Danube Frontier, 1420–1424,” Harvard Ukrainian Studies 7 (1983): 684–686. The last-named study has been all but ignored in Romanian historiography, despite some important conclusions. 17 Cronici turceşti, 32; another Ottoman chronicler, Oruc, writing of the same time, does not mention anything except the tribute payment and one of the prince’s sons being sent as a hostage; cf. ibid., 51. 18 Constantin Rezachievici, Constantin Brâncoveanu, Zărneşti 1690 (Bucharest, 1989). 19 Şerban Papacostea, “La Moldavie, état tributaire de l’Empire Ottoman au XVe siècle: le cadre international des rapports établis en 1455–1456,” Revue Roumaine d’Histoire 13 (1974): 445–461. 20 Alexandru Simon, “Bisericile Turcului: Valahii lui Spandounes şi geneza mitropoliilor Ţării Româneşti şi Moldovei” [The churches of the Turk: Spandounes’ Wallachs and the genesis of Wallachian and Moldavian metropolitan sees], Studia Universitatis Babeş-Bolyai: Series Theologiae Orthodoxae 17 (2010): 91–97. 258 ovidiu cristea

interruptions). In a letter, probably dating to 1432, Wallachian Prince Alexander Aldea (Alexandru Aldea, 1431–1436) wrote to István Rozgonyi, comes of Timişoara, that the Ottomans were preparing a new campaign against Transylvania. The letter also mentions Stephen II as being among those called upon to supply troops to the sultan: “[The Turks] have also sought aid from Moldavia; as for that, the prince of Moldavia is of the same mind as I am toward the lord king.” The last part of the phrase is Alexandru Aldea’s assurance to the recipient of the letter that even though he and his Moldavian counterpart must perforce form part of the Otto- man troops, both were ready to switch sides and join the Christians once battle was joined.21 Although this document was published some time ago, it has not received the attention it deserves in the discussion of how Moldavia submitted to Ottoman hegemony. Although the text merely mentions a request from the Porte and does not say that this was met, we might suppose that no such request would have been put unless Moldavia were a vassal state in the same way as Wallachia. It is most likely that this state of affairs did not last long and that Mehmed II Fatih was able to impose a clear relationship of sovereignty on Moldavia. This relationship was established in 1456, during the reign of Petru Aron (1451–1457, with interruptions), and persisted through the first years of the reign of Stephen the Great (Ştefan cel Mare, 1457–1504), when the first joint Moldavian–Ottoman campaign is chronicled. In 1462, Moldavian troops joined Ottoman forces in the unsuccessful siege of the fortress of Chilia. Three different sources record this joint expedition: the chronicle of Laonikos Chalkokondyles, a report from the Venetian bailo at Constantinople dated 28 July 1462, and a letter from the Genoese of Caffa to Casimir IV Jagiello of Poland.22 The siege was part of Sultan Mehmed II’s campaign against the Wallachian prince, Vlad the Impaler (Vlad Ţepeş, 1448, 1456–1462, 1476), and was intended to secure Chilia as an important operational base.23

21 Documenta Romaniae Historica D. Relaţii între Ţările Române [Relations between Romanian principalities], ed. Ştefan Pascu, et al. (Bucharest, 1977), 292. 22 All three sources are collected and presented by Nicolae Iorga, Studii istorice asupra Chiliei şi Cetăţii Albe [Historical studies on Chilia and Cetatea Alba] (Bucharest, 1899), 124–125. 23 For the importance of Chilia in crusade plans, see Francisc Pall, “Byzance à la veille de sa chute et Janco de Hunedoara (Hunyadi),” Byzantinoslavica 30 (1969): 240–248; Şerban Papacostea, “Kilia et la politique orientale de Sigismond de Luxembourg,” Revue Roumaine d’Histoire 15 (1976): 421–436. romanian participation in ottoman campaigns 259

Romanian historians have tried to separate Stephen’s actions from Mehmed II’s attempts to occupy this strategic point at the mouth of the Danube and have suggested that the Moldavian prince was even trying to prevent Chilia from falling into the sultan’s hands.24 Even if Stephen really was trying to incorporate the stronghold into Moldavian territory, all external sources on the period either explicitly state, or at the very least suggest, that the campaign of 1462 was undeniably the result of a pact with Sultan Mehmed II.25 After 1462 Moldavia drifted further away from the Ottoman Empire’s political goals and by 1473 was in open conflict with it. When this war was concluded in the truce of 1486,26 the path was open once more for military cooperation, next seen in practice in 1497–1498. It is a paradox, however, that we know of no single situation in which Stephen sent help to the sultan. In contrast, in 1497, Bayezid II sent an important contingent of troops under Mesih Pasha to help Moldavia in its conflict with King John I Albert of Poland. The peculiarities of Moldavia’s status during Stephen the Great’s reign led to a number of episodes in which the Porte, rather than requiring direct involvement, sought “benevolent neutrality” or more specifically, permission to cross Moldavian territory. In at least two cases, the response was far from what might have been expected from an obedient vassal.

24 A fully-supported criticism of this argument is Şerban Papacostea, “Relaţiile internaţionale ale Moldovei în vremea lui Ştefan cel Mare” [Moldavia’s international relations in the time of Stephen the Great], Revista de Istorie 35 (1982): 607–638; idem, “Comerţ, alianţe şi acţiune militară în politica lui Ştefan cel Mare la începuturile domniei, 1457–1462” [Commerce, alliance and warfare in Stephen the Great’s policy at the start of his reign, 1457–1462] in Ştefan cel Mare şi Sfânt: Atlet al credinţei creştine [Saint Stephen the Great: Athlete of Christendom] (Putna, 2004), 445–454. The same perspective is shared by Matei Cazacu, “Du nouveau sur le rôle international de la Moldavie dans la seconde moitié du XVe siècle,” Revue des Etudes Roumaines 16 (1981): 41. 25 Internal sources mention the failed siege without any reference to Ottoman collaboration. This lack of reference is not surprising when we remember the “Constantinian model” used in writing these texts. See Andrei Pippidi, Tradiţia politică bizantină în ţările române în secolele XVI–XVIII [The Byzantine political tradition in the Romanian principalities, sixteenth to eighteenth centuries] (Bucharest, 1983), 69, 146; Liviu Pilat, Între Roma şi Bizanţ: Societate şi putere în Moldova (sec. XIV–XVI) [Between Rome and Byzantium: Society and power in Moldavia, fourteenth to sixteenth centuries] (Iaşi, 2008), 461–474. 26 For the dating and conditions of the treaty of 1486, Ştefan S. Gorovei and Maria Magdalena Székely, Princeps omni laudae maior: O istorie a lui Ştefan cel Mare [Prin- ceps omni laudae maior: A history of Stephen the Great] (Sfânta Mănăstire Putna, 2005), 238–248. 260 ovidiu cristea

In 1480–1481 the Porte asked the prince’s acquiescence in allowing their troops to attack Poland, though we do not know Stephen’s reply.27 In contrast, in 1492 the Venetian bailo, Ieronimo Marcello, mentioned in one of his reports that the prince had refused the sultan’s request that the Tatars be allowed to cross his realm;28 a similar situation dating to May 1498 should be seen as a result of the Polish–Moldavian conflict that had broken out the previous year. According to the Moldavian Anony- mous Chronicle, the Moldavian prince allowed the sancakbeyi of Silistra, Malkoçoğlu Balı Bey, to lead his troops across the territory en route to Poland, where the Ottomans plundered the south of the kingdom as far as L’viv and even beyond.29 A similar raid, with an unexpected outcome, took place at the end of the same year. The Moldavian German Chronicle (which dates the events to February 1499) records that on this occasion the invading force suffered cruelly from cold and hunger and that those who escaped were crushed on their way home by Stephen’s army. However unlikely this may seem, this version is confirmed by the Polish chronicler Bielski, who also adds a new element. In order to avoid conflict with the Porte, the Moldavian prince supposedly dressed his troops in the Polish fashion. Even if this detail is merely the result of the chronicler’s imagination, it is certain that this hostile act provoked no reaction from the Porte, which maintained peaceful relations with Moldavia for a long time afterward. A comparable episode to that of 1499 took place in 1524, during the reign of Stephen the Young (Ştefan cel Tânăr, 1517–1527), when a Turkish– Tatar army that had been plundering the south of Poland was defeated at Tărăsăuşi on the Prut.30 Such engagements seem to have been enabled by

27 The document is published by Ioan Bogdan, Documentele lui Ştefan cel Mare (Bucha- rest, 1913), vol. 2, 365; for the chronological context and an assessment of its importance, see Ştefan S. Gorovei, “La paix moldo–ottomane de 1486 (Quelques observations en marge de textes),” Revue Roumaine d’Histoire 21 (1982): 405–421. 28 Documente privitoare la Istoria Românilor culese de Eudoxiu de Hurmuzaki [Docu- ments concerning the history of the Romanians collected by Eudoxiu de Hurmuzaki], vol. 8, 1376–1650, ed. Ioan Slavici (Bucharest, 1894) [henceforth Hurmuzaki VIII], 28. 29 Cronicile slavo-române din sec. XIV–XVI publicate de Ioan Bogdan [Romanian Slavonic chronicles of the fourteenth to sixteenth centuries, published by Ioan Bogdan], ed. Petre P. Panaitescu (Bucharest, 1959), 23. 30 Cronicile slavo-române, 94; for a commentary on this episode, see Andreescu, Res- titutio Daciae, 54, who suggests that Ottoman difficulties in maintaining their control of Wallachia explain why Stephen the Young’s action went unpunished. It was probably rele vant here that Süleyman the Magnificent was preparing a campaign against Rhodes that promised to be extremely difficult. On this last aspect see Nicolas Vatin, “La conquête de Rhodes,” in Soliman le Magnifique et son temps, ed. Gilles Veinstein (Paris, 1992), 435–454. romanian participation in ottoman campaigns 261 the delicately balanced regional political context. The situation changed drastically after the Hungarian defeat at Mohács in 1526. After this point, insubordinate acts became much rarer and the duty to send military aid to the Porte became a recognized custom. The first campaigns date from June and October–November 1529, when the Mol- davian prince, Petru Rareş (1527–1546, with interruption), joined the wars in Transylvania to support János Szapolyai at the Porte’s request.31 After Süleyman the Magnificent marched on Moldavia in 1538 there are many more examples of participation in the Porte’s campaigns. Hostile or insubordinate actions, such as we find in Stephen the Great’s time or Stephen the Young’s, became hard to imagine and implement.

The Theaters of War

Another matter for discussion is whether the obligation of the Romanian princes to be “the friend of our friend and enemy of our enemy” applied to all of the Porte’s military expeditions. It seems that only neighboring realms were understood under these duties. Wallachian troops took part in the campaigns in Hungary (until 1541), the Habsburg lands, Moldavia, and Poland, while Moldavia is mentioned as taking part in expeditions against Hungary, the Habsburgs, Wallachia, and Poland. It is hard to establish exactly why these two countries only took part in campaigns in their immediate region. It is likely that the most cogent motive was the Romanian princes’ unwillingness to leave their countries undefended. Ottoman chronicles frequently report that the Wallachian voievods excused themselves from marching with the sultan because they feared invasions by other powers. This is given as a reason for refusing a request for military duty and for failing to appear in person with tribute.32 The Moldavian prince, Stephen the Young, offered such an explanation in 1521, invoking Tatar and Polish threats on his borders, when he refused to attack Transylvania.33

31 Andreescu, Restitutio Daciae, 79, argues that although acting on the orders of the Porte, Petru Rareş was also following his own interests in Transylvania, in particular in securing the citadels of Ciceu and Cetatea de Baltă. 32 E.g., Aşık Paşazade in Cronici turceşti, 92. 33 The document is published and annotated by Tahsin Gemil, “Din relaţiile moldo– otomane în primul sfert al secolului al XVI-lea” [On Moldavian–Ottoman relations in the first quarter of the sixteenth century], Anuarul Institutului de Istorie şi Arheologie “A.D. Xenopol” 9 (1972): 142. 262 ovidiu cristea

From the sixteenth century onward this argument was strengthened by the further fear of domestic political upheaval. The absence of the prince could favor a rival boyar faction seeking to place its own candidate on the throne. A well-known example dates to 1683, when the Moldavian prince, Gheorghe Duca (1665–1683, with interruptions), returning from the unsuccessful siege of Vienna, found Ştefan Petriceicu (1672–1684, with interruptions) on the throne in Moldavia and ended by being taken prisoner and held captive in Poland. Further research is needed on the question of whether armies from these two realms took part in expeditions farther afield. There are some hints in this direction, but only in the case of Wallachia. Although—as mentioned above—it is hard to accept the suggestion that Wallachian troops fought at the battle of Ankara (1402), nevertheless several narrative sources for the year 1473 mention Wallachians among Mehmed II’s allies at the battle of Başkent. Two Oriental sources, the chronicles of Ebu Bekr-i Tihrani and Hasan-ı Rumlu, recording the Ottoman war against the Ak Koyunlu khan Uzun Hasan in 1473, report: “A great host had assembled at the imperial palace [that is, Constantinople] from the regions of Morea, Bosnia, Wallachia, Alanya, Anatolia, Menteshe, Karamania, Kasta- monu, and other parts of Rum.”34 The reference to Wallachian troops is all the more interesting because at the time in question the principality was struggling in an ongoing war with Stephen the Great in Moldavia. In November of that year, the Wallachian prince, Radu the Handsome (Radu cel Frumos, 1462–1475, with interruptions), was defeated by his Moldavian counterpart, his capital was occupied, and his family captured. Due to the timing, there is good reason to question the accuracy of our information, were it not confirmed by two Western sources, the His- toria Turchesca of Giovanni Maria Angiollelo and an account of Caterino Zeno’s embassy in Persia. These texts mention Wallachian participation in the campaign, give their numbers (12,000 men) and their position in the Ottoman troop deployment (on the left flank, led by Mustafa, one of the sultan’s sons who later distinguished himself in the fighting at Başkent).35

34 The citation is from Ebu Bekr-i Tihranî, Kitab-ı Diyarbekiriya (Akkoyunlu, Kara- koyunlu ve Çagatay Dönemleri Tarihi) [A book about Diyarbekir: The history of the epochs of Akkoyunlu, Karakoyonlu, and Çagatay], ed. Mehmet Demirdag (Istanbul, 1999), 323; Hasan-ı Rumlu, Ahseni’t-tevarîh [The most beautiful chronicle], ed. Mürsel Öztürk (Ankara, 2006), 511, adds Trebizond to the list. I am grateful to my colleague Nagy Pienaru for drawing my attention to these passages and translating them from the original. 35 Donado da Lezze, Historia Turchesca, 1300–1514, ed. Ion Ursu (Bucharest, 1909), 46; Dei Commentarii del Viaggio in Persia di M. Caterino Zeno il K[avaliere] et delle guerre fatte romanian participation in ottoman campaigns 263

Even if the number of soldiers cited is exaggerated, the reference offers solid support for the argument that the sultan’s ally in this battle was indeed the prince of Wallachia, not a pretender at the Ottoman court. Such a presence on the Asian front is quite exceptional and we might suppose that the Wallachian Prince Radu the Handsome felt that he had to honor his military obligations precisely because he greatly needed Ottoman support in his own dispute with his neighbor. He hoped that by demonstrating his loyalty he might gain the sultan’s effective support in the long run. Besides this campaign in 1473, there is no mention of Wallachian contingents serving in any subsequent Ottoman campaign against Egypt or Safavid Persia. We might suppose that the Wallachian troops fighting on such a remote front in 1473 were fairly ineffective and that Mehmed II’s heirs consequently decided not to call upon soldiers little prepared for the rigors of a campaign on the Eastern front.

Forces, Tactical Roles, Effectiveness

The exact number of forces from Wallachia and Moldavia fighting for the Ottoman Empire is a vexing question, with historians estimating numbers either from documents that exaggerated the two countries’ military potential or by using inexact methods to calculate their figures.36 For the fifteenth century, generally accepted numbers are around 40,000 soldiers for Moldavia and between 10,000 and 30,000 for Wallachia. For the following century, the sources range between 15,000 and 60,000 men in arms for Moldavia and between 10,000 and 30,000 for its neighbor. Finally, in the first half of the seventeenth century, Moldavia was able to muster between 20,000 and 37,000 men while Wallachia could furnish 30,000 to 40,000 men.37 Such figures would have put the principalities on an equal footing with the great Western kingdoms, but the costs of recruiting,

nell’Imperio Persiano dal tempo di Ussuncassano in qua libri due (Venice, 1558), 16 recto: “dodici mila Valacchi condotti da basaraba lor Capitano, che venne in aiuto del Turco in quella guerra.” 36 For the latter problem, see Radu Rosetti, Istoria artei militare a românilor până la mijlocul veacului al XVII-lea [A History of Romanian warfare until the mid-seventeenth century] (Bucharest, 1947), 136–137. He considers several sources claiming that Stephen the Great could field 40,000 men as credible and realistic and argues that in later fifteenth- century Moldavia there were around 2,000 villages with about 20 houses in each. 37 For these figures, and the sources on which they are based, see Rosetti, Istoria artei militare, 333–334. 264 ovidiu cristea equipping, and training such armies would have required exorbitant sums of money. A well-known and well-documented case from 1599 shows the Wallachian prince invading and taking Transylvania with an army numbering 13,281 mercenaries among a total force of around 30,000 men. Accounts show that their wages amounted to almost one million thalers per year.38 In the absence of any exact figures for the two countries’ forces, we are equally uncertain how many soldiers the Romanian princes were obliged to supply to the Porte in time of war. There are no continuous records on the subject, and even where partial records exist for some periods, these all come from different authors. Often we do not know what sources the authors themselves were drawing from, how reliable they may have been or how we should interpret them. An anonymous history of the Ottoman Empire from 1556 gives the following figures: Moldavia was required to furnish the sultan with 35,000 cavalry (among a total of 60,000 men), while Wallachia’s obligation came to 25,000 cavalry (from a total of 150,000[!]), and Transylvania supplied only 10,000 cavalry. These figures may be hard to accept, but are also found in other Western sources (diplomatic reports, travelers’ accounts, chronicles).39 Internal and Ottoman sources in turn are not much more informative. There are references to military participation in the Porte’s campaigns, but few details on troop numbers. The orders sent to Transylvania and Wallachia in 1574 are preserved, in which the principalities were asked to join the sultan’s armies on the campaign in Moldavia. The order of 25 April requires some cannon and an expeditionary force, though the size of the latter is not specified; on 18 May a plan of attack against Mol- davia is worked out, marching orders are given, and the orders are confirmed on 13 June.40 None of these documents mentions actual numbers; they only confirm that Wallachia was indeed involved in the Ottoman

38 Nicolae Stoicescu, “Oastea lui Mihai Viteazul” [Michael the Brave’s army], in Mihai Viteazul: Culegere de studii, ed. Paul Cernovodeanu and Constantin Rezachevici (Bucha- rest, 1975), 86. 39 Andrei Pippidi, “Descrieri ale Ţărilor Române din secolul XVI” [Descriptions of the Romanian principalities in the sixteenth century], in Românii în istoria universală, ed. Gheorghe Buzatu (Iaşi, 1988), 91; and Călători străini despre Ţările Române [Foreign travelers in Romanian principalities], vols. 1–3, ed. Maria Holban, Paul Cernovodeanu, and Maria Matilda Alexandrescu-Dersca Bulgaru (Bucharest, 1968–1971), passim. 40 The documents are in Başbakanlık Osmanlı Arşivi, Mühimme Defteri, no. 24, doc. 474, p. 176; Mühimme Defteri, no. 24, doc. 236, p. 610; Mühimme Defteri, no. 26, doc. 36, p. 13. I am indebted for these details to my colleague, Turcologist Nagy Pienaru, who was able to locate and elucidate them for me. romanian participation in ottoman campaigns 265 campaign against Moldavia, along with Tatar and Transylvanian forces. The campaign ended in Ioan the Terrible’s (Ioan cel Cumplit, 1572–1574) defeat at Cahul (10 June 1574). In certain exceptional situations we have precise and credible data, but it is extremely risky to extrapolate from these to an entire century, let alone the whole period of the fifteenth to seventeenth century. Further research is needed on whether the two principalities were obliged to bring their armies to any given Ottoman campaign, or whether this obligation was restricted to expeditions led by the sultan himself. Finally, the role of these vassal contingents in the Ottoman army must be investigated. Did they have a well-defined tactical role? Were they simple auxiliary units? Were they merely a means for the Porte to ensure the loyalty of the two rulers? The answers vary depending on the events we examine and the sources in which they are recorded. The first mentions of participation in Ottoman campaigns are unclear, with sources suggesting that Mircea the Elder sent only one detachment of troops, albeit led by one of his sons, to support Musa Çelebi in his fight for the Ottoman throne. The situation in 1409 is special, however. The Wallachian troops were fighting for a pretender rather than for an established emperor and thus this instance of military aid cannot be used to draw clear conclusions. For two of Mircea’s successors, Alexandru Aldea and Vlad Dracul, it seems that the prince’s presence was obligatory when the sultan himself was leading the expedition. The above-mentioned letter of Alexandru Aldea to István Rozgonyi seems to suggest this: “if this [Ottoman] army arrives in Transylvania, I, too, will be with them; if not me, then I will send Albul.”41 Alexandru Aldea’s uncertainty as to who would lead the Wallachian detachment is not explained, but we suppose that if Murad II did not come himself then the prince would delegate leadership to his highest-ranking boyar.42 Six years later, Vlad Dracul rode with the same sultan’s army, which laid waste to Transylvania in a lightning raid that lasted twenty-three days.43 According to the chronicler Tursun Bey, Vlad Dracul had the choice of obeying the call or being attacked himself: “Call

41 Documenta Romaniae Historica, D, 292. 42 For the career of the boyar Albu, see Nicolae Stoicescu, Dicţionar al marilor dregători din Ţara Românească şi Moldova, sec. XIV–XVII [Dictionary of the high officials of Walla- chia and Moldavia, fourteenth to seventeenth centuries] (Bucharest, 1971), 15. 43 For a discussion of the chronology of the campaign, see Ciocîltan, Între sultan şi împărat, 1768–1771. 266 ovidiu cristea your army together quickly and come,” the sultan supposedly told him, adding: “If you do not come, I will not hesitate to seek you out.”44 The same chronicler adds that on the return march, the Wallachian troops were in the vanguard, which the Serbian despot Đorđe Branković had taken on the outward march. We have no data on the number of troops in any of these instances. The first numerical data are from the time of Vlad Dracul’s son, Radu the Handsome, concerning Mehmed II’s wars with Uzun Hasan. According to Giovanni Maria Angiollelo, the Wallachian contingent numbered no fewer than 12,000 men, a number which historians have taken as an estimate of the size of the entire Wallachian army of the period.45 Though the number might at first sight seem greatly exaggerated, it is confirmed by another Venetian source, the account of Caterino Zeno’s embassy to Uzun Hasan. This contingent supposedly formed an important part of Prince Mustafa’s force, which the same author estimates at around 30,000 men, and which formed the left flank of Mehmed II’s army at the battle of Başkent (11 August 1473). The instance of 1473 is unusual. It is the only documented case of a Romanian ruler taking part in an expedition so far away from his own borders with a force representing almost the entirety of his country’s army. The force of 12,000 Wallachians is also mentioned by Angiolello in connection with Mehmed II’s campaign in Moldavia in 1476; in this instance, the author reports that the Wallachian contingent no longer took the Ottoman left flank, but encamped separately from the Ottoman force (“et allogiava separatemente dietro il campo”).46 Other fifteenth-century engagements saw smaller forces taking part in campaigns much closer to their own borders. In 1475, Laiotă Basarab (1473–1477, with interruptions) brought Wallachian troops to the army of the beylerbeyi of Rumelia, Süleyman, which had suffered a defeat at Vaslui (24 January).47 Four years later, Basarab Ţepeluş (1477–1482, with interruption) sent 5,000 men to join the Ottoman army that invaded Transylvania, which was defeated by István Báthory and Pál Kinizsi at Câmpul Pâinii (13 October 1479).48 Wallachian troops also fought alongside the Ottomans

44 Cronici turceşti, 87. 45 Rosetti, Istoria artei militare, 138. 46 Donado da Lezze, Historia Turchesca, 89. 47 Kemalpaşazade, in Cronici turceşti, 209. 48 A discussion of the sources for this battle can be found in Ioan Haţegan, Pavel Chi- nezu (Timişoara, 1994), 139. romanian participation in ottoman campaigns 267 in Moldavia in 1484–1486. Tursun Bey mentions Wallachian troops under Vlad the Monk (Vlad Călugărul, 1481–1495, with interruption) serving as Bayezid’s vanguard in the expedition of 1484 against Chilia and Cetatea Albă.49 Oruc in turn reports that the same prince aided the Ottomans in battles against Stephen the Great at Şcheia (1485) and Cătlăbuga (1486).50 An anonymous Ottoman chronicle from the reign of Bayezid II mentions a Wallachian detachment fighting alongside Mesih Pasha’s troops in the Moldavian–Polish war of 1497, although this source gives a doubtful account of how events played out.51 The situation is even more unclear for Moldavia in the fifteenth century. Other than the episode of 1432, when indirect testimony reports that the Moldavian ruler summoned his army for a campaign in Transylvania, the only occasion on which Moldavian troops supported the sultan’s army was in 1462. According to the chronicler Laonikos Chalkokondyles, Sul- tan Mehmed II ordered Stephen the Great to attack Chilia from the land while the Ottoman fleet besieged the fortress from the water. There are no precise numbers and the Moldavian chronicles that do mention the siege offer no further details than that Stephen was wounded in the leg. Nor is the situation regarding troop numbers much clearer in the sixteenth century. Some sources give exact numbers, but these relate only certain points in the history of Romanian relations with the Porte. Alvise Gritti’s 1534 report to Sultan Süleyman the Magnificent mentions Wal- lachia’s duty to supply 3,000 cavalry;52 according to a Venetian account from Constantinople, both Wallachia and Moldavia sent this number in 1532, when Gritti took part in the sultan’s campaign in Hungary and the unsuccessful siege of Esztergom.53 These details led the account’s editor, Aurel Decei, to suppose that Wallachia and Moldavia’s military obligations to the Porte in the sixteenth century came to 3,000 cavalry. It is, however, hard to generalize from the events of 1532–1534 to the whole of the first half of the sixteenth century. A comparison with Moldavia’s obligations to Poland is instructive. As vassals of the Polish crown, the Mol- davian princes were obliged to furnish their lords with 4,000 men in the

49 Cronici turceşti, 99; see an analysis of Ottoman sources and of the troops in the campaign of 1484 in Nicoară Beldiceanu, “La conquête des cités marchandes de Kilia et de Cetatea Albă par Bayezid II,” Südost-Forschungen 23 (1964): 36–90, especially 62–64. 50 Cronici turceşti, 63. 51 Cronici turceşti, 137. 52 Aurel Decei, “Aloisio Gritti au service de Soliman le Magnifique d’après des documents Turcs inédits, 1533–1534,” Anatolia Moderna 3 (1992): 24. 53 Hurmuzaki VIII, 61. 268 ovidiu cristea fifteenth century and with 7,000 men according to some sixteenth-century accounts.54 Wallachia took part in several Hungarian campaigns against the Ottomans in the fourteenth and fifteenth centuries, although, unfortunately, there are no exact numbers. Only two years after Alvise Gritti’s account, Mark Pemfflinger and Boldizsár Bánffy, Ferdinand of Habsburg’s emissaries in Moldavia, offer a different picture of the Romanian princes’ military duties to the Porte. The Moldavian prince, Petru Rareş, apparently said that he was bound to send 6,000 men to the aid of János Szapolyai, while Wallachia was asked for only 4,000.55 It is likely that the number of men sent as support to the Ottomans varied depending on the scale of a campaign, its strategic aims, and the other duties of the country (supplying oarsmen for the Danube fleet, paying to arm the fleet,56 furnishing fodder and provisions,57 etc.). Aurel Decei’s figure of 3,000 horsemen represents only one occasion, which should not be used as a basis for extrapolation in a period when, as historians generally accept, Ottoman suzerainty was growing stronger and the principalities’ military obligations grew correspondingly. References abound in the seventeenth century, although without more exact figures. Internal chronicles use standard phrases which may mean

54 For these numbers see Andrei Pippidi, “Moldavie et Pologne: la fin de la vassalité,” Acta Poloniae Historica 83 (2001): 65–68. 55 Călători străini despre Ţările Române [Foreign travelers in the Romanian principalities], vol. 1, ed. Maria Holban (Bucharest, 1968), 376. 56 In a report of 27 July 1551, the Venetian bailo at Constantinople, Bernardo Navagero, mentions that the prince of Moldavia, Ştefan Rareş, had been called upon to provide 10,000 ducats to arm the Ottoman fleet. The document is published in Ovidiu Cristea, “Si e fatto Turcho: di ricco povero, di signor schiavo: Bailul veneţian Bernardo Navagero despre turcirea lui Iliaş Rareş” [The Venetian bailo Bernardo Navagero on Iliaş Rareş’ apostasy], Studii şi Materiale de Istorie Medie 27 (2009): 97. 57 Many sources mention that during military campaigns food was scarce in Con- stantinople and prices rose precipitously. There are two opposing viewpoints in recent Romanian historiography on the Romanian role in provisioning Constantinople. Bogdan Murgescu (“Did Moldavia and Walachia export grain during the 16th century?” in Miscel- lanea in honorem Radu Manolescu emerito, ed. Zoe Petre and Stelian Brezeanu (Bucharest, 1996), 190–196) argues that the two principalities did not play a significant role in supplying Constantinople in the sixteenth century, while Ştefan Andreescu (“Răscoala Ţărilor Române şi chestiunea aprovizionării Constantinopolului” [The Romanian uprising and the question of supplies for Constantinople], Revista Istorică 8 (1997): 591–614) argues the opposite. There is no such debate for the seventeenth century, although sources continue to link supply problems in the capital with Romanian obligations to the sultan’s army; see, for example, the report of the Venetian bailo, Pietro Foscarini, of 24 February 1634, published in Documente privitoare la Istoria Românilor culese de Eudoxiu de Hurmuzaki [Documents concerning the history of the Romanians collected by Eudoxiu de Hurmu- zaki], vol. 4, part 1, 1600–1650, ed. Ioan Slavici (Bucharest, 1884), 474. romanian participation in ottoman campaigns 269 anything or nothing: “the armies marched, and the sultan was with them with all his strength,” “the armies marched (. . .) together with a great mul- titude of Turks,” “they marched with armies at the emperor’s bidding.” We might suppose that in a period of Wallachian military decline and with the numbers of troops falling, the forces sent to the aid of the Porte fell correspondingly. Such an assumption is not, however, supported by sources that estimate Wallachian and Moldavian contingents at several thousand men.58 The significant change was in the efficacy of these troops, which were less and less suited to the conditions of seventeenth- century warfare. Giovanni Luigi Zani, a Bolognese officer in Habsburg service, observed in 1663 that there were 4,000 Wallachs among the Ottoman forces that invaded Transylvania, armed with hand weapons but with few firearms.59 The same observer noted that their discipline left a great deal to be desired and that these troops could not be considered as fearful adversaries.60 This is supported by the fact that the Porte began to use the troops sent from the two principalities more often for works and corvée labor (building bridges and fortifications), although without entirely giving up on using them in battle. This summary overview certainly contains numerous question marks, but nevertheless allows a few conclusions, hypothetical though these must be. Moldavian and Wallachian contingents varied in number from the fifteenth to the seventeenth century, depending on the magnitude of the Ottoman campaign, the presence of the sultan, and the other services that the two principalities performed for the Porte. The size of these forces probably never exceeded several thousand soldiers and sources agree that they were most often cavalry. It is difficult to estimate exactly how effective these units were in the outcome of a campaign. In expeditions to Transylvania or Moldavia, Wallachian troops served as the vanguard of the sultan’s army, with their local knowledge being decisive in this role. We can suppose that the Ottomans were more concerned

58 E.g., Documente privitoare la Istoria Românilor: Urmare la colecţiunea lui Eudoxiu de Hurmuzaki [Documents concerning the history of the Romanians. A continuation of Eudoxiu de Hurmuzaki’s collection], Suppl. 1, vol. 1, 1518–1780, ed. G.G. Tocilescu and A.I. Odobescu (Bucharest, 1883) [henceforth, Hurmuzaki, Suppl. I/1], 229: the French ambassador in Istanbul estimated the Moldavian troops sent to support the Ottomans in 1628 from 8,000 to 10,000 men. 59 Ovidiu Cristea, “Un viaggiatore italiano nella Transilvania del Seicento: il bolognese Giovanni Luigi Zani,” in L’Italia e l’Europa Centro-Orientale attraverso i secoli, ed. Cristian Luca, Gianluca Masi, and Andrea Piccardi (Brăila and Venice, 2004), 290: “armati di dardi, freccie e sabla.” 60 Cristea, “Un viaggiatore,” 291. 270 ovidiu cristea to secure the loyalty of the two princes than to actually reinforce their troop numbers. It is hard to say exactly how useful the Wallachian or Mol- davian troops were; a source from the mid-sixteenth century seems to say that one positive feature of the sultan’s troops in a battle against Ferdi- nand of Habsburg was the presence of Tatar, Wallachian, and Moldavian auxiliaries.61 Although we have no particular reason to doubt these words of the Venetian bailo, Bernardo Navagero, their relevance should not be exaggerated. The context of this discussion shows that the Ottoman dignitary was trying to extract information from his interlocutor about the Habsburg’s strengths and weaknesses. Impressed by the number of ships that Charles V was able to put under sail and under arms in the Mediter- ranean in peacetime, the grand vizier sought to impress the Venetians in turn with the size of the sultan’s land armies. It must be said that, apart from this episode, the Ottomans seem to have doubted the loyalty of the two Romanian princes throughout the period under discussion. This distrust was not unfounded. Wallachian and Moldavian sources record the two rulers’ desire to join the side of the Christian powers when they were at war with the Ottoman Empire. Alex- andru Aldea’s letter of 1432, quoted above, leaves no trace of doubt on this subject: “When your army arrives, I will take my army away at that same hour. The Turks have deceived me many times, and I will do the same, by the Lord, so that they may be entirely rooted out.”62 The Moldavian prince, Stephen the Young, took a similar stance in the treaty of Hârlău (4 May 1518) concluded with King Sigismund I of Poland.63 Lastly, in 1663 Zani relates an almost identical conversation with an envoy from the Wal- lachian Prince Grigore Ghica (1660–1664, 1672–1674): [the messenger] told us precisely that their voievod was forced to fight against us, and that his troops were poorly armed and ill-disciplined. He told us that we should advise our generals to bring five or six thousand soldiers and attack decisively, and they would be sure to have the victory. And that they would all gladly die, if they knew that Christendom would keep the advantage thereby.64

61 Cristea, “Si e fatto turcho,” 94–95: “perche oltra le nostre forze havemo adesso Tartari al nostro comando Valacchi e Bogdani sogiongendo.” 62 Documenta Romaniae Historica, D, 292; the same undertaking is given in a letter to the town magistrates of Sibiu, ibid., 296. 63 Mihai Costăchescu, Documente moldoveneşti de la Ştefăniţă voievod, 1517–1527 [Molda- vian documents on the voievod Ştefăniţă, 1517–1527] (Iaşi, 1943), 500. 64 Cristea, “Un viaggiatore,” 291–292. romanian participation in ottoman campaigns 271

The prince of Moldavia showed the same attitude to this campaign when he told Anna Bornemisza, wife of the prince of Transylvania, that he had ordered his armies not to cause damage in Transylvania “for we are moved by that same Christian sentiment (. . .) we did not come by our own will, rather others have forced us to come.”65 An example showing that such protestations were not merely given for form’s sake is the Wallachian– Moldavian desertion during the campaign of Hotin (1673) as well as their friendly attitude to the Habsburgs during the siege of Vienna in 1683.66 This ambiguous or even duplicitous attitude emerges in other episodes, too, when the Porte sought strength from the two countries. In such situations, the princes used various excuses for not joining the sultan’s armies. Thus, according to the Venetian bailo in Constantinople, Girolamo Mar- cello, in 1492 Bayezid II ordered the Wallachian prince, Vlad the Monk, to send troops for a campaign against Hungary, and he also required Ste- phen the Great to allow Tatar forces to cross Moldavian territory to join the Ottoman army. Both princes refused. Vlad the Monk claimed that if he were to obey, his country would be attacked while his armies were away in Hungary, while Stephen simply refused without explanation. The bailo’s report notes that the sultan thought the Wallachian ruler’s reply was justified, but does not mention what reaction the Moldavian response provoked.67 We might suppose that Stephen the Great’s attitude was due to his comparatively privileged status in dealing with the Porte, in contrast to that of Wallachia, which is documented by both Ottoman and Western sources.68 Such episodes were not unusual. In 1521, the Moldavian prince Stephen the Young refused to join the sultan’s army, speaking of a Polish and Tatar threat on his own borders.69 Five years later, both Wallachia and Moldavia

65 Andrei Veress, Documente privitoare la istoria Ardealului, Moldovei şi Ţării Româneşti [Documents on the history of Transylvania, Moldavia, and Wallachia], vol. 11, 1661–1690 (Bucharest, 1939), 40. 66 Victor Papacostea, Civilizaţie românească şi civilizaţie balcanică: Studii istorice [Roma- nian civilization and Balkan civilization: Historical studies] (Bucharest, 1983), especially chapter V, Românii la asediul Vienei [The Romanians at the siege of Vienna]. 67 Hurmuzaki VIII, 28. 68 Ovidiu Cristea, “Pacea din 1486 şi relaţiile lui Ştefan cel Mare cu Imperiul Otoman în ultima parte a domniei” [The peace treaty of 1486 and the relations of Stephen the Great with the Ottoman Empire in the last years of his reign], Revista Istorică 15 (2004): 25–36. 69 See above, note 33. 272 ovidiu cristea excused themselves from taking part in the campaign against Hungary, explaining that they were at war with one another.70 Another interesting case dates from 1534. Sent to demarcate the boundary between Habsburg and Ottoman territories, Alvise Gritti was also ordered to collect the Wallachian tribute and to require Vlad Vintilă (1532–1535) to send a contingent of 3,000 cavalry. Alvise Gritti’s adventures and misadventures on his Wallachian journey are sufficiently known that they need not be recounted here in detail.71 Suffice it to say that in the end this emissary of the Porte did not obtain the money or the troops for which he had been sent. Of the 3,000 cavalry, the prince sent only a much smaller number and the Venetian reckoned that even these were hardly reliable. In this instance the lack of any response from the Porte may be explained by the unfolding war with Persia, or, more likely, by the intrigues in the imperial divan around Alvise Gritti and his protector, İbrahim Pasha, who lost the power and influence he had enjoyed up until that time. In turn, the refusal of Petru Rareş, prince of Moldavia, to send troops to support János Szapolyai in 1536, as ordered by the sultan, was one reason for Süleyman the Magnificent’s campaign in Moldavia in 1538.72 From the second half of the sixteenth century, the rise of Ottoman power seriously reduced the room for maneuver available to the rulers of the two Romanian principalities. The Porte was not inclined to allow Wal- lachia and Moldavia any freedom of initiative in foreign affairs, and was even less willing to see them wage war on their own account. Even in the new balance of power, though, some princes were able to take advantage of the ebb and flow of political events and refused to obey the sultan’s will. In 1628, the Moldavian prince, Miron Barnovschi (1629–1633, with

70 Andreescu, Restitutio Daciae, vol. 1, 56–58, considers that this was a trumped-up conflict which the two voievods used as a pretext in their dealings with the Porte. This hypothesis may be ruled out since the Porte had plenty of informants in both countries and the risk of being discovered faking a war would have been enormous. It is much more likely that this was a flare-up of ongoing conflicts that lasted through Stephen the Great’s reign and throughout the first half of the sixteenth century. 71 Heinrich Kretschmayr, “Ludovico Gritti: Eine Monographie,” Archiv für Ostereichische Geschichte 83 (1897): 1–106; Decei, “Aloisio Gritti,” 10–60; Cristina Feneşan and Jean-Louis Bacqué-Grammont, “Notes et autres documents sur Aloisio Gritti et les Pays Roumaines,” Anatolia Moderna 3 (1993): 61–103; Ferenc Szakály, Lodovico Gritti in Hungary, 1529–1534: A Historical Insight into the Beginnings of Turco–Habsburgian Rivalry (Budapest, 1995); Adriano Papo, “Ludovico Gritti e i Principi Romeni,” Annuario: Istituto Romeno di Cultura e Ricerca Umanistica (Venezia), 6–7 (2004–2005): 353–362. 72 On this act of disobedience, see the account by Ferdinand of Habsburg’s ambassador to Petru Rareş’ court, Călători străini, vol. 1, 377. romanian participation in ottoman campaigns 273 interruption), failed to appear with the 8,000 to 10,000 cavalry required of him, sending instead one of his boyars with a few hundred men.73 In other situations, rather than assigning an active role to troops, the Ottomans preferred to give their vassals a defensive role, especially when the Ottoman Empire was in the field against its enemies in Asia. One example of this is offered by an English document of 1585, which mentions that the garrison troops from the Danube border had been sent to join the armies on campaign against the Safavids. The Porte then asked Transylvania, Moldavia, and Wallachia to take over the role of these garrisons on the empire’s frontier: [the Sultan] hath sent [to Persia] all the people in pay, in so much as upon the borders of Transylvania, Moldavia and Wallachia there now remeineth no garrison as accustomed but they themselves are commanded to keep their and the adjoining confines against the Hungar, Jermayn, Powle and Tartar.74

Instead of a Conclusion

The present study attempts to sketch the outlines of a problem that deserves more attention. Although Ottoman suzerainty over the Roma- nian principalities is not in doubt, the participation of these two countries in Ottoman campaigns—an essential aspect of the ahdname capitulations that the sultans granted to the Romanian princes—still needs considerable and detailed research. There is evidence for such agreements from the first decades of the fifteenth century in the case of Wallachia and from the second half of the same century for Moldavia, but they only became routine in the following century, lasting until the end of the seventeenth century. The troops fielded seem to have varied depending on total military capacity, the Porte’s strategic needs, and perhaps the troops’ effectiveness, which fell during the course of the “military revolution.” Wallachian and Moldavian troops tended to be used in neighboring theaters of war, with the exception of 1473, when Radu the Handsome took part in the battle of Başkent. Most often the troops joined the sultan’s main army, but there were also instances—especially during the reign of Süleyman the

73 Hurmuzaki, Suppl. I/1, 229. The broad context is analyzed by Tahsin Gemil, Ţările Române în contextul politic internaţional, 1621–1672 [The Romanian principalities within the international political context, 1621–1672] (Bucharest, 1979), 70–71. 74 Eric Dietmar Tappe, Documents Concerning Rumanian History 1427–1601 Collected from British Archives (London, The Hague, and Paris, 1964), 43. 274 ovidiu cristea

Magnificent—when they were entrusted with separate actions. We may count among such cases the order to send troops to support the khan of Crimea when the Tatars invaded Podolia in 1655.75 The Porte seems to have preferred to draw cavalry troops from the two realms, although other categories are also mentioned (oarsmen, corvée labor) from the second half of the sixteenth century onward. It is hard to judge the effectiveness of the Wallachian and Moldavian troops, since the Ottomans frequently questioned their loyalty. In many cases Wallachian troops were called on to serve as the vanguard in campaigns against Tran- sylvania and Moldavia. All this information helps build an ambiguous picture of the two realms. The wish to survive even while surrounded by much stronger neighbors led them to develop a policy of equilibrium, where obeying the Porte’s orders was balanced by a simultaneous effort to seek the goodwill of the Christian powers. This policy is admirably summed up in a conversation between the grand vizier and the chronicler Miron Costin. Asked whether the Moldavians were pleased that the sultan had conquered Kamieniec Podolski, the boyar answered, “We Moldavians are happy to see him spread out all over, though we are not glad to see him spread in our country.”76

75 Documente privitoare la Istoria Românilor culese de Eudoxiu de Hurmuzaki [Docu- ments concerning the history of the Romanians collected by Eudoxiu de Hurmuzaki], vol. 5, part 1, 1650–1699, ed. Ioan Slavici (Bucharest, 1885), 29. 76 The response is recorded by the chronicler Ion Neculce, Opere: Letopiseţul Ţării Moldovei şi O samă de cuvinte [Works: The chronicle of Moldavia and a collection of tales], ed. Gabriel Ştrempel (Bucharest, 1982), 221. The Military Co-operation of the Crimean Khanate with the Ottoman Empire in the Sixteenth and Seventeenth Centuries*

Mária Ivanics

Among the European vassals of the Ottoman Empire there was only one Muslim state: the Khanate of Crimea. The international literature generally takes 1475, the conquest of the Genoese colonies on the southern coast of the Crimean Peninsula, as the year when its vassal status was established. However, recently a growing number of historians have expressed the opinion that the khanate should be seen as an ally and not as a vassal until the mid-seventeenth century, as it was only in the early eighteenth century that the state’s political scope became so narrow that it depended unilaterally on the Sublime Porte.1 The contacts between these two Muslim states were characterized by common religious ground and a centuries-long mutual geopolitical dependence. Until the 1570s, the political interests of the Crimean Tatars were linked to the steppe, but the emerg- ing power of Muscovy pushed them out of the sphere of steppe politics. The Crimean Tatars had no other direction left than to establish a closer cooperation with the Ottoman Empire and acknowledge the interests of the Sublime Porte. The military potential of the Crimean Khanate, freed from its earlier involvements, fortunately met the needs of the Ottoman Empire, which stood at the height of its power and was enjoying a period of expansion on both land and sea. Apart from military cooperation, the

* The research for this paper has been carried out with the support of the Research Group of Turcology of Hungarian Academy of Science at the University of Szeged. 1 On the history of the Crimean Khanate, see V.D. Smirnov, Krymskoe Khanstvo pod ver- hovenstvom Otomanskoi Porty [The Crimean Khanate under Ottoman sovereignty], vol. 1 (Sanktpeterburg, 1887); and vol. 2 (Odessa, 1889), both reprinted Moscow, 2005; Alexander Benningsen, et al. Le Khanat de Crimée dans les Archives du Musée du Palais de Topkapı (Paris and The Hague, 1968); Alan W. Fisher, The Crimean Tatars, Studies of Nationalities in the USSR, ed. Wayne S. Vucinich (Stanford, 1978); Alan W. Fisher, “Crimean Separatism in the Ottoman Empire,” in Nationalism in a Non-National State: The Dissolution of the Ottoman Empire, ed. William W. Haddad and William Ochsenwald (Columbus, OH, 1977), 57–76; Halil İnalcık, “The Khan and the Tribal Aristocracy: The Crimean Khanate under Sahib Giray I,” Harvard Ukrainian Studies 3–4 (1979–80), 445–466. 276 mária ivanics slave trade also played an important part in this process; without the galley slaves provided by the Crimean Tatars, the Ottoman fleet would not have been capable of maritime activities. The geographic proximity of the khanate could also be easily used to keep the Ottoman Empire’s Christian vassals (Transylvania, Moldavia, and Wallachia) at bay.

The Legal Bases of the Cooperation between the Ottomans and Tatars

It was the renowned Turkish historian of Crimean origin, Halil İnalcık, who addressed the problem of the establishment of Ottoman protection over the Crimea for the first time in 1944, along with the question of whether there was a treaty (‘ahdname) between these two Muslim powers.2 The fact that, although it has been touched upon in several works, no one has taken up the task of writing a new analysis of the problem tells us a great deal about the complexity of the question.3 In his long essay, İnalcık consulted the relevant narrative and documentary sources and concluded that neither the Turkish nor the Tatar sources suggest that a document was drafted after the occupation of the Genoese colonies in 1475 to formalize the relationship of vassalage between the Tatars and Ottomans. The single contemporary reference to a treaty can be found in a letter sent by Khan Mengli Giray to the Porte on 15 July 1475: “We concluded an agreement with Ahmed Pasha [that is, Gedik Ahmed Pasha, the commander of the Ottoman fleet that was besieging Kaffa] that we shall be a friend of the friend of the Padishah, as well as the enemy of His enemy.”4 Evliya Çelebi—referring to an unknown Eastern Turkish (Chagatai) chronicle, the tarih of Tohta Bey—also reports that Mengli Giray concluded an agreement with Sultan Bayezid [sic],5 according to

2 Halil İnalcık, “Yeni vesikalara göre Kırım Hanlığının Osmanlı tâbiliğine girmesi ve ahidname meselesi” [The establishment of Ottoman sovereignty over the Crimean Khanate and the problem of the treaty according to new documents], Belleten 8 (1944): 185–229. 3 Alexander Bennigsen and Chantal Lemercier-Quelquejay, “Le Khanat de Crimée au début du XVIe siècle: De la Tradition Mongole à la Suzeraineté Ottomane d’après un document inédit des Archives Ottomanes,” Cahier du Monde Russe et Sovietique 13, no. 3 (1972): 321–337; Iliia V. Zaitsev, Krymskaia istoriograficheskaia traditsiia XV–XIX vekov [Crimean historiographic tradition, fifteenth to nineteenth centuries] (Moscow, 2009), 142–146. 4 “Aḥmed paşa birle ‘ahd ü şart ḳılduḳ kim pādişāhning dostına dost düşmenine düşmen bolġaymız” A. Melek Özyetgin, Altın Ordu, Kırım ve Kazan sahasına ait yarlık ve bitiklerin dil üslûp incelemesi [Analysis of the language and style of decrees and documents related to the Golden Horde, the Crimea, and Kazan] (Ankara, 1996), 118. 5 Sultan Bayezid II ruled between 1481 and 1512, thus he could not have concluded a treaty with Mengli Giray in 1475. the military co-operation of the crimean khanate 277 which the mountains and coastal castles of the Crimean Peninsula would be in Ottoman hands, the flatlands in Tatar hands, and a son of the khan would be staying in Yanbolu as a hostage.6 İnalcık, however, deems Evliya Çelebi’s story unreliable; he suggests that Evliya Çelebi described the situation in the seventeenth century and ascribed it to a date two hundred years earlier.7 The articles of an assumed treaty between the sultan and the khan appear for the first time in a Tatar chronicle from the early eighteenth century, when the khanate needed the full support of the Porte against Russia’s forceful expansion. The earliest occurrence of the treaty appears in the chronicle of Mehmed Giray, dated to 1704. The articles include assurances that they will not kill or cause damage to each other; they will help each other in every way, lead a war of faith (gaza) against the unfaithful and take revenge; and that from that time on the khans will be appointed by the sultan and the sultan’s name will come first in the hutbe (the Friday prayer), preceding that of the khan.8

6 Evliya Çelebi b. Derviş Mehemmed Zılli, Seyahatnâmesi, vol. 8, ed. Seyit Ali Kahraman, Yücel Dağlı, and Robert Dankoff (Istanbul, 2003), 20. In his work on Crimean historiography, Zaitsev concluded from the distortion of personal names in the parts of the Seyahat- name [Book of travels] referring to Tohta Bey that Evliya Çelebi’s source must have been a chronicle in Arabic, probably a lost work of Ibn Arabshah (Zaitsev, Krymskaia istoriograficheskaia traditsiia, 224). As Ibn Arabshah died in 1450, it is more probable that the part about the agreement comes rather from the Arabic universal chronicle of Mustafa Cenabi (Jenābī, died 1590/91, see www.ottomanhistorians.com, consulted 19 December 2010). 7 İnalcık, “Yeni vesikalara.” 8 “Hilaf-i şer‘-i mutahhire biribirimize fırsat buldugımız takdirce katl ve ziyan eylemeye- lim; her umurda mu‘avenet üzere olup, imdad-i din küffar-i bedayin üzerine gazalar eyleyüp, intikam alalım; ve ba‘de’l-yevm silsile-i tahirleriñ benden bey‘at eyleyüp, ibtidaen benim ismim hutbesinde yad ve ba‘de han olan zi-şanı zikr ede” (Tarih-i Mehmed Giray Vienna, Österreichische Nationalbibliothek, nr. H. O. 86, 103b–104a). A longer version (of seven points) survived in an anonymous chronicle finished in 1782, which was also published in a French translation: M. M. Kazimirski – A. Jaubert, “Précis de l’histoire des Chans de Crimée, depuis l’an 880 jusqu’à l’an 1198 de l’hérige, traduit du turc par M. Kazimirski,” Journal Asiatique 12 (1833): 349–380, 428–458. This, in turn, was the source of Joseph Ham- mer von Purgstall: “Mengli erkannte die Oberherrschaft des Sultans unter folgenden Bed- ingungen an: dass der Sultan nur einen Prinzen aus dem Geblüte Dschengis-Chan’s mit der Herrschaft der Krim belehnen könne; dass der Sultan nie und aus keinem Grunde einen Abkömmling des Hauses Dschengis-Chan’s zu tödten berechtigt sei; dass die Län- der des Chans, oder der Prinzen vom Geblüte Dschengis-Chan’s eine unverletzliche Fre- istätte seien; dass am Freitage in der feierlichen Bühnenrede (Chutbe) nach dem Namen des Sultans der des Chans genennet werde; dass eine schriftliche Bitte des Chans nie eine abschlägige Antwort erhalte; dass der Chan fünf Rossschweife führen dürfe; dass zu jedem Feldzuge der Chan hundert zwanzig Beutel für den Unterhalt seiner Leibwache und achtzig Beutel für seine Mirsen und Kapikuli (Pfortendiener) erhalte.” Geschichte der Chane der Krim unter osmanischer Herrschaft: Aus türkischen Quellen zusammengetragen mit der Zugabe eines Gasels Schahingerai’s (Vienna, 1856), 35. İnalcık writes about “a pouch 278 mária ivanics

İnalcık proved that the points of this treaty were later compilations. If there had been any agreement after 1475, according to him, it would have treated the legal status of Kaffa and other seaports conquered by the Ottomans, cooperation against enemies, and the subtleties of the relationship of vassalage created between the two powers. The idea that a vassal status was established is also questionable because of the fact that the Crimean khans disposed of both signs of the sovereignty of a Muslim ruler, the mentioning of their name in the hutbe (the sultan’s name preceded that of the khan’s only after 1584), and the sikke, that is, the right to mint coinage, which they maintained throughout the entire existence of the khanate. There is also no trace of vassalage in the berats given to the khans appointed by the sultans.9 In contrast to the other European vassals, the khan never paid tribute to the Porte, although sometimes he sent captives to Istanbul. İnalcık concludes that if there were any agreement at all, the Tatars accepted the protection of the sultan and the Porte, as the leading power of Islam, in the summer of 1475.10 The vassal status and requirements for military service only formed gradually. The participation of Tatar auxiliary troops in Ottoman campaigns was based more on the similarities of the two powers’ political and economic interests and religious prescriptions than on legal documents. Remmal Hoca, in his history of Khan Sahib Giray, notes the only known example of a Crimean khan demanding material benefit before a campaign. According to the chronicler, the khan gave the following answer to the sultan’s invitation to the 1547 campaign against Iran: My sultan! The Tatar military is poor [literally: naked], has no means to go to a campaign so far away. But you [alone] can also not bring the Otto- man army to victory. Your army has heavy weaponry, whereas the troops of the kizilbaşis [that is, the Persians] consist of light cavalry, just as those of the Tatars. They can cover in one ride [or: during one day] the distance for which you need five days. But if you dedicate 5,000 akçes per capita from the treasury, I shall choose 10,000 brave warriors from among my soldiers, spend the 5–5,000 akçes to equip them and send them with my son, or take their lead myself. We shall cross the Iron Gate [Derbent] and crossing the

of gold,” however, in the original text, only “pouch” (kese, in Hammer’s text, Beutel) is mentioned, which was also an Ottoman fiscal accounting unit, 1 kese = 500 akçe (piaster). Cf. İnalcık, “Yeni vesikalara,” 226. 9 Feridun Bey, Mecmu‘a-i Münşeat-i Selatin [A collection of the sultan’s letters] (Istan- bul, 12752 [1858–59]), vols. 1–2. The Porte always tried to send a candidate to the Crimea who would be welcomed by the Tatar aristocracy. 10 İnalcık, “Yeni vesikalara,” 228. the military co-operation of the crimean khanate 279

province of Shirvan, I shall join the Ottoman army in the country of the kizilbaş [that is, the shah of Persia]. I shall bring the quick cavalry of his army to my side, and when I receive news where the kizilbaş is to be found, I shall send raiders after him, attack him during the night and make them flee. Wherever they should go, I shall attack them at night, I [will] not even allow them to open their eyes, [but will] crush them and defeat them; and I shall give him [that is, the shah] into your hands. If you do otherwise, you shall spend your money from your treasury in vain and will suffer many hardships. You are the world’s master, [that is: You know it better].11 In the end, Sahib Giray did not participate in the campaign in Iran in 1574, but rather took the field against the Circassians with the Tatar army. According to İnalcık, this was one of the main causes of the rift between the khan and the Ottomans.12

The Phases of the Ottoman–Tatar Military Cooperation

The first Tatar auxiliary troops ever sent by a khan joined the Moldavian campaign of Sultan Mehmed II in 1476. The goal of this campaign was to conquer Akkerman and Kilia, the two main ports that secured the north–south trade route, to repulse Polish influence from the region, and to bring the western part of the Black Sea basin under Ottoman control. The Crimean auxiliary troops could not have been large in number, as they were led not by the khan, but by the head of the Tatar aristocracy, Eminek Bey from the Şirin clan. Ultimately, no common military action took place because the khan of the Great Horde invaded the Crimea and the Tatars returned home after suffering huge casualties.13 It was only in 1484 that actual cooperation between Sultan Bayezid II and Khan Mengli Giray took place, when they managed to conquer the two fortresses mentioned above. During the sixteenth and seventeenth centuries, the Crimean Tatars participated in every important war in Central and Eastern Europe, and also in Western Asia. The sixteenth century was characterized by the struggle between the Habsburgs and Ottomans over control of the King- dom of Hungary and also between the Safavids and the Ottomans over the

11 Tārīh-i Ṣāḥib Giray Ḫān: Histoire de Sahip Giray, Khan de Crimée de 1532 à 1551, ed. and trans. Özalp Gökbilgin (Ankara, 1973), 114. 12 İnalcık, “The Khan and the Tribal Aristocracy,” 450–451. 13 The apologetic letter of Eminek Bey to the sultan was published recently by Özyetgin, Altın Ordu, 120–122. 280 mária ivanics rule of Azerbaijan. At the end of the century, both conflicts culminated in long wars. The Tatars first crossed the Hungarian border in 1521 as auxiliary troops of the Ottoman army and their last major attack took place in 1718, against Transylvania. During these two centuries there were intervals between campaigns, sometimes as long as half a century, but at the end of the sixteenth and into the seventeenth century the Tatars appeared in the Hun- garian theater of war each year; sometimes they even spent the winter there.14 In 1683, they shared in the Ottoman defeat at Vienna and they suffered heavy casualties in the subsequent anti-Ottoman wars in Hungary. In the Persian theater of war, during the course of the struggle over the east Anatolian territories and Azerbaijan, Sultan Selim I won great victories in 1514 at Chaldiran and Tebriz against Shah Ismail of Persia (1501– 1524). His son, Süleyman the Magnificent, consolidated Ottoman rule over these lands. Between 1578 and 1590, the armies engaged in fighting again to divide the Caucasus into Ottoman and Persian spheres of influence. Tatar auxiliary troops were regularly ordered to appear in the war zone and sometimes even spent the winter there. In 1603, war broke out again and continued, with short intervals of calm, until 1639.15 In the Eastern European theater of war, Ottoman military leaders used the Tatars regularly from 1538 on. After the Moldavian campaign, the first major offensive, the failed siege of Khotyn, began in 1621, followed by a series of others. The greatest victory was celebrated in 1672, when Podolia was incorporated into the Ottoman Empire as a vilayet, and part of the Cossack territories also came under Ottoman influence. They had to cede them, however, in the peace of Constantinople (1700).16 Although the Otto- mans had defeated the armies of Peter the Great at the Prut River in 1711, the Russian advance toward the Black Sea could no longer be stopped.

14 Mária Ivanics, A Krími Kánság a tizenöt éves háborúban [The Crimean Khanate in the fifteen-year war 1595–1606] (Budapest, 1994). For an English summary of the book, see Mária Ivanics, “The Role of the Crimean Tatars in the Habsburg-Ottoman War (1595– 1606),” in The Great Ottoman Turkish Civilisation, vol. 1, ed. Güler Eren, Ercüment Kuran, Nejat Göyünç, İlber Ortaylı, and Kemal Çiçek (Ankara, 2000), 302–310. See also, Jan Paul Nieder- korn, Die europäischen Mächte und der “Lange Türkenkrieg” Kaisers Rudolfs II. (1593–1606) (Vienna, 1993). 15 Bekir Kütükoğlu, Osmanlı-İran Siyâsî Münâsebetleri [Political contacts between the Ottomans and Iran] (Istanbul, 1993). 16 Dariusz Kołodziejczyk, Podole pod panowaniem tureckim: Ejalet Kamieniecki 1672–1699 [Podolia under Ottoman rule: The eyalet of Kamieniec] (Warsaw, 1994); idem, Ottoman– Polish Diplomatic Relations (15th–18th Century): An Annotated Edition of ‘Ahdnames and Other Documents, The Ottoman Empire and Its Heritage, no. 18 (Leiden, 2000). the military co-operation of the crimean khanate 281

The Tatar Army

In the second half of the sixteenth century the Tatar army consisted of three major parts: the troops of the khan and the ruling family; the cohorts of the Crimean Tatar chieftains; and the units of the Nogays. Two units were placed closest to the khan: the sekbans and the tüfenkcis. The sekbans were Ottomans, members of the guard sent by the Porte; originally a separate unit of mostly infantry, they were later incorporated into the janissary army. Apart from their military service, they also carried out administrative and diplomatic tasks; the approximately 1,500-man-strong unit received its salary from the Sublime Porte.17 The tüfenkcis (muske- teers) originally came from among the Tatars and Circassians and were in the direct service of the khan. Their numbers increased to 500 in the late sixteenth century. The backbone of the Tatar army was made up of the 10,000-man-strong light cavalry coming from the ranks of the four most important ruling tribes, twice as many in case the khan went to war in person.18 There is no reliable data on the size of the Nogay troops, but as this group lived mostly on their booty, their numbers must have been relatively large. Apart from the soldiers left behind to guard the Crimea, usually the entire Tatar army participated in the Eastern European campaigns. Accord- ing to Evliya Çelebi, 80,000 soldiers left the Crimea when headed by the khan, 50,000 with the kalga, and 40,000 with the nureddin; however, these data are not supported by any other source.19 Based on Russian sources,

17 Halim Giray, Gülbünü Hanan yahud Kırım Tarihi / Rozovyi kust khanov ili istoria Kryma [The rose-bush of the Crimean khans, or The history of Crimea], ed. Ablakim Ilmi (Simferopol, 2004), 38–39; Seyyid Muhammed Riza, Asseb o-sseyyar ili sem’ planet soderzhavshchii istoriiu krymskikh khanov ot Mengli Girej Khana piervogo do Mengli Girej Khana vtorogo t. e. s 871/1466 po 1150/1737 [Asseb o-sseyyar or The seven planets, covering the history of the Crimean khans from Khan Mengli Giray I to Khan Mengli Giray II, i.e., from 871/1466 to 1150/1737], ed. Mirza Kazembeg (Kazan, 1832), 91. 18 In 1547, the head of the Şirin clan, Baba Bey, joined the campaign of Khan Sahib Giray I against the Circassians with 5,000 men under his own banner, together with his military orchestra. The Barın, Argın and Kıpçak clans altogether mustered the same number of soldiers. The khan participated in the campaign with a cavalry of similar size and 300 janissaries. Tārīh-i Ṣāḥib Giray Ḫān, 73. 19 Evliya Çelebi b. Derviş Mehemmed Zılli, Seyahatnâmesi, vol. 7, ed. Dağlı Yücel, Seyit Ali Kahraman, and Robert Dankoff (Istanbul, 2003), 204. Following the old Mongol custom, the khan shared his power with several members of his dynasty. After his election, the khan appointed the kalga and the nureddin from among his brothers or cousins (sometimes appointing one of his sons). The kalga was the leader of the army’s right wing, and if the khan was not present in person, he took charge of the entire army. The existence of the office is documented from the early sixteenth century. The nureddin was the leader 282 mária ivanics the military historian V.V. Penskoi estimates the Tatar military potential at 30–40,000 in the sixteenth century and 40–60,000 in the seventeenth century;20 however, the khans could only have taken part of this force to the campaigns, as they also had to secure the defense of the Crimea. The numbers were significantly lower if the Tatars participated in a campaign as auxiliary troops of the Ottoman army; the size of the cohort did not depend on the wish of the center in Istanbul, rather on the person who led the Tatar troops. If the khan participated in person, the army was larger and better equipped; cases are mostly known from the long wars at the end of the sixteenth and seventeenth centuries, but even in these periods it often happened that—with a reference to the defense of Crimea—the khan did not join the campaign and sent his troops led by the kalga to the Hungarian or Caucasian theater of war. The khan was invited to the campaigns by imperial letters (name-i hüma yun) or noble orders (hükm-i şerif) issued by the sultan’s divan. The documents were delivered together with swords inlaid with gems and valuable kaftans for the khan and the highest aristocrats (mirzas) participating in the campaign.21 The khan was also granted a sum, which varied, for his

of the left wing. His office is known in the sources only after 1584; its creation must have taken place in connection to the settlement of the Nogays in the Crimea. Following the chronicle of Hezarfen, the kalga is known in the literature as the first, the nureddin as the second in the line of succession to the throne, cf. Hezarfen Hüseyin Efendi, Telhîsü’l- Beyán fî Kavânîn-i Âl-i Osmân [Explanation of the Ottoman dynasty’s essence], ed. Selim İlgürel (Ankara, 1998), 170. As a matter of fact, these offices only secured a chance for their holders to be seen as competitors for the title of khan. On the origins of the offices, see Josef Matuz, “Qalġa,” Turcica 2 (1970): 101–129; V.V. Trepavlov, “Nuradiny Noganskoi Ordy” [Nureddins of the Nogay Horde], in Istoriko-geograficheskie aspekty razvitiia Nogaiskoi Ordy, ed. S.G. Agadzhanov et al. (Mahachkala 1993), 43–61. 20 V.V. Penskoi, “Voennyi potentsial Krymskogo Khanstva v konce XV–nachale XVII v.” [The military potential of the Crimean khanate from the end of the fifteenth to the beginning of the seventeenth century], Vostok (Oriens) [20], no. 2 (2010): 56–66. In the second half of the seventeenth century one has to take into account that Tatar armies frequently undertook joint campaigns with Ottoman and Cossack troops, therefore their larger numbers did not necessarily mirror growth in the potential of the Crimea, cf. Evliya Çelebi b. Derviş Mehemmed Zılli, Seyahatnâmesi, vol. 5, ed. Ali Kahraman and Yücel Dağli (Istan- bul, 2001), 72–77. See also my “Melek Ahmed pasa 1657. évi lengyelországi hadjárata II. Rákóczi György ellen Evlia Cselebi útleírásában” [The 1657 campaign of Melek Ahmed Pasha to Poland against György Rákóczi II in the travelogue of Evliya Çelebi], in Szeren csének elegyes forgása: II. Rákóczi György és kora, ed. Gábor Kármán and András Péter Szabó (Budapest, 2009), 171–203. 21 The word mirza derives from the Ar. amīr and Per. zāde “child of the emir, nobleman.” In the Tatar khanates of Kasan, Astrahan, and Crimea—as in Russia—it meant “a member of the highest aristocracy.” the military co-operation of the crimean khanate 283 military expenses.22 In the registry of the sultan’s council (Mühimme defter leri), a draft of the letter inviting Khan Gazi Giray II to war survived: You should send an army of the size you used to send before to my deceased predecessor, Sultan Süleyman, selecting more than 50–60,000 soldiers from among the enemy-hunting (tatar-i ‘aduvv şikar) and windstorm-quick Tatars (tatar-i saba reftar), led, if possible, by yourself, or by your brother, Kalga Feth Giray, against the country of the infidels.23 Scholars consider this data about 50–60,000 soldiers as reliable because of the reference to the “usual size of the army sent in the time of Sultan Süleyman”;24 nevertheless, this should rather be seen as a literary topos. We have no evidence that the Ottoman military leadership would ever have entered into negotiations with the Tatars about the size of the auxiliary troops and the numbers requested are more likely to mirror the size of the cohorts wished for rather than those expected. This number also grew considerably later; in 1596, for instance, an army of 70–80,000 was requested for the Hungarian theater of war.25 The reports of Turkish chronicles, as well as contemporary European pamphlets and newsletters about Tatar armies numbering 80–100,000 soldiers, are taken to be overstatements. Somewhat unique among his contemporaries with his critical attitude, İbrahim Peçevi, the chronicler of the late sixteenth-century wars, delivered some money in 1603 from the Ottoman military headquarters to Khan Gazi Giray, who had been spending the winter in Hungary. The khan claimed that he entered Hun- gary with 50–60,000 Tatars, on which the chronicler commented that they might have been 50–60,000, but it is much more likely that there were only 30–40,000.26 On the basis of reliable archival sources, it can be said that the Otto- man army in the Hungarian theater of war was complemented with a

22 The Ottoman chronicler Selaniki, based on the accounts in defters at the Porte, states that Khan Gazi Giray II and his brother, Kalga Feth Giray, were sent 40,000 gold coins and 129 honorary robes between 1593 and May of 1596. In early May 1596, a further 30,000 gold coins, a sword with gilded grip, and honorary kaftans—two from the private treasury of the sultan and four from the state treasury—were granted to the khan. See Selânikî Mus- tafa Efendi, Tarih-i Selânikî I. (971–1003 / 1563–1595) II. (1003–1008 / 1595–1600) [The history of Selânikî], ed. Mehmet İpşirli (Istanbul, 1989), 593. 23 Başbakanlık Osmanlı Arşivi, Mühimme Defterleri 71, Nr. 571 [henceforth BOA MD]. 24 See, for instance, Caroline Finkel, The Administration of Warfare: The Ottoman Mili- tary Campaigns in Hungary, 1593–1606, Beihefte zur Wiener Zeitschrift für die Kunde des Morgenlandes, no. 14 (Vienna, 1988), 100. 25 BOA MD 73, No. 1146. 26 İbrahim Peçevi, Tarih-i Peçevi [Peçevi’s history] (Istanbul, 1283/1866), vol. 2, 269. 284 mária ivanics

Tatar contingent of 22–25,000 soldiers if it was under the command of the khan—under the command of the kalga, their number was only half of this.27 In the case of the Persian campaigns, Turkish sources mention auxiliary troops of the same size. For instance, the Tatar contingent that joined the Ottoman army’s Persian campaign in 1578 under the command of Kalga Adil Giray was estimated to be 10,000, 15,000 or 30,000 men strong (the last number is surely an exaggeration). The Porte sent a further 500 janissaries, fifteen artillerymen, and twelve gun-carriage drivers to Adil Giray.28 When heading to the Hungarian theater of war, the Tatar troops usually rode through Wallachia and met the Ottoman army near Belgrade, from where they continued their journey along the Danube. If the Tatars arrived through Transylvania, they crossed the Tisza at Szolnok and made camp on the lowlands close to Pest. A marching army always causes damage to the local population and as the Tatars could only reach Hungary through tributary states of the Ottoman Empire, the khans—in order to avoid later blame—always asked for an order from the Porte when choosing a route. In 1600, Khan Gazi Giray shared the following possibilities with his envoy to be sent to the Sublime Porte: “If I went to Dobruja, the province would not recover for ten years afterwards. Wallachia and Tran- sylvania would also be laid ruin, and would be uninhabitable for years. Hurry back with the permission, which route would be the right one.” In Istanbul, he received the answer that he could choose any way he liked, but he had to be in Transylvania at the side of the serdar as soon as possible. The khan then chose the Dobruja road.29 No such deliberations were necessary concerning the route to the Caucasus, as there was only one: the Tatars turned south at the Azov fortress at the mouth of the Don and crossed the mountains through Derbent in Dagestan.30 Supplying food for the Tatar armies until they reached the theater of war was the responsibility of the Christian vassals. In the second half of the seventeenth century the princes of Transylvania appointed a com- missioner who stayed permanently with the Tatar troops and supplied

27 See my results on the basis of Habsburg, Hungarian, Turkish, and Tatar sources: Iva nics, A Krími Kánság, 176–179. 28 Kütükoğlu, Osmanlı-İran, 86–87, nn. 22, 25, 26. 29 “(. . .) meşveret buyurup yol ta‘yîn buyursunlar. Dobrica vilâyetine girersem ol vilâyet on yıla dek yerine gelmez. Ve Eflâk ve Erdel vilâyetleri kezâlik harâb ve yebâb olup, nice yıllar âbâdân olamaz. Bana toğrı yol kangısı ise izn ü icâzet-i hümâyûnları ile gele yetişesin.” Selânikî, Tarih-i Selânikî, 748, 752. 30 Kütükoğlu, Osmanlı-İran, 86–88. the military co-operation of the crimean khanate 285 them with provisions that awaited them at previously defined camp sites. In this way they could prevent the plundering of territories close to the deployment area.31 Supposedly, the voievods of Wallachia and Moldavia also used similar practices. These rulers also had to contribute during the Persian campaigns. According to the sultan’s order, Moldavia had to deliver, from the harbor of Brăila to Gözleve, 10,000 kiles each of wheat, barley, and flour before the 987/1578 campaign and Wallachia had to bring 10,000 kiles of flour, barley, and millet to Kaffa.32 In enemy territories, the Tatars took care of their own provisions through plundering. In the Hungarian theater of war, after a successful raid, they could even sell their surplus to the Ottoman army. The literature has as yet dedicated meager attention to the role of the Tatars in the logistics of the Ottoman army and border fortresses during their overwin- tering, despite the fact that it would have been hard to solve the problem of transporting and collecting victuals in the border fortresses without the Tatars.33

31 Princes of Transylvania tried regularly—often without success—to convince the leader of the Tatar army to choose the route along the Lower Danube for deployment instead of the one through the principality. In 1621, Gábor Bethlen offered 1,000 gold coins to the high mirza through the Tatar envoy sent to him in order to save Transylvania from damage, see Áron Szilády and Sándor Szilágyi, eds., Török-magyarkori állam-okmánytár [State documents from the Turkish-Hungarian age], vol. 1 (Pest, 1868) [henceforth, TMÁO I], 338. The Ottoman military command, however, had already defined the route of deployment for the Tatars, which went in most cases along the Tatros River to Transylvania, then through Brassó, Nagyszeben, and Jenő until it reached Szolnok and the plains around Pest. The Ottoman army, in its turn, marched on the right bank of the Danube. See the letter of Murad Giray to Mihály Apafi, 10–20 Rebiülevvel 1094 (9–19 March 1683). Original in the Magyar Országos Levéltár (Hungarian State Archives; henceforth MOL), Cista Diplomatica F 126, no. 247. See also its nineteenth-century Hungarian translation in Áron Szilády and Sándor Szilágyi, eds., Török-magyarkori állam-okmánytár [State documents from the Turk- ish-Hungarian age], vol. 6 (Pest, 1871), 319–321. See also the letter of Ahmed Ağa, the vizier of the khan to the locum tenens of Mihály Apafi, sine dato, ibid., 233–234. 32 “Bogdan’dan ihracı emr edilen onar bin kile buğday, arpa ve unun İbrail’den Gözleve iskelesine gönderilip Kırım Hanı’na teslimi hakkında” BOA MD 37/5 (Safer 987/1578), quoted by Kütükoğlu, Osmanlı-İran, 34n5; 36n6. The kile was used for measuring the volume of grain. In sixteenth-century Istanbul, a kile of wheat weighed 25.65 kg, whereas a kile of barley weighed 22.25 kg. The Ottoman army conquered Kaffa in 1475 together with the other Genoese colonies in the Crimea. Gözleve was the only harbor on the peninsula which was left at the hands of the Tatars by Sultan Süleyman at the request of Khan Sahib Giray. This proves that the Moldavian delivery was destined for the provision of Tatar troops. The victuals for the Ottoman army were transported by sea to Trabzon or Batum, from where they were taken further on pack-animals toward Erzurum. 33 The Tatars, who could be mobilized rapidly, either escorted the food shipments themselves or, as in 1602, conducted larger campaigns of distraction so that Ottomans could supply Eger with provisions in the meantime. (Erich Lassota to Archduke Matthias, Kassa 20 May 1602. Kriegsarchiv [Vienna], Hofkriegsratregistratur [henceforth, KA HKR] 1602 286 mária ivanics

During the campaigns, the Tatars operated under the command of their own leaders as separate units—under the khan or, more often, the non-official heir to the throne, the second man of the khanate, the kalga sultan. As a rule, the serdar of the campaign met the head of the Tatar troops personally only when the army arrived; after these negotiations, all communication took place through letters by intermediaries. Such was Peçevi’s visit to the khan; but we also have another piece of data from the personal correspondence of Khan Gazi Giray II. Narrating the siege of Győr, he wrote the following to Hüseyin Efendi, his trustee in Istanbul: His Highness the serdar [Grand Vizier Sinan Pasha] suggested that it would be useful to send 5–10,000 Tatars across the river [Danube]. We answered that we could transfer 2,000 janissaries, and were ready to offer our lives for our faith in the way seen best by His Highness; but the Tatars would only with difficulty withstand guns and cannons. So we delayed [getting across the river] that day. The next day, the serdar sent a tezkere in his own hand with the defterdar efendi. This made it clear to us that he interpreted our words as negligence. We saw that if we delayed [crossing the river] any more, we should surely affront him. Thus, we called the Tatars together and set out to make rafts.34 During the Érsekújvár campaign of 1663 the army was led by the khan’s son, Ahmed Giray. On this occasion, the letters between the Ottoman troops besieging the castle and the khan’s son, who camped at Beszterce- bánya, 150 kilometers away[!], were carried by a certain Paul the Cossack, a spy in the khan’s service.35

Expedit No. 5). The Tatars, having taken up their winter quarters in and around Pécs in 1602–1603, also supplied Kanizsa. In news from early 1603, it was reported that the soldiers in Kanizsa were starving and that a total of 50 janissaries and three wagons of food had been sent there (Siegfried Kollononits to Archduke Matthias KA HKR 1603 Expedit Nr. 69). During the Tatar raids in Međimurje, they plundered livestock and captives in such abundance that they were unable to sell all of them in Kanizsa and brought them to Buda. Besides, in June, a further 120 wagons of food were taken to Kanizsa by 2,000 Tatars and 1,500 Turks (KA HKR 1603 Expedit Nr. 164). 34 Nuru Osmaniye Kütüphanesi Nr. 4292, 45–54. The letter, with some minor changes, was also included in the chronicle of Selânikî, see Tarih-i Selânikî, 399–404. 35 Mária Ivanics, “Krimtatarische Spionage im osmanisch-habsburgischen Grenzgebiet während des Feldzuges im Jahre 1663,” Acta Orientalia Academiae Scientiarum Hungaricae 61, nos. 1–2 (2008): 119–133. the military co-operation of the crimean khanate 287

Characteristics of Tatar Warfare and the Military Value of the Troops

Although the Crimean Khanate was a successor state of the Golden Horde, it did not follow its predecessor’s military traditions. The Golden Horde had heavy cavalry; on the miniatures depicting its warriors one can clearly see their armor, made with various techniques, and also that of their horses.36 The heavy cavalry came from the ranks of the nökers, who—leaving the bonds of their clans behind—took up the service of the khan of the Golden Horde and constituted his personal entourage. They received their equipment from the khan’s armory. Apart from them, it was the members of the ruling dynasty and the highest-ranking heads of the clans who had at least chainmail. War stallions (argamak) were specifically bred for the heavy cavalry. Most of the army, however, consisted of light cavalry equipped with bows, arrows, and sabers. With the dissolving of the Golden Horde, heavy cavalry disappeared from among the military units of the successor states. Penskoi attributes this change to two factors. First, when Timur Lenk led two campaigns against the Golden Horde at the end of the fourteenth century he reset- tled the artisans from the conquered cities on his lands. Also, the direct contacts between the steppe and the Central Asian trade centers were cut; thus, heavy armor was no longer available in the necessary quantities. The spread of firearms also contributed to the disappearance of the heavy cavalry. Successor states could not compete with the Polish and Russian states, which had more developed industry and better military technology.37 The army of the khans had an artillery force of some smaller guns at its disposal, which they received from the Sublime Porte, but the Ottoman military command was careful to prevent the Tatars from having significant firepower, in order to prevent them from attacking the coastal towns that were under Ottoman control.38 In any case, the Tatars had problems even with this smaller artillery; they were only able to use it effectively in

36 L.A. Bobrov and Iu. S. Hudiakov, “Zashchitnoe vooruzhenie sredneaziatskogo voina perioda pozdnego srednevekov’ia” [Defence equipment of the Central Asian warrior in the late Middle Ages], in Voennoe delo nomadov Severnoi i Tsetral’noi Azii (Novosibirsk, 2002). 37 Penskoi, “Voennii potentsial,” 60. 38 The siege of Kaffa was attempted by two khans revolting against the Ottoman sultan, Mehmed Giray II (1577–1584) in 1584 and Mehmed Giray III (1623–1628, with an interruption) in 1624, both times without or at least with little success. 288 mária ivanics the mountains, in the campaigns against the Circassians, and combined with the strategy of encampments made up by carts.39 From the end of the sixteenth century on, the Tatars concentrated exclusively on using light cavalry and perfected its tactics. The effectiveness of their equipment did not lag behind hand firearms; their bows allowed them to shoot farther than the Cossacks’ guns and reloading was also much quicker. Their speed, training, discipline, and cunning proved to be their best weapon against Western technical superiority, as Penskoi puts it.40 The Tatar troops needed speed and discipline mostly to take and keep booty. Guillaume le Vasseur de Beauplan, a French military engineer in the service of the Polish king, noted that when the Tatar army reached the border they rested for two to three days. They divided the army into three units, which, however, moved together until they were deep into enemy territory. On the way in, they did not cause damage, but when they turned back, two-thirds of the army stayed together and proceeded slowly, while the third, functioning as right and left wings, looted the territory. The wings had to return to the main army at a given time, thus making it possible for the other units to replace them as looters; at the same time, the army was secure and proceeded continuously toward the Crimea with the booty. During the raids, Tatars surrounded the villages along the way and drove the inhabitants and the animals away, except for the swine, which were driven into the stables and burned, together with the building. The Tatar army was constantly on the watch to avoid—or, if that was not possible, defeat—Polish and Cossack sentries, and out of caution always took a different route leaving an area than they had used entering the country. When they were far enough from the border, well into Crimean territory and felt secure, they had a longer rest, collected the booty, and redistrib- uted it among themselves.41 In all likelihood the same order of looting was also followed when Tatar armies spent the winter in Hungary as auxiliary troops of Ottoman armies; according to the reports sent to the Imperial

39 Victor Ostapchuk, “Crimean Tatar Long-Range Campaigns: The View from Remmal Khoja’s History of Sahib Gereby Khan,” in Warfare in Eastern Europe, 1550–1800, ed. Brian Davies (Leiden and Boston, 2011), 147–172. It is characteristic that the Tatar army lost all of its eighteen cannons at the siege of Tula in 1552, see Penskoi, “Voennii potentsial,” 62. 40 Penskoi, “Voennii potentsial,” 61. 41 Guillaume le Vasseur de Beauplan, Opisanie Ukrainy [Description of Ukraine] (Mos- cow, 2004), 230–237. the military co-operation of the crimean khanate 289

War Council, one third of the Tatar camp at Dunaföldvár was constantly on the move raiding neighborhoods.42 Sources are as good as silent about further characteristics of Tatar warfare; however, many of them mention the dreaded shower of arrows. The maneuver, called çarhçılık (whirl), during which the Tatars rode around the enemy camp and maintained a constant shower of arrows, is a typical form of nomadic warfare and was practiced on the common hunts staged each year during which game was surrounded and hunted down by gradually narrowing the circle around them. The goal of the çarhçılık was to threaten the enemy and find the weak points of the camp. During the siege of Győr in 1594, the whirl was commanded by Kutlu Giray Bey and Arslanay Bey, the chieftains of the Şirins and the Mangits (Nogay), that is, the two greatest clans; they successfully forced the camp of Arch- duke Matthias to flee.43 Two years later, when Zsigmond Báthory, prince of Transylvania, laid siege to Temesvár, the Tatars rode around the besiegers’ camp for three hours, but their tactics were unsuccessful this time.44 There are also scattered data in the sources about other military customs of the Tatars. After the campaigns, they counted the human losses of the enemies by cutting marks on the backs of the carcasses. If a dignitary died during the campaign, his body was salted and then transferred to the Crimea.45 After the death of an illustrious person, his relatives, servants, and even horses did not touch any food or drink for two days, and the horses’ tails were cut as a sign of mourning.46 The enemy was not allowed to have the bodies of Tatar soldiers fallen in battle; these were heaped in a mound, enclosed with a palisade of logs, and burned.47 Concerning the military value of the Tatar army, the khan was well aware of the limitations of his troops. After the failed siege of Várad in 1598, Khan Gazi Giray II himself apologized in a letter to Sa‘adeddin Hoca, the former tutor of Sultan Murad III (1574–1595): “Besieging castles is not

42 However, they were not cautious enough, or their numbers were not great enough, because during one night Hungarian soldiers from the nearby border fortress invaded the camp, killed the Tatars, took the booty accumulated there, and set the captives free. KA Alte Feldakten 1597/4/11. 43 Nuru Osmaniye Kütüphanesi No. 4292, 48. Kutlu Giray Bey himself was no Genghisid, he could only use the name of the dynasty because he had married the khan’s daughter. 44 István Szamosközy, Történeti maradványai [Historical legacy], vol. 4, Vegyes föl jegyzések [Diverse notes], ed. Sándor Szilágyi (Budapest, 1880), 58. 45 Both data in Evliya Çelebi, Seyahatnâmesi, vol. 5, 76. 46 Szamosközy, Történeti maradványai, vol. 4, 157. 47 Sándor Takáts, Régi idők, régi emberek [Olden times, olden people] (Budapest, 1922), 224. 290 mária ivanics the task of Tatars; it is rather to ravage, desolate, and ransack the infidels’ country with their raids, and also to collect captives and provisions.”48 Although the Tatars never participated in sieges or major field battles, they had a significant impact on military activities. In 1596, for instance, they changed the outcome of the battle of Mezőkeresztes by attacking the Christian troops, who prematurely thought themselves winners and started to loot the Ottoman camp too early. The Ottoman units, who were already on the run, re-appeared on the trail of the Tatars and their united efforts won the battle for them.49 The failure of the imperial army sent to relieve Kanizsa in 1600—which had the best equipment and the most up-to-date military technology of the entire Long Turkish War of the sixteenth century—was caused by Tatar and Ottoman troops, which captured the caravan of wagons that transported victuals to the Christian army.50 When the banner of the Prophet (sancak-i şerif) was endangered during the siege of Vienna in 1683 it had greater psychological relevance than strategic importance. From the end of the sixteenth century on, the sultans had taken it to major campaigns; and when they no longer led their forces personally the banner was placed beside the tent of the grand vizier. The banner of the Prophet was supposed to secure victory at Vienna, but was threatened by a Polish unit that tried to seize it. Haci Giray Sultan from the Tatar army managed to repulse the Polish attack with 5–600 cavalry and thus saved the Ottoman army laying siege to Vienna from psychological collapse.51 The ranks of the Ottoman border fortresses in the territory of Hungary were completed by the Tatar troops.52 Although the Tatars were much

48 Mustafa Na‘ima, Tarih-i Na‘ima [Na‘ima’s history] (Istanbul, 1734), 112. 49 Ibid., 88. 50 Günter Cerwinka, “Die Eroberung der Festung Kanizsa durch die Türken im Jahre 1600,” in Innerösterreich 1564–1619, ed. Alexander Novotny and Berthold Sutter (Graz, 1967), 434. 51 Tarih-i Mehmed Giray f. 5r. In the Crimean Khanate members of the ruling dynasty of both sexes used the title sultan attached to their names. 52 Numerous data on this can be found in the records of the Vienna War Council, which coordinated the war against the Ottomans, since the first question in interrogating captives always regarded the size of the garrisons in these fortresses. We know from these that in 1596, 50 Tatars were posted in the castle at Hatvan in addition to 120 beşli, 180 sipahi, 300 infantry, 80 gunsmiths, and 40 marauders. In 1597, 30 Tatar cavalry and 50 infantry served in Becskerek, while in the same year 500 Tatars were posted in Buda. In May 1605, the lower castle of Visegrád was guarded by 200 Turks and Tatars. In wintertime, the number of Tatars posted in Turkish border fortifications may have multiplied, especially in the districts designated for wintering. Although the data available are rather limited, they clearly show a tendency on the part of the Ottoman command to use southern Slav the military co-operation of the crimean khanate 291 inferior to the Ottoman military in terms of armor, they were useful, not least due to their effective and wide-ranging espionage network.53

The Economic Aspect of the Campaigns

The campaigns were economic enterprises for the Tatars. The large numbers of Tatar military had to be kept continuously busy. The Sublime Porte paid the khan 5,000 to 15,000 gold coins from the income of the peninsula’s harbors as preparation costs (harclık).54 The khans and the ruling elite were motivated to join the campaign by the money they received from the Porte for their participation. For the simple soldiers, the attraction lay in the booty and the opportunity to acquire new slaves, who were primarily obtained in campaigns against Christian lands. In the Cauca- sian and Persian theaters of war the opportunities were limited, as Islamic law prohibits the enslavement of Muslims. For economic reasons, they were also not supposed to take the inhabitants of other tributary states into slavery, although this eventually could not be avoided. The common

marauders and even Tatars to replace losses at these positions if they could not send enough Ottoman soldiers. 53 Before the Ottoman campaign of 1663, the Crimean khan took into his service several spies who could speak various European languages in order to become better informed about the positions and numbers of the defensive forces in Moravia and Silesia, see Ivanics, “Krimtatarische Spionage” (with an edition of a report from a Tatar spy). 54 In 1593, the divan sent the following instruction to the supervisor of Kaffa harbor: “Macar seferi tedarüki içün Kefe naziri Ebu Bekr çavuş . . . emr-i şerif mucebince Kefe iske- lesinden ta‘yin olunan salyanelerin ve her seferde mu‘ayyen olunan beş bin filori muhase- besin verüb . . .” [For the preparations of the Hungarian campaign, Ebu Bekr çavuş, the supervisor (of the customs) of Kaffa, should, according to the noble order, pay from the income of the harbor the apanage (to which the khan was entitled) and the 5,000 gold coins usual for campaigns], BOA MD 71, no. 320. This sum was probably given to Khan Gazi Giray II in the Hungarian theater of war; at least we have an item of data that in 1594, 5,000 gold coins were given to the khan during his entry of the fortress of Győr, see Peçevi, Tarih-i Peçevi, vol. 2, 150. The same sum was given to Canibek Giray in 1617 during the Persian campaign, see İsmail Hakki Uzunçarşılı, Osmanlı Tarihi (Ankara, 1988), part 3, vol. 2, 10. In 1602, Khan Gazi Giray II demanded 30,000 gold coins for his support of the Ottoman army and for staying in the Hungarian war zone for the winter; he received almost all of this, according to the statements of accounts. See Orhonlu Cengiz, Osmanlı Tarihine Aid Belgeler—Telhisler (1597–1607) [Documents relating to Ottoman history—the telhis] (Istanbul, 1970), 55; Finkel, The Administration of Warfare, 105. In 1671, Khan Selim Giray received 15,000 gold coins in addition to the regular gifts given to him, together with an order to join the Polish campaign; see Uzunçarşılı, Osmanlı Tarihi, 20. See also the data quoted in note 22. 292 mária ivanics slaves were sold to slave traders.55 Sometimes they were also exchanged for Tatar noblemen who had fallen captive to the Christians.56 The Tatars kept their more valuable captives in the fortresses of the Crimea for a long time in order to get a higher ransom. A distinguished captive could also be used to establish political influence; this can be illustrated with the example of Transylvania. Two princes of the country, János Kemény and Mihály Apafi, spent some time in the Crimea as captives, and the diary of the latter shows that he maintained the contact with his former captor, Karaş Bey (who had kept Apafi in the fortress of Or) even after he had been elected to the princely title.57 The Hungarian, German, Russian, and Caucasian slaves were a considerable source of income. The Crimean khan kept one-fifth of the slaves (the kalga and the nureddin kept one-tenth).58 The khan received revenues directly from selling his own slaves, but also indirectly from the annual rent (saliyane) paid from the income of the slave trade in the Crimean Ottoman harbors.59 During the raids, the primary goal of the common Tatars was to collect slaves, but they were also satisfied with cattle or even small livestock; these could be driven home easily after campaigns

55 An account by the Tatar chronicler Remmal Hoca allows us to estimate the number of prisoners owned by the khan; according to him, Khan Sahib Giray had around 3,000 slaves (kul) left after his death (1551), see Tārīh-i Ṣāḥib Giray Ḫān, 138. In the second half of the sixteenth century, a special “handbook” for slave traders was compiled, in which the most important characteristics of the individual peoples were described in order to minimize risk in the business. The Hungarian slaves are described thus: “The Hungarians are clever, sensible, capable of all kinds of handicrafts, and also adept; but also invidious, bad and aggressive. They like killing, wounding and running away. One has to proceed cautiously when one puts them into service. Most of them have a good stature and a white color of skin.” Hans Müller, Die Kunst des Sklavenhandels (Freiburg, 1980), 181–187. 56 İslam Giray, the later khan (1644–1654) was, for instance, released from Polish captivity in exchange for ten distinguished Poles, see Benningsen, Le Khanat, 341. 57 Ernő Tóth, ed., I. Apafi Mihály és II. Apafi Mihály erdélyi fejedelmek naplója [The dia- ries of Mihály Apafi I and Mihály Apafi II, princes of Transylvania] (Kolozsvár, 1900). Both Kemény and Apafi were taken captive in 1657 during the Polish campaign of Prince György Rákóczi II. Kemény was kept at the fortress of Čifut-kale, near Bakhchisaray, until 1659. Apafi returned to Transylvania in 1660. See also Mária Ivanics, “Enslavement, Slave Labour and Treatment of Captives in the Crimean Khanate,” in Ransom Slavery along the Ottoman Borsers (Early Fifteenth–Early Eighteenth Centuries), ed. Géza Dávid and Pál Fodor (Leiden and Boston, 2007), 193–218. 58 Hezarfen Hüseyin Efendi, Telhîsü’l-beyân fî kavânîn-i âl-i Osmân ed. Sevim İlgürel (Ankara, 1998), 170. 59 Detailed saliyane lists are published in A.W. Fisher, “Les rapport entre l’Empire Ottoman et la Crimée: L’aspect financier,” Cahiers du Monde Russe et Soviétique 13 (1972): 368–382. the military co-operation of the crimean khanate 293 in Moldavia or Poland.60 In the Hungarian theater of war, the Tatars regularly sold the booty they collected to the Ottomans. For instance, in spring 1596, the Turks from Győr purchased 600 head of cattle from the Tatars in Buda, which they could safely herd home.61 The Tatars and Turks, following their 1603 joint adventure to Međimurje, divided the large amount of small livestock evenly, while the khan had the grazing stock herded to the meadows of the Požega sancak in Slavonia for later sale.62

Ottoman–Crimean Tatar Contacts during the Campaigns

The Ottomans usually looked down on the Tatars and considered them incapable of completing military missions of high importance. Their armament (sabers, bows, and arrows) was clearly inferior to that of the Turks. Only the highest ranking officers had firearms (usually only pis- tols), together with helmets, three-quarter armor, and vambraces.63 The contacts between the two armies were not without conflict, sometimes murder even took place;64 thus, during the campaigns camps were kept

60 “. . . but if it be easy to conceive the utility of this expedient, a circumstance one would scarcely believe on seeing it, is the care, the attention, the patience, the extreme agility of the Tartars in keeping what they have taken. Five or six slaves, of different ages, sixty sheep, and twenty oxen, the prize of a single man, do not embarrass him. The children, with their heads peeping out of a sack, hanging by the pommel of the saddle; a young girl sitting before, leaning on her left hand; the mother behind on the crupper, the father on a led horse, the son upon another, sheep and oxen before them, every thing in a march, and nothing goes astray from under the vigilant eye of the shepherd of this flock. To collect, to conduct them, to provide for their subsistence, to go on foot to ease his slaves, no trouble is too great for him; and this scene would be truly interesting, did not avarice, and the most cruel injustice furnish the subject of the picture.” François Tott, Memoirs of the Baron de Tott on the Turks and the Tartars, translated from the French, by an English gentleman at Paris, under the immediate inspection of the Baron (Dublin, 1785), 198. 61 Interrogation KA AFA 1596/5/35. 62 KA HKR 1603 Expedit No. 66. 30 April 1603; Pál Fodor, “Tatár marhakereskedelem a tizenöt éves háború idején” [Tatar cattle trade during the Fifteen Years War], Keletkutatás (2011) tavasz, 145–147. 63 “Eyü cebe ve küçük el tüfenki diyarımızda azim makbuldur, buyurdum ki yahşı cebe ve yahşı el tüfenklerinden gönderesiz.” [The good armor and small handguns are much liked in our country, so I order you to send some good armor and small handguns], Letter of Khan Canibek Giray to Gábor Bethlen, arrived at Kolozsvár on 15 September 1629, MOL Cista Diplomatica F 126 no. 5. Gábor Bethlen, apart from other presents, also sent armor to Kalga Devlet Giray, see TMÁO I: 287. 64 In the 1660 Transylvanian campaign there was a skirmish between the janissaries and the Tatars over booty, with casualties on both sides. A detailed account of the case is given by Evliyâ Çelebi b. Derviş Mehemmed Zılli, Seyahatnâmesi, Topkapı Sarayı Kütüphanesi Revan 1457 Numaralı Yazmanın Transkripsiyonu–Dizini, vol. 6, ed. Seyit Ali Kahraman and Yücel Dağlı (Istanbul, 2001), 40–41. 294 mária ivanics separated. By the time of Süleyman the Magnificent, the Tatar khans were still treated according to their rank, because they—as successors of Genghis Khan—were regarded as equals of the Ottoman sultans. However, during the Long Turkish War at the end of the sixteenth century, Grand Vizier Sinan Pasha no longer gave them the respect due to a Genghisid. On his arrival at the Ottoman camp he did not allow a separate tent, which would have been fit for a sovereign, to be raised for Khan Gazi Giray II, rather he entertained the khan in his own tent. Peçevi notes that the members of the Ottoman elite who were at the grand vizier’s camp condemned this treatment.65 At the end of the seventeenth century, the khans played a more important role again, as the Ottoman sultans rarely participated in the campaigns in person. It was during this period that unfounded rumors spread that the Giray dynasty would take over the throne of the Ottoman Empire after the House of Osman had died out.66 The khans were not in the least inferior to the Ottoman sultans in terms of intellectual and cultural pursuits. They had a large library, which burned down in 1736 during the Russian siege of Bahçesaray,67 and they composed literary and historical works.68 The literary activities of the above-mentioned Gazi Giray II can mostly be dated to the period he spent in winter quarters in Hungary. The poetic disposition of Gazi Giray, as well as his interest in the sciences, was also known to his Hungarian contemporaries. The Hungarian historian, István Szamosközi, noted in his chronicle that this Crimean khan had a camel-load of books by his side on every campaign. He also added that Gazi Giray II had a “Latin” education, as he had also mastered, apart from his Tatar mother tongue, Ottoman Turk- ish, Arabic, Persian, and the East Turkic literary language (Chagatai)— that is, every major language of the sophisticated Orient of that time—at

65 Peçevi, Tarih-i Peçevi, vol. 2, 150–151. 66 “Selim-Giray, who reigned at the end of the past, and at the beginning of the present century, after saving by his valor the Turkish army, on the point of sinking under the combined force of the Germans, the Poles and the Muscovites, refused the Ottoman throne, on which the soldiery wished to elevate him; and the Grand Signor, to recompense the bravery and disinterestedness of the deliverer, settled the throne of the Tartars on his descendants, to the prejudices of the other Princes of the Jengiz-Kan family,” Tott, Mémoires, 251. Selim Giray held the office of the khan four times: 1671–1678, 1684–1691, 1692–1699, 1702–1704. 67 Ilia V. Zaitsev, “Ad Fontes: Krymskaia istoriografiia epokhy Srednevekov’ia i Novogo vremeni” [Crimean historiography in the Middle Ages and modernity], in Tiurkologicheskii Sbornik 2003–2004: Tiurkskie narody v drevnosti i srednevekov’e (Moscow, 2005), 48–85. 68 Bursalı Mehmed Tahir, Kırım Müellifleri [Writers of the Crimea] (Istanbul, 1335/1916). the military co-operation of the crimean khanate 295 the same level. From his writings in Chagatai, one epic poem and two philosophical works are still known today. His literary correspondence with the higher officers of the Sublime Porte has also survived. Musical education also played a formative part of this multi-talented personality. Contemporary chronicles narrate that he could play artistically on various instruments and he was also a well-known composer. His poems were taken up in literary chrestomathies and his musical pieces are still being played today.69

The Relevance of the Crimean Tatars in East Central European Politics

The Crimean Tatars were the last people coming from the steppe, who— as heirs of the Golden Horde—enforced the payment of tribute on their neighbors (Russia, Poland, Moldavia, Wallachia) and offered their military force to other states in the region for money or booty.70 Apart from the Ottomans, various contingents of their auxiliary troops also supported the princes of Transylvania, the tsars of Russia, the kings of Poland, the Prussians, and the Cossacks. In the first half of the seventeenth century, Crimean Tatars played a great role in the making of the power structures in Eastern Europe. They formed a foreign policy of their own, although they had to remain sensitive to the interests of the Sublime Porte. For a yearly tribute—called pominki (a gift given out of reverence)—the Russians

69 Mária Ivanics, “Khan Gazi II. Giray, der Dichter und der Heerführer,” in Geteilt— Vereinigt: Beiträge zur Geschichte des Königreichs Ungarn in der Frühneuzeit (16.–18. Jh.), ed. Krisztián Csaplár-Degovics and István Fazekas (Berlin, 2010), 250–259. 70 From the time of the rule of Khan Mehmed Giray (1515–1523) on, Muscovy and Poland sent the former Tatar tribute to the Crimea. For the first half of the seventeenth century, we have a precise yearly list of the Russian tribute, see A. A. Novosel’skii, Bor’ba Mosk- ovskogo Gosudartsva s tatarami v pervoi polovine XVII veka [The struggles of Muscovy with the Tatars in the first half of the seventeenth century] (Moscow and Leningrad, 1948). For the development of the Russian tribute between 1485 and 1685, see Sagit Faizov, Pominki “tysh” v kontekste vzaimootnoshnii Rusi-Rossii s Zolotoi Ordoi i Krymskim Iurtom [The tribute “tysh” in the context of the contacts between Rus–Russia and the Golden Horde and Crimean Khanate]. See http://kirimtatar.com/index.php?option=com_content&task=view &id=132&Itemid=381 (downloaded 29 March 2010). For the sums paid by Poland in the mid-seventeenth century, see Zbigniew Wójczik, “Some Problems of Polish–Tatar Rela- tions in the Seventeenth Century: The Financial Aspect of the Polish–Tatar Alliance in the Years 1654–1666: Polish–Tatar Relations in the Seventeenth Century,” Acta Poloniae Historica 13 (1966), 87–102. Moldavia and Wallachia paid their duties for the khan in natural products (mostly in honey and wax), see Vasile Mihordea, “Raporturile Moldovei şi Ţării Româneşti cu tătarii în secolele XVI–XVIII” [The contacts of Moldavia and Wallachia with the Tatars in the sixteenth to the eighteenth centuries], Revista de Istorie 32 (1979): 1069–1095. 296 mária ivanics could use their services against Poland.71 They also offered Poland their help against Muscovy in exchange for five thousand gold coins.72 In the middle of the century, they became involved in the First Northern War (1655–1660), a result of the escalation of the Polish-Cossack conflict. On the side of the Polish army, Tatar troops engaged in military activities as far away as in Brandenburg.73 Tatar auxiliary troops helped the prince of Transylvania, Gábor Bethlen (1613–1629), to power and also participated in his anti-Habsburg wars.74 In 1657, on the order of the Sublime Porte, the Tatars stopped the military expedition of György Rákóczi II against Poland and thus hindered the successful expansion of the Transylvanian prince’s area of influence.75 Between 1598 and 1682 they also maintained diplomatic contacts with the Habsburgs; after 1662 their envoys visited the Viennese court each year. Their activity in the realm of foreign policy can be compared to that of the princes of Transylvania. In times of the rule of weaker sultans and under favorable political constellations they also tried to extend their influence to the Christian vassals of the Ottoman Empire.76 In the mid-seventeenth century the Crimean khan was one of the key players in the politics of the region.

71 Mária Ivanics, “Entstehung und Quellenwert der krimtatarischen tiyiş defters,” Acta Ori- entalia Academiae Scientiarum Hungaricae 47, nos. 1–2 (1994): 105–112. 72 Letter of Khan Gazi Giray II to the Polish King Sigismund III, 1592. Dariusz Kołodziejczyk, The Crimean Khanate and Poland-Lithuania: International Diplomacy on the European Periphery (15th–18th Century): A Study of Peace Treaties Followed by Annotated Documents (Leiden and Boston, 2011), no. 34. 73 Hans Saring, “Tatarische Gesandtfschaften an Kurfürst Friedrich Wilhelm während des Ersten Nordischen Krieges,” Forschungen zur brandenburgischen und preußischen Geschichte, 46 (1934), 374–380. 74 Gábor Bethlen gave a bill of debt of 5,000 gold coins to the Kalga Şahin Giray as thanks for escorting him into the principality and helping him on the throne. For the payment of his debt, the Tatar office-holder expected gunpowder worth one thousand guldens, expensive cloth worth two thousand, and the rest in cash. TMÁO I: 425–427. See also the study of János B. Szabó in this volume. 75 On the First Northern War, see Jan Wimmer, ed., Wojna polsko-szwedzka 1655–1660 [The Polish–Swedish war 1655–1660] (Warsaw, 1973) and Eckhardt Opitz, Österreich und Brandenburg im schwedisch–polnischen Krieg 1655–1660 (Boppard am Rhein, 1969). 76 Report of Bailo Marco Venier from Pera, 12 November 1594. “Il disparere tra Sinan et il Tartaro dicono che sia nato pechè voleva il Tartaro che due suoi fratelli fossero posti al governo, l’uno della Valachie et l’altro della Bogdania.” Ioan Slavici, ed. Documente privitoare la istoria românilor culese de Eudoxiu de Hurmuzaki [Documents for the history of the Romanians collected by Eudoxiu de Hurmuzaki], vol. 3, 1576–1600 (Bucharest, 1880), 463; Pečevī, Tārih, 174. the military co-operation of the crimean khanate 297

The Diplomacy of the Crimean Tatars

The Khanate of Crimea maintained contacts with a number of European and Asian powers during the sixteenth and seventeenth centuries. Russia, Poland, the Cossacks, Sweden, Denmark, Brandenburg, the Holy Roman Empire, Transylvania, Georgia, and Persia were the most important des- tinations of Tatar envoys—all connections that are documented in surviving archival sources. The Crimean Tatars conducted an independent foreign policy toward Russia and Poland. The most important task of Crimean Tatar diplomacy was having their khans acknowledged as the heirs of the Golden Horde.77 The Tatar envoys negotiated in Moscow and in Warsaw about the sum and the composition of the tribute to be paid to the khan, his family, and his court officers; about the designation of the ransom for people who had fallen into Tatar captivity; and about the number of the auxiliary troops the Crimeans were supposed to send.78 They had diplomatic contacts with Sweden from 1591 and with Denmark and Brandenburg from the mid-seventeenth century, though this nevertheless remained a rather one-sided series of visits.79 The experiences of Crimean Tatar diplomacy with foreign powers made it possible for them to serve as mediators at peace negotiations between the Ottoman Empire and Christian rulers. Khan Gazi Giray II was the first to take on such a

77 The only military cooperation between the Ottoman and Tatar armies on the steppe (which, however, eventually failed) was arranged exactly to achieve this goal. In a common enterprise, a channel was to be built in order to facilitate the transportation of military supplies against the Russians, primarily the cannons necessary for the siege of Astrakhan. Through the recapturing of this important merchant town on the Lower Volga and the Caspian Sea, they hoped to cut off the Volga region from the Russian sphere of influence. This plan, however, could not be put into practice. See Halil İnalcık, “The Origins of the Ottoman-Russian Rivalry and the Don-Volga Canal,” Ankara Üniversitesi Dil- Tarih- Cografya Fakültesi Dergisi 1 (1946–1947): 47–106; Lajos Tardy and István Vásáry, “Andrzej Taranowskis Bericht über seine Gesandschaftsreise in der Tartarei (1569),” Acta Orientalia Academiae Scientiarum Hungaricae 28 (1974): 213–252. 78 The most important publications about this are: D.V. Liseitsev, “Russko-krymskie diplomaticheskie kontakty v nachale XVII veka” [Russian–Crimean Tatar diplomatic contacts in the early seventeenth century], in Tiurkologicheskii Sbornik 2005, ed. S.G. Kliash- tornyi, S.G. Sultanov and V.V. Trepavlov (Moscow, 2006), 234–282; Dariusz Kołodziejczyk, Ottoman–Polish Diplomatic Relations (15th–18th Century): An Annotated Edition of ‘Ahd- names and Other Documents, The Ottoman Empire and its Heritage, no. 18 (Leiden, 2000); Josef Matuz, “Les relations étrangères du Khanat de Crimée (XVe–XVIIIe siécles)” Revue d’ Histoire Diplomatique 3–4 (1988): 233–249; Özyetgin, Kırım Yurtında. 79 Karl Vilhelm Zetterstéen, Türkische, tatarische und persiche Urkunden im Schwedi schen Reichsarchiv (Uppsala, 1945); Josef Matuz, Krimtatarische Urkunden im Reichsarchiv zu Kopenhagen: mit historisch-diplomatischen und sprachlichen Untersuchungen, Islam- kundliche Untersuchungen, no. 37 (Freiburg im Breisgau, 1976). 298 mária ivanics function, at the making of the peace of Ţuţora with Poland in 1595. The Ottoman–Polish peace treaty of Žuravno in 1676 was also concluded by Khan Selim Giray I as the authorized delegate of the sultan.80 The Long Turkish War (1593–1606) meant opportunities for the Tatars to establish diplomatic contacts with the imperial court in Vienna. They maintained continuous, although mostly one-sided, diplomatic relations for almost a century (1598–1682).81 The establishment of these contacts was closely connected to the events of the Long Turkish War. Contacts were established for the first time with the mediation of the prince of Transylvania and the voievod of Wallachia, that is, the allies of Emperor Rudolph II. Khan Gazi Giray II proposed to Zsigmond Báthory, prince of Transylvania, a mutual agreement for support and an alliance against the assaults of the Sublime Porte. At the same time, he asked the prince to mediate peace between Vienna and Constantinople.82 In 1599 the delegation of the sultan to the peace negotiations was led personally by Khan Gazi Giray II, who stayed in Hungary for the winter with his army in that year. Although these negotiations ended without success, a core of the later peace of Zsitvatorok (1606) can already be found in the documents of the negotiations.83 After the peace of Zsitvatorok, diplomatic contacts intensified again during the First Northern War, when the Tatars and the emperor supported Poland in its struggle against the Swedish– Transylvanian alliance. Between 1666 and 1682, Tatar envoys came to Vienna each year, despite the fact that their ruler did not have any concrete political goals of his own with these embassies; neither did he act as a mediator. They came to secure economic goals, namely, the maintenance of customs’-free trade.

80 Kołodziejczyk: Ottoman–Polish Diplomatic Relations, documents no. 26 and no. 53. 81 Mária Ivanics, “Krími-tatár követjárások a bécsi udvarban (Vázlat a krími-tatár diplomá- cia 16–17. századi történetéről)” [Tatar envoys at the imperial court: A sketch of the history of the Crimean Tatar diplomacy in the sixteenth and seventeenth centuries], Aetas 14, no. 4 (1999): 41–50; Christoph Augustynowicz, “Tatarische Gesandschaften am Kaiserhof des 17. Jahrhunderts—Protokoll und Alltag,” in Das Osmanische Reich und die Habsburgermonar- chie: Akten des internationalen Kongresses zum 150-jährigen Bestehen des Instituts für Öster- reichische Geschichtsforschung, Wien, 22.–25. September 2004, ed. Marlene Kurz, Martin Scheutz, Karl Vocelka, and Thomas Winkelbauer (Vienna and Munich, 2005), 315–340. 82 For the planned alliance between Khan Gazi Giray II and Prince Zsigmond Báthory in 1598, see Hungarian–Ottoman Military and Diplomatic Relations in the Age of Süleyman the Magnificent, ed. Géza Dávid and Pál Fodor (Budapest, 1994), 183–198. 83 Mária Ivanics, “A zsitvatoroki békéhez vezető út: Az 1599. évi béketárgyalások” [The road to the peace of Zsitvatorok: Peace negotiations in 1599], Történelmi Szemle 35, nos. 3–4 (1993): 297–311. the military co-operation of the crimean khanate 299

* * * The various aspects of the military cooperation between the Ottomans and Tatars, the political and economic background, and the diplomatic consequences of the campaigns discussed in this paper make clear that the Crimean Khanate, as a marginal state in the sphere of interest of the Ottoman Empire, enjoyed a relatively significant independence in its military activities during the sixteenth and seventeenth centuries and also had a great deal of space to maneuver in the sphere of foreign policy. The permanent contacts of the Tatars with the Habsburgs, the arch-enemy of the Ottomans, served to counterbalance the often incalculable power relations and intrigues at the Sublime Porte.

‘Splendid Isolation’? The Military Cooperation of the Principality of Transylvania with the Ottoman Empire (1571–1688) in the Mirror of the Hungarian Historiography’s Dilemmas

János B. Szabó

Historiographic Introduction

One of the great paradoxes of modern Hungarian historiography is that it gives a prominent place in the discussions on the sixteenth and seventeenth centuries to the Principality of Transylvania as an independent Hungarian state formation that seceded from the Kingdom of Hungary; at the same time for many decades it entirely neglected any other context for analyzing the frames of the Transylvanian state’s development.1 One such context is the state’s contacts with the Ottoman Empire, a very seldom noted theme in historical overviews. The only Hungarian monograph concerning this issue was published in 1907, and in German the most recent short overview dates from 1923. In contrast, modern Romanian historiography dedicates considerable attention to the topic, including several monographs in the last two decades.2 Indeed, it can be assumed that the

1 Zsolt Trócsányi, Erdélyi központi kormányzata 1540–1690 [Central Government in Transylvania 1540–1690] (Budapest, 1980); István Sinkovics, “Az erdélyi fejedelemség államszervezete” [State administration of the Principality of Transylvania], in Magyar ország története 1526–1686 [The history of Hungary 1526–1686], ed. Zsigmond Pál Pach and Ágnes R. Várkonyi (Budapest, 1985), 439–453; Katalin Péter, “Az erdélyi országgyűlés a kora újkori magyar fejlődésben” [The diet of Transylvania in the development of early modern Hungary], in Kálmán Benda and Katalin Péter, Az országgyűlések a kora újkori magyar történelemben [Diets in early modern Hungarian history] (Budapest, 1987), 13–23. 2 Gyula Vajda, Erdély viszonya a Portához és a római császárhoz mint magyar királyhoz a nemzeti fejedelemség korszakában [The relationship of Transylvania to the Porte and the emperor as king of Hungary in the epoch of the national principality] (Kolozsvár, 1891); a shorter overview in Hungarian with comparative parts was published by Vencel Bíró, Erdély követei a Portán [Transylvanian envoys at the Porte] (Kolozsvár, 1921), 58–60, and “Erdély és a porta” [Transylvania and the Porte], Századok 57 (1923): 76–93. In German: Georg Müller, Türkenherrschaft in Siebenbürgen: Verfassungsrechliches Verhältnis Sie- benbürgens zur Pforte 1541–1688 (Hermannstadt-Sibiu, 1923); for a summary of the more recent literature, see Gerald Volkmer, Das Fürstentum Siebenbürgen: Aussenpolitik und 302 jános b. szabó reason for this oversight and silence is the topic’s relevance and sensitive nature. In 1866 Sándor Szilágyi wrote the first modern synthesis on the principality’s history, in his foreword defining the perspectives that served as a guide for Hungarian historiography for more than a century: It is true that recently historiographers from Hungary give adequate respect to the relevance of Transylvania, study and analyze the sources of its history and write about it much more widely and fairly than their predecessors (. . .) they have neither space nor opportunity to destroy the bulk of old prejudices and to show the full picture of the institutions of the small state, which were so functional, its struggles, in which they had shown so much persistence, and its sufferings, which in some cases are unique in universal history (. . .) Today, no one would deny that it [the Principality] had the mission of maintaining the Hungarian nation and its constitution, even in cases when it brought its weapons to the other shore of the river Tisza [that is, when it waged war in Hungary] (. . .) It is a futile endeavor to make these movements look small by labeling them as revolts and a vain ambition to degrade the princes to simple leaders of specific parties. There was a time when Transylvania played a part in European politics and the foreign powers saw it as a factor in their calculations; and [the fact] that it was so is not simply the work of one or two ambitious men, regardless of whether or not some people want the world believe it is the case.3 This work, like many contemporaneous ones, was a typical product of the anti-Habsburg Hungarian war for independence of 1848–1849, a war that brought about fierce attacks on the dynastic perspective of earlier historio graphy. Although the specific political situation changed with the Austro- Hungarian Compromise of 1867, the earlier situation was imprinted and over time affirmed by aspects of political debates from the period. It has had a large influence on historical thinking and the direction of research that has been sustained until the present time.4

Völkerrechtliche Stellung (Heidelberg and Braşov, 2002), 40–61. In Romanian: Călin Felezeu, Statutul Principatului Transilvaniei în raporturile cu Poarta Otomană 1541–1688 [The status of the Principality of Transylvania in relation to the Ottoman Empire] (Cluj-Napoca, 1996); Cristina Feneşan, Constituirea Principatului autonom al Transilvaniei [The formation of the autonomous Principality of Transylvania] (Bucharest, 1997). 3 Sándor Szilágyi, Erdélyország története, tekintettel mívelődésére [The history of Tran- sylvania, with special attention to its culture] (Pest, 1866), v–vii. 4 Géza Pálffy, The Kingdom of Hungary and the Habsburg Monarchy in the Sixteenth Century (Boulder, CO, 2009), 1–6. See also the short summary of the viewpoints of Géza Pálffy and Sándor Őze in the most recent historians’ debate on the topic: Péter Illik, Tör ténészek, viták a 16–17. századi magyar történelemről [Historians and debates on sixteenth and seventeenth-century Hungarian history] (Budapest, 2011), 67–89. transylvanian-ottoman military cooperation 303

The vigor with which these earlier perspectives survived can be measured in the recently published English language summaries about the history of Hungary.5 In any case, the old question remains: how exactly should this mission of Transylvania as a “freedom fighter” or “unifier of the country” be understood, when “these princes, particularly István Bocskai, Gábor Bethlen, and Imre Thököly fought for an ‘independent Hungary,’ albeit with the help of several thousand Ottoman and Tatar troops.”6 For a long time the “solution” to this problem was to avoid the topic: the fact that the princes of Transylvania were vassals of the Ottoman Empire was not denied, but its relevance underwent a perceptible relativization.7 Ambitious undertakings, such as the three-volume History of Transylva nia were published during the 1980s in Hungary, by their sheer bulk and lack of a unified structure they lost themselves in the details of political, economic, and cultural history and failed to address structural questions

5 “Transylvania also maintained the hope that the unity of the kingdom would be re- established quite soon (. . .) The institutions and the court of royal Hungary being outside the country, those of Transylvania were cherished as the repositories of Hungarian statehood and culture.” László Kontler, Millennium in Central Europe: A History of Hungary (Budapest, 1999), 150; “There is no doubt, that the autonomy of Transylvania (. . .) became an important locus of Hungarian statehood in the midst of the German and Ottoman domination, and a haven for Hungarian culture and identity that was significant for the nation’s future.” Paul Lendvai, The Hungarians: A Thousand Years of Victory in Defeat (Lon- don, 2003), 106; “The century of virtual Transylvanian autonomy had thus been character- ised by more than the Principality’s leadership of the struggle for Hungarian unity and independence (. . .) Transylvania had offered a beacon and pride to all Hungarians during one of the darkest periods of their history.” Bryan Cartledge, The Will to Survive: A History of Hungary (London, 2006), 96. 6 Pálffy, The Kingdom of Hungary, 4. For the “missionary” interpretation, see e.g., Béla Köpeczi, “The Hungarian Wars of Independence of the Seventeenth and Eighteenth Cen- turies in Their European Context,” in From Hunyadi to Rákóczi: War and Society in Late Medieval and Early Modern Hungary, ed. János M. Bak and Béla K. Király (New York, 1982), 445–455. When they had to reflect on the principality’s position toward Habsburg rule in Hungary, seventeenth-century contemporaries instead emphasized the state’s role as a counter balance. See Gábor Kármán, “Transylvania Between the Ottoman and Habsburg Empires,” in Statehood Before and Beyond Ethnicity: Minor States in Northern and Eastern Europe, 1600–2000, ed. Linas Eriksonas and Leos Müller (Brussels, 2005), 151–158. 7 If the question was raised at all, studies were conducted and written from the perspective of excusing the princes of Transylvania from accusations of cooperating with the Ottomans, placing emphasis upon the coercive force of circumstances. For example, László Nagy, “A XVII. századi Habsburg-ellenes függetlenségi harcok értékeléséhez (A török szövetség problematikája a Bocskai, Bethlen és I. Rákóczi György vezette küzdelmekben)” [To the interpretation of seventeenth-century anti-Habsburg wars for independence: The problem of Turkish alliance in the struggles led by Bocskai, Bethlen, and György Rákóczi I], Hadtörténelmi Közlemények n.s. 10, no. 2 (1963): 185–241. 304 jános b. szabó concerning the long-term development of the principality.8 In shorter summaries the dependent status is sometimes only described as “nominal” vassalage,9 or even as a “protectorate,”10 and the establishment of a “real” vassalage is traced to the 1660s, after the fall of the house Rákóczi and the Ottoman conquest of Várad.11 One explanation for this long and diplomatic silence is that an analysis of the contacts with the Ottoman Empire could be associated with the painful political problems related to the presence of Soviet occupation forces in post-World War II “socialist” Hungary.12 Since the late 1980s, research has also been less hindered by the “international harmonizing” of the cultural politics of Eastern European regimes dependent on the Soviet Union. The rather oddly interpreted “solidarity” of the “socialist” countries led to a system in which the research of Hungarian scholars into the past of neighboring “friendly” countries was directly and indirectly restricted, apart from the topic of the anti-Habsburg, or generally anti-German (that is, anti-Western) “wars of independence.” The history of Transylvania fell into this restricted category, as clearly shown in the vehement reactions

8 See the full English translation: History of Transylvania, vols. 1–3, ed. Béla Köpeczi (Boulder, CO, 2002). 9 Peter F. Sugar, “The Principality of Transylvania,” in A History of Hungary, ed. Peter F. Sugar (London and New York, 1990), 129. 10 Péter Szabó, Az Erdélyi Fejedelemség [The Principality of Transylvania] (Budapest, 1997), 39–40; Sinkovics, “Az erdélyi fejedelemség államszervezete,” 442, 453. 11 “The country that shrank into a small state lived from this time on really as an Otto- man vassal.” Gábor Barta, Erdély [Transylvania] (Budapest, 1989), 54. “(. . .) the subsequent installation by the Porte of a puppet prince, Ákos Barcsay, to rule a devastated Transyl- vania marked the end of Principality’s existence as an independent force in European politics,” Cartledge, The Will to Survive, 112. 12 In the contemporary atmosphere, where everything was seen in the light of its political relevance, it was clear how a historical essay published in 1983, on the 400th anniver- sary of the birth of Gábor Bethlen, later turned into a film (Ágnes Hankiss’ Érzékeny búcsú a fejedelemtől [Sentimental farewell to the prince]) suggested strong parallels between the activities of this prince, widely regarded as the most successful ruler of Transylvania, who gained his throne with Ottoman military support in 1613, and that of János Kádár, the chief secretary of the Party, who also won his power with Soviet weapons in 1956. See Lendvai, The Hungarians, 117–118, 124. This highlights the significance of remarks such as that of Katalin Péter in the three-volume History of Transylvania (1986): “Had Gabriel Bethlen died at this moment, he would have been remembered as one of the most sinister figures in the annals of Transylvania. Instead, he continued to reign for thirteen years and took his place in history among Transylvania’s outstanding leaders.” Katalin Péter, “The Golden Age of the Principality (1606–1660),” in History of Transylvania, vol. 2, From 1606 to 1830, ed. László Makkai and Zoltán Szász (New York, 2002), 40. transylvanian-ottoman military cooperation 305 to the publication of the three-volume Hungarian History of Transylvania in 1986.13 It was only during the 1990s that circumstances arose in which work could begin to place the Principality of Transylvania into a wider international system of coordinates. The changes in Hungarian historiography after 1990 are well illustrated by the works of László Nagy, who before the transition wrote a dozen books about the anti-Habsburg military endeavors of the Transylvanian princes and the principality’s military. These were based mostly on nineteenth-century source editions, because—apart from some local archives in Czechoslovakia—he did not have access to the relevant manuscript collections. In a study published in 1992 he was able to establish a new perspective that was exactly the opposite of his former anti-Habsburg viewpoint.14

13 See Béla Köpeczi, “Introduction to the English Edition,” in History of Transylvania, vol. 1, From the Beginnings to 1606, ed. László Makkai and András Mócsy (Boulder, CO, 2002), 9–13; idem, “Erdély története harminc év távlatából” [The history of Transylvania in the perspective of thirty years], Kisebbségkutatás 15, no. 1. (2006): 47–59. 14 See László Nagy, “A magyarországi Habsburg-uralom a török hódoltság idején” [Habs burg rule in Hungary in the period of the Ottoman occupation], Hadtörténelmi Közlemé nyek 105, no. 2 (1992): 3–29. The first Hungarian historian to be mentioned in the context of the new research is Gábor Barta, who studied the collapse of the Kingdom of Hungary and explained the formation of the Principality of Transylvania through the circumstances of the rule of John I Szapolyai and his son, John Sigismund. See his “An d’illusions (Notes sur la double éléction de rois après la défaite de Mohács),” Acta Historica Academiae Scien tiarum Hungaricae 2 (1978): 1–40; Az erdélyi fejedelemség születése [The birth of the Princi- pality of Transylvania] (Budapest, 1979); La route qui mène à Istanbul 1526–1528 (Budapest, 1994). More recently, Teréz Oborni focused upon the contacts of the principality toward the Kingdom of Hungary and the Habsburg dynasty on the basis of the treaties concluded between them. Apart from her chapter in this volume, see also her previous studies: “From Province to Principality: Continuity and Change in Transylvania in the First Half of the Sixteenth Century,” in Fight Against the Turk in Central-Europe in the First Half of the 16th Century, ed. István Zombori (Budapest, 2004), 165–180; “A gyalui szerződés” [The treaty of Gyalu], in A magyar államiság első ezer éve [The first thousand years of Hungarian statehood], ed. Márta Font and István Kajtár (Pécs, 2000), 133–146; “Die Plane des Wiener Hofes zur Rückeroberung Siebenbürgens 1557–1563,” in Kaiser Ferdinand I: Ein mitteleuropäischer Herrscher, ed. Martina Fuchs, Teréz Oborni, and Gábor Ujváry (Munster, 2005), 277–298; “Erdély közjogi helyzete a speyeri szerződés után (1571–1575)” [The legal status of Transyl- vania after the treaty of Speyer], in Tanulmányok Szakály Ferenc emlékére, ed. Pál Fodor, Géza Pálffy, and István György Tóth (Budapest, 2002), 291–305; “Báthory Gábor megál- lapodásai a Magyar Királysággal” [The agreements between Gábor Báthory and the King- dom of Hungary], in Báthory Gábor és kora, ed. Annamária Jeney-Tóth, Klára Papp, and Attila Ulrich (Debrecen, 2009), 111–122; and “Bethlen Gábor és a nagyszombati szerződés (1615)” [Gábor Bethlen and the treaty of Nagyszombat], Századok 145 (2011): 877–914. The political and dynastic contacts of the principality with the Polish–Lithuanian Com- monwealth again came under the spotlight, see Ildikó Horn, Hit és hatalom: Az erdélyi uni tárius nemesség 16. századi története [Faith and power: The history of the Unitarian nobility 306 jános b. szabó

This new research trend scrutinized several aspects of the contacts with the Ottoman Empire. Pál Fodor drew an entirely new picture about the division of the sixteenth-century Kingdom of Hungary, shedding light on the involvement of the precursors of the Principality of Transylvania in the Ottoman system of sancaks; he introduced the term “virtual sancak” into Hungarian historiography. This was applied to those territories that were not occupied, but which the Ottomans granted charters formally integrating them into the empire. Included among these were those regions that later constituted the Principality of Transylvania.15 Later on, Gábor Ágos- ton drew a parallel between this form of organization—traditionally seen as unique—and the system of the hereditary sancaks known from the eastern periphery of the empire. This comparison clarified the versatile character of the Ottoman structure of administration.16 Simultaneously, new research aimed at analyzing the procedure of the Transylvanian princes’ appointment, inauguration, and their insignia.17 Sándor Papp’s

of Transylvania in the sixteenth century] (Budapest, 2009); eadem, A könnyező krokodil: Jagelló Anna és Báthory István házassága [The weeping crocodile: The marriage of Anna Jagiełło with István Báthory] (Budapest, 2007); eadem, Báthory András (Budapest, 2002); Sándor Gebei, “Der Fürst von Siebenbürgen Gábor Bethlen und der polnische Thron,” in The First Millenium of Hungary in Europe, ed. Klára Papp and János Barta (Debrecen, 2002), 197–207; idem, Az erdélyi fejedelmek és a lengyel királyválasztások [The princes of Transylvania and royal elections in Poland] (Szeged, 2007). New light was shed on topics that had been popular earlier, such as the principality’s diplomacy toward Western Europe, see Gábor Kármán, Erdélyi külpolitika a vesztfáliai béke után [The foreign policy of Transylvania after the peace of Westphalia] (Budapest, 2011); idem, “The Hardship of Being an Ottoman Tributary: Transylvania at the Peace Congress of Westphalia,” in Frieden und Konfliktmanagement in interkulturellen Räumen, ed. Norbert Spannenberger and Arno Strohmeyer (Stuttgart, 2013), 163–183. 15 Pál Fodor, “Ottoman Policy towards Hungary, 1520–1541,” Acta Orientalia Academiae Scientiarum Hungaricae 45, nos. 2–3 (1991): 271–345; idem, “Das Wilajet von Temeschwar zur Zeit der osmanischen Eroberung,” Südost-Forschungen 55 (1996): 25–44. 16 Gábor Ágoston, “A Flexible Empire: Authority and its Limits on the Ottoman Fron- tiers,” International Journal of Turkish Studies 9 (2003): 15–31. 17 For the typology of the insignia sent to the princes by the Ottomans; their parallels elsewhere; the procedure of their delivery; and the inauguration ceremonies of the princes, see János B. Szabó, “Insignia of the Transylvanian Princes,” Majestas 4 (1996): 85–105; János B. Szabó and Péter Erdősi, “Ceremonies Marking the Transfer of Power in the Principal- ity of Transylvania in East European Context,” Majestas 11 (2003): 111–160; János B. Szabó, “The Insignia of the Princes of Transylvania,” in Turkish Flowers: Studies on Ottoman Art in Hungary, ed. Ibolya Gerelyes (Budapest, 2005), 131–142. On the phases of the Ottoman inauguration of the princes, and the types of the documents of inauguration, see Sándor Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden der Osmanen für Ungarn und Siebenbürgen: Eine quellenkritische Untersuchung (Vienna, 2003); idem, “Critical Survey of Letter of Donation, Confirmation and that of Agreement Issued by the Ottoman Sultans for the Ruler of Hungary and Transylvania,” Chronica (Szeged) 4 (2004): 126–128. On the use of Ottoman military music as a symbol of power, see Balázs Sudár and Rumen István transylvanian-ottoman military cooperation 307 systematic research in the diplomacy of the period was especially important, outlining for the first time the phases of the appointment procedure of the princes, the types of documents used for their inauguration, and the place of the phenomenon in the contexts of Ottoman legal theory.18 Apart from research on the economic contacts of the principality, the most recent studies also address the networks of personal contacts between the princes and the leading Ottoman politicians of the region. Their description of the former’s role in the faction struggles of the empire’s political life offers valuable insight into the hitherto rather mod- estly known history of the complex interactions between the two state entities.19 There is an absence of information regarding cooperation in the military field; we still substantially lack research that would uncover the Turk- ish source material. Consequently I am limited here to a summary of the data available at the moment.20 On the basis of this, we can take note of the (mostly implicit) assumption of Hungarian historiography, that there was a “golden age” of the Principality of Transylvania—usually placed at the reign of István Báthory (1571–1586) or Gábor Bethlen (1613–1629). The chief notion here is that the principality’s space for maneuvering was so broad that, except for the payment of the tribute, each of its obligations

Csörsz, “Trombita, rézdob, tárogató. . .” (A török hadizene és Magyarország) [“Trumpet, copper drum, Turkish pipe . . .”: Turkish military music and Hungary] (Enying, 1996). 18 See his studies quoted in the previous note, as well as his chapter in this volume. 19 Sándor Papp, “Homonnai Drugeth Bálint, Rákóczi Zsigmond és Báthory Gábor küzdelme az Erdély feletti hatalomért 1607-ben” [The stuggle between Bálint Homonnai Drugeth, Zsigmond Rákóczi, and Gábor Báthory over Transylvania in 1607], Századok 148 (2008): 807–848; idem, “Bethlen Gábor, a Magyar Királyság és a Porta” [Gábor Bethlen, the Kingdom of Hungary and the Porte], Századok 145 (2011): 915–974; Balázs Sudár, “Iskender and Gábor Bethlen: The Pasha and the Prince,” in Europe and the Ottoman World: Exchanges and Conflicts (Sixteenth-Seventeenth Centuries), ed. Gábor Kármán and Radu G. Păun (Istanbul, 2013), forthcoming; János B. Szabó and Balázs Sudár, “‘Independens fejedelem az Portán kívül’: II. Rákóczi György oszmán kapcsolatai. Esettanulmány az Erdélyi Feje- delemség és az Oszmán Birodalom viszonyának történetéhez. (1. rész)” [“An independent prince outside the Porte”: The Ottoman relations of Prince György Rákóczi II: A case study for the history of the relationship between the Principality of Transylvania and the Otto- man Empire (Part one)], Századok 146 (2012): 1017–1048. For economic contacts, see Mária Pakucs-Willcocks, Sibiu–Hermannstadt: Oriental Trade in Sixteenth Century Transylvania (Cologne, 2007); eadem, “Oriental Trade and Merchants: Economic Relations Between the Ottoman Empire and Transylvania in the Sixteenth Century,” in Osmanische Orient und Ostmitteleuropa, ed. Robert Born and Andreas Puth (Stuttgart, 2013), forthcoming. 20 Surprisingly, the military dimension has regularly been ignored in recent research on the interaction between the Ottoman Empire and the Principality of Transylvania. 308 jános b. szabó

(among them, the military) to the Ottoman Empire was nominal and could be avoided.21 The starting point of my analysis is 1571, and the election of István Báthory as a ruler. This date can be seen as marking the end of the principality’s formation period after many earlier stages, as the state continued to function and implement the principles laid down during the rule of the Szapolyai dynasty (even after their extinction).22 The end point of the study is 1688, when the Principality of Transylvania was definitively occupied by the forces of Leopold I, king of Hungary and emperor of the Holy Roman Empire. Prince Mihály Apafi I then signed the declaration of Fogaras, denouncing all his obligations toward the Ottoman Empire

21 In some cases Hungarian authors did emphasize the relevance of Transylvania’s vassal status, in order to point out the sharp differences toward the two Romanian voievodates. See Kálmán Benda, “Erdély politikai kapcsolatai a román vajdaságokkal a XVI–XVII. században” [The political contacts of Transylvania with the Romanian voievodates in the sixteenth and seventeenth centuries], in Magyarok és románok, vol. 2, ed. József Deér and László Gáldi (Budapest, 1944), 51–53; idem, “L’idée d’empire en Europe carpathique à la fin du Moyen Age,” Revue d’Histoire Comparée: Études Hongroises 22 (1944): 54–80. At the same time, Romanian historiography almost always made a point of underlining the similar status of the “three Romanian states” (i.e., Transylvania, Moldavia, and Wallachia) toward the Ottoman Empire. For example: “Transylvania and Partium consolidated their status of autonomous principality under Ottoman suzeranity, a status similar to that of Wallachia and Moldavia,” or “Ottoman hegemony was uneven across the three countries, even though the differences in status were unessential. Transylvania enjoyed lighter suzeranity for a number of reasons (greater distance to the center and closeness to the west; its subjugation was accomplished later, fear of the Habsburgs, who pleaded the “liberation” of the province, etc.).” Ioan-Aurel Pop, “Romanians in the 14th–16th Centuries: From the ‘Christan Republic’ to the ‘Restoration of Dacia,’ ” in History of Romania (Compendium), ed. Ioan-Aurel Pop and Ioan Bolovan (Cluj-Napoca, 2006), 281, 288. At the same time, for many it was important to prove the contractual character of the Romanian voievodates’ status toward the Ottoman Empire. For the debate, see Viorel Panaite’s chapter in this volume, as well as Sándor Papp, “Keresztény vazallusok az Oszmán Birodalom észak-nyugati határainál (Diplomatikai vizsgálat a román vajdák szultáni ‘ahdnâméi körül)” [Christian vassals at the northwestern borders of the Ottoman empire: Diplomatic analysis of the ‘ahdnames for the Romanian voievods], Aetas 17, no. 1 (2002): 67–94; idem, “Christian Vas- sals on the Northwest Border of the Ottoman Empire,” in The Turks, vol. 3, Ottomans, ed. Hasan Celal Güzel, C. Cem Oguz, and Osman Karatay (Ankara, 2002), 719–730. Cf. Mihai Maxim, “Le statut des Pays Roumains envers la Porte Ottomane aux XVI–XVIIIe siècles,” Revue Roumaine d’Histoire 24, nos. 1–2 (1985): 29–50; Tahsin Gemil, “’Capitulaţiile’ Tran- silvaniei de la jumătatea secolului XVII” [The “capitulations” of Transylvania from the mid-seventeenth century], Anuarul Institutului de Istorie şi Arheologie “A. D. Xenopol,” Iaşi 23 (1986): 717–721; Mihai Maxim, “Din istoria relaţiilor Româno-Otomane: ‘Capitulaţiile’” [From the history of the Romano–Ottoman relations: “Capitulations”], Anale de Istorie 6 (1982): 31–68. 22 Oborni, “Erdély közjogi helyzete.” transylvanian-ottoman military cooperation 309 finally and completely.23 I do not address those cases when “auxiliary troops” from the Ottoman Empire repeatedly forced specific princes (who lacked strong inner support) upon the principality, as happened in the cases of Mózes Székely (1603), István Bocskai (1606), Gábor Bethlen (1613), Ákos Barcsai (1658), and Mihály Apafi I (1661).

The Army of the Principality of Transylvania

The existence of the Transylvanian principality was roughly simultaneous with the first phase of the military revolution took place in Western Europe, which had its versatile impacts on Central Europe, including Tran- sylvania.24 Although throughout its existence the principality retained its medieval military organization, inherited from the Kingdom of Hun- gary—based upon a system of privileges—the princely court maintained a 2–3000-man-strong army of mercenaries as well.25 In the sixteenth century the Polish were the most important foreign troops, and in the seventeenth their place was taken by Germans.26 At the Western borders of the country, which faced most of the dangers, Italian-style fortresses were built and the fortified places were supplied with artillery from the workshops in the territory of the principality. The use of firearms was widespread: At the end of the sixteenth century Hungarian and Transyl- vanian infantry consisted almost exclusively of riflemen. This was far from the norm in the region outside the medieval Kingdom of Hungary.27

23 Ágnes R. Várkonyi, “The Last Decades of the Independent Principality (1600–1711),” in History of Transylvania, vol. 2, From 1606 to 1830, ed. László Makkai and Zoltán Szász (Boulder, CO, 2002), 233–512. 24 Jeremy Black, A Military Revolution? Military Change and European Society, 1550–1800 (London, 1991); idem, European Warfare, 1660–1815 (London and New York, 1994). 25 János B. Szabó and Győző Somogyi, Az Erdélyi Fejedelemség hadserege [The army of the Principality of Transylvania] (Budapest, 1996), 40–52, 107–108. 26 Ibid., 56–60, 109. 27 Ibid., 63–74, 109–111. It is no coincidence that Sultan Murad III emphasized exactly this phenomenon in his letter to Voievod Kristóf Báthory in February 1578, in relation to the ongoing conflicts in Moldavia: “But those lords in Moldavia, they do not have rifles, but those who came in [Cossacks from Poland], they all have rifles. Due to these issues, you should quickly send in two thousand people, as many as needed, of good soldiers and lords, who could fight them. (. . .) And the army that you send should consist mostly of riflemen; there should be no neglect of this issue.” Áron Szilády and Sándor Szilágyi, ed., Török-magyarkori állam-okmánytár [A collection of state papers from the Turkish– Hungarian age], vol. 1 (Pest, 1868) (henceforth TMÁO I), 19. See also the chapter of Ovidiu Cristea in this volume, which points out the growing competitive disadvantage of the soldiers from the Romanian voievodates with regard to military technology and the limited availability of firearms. 310 jános b. szabó

The Principality of Transylvania was thus able to transmit several elements of the changes in Western European military technology to the two Romanian voievodates and to Poland. In the so-called “fortress wars” in Northern Hungary from 1557 to 1566, the Habsburg armies were led by the best German officers of the time. The principality’s army faced severe challenges, and these wars brought a tremendous amount of experience for both the officers’ corps and the common soldiers. István Báthory, voievod of Transylvania, was elected king of Poland in 1576, thus his knowledge also enriched the Polish military. He introduced important reforms, based on Transylvanian and Hungarian examples, in equipment, weaponry, tactics, as well as on recruiting and finances. As a result, the previously troubled Polish–Lithuanian Commonwealth became and remained until the 1620s a prominent military power in the region.28 According to an estimate of King Ferdinand I of Hungary, the voievodate of Transylvania and the Hungarian counties east of the river Tisza—that is, the territory later ruled by the princes—could raise 40,000 soldiers for their own defense, and 16–20,000 for deployment in Hungary.29 These numbers appear realistic: fragmentary data suggest that until the mid-seventeenth century, in the event of an offensive war, 10–12,000 soldiers could be raised from the narrowly defined territory of the principality (which had a population of approximately one million people), and another 4,000 from the Partium (which, with its population of 300,000 people, formally belonged to the Kingdom of Hungary during the entire period). This constituted a force that can be considered, for this region, relatively powerful.30 However, the Transylvanian army was, in many respects, in a disad- vantageous position when compared to a Western European type mercenary army of the same size. It lagged behind as a result of its structure of

28 For the impact of Báthory’s reforms, see Dariusz Kupisz, “The Polish–Lithuanian Army in the Reign of King Stefan Bathory (1576–1586),” in Warfare in Eastern Europe 1500–1800, ed. Brian J. Davies (Leiden, 2012), 63–79. On the Polish infantry, see Jan Wim- mer, “L’infanterie dans l’armée polonaise aux XVe–XVIIIe siècles,” in Histoire militaire de la Pologne: problèmes choisis: dissertations, études, esquisses, ed. Witold Biegannski, Piotr Stawecki, and Janusz Wojtasik (Warsaw, 1970), 78–94; on his impact on the Polish cavalry, see Richard Brzezinski and Velimir Vukšić, Polish Winged Hussar 1576–1775 (Oxford, 2006), 6, 17. 29 Karl Lanz, ed., Correspondenz des Kaisers Karl V. aus dem königlichen Archiv und der Bibliothèque de Bourgogne zu Brüssel, vol. 3, 1550–1556 (Leipzig, 1846), 11–15. 30 In the last phase of the Thirty Years War, Swedish and French diplomats had doubts about accepting the Transylvanian offers to cooperate and enter an alliance. They were afraid that the expectations of the prince, with an army significantly larger than the other allies of the two crowns, could become an obstacle to the relatively quick closing of peace negotiations. See Kármán, Erdélyi külpolitika, 87. transylvanian-ottoman military cooperation 311 branches, weaponry, and tactics, which adhered to earlier Hungarian traditions and were adjusted to the border regions of the Ottoman Empire. Moreover, it suffered from a relative lack of experience and the short- comings of older equipment. Yet despite all this it nevertheless managed to serve the political goals of the princes relatively successfully until the mid-seventeenth century.31 That being said, it is also clear that by the second half of the century Transylvania could not keep pace with the new developments. In Western Europe large well-trained permanent armies were formed, a change that could not be followed by the principality, which had lost a third of its territory in the last years of the 1650s. A spy report written shortly after 1677 estimated the total strength of the army at 13,900 soldiers, of which the prince’s permanent troops numbered only 1,500.32 When the military potential was surveyed on the occasion of the election of Ferenc Rákóczi II as a prince in 1704, only 16,511 soldiers were registered.33 The army that was available for use in interventions in foreign lands was reduced to just 7–8,000. This Transylvanian military force was inadequate for achieving local goals; in the wars from 1664 to 1686 its role was limited to a relatively successful defense of the principality’s borders.

31 Neither the anti-Habsburg campaigns of Gábor Bethlen in 1618–1621, 1623, and 1626, nor that of György Rákóczi I in 1644–1645 ended in failure. Transylvanian armies won significant victories against the armies of the Romanian voievods and the Cossack troops supporting them in Wallachia (1611, 1655) and in Moldavia (1653). They defeated Ottoman armies at Giurgiu in 1595, and had the lion’s part in the initial tactical successes of the Christian army at Mezőkeresztes in 1596. In 1636 and 1658 they were also victorious against the armies of the beylerbeyi of Buda. It must be acknowledged that they could not collect many trophies in the Ottoman wars between 1658 and 1661. This war, along with the failure of the Polish campaign in 1657, critically undermined the principality’s military abilities and prestige. For discussion of the parallel Polish development, see Dariusz Kołodziejczyk, “Az oszmán ‘katonai lemaradás’ problémája és a kelet-európai hadszíntér” [The problem of the Ottoman “military backwardness” and the eastern European theater of battle], Aetas 17, no. 4 (1999): 142–148. 32 István Czigány, Reform vagy kudarc? Kísérletek a magyarországi katonaság beillesz tésére a Habsburg Birodalom haderejébe 1600–1700 [Reform or failure? Attempts at the integration of the military in Hungary into the army of the Habsburg Empire] (Budapest, 2004), 158. 33 András Magyari, ed., II. Rákóczi Ferenc erdélyi hadserege [The Transylvanian army of Ferenc Rákóczi II] (Bucharest, 2004), 79–80. 312 jános b. szabó

The Framework of Military Cooperation with the Ottoman Empire

The texts of the early agreements between the Szapolyai dynasty and the Porte did not survive. However, it is possible, utilizing Ottoman documents issued in the Báthory-period (the last three decades of the sixteenth century) to reconstruct in some detail the framework which the empire provided for the cooperation in the military sphere with its Transylva- nian vassal. The sultan, in his letter of order (hükm-i hümayun) issued on 21 May 1571, promised to send Ottoman border troops in case of an enemy attack; a provision that was completed in the ‛ahdname-i hümayun issued the next year affirming that the Transylvanian prince was obligated to provide these Ottoman troops with victuals and equipment.34 The second ‛ahdname-i hümayun given to István Báthory in 1575— when he was reinforced in his rule by the new sultan—repeated the conditional promise of aid and went into more detail concerning the military cooperation: Similarly, if there should be any necessity for a service at the borders of our empire, when it is announced by the pashas and beys, the voievod and the dignitaries of the country should—according to their enduring subservience—show their readiness personally or through sending soldiers, equipment or victuals in the necessary extent.35 This document thus clearly prescribed mutual assistance, which was later further generalized in the charter (hükm-i hümayun) sent to Kristóf Báthory—who replaced his brother as a supervisor of Transylvania (with the title of voievod)—using the formula “be the enemy of my enemy and a friend of my friend,” later a regular part of every seventeenth-century

34 See their new editions in Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 190–194, respectively 214–219. They are also available in Hungarian translations, the first (in early modern translation) in Sándor Szilágyi, ed., Erdélyi országgyűlési emlékek történeti bevezetésekkel [Documents of the diets of Transylvania, with a historical introduction], vol. 2, 1556 sept.–1576 (Budapest, 1876) 459–461; the second (in early twentieth-century translation) in Imre Karácson, et al., ed., Török-magyar oklevéltár 1533–1789 [Turkish– Hungarian collection of documents] (Budapest, 1914), 100. 35 Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 221. The original Turk- ish text is unknown; the sixteenth-century Latin translation was also further translated to Hungarian by László Szalay, Erdély és a Porta 1567–1578 [Transylvania and the Porte] (Pest, 1862), 202. Another ‛ahdname given to Báthory in late 1575 practically repeats the codification of mutual military help, see Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkun den, 223–227; Szalay, Erdély, 256. transylvanian-ottoman military cooperation 313

‛ahdname.36 During the Long Turkish War at the end of the sixteenth century the Ottoman promises to support the principality were repeated, but there were also changes in the field of the obligations. On one hand, the ‛ahdname given to András Báthory in 1599 gave the prince—in a rather unconventional manner—a free hand to conclude a separate peace with the kings of Hungary; but it also included specific obligations for military cooperation: “Should Poles, Cossacks or other enemies attack Wallachia and Moldavia, the Transylvanians shall defend it if necessary.”37 The reverse regulation—that the two Romanian voievodates should give military support to Transylvania—appeared in the ‛ahdname given to Zsigmond Báthory when he returned to the side of the sultan in 1601.38 By the early seventeenth century the military obligations of the Principality of Transyl- vania to the Ottoman Empire were routine and taken for granted. When the Porte promoted Mózes Székely as a candidate for the princely title in 1603, he was not only promised some alleviation in tribute payment due to the devastations of the country in the war, but also that he would be excused from sending auxiliary troops to the Ottoman army.39 In the later period this question was discussed in general terms in the ‛ahdname of Gábor Bethlen from 1614. It was, however, concrete in explicitly discussing the various fields of military cooperation, such as the respective conditions of support for Transylvania and for the Ottoman Empire. Considerable attention was also drawn to the specific orientation toward the voievodates: [The prince of Transylvania] should have an understanding and live in a friendly manner with the voievods of Wallachia and Moldavia according to the obedience they owe to my Sublime Porte, they should hold good neighborhood in esteem, so that when any evil or an inimical intention should be heard from any direction, which could bring danger, if necessary, they

36 The text is only known in a sixteenth-century Hungarian translation, see Szalay, Erdély, 275; Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 230–231. The exact relevance of the term is, however, not clear, see Papp, “Keresztény vazallusok,” 78. 37 Gábor Kármán, “Báthori András ahdnáméja” [The ‛ahdname of András Báthori], Fons (Forráskiadás és Segédtudományok) 14 (2007): 348. The text is only known in sixteenth- century Hungarian translation. 38 “The voievods of Wallachia and Moldavia should, when the need occurs, help him with their entire armies.” The recent edition of the only surviving version (again, a sixteenth-century Hungarian translation) can be found in Papp, Die Verleihungs-, Bekräfti gungs- und Vertragsurkunden, 253. 39 Sándor Papp, “Székely Mózes erdélyi fejedelem hatalomra kerülésének diplomáciai tanulságai és egy nagyvezíri előterjesztés (telhis) keletkezése” [Conclusions for diplomacy from the process of the coming to power of Mózes Székely, prince of Transylvania, and the formation of a proposal (telhis) by the grand vizier], Aetas 14, no. 4. (1999), 79. 314 jános b. szabó

should support each other with their personal insurrection, and if there is no need for that, help each other by sending sufficient troops, and try to divert the ruin from each other.40 For the entire variety of obligations found in the documents concerning military cooperation between the principality and the empire, we note examples from the rule of the Szapolyai dynasty. Ever since King John of Hungary paid homage to the sultan in 1528, the Ottoman beys from the border were instructed to give him support.41 Victuals were provided by Transylvania to Ottoman armies in 1545, when the Sublime Porte planned a large-scale campaign in Hungary and requested supplies of forage to be sent to Buda in the value of 10,000 guldens tribute from the principality.42 In 1532 we even find examples of King John’s soldiers participating at the siege of Esztergom, in cooperation with minor Ottoman troops, but far from the main Ottoman army. This of course opened a “second front” for the sultan’s campaign directed against Vienna.43 There is, however, only a late example from the sometimes rather chaotic period from 1528 to 1571 for the intervention—in the interest of the sultan—of a Transylvanian prince into the affairs of the two Romanian voievodates. John Sigismund sent 1,000 cavalry and 1,000 infantry led by László Radák to Moldavia in order to help the struggle of Ştefan Tomşa against Ioan Iacob Heraclid (Despot-vodă). In order to maintain mutuality, two years later the Sublime

40 The seventeenth-century Hungarian translation is published in Erdélyi történelmi adatok [Transylvanian historical data], vol. 2, ed. Imre Mikó (Kolozsvár, 1856), 340–341. Bethlen sought to eliminate at least a part of these burdensome obligations in 1617 through his diplomats at the Sublime Porte. He hoped that in relation to Christian opponents, only the regulations for common defense should remain in force and he should not be forced to participate in offensive campaigns: “If a Christian country or prince had not offended His Highness, us or our country, His Highness should have no power to force us against our own will to attack such a country or prince with weapons.” Instructions of Bethlen for János Gáspár, his envoy in Istanbul (25 September 1617) Áron Szilády and Sándor Szilágyi, eds., TMÁO I, 187. 41 For the precedents in 1528, see Sándor Papp, “Hungary and the Ottoman Empire (From the Beginning to 1540),” in Fight Against the Turk in Central-Europe in the First Half of the 16th Century, ed. István Zombori (Budapest, 2004), 71. This way of giving Ottoman support must have been very frequent in the period between 1528 and 1571. The Ottoman armies stationed in Syrmia gave substantial assistance to John I in defending Buda in 1530, 1540, and 1541; he also received significant Ottoman aid in defeating the Habsburg coun- terattack after the failed Kőszeg campaign of Süleyman I in 1532. 42 Géza Dávid and Pál Fodor, eds., “Az ország ügye mindenek előtt való:” A szultáni tanács Magyarországra vonatkozó rendeletei (1544–1545, 1552) [“The affairs of the country stand before everything else”: The ordinances of the sultan’s council concerning Hungary] (Budapest, 2005), 18–21, 24–25. 43 Zoltán Korpás, V. Károly és Magyarország, 1526–1538 [Charles V and Hungary] (Buda- pest, 2008), 150, 232–233. transylvanian-ottoman military cooperation 315

Porte ordered the Moldavian voievod to send 5–6000 soldiers and the Wal- lachians another 5,000 to support John Sigismund’s wars in Hungary.44 This means that both the prospects and obligations included in the ‛ahdname-i hümayuns granted to Transylvanian princes after 1571 had their forerunners during the reign of the Szapolyai dynasty. Although their formulation over time became more precise, their content remained highly stable—as a matter of fact, their inclusion did not depend upon who was the actual prince of Transylvania.45

Designations of Transylvanian Troops to the Romanian voievodates

In Hungarian historiography it is well-known that from 1529 on the Sub- lime Porte frequently delegated the troops of the Romanian voievods to support the Szapolyai dynasty.46 Much less attention was devoted to the fact that Transylvania was not only relevant to the Sublime Porte’s expansion toward the west, but that the sultans also gave it an important role in the security system of the northern borders of the empire (the Black Sea region). Here, following Genoese and Venetian patterns, the Ottomans developed a maritime empire surrounded by a ring of vassal territories.47

44 Géza Dávid and Pál Fodor, eds., “Ez az ügy fölöttébb fontos:” A szultáni tanács Magyarországra vonatkozó rendeletei (1544–1545, 1552) [“This is a very important issue”: The ordinances of the sultan’s council concerning Hungary] (Budapest, 2009), 249, 256–257, 312, 316–318. See also the corresponding parts of Ovidiu Cristea’s chapter in this volume. Hungarian narrative sources suggest that the quick conclusion of the military activities and the capture of the voievod under siege were made possible by an agreement between the soldiers of Radák and the Hungarian mercenaries of the distressed Despot-vodă. See Ferenc Forgách, “Emlékirat Magyarország állapotáról Ferdinánd, János, Miksa királysága és II. János erdélyi fejedelemsége alatt” [Memorial about the state of Hungary under the rule of kings Ferdinand, John, and Maximillian, and the prince John II in Transylvania], in Humanista történetírók, ed. Péter Kulcsár (Budapest, 1977), 804–805; Istvánffy Miklós, Magyarok dolgairól írt históriája Tállyai Pál XVII. századi fordításában [The chronicle about the Hungarian things in Pál Tállyai’s seventeenth-century translation], vol. 1/1, ed. Péter Benits (Budapest, 2003), 345–346; Farkas Bethlen, Erdély története [The history of Transylvania], vol. 2, A váradi békekötéstől János Zsigmond haláláig (1538–1571) [From the peace of Várad until the death of John Sigismund], ed. András Bodor and Tamás Kruppa (Budapest and Kolozsvár, 2002), 238–239. 45 A final word on this question can, however, only be given after Sándor Papp pub- lishes the second part of his endeavor, the collection of seventeenth-century inauguration documents given to princes of Transylvania. 46 László Makkai, Magyar–román közös múlt [Common Hungarian–Romanian past] (Budapest, 1989), 89–93. 47 Gáspár Katkó, “Campaigns of Zaporozhian Cossacks against the Ottoman Empire between 1492 and 1642,” in Rzeczpospolita wobec Orientu w epoce nowożytnej, ed. Dariusz Milewski (Zarbrze, 2011), 37–43; Feridun M. Emecen, “Az Oszmán Birodalom sztyeppei 316 jános b. szabó

Beyond the Transylvanian interventions into the unstable political relations of the Romanian voievodates, Hungarian historiography in the main failed to explore the principality’s involvement in the management of the Ottoman Empire’s problems with the Ukraine Cossacks, who exerted pressure on the Crimea and Moldavia.48 Traditionally, these territories were in the Polish sphere of influence; the voievods were the kings’ vassals until the Ottomans pushed back the Rzeczpospolita from the Black Sea region.49 However, the unsettled conditions of Ukraine made it possible for various ambitious and violent local lords or Cossack leaders to aim beyond leading routine raids on the Ottoman borderlands, and rather try to achieve rule over Moldavia.50 In contrast to the traditional slow Polish armies, the Cossacks raiders’ troops were relatively small and quick, and by the end of the sixteenth century frequently used firearms. As their activities covered a large region it was very difficult to organize defenses against them.51 It was in this defensive system that Transylvania was given an important role, in part due to its geographical position and military capacity, and in part because of the advantageous political situation after the 1576 election of István Báthory as king of Poland.52 határai és a moldvai kijáró” [The steppe borders of the Ottoman Empire and the exit through Moldavia], Aetas 18, no, 2 (2003): 21–29. 48 Katkó, “Campaigns,” 45–46; Sándor Gebei, “A lengyel végek katonaparasztsága a 16–17. században” [The armored peasantry at the Polish borderlands in the sixteenth and seventeenth centuries], in Emlékkönyv Rácz István 70. születésnapjára, ed. Ágnes Kovács (Debrecen, 1999), 69–82. On the Transylvanian princes’ involvement with the political issues of Romanian voievodes, see Benda, “Erdély politikai kapcsolatai”; Makkai, Magyar– román közös múlt, 94–95. 49 Marek Plewczyński, “Söldnertruppen in Polen um die Wende des 15. und 16. Jahrhun- derts,” in Von Crécy bis Mohács: Kriegswesen im späten Mittelalter (1346–1526): Acta XXII. Kongress der Internationalen Kommission für Militärgeschichte, Wien, 9–13. September 1996 (Vienna, 1997), 345–349, 353–358; Ilona Czamańska, “Poland and Turkey in the 1st Half of the 16th Century—Turning Points,” in Fight against the Turk in Central-Europe in the first Half of the 16th Century, ed. István Zombori (Budapest, 2004), 91–102; Andrei Pippidi, “Mol- davie et Pologne: la fin de la vassalité,” Acta Poloniae Historica 83 (2001): 59–78. 50 The notorious Cossack leader from the mid-sixteenth century Ukraine, Dimitry Wiśniowiecki, also fell into Ottoman captivity when he interfered in the competition for the Moldavian throne in 1563. Forgách, “Emlékirat,” 799–803. 51 Mária Ivanics, “Kozák segédcsapatok Habsburg szolgálatban (1593–1606)” [Cossack auxiliary troops in the service of Habsburgs], in Az értelem bátorsága: Tanulmányok Perjés Géza emlékére, ed. Gábor Hausner (Budapest, 2005), 291–295. 52 Kazimierz Dopierała, Stosunki dyplomatyczne Polski z Turcją za Stefana Batorego [Poland’s diplomatic relations with Turkey under István Báthory] (Warsaw, 1986). See also a short English summary in Krzystof Wawryzniak, “Ottoman–Polish Diplomatic Relations in the Sixteenth Century” (MA diss., Bilkent Univerity, Ankara, 2003), 62–63 (http://www. thesis.bilkent.edu.tr/0002291.pdf, accessed 22 December 2011). The Sublime Porte could be transylvanian-ottoman military cooperation 317

Each second year a letter came from the Sublime Porte to Transylvania reminding its rulers to keep their troops ready for action, given the threat of turmoil due either to the changing voievods in Moldavia or Wallachia or the Cossack raids. It was in 1574 during István Báthory’s rule that for the first time a Transylvanian army was sent, according to the order to Moldavia, to defeat Voievod Ioan the Terrible (Ioan cel Cumplit), who had revolted against the sultan. Báthory sent Szekler troops to support the candidate of the Porte, and the united forces defeated Ioan in June of the same year.53 In 1577 a subject of King István Báthory, Ivan Pidkova (rumored to be a half-brother of Voievod Ioan) tried to establish rule over the voievodate. After the conquest of Iaşi, his Cossack army also defeated the Ottoman army coming to support the deposed voievod, Petru the Lame (Petru Şchiopul). When the news reached the Transylvanian diet at Torda in October, they ordained general mobilization for November–December. During the summer of the next year an army consisting of Szeklers and the princely mercenary troops was united with the Ottoman and Molda- vian troops, and besieged Pidkova at Iaşi. The Cossack leader was expelled from the country; he fled to Poland, where he was captured by István Báthory, who had him decapitated in Lemberg.54 The relevance of sending

certain that for Báthory as king of Poland (and thus ally of the Ottoman Empire) and the prince of Transylvania to maintain good relations, he would have to act in a determined manner against those who disturbed the peace. 53 Farkas Bethlen, Erdély története [The history of Transylvania], vol. 3, Báthory István trónra lépésétől Báthory Zsigmond uralkodásáig, 1571–1594 [From István Báthory’s succession to the throne till the reign of Zsigmond Báthory], ed. András Bodor and Tamás Kruppa (Budapest and Kolozsvár, 2004), 35. See also the orders of Sultan Selim II and the letter of Grand Vizier Sokollu Mehmed Pasha to István Báthory (April–May 1574) Sza- lay, Erdély, 146–148, and Ovidiu Cristea’s chapter in this volume. On the Szeklers of the Transylvanian army, see B. Szabó and Somogyi, Az Erdélyi Fejedelemség hadserege, 25–33, 105–106. The army was led by György Daczó and Mihály Rácz, the latter being a captain of the court infantry. See Attila Sunkó, “Az erdélyi fejedelmi testőrség archontológiája a XVI. században” [The archontology of the princely court guard in the sixteenth century], Fons (Forráskiadás és Segédtudományok) 1, no. 2 (1994): 102, 104. 54 Farkas Bethlen, Erdély története, vol. 3, 115–116. The decree of general mobilization: Sándor Szilágyi, ed., Erdélyi országgyűlési emlékek történeti bevezetésekkel [Documents of the diets of Transylvania, with a historical introduction], vol. 3, 1576–1596 (Budapest, 1877) [henceforth EOE III], 121. The order of Sultan Murad III to Kristóf Báthory (February 1878) and the report of Davut Bey on military activities: TMÁO I, 18–20. On the princes’ mercenary troops, see B. Szabó and Somogyi, Az Erdélyi Fejedelemség hadserege, 40–43, 107–108; Attila Sunkó, “Az Erdélyi Fejedelemség udvari hadai a XVI. században” [The court troops of the Principality of Transylvania in the sixteenth century], Levéltári Közlemények 69, nos. 1–2 (1998): 99–131. The leader of the troops this time was the king’s nephew, István Báthory Jr., but Mihály Rácz again played an important part in the supervision of the army. 318 jános b. szabó T ransylvania and the surrounding territories in the seventeenth century P rincipality of Map 1. T he transylvanian-ottoman military cooperation 319 mercenaries was highlighted by István Báthory on a later occasion, when an earlier brother-in-arms, Samuel Zborowski, who had in the meantime joined the Cossacks, invaded Moldavia in 1583. He cautioned the three governors of Transylvania from a distance: “Your Lordships should not send only Szeklers to such a place, because nothing else would follow than public damage and shame, but they should be mixed with mercenaries serving at the court, and an experienced and prudent person should be appointed as their leader.”55 In 1592 Prince Zsigmond Báthory had already received an order from Sultan Murad III to assist in the repression of Petru Şchiopul’s attempt, supported by the Cossacks, to regain lost power over Moldavia. The army consisted of 2,000 soldiers (1,000 cavalry and 1,000 infantry) and brought some field cannons with them.56 The prince ordered his captain to “try to enable our army to have its own ways, own quarters and not mix with foreign troops, so that there should be no conflict or concourse between them.”57 The battle between the armies of Petru and the principality took place near Iaşi, and according to Transylvanian sources the voievod was betrayed by his army. His cavalry fled at the beginning of the fight and his infantry (comprised mostly of Transylvanians) sounded a parley to Gáspár Sibrik, the leader of the prince’s army. Petru himself was captured and handed over to the Ottomans.58 The Long Turkish War at the end of the sixteenth century was a turning point in this kind of cooperation. The Ottomans became much more cautious in using the Transylvanian army for such services in the seventeenth century as Zsigmond Báthory managed to greatly increase his influence in both voievodates and in coalition with him, Moldavia and Wallachia lost allegiance to the sultan. In any case, in this period princes of Transylvania

55 István Báthory to the three governors (24 August 1583), Andrei Veress, ed., Docu mente privitoare la istoria Ardealului, Moldovei şi Ţării Româneşti [Documents concerning the history of Transylvania, Moldavia, and Wallachia], vol. 2, Acte şi scrisori (1573–1584) [Charters and letters] (Bucharest, 1930), 266. See also the order of Sultan Murad III to Voievod Zsigmond Báthory (22 June 1583), ibid., 251. 56 Of the cavalry, 200 were the mercenaries of the prince, while the rest came from the Szekler region of Háromszék; the infantry’s origins were Háromszék (446), Szeklers from Csík-, Gyergyó- and Kászonszék (354) and 200 riflemen from the Saxon town of Kronstadt. 57 Zsigmond Báthory’s instructions to Gáspár Sibrik (25 September 1592) EOE III, 412– 415. See also Farkas Bethlen, Erdély története, vol. 3, 229–230. For the Transylvanian Saxon troops, see B. Szabó and Somogyi, Az Erdélyi Fejedelemség hadserege, 36–39, 106–107. 58 Farkas Bethlen, Erdély története, vol. 3, 231. See also Zsigmond Báthory’s report on the victory to the sultan (2 October 1592), Karácson, ed., Török-magyar oklevéltár, 173. On Sibrik, see Sunkó, “A fejedelmi testőrség,” 200. 320 jános b. szabó regularly interfered in the struggles for the voivodates’ thrones, even with armed forces.59 Some of these endeavors, such as one in 1655, may have had the support of the Porte as well. After the death of Matei Basarab, voievod of Wallachia, a military uprising started in the country. Matei had a large army of mercenaries, the so called seimeni troops from the Balkans, whom his successor Constantin Şerban wanted to release from service and deprive of their traditional privileges.60 The military revolt turned into a popular uprising. When Constantin Şerban, after being held almost as hostage by the mercenaries, escaped to the pasha of Silistria, the rebels elected a popular boyar to be their new voievod, the Vornic Hrizea. These events seemed to become increasingly unpredictable, causing considerable anxiety in the neighboring countries. The deposed Constantin requested the help of Prince György Rákóczi II of Transylvania, whose intervention was not only supported by the pasha of Silistria, but for a while even by the Sublime Porte.61 In May 1655 the prince collected his forces under his personal command and was joined by Gheorghe Ştefan, voievod of Moldavia, who personally brought his troops to Wallachia. The war was won in one decisive victory between Şoplea and Ploieşti with the rebels suffering a crushing defeat. Rákóczi left 1,500 Transylvanian soldiers in support of Constantin Şerban and marched back home, taking thirty-two cannons as booty.62

59 During the rule of Gábor Bethlen, many such cases are known. See Lajos Demény, Bethlen Gábor és kora [Gábor Bethlen and his age] (Bucharest, 1982), 40–64. 60 For the uprising, see Lidia A. Demény, Lajos Demény, and Nicolae Stoicescu, Răscoala seimenilor sau răscoală populară? 1655, Ţară Românească [The revolt of the sei mens or a popular revolt? 1655, Wallachia] (Bucharest, 1968), 297–300 (English summary); Lajos Demény, “Die Rolle des Soldatenelements in den Volksbewegungen des 17. Jahrhun- derts: Der Charakter des Sejmen-Aufstandes in der Walachei im Jahre 1655,” in Aus der Geschichte der ostmitteleuropäischen Bauernbewegungen im 16–17. Jahrhundert: Vorträge der internationalen wissenschaftlichen Konferenz aus Anlass der 500. Wiederkehr der Geburt von György Dózsa, Budapest, 12–15. September 1972, ed. Gusztáv Heckenast (Budapest, 1977), 91–101; Lajos Thallóczy, “II. Rákóczy György és az oláh szemények” [György Rákóczi II and the Wallachian seimens], Századok 26 (1892): 449–456. 61 For the changes in the Porte’s verdict, see Kármán, Erdélyi külpolitika, 256–258; idem, “György Rákóczi II’s Attempt to Establish a Local Power Base among the Tributaries of the Ottoman Empire 1653–1657,” in Power and Influence in Southeastern Europe, 16th–19th Centuries, ed. Maria Baramova, et al. (Berlin, 2013), 229–243. 62 For the 1655 campaign to Wallachia, see Georg Kraus, Erdélyi krónika 1608–1665 [Transylvanian chronicle], ed. and trans. Sándor Vogel (Budapest, 1994), 233–235; János Szalárdi, Siralmas magyar krónikája [Hungarian chronicle of laments], ed. Ferenc Szakály (Budapest, 1980), 337–341; Anonymous, Rákóczi-eposz [The Rákóczi epic], ed. Csaba Szigeti (Budapest, 1988), 76–108. For the battle, see György Rákóczi II to Zsuzsanna Lorántffy (9 July 1655), Sándor Szilágyi, ed., A két Rákóczy György családi levelezése [Family letters transylvanian-ottoman military cooperation 321

The Transylvanian Role in the Ottoman–Polish Conflicts

In the Long Turkish War that began in the early 1590s Transylvania and the Polish–Lithuanian Commonwealth took different stances. Under the rule of Zsigmond Báthory, the principality turned against the Porte and joined the Christian coalition together with the two Romanian voievodates. The Rzeczpospolita, on the other hand, followed the tradition of István Báthory’s politics and remained neutral. In 1595, by taking advantage of the war and the way it preoccupied the two countries, the Rzecz- pospolita was able to extend its sphere of influence over Moldavia at the expense of Transylvania. From this time on, the voievodate functioned as a triple condominium, causing many conflicts between the three powers— the Ottomans, the Crimean Tatars, and the Commonwealth.63 It was at a later phase of this series of conflicts that the principality was ordered to send support to the sultan’s armies. In 1615, rebelling boyars, with support from Polish lords from the borderlands, sought to install Alexander Movilă upon the throne. Gábor Bethlen received an order from the Porte in October the same year to help the earlier voievod, Ştefan Tomşa. Bethlen, however, also favored Movilă and was not eager to obey the order. When the latter’s armies defeated Ştefan Tomşa, the prince of Transylvania was accused of treason at the Sublime Porte.64 As the attempt of Sarhoş İbrahim Pasha to support the expelled voievod failed, the Porte appointed Iskender, the beylerbeyi of Bosnia, to be the chief general (serdar) of the armies sent to Moldavia. He mustered his armies at Temesvár and called the prince of Transylvania to send his troops as well.65 Bethlen was in a tight corner, as he was at the same time threatened by an attack from the Kingdom of Hungary, and also pressured by the Porte to hand over the castle of Lippa. He decided to send a smaller auxiliary troop to the serdar’s army, he added to the beylerbeyi of Temesvár’s 500 cavalry

of the two princes György Rákóczi], Monumenta Hungariae Historica: Diplomataria, 24 (Budapest, 1875), 494. 63 For the entry of Transylvania into the war, see Horn, Báthory András, 178–184, 189– 191. For the politics connected to the Polish Chancellor Jan Zamoyski, see Mária Ivanics, A Krími Kánság a tizenöt éves háborúban [The Crimean khanate in the Fifteen Years War 1595–1606] (Budapest, 1994), 83–88; Gábor Várkonyi: “Angol békeközvetítés és lengyel– török tárgyalások a tizenöt éves háború időszakában” [English mediation and Polish– Ottoman negotiations during the Fifteen Years War], Aetas 18, no. 2 (2003): 44–62. 64 Demény, Bethlen, 44–45; Sudár, “Iszkender és Bethlen Gábor,” 984; Katkó, “Cam- paigns,” 47–50. 65 In July 1616 the Porte ordered that a 19,000 man strong Transylvanian army be sent to Moldavia, see TMÁO I: 169. 322 jános b. szabó armed with pikes another 500 from his own military and then used his main army to lay siege to his own Lippa castle, in order to expedite the handover. Then, on learning of Iskender Pasha’s victory on 3 August, he was able to use the excuse of the threat of György Homonnai Drugeth (a pretender of his throne from Hungary) to withdraw his troops from the Transylvanian–Moldavian border.66 Iskender—who in the meantime had been appointed pasha of Silistria— had to bring his armies into the voievodate again in the next year, as the Polish–Lithuanian Commonwealth did not easily cede its influence on Moldavia. Bethlen had already received an order to join the Moldavian campaign in April, and after long preparations and preliminary peace talks, on 29 August he crossed the border with approximately 12,000 soldiers.67 They arrived in Suceava on 12 September, and with his mediation the peace of Busza was concluded on 23 September, after an eight-day negotiation.68 In a few years, the intervention of the prince of Transylvania in Molda- vian and Polish affairs was requested again, but this time he managed to avoid sending his troops. In 1620, Gaspar Graziani, the voievod of Molda- via, lost the Porte’s favor and was replaced—a decision he was not ready to accept. He invited Polish troops into the country but his armies suffered a crushing defeat at Cecora on 5–6 October from Iskender Pasha.69 Although Graziani was his personal adversary, Bethlen contradicted the order of the Porte to send his troops to the Polish front with the excuse of his military involvement in Hungary (where he participated in the first phase of the Thirty Years War).70 He used the same reasoning the next

66 Sudár, “Iszkender és Bethlen Gábor,” 984–986; Demény, Bethlen, 107. According to the chronicle of Máté Szepsi Lackó, the Transylvanian armored cavalry, armed with pikes, played a decisive role in Iskender’s success. See Erdélyi történelmi adatok [Transylvanian historical data], vol. 3, ed. Imre Mikó (Kolozsvár, 1858), 161–162. 67 The Ottoman army sent to Moldavia was at the time only 4,000 men strong, with 8–9,000 Tatars from the Bucak. See Bethlen’s letter to János Gáspár (25 September 1617) TMÁO I: 185. 68 Sudár, “Iszkender és Bethlen Gábor,” 987; Demény, Bethlen, 107; Dariusz Kołodziejczyk, Ottoman–Polish Diplomatic Relations (15th–18th Century): An Annotated Edition of ‘Ahd names and Other Documents (Leiden, 1999), 345–353. 69 Demény, Bethlen, 55–56; Gebei, Az erdélyi fejedelmek, 82; Manfred Stoy, “Gas- par Graţiani, Fürst der Moldau 1619–1620: Seine marginale Rolle in den Anfängen des Dreißigjährigen Krieges,” Mitteilungen des Instituts für Österreichische Geschichtsforschung 112 (2004): 306–315. 70 Demény, Bethlen, 55; Sudár, “Iszkender és Bethlen Gábor,” 992. See also the order of Sultan Osman II to the governor of Transylvania, István Bethlen (1620), TMÁO I, 251, and Gábor Bethlen’s letter to the judge of Beszterce (20 July 1620) Nicolae Iorga, ed., Documente transylvanian-ottoman military cooperation 323 year to persuade the Porte to be content with his sending 600 carriages of victuals to the camp of Sultan Osman II, who faced acute logistical problems during his ultimately failed attempts to defeat the Polish army entrenched near Chocim. The victuals actually sent were even less than promised, and did not exceed 560 carriages.71 In 1633, animosity intensified again between the Ottomans and the Polish–Lithuanian Commonwealth, a result of the increasingly audacious raids of the Cossacks from Ukraine. Abaza Hasan, pasha of Silistria, tried to exploit the situation of the Rzeczpospolita’s war against Muscovy, but his first attempt did not bring the expected results. With the Polish victory at Smolensk the following year, the probability of an easy victory disappeared; originally a great campaign with the sultan’s participation was planned, but the Ottomans decided not to risk opening an additional second front given their ongoing war against Persia. A peace treaty was thus concluded with the Polish king.72 In any case, preparations for the war had included the mobilization of the Transylvanian army as well. In March 1634 an order was sent to Prince György Rákóczi I, who did not dare to resist. Although he dedicated much energy to the mediation between the two powers, he kept his army ready between 15 May and mid-October, despite formidable logistical problems during the summer due to drought and plague.73 The conclusion of the Polish–Russian peace in 1667 opened a new period of conflict between the Ottoman Empire and the Polish–Lithuanian Com- monwealth. The Rzeczpospolita’s attention could now again be turned privitoare la istoria românilor [Documents on the history of the Romanians], vol. 15, Acta şi scrisori din arhivele oraşelor ardelene (Bistriţa, Braşov, Sibiu) [Documents and letters from the archives of Transylvainan towns (Bistriţa, Braşov, Sibiu)] part 2 (1601–1825) (Bucharest, 1913), 868. 71 Demény, Bethlen, 53–56, 115–116; Gebei, Az erdélyi fejedelmek, 81–84; Sudár, “Iszkender és Bethlen Gábor,” 992. For the adventurous journey of the caravan of carriages, see the memorial of Ferenc Nagy Szabó from Marosvásárhely in Erdélyi történelmi adatok [Transyl- vanian historical data], vol. 1, ed. Imre Mikó (Kolozsvár, 1855) [henceforth ETA I], 128–139. The Szekler soldiers in the caravan’s retinue were also used for engineering tasks, and on their way they quickly repaired one of the damaged Ottoman bridges in Moldavia, “as they had all the equipment [axes and drills] for it,” as the eyewitness put it. 72 Kołodziejczyk, Ottoman–Polish Diplomatic Relations, 432–436. 73 The considerations of the prince’s council were summarized in a contemporary chronicle as follows: “This issue seemed to be very grave to the prince as well as to his councilors, mostly because this warfare was supposed to be directed against a Christian country that had been good neighbors. However, after lengthy deliberations it was decided that offending the Porte should be avoided at any cost, and that it is necessary to show and eagerly maintain fidelity,” see Szalárdi, Siralmas magyar krónikája, 157. On the background, see also Kraus, Erdélyi krónika, 145–146; ETA I, 146. 324 jános b. szabó toward the Cossacks of Ukraine. The Crimean Khanate, afraid of their cooperation with its Christian neighbors, managed to involve the sultan in the conflict. At one peak in the extended warfare Sultan Mehmed IV himself led a campaign to conquer Kamieniec Podolski, the most important Polish castle in the border zone. This time the Sublime Porte referred back to the example of Bethlen: “as when my divine predecessors waged war in these regions, the princes of our hereditary country, Transylvania, regularly sent 7–800 carriages filled with barley and flour as an assistance to the Turkish army,” and so he ordered 600 carriages of victuals to be sent to the Ottoman camp—a number which was later mercifully reduced to 500.74 After the successful siege, the Polish diplomats invited the prince of Transylvania to be a mediator at the peace talks. They later tried to turn Mihály Apafi I to their side in the conflict; he even appeared among the candidates in the royal election in 1674.75 In the same year, Apafi was once more ordered to deliver 600 carriages of victuals to another campaign led by the sultan against the Polish territories.76

The Role of Transylvania in the Hungarian Theater of War

After the peace of Adrianople in 1568, which put an end to the long wars in Hungary, it was only in the beginning of the 1590s that the idea reap- peared that the weapons of the Ottoman Empire should be turned against Central Europe. Until the middle of 1593, however, it was not clear whether the war against Christian powers should start on land or sea. Due to the hesitation of the Ottoman dignitaries, the army under Grand Vizier Sinan

74 János Bethlen, Erdély története 1629–1673 [The history of Transylvania], trans. Judit Vásárhelyi, ed. József Jankovics (Budapest, 1993), 465–473. See also the ordinance distrib- uting the obligation among the Transylvanian administrative regions: Sándor Szilágyi, ed., Erdélyi országgyűlési emlékek történeti bevezetésekkel [Documents of the diets of Tran- sylvania, with a historical introduction], vol. 15, 1669–1676 (Budapest, 1892) [henceforth EOE XV] 279–280. For the conflict generally, see Vojtech Kopčan, “A török Porta Thököly- politikája” [The politics of the Turkish Porte concerning Thököly], in A Thököly-felkelés és kora, ed. László Benczédi (Budapest, 1983), 121 122; Peter Bartl, “Der Kosakenstaat und das Osmanische Reich im 17. und in der ersten Hälfte des 18. Jahrhunderts,” Südost-Forschungen 33 (1974): 183–188; Ekkehard Eickhoff, Venedig, Wien und die Osmanen: Umbruch in Südost europa, 1645–1700 (Munich, 1970), 260–279; as well as Victor Ostapchuk’s chapter in this volume. 75 János Bethlen, Erdély története, 528, 561; Gebei, Az erdélyi fejedelmek, 227. 76 EOE XV, 385–387. transylvanian-ottoman military cooperation 325 could not start its march toward Hungary until 29 July 1593.77 The order of Sultan Murad III that the Transylvanian army should join the attack against the Habsburg-ruled country also came quite late, but included further demands: 100,000 kiles of wheat flour, the same amount of barley, a large quantity of honey, butter, and salt; eventually a large number of pikes, cannons with powder, equipment, and personnel for trail blazing. Also, the sultan demanded that Transylvanians should be obliged under penalty of death to join the army and be forbidden to take up service in the troops of the king of Hungary.78 At their diet in September the Transylvanian estates discussed what sort of position they should take in this unprecedented situation. The dignitaries accepted the advice of the Polish Chancellor Jan Zamoyski, and adopted the tactic of stalling for time and waiting for developments. They sent an envoy to the grand vizier’s camp and petitioned for a reduction in demands. Support was also sought from the English ambassador at the Sublime Porte, for they claimed that Transylvania should be granted an exemption from warfare against Christian powers.79 According to the Ottomans’ operational plans, the Transylvanian army, supported by the beylerbeyi of Temesvár, should open a second front in Northern Hungary. Serious attempts were being made to convince the Hungarian magnates in that territory to pay homage to the sultan, in effect to establish another vassal state similar to Transylvania.80 The prince did not yet dare to openly oppose the sultan’s order, and personally joined

77 Pál Fodor, “Prelude to the Long War (1593–1606): Some Notes on the Ottoman Foreign Policy in 1591–1593,” in The Great Ottoman Turkish Civilization, ed. Kemal Çiçek (Istanbul, 2000), 297 301. 78 Farkas Bethlen, Erdély története, vol. 3, 221–222. 79 Ibid., 237–248; 263–265. The instructions for the envoy to England have been published by Andrei Veress, ed. Documente privitoare la istoria Ardealului, Moldovei şi Ţării Româneşti [Documents concerning the history of Transylvania, Moldavia, and Wallachia], vol. 4, Acte şi scrisori (1593–1595) [Charters and letters] (Bucharest, 1932), 25–28. See also Fodor, “Prelude”; Gábor Várkonyi, “A konstantinápolyi angol politika a tizenötéves háború időszakában: Edward Barton angol portai követ jelentései Konstantinápolyból, 1593–1597” [English politics in Constantinople during the Fifteen Years War: The reports of Edward Barton, English Ambassador at the Porte from Constantinople, 1593–1597], Aetas, 15, no. 4. (2000): 119–120. 80 Sándor László Tóth, A mezőkeresztesi csata és a tizenöt éves háború [The battle of Mezőkeresztes and the Fifteen Years War] (Szeged, 2000), 132–133, 136; Farkas Bethlen, Erdély története, vol. 3, 236. See also the 1593 order of Murad III concerning Transylvania’s entry to the war, EOE III, 295. Similar joint warfare between the ruler of Transylvania and the beylerbeyi of Temesvár had already taken place in 1566. The aims were, just as in the earlier case, territories once under the rule of the Szapolyai dynasty: Ecsed, Nagykálló, Tokaj, and Kassa. 326 jános b. szabó his troops in October. However, he managed to keep his armies in the Partium, under the pretext that he sought to defend against a likely attack on his lands from Hungary.81 At the end of the Ottoman campaign, the grand vizier gave his permission to the Transylvanians to return home. When the military successes of the Christian armies during the winter caused a great anxiety in Ottoman Hungary, Zsigmond Báthory excused himself by stating that the troops were already in their winter quarters and unable to interfere.82 The following year the Sublime Porte again notified the prince to “prepare himself (. . .) with sixty thousand men and well-equipped ships (. . .) till the trees start to blossom (. . .) and be ready to march if the situation requires or the sultan orders it.”83 There were even attempts to threaten the hesitating Transylvanians with the advance of the Crimean Tatars. Although Zsigmond Báthory failed to join the Ottoman campaign in 1594, it was not until early the next year that the Sublime Porte became certain that Transylvania and the two Romanian voievodates had not only backed out of the war, but even changed sides and joined the Christian coalition in the unfolding military conflict.84 The end of the Long Turkish War in 1606 brought with it a long period of peace—at least officially—to the Kingdom of Hungary. But when, in 1663, the prolongation of the peace between the Sublime Porte and Leopold I proved impossible, despite long negotiations, the armies of the Ottoman Empire again invaded the country. Grand Vizier Köprülü Ahmed marched with a sizeable army to conquer Érsekújvár and ordered the prince of Transylvania, Mihály Apafi I, to join him with his troops.85 After the

81 Farkas Bethlen, Erdély története, vol. 3, 267–268; EOE III: 300. 82 Farkas Bethlen, Erdély története, vol. 3, 270, 274–276. 83 Ibid., 270. 84 The Porte tried to mobilize the Transylvanians again during the summer of 1594. See the letters of Sultan Murad III to Zsigmond Báthory in Karácson, ed., Török-magyar oklevéltár, 175–179. Pál Fodor, “A váci ‘harmincad’ és egy hódoltsági főember a 16. század- ban: Oszmán aga, cselebi és bég” [The custom duty of Vác and a dignitary from sixteenth- century Ottoman Hungary: Osman Ağa, Çelebi and Bey], Történelmi Szemle 43 (2001): 338; Tamás Kruppa, “Erdély és a Porta 1594–1597. évi békealkudozásainak történetéhez” [On the negotiations between Transylvania and the Sublime Porte in 1594–1597], Századok 137 (2003): 604–607; Ivanics, A Krími Kánság, 68–74; Géza Dávid and Pál Fodor, “Magyar vonatkozású török államiratok a tizenötéves háború korából” [Ottoman state papers related to Hungary from the period of the Fifteen Years War], Hadtörténelmi Közlemények, n.s. 30, no. 2 (1983): 283. 85 Ferenc Tóth, Saint Gotthard 1664: une bataille Européenne (Paris, 2007), 43–54; Ágnes R. Várkonyi, “The Principatus Transylvaniae and the Genesis of the Anti-Turkish Alliance,” in Études historiques hongroises 1985: Publiees a l’occasion du XVIe Congres international des transylvanian-ottoman military cooperation 327

previous five years of bloody wars, the principality had no significant military power, and its surviving capacity was bound by the imperial troops stationed in the country. The grand vizier took this weakness, along with the financially depleted state of Transylvania, into account and agreed that the prince should join him with only his personal retinue, a cavalry of 100 men and eighteen courtiers.86 Apafi waited until 10,000 Crimean Tatars and troops of the Moldavian and Wallachian voievods (1,000/600 infantry and 3,000/5,000 cavalry) crossed the country, and then marched with his retinue through Ottoman territory to the camp at Érsekújvár. When they reached the fortress on 18 October it had already fallen. His military presence was thus inconse- quential, but the Ottomans tried to make use of him otherwise, publishing manifests in his name promising full freedom and security of property for the inhabitants of the country if they were ready to accept Ottoman sovereignty. The effect of this contribution, however, was counterbalanced by the prince’s secret declaration to the dignitaries of the Kingdom that he did not aim to become a puppet king of Hungary in the sultan’s service. In December at the close of the grand vizier’s campaign, Apafi was allowed to return to his country.87 In the course of the counteroffensive during the winter, the Transylvanian army managed to miss the relief of the castle Nyitra, and remained in the country in 1664 with the excuse of an attack upon Várad and a raid of hajdú troops on Transylvanian soil (which had been staged for this purpose).88 The anti-Ottoman wars of 1663–1664 ended in failure for the Habsburg Empire, and afterward, in Hungary, there was growing discontent with Leopold I that peaked in an armed uprising in the 1670s. Transylvania tried to support and control this movement with its modest means, but in the early 1680s the leading role of the anti-Habsburg activities was taken over from the principality’s representatives by the so-called kuruc leaders

sciences historiques par le Comite national des historiens hongrois, ed. Ferenc Glatz, et al. (Budapest, 1985), 593–612. The letters calling Apafi to the Ottoman camp are published in János Bethlen, Erdély története, 152–153, 162–163. 86 Kraus, Erdélyi krónika, 565. 87 Ágnes R. Várkonyi, “The Last Decades of the Independent Principality (1660–1711),” in History of Transylvania, vol. 2, From 1606 to 1830, ed. László Makkai and Zoltán Szász (New York, 2002), 252–254. See also János Bethlen, Erdély története, 156–157. 88 János Bethlen, Erdély története, 215–222; Zsolt Trócsányi, Teleki Mihály (Erdély és a kurucmozgalom 1690-ig) [Mihály Teleki: Transylvania and the kuruc movement until 1690] (Budapest, 1972), 44; Sándor László Tóth, “Kücsük Mehmed magyarországi tevéke- nységéhez” [To the activities of Küçük Mehmed Pasha in Hungary], Aetas 17, no. 4 (1999): 93–94. 328 jános b. szabó from the group of Hungarian exiles. By this time, the Sublime Porte no longer saw in Mihály Apafi I, prince of Transylvania, his potential as an instrument to be used against the Habsburg Empire, but rather turned to the young rebel leader Imre Thököly.89 This change of preference was formalized by the sultan in 1682, with the withdrawal of the title of chief leader (serasker or serdar) granted to Apafi the year before; on 16 August 1682 in the camp at Fülek, Thököly was appointed “king” of a newly founded vassal state, the “Orta Macar” (Middle Hungary).90 In the campaign led by İbrahim, beylerbeyi of Buda, an Otto- man army of 17–20,000 (according to others 25–30,000) men, 10–12,000 Hungarian kurucs of Thököly, and 5–8,000 soldiers from Transylvania took part.91 The latter marched through Ottoman territory to the border castle of Fülek, and joined the ongoing siege. For the final charge, the serasker requested from the Transylvanians 2,000 infantry, 500 men to work in the trenches, and powder and balls for twelve cannons. They were also commissioned to destroy the walls of the fortress after its conquest by the Ottomans, after which the Transylvanian army returned home.92 In the following year, the Ottoman Empire resolved to deal a decisive blow to the Habsburgs: Grand Vizier Kara Mustafa mobilized the forces of the European vassal states and led his army against Vienna.93 Prince

89 János J. Varga, “Varianten der türkischen Orientierung in Ungarn: Wesselényi— Apafi—Thököly 1663–1683,” in The First Millennium of Hungary in Europe, ed. Klára Papp and János Barta (Debrecen, 2002), 208–215. 90 On the term Orta Macar, see János J. Varga, “Az ‘Orta Madzsar’ szerepe Perényi Pétertõl Thököly Imréig: A nyugati irányú török hódítás metodikájához” [The role of the ‘Orta Macar’ from Péter Perényi to Imre Thököly: Comments on the methods of the Turk- ish expansion toward the west], in Tanulmányok Szakály Ferenc emlékére, ed. Pál Fodor, Géza Pálffy, and István György Tóth (Budapest, 2002), 415–423; also Sándor Papp’s chapter in this volume. The document of appointment was recently edited by Sándor Papp, “Thököly Imrének és Közép-Magyarország népének kiállított szultáni szerződéslevél / İmre Thököly ve Orta Macaristan halkına verilen ahidname-i hümayun” [Treaty letter granted by the sultan to Imre Thököly and the people of Middle Hungary], in István Seres, Thököly Imre és Törökország/Imre Thököly ve Türkiye (Budapest, 2006), 278–289. 91 At the beginning of August there were only 3,000 Szeklers, as well as 700 infantry and 700 cavalry from the Hungarian counties of Transylvania present at the prince’s camp; Apafi was, therefore, worried about how he would manage to collect 5,000 able soldiers. It was this army that left Szamosújvár on 5 August, but on the 10–11 it was joined by eleven battalions of Hungarian kuruc rebels, ten battalions of Hungarian hajdús, and two compa- nies of German mercenaries. This was how their numbers reached 8,000. See Trócsányi, Teleki Mihály, 272. 92 János J. Varga, Válaszúton: Thököly Imre és Magyaroszág 1682–1684-ben [At the cross- roads: Imre Thököly and Hungary in 1682–1684] (Budapest, 2007), 27–36, 175–176. 93 See Thomas M. Barker, Double Eagle and Crescent: Vienna’s Second Turkish Siege and Its Historical Setting (Albany, NY, 1967). transylvanian-ottoman military cooperation 329

Mihály Apafi ordained that the insurrection of the army should begin as early as 23 February at Segesvár, and expected everyone to appear with good garments, weapons, and horses, “unlike last year and before, with wooden stirrups and in moccasins, because this time we shall have to meet our lord the grand vizier.”94 The Sublime Porte also ordered the principality to send 800 carriages of victuals to Belgrade; the Transylvanian diplomats, however, managed to convince them to reduce this demand in exchange for the participation of the country’s troops in the campaign and a contribution of 10,000 thalers.95 The army of the Crimean Tatars, as well as the auxiliary troops of Mol- davians and Wallachians (2,000, respectively 4,000 men strong) again used the route through Transylvania to the theater of war in Hungary, causing massive problems.96 Once again it was only after their departure that the Transylvanian army of 6,000 men started their march on 8 July. They traversed Ottoman Hungary and then followed the Danube, arriving on 19 August at the encampment of the main Ottoman army at Bruck. The prince visited Grand Vizier Kara Mustafa on 22 August and was commissioned to blockade the fortress of Győr, which had a key position in the supply of reserves; together with Mustafa Pasha of Silistria guarded the bridges on the rivers Rába and Rábca. Thus Kara Mustafa managed to reposition the elite troops of İbrahim, beylerbeyi of Buda to Vienna, and keep his little-trusted Transylvanian vassal from making trouble. This was all the more reasonable, as the secret diplomacy of the prince throughout the entire campaign reveals that he was searching for ways to join the Christian powers.97 Thus, after the relief of Vienna by Jan Sobieski, the Transylvanian troops withdrew, relatively undisturbed, to Buda as a rearguard for the retreating Ottoman army. From there, on 13 October, the grand vizier released Apafi’s troops to return home, but they suffered from epidemics during the encampment in Hungary. When they finally

94 Sándor Szilágyi, ed., Erdélyi országgyűlési emlékek történeti bevezetésekkel [Docu- ments of the diets of Transylvania, with a historical introduction], vol. 18, 1683–1686 (Buda- pest, 1895), 95. 95 Varga, Válaszúton, 132. 96 Maria Matilda Alexandrescu-Dersca-Bulgaru, “Die Rumänen bei der Belagerung Wiens,” in Die Rumänen und die Belagerung Wiens, 1683: Aufsätze, ed. Nicolae Stoicescu, et al. (Bucharest, 1983), 17–29. 97 Varga, Válaszúton, 132–135. In secret, Apafi had already recommended himself as a mediator for peace negotiations to Emperor Leopold I through Pál Esterházy, the Palatine of Hungary. See Trócsányi, Teleki Mihály, 278–280. 330 jános b. szabó arrived back on Transylvanian soil at Déva on 4 November almost half of the soldiers had expired.98

Auxiliary Troops of the Ottoman Empire at the Princes’ Side

It is evident from the example of the Romanian voievodates that the rulers and pretenders of the vassal states who had enjoyed the trust of the Sublime Porte could count upon reciprocal military assistance from the empire. It sent the neighboring vassals as well as border troops for support when necessary. Beside the demands made on the princes, from 1572 to 1688 there was also a continuity of military help given to the Szapolyai dynasty under the rule of their successors. The Sublime Porte did acknowledge the right of the Transylvanian diet to freely elect their prince, and confirmed the rule of István Báthory, who had been appointed by the estates’ assembly. However, another candidate, Gáspár Bekes, also enjoyed some Ottoman support, in addition to the favors of the Habsburg court in Vienna. But when Bekes opted for an armed revolt in 1575, the Sublime Porte ordered its forces in Hungary and the voievods of Moldavia and Wallachia to assist Báthory.99 The size of the forces reported in the sources varies, but even if they were limited, the narrative sources note the bravery of the Turks in the decisive battle of Kerelőszentpál (8 July 1575). In addition, the diversions they started blocked politicians from the Kingdom of Hungary from sending military support to Bekes.100 In 1611 there was a further case in which the rule of the prince confirmed by the Sublime Porte was seriously endangered. From the Kingdom of Hungary, Bálint Homonnai Drugeth, who enjoyed the support of certain Ottoman politicians, raised his flag as a pretender against Gábor Báthory.

98 Ibid., 197–198. 99 Horn, Hit és hatalom, 87–97. 100 Narrative sources claim that the Ottoman auxiliary army consisted of 400 soldiers sent by the pasha of Temesvár, and 200 light cavalry from Wallachia. According to the reward registers, the numbers of the Ottoman soldiers should have been up to 1,000–1,500, and the order of the Porte prescribed the Moldavian and Wallachian voievods to send troops of 5,000 each. See Farkas Bethlen, Erdély története, vol. 3, 49, 56; Istvánffy, Magyarok dolgairól, vol. 3, 27–28; Szalay, Erdély, 227–229; Pál Fodor, “Making a Living on the Fron- tiers: Volunteers in the Sixteenth-Century Ottoman Army,” in Ottomans, Hungarians, and Habsburgs in Central Europe: The Military Confines in the Era of Ottoman Conquest, ed. Géza Dávid and Pál Fodor (Leiden, 2000), 229–263. During the diversions, Ottoman troops conquered the castles of Kékkő, Divény, and Somoskő, belonging to Bekes’ supporter, János Balassa. transylvanian-ottoman military cooperation 331

Báthory was also threatened by the former voievod of Wallachia, Radu Şerban, as well as by his own Saxon subjects. His enemies managed to confine him to the town of Nagyszeben, from which the prince asked for Ottoman support.101 According to the chronicle of Tamás Borsos, “there came 30–40,000 Turks from the direction of Wallachia under Ömer Pasha for Báthory’s help, and 10,000 Turks from Temesvár under Zülfikar Pasha.”102 No battle took place because the invaders fled upon learning of the Otto- man support, and after five days Ömer Pasha’s army was sent away from Transylvanian territory by the prince, because of the severe damage they had done during this time.103 In 1613 a new phase of Ottoman support for Transylvania began with the rule of Gábor Bethlen. The prince who had seized power with the help of Ottoman armies recognized that he was highly unpopular at the Viennese court, and to counter designs against him tried to convince the Porte to refresh its Western expansionist policies.104 After 1618 he quickly joined the anti-Habsburg struggle of the Bohemian estates, and wished to involve the Ottoman Empire in the conflict as well, but initially without success. It was, however, a great loss of prestige for him in Octo- ber 1620 when Karakaş Mehmed, the beylerbeyi of Buda, annexed Vác into Ottoman territory.105 In 1614 Bethlen still believed his armies were not prepared to march together with the Ottomans into an anti-Habsburg war, so he only asked

101 Papp, “Homonnai”; Péter Vörös, “Forgách Zsigmond erdélyi expedíciója 1611-ben” [Zsigmond Forgách’s expedition to Transylvania in 1611], Hadtörténelmi Közlemények 124 (2011): 864–887. 102 Tamás Borsos, Vásárhelytől a Fényes Portáig [From Vásárhely to the Sublime Porte], ed. László Kocziány (Bucharest, 1972), 59. 103 Sándor Mika, Weiss Mihály (Budapest, 1893), 141–143. 104 See the detailed plans for an Ottoman attack against the Habsburg countries in Gábor Bethlen’s instructions to his envoy to the Sublime Porte: Sándor Szilágyi, ed. Beth len Gábor kiadatlan politikai levelei [The unpublished political letters of Gábor Bethlen] (Budapest, 1879), 15–21. At the beginning of his rule the prince would have been more content if a cavalry of 2,000 Ottomans, 2,000 Romanians, and 1,000 Tatars, as well as 1,000 Ottoman infantry were left in his command, ibid., 65. On the Habsburg attempts to depose Bethlen, see Oborni, “Bethlen Gábor”; Zsuzsanna Cziráki, “ ‘Senkinek pénzen vött rabjai nem voltunk sem nem vagyunk’: Brassó és a fejedelmi hatalom viszonya a város fejedelmi szolgáltatásai tükrében, Bethlen Gábor uralkodása idején” [“We are no slaves of anyone”: The relationship between Brassó and the princely power as reflected in the town’s services under the rule of Gábor Bethlen], (PhD diss., Szeged, 2010), 109–152. 105 The Hungarian castle had secretly been promised by both sides, Bethlen and the Habsburgs, in exchange for Ottoman support. See Sándor Papp, “Bethlen Gábor, a Magyar Királyság és a Porta (1619–1621)” [Gábor Bethlen, the Kingdom of Hungary and the Porte], Századok 145 (2011): 953, 958; Sudár, “Iszkender és Bethlen Gábor,” 992–994. 332 jános b. szabó for financial support from the Sublime Porte. By 1620–1621 the deterioration of his military situation motivated him to ask for more direct assistance, and he managed to gain some imperial troops in early 1621. In October of the same year he also obtained explicit authorization for the beylerbeyis of Eger and Temesvár to help him.106 The few hundred cavalry coming from Ottoman Hungary played no significant role in the military activities, which were already winding down. Overall, Bethlen was able to use the visible Ottoman presence as an excellent means of putting pressure upon the court in Vienna.107 Bethlen came with a novel idea—to invite not only the Ottoman, but also the Romanian and Tatar troops to participate in the Transylvanian military activities in Hungary; this was unknown since the time of the Szapolyais. His letter, sent with the Tatar envoys to Khan Canibek Giray II, caused a considerable uproar, as the envoys were captured by Hungarian lords, who published Bethlen’s offer to send 10,000 florins in exchange for 15,000 Tatar cavalry.108 Although the prince’s reputation was seriously damaged by his “fraternizing with the pagans,” he refused to alter the plan and repeated the offer once again. As a result, 3,000 Tatars came to North- ern Hungary from the Ottoman camp in Moldavia during the autumn of 1621. In the winter of the same year another 10,000 Nogay Tatars were positioned to follow under Kantemir Mirza. The prince, however, did not want to risk damage to his own country as a result of their march, and he needed the Tatars less, as peace negotiations had already begun with Ferdinand II. This unfolding context led him to allow the blockade of the mountain passes of the Carpathians, thus shutting the Nogays out of Transylvania.109 During Bethlen’s next campaign in Hungary in 1623,

106 TMÁO I, 316–317, see Papp, “Bethlen,” 963–968. His attempts to gain financial support in 1614 had all the more chance as the Ottoman Empire had already paid large sums for István Bocskai’s soldiers in 1605–1606. See Nagy, “A XVII. századi Habsburg-ellenes függetlenségi harcok,” 216. 107 Papp, “Bethlen,” 956–960. See also Gábor Bethlen’s letters to his wife Károly Szabó, “Bethlen Gábornak és nejének Károlyi Zsuzsannának levelezése (Az Erdélyi Múzeum- Egylet birtokában lévő eredeti és saját kezü példányokról)” [The correspondence between Gábor Bethlen and his wife, Zsuzsanna Károlyi (from the original and autographed copies owned by the Transylvanian Museum Society)], Történelmi Tár [2] (1879): 206; TMÁO I, 300; Samu Gergely, “Bethlen Gábor levelei feleségéhez Károlyi Zsuzsannához” [Gábor Bethlen’s letters to his wife, Zsuzsanna Károlyi], Történelmi Tár [5] (1882): 129. 108 See the report of György Széchy to Ferdinand II (26 April 1621) Szilágyi, ed., Beth len Gábor, 163; Gyula Szekfű, Bethlen Gábor (Budapest 1983), 136–137; Papp, “Bethlen,” 963–964. 109 See the diary of Mihály Tholdalagi: ETA I, 233–234; Andrei Veress, ed., Documente privitoare la istoria Ardealului, Moldovei şi Ţării Româneşti [Documents concerning the transylvanian-ottoman military cooperation 333

2–3,000 Tatar troops again joined his army and fewer also participated in his 1626 war.110 The Tatars were not trained to fight against firearms; their equipment of sabers, bows, lassos, and short lances made them only suitable for raiding activities. Their strategic relevance lay mostly in their infamy, which was deliberately used by Bethlen to threaten his opponents: if 4,000 Tatars should come—he wrote to his envoy at the Porte in 1624— the news will report about 14,000, and Your Lordship should believe that the presence of 4,000 Tatars is more effective for negotiating peace with the Germans than 20,000 Ottomans, because they are feared so much.111 It also became a normal practice for the Romanian voievods to send auxiliary troops at the request of the prince of Transylvania; both countries sent him 1,000 cavalry each in 1623, 1626, and again in 1644.112 The cur teani, a light cavalry equipped with bows, were usually used as advance guards of the Transylvanian troops, and as such did a great service.113 But the Romanian cohorts, like the Ottomans and Tatars, also represented a burden for the prince. While they were supposed to be paid by the voi evods, their wages were paid often late. This led to problems, as in 1644 when uncompensated Romanian soldiers tried to return home through Poland.114 Although Gábor Bethlen asked for the rank of serdar (Ottoman commander) and its attributes from the Sublime Porte in 1623, the rank for this campaign was given to İbrahim, beylerbeyi of Bosnia—known as Deli (“insane”) or Sarhoş (“drunkard”). He marched to Upper Hungary in Octo- ber with an army of more than 10,000 Ottoman and Tatar soldiers from

history of Transylvania, Moldavia, and Wallachia], vol. 9, Acte şi scrisori (1614–1636) [Char- ters and letters] (Bucharest, 1937), 240; TMÁO I, 336, 341. Ferenc Nagy Szabó notes, about the Tatar troops arriving in the autumn, that “these were taken there by the prince only for the sensation they cause,” see ETA I, 139. 110 For both campaigns, see János Kemény, “Önéletírása” [Autobiography], in Kemény János és Bethlen Miklós művei, ed. Éva V. Windisch (Budapest, 1980), 49–50 and 68. 111 TMÁO I, 410. On the Tatar military, see the study of Mária Ivanics in this volume. 112 According to János Kemény, in 1623 their numbers were as high as 4,000. See Kemény, “Önéletírása,” 43. 113 Bethlen had already asked (in 1620) for the “red cavalry” (“ross”) from Wallachia with the explanation: “we have enough heavy troops, but are in great need of light ones.” TMÁO I, 234. See also ETA I, 237. 114 Kemény, “Önéletírása,” 262. The Transylvanian higher command in 1681 tried to avoid routing the Romanian auxiliary armies through the principality’s territory, but they failed to acquire the Sublime Porte’s support. See Trócsányi, Teleki Mihály, 258. 334 jános b. szabó the eyalets of Bosnia, Kanizsa, Buda, Eger, Temesvár, and Silistria.115 How- ever, their military contribution was limited because neither the prince nor the pasha wanted to attack the army of Ferdinand II, then entrenched near Hodonín. Bethlen could not even use the Ottoman threat during his peace negotiations, as the pasha’s troops—without informing him— returned home after the truce was concluded on 20 November. This was long after the standard end of the Ottoman campaigning season, the day of Kasim (23 October). The prince displayed another proof of the ambiguity of Transylvanian–Ottoman relations. He informed the captains of the Hungarian border fortresses of the route taken by İbrahim’s runaway armies, and several smaller troops were thus defeated before reaching Ottoman territory.116 In the summer of 1626 Bethlen again managed to gain the Sublime Porte’s support for his anti-Habsburg campaign. This time the commander of the Ottoman troops—assembled in a manner similar to 1623—was Murteza Pasha, who was also appointed beylerbeyi of Buda. He began his campaign by besieging the castle of Nógrád.117 Bethlen sought to lure him away (in order to avoid the repetition of Vác’s case of 1620) but succeeded only at the end of September, when the Ottoman pasha united his forces with those of Bethlen and his ally, Ernst Count of Mansfeld. No decisive battle took place—neither the Transylvanians and their allies, nor Albrecht von Wallenstein, the leader of the imperial troops, dared to risk it. But Bethlen at least managed to convince the pasha to station his armies in Hungary

115 The order of Sultan Mustafa I about the marching of the troops (1623): TMÁO I, 381–382. See also the letters of Abdul Kerim, beylerbeyi of Silistria, and İbrahim Pasha to Gábor Bethlen (27 May 1623), ibid., 381–382, and 400. For the Tatars, probably from the Bucak, see Kemény, “Önéletírása,” 49–50. On İbrahim Pasha, see Balázs Sudár, A pécsi Jakovali Haszan-dzsámi [The Jakovali Hasan cami in Pécs] (Budapest, 2010), 20–22. For Bethlen’s request concerning the title of serdar, see his instruction to his envoy at the Sublime Porte: TMÁO I, 392. 116 Gábor Bethlen’s letter to Gáspár Illésházy (18 October 1623), Szádeczky, “Bethlen Gábor,” 54–55; Kemény, “Önéletírás,” 43–50; diary of László Révai, entries 27 and 30 Novem- ber 1623, in László Szalay, Galántai Gróf Eszterházy Miklós Magyarország nádora [Count Miklós Eszterházy, palatine of Hungary] (Pest, 1866), vol. 2, 86–87. On the entire campaign, see László Nagy, Bethlen Gábor a független Magyarországért [Gábor Bethlen for the independent Hungary] (Budapest, 1969), 323–351. 117 ETA I, 238; Gábor Bethlen’s letter to Gáspár Illésházy (26 August 1626) Lajos K. Szá- deczky, “Bethlen Gábor levelei Illésházy Gáspárhoz 1619–1629” [The letters of Gábor Beth- len to Gáspár Illésházy], Magyar Történelmi Tár 27 (1915), 90. The most important Ottoman source on the campaign is available in a recent modern edition: Nedim Zahirović, Murteza Pascha von Ofen zwischen Panegyrik und Historie: Eine literarisch-historische Analyse eines osmanischen Wesirspiegels von Nergisi (El-vasfü l-kamil fi-ahvali l-veziri l-‘adil) (Berlin, 2010). transylvanian-ottoman military cooperation 335 for winter quarters, thereby gaining considerable advantage for his peace negotiations.118 Like 1623 and 1626 when the campaigns of Bethlen in Hungary received fairly significant Ottoman support, in 1643 Prince György Rákóczi I also managed to acquire the consent and help of the Sublime Porte for his anti-Habsburg designs in Swedish and French alliance.119 The rather small troops, only 1,000–1,500 men strong, did not play a large part in the military activity, and Rákóczi could not convince the leaders of Ottoman Hungary to start a diversion with the 7–8,000 soldiers they collected around Buda.120 Rákóczi had problems keeping discipline among his Ottoman auxiliary troops, and his attempts to do so were interpreted instead as favoring the “common enemy,” the royal Hungarian armies. At the end of August 1644 one such disciplinary measure was used as a pretext for a temporary withdrawal of the Ottoman troops from the Hungarian theater of war. The preparations for the War of Candia motivated the Ottoman command to reduce their numbers and by 1645 entirely cease their activity on behalf of the Transylvanian prince.121

118 The prince had to use force to suppress a revolt in the Ottoman camp among those soldiers who preferred to leave the theater of war after the day of Kasim. See Kemény, “Önéletírása,” 76–78. On the campaign, see ibid., 66–73; Gábor Bethlen’s letters to Gáspár Illésházy: Szádeczky, “Bethlen Gábor,” 95–97, 101–102, 104; and to Palatine Miklós Esterházy (29 September 1626) Szalay, Galántai Gróf Eszterházy Miklós, 398; Nagy, Bethlen Gábor, 383–437. 119 Although these campaigns have the greatest number and most detailed descriptions in the Hungarian literature on Transylvanian military history, the political background of the Ottoman involvement remains an open question. See, for example, Nagy, Bethlen Gábor; László Cseh Szombathy, “I. Rákóczi György 1644-es hadjárata” [The 1644 campaign of György Rákóczi I], Hadtörténelmi Közlemények n.s. 3, no. 1 (1956), 43 76, and n.s. 4, nos. 1–2 (1957), 179–203; idem, “I. Rákóczi György 1645. évi hadjárata” [The 1645 campaign of György Rákóczi I], Hadtörténelmi Közlemények n.s. 4, nos. 3–4 (1957), 101–135. 120 According to the chronicle of Katib Çelebi, Rákóczi asked for as many soldiers as had been sent from the beylerbeylik of Eger to support Bethlen, which numbered 1,500. See Imre Karácson, ed., Török történetírók [Turkish historiographers], vol. 3, 1566–1659 (Buda- pest, 1916), 340. On Rákóczi’s attempts to start the diversion, see István Szentpáli’s letter to the prince (21 June 1644) Sándor Szilágyi, ed., Levelek és okiratok I. Rákóczy György keleti összeköttetései történetéhez [Letters and documents of the history of the eastern contacts of György Rákóczi I] (Budapest, 1883), 790. 121 See Cseh-Szombathy, “I. Rákóczi György 1644-es hadjárata”; Cseh-Szombathy, “I. Rákóczi György 1645. évi hadjárata”. On the motivation of the Ottoman withdrawal in 1644, see György Rákóczi I s letter to Mihály Maurer (1 September 1644), Antal Beke, and Samu Barabás, ed., I. Rákóczy György és a Porta: Levelek és okiratok [György Rákóczi I and the Porte: Letters and documents] (Budapest, 1888), 709–710. On further disciplinary problems concerning the Ottoman troops, see the prince’s letter to the same (11 and 20 Decem- ber 1644), Ágoston Ötvös, ed., Rejtelmes levelek első Rákóczy György korából [Secret letters from the age of György Rákóczi I] (Kolozsvár, 1848), 122–126. 336 jános b. szabó

The next opportunity for a prince of Transylvania to intervene in the affairs of the Kingdom of Hungary in the vein of Bethlen and Rákóczi was presented by the anti-Habsburg Hungarian uprisings in the late 1670s. The Porte chose to decline to become involved in the conflicts in Hun- gary due to their Polish and Russian wars (1672–1676 and from 1677 on). Prince Mihály Apafi I only received the sultan’s consent and the title of serdar in November 1680, and the ferman reached Transylvania in May 1681. Although the size of the army under Apafi’s command was impressive, the campaign could not boast significant results.122 Because of the slow mobilization of the Transylvanian army and the discord between Apafi and the young leader of the kuruc rebels, Imre Thököly, they failed to conquer the important border fortress of Szatmár. The negotiations to abandon the fortress were stalled by the pillage of the town by the Otto- man auxiliaries. Contrary to his promises, the cohorts of the beylerbeyi of Várad left the camp with their captives; and when the demoralized Tran- sylvanian army also left the theater of war, even the two smaller castles of Kálló and Nagykároly (conquered earlier by Apafi) were recaptured by the imperial troops.123

Conclusions

When assessing the relatively limited scope of the military cooperation between Transylvania and the Ottoman Empire, we should also take into account that during the time frame in question there were rather long periods of peace at the western and northern borders of the Otto- man Empire, where a Transylvanian involvement could have been called upon and put to use. The Ottoman Empire had been in open war with the Habsburg Empire between 1568 and 1683, for only sixteen years, and with

122 In September 1681, Apafi had 7,000–8,000 Transylvanian, 7,000 Ottoman, 2,000 Mol- davian and Wallachian troops, 6,800 kurucs under Imre Thököly’s command, and 1,200 kurucs under Transylvanian command. In sum, approximately 24–25,000 soldiers and sixteen cannons. This was the largest army since the 1657 Polish campaign of György Rákóczi II. On the campaign, see Trócsányi, Teleki Mihály, 253–264; on the attempts to acquire the sultan’s consent in the late 1670s, ibid., 87–233. 123 Trócsányi, Teleki Mihály, 253–264. On the role of Szatmár, see Géza Pálffy, “A felső- magyarországi főkapitányság és Erdély Báthory István uralkodása idején (1571–1586) (A Báthory kutatás egy feldolgozatlan kérdésköréről)” [The captain generalcy of upper Hungary and Transylvania during the rule of István Báthory (1571–1586): On a question not addressed earlier by the research concerning Báthory], Mediaevalia Transilvanica 1, nos. 1–2 (1997): 113–126. transylvanian-ottoman military cooperation 337 the Polish–Lithuanian Commonwealth for only eight. In contrast, in the Mediterranean the wars against Venice and Spain lasted thirty-five years altogether, and against Persia no less than forty-six. The fact that wars taking place in faraway theaters did not compel the direct involvement of Transylvania does not indicate that the principality had a special status. When the Ottoman Empire waged war on a power that fell into the land’s radius of action, it was asked to participate in one way or another. The preceding overview suggests there was no period in the principality’s history when the mechanisms of mutual support were not present. The ‛ahdnames prescribed such commitments for both sides, and it seems that practice even preceded the normative texts. For example, active defense against Cossack interventions in Moldavia was routine even before the codification of the obligation to help in official documents. The demands of the Sublime Porte clearly show the Ottoman political elite’s method of using pressure with reference to precedents, which was also mirrored by the sultan’s orders. In 1593 the Sublime Porte described its expectations concerning the participation of Transylvania in the Hungarian war with reference to John Sigismund’s campaigns of 1566. In 1672 they used the fifty-year-old exemplar of Gábor Bethlen to demand victuals for the Polish campaign from Prince Mihály Apafi I. The size of the cohorts also followed precedents: it was normal for the Romanian voievodates to regularly send 2,000 auxiliary troops to Transylvania while the Ottomans sent 1,000–1,500 soldiers. When looking at the problems from the side of Transylvania, one can see that they were relatively successful at avoiding the obligations, even in the post-1660 period. This is usually seen as a period of decline in the principality’s power, as illustrated by the years 1663 and 1664. At the same time, it was not an exception to the rule that the main Transylvanian army had to participate in the Ottoman campaigns of 1682 and 1683. During the Long Turkish War of the late sixteenth century Zsigmond Báthory was able to avoid a similar situation only by joining the Christian side. György Rákóczi I had to endure an analogous conjuncture in the shadow of the Polish war in 1634, and Gábor Bethlen was unable to resist similar Otto- man demands in 1617. In their wars between 1529 and 1568 the Ottomans could only achieve and maintain the rather ambiguous backing of the Szapolyai family because of the high stakes for the dynasty itself in their struggle against 338 jános b. szabó the Habsburgs. After 1571 the situation was consolidated, as noted by the chronicler János Baranyai Decsi: Neither the goals, nor the motives of the war are the same as they were back then, when John and his son, with the help of Turkish arms, had their rivalry with the Austrian princes for the possession of the country. There is but one law in Hungary now and it sees the Turk as a violent conqueror, the enemy of law, freedom and faith, and it tries with all its power to repulse it.124 It is in light of this lack of motivation that the Porte’s ability to advance its own interests should be judged: far more sultanic orders arrived in the principality than those that were actually followed in practical terms. It depended upon the political wisdom of the Ottoman elite to force their subjects, prone to “disobedience,” to live up to their obligations. They had to decide what kind of task could be trusted to the Transylvanians, who were known to regularly collaborate with the Porte’s adversaries. Nevertheless there were two peculiar aspects to the cooperation. On one hand, the Hungarian influence over Wallachia and Moldavia—earlier vassals of the Kingdom of Hungary—was exercised between 1571 and 1594 by Transylvania, the state then seen as the Kingdom’s successor, with the consent of the Ottoman sultans.125 The peace and stability of this region, connected to Transylvania through a thousand channels, and the good relations between their rulers, was, traditionally, the primary interest of the principality. Nevertheless, in the Long Turkish War the prince of Transylvania built on these close contacts and managed to compel the voievods to break away from the Porte; thus, the Ottoman elite began to disapprove of the signs of Transylvanian influence in the voievodates, and after the wars between 1658 and 1662 they managed to almost entirely eliminate it.

124 Baranyai Decsi, Magyar históriája, 90. 125 Benda, “Erdély politikai kapcsolatai”; Klára Jakó, “Hogyan láttak minket? Magyar- ságkép a 17. századi és 18. század eleji moldvai és havasalföldi elbeszélő forrásokban” [How did they see us? The image of Hungary in the narrative sources from Moldavia and Wallachia from the seventeenth and early eighteenth centuries], Történelmi Szemle 53, no. 2 (2011): 163–182. Time and again, there was even a Transylvanian military presence in the voievodates. See the order of Sultan Murad III to Zsigmond Báthory to allow the voievods to recruit soldiers in Transylvania (30 September 1591) Karácson, ed., Török-magyar oklevéltár, 172–173; and the correspondence between them, in which the sultan calls upon the prince not to keep more than 800 soldiers at the voievods’ court, “according to the old customs,” Veress, ed., Documente, vol. 4, 15–17. For the presence of Transylvanian soldiers at the voievods’ courts in the first half of the seventeenth century, see Demény, Bethlen Gábor, 49, 59–61, 111; Kármán, “György Rákóczi II’s Attempt,” 239–240. transylvanian-ottoman military cooperation 339

On the other hand, after the attempts of the Szapolyai dynasty to unify the Kingdom of Hungary, under the rule of Gábor Bethlen and Mihály Apafi I the opportunity arose to merge the old Hungarian territories not under Ottoman control. This required the help of the sultan’s weapons and meant acknowledging his suzerainty, but was in the interests of the Transylvanian princes. None of these rulers could fulfil this goal—and we are not even sure whether György Rákóczi I, who was also fighting an anti-Habsburg war, ever aimed at it—and in most cases they did not even gain the necessary Ottoman support for it. The contribution of several thousand soldiers from the borderlands was sufficient for pillaging the theater of war, but not for achieving serious strategic goals. At the same time, the campaigns of Gábor Bethlen in 1623 and 1626 might not have ended with compromise if he had not had at his side the entire army of the Hungarian Ottoman borderlands, which in size almost rivaled the prince’s very own troops.

The Defensive System of the Ragusan Republic (c. 1580–1620)

Domagoj Madunić

Introduction

The theme of Dubrovnik, or Ragusa as “the city between the East and the West,” a Christian city-state deeply integrated into the economic life of the Ottoman Empire, is well known and researched. The longevity of this small aristocratic republic, situated on the eastern Adriatic coast with a population ranging from 30–50,000 and a territorial span of no more than 1,100 square kilometers, wedged between the Venetian and Ottoman empires,1 is commonly explained by its success at convincing other states of the necessity of its continued existence. Accordingly, it is argued that Ragusa served as an important center of commerce, especially in times of war, when Ottoman goods were denied to Christian merchants. It was a center of various intelligence activities and functioned as one of the major sources of information for the Ottomans and the Christian powers alike. It also served as a convenient place for the exchange and ransom of war prisoners, to mention just few of the intermediary roles most commonly attributed to Ragusa. And indeed, as such, the Ragusan republic managed to enlist several European powers in its cause, as protectors and guarantors of the republic’s existence. In a time of crisis, it could count on the support of the Spanish monarchy and the Papacy in the West, and on the patronage of the vast Ottoman Empire in the East. The tributary status of the Ragusan republic not only saved this small Christian state from being absorbed by the Ottomans during their conquest of the Balkans, but more importantly, it also ensured Ottoman protection against the Republic of Venice, Ragusa’s more powerful economic rival. Moreover, the prospect

1 Francis W. Carter, Dubrovnik (Ragusa): A Classic City-State (London, 1972), 15. Nenad Vekarić, “The Population of the Dubrovnik Republic in the Fifteenth, Sixteenth, and Sev- enteenth Centuries,” Dubrovnik Annals 2 (1998): 7–28. 342 domagoj madunić of enlisting Ottoman military aid in case of emergency formed one of the axes on which the republic’s overall defensive strategy rested. Although the Holy See was occasionally instrumental in preventing a Venetian military assault on Ragusa—for instance in the critical period of the second war of Holy League (1570–1573)2—in the time of formal peace between the Western powers and the Sublime Porte, the Ragusan government mostly relied on the Ottoman Empire as an effective shield against the very real pretensions of the Republic of St. Marc. The full power of Ottoman protection was most clearly demonstrated during the two crises that threatened the republic in the decades covered by this study. First, during the events known as the “Lastovo crisis” (1602–1606), when, after the outbreak of a rebellion on the Ragusan island of Lastovo, Venice occupied this part of the Ragusan state under the pretext of maintaining law and order. The Venetian occupation of the island lasted for almost four years, and only after the active involvement of the Sublime Porte and open threats of military intervention was the island restored to Ragusan jurisdiction. The second crisis that revealed the importance of Ottoman protection took place during the period between 1617 and 1619, when Ragusa found itself caught in the middle of the armed conflict between Venice and Spain, and for more than a year, Venetian forces practically conducted a naval blockade of the town (October 1617–November 1618). In both cases, it was Ottoman pressure that finally forced the Republic of Venice to back down, and secured a positive outcome for the Ragusan state.3 Yet, small early modern states, even those with the mightiest protectors, needed to be able to defend themselves on their own: protect their subjects from bandits or pirate raids, or the passing armies of their “allies and protectors”; maintain law and order; suppress civil disorders; and fend

2 For more on Ragusa in the time of the “War for Cyprus,” see Robin Harris, Dubrovnik: A History (London, 2003), 112–114. 3 For a general overview of the “Lastovo crisis” and the Ragusan–Venetian conflict, see Harris, Dubrovnik, 117–120; for more detailed coverage of the political developments, see Antonije Vučetić, “Dubrovčani na obrani svog teritorija i slobodne plovidbe prema Mlečanima početkom XVII. vijeka” [Ragusans at the defense of their territory and freedom of sail against Venice at the beginning of the seventeenth century], Glasnik dubrovačkog učenog društva ‘Sveti Vlaho’ 1 (1929): 84–117; on the political and social context and the causes of the rebellion, see Nenad Vekarić, “Lastovski pobunjenici 1602. god.” [The Lastovo rebels, 1602], Anali Zavoda za povijesne znanosti HAZU u Dubrovniku 43 (2005): 43–73. Con- cerning the Venetian–Spanish conflict in 1617–1619, see Vuk Vinaver, “Turska i Dubrovnik u doba španske invazije Jadranskog mora, 1617–1619” [The Ottoman empire and Ragusa in the time of Spanish invasion of the Adriatic, 1617–1619], Istorijski glasnik 1–4 (1952): 21–60. the defensive system of the ragusan republic (c. 1580–1620) 343 off adventurers and conspirators who aimed to overthrow the government and usurp rule. All of this is also true in the case of the Republic of Ragusa. Undoubtedly, the long survival of this city-state was more a consequence of its peculiar geopolitical position and skillful diplomatic maneuvering than its military strength. Nevertheless, in order to ensure its continued existence and the protection of its territory and its subjects, the Republic of Ragusa maintained its own armed forces charged with the difficult and often quite delicate task of the state’s physical protection. The armed forces of Ragusa served two main functions. First, they were to provide the republic with the capability to conduct its own small-scale police operations, fight off pirate and corsair raids, and pacify the rebellious parts of the republic. Second, they served as a means of deterrence. The system of well-maintained and supplied fortifications was intended to discourage and fend off any surprise attacks. In the event of a major formal attack, however, the forces were expected to hold off the attackers and buy enough time for the diplomatic network of the republic to rally practical support from some of its powerful protectors. Keeping these basic issues in mind, the aim of this paper is to explore three main questions. First, how the Republic of Ragusa organized its defense with regard to its peculiar international status; second, what strategic considerations influenced the formulation of its defensive policy; third, what was the relevance of its armed forces and military capacities in their local context, that is, in the eastern shore of the Adriatic and its hinterland.

Historical Context: Between the Uskoks and the Ottomans (1580s–1620s)

(. . .) between the Turks and the Uskoks we are caught in the middle: we can neither defend ourselves from their abuse of power nor satisfy them in such a way as to bind them to the promise of peace. [Thus] we are going to fall into great ruin. Letter from the Ragusan government to its ambassadors in Belgrade, 16 July 16054 The decision to choose the last two decades of the sixteenth and the first two of the seventeenth century as the time frame for this study was

4 “(. . .) fra i turchi e fra gli scocchi essendo noi di mezzo, e non potendo, ne diffenderci dalla superchiaria loro, ne meno sodisfargli talmente, che ci possiamo prometter della pace e con gli uni, e con gli altri stiamo per incorrer in qualche rouina notabile.” Državni arhiv u Dubrovniku (Ragusan State Archives, henceforth DAD), Lettere di Levante, vol. 41 (1604–1608), fol. 55v. 344 domagoj madunić M ap 1. The territory of the Ragusan Republic in the seventeenth century the defensive system of the ragusan republic (c. 1580–1620) 345 mainly influenced by two factors. First, the availability of sources: almost half of the archival material available in the State Archive of Dubrovnik in the series “Guardie et Armamento,” which contains documents related to the functioning of the defensive system of the republic, originates from these four decades. Even more importantly, these decades also represent a turbulent and perilous period of Ragusan history. Although no major engagement of Christian and Ottoman forces took place in the Adriatic in this period, it was nevertheless a time of crises, minor and major alike, which seriously threatened the existence and territorial integrity of the Ragusan republic, and tested its defensive policies. In addition to the two previously mentioned major crises involving the Republic of Venice, the main threat for the Ragusan republic in the period of this study came from an enemy of a much lower profile than the Serenissima Republica—the Uskoks of Senj. Operating from their base in the fortified port of Senj, these imperial privateers raided not only the Ottoman settlements along the coast and in the immediate hinterland, but also, under the pretext of seizing goods of the infidels, targeted merchant shipping throughout the Adriatic.5 Their activity reached its first peak during the War for Cyprus (1570–1573), when the Uskoks launched a campaign of unrestrained attacks against Ottoman targets with the blessing of the Venetian republic, which otherwise suppressed their activity. Unfortunately for Ragusa, not only did the Uskoks show a complete disregard for the proclaimed Ragusan neutrality in this war, but—because of Ragusa’s status as a tributary state of the empire—they also considered its territory and ships legitimate targets for attack. In 1570 alone, the Uskoks conducted thirty raids against targets on Ragusan territory, plundering and taking hostages. The ill-feeling between the Uskoks and Ragusa was further confirmed in 1571, when the Ragusan government, after capturing him by trickery, murdered the famous Uskok leader Juraj Daničić. Although some sort of reconciliation was reached between the sons of Daničić and the Ragusan republic in 1580 through Papal mediation, Uskok attacks against Ragusa never ceased, and the renewal of the

5 For a good study on the Uskoks of Senj in English, see Catherine W. Bracewell, The Uskoks of Senj: Piracy, Banditry, and Holy War in the Sixteenth-Century Adriatic (Ithaca, NY, 1992). In addition, for a detailed study of the Uskoks in the context of political events of the sixteenth century, see Gligor Stanojević, Senjski uskoci [The uskoks of Senj] (Belgrade, 1973). 346 domagoj madunić

Habsburg–Ottoman hostilities in 1593 marked the beginning of a new series of Uskok raids that did not stop until after 1613.6 An even greater danger for the Ragusan government than the direct consequences of Uskok attacks was the use, by the Uskoks, of the territory of the republic as a base of operations for their raids on neighboring Ottoman lands. Moreover, the involvement of some Ragusan subjects as accomplices to the Uskoks, providing them with shelter and guidance, additionally compromised Ragusans in the eyes of local Ottoman officials. This not only led the Porte to accuse the Ragusan government of supporting the Uskoks and neglecting to prevent their exploits, but also led to retaliatory attacks against Ragusan targets conducted by the inhabitants of nearby Ottoman lands. Thus, from the 1580s on, relations between Ragusa and its hinterland rapidly deteriorated as a direct consequence of Uskok raids. Although Ragusan diplomats could count on finding a sym- pathetic ear for their cause in the center of the empire, officials of nearby provinces became ever more hostile to the Ragusan state. Additionally, rumors of Ragusan “infidelity” provided Ottoman corsairs from Tunis to Valona with a pretext to attack Ragusan targets. All in all, for the Repub- lic of Ragusa, caught in the crossfire between the Uskoks and Ottomans, these years resembled more a state of war than of peace.7 The seriousness of the situation can best be illustrated by the events of 1605, when the crisis reached a peak. At the end of June, an Uskok party of eight vessels, carrying 360 men, arrived in the Ragusan waters. After land- ing on the Ragusan coast, the Uskoks crossed over into Ottoman lands unhindered, where they captured many prisoners, animals, and a variety of goods near the town of Trebinje in eastern Herzegovina. After breaking camp, instead of just sailing away with their booty, the Uskoks, drunk on

6 For more on Uskoks and Ragusa, see Harris, Dubrovnik, 114–116; for the affair concerning the murder of Juraj Daničić, see Stanojević, Senjski uskoci, 94; Bracewell, The Uskoks of Senj, 165–166; Jorjo Tadić, “Pogibija uskočkog vojvode Djura Daničića 1571” [The death of the Uskok voievod Djuro Daničić], Novosti 24 (1931): 8; Vinko Foretić, Povijest Dubrovnika do 1808 [History of Ragusa until 1808], vol. 2 (Dubrovnik, 1980), 61–63. 7 How deeply Uskok raids troubled Ottoman–Ragusan relations can be seen from the number of fermans dealing with the problem of Uskoks. In the period between 1588 and 1618, no less than 57 fermans were issued. Out of these, 22 were against Ottoman subjects who conducted attacks on Ragusan territory; 11 stating that Ragusans were not guilty of Uskok raids; 17 written against Ottoman corsairs and ordering the release of Ragusan prisoners and the return of stolen goods; 9 forbidding Ottoman investigations on Ragusan territory, and so on. See Vesna Miović, Dubrovačka Republika u spisima osmanskih sultana [The republic of Ragusa in the documents of the Ottoman sultans] (Dubrovnik, 2005), 185, 191–199, 206–221, 227, 230. the defensive system of the ragusan republic (c. 1580–1620) 347 success and wine, inexplicably attacked the armed Venetian vessels guarding the nearby Korčula channel. The superior Venetian forces defeated the Uskoks, inflicting heavy casualties. The Uskoks suffered 25 dead and 50 wounded, while the rest managed to save themselves by retreating to Ragusan territory. As news of this Uskok defeat spread, a 400-man-strong Ottoman force from the delta region of the river Neretva crossed into Ragusan territory, and attacked the Uskoks near the fortress of Ston. Unfortunately for the Ragusan government, this time fortune favored the Uskoks, who routed the attackers completely, killing more than 150 and capturing many. The fleeing Ottomans found succor by retreating to the protection of the Ragusan fortress complex at Ston, where they received shelter and medical aid. As a final act of insult, the Uskoks organized a ransom of prisoners in the vicinity of the fortress of Ston and after all ransoms had been settled, finally sailed away. However, this was just the beginning of the troubles for the Ragusan government.8 Fearing Ottoman retaliation for this Uskok raid, the Ragusan government started simultaneous preparations on both the diplomatic and the military fronts. Envoys were sent both to Istanbul and to the grand vizier in Belgrade, with instructions to refute any imputations of a Ragu- san involvement in this affair. At the same time, the city garrison was strengthened and militia levies were raised throughout the republic. Indeed, the anticipated attack finally came in late July, when a party of raiders from the Ottoman stronghold of Herceg-Novi intruded into Konavle, killing four and capturing 62 Ragusan subjects (including two nobles) and 2,200 animals. This, however, turned out to be the least of the Ragusan problems. Although Ragusan envoys in Istanbul managed to obtain fermans con- demning the attackers,9 voices accusing the Ragusans of being Uskok accomplices found fertile ground in the headquarters of the grand vizier in Belgrade. Already in July, Ragusan ambassadors in Belgrade informed their government that they had fallen out of favor with the grand vizier, who turned against them after hearing reports of this last Uskok raid. Most importantly, the Ragusan inability to prevent Uskoks from crossing

8 Stanojević, Senjski uskoci, 218–219. For the Ragusan version of events, according to which they did everything in their power to prevent Uskoks, sent prompt warnings of their movements to local Ottoman officials, and provided every possible help to the Ottomans after the battle, see the instruction for the Ragusan ambassadors in Belgrade and Istanbul: DAD, Lettere di Levante, vol. 41 (1604–1608), fols. 47v–51v, 53r–56v. 9 Miović, Dubrovačka Republika, 210–212. 348 domagoj madunić their territory revived a plan to take away the republic’s easternmost part, the region of Konavle, and organize it as a new sancak.10 In the following months, an intensive diplomatic struggle went on in the triangle of Ragusa, Istanbul, and Belgrade. The crisis reached its culmination in December 1605, when a ferman was issued in Belgrade, ordering an imperial envoy to Ragusa to conduct an investigation on Ragusan territory and establish the extent of their culpability in this affair. A second decree followed, according to which Konavle was to be ceded and made a sancak, appointing the same imperial investigator as its sancakbeyi.11 Although Ragusan envoys managed to have both of these decrees revoked over the following few months, the events clearly show the dire consequences Uskok raids could have.12 When the Ragusan government was relying heavily on Ottoman backing to force Venice to return the island of Lastovo to its jurisdiction, it not only almost lost this support, but also came rather close to losing another part of its state, this time to the very power it relied on for protection and assistance. This sketch of the geopolitical situation of Ragusa provides a sample of the challenges its defensive system faced. Now, it is time to take a closer look at the republic’s military capacities. The defensive system of the republic was made up of three main components: (1) the armed forces; (2) a system of fortifications that covered all the key points of its territory; and (3) a small supportive military infrastructure and a set of well- supplied magazines and stores of war materials.

Armed Forces

The number of regular/salaried troops in Ragusan service varied from year to year, depending on the circumstances and the military needs of the republic, but it almost never exceeded 350 men. These forces were stationed in two main garrisons. The first—and more numerous—one, which

10 For the first news of Ragusans falling out of grace with the grand vizier, and the reac- tivation of the plan to organize Konavle as a sancak, see DAD, Lettere di Levante, vol. 41, fols. 56v–59v. The first attempt to seize this part of the Republic and turn it into a sancak took place in 1588–1590, also as a consequence of the deteriorating relationship between Ragusa and its hinterland, due to uskok raids. For more detail, see Harris, Dubrovnik, 108–109; and Miović, Dubrovačka Republika, 66–68; Toma Popović, Turska i Dubrovnik u XVI veku [Turkey and Ragusa in the sixteenth century] (Belgrade, 1973), 348–361. 11 For the fermans, see Miović, Dubrovačka Republika, 211–212. 12 For the decrees revoking the appointment of the sancakbeyi, and later for orders to arrest culprits from Herceg-Novi guilty of attacking Konavle, see ibid., 213. the defensive system of the ragusan republic (c. 1580–1620) 349 numbered around 200 men, was stationed in the town of Dubrovnik itself. The second was in the complex of the Ston fortifications in the western part of the republic, guarding the Pelješac peninsula: here a further 70–100 men were posted. In addition to these two main garrisons, the republic also maintained a small unit (10–20 men strong) in the fortress of Sokol in the easternmost part of its territory, guarding important caravan routes that led from its territory into the Ottoman lands.13 The core of the Ragusan forces consisted of a company of foreign mercenaries called the barabanti.14 The ranks of the barabanti were filled exclusively by foreigners (mainly from Croatia and Hungary) with military experience and served as a sort of elite force that provided a backbone for the units of local Ragusan soldiers. Without any personal ties to the local population or the ruling aristocracy, this unit was a perfect choice for police and security operations, and represented a major asset for sup- pressing civil disorders or rebellions. One of the most important duties entrusted to the barabanti was the daily and nightly patrolling of the town and its fortifications.15 The majority of the barabanti, 30–40 of them, served in the town of Ragusa, while a smaller detachment of 10–20 men was stationed in Ston. The bulk of the Ragusan military was made up by the so-called soldati, Ragusan subjects who had chosen the military profession as their trade. While the barabanti were organized as a single company under one captain, soldati were divided into six squadrons (squadre), commanded by two captains, helped by a dozen corporals (capi). Of those two captains, one was in command of the five town squadrons of various sizes (see Table 1), each in charge of a single guard post, whereas the second captain commanded the so called squadra di ormaniza, a unit of a soldati serving as a marine infantry on the armed vessels of the republic.16

13 Ilija Mitić, Dubrovačka država u međunarodnoj zajednici (od 1358. do 1815.) [The Ragu- san state in the international community from 1358 to 1815] (Zagreb, 1988), 83–91. 14 The term “barabant” is the Ragusan derivative of the Hungarian word darabont, origi- nating from the German Trabant, meaning “escort.” For more on the origin of this unit, see Trpimir Macan, “Dubrovački barabanti” [Ragusan barabants], Anali historijskog instituta JAZU u Dubrovniku 8–9 (1962): 301–323. 15 The unit in charge of patrolling the town was called soprazonta and consisted of 12–18 men. From 1597, the town and guard posts on the city walls were patrolled four times during the day and four times during the night, see Macan, “Dubrovački barabanti,” 318. 16 Ormaniza (Ormanica, Ormanizza) was the name of a specific type of Ragusan oar- propelled vessel from the branch of the small galleys, known in Italian sources as galeotte. It had approximately 12 banks for oarsmen and a single sail. It was usually armed with a single cannon positioned at the prow. For more detail, see Josip Luetić, Pomorci i 350 domagoj madunić

Table 1. Size and the structure of the Ragusan garrison for the years 1612–161717 Formation Number of men Formation 1612 (a) 1613 (b) 1614 (c) 1617 (d) La squadra di Logia 41 — 33 42 La squadra di Ponta 15 — 16 14 La squadra di Ploce 19 — 17 22 La sqadra di Pile 17 — 15 21 La squadra di S.to Lorenzo 10 — 16 13 Soldati di ormaniza 29 — 27 30 Unassigned soldati 5 — 14 0 Total number of soldati 136 138 138 142 Compagnia di bombardieri 23 24 22 26 Compagnia di barabanti 41 43 39 34 Town total 200 205 199 202

As a supplement to these two formations, the republic also kept a company (30–40 men) of trained artillery specialists called bombardieri on its payroll. Unlike the soldati or the barabanti, who were professional soldiers per se and supported themselves strictly through such activities, the ranks of the bombardieri were also open to practitioners of other professions, like blacksmiths or shipbuilders. Based on the Venetian model, in the early modern period the bombardieri were organized in their own fraternity (scuola) under the protection of St. Barbara, the patron saint for handlers of explosives.18 In accordance with the semi-military status of the gunners, their commander did not hold the title of captain, but rather that of prota, the head of a fraternity.

jedrenjaci Republike Dubrovačke [Sailors and sail ships of the Ragusan republic] (Zagreb, 1984), 237–242. 17 Data for the columns are derived from the following sources: (a) the payment records for November 1612: DAD, Guardie et Armamento, vol. 37. Guardie e Barabanti 1612, f. 1–42; (b) January 1613: DAD, Guardie et Armamento, vol. 52. Libro di soldati, Barabanti e bon- bardieri per messi di agosto–genaro 1612 in 1613; (c) June–July 1614: DAD, Guardie et Arma- mento, vol. 45, Registro de pagamenti de Salarij da Proveditori delle Guardie dell’ anno 1614; (d) the payment records for January 1617: DAD, Guardie et Armamento, vol. 3. Salarii dei Soldati e Guardie dell’ anno 1617, f. 1–8. 18 On this fraternity and its statute, see Tonko Marunčić, “Matrikula bratovštine dubrovačkih topnika (bombardijera) u prijepisu iz 1697. godine” [The statute of the fraternity of Ragusan gunners (bombardieri) in a transcript from the year 1697], Anali zavoda za povijesne znanosti HAZU u Dubrovniku 45 (2007): 285–318. On the development of scuola di S. Barbara in Venice, see M.E. Mallett and J.R. Hale, Military Organization of a Renaissance State: Venice c. 1400 to 1617 (Cambridge, 1984), 403–408. the defensive system of the ragusan republic (c. 1580–1620) 351

In the event of an emergency, this core, consisting of 300–350 regular troops—that is, around 200 in the city and some 150 in the surrounding areas—was expanded by levies of territorial militia. Drafted Ragu- san subjects were equipped with arms and ammunition from the state armories and were paid the same wages as regular Ragusan soldati for each day of service. In the perilous years covered by this study, these territorial levies were almost a regular occurrence. On any news or at the prospect of an Uskok or Ottoman raid, a few hundred men from the threatened regions were mobilized.19 Moreover, the Ragusan government also relied on these levies to strengthen its regular forces for offensive operations. For example, during the Lastovo crisis, when the government organized two expeditions against the rebellious island in July 1602 and January 1603, the core of the regular troops was expanded by the addition of several hundred of these militiamen.20 Probably the most extensive operation of levying the territorial militia occurred during the years of the Venetian–Spanish conflict (1617–1619), when each and every time a Spanish or Venetian fleet arrived in Ragusan waters, it provoked the strengthening of the town garrison.21 This operation reached its height in the period between September 1617 and September 1618, when the Venetian fleet undertook an informal blockade of the town. As a response, the Ragusan garrison was strengthened with as many as 500 additional levies in October 1617; in November of the same year, after the Venetian fleet chased off the fourth Spanish incursion into the Adriatic, the Senate

19 For example, in the aftermath of the uskok raid of 1605, while expecting an Ottoman retaliation, the Ragusan government mobilized 40 additional town citizens to strengthen the town watch. DAD, Acta Consilii Rogatorum [henceforth ACR], vol. 80, fol. 57r. Further- more, from June to August 1605, the Ragusan government ordered levies of a few hundred of its militia from various parts of the Republic. DAD, ACR, vol. 80, fols. 53v, 62v, 63r. 20 In July 1602, the Ragusan government sent 500 men, and in January 1603, a force of 600 men accompanied by artillery was sent to reconquer the island from the rebels. Unfortunately, due to the nature of the available sources, it is not possible to determine how many of these 600 men were regular troops and how many militia levies. For more detail on these events, see Vučetić, “Dubrovčani,” 86–87. For the decision to send a second expedition in January 1603, see DAD, ACR. vol. 78, fol. 140r. 21 In January 1617, 100 men were brought to Ragusa to strengthen the town garrison; they were followed in April by further 100 men. In the same month, similar orders were issued to the captain of Ston to reinforce its garrison with a levy of 100 men. When the Spanish fleet returned to Ragusan waters in June 1617, the government raised a further 150 men to guard the town. The third arrival of the Spanish fleet was also accompanied by orders to raise 250 men. See Vinaver, “Turska i Dubrovnik,” 28–29, 31–32, 42, 48. For the political context of the Spanish incursion into the Adriatic, see also Radovan Samardžić, Veliki vek Dubrovnika [The grand century of Ragusa] (Belgrade, 1983), 70–82. 352 domagoj madunić issued a general call to arms for the town’s inhabitants.22 Throughout the crisis, the Ragusan government kept hundreds of its subjects under arms (the town garrison reached its highest level of manpower in September 1618, when it amounted to 1,000 men), rotating them so that the burden of service would be evenly spread among the various parts of the republic. According to the preserved records, nearly 8,000 ducats were spent on the salaries of these additional troops in 1618 alone.23 In order that Ragusan subjects would be attracted to volunteer for this service—rather than be pressed into the ranks by means of coercion—the city’s government granted various privileges to those ready to serve, in addition to regular pay. For example, during the second Lastovo expedition, the Senate declared that soldiers who participated in this operation would be granted liberty to plunder.24 In times when levies were needed for the more unpopular duty of confronting Uskok raiders, the government went even further. In August 1605, five Provisores contra Uscocchos were given special powers to pardon those willing to enlist, in both public and private cases, with the exception of capital crimes.25 Concerning the practice of employing and paying its military personnel, the Republic of Ragusa diverged from contemporary European trends. Unlike the majority of European states that contracted private individuals, captains or colonels to raise units for their service, the Ragusan government did not use any intermediaries. The decision to receive a particular individual as captain, barabant, soldat or bombardiere into service was in the jurisdiction of the Senate, and the Ragusan state was the sole direct employer of all of its military personnel.26 The republic differed from the majority of European states in yet another respect: as the preserved payment books of soldiers from the period of this study reveal, Ragusa was a very good employer from the perspective of the ordinary soldier, as, uncharacteristically for that time, the troops were paid quite regularly.27

22 Vinaver, “Turska i Dubrovnik,” 49–52. 23 DAD, Guardie et Armamento, vol. 4, Libro delli pagamenti de soldati di Sopragionta 1618, fol. 215r. 24 DAD, ACR, vol. 78, fol. 140r. 25 DAD, ACR. vol. 80. fol. 72v 26 On the acceptance of barabanti into service, see DAD, ACR, vol. 80, fols. 13r, 73r, 181r; vol. 81, fol. 4r–4v. For bombardieri: DAD, ACR, vol. 80, fols. 56r, 63v, 175v. Like the barabants, their commanders were also recruited from abroad. For example, in October 1603 it was decided to find a suitable person “de partibus Hungariae, vel Croatiae” for the office of the captain of barabanti: DAD, ACR, vol. 80, 23v–24r. 27 See, for example: DAD, Guardie et Armamento, vol. 45, Pagamenti dei Provveditori alle Guardie 1614; or DAD, Guardie et Armamento, vol. 53. Libro Salari delle guardie 1617. the defensive system of the ragusan republic (c. 1580–1620) 353

The soldati, the barabanti, and the bombardieri also differed with respect to the way in which their salaries were calculated. While for the barabanti and the bombardieri salaries were determined on a monthly basis (as was customary for the majority of seventeenth-century armies), the salaries of the soldati were counted on a daily basis and thus varied from month to month. Soldiers’ wages also varied according to the equipment they wore and their experience. The remuneration received by the barabanti ranged from 2:15 to 4:15 ducats and grossi, with the majority receiving a pay of 3 ducats.28 The salaries of the bombardieri were more closely bound to the scudo and equaled 3, 4 or 5 scudi per month (or 2:28, 3:24, and 4:20 ducats and grossi respectively).29 Unlike the barabanti, among whom no distinction of rank was made, several Ragusan soldati served as non- commissioned officers and carried the rank of corporal (capo). In 1617, out of 140 men serving as soldati, 15 were enlisted as corporals and two more as drummers (tamburino). The daily wages of corporals and drummers were 5 grossi (or 3:30–35 ducats and grossi monthly), while the majority of the soldiers received 4 grossi, with a few exceptions of those who served for 3 grossi per day (or 3, and 2:10 ducats and grossi respectively). Due to the harsher conditions of their service in contrast to garrison duties, 5 grossi per day were also granted to soldiers who served as marine infantry in the squadra di ormaniza. The fact that officers in Ragusan service commanded units that they did not raise, but which were recruited by the Ragusan republic, also manifested itself in their salaries. While the wages of soldiers in Ragusan service were more or less in line with those received by soldiers in Venetian service,30 those of the captains were considerably smaller. While a captain of a company of fanti Italiani (a unit of troops raised in Italy) in Venetian service could expect to earn 20 ducats per month,31 no Ragusan officer

28 For example, out of 40 enlisted barabants in 1612, 8 received a salary of 4:15 ducats, 6 received 3:24, 9 received 2:28, while 17 served for a monthly pay of 3 ducats. The sums are given in the units of ducat: grossi. The former was a money used in Ragusan account books; whereas the latter (also called grossus or dinar) was a small Ragusan silver currency, coined in the mints of the Republic, which represented the main currency used in daily life. The exchange rate in the period of this study was 1 ducat: 40 grossi. For a short overview of Ragusan coins, see Carter, Dubrovnik, 555–566. 29 Scudi were silver coins originally introduced by the Papal State at the end of sixteenth century, and were called so because of the mark of the shield imprinted on them. In the period of this study, one scudo was valued at 36 grossi; see Carter, Dubrovnik, 555–566. 30 On infantry wages in the sixteenth and early seventeenth century in the Venetian army, see Mallett and Hale, Military Organization, 494–501. 31 Ibid., 394. 354 domagoj madunić was paid that much. The highest salary was given to the captain of the barabanti, who received 8 ducats a month, followed by the commander of the gunners, the prota (at 7 ducats per month). The wage of the captain of soldati, in charge of the town’s guard posts, equaled 8 or 10 grossi per day (approximately 6–7:30 ducats a month),32 and finally, the lowest salary was given to the captain in charge of the squadra di ormaniza, who was paid 6.5 grossi per day (roughly 6 ducats a month).33 In spite of the Venetian claim to an exclusive right to maintain armed ships in the Adriatic, the Republic of Ragusa did keep a small fleet of armed ships. This war navy consisted of a variety of ships: galleys, frigates, filjuga ( felucca), ormaniza or barca longa, and small galleys (galeotte). These were all fast oar-propelled vessels, fit for coastal service and the protection of the Ragusan littoral against pirate attacks. The number of ships in service depended on the level of threat, and usually ranged between two and four.34 Additionally, this squadron of state-owned ships was, when necessary, strengthened by the commissioning of private vessels, which were then armed with weapons from the state arsenals.35 In the 1580s and 1590s, the core of this small fleet consisted of frigates, but as a consequence of the increased pirate activity, in 1605–1606 the Ragusan government modernized its fleet by replacing the frigates with two newly built small galleys (galeotte).36 Unlike frigates, which were

32 Payment records for 1612–1613 show that Captain Pasqualino de Andrea received 8 grossi per day, while in the records for 1614 and 1617, his successor, Captain Piero Tasi is recorded with a daily wage of 10 grossi per day. 33 See the list of the relevant archival sources in note 18 above. 34 For an overview of the long Ragusan tradition of employing the galleys and similar oar-propelled ships as the Republic’s main warships, see Antun Ničetić, “Galije trireme i bireme bile su tijekom više stoljeća ratni brodovi Dubrovnika” [Galleys, triremes, and biremes were in the course of centuries the warships of Ragusa], Anali HAZU Dubrovnik 40 (2002): 9–56. 35 For example, the Senate commissioned two large private ships to transport the troops sent to crush the rebellion on the island of Lastovo in 1603. DAD, ACR, vol. 78, fol. 140r. In 1608, due to the higher level of threat, three private ships (two frigates and one tartana) were commissioned and armed with weapons from state arsenal. See Josip Luetić, “Galijica Dubrovačke Republike” [Small galley of the Ragusan republic], Mornarički glasnik 1 (1959): 54. On the practice of arming Ragusan merchant ships with state guns, see Lukša Beritić, Dubrovačka artiljerija [Ragusan artillery] (Belgrade, 1960), 120–124. 36 The decision to build these new vessels was reached in 1601, when the Ragusan Sen- ate asked Vice Bune, its consul in Naples, to try and find a master shipbuilder willing to come to Ragusa to supervise the construction of these ships. In December 1604, two master shipbuilders arrived and work began. During 1605 and 1606, two small galleys were built, one with 15 benches, the other with 20. See Luetić, “Galijica Dubrovačke Republike,” 55; Ničetić, “Galije trireme i bireme,” 34–35; DAD, Guardie ed Armamento, vol. 29a, Officiali dell’ armamento dell’ anni (1605–1607), fols. 3v, 6r, 17v, 27r, 29r, 31v, 34v, 34r, 35v. the defensive system of the ragusan republic (c. 1580–1620) 355

basically multipurpose transport ships fitted with guns for military service, the galeotte were from the very start built and designed as warships.37 Seen from the perspective of the Ragusan government, the main advantage of these small warships were their very low maintenance costs and their ability to be quickly manned, equipped, and put into service; all of which was a result of the small size of their crews. The crew of a Ragusan frigate, for example, consisted of a core of 10–15 sailors on more or less permanent payroll (in summer months five soldiers were usually added to the crew), and approximately 20–30 rowers, who were employed as needed. Ships would usually begin to assemble their crews to full size in May, and with the coming of late fall, rowers would be decommissioned and ships drawn ashore for winter maintenance. Rowers on Ragusan ships were either volunteers who were paid two grossi per day, or convicts sentenced by magistrates to this service for a specific period of time (from seven days up to five years).38 In addition to these small warships, the Ragusan government also kept four galleys (galere) in readiness, as a strategic reserve. However, launching these ships required not only a greater expenditure of time and resources (large quantities of food, ammunition, and piles of various other materials), but also required significantly larger crews (200–280 men), which could not be assembled quickly. All of this, coupled with very high maintenance costs once a galley was put into service, rendered this otherwise first-rank Mediterranean warship a very cumbersome and ineffective tool against pirate attacks for the defense of a small state such as Ragusa.39 For the Republic of Ragusa, these ships represented strategic weapons, to be used only in times of dire need. By organizing its war navy in this manner, the Ragusan government not only ensured a relatively inexpensive naval defense and reduction of

37 A galeotta not only carried a slightly bigger crew than the frigate; it also mounted higher firepower. Unlike frigates, which usually carried 2–4 pedriere (a model of gun that used cartridges and a breech-loading technique), a galeotta carried as many as 6 pieces of artillery. For a full inventory of equipment and armament of the galeottas, see Luetić, “Galijica Dubrovačke Republike,” 60–61. For a more detailed analysis of the armament of the Ragusan small galley, see also Ničetić, “Galije trireme i bireme,” 47–50. 38 Ničetić, “Galije trireme i bireme,” 36–37. For the dynamics of providing the crews for the two frigates in 1581, see, for example, DAD, Guardie ed Armamento, vol. 18, Officiali della fregata. Convicts sentenced for longer detention were usually sent to the galleys of Naples; see, for example, DAD, ACR. vol. 80, 186v–187r, 198v. 39 For example, the monthly operating costs of the Genovese galley from 1646 are estimated at 9,042 Genovese lire, or roughly 2,500 ducats. Thomas A. Kirk, Genoa and the Sea: Policy and Power in an Early Modern Maritime Republic, 1559–1684 (Baltimore, 2005), 203. 356 domagoj madunić expenditures to the lowest possible level,40 but it was also able to finely calibrate its military responses to match the level of threat. Thus, it was possible for the Ragusan government to have just a fraction of its warships in active service, manned by skeleton crews, while keeping other ships disarmed in the arsenal, ready to be fitted and launched on short notice. In 1614, for example, the Ragusan war navy totaled nine vessels: four galleys, three small galleys of different sizes (15, 20 and 22 benches), and two barche longhe—how many of these were put into active service, however, depended on the circumstances.41

Fortifications and Armaments

The second key factor of the Ragusan defense system was its fortifications. The determinants of Ragusan geostrategic setting, that is, a city-state located in a region with a high level of everyday violence, exposed to the frequent raids of Uskoks and Ottoman pirates, were also reflected in the Ragusan landscape. In addition to the two major fortification complexes, Ston and the town of Dubrovnik itself, there were other smaller fortified fortresses or towers scattered over the entire territory of the republic, the most important of which was the small fortress of Sokol. These smaller facilities served a dual function: they provided shelter for the local population in the event of pirate raids; but, due to their positions at important strategic places, usually dominating the particular settlement, they also served as centers of state control over the local population. Impressive as they seem to us today, the round Ragusan bastions, built at great expense during the fifteenth and the first half of the sixteenth century, had already passed their zenith by 1600.42 Technological advances in the field of artillery rendered tall and thin fortifications less effective against improved firepower, and led to the development of a new style of low-lying, thick walls, and fortifications based on star-shaped bastions,

40 For example, the total cost of crews’ salaries of two frigates in 1581, during the service period, amounted to less then 390 ducats. DAD, Guardie et Armamento, vol. 18, Officiali della fregata. 41 DAD, Guardi et Armamento, vol. 2, Inventario delle armi e munizioni dell’ armamento 1614, fol. 52v, 53v, 55v, 57v. 42 For a more detailed overview of the Ragusan fortifications and their development, see Lukša Beritić, Utvrđenja grada Dubrovnika [Fortifications of the town of Dubrovnik] (Zagreb, 1955); Harris, Dubrovnik, 290–297. the defensive system of the ragusan republic (c. 1580–1620) 357 M ap 2. The fortifications of Ragusa 358 domagoj madunić known as trace italienne.43 Unlike the majority of other early modern states that spent lavishly on the modernization of their fortifications, after 1540 when the fortresses of Revelin and St. John were completed, the Ragusan government did not embark on any other extensive construction projects. Their reasons could certainly be attributed to the extremely high costs of such projects and the decline of Ragusan commerce in the period of this study, but they could hardly be ascribed to financial factors only. If the survival of the republic had depended on surrounding the town with new star-shaped fortifications, money would have undoubtedly been found for such an enterprise.44 It is more likely that town fortifications were never brought up to seventeenth-century standards because they did not fit Ragusan defensive strategy, which rested primarily on the ability of the republic to ensure outside military and diplomatic support in case of a major attack. In spite of the undoubtedly increased effectiveness of the new-style fortifications against artillery attacks, nevertheless it can be argued that, throughout the sixteenth and (at least) the first half of the seventeenth century, the human factor rather than technical considerations remained key to the successful defense of a fortress. Even the most technologically advanced fortifications needed to be manned by a sufficient number of soldiers, who were willing and adequately provisioned to defend them. One of the proponents of this approach, David Parrot, writes about the utility of early modern fortifications in the Italian context, arguing that the key issue during a siege was the morale of defenders rather than the improved technological features of new-style fortifications. Furthermore, according to Parrot, the morale of the defenders was closely linked to the prospect of being relieved by a field army that would drive off the besiegers, or the availability of reinforcements of men, munitions, and supplies in quantities that would discourage attackers from continuing with the siege.45

43 Christopher Duffy, Siege Warfare: The Fortress in the Early Modern World 1494–1660 (London, 1996), 25–34; See also John R. Hale, “The Early Development of the Bastion: An Italian Chronology, c. 1450–1534,” in Renaissance War Studies, ed. John R. Hale (London, 1983), 1–29. 44 The decline of Ragusan commerce was accompanied by the withdrawal of Ragusan capital and its investment in the more secure, although much less profitable, deposits in Italian banks. In 1575, deposits of Ragusans amounted to 262,140 ducats, while by the end of the seventeenth century they reached 700,000 ducats. Carter, Dubrovnik, 578. 45 David Parrot, “The Utility of Fortifications in Early Modern Europe: Italian Princes and Their Citadels, 1540–1640,” War in History 7 (2000): 133–135. Although Parrot based his arguments on examples taken from French–Habsburg confrontations and warfare in the Netherlands, similar examples can also be found in a context more relevant for the Ragusan case, that of Ottoman–Christian conflicts in the Mediterranean. For example, the defensive system of the ragusan republic (c. 1580–1620) 359

Although by the 1600s Ragusan fortifications were no longer considered state of the art, they still represented a formidable obstacle, because of their thickness. They were obstacles that, if well manned and supplied, could be counted on to keep out attackers until military aid from some of the republic’s protectors was ensured. But, if such help could not be obtained and the attackers were willing to press the siege, even the most modern fortifications would (probably) not have saved the city. There- fore, the question to address is: how well was the Ragusan government prepared in case of attack? On 7 April 1616, two Ragusan nobles conducted an inspection of all the artillery, those positioned on the city walls and those stored in magazines. According to their report, the town fortifications had 145 pieces mounted and a further 101 pieces stored in magazines, giving an impressive total of 246 guns (see table 2) available for the defense of the town.46 Con- cerning personal weapons, inspections of the inventories in the Ragusan armories and magazines from the years 1614–1617 reveals equally interesting data. The stockpiles of weapons stored there (see table 3) included more than 1,800 arquebuses, 680 muskets, and hundreds of cold weapons (swords, halberds, pikes, and other polearms) of various types. All in all, enough weapons were available to arm more than 4,000 additional defenders if necessary. Importantly, as mentioned previously, the Ragu- san government was not afraid to arm large numbers of its subjects in times of crisis: a fact that best testifies to the stability of the town’s aristocratic regime. We must also mention that the Ragusan government went to great lengths to ensure that the town was always well supplied with

Malta managed to endure an almost four-month siege of the Ottoman army in 1565 with a mixture of old and new fortifications, until it was finally saved by the arrival of the Spanish relief force. Yet, in 1570, in spite of all the money spent on building state of the art fortifications, Cyprus fell to the Ottomans almost without a fight, because of the Venetian Nicosia’s ill-prepared defense and their lack of will to resist. An even more relevant example can be found in the vicinity of Ragusa: in 1648, the Dalmatian town of Šibenik—whose position in the depression of the hills very much resembles that of Dubrovnik—was defended by only 4,000 soldiers and a system of quickly erected earthen ramparts as an addition to the old town fortifications. Nevertheless, they managed to endure twenty-eight days of siege against an Ottoman force 30,000 strong, until the outbreak of an epidemic and the arrival of a Venetian relief force broke the morale of the attackers and forced them to lift the siege. For more detail, see Duffy, Siege Warfare, 193–196. For the 1648 siege of Šibenik, see Gligor Stanojević, “Dalmacija u doba Kandijskog rata: 1645–1669” [Dalmatia in the time of the Cretan war 1645–1669], Vesnik vojnog muzeja 5 (1958): 111–112. On the composition of the defenders, see Archivio di Stato di Venezia [henceforth ASVe], Senato, Dispacci, b. 471, no. 32, Zara, 25. Maggio 1651 (attachment to the letter). 46 DAD, Guardie et Armamento, vol. 2, Inventarii delle armi e munizioni dell’ armamento dell’ anno 1614, fol. 76r–79v. 360 domagoj madunić weapons and armament. For example, when lightning set the armory in flames in 1610, resulting in the loss of large quantities of arms, the government immediately ordered the purchase of new weapons to replace the ones that had been lost.47

Table 2. Inventory of the Ragusan artillery train, including all the pieces positioned on the town’s fortifications and those stored in magazines, from April 161648 Artillery model Number of pieces Moratari 2 Falconi di 6 5 Falconi di 3 2 Falconeti da 5 5 Falconeti da 3 24 Cannoni di 100 2 Cannoni di 50 3 Cannoni di 30 2 Cannoni di 25 7 Cannoni di 20 3 Cannoni di 15 4 Mezzocannoni 2 Cannone Marcocco 1 Columbrine 3 Mezzacolumbrine 1 Sacri di 8 6 Sacri di 7 5 Sacri di 6 1 Passavolanti 5 Pedriere 17 Pedrale 8 Aspide 3 Bastarde 3 Smirigli et Saltamarini 115 Saltamarini 6 Smirigli 11 Total pieces 246

47 The purchase order set forth by Senate included 1,000 arquebuses, 200 muskets, 100 wall muskets, 100 sabers, 200 halberds, 600 helmets, and 400 pikes. Beritić, Dubrovačka artiljerija, 86–87. 48 The data are taken from DAD, Guardie ed Armamento, Inventatrio, vol. 2, Inventarii delle armi e munizioni dell’ armamento dell’ anno 1614, fol. 76r–79v. the defensive system of the ragusan republic (c. 1580–1620) 361

Table 3. Inventory of the weapons stored in the armories, fortifications, and magazines of the city of Ragusa and in various other localities on the territory of the republic49

Location / storage G rande) M uskets A rquebuses ( A rchibugi) G reat ( M oschetti M uskets ( M oschetti) H alberds ( A labarde) P ikes ( P iche) S pears ( L anze) P artisans ( S pontone) S words ( S pade schiavone) S words ( S pade bresciane) S cimtars ( S cimitare)

Main armory in Ragusa 1332 151 678 463 463 125 100 104 12 (armamento grande) Small armory in Ragusa 276 (armamento piccolo) Town forts and guard 52 6 58 30 12 posts (A) Total in town of 1660 151 684 521 493 125 100 12 104 12 Ragusa Armory of Ston 136 11 22 14 14 Other smaller storages 80 7 outside the town (B) Total armaments 216 11 29 14 14 outside the town Summary total (A + B): 1876 162 684 550 493 125 100 26 104 26

A closer look at the Ragusan artillery arsenal (table 2) reveals an interesting phenomenon common to the ordinances of the early modern period: the lack of standardization clearly visible in the large number of models ( falconi, falconetti, sarci, cannoni, passavolanti, smirigli, apside . . .) and calibers. Although these pieces were mainly cast in bronze, the balls they fired were made of diverse materials: stone, lead, and iron. All this diversity did not make the job of ammunition supply easy or simple. Consequently, a particular bastion, or a strong post at the town walls, would usually hold several different pieces of ordinance, a few barrels of gunpowder, and

49 Source for the data: DAD, Guardie et Armamento, vol. 2. Inventarii delle armi e munizioni dell’ armamento 1614. 362 domagoj madunić a collection of different types of cannon balls, of all sizes and materials.50 To keep the town ready in case of attack, large quantities of raw materials—saltpeter, iron, lead, fuses, and cannon balls—were stockpiled in state magazines in the customs house. Additionally, at various points along the town fortifications—for example at the bastions of St. Luke and of the Holy Savior, and in the fortresses of St. Lawrence, St. John, and Revelin—stores of gunpowder and ammunition were produced.51 In light of the above considerations, it can be argued that Ragusan fortifications, as they were, fitted their role quite adequately. Furthermore, whether consciously or not, by adopting a “low profile” strategy and not modernizing their fortifications, Ragusans managed to escape a deadly strategic trap. As pointed out by David Parrot, the construction of modern fortresses by second- or third-rank European powers, such as the Dukes of Gonzaga or Farnese, who by the construction of these fortifications meant to compensate for their military weakness, usually resulted in pre-emptive attacks by the great powers (France or Spain), who had, until that time, been ready to tolerate their independence. These actions were motivated by the strategic reasoning of the great powers and based on the fear that the balance of power would be seriously tipped if one of their rivals came into possession of such a valuable military asset.52 In the atmosphere of the Ottoman–Venetian–Spanish rivalry in the Adriatic, it is not difficult to envision a similar fate for the Ragusan state, had such a project ever been undertaken.

Supportive Infrastructure: Arsenals, Armories, Gun Foundries

Over the course of the centuries, in addition to its impressive fortifications, Ragusa also acquired a substantial infrastructure of supportive military facilities. Probably the oldest and most famous of these was the

50 The logistical problems are amply illustrated, for example, by the inventory of the bastion of St. Pietro, which was armed with six ordinances: three cannons (one of 40, one of 35, and one of 18) one columbrina, one passavolante, and one falconetto, and was supplied with 42 iron cannon balls, 14 balls of iron for columbrina, 28 balls of lead for falconetto, and 15 balls of lead for sacri. DAD, Guardie et Armamento, vol. 2, Inventario delle armi e munitioni dell’ armamento dell’ anno 1614, fol. 45v. 51 For example, on 1 December 1614, the ammunition stock at the Bastion of the Holy Savior consisted of 108 barrels and 10 cassettes of fine gunpowder, and 25 barrels and 9 cassettes of rough gunpowder. DAD, Guardie et Armamento, vol. 2, Inventario delle Armi et Munitioni dell’ Armamento dell’ anno 1614, fol. 63v. 52 Parrot, “The Utility of Fortifications in Early Modern Europe,” 135–137. the defensive system of the ragusan republic (c. 1580–1620) 363

Great Arsenal situated in the old town harbor. The exact date of construction of this facility is unknown, but the Ragusan Great Arsenal was undoubtedly—even in the fourteenth century—the largest such structure on the eastern Adriatic coast. Over the centuries, the Great Arsenal underwent several expansions, until, by the sixteenth century, it reached its current size and capacity to hold up to four galleys. In 1409, the Ragu- san government expanded its naval facilities by building a new, so-called Small Arsenal next to the old one. This building was designed to hold smaller warships such as brigantines or galeotte, and had the capacity to accommodate three such vessels. As a supplement to these two town arsenals, in the sixteenth century, the Ragusan government also began the construction of a small arsenal in Ston. In the fifteenth century, this facility was expanded to accommodate three smaller vessels, similar to the Small Arsenal in Ragusa.53 Like their much larger counterparts in Venice or Naples, the Ragusan arsenals were not simply state-owned shipyards that specialized in the construction of warships. They also served as depots for large quantities of naval stores and ammunition, necessary to maintain ships and keep the Ragusan war fleet prepared. As such, these arsenals represented the organizational and manufacturing foundations of Ragusan naval military power for centuries. This was where Ragusan warships were built, repaired, maintained, and stored, and without them, the existence of the state-owned war fleet, capable of being quickly put into action when needed would not have been possible. The large Ragusan artillery train from the 1610s was neither a novelty nor a recent acquisition. As Jan Glete points out, “guns had much longer lives than ships,”54 and so it happened that over the centuries, like other states, Ragusa acquired a variety of pieces, old and new. The Ragusans joined the European gunpowder revolution very early. The first artillery pieces were commissioned in 1351, and by the fifteenth century, Ragusa already owned a large number of guns. Moreover, the republic not only functioned as an important center for the proliferation of firearms in the Balkans, but also organized its manufacture. From 1410, a small foundry operated in Ragusa, and the massive fortification works of the fifteenth

53 Harris, Dubrovnik, 294–295; For an overview on the debate about the origin of the Grand Arsenal, see Ničetić, “Galije trireme i bireme,” 22–27. See also Lukša Beritić, “Dubrovački arsenali” [Ragusan arsenals], Mornarički glasnik 5 (1956): 577–586. 54 Jan Glete, Warfare at Sea, 1600–1650: Maritime Conflict and the Transformation of Europe (London, 2000), 24. 364 domagoj madunić and sixteenth centuries were supplemented by activities connected to the expansion and upgrade of the artillery train. The Ragusan gun foundry saw its golden age in the first four decades of the sixteenth century, under the management of Master Ivan Krstitelj (John the Baptist), originally from the island of Rab. From 1585, however, the position of master founder was left vacant, and the Ragusan government turned mainly to importations to satisfy its needs; it revived the casting of artillery pieces only occasionally during the seventeenth century.55 As a supplement to the gun foundry, two armories were organized, a large and a small one, where stocks of arms and armaments were repaired and stored.

15000 13345

erperi) 11250

7500

4178 3750 2821 2219 2221 early Expenditures (in Ragusan P

Y 1300 1052 425 886 0 1589 1594 1595 1599 1603 1605 1611 1612 1617 Year Graph 1. Overview of the yearly spending on works in arsenals and armories based on the available spese di armamento for the years 1589–161756

55 Harris, Dubrovnik, 297–298; Beritić, Dubrovačka artiljerija, 31–90, 157–160. 56 Data taken from DAD, Guardie et Armamento, vol. 21. Libro dell’ Armamento 1589; vol. 23, Libro dell’ Armamento del Anno 1594; vol. 25, Libro dell’ Armamento del Anno 1595; vol. 27, Armamento 1599; vol. 28, Scrivano del’ Armamento del’ anno 1603; vol. 29a, Spese di armament 1605–1607; vol. 36, Spese dell’ Armaento dell’ anno 1608; vol. 51, Libro spese dell’ Armamento 1612; vol. 53a, Libro del’ Armamento et Arsenale 1617. the defensive system of the ragusan republic (c. 1580–1620) 365

Like any other large military complex, Ragusa also required ongoing maintenance. Artillery pieces needed to have wooden carriages, which rotted, fixed; broken gunpowder cassettes and barrels had to be replaced; ships were drawn ashore and repaired during the winter months; stored weapons required cleaning and repair from time to time, and so on. Con- sequently, the Ragusan arsenals and armories employed dozens of craftsmen every year for these tasks. Although not all the accounting books that contain the records of payments to the craftsmen employed (the so-called spese di armamento) have been preserved, the surviving ones provide good orientation data on these activities. A closer look at the spese (Graph 1) reveals that spending differed significantly from year to year, depending on whether any major construction or repair projects were undertaken for that year, in addition to the usual maintenance works. As can be seen from the graph, the lowest recorded amount (425 Ragusan perperi 57 or roughly 127 ducats) is from the year 1589, and the highest (13,345 perperi) are recorded for the years 1605–1606, when two new galeotte were constructed.58 If one disregards the data for the years with the lowest and highest recorded spending (1589 and 1605 respectively), the average expenditure of the Ragusan government on the works in arsenals and armories amounted to the rather low sum of 2,090 perperi a year (roughly 600 ducats). As an illustration of what the structure of costs looked like, let us consider the data from the year 1603, that is, when the second highest amount was recorded. In that year, 884 perperi (21.16%) went to artillery maintenance and repairs, 537 (12.87%) to maintenance works in the armories, 348 (8.34%) were spent on small construction works in the arsenal, and a further 345 perperi (or 8.26%) on the repair of the great chain that closed the entry to the town harbor. Since there were no major shipbuilding projects in that year, a rather low sum of 342 perperi (8.20%) was spent on the maintenance of the warships. The production of gunpowder and the construction of the gunpowder barrels and chests consumed an even smaller amount of 138 perperi (3.31%). Finally, some 1,204 perperi (28.90%) went to the purchase of various materials (iron, lead, rope, wood, stone, etc.)

57 Perperi were silver coins in circulation in Ragusa; in the period of this study, one perper was valued at 12 grossi. Carter, Dubrovnik, 560. 58 Work on the construction of small galleys continued well into 1607, when a third galeotta, one with 22 benches—thus the largest of the three—was built. Expenditures related to these works were kept in a separate book, and contain data only for the period between June and August 1607, recording a spending of an additional 2,858 perperi. DAD, Guardie et Armamento, vol. 50, Spese delle Galeotte 1607. 366 domagoj madunić used by arsenals and armories, while the rest (9%) of the expenses represent various uncategorizable spending (mainly transport costs, expenditures on the festivities of St. Blaise, and so forth).59 All the tasks outlined above required skilled craftsmen. According to spese, the highest demand was for the various wood workers, such as carpenters, shipbuilders or sawyers, who represented the main workforce in a large variety of tasks, from small construction repairs of buildings or fortifications to artillery maintenance and complex shipbuilding projects. For instance, in 1594, woodworkers constituted almost 63 percent of the craftsmen employed. The second category of multipurpose workers were blacksmiths or iron workers, whose services were also required in a wide range of jobs (in 1594, they made up around 10 percent of the craftsmen employed), followed by more specialist professions: weapon smiths, rope makers, oar makers, barrel makers, and so forth. At this point, a word or two concerning wages are in order. Manual workers or apprentices, like the rowers on ships and those drafted into military service, were paid two grossi a day, an amount that represented the standard daily wage for unskilled laborers in seventeenth-century Ragusa.60 The daily wage of carpenters or shipbuilders was between 6–8 grossi per day, while those of various other master craftsmen ranged between 10 and 12 grossi. However, not all craftsmen were paid on a per day basis. Blacksmiths and oar makers, for example, were paid upon the completion of the specific job assigned to them. For instance, in 1603, the blacksmiths Georgio de Andrea and Marco and their apprentices were paid a total of 185 perperi for the repair of the great harbor chain.61 Black- smiths, however, were more often than not paid for melting the iron, and the amount paid to them depended on the quantity of the iron melted and on whether the raw material was provided by the state or not. Simi- larly, oar makers were paid for each oar manufactured and so on. It is also interesting to note that the wages of a particular profession were almost inversely proportional to the number of craftsmen employed.

59 Such a large percentage of the expenditure for the purchase of materials for arsenal and armories (28.9%) is a consequence of the source, which does not clearly separate purchases of materials for a specific project and those that represent the general replenish- ment of the stockpile of naval stores and munition. DAD, Guardie ed Armamento, vol. 28, Scrivano del’ Armamento del’ anno 1603. 60 See, for example, DAD, Guardie et Armamento, vol 23, Libri deli Armamento del Anno 1594, passim; or DAD, Guardie et Armamento, vol 29a, Officiali dell’ armamento dell’ anni (1605–1607), passim. 61 DAD, Guardie ed Armamento, vol. 28, Scrivano del’ Armamento del’ anno 1603, fol. 2r. the defensive system of the ragusan republic (c. 1580–1620) 367

This fact is most visible in the rather low salaries of carpenters; their salaries were just a little higher than the average daily wage of 4–5 grossi paid to the Ragusan soldati. By contrast, the wages of less common specialists, such as weapon smiths (spadori), or bombardieri, who had the skills necessary to be employed to repair firearms, were considerably higher, 8–12 grossi a day. Moreover, foreign experts imported to Ragusa for a specific job and tasked with transferring knowledge to local craftsmen, could expect to earn quite handsome salaries. Such, for example, was the case of the two previously mentioned master shipbuilders brought from Naples to supervise the manufacture of Ragusan small galleys: they were (together) paid 33 ducats (or 110 perperi) per month.62 Oddly, the aforementioned spese from 1603 show very low expenditures for gunpowder: only 138 perperi. For the supply of this important resource, Ragusa was self-sufficient, and the spending related to its production and storage usually amounted to much larger sums. For example, in 1594 around 2.4 metric tons of gunpowder were produced at the cost of 502:3 perperi or 25 percent of the entire yearly expenses recorded in the spese.63 For the manufacture of gunpowder, specialists from the ranks of the bombardieri were commissioned (usually by their prota), who were paid extra for each order. The importance of this task can best be deduced from the fact that from 1612 onward, the Ragusan government employed a foreign specialist among the bombardieri, the so-called pulverista, who was charged with the production of gunpowder and was paid 10 ducats a month (in addition to the usual bonus given on the per order basis). This was the highest salary given to any Ragusan military personnel.64 Without all these people, the proper functioning of the Ragusan military infrastructure would not have been possible. Fortunately, Ragusan arsenals and armories could call upon a centuries-old tradition of urban life, rich with industry and manufacturing, and turn to the local private sector in order to satisfy their needs for workforce. The presence and availability of these skilled workers made Ragusa almost self-sufficient in this regard and minimized (although did not entirely eliminate) the need to import

62 DAD, Guardie et Armamento, vol 29a, Officiali dell’ armamento dell’ anni (1605– 1607), fols. 3v, 6r, 13v, 17v, 27r, 29r, 31v, 34r. 63 DAD, Guardie et Armamento, vol. 23, Libri deli Armamento del Anno 1594, fol. 29v. 64 Beritić, Dubrovačka artiljerija, 87. The decision to bring to Ragusa, with the help of Vice Buna, the Ragusan consul in Naples, a magister pro facendum pulverum was reached in the Senate meeting of 24 November 1611. DAD, ACR, vol. 83, fol. 39v. 368 domagoj madunić foreign technical experts. Moreover, as we argue later, this represented one of the greatest strategic assets of the Ragusan state.

Final Remarks: Regarding the Military Importance of Ragusa

Although the Republic of Ragusa managed to remain neutral and avoid involvement in any major European conflict during the entire early modern period, it was, as this paper shows, neither a demilitarized, nor an unarmed state. Compared to other major powers present in the Adriatic—the Republic of Venice, the Spanish monarchy and the Otto- man Empire—the resources of this small republic may appear quite small and insignificant. Ragusans themselves were acutely aware of their weakness and of the disproportionate military superiority of neighboring states. However, this military weakness, when transferred to the sphere of diplomacy, was in fact easily transformed into an advantage. It served as an effective excuse to refrain from any kind of military involvement on anyone’s behalf, as is well illustrated by the following case. In June 1617, during the Spanish–Venetian war, the Ragusan government received commands from the grand vizier to ensure that no harm should reach the empire through their territory, neither to the lands of the Grand Signor, nor to that of the Venetians, and that no excuses would be tolerated in regard to this issue. In response, the Ragusan government instructed its ambassadors in Constantinople to counter this demand, by stating that: “such feats are not in our power, but rather in the power of the Gran Signore and the signori Veneziani, who possess armies and fleets; in respect to them we are just like a worm(!)”; adding that “we have put ourselves under the protection of the Gran Signore, to that end that He should be the one defending us, and not this tiny city defending His Majesty.”65 Yet we must also take into account that in spite of their apparent inex- haustibility, the resources of the major powers, like Spain or the Ottoman Empire, were far from endless. Quite the contrary, because of their ongoing engagements on several fronts, major powers almost always found

65 “In detti commandamentim e lettera, ci uiene aduertito che non sequa nessun danno ai paesi del Gran Signore nè a i Signori Venetiani, perche poi non ci valera nessuna escusa- tione, pero direte che questo non sta in potere nostro, ma nel Gran Signore, e nei Signori Venetiani, che hanno armate; perche noi siamo un verme; rispetto alla gran potenza loro; e pero ci siamo messi nella prottettione del Gran Signore (f. 88v) a fine che egli difenda noi, e non questa piciol citta à sua Maestà.” DAD, Lettere di Levante, vol. 43 (1615–1622), fols. 88r–v. the defensive system of the ragusan republic (c. 1580–1620) 369 themselves overstretched, and thus their military presence in secondary theaters of operation was almost always very limited. In such circumstances, offers of alliances and prospects of modest military support by minor states in the border zones of their influence came to play an important role in the strategic considerations of major powers. The above also applies in the case of the Ottoman–Ragusan relations. Apart from a few brief periods between 1594 and 1600, when the Ottomans attempted to levy a few hundred azaps from the territory of Konavle, their calls for military aid from the Ragusan government consisted mainly of requests for military specialists or logistical services.66 In the 1620s, Ragusan ships were regularly called on to provide transportation for large quantities of tar from the port of Valona—where it was produced—to the shipyards of Istanbul.67 After the defeat at Lepanto, the Ragusan government was asked to send several master shipbuilders with their apprentices to the Ottoman port of Herceg-Novi.68 Similar requests were repeated during the long Ottoman–Venetian war over the island of Crete (1645–1669). For example, in October 1649, after the successful sabotage attack on Ottoman ships docked in the port of Herceg-Novi that resulted in the destruction of three fustas, the Ragusan government was asked for assistance in the form of skilled master shipbuilders and workers to help to rebuild the lost ships.69 Of even greater strategic importance were the Ragusan artillery pieces and armed vessels. As the war dragged on and both Venetian and Otto- man resources began to dwindle, these untapped Ragusan resources grew in importance. Throughout the course of the war, Venetian officials in Dalmatia kept a close watch over Ragusan naval activities. Each time the Ragusan government armed itself and launched a ship, a report was sent to Venice, and the Ragusan government was duly informed by the Venetian governor-general (provveditore generale) in Dalmatia and Alba- nia that the Serenissima considered the Gulf as its dominion and that it

66 All such attempts were initiated by the local Bosnian officials and were quickly met with Ragusan diplomatic counter-action in Istanbul, which prevented their execution. In May 1600, again in May 1601, and finally in July 1604, Ragusans managed to obtain several imperial decrees that forbade such levies of their subject, since “this is in opposition to the Ragusan capitulations and is against the will of the sultan”; Popović, Dubrovnik i Turska u XVI veku, 371; Miović, Dubrovačka Republika, 202–204. 67 Miović, Dubrovačka Republika, 55–57. 68 Popović, Turska i Dubrovnik u XVI veku, 290–292. 69 ASVe, Senato, Dispacci, PTM, b. 469, no. 635. Zara, 16. Ottobre 1649. 370 domagoj madunić would not tolerate armed vessels in it, other than its own.70 Moreover, the risk of Ottomans activating Ragusan naval resources and thus tipping the balance in favor of the steadily increasing number of Ottoman corsairs in the Adriatic—an event that would have seriously threatened Venetian supply lines to the Levant—prompted Venetian officials not only to keep track of Ragusan activities, but also to diligently gather information on Ragusan plans to build new ships.71 Equally troubling for the Venetian high command was the Ragusan artillery arsenal. The problem faced by the Ottomans in the Dalmatian theater of operations was not lack of artillery, but rather its availability on the spot. Transporting large siege pieces on land was both slow and costly. It required a large number of draft animals, which in turn consumed huge quantities of fodder, thereby making this kind of operation possible only during the summer months, when, in addition to the supplies they had to carry, an advancing army could also count on forage from the coun- tryside. Furthermore, in the case of inhospitable, barely passable terrain and/or lack of roads, both the cost of transport and the time required for travel grew exponentially. In the early modern period, every military commander attempting a siege struggled against the same equation: the longer the time of the travel, the shorter the available time for the siege. In this regard, the proximity of Ragusa—and its artillery train of 240 guns—to the Venetian stronghold of Kotor (less than 60 kilometers away), situated in the Bay of Boka and otherwise distant from any major Ottoman logistical base, represented a real strategic risk for the Venetian war effort. Thus, it comes as no surprise that the Venetian government followed with great interest Ottoman plans to request that Ragusans supply guns,

70 For example, in 1651, a report was sent to Venice with the news that Ragusa, alarmed by the appearance of Berber corsairs in their waters, was in the process of arming “five felucche con sedici huomini per una, et una fregatta cattarina con trenta huomini.” ASVe, Senato, Dispacci, PTM. b. 472, no. 56. Di Galea in Viaggio per trau, 2. Settembre 1651. Simi- lar news reached Venice in 1655, namely that the Ragusans were arming one small galley (galoetta) and two bregantins. These ships were launched one month later, and in accordance with the orders of the Venetian Senate from 10 September, the governor-general informed the Ragusan government that guarding the Gulf was the sole responsibility of Venice. ASVe, Senato, Dispacci, PTM, b. 478, no. 52. Zara, 3 Luglio 1655; ASVe, Senato, Dispacci, PTM, b. 478, no. 56. Zara, 14 Agosto 1655. 71 How well the Venetian government was informed is illustrated by the events of June 1663, when the governor-general of Dalmatia informed his superiors that he had received information concerning a decision of the Ragusan Senate to build two galleys with the help of a foreign master shipbuilder, Protto dell’ Arsenal di Sicilia, who was expected to arrive in Ragusa soon. ASVe, Senato, Dispacci, PTM, b. 490, no. 76. Sebenico, 11. Giugno 1663. the defensive system of the ragusan republic (c. 1580–1620) 371 ammunition, and artillery crews as a part of the preparations for the attack on Kotor. In the spring of 1663, the Venetian Senate, prompted by rumors that the Ottomans were planning a major attack on Kotor and that the Ragusans had been asked to provide the artillery, ordered its governor-general in Dalmatia to find out the exact number of ships and cannons available in Ragusa, and to investigate the attitude of the Ragusan government on this issue.72 Soon, the governor-general informed the Senate that the Signori Ragusei had instructed their emissaries to the Porte in the following way: should the grand vizier express his intention to personally supervise the operation and designate Ragusa as a place for his accommodation, they should dissuade him from such intent by any means, from offering expensive gifts to explaining that, if they accom- modated him, after his departure, since it is located on the coast, Ragusa would be destroyed by Venetian arms.73 Fortunately, both for Venice and Ragusa, nothing came of this. That year the Ottoman army went on a campaign in Hungary, and the only shots fired from Ragusan guns in the “service” of the Ottomans during this war remained the honorary salvos from the city walls that commemorated the succession of the new sultan to his throne.74 Whether these rumors held any truth or not is of no importance for this paper; they provide a good illustration of the logic of contemporary strategic thinking and reveal the military importance of Ragusa in the Adriatic basin. The real military value of the Ragusan republic was not in its effective military force, but rather in its military facilities and stocks of war materials. These three factors: (1) a small supportive military infrastructure organized around three arsenals and two armories, (2) the availability of skilled workers needed for military efforts, and (3) large depots of munitions, arms, and other war materials made Ragusa one of the most valuable military assets in the region. In the first half of the seventeenth century, Ragusan shipyards were still among the largest on the eastern coast of the Adriatic, surpassing any nearby Ottoman port in size and capacity and matching the Venetian facilities at Korčula. Its fortifications boasted the highest level of firepower in the region, whereby the 140 guns

72 ASVe, Senato, Dispacci, PTM, b. 490, no. 60. Zara, 17. Marzo 1663. 73 ASVe, Senato, Dispacci, PTM, b. 490, no. 66. Zara, 11. Aprile 1663; Correspondence between the governor-general and the Ragusan government: ASVe, Senato, Dispacci, PTM, b. 490, no. 70. Zara, 10. Maggio 1663. 74 Salvoes were fired in honor of the enthronement of Mehmed III (1595–1603), and Ahmed I (1603–1617). Miović, Dubrovačka Republika, 59–60. DAD, Guardie ed Armamento, vol. 25, Libri deli Armamento del Anno 1595, fol. 3r. 372 domagoj madunić positioned along the walls of Ragusa by far surpassed the 95 pieces positioned on the walls of Zadar, which was regarded as the strongest Vene- tian stronghold in Dalmatia.75 Furthermore, although Venetian arsenals and armories in Zadar were similar in size to those of Ragusa, they lacked any local support industry. In contrast to Ragusa, where oars, nails, sails, and other necessary products could be produced locally, the facilities in Zadar, because of the lack of local productive capabilities, had to import everything, from nails to sails, from Venice.76 This was the prism through which the military strategists of the seventeenth century saw the Ragusan republic and what made it an interesting item in their calculations.

75 The figure is taken from the survey of the military and economic resources of Dal- matia assembled by Giovvani Battista Grimani, the Provveditore Generale in Dalmatia and Albania, in 1644. For the document known as “Inserto Grimmani,” see Šime Ljubić, Com- missiones and Relatione Venetae (Zagreb, 1972), vol. 6, 220. 76 For more detail on the arsenal in Zadar, see Michela Dal Borgo and Gugliemo Zanelli, Zara: Una fortezza, un porto, un arsenale (secoli XV–XVIII) (Rome, 2008), 38–48. Section four

Instead of a Conclusion: on the “Compositeness” of the Empire

The System of Autonomous Muslim and Christian Communities, Churches, and States in the Ottoman Empire*

Sándor Papp

Should a people break the treaty, Allah will give power to his enemies upon them Molla Hasan El-Kafi1 The historiography on the Ottoman Empire draws a clear dividing line between the vilayets and sancaks. These units were characterized by a government that can be labeled classical, in contrast to the territories that kept their administrative structure, or even their ruling dynasty, from before the time of conquest. European literature earlier generally used the term “vassal state”; this has been replaced by the concept of “tributary state”—which can be traced back to the Ottoman word haracgüzar—even for the Muslim and Christian vassal states that never paid tribute. In Ottoman terminology, the terms used for vassals are teba‛a or teba‛iyet. The Ottoman Empire saw the vassal states as its own territory in the well-protected empire (memalik-i mahruse). Their subordination was manifested by compulsory services (apart from the incidental tribute payment) and the sultan’s right of investiture over the vassal ruler. From

* While writing this study I received help from many colleagues, whom I want to thank here. My research in Istanbul and Vienna was kindly assisted by Feridun M. Emecen and István Fazekas, respectively. Part of my research was funded by GWZO (Leipzig), and for their help I am grateful to each member of the Ottoman project, including the head of the project, Robert Born, Nedim Zahirović, Norbert Spannenberger, and, last but not least, Gábor Kármán, who also gave me useful advice for the final written version of this paper. I would not have been able to track down Czech literature without Péter Kónya, professor at the University of Prešov. During my research I also received generous financial help from the Andrew W. Mellon Foundation, the Humboldt Stiftung, and the Habsburg Insti- tute (Budapest). 1 Imre Karácson, Az egri török emlékirat a kormányzás módjáról. Eger vára elfoglalása alkalmával az 1596. évben írta Molla Haszán Elkjáfi [A Turkish memorial on the modes of government in Eger: Written on the occasion of the occupation of the castle Eger in 1596 by Molla Hasan El-Kafi] (Budapest, 1909), 24; Hasan El-Kafi, Nizamü l-‛Alem [The order of the world], Magyar Tudományos Akadémia, Keleti Gyűjtemény, Török, O.420/I. 376 sándor papp the Ottoman viewpoint, the core of the relationship was the premise that the vassal state was under the sultan’s protection; this was conceptualized with the terms himaye, ‛inayet, emn u aman, ‛ahd u misak (variations on the words for “protection” and “treaty”). I was drawn into this field of problems by the debates surrounding the relationship of Transylvania to the Ottoman Empire in Hungarian historiography. The analysis of this topic is supported by a large number of source publications from the eighteenth and nineteenth centuries; however, the discussion has been hindered by the fact that most students of this largely Hungarian material lack the background in Ottoman history and the Islamic legal system, background that is necessary for their interpretation. Because of this, the idea was established that the prince of Transylvania enjoyed full internal freedom as long as he paid the tribute faithfully and did not act against the interests of the sultan. Many analyses concentrated on cases in which the activities of the princes who waged wars against Habsburg rulers coincided with the political goals of the Porte. During these periods, the level of the princes’ subordination was not clearly visible, and this led historians to promote a broadly conceived though erroneous thesis of a quasi-independent Transylvania. In the last few years I have dedicated a series of works to clarifying this question and have also started to publish the documents of inauguration granted by the sultans to the princes of Transylvania. These show a fairly clear picture of sixteenth-century developments.2 My recent work aims to map the system of autonomies in the Ottoman Empire—those granted on religious, communal, or on a state basis—in order to better understand Transylvania’s place in it.

2 Sándor Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden der Osmanen für Ungarn und Siebenbürgen: Eine quellenkritische Untersuchung (Vienna, 2003). See also my other studies on the topic: “Macar ve Erdel Kralı olan Bocskai’nin Halefleri (Tadil Edilen Bir Berat-ı hümayun)” [The successors of Bocskai, king of Hungary and Transylvania: A cor- rected berat-i hümayun], in CIÉPO Osmanlı Öncesi ve Osmanlı Araştırmaları Uluslararası Komitesi XIV. Sempozyumu Bildirileri (18–22. Eylül 2000, Çeşme), ed. Tuncer Baykara (Ankara, 2004), 557–580; “Die Osmanische Diplomatie des Rákóczi-Freiheitskampfes (1703–1711),” in Studii şi cercetări de Turkologie contemporană: Omagiu Professorului Mihai Maxim, ed. Călin Felezeu (Cluj-Napoca, 2004), 73–87; “Critical Survey of Letter of Donation, Confirma- tion and That of Agreement Issued by the Ottoman Sultans for the Ruler of Hungary and Transylvania,” Chronica (Szeged) 4 (2004): 126–128; “Thököly Imre és a törökök: A kuruc állam helye az Oszmán Birodalomban” [Imre Thököly and the Turks: The place of the kuruc state in the Ottoman empire], in Gróf Imrich Thököly a jeho povstanie / Thököly Imre gróf és felkelése, ed. Péter Kónya (Prešov, 2009), 255–267; “Thököly Imre erdélyi fejedelem- sége és a Porta” [Imre Thököly as prince of Transylvania and the Porte], in Az üstökös kegyeltje: Gróf Késmárki Thököly Imre, ed. Sándor Gebei (Hajdószoboszló, 2010), 191–208. the system of autonomous muslim and christian communities 377

Any analysis of the vassal states of the sultan faces a larger problem: we still lack systematic, comparative research on the topic. The works in historical geography by Donald Pitcher and Andreas Birken do discuss every vilayet, sancak, and vassal territory, but on the basis of older literature and sources published in Western languages; no targeted archival research has yet been done on the question.3 On the category of Christian vassals to the seventeenth century, such research should take into consideration the following state formations (the dates are sometimes hard to establish due to the scarcity of sources): Byzantium (1362/63–1453), Bulgaria (1364–1393), Serbia (1389–1444), Wallachia (1394/95–1877), Moldavia (1455–1877), Hun- gary (1529–1541), Transylvania (1541–1699), Middle Hungary (Orta Macar) led by Imre Thököly (1682–1686), the Cossack hetmanate (1650–1654, 1669–1676), and in some periods even the Caucasian states of Mingrelia, Georgia, and Kartil. As my fundamental aim is to discuss the vassal status on the basis of the treaties and documents of confirmation and inauguration that were created during the practice of the sultan’s right of investiture, I have limited my examples to those for which I was able to find one or more such document. I hope to broaden the circle by continuing my research, but I believe that the data I have at the moment already allows me to draw a general picture. In this study, I discuss the Muslim and non-Muslim vassal states separately, because—as I argue in the following—their status originated from different legal and historical sources. In this respect, I follow a scheme different from that used in earlier literature, notably that of Andreas Birken. The structure of the Ottoman Empire, despite a strong aspiration for centralization among its rulers, had a certain “conglomerate” character throughout the entire period of its existence. As theoretically all power came from the person of the sultan, each officeholder—from the most menial to the most powerful, including the heads of autonomous communities and vassal states—was connected to the center through the inauguration procedure (where it was applicable) and the documents canonizing the community’s rights and obligations. In the following examples, I refer to a variety of document types following the terminology of the Ottoman chancellery. To make my analyses easier to follow, I explicate the general

3 Donald Edgar Pitcher, An Historical Geography of the Ottoman Empire from the Earliest Times to the End of the Sixteenth Century (Leiden, 1972); Andreas Birken, Die Provinzen des Osmanischen Reiches (Wiesbaden, 1976). 378 sándor papp character and role of the document types used in the practices related to the vassal states; I do not go into detail here concerning the large number of variations, which are duly commented upon during the analyses of individual cases. The document type most often used in the vassal relationship was the ‘ahdname. Its application to the cases of tributary rulers is important, as the trade agreements and peace treaties (capitulatios) with sovereign rulers, such as the Habsburg Empire, Poland, France or Venice, were also labeled as ‘ahdname (or one of its variations, sevgend-name, sulhname, mu‛ahede) in Ottoman Turkish diplomatic language. There were various types of ‘ahdname, used according to the purpose of their issuance or the Porte’s relationship with the communities to which they applied. Perhaps the most common was the nişan-type ‘ahdname, which received its name from its introductory nişan formula.4 In contrast to this, the name-type ‘ahdname was seldom used for vassals—it was mainly reserved for foreign rulers.5 The text of this document type, which was usually issued in a decorative format, differs from the nişan type by the two-part structure of the invocatio (da‘vet) formula. This includes a smaller section at the top of the document stating that it is issued in the name of God, with “Hüve [He, God]” or its variations. It is then followed by an invocatio of three to

4 “Nişan-i şerif-i ‘ali-şan-i sultani ve tuga-yi garra-yi giti-sitan-i hakani nefeze bi-l-‘avni r-rabbani ve-s-savni l-samedani ve-l-menni l-sübhani hükmi oldur ki” [The order of the noble, renowned and high-ranking sultan’s signature and the splendid country-conquering tuğra—should it receive divine support and celestial help—is the following]. In his 1991 dissertation, Hans Peter Alexander Theunissen collected the structural elements of the ‘ahdnames given to Venice, from which it is clear that the formula remained relatively unchanged between 1482 and 1641; see his “Ottoman–Venetian Diplomatics: The ‘Ahd-names: The Historical Background and the Development of a Category of Political- Commercial Instruments Together with an Annotated Edition of a Corpus of Relevant Documents,” Electronic Journal of Oriental Studies 1, no. 2 (1998), 258–259. A variant with similar meaning is found in the sultan’s treaty letter to István Báthory: “Tevki‘-i refi‘-i hümayan vasıl olıcak ma‘lum ola ki” [As the majestic signature of the great lord arrives, it should be acknowledged] Feridun Ahmed Bey, Mecmu‘a-i münşe’atü s-selatin [A collection of sultans’ documents], vol. 2, 2nd ed. (Istanbul, 1275 (AH)), 461; Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 214–219 (with German translation). It must be noted that the most characteristic nişan-type ‘ahdnames with the purest structure are the ones given to Dubrovnik, where the nişan formula is followed immediately by the expositio/ narratio part (in Turkish, iblag), which usually starts with the words bundan akdem, bundan evvel [earlier, before this]. 5 Some examples include Anton C. Schaendlinger and Claudia Römer, eds., Die Schrei- ben Süleymâns des Prächtigen an Karl V., Ferdinand I. und Maximilian II. aus dem Haus-, Hof- und Staatsarchiv zu Wien, vol. 1, Transkriptionen und Übersetzungen, and vol. 2, Fak- simile (Vienna, 1983); Dariusz Kołodziejczyk, Ottoman–Polish Diplomatic Relations (15th– 18th Century): An Annotated Edition of ‘Ahdnames and Other Documents (Leiden, 2000). the system of autonomous muslim and christian communities 379 five lines, stating that the sultan’s rule enjoys the sacred benevolence of God, the Prophet, the first four caliphs, and every saint.6 The berat—also called nişan or menşur—is the document type that was generally used in the Ottoman Empire to acknowledge inaugura- tions; it was issued when a privilege was conferred, ranks, titles or military estates were granted, and when state offices were filled. As in the case of the ‘ahdname, there were also various types of berat; I know of only one case when something other than the nişan type was used by the Porte in connection to vassals.7 The starting formula of these berats is similar to those of the nişan-type ‘ahdname, but the following intitulatio/‘unvan is radically different. Starting with the word çun [as], the sultan does not list his name and titles, but rather states that by divine power, for which he is grateful, the center of his empire was made to be the refuge of the world and the defense of the persecuted. He also stresses that the Porte stands open to all who turn to it and that those who plead for help will not be turned away. Other differences are rooted in the contents of the documents; they cannot, however, provide a decisive argument because demarcating them as the fields where one or the other was applied was also not strictly prescribed. In the case of berats, the dispositio/emir or hükm section—apart from inaugurating the new ruler—lists his obligations, whereas the corresponding parts of the ‘ahdnames regulate the addressee’s relationship with the Sublime Porte; some cases are known, however, where the dispositio section of the berat shows elements characteristic of an ‘ahdname.8 The best solution is to define the document type through the word by which their texts refer to themselves, even if some sources contain more than one designation used as synonyms; in these cases, where the words ‘ahdname and berat are found in the same document, it can be regarded

6 “Hazret-i rebbü l-‘izzeti cellet kudretuhu ve ‘alet kelimetuhunun ‘inayet bi-gayeti ve mihr-i sipihr-i nübüvvet ve ahter-i burc-i risalet pişva-yi firka-i asfiya ve mukteda-yi zümre-i enbiya iki cihan fahrı Muhammed Mustafa salla llahu te‘ala ‘aleyhi ve sellemün mu‘cizat-i kesiretü l-berekatı ve çehar-yar-i güzin-i rizvanu llahu te‘ala ‘aleyhim ecma‘in ve sayir evliya-yi ‘izam ve etkiya-yi kiramün ervah-i mukaddiseleri mürafakatıyle.” Franz Babinger, “Zwei türkische Schutzbriefe für Georg II. Rákóczi, Fürsten von Siebenbürgen, aus dem Jahre 1649,” Le Monde Oriental 14 (1920): 126–149. 7 This was the document that appointed István Bocskai king of Hungary and prince of Transylvania in 1604, see Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 261–263. 8 See the berats given to the voievods of Moldavia and Wallachia and Mihály Apafi I, prince of Transylvania in the relevant chapter sections. 380 sándor papp as a transitional type.9 Although these criteria do not always allow for a clear distinction, it is important to differentiate between these two types because—as we see below—many data suggest that the choice among them was important to the Ottoman chancellery.10

Muslim States and Communities under the Rule of the Ottoman Empire

Caramania From the early Ottoman period the example of Caramania promises the most comprehensive documentation, because the text of the 1444 peace treaty between Karamanoğlu II İbrahim Bey and Sultan Murad II is preserved, even in two copies. This document, the earliest one acknowledging a vassal relationship with the Ottomans, was created in 1443–44 after Caramania attacked the Ottoman Empire as an ally of the Kingdom of Hungary.11 Although the first publisher of the text, İ. Hakki Uzunçarşılı, discussed it as a sevgend-name (written oath), the text refers to itself unambiguously as mu‛ahede and ‛ahd, thereby defining its own genre as ‛ahdname. The copy preserved in Konya has 46 lines, out of which the first seven express the position of the Karaman bey, İbrahim II, as a subject of the sultan. The making of the treaty is described in the narrative part: İbrahim Bey sent a letter to Murad II begging for mercy, asking that the sultan should pardon his animosity. This first offer of peace was declined by the sultan and thus it had to be repeated. Surprisingly, the document mentions not only Sultan Murad II as the contracting party, but also his

9 I have only seen such documents in the case of the princes of Transylvania in the first half of the seventeenth century, see the relevant chapter section. 10 See the analyses concerning the changes in the document type given to the princes of Transylvania, the appointment of Imre Thököly in 1682, the appointment of György Rákóczi II during the life of his father, and the collective ‘ahdname given to the Kurdish dignitaries. 11 Halil İnalcik and Mevlud Oguz, Gazavat-i Sultan Murad b. Mehemmed Han: İzladi ve Varna Savasları (1443–1444) Üzerinde Anonim Gazavatname [The conquests of Sultan Murad b. Mehemmed han: An anonymous chronicle about the campaigns of Slatica and Varna, 1443–1444] (Ankara, 1989), 33; Gábor Ágoston, “Az 1444. évi török követjárás (Adalékok az 1444. évi török–magyar békekötés történetéhez)” [The Ottoman diplomatic mission of 1444: Contribution to the history of the Ottoman–Hungarian peace of 1444], Történelmi Szemle 29, no. 2 (1986): 262–263; Sándor Papp, “II. Murád szultán és I. Ulászló lengyel és magyar király 1444. évi békekötése” [The peace treaty between Sultan Murad II and Wladyslaw I, king of Poland and Hungary, in 1444], Acta Historica (Szeged) 109 (1999): 47–62; for a German version of the article, without the edition of the text: idem, “Der ungarisch–türkische Friedensvertrag im Jahre 1444,” Chronica (Szeged) 1 (2001): 67–78. the system of autonomous muslim and christian communities 381 son, Mehmed II. The Karaman bey promised the sultan that he would deliver every document of the confederacy against him and that he would not accept anyone else’s offer to return the territories taken by the Otto- mans. He also vowed to send the sultan one of his sons with auxiliary forces each year. This document deserves special attention from a diplomatic point of view. Although it does not prescribe the payment of tribute, all of the conditions show that the Karaman bey became a subject of the sultan. At the same time, it is clear that in this early Ottoman period the agreement with the vassal ruler was bilateral, as both preserved copies are the Kara- man versions of the contract. It should also be noted that the element later found only in the inauguration documents of the subjected Christian rulers, that the vassal must be the friend of the sultan’s friends and the enemy of his enemies (“dostlarına dost ve düşmenlerine düşmen oldum”), also appears in this treaty. The structure of the document is probably related more to the pattern of treaties with Anatolian beyliks than to the later practice of the Ottoman chancellery.12

Crimea The khans of the Crimea had a peculiar place among the states belonging to the Ottoman Empire, as the dynasty of the khans, the Giray family, had a much more prominent lineage than the House of Osman.13 On the basis of the rather limited material, one can conclude that originally there was a contractual relationship between the sultans and khans, which means that these two Islamic powers maintained equality in their diplomatic contacts. The nature of their relationship changed to a certain level of vassalage in 1475, after the Ottoman occupation of Kaffa and the establishment of a vilayet there. Until the late sixteenth century, the khans could

12 İsmail Hakkı Uzunçarşılı, “Karamanoğulları Devri Vesikalarından İbrahim Beyin Kara- man İmareti Vakfiyesi” [A document from the time of Karamanoğulları: The foundation charter of a public kitchen by Karaman İbrahim Bey], Belleten 1, no. 1 (1937): 1–127; Aladdin Akgöz, “Karamanoğlu II. İbrahim Beyin Osmanlı Sultanı II. Murad’a Verilmiş Olduğu Ahid- nâme” [The ‘ahdname given by Karaman İbrahim Bey II to the Ottoman sultan Murad II], Tarih Araştırmaları 24 (2005): 159–178. 13 On the position of the khans, see Mária Ivanics, A Krími Kánság a tizenöt éves háborúban [The khanate of Crimea in the Fifteen-Years’ War] (Budapest, 1994), as well as her chapter in this volume. On the independent international diplomacy of the khanate, see also eadem, “Der Bündnisplan zwischen dem Khan Gazi Giray II. und dem Fürsten Sigismund Báthory aus dem Jahre 1598,” in Hungarian Ottoman Military and Diplomatic Relations in the Age of Süleyman the Magnificent, ed. Géza Dávid and Pál Fodor (Budapest, 1994), 183–193. 382 sándor papp nominate their successors, who were confirmed on the Crimean throne by the sultans with a berat-i hümayun. After the 1624 revolt of Khan Mehmed Giray IV and his brother, Kalga Şahin Giray, relieving and appointing the khans became primarily dependent on the will of the sultans.14 It must be emphasized, however, that later there were also chances to increase the khans’ power, especially in those periods when the military-geopolitical role of the Crimea became important. The first letter of appointment given to Khan Canibek Giray that has been preserved dates from 1628.15 My archival research uncovered three further inauguration berats given to the khans; these are quite similar in their structure and wording.16 The berat-type of document was originally used for donation letters in the case of private lands (mülkname) and state offices, but later was also used as a diploma transferring power to vassal rulers. The berats for the khans differ from those given to other vassals, in that they have a significantly higher number of quotations from the Qur’an. After the especially eloquent introductory formula, their contents are fairly simple: the khan, following the şeriat, should not be unjust to his subjects and re‛ayas, and the magistrates of the country should respect him as their ruler and not revolt against him. The most important specific point regards the incomes of the khans: these were included in the letters in the manner of donations. For instance, Canibek Giray was entitled to receive a yearly revenue of one million akçes from Kaffa (“han-i sabık mutasarrıf olduğı on kere yüzbin keffevî akçe saliyaneye mutasarrıf olub”). The sum changed, but the revenue was received until the fall of the khanate. It was customary that the kalga, the khan’s first deputy, was appointed together with the khan, although in some cases it was left for

14 Sándor Papp, “Die Inaugurationen der Krimkhane durch die Hohe Pforte (16.–18. Jahr hundert),” in The Crimean Khanate between East and West (15th–18th Century), ed. Denise Klein (Wiesbaden, 2012), 75–90; Halil İnalcık, “Yeni Vesikalarına Göre Kırım Hanlığının Osmanlı Tâbiliğine Girmesi ve Ahidnâme Meselesi,” Belleten 8 (1944): 185–229; Alexan- dre Bennigsen, Pertev Naili Boratav, Dilek Desaive, and Chantal Lemercier-Quelquejay, Le Khanat de Crimée dans les Archives du Musée du Palais de Topkapı (Paris, 1978). 15 Jan Rypka, “Briefwechsel der Hohen Pforte mit den Krimchanen im II. Bande von Ferīdūns Münšeāt (Regesten und Chronologie),” in Festschrift Georg Jacob zum siebzigsten Geburtstag, 26. Mai 1932: gewidmet von Freunden und Schülern (Leipzig, 1932), 263; Feridun Ahmed Beg, Mecmu’a, 135–136; Papp, “Die osmanischen Inaugurationen.” 16 Under the following dates: 1–10 Ramazan 1076 (7–16 March 1666) Niedersächsische Nationalbibliothek (Göttingen, henceforth NSNB) 4o Cod. MS. Turcica 29. fol. 179v–180v; 1–10 Şevval 1094 (23 September–2 November 1683) NSNB 4o Cod. MS. Turcica 30. fol. 65v– 66r; 1–10 Receb 1095 (24 June–7 July 1684) NSNB 4o Cod. MS. Turcica 30. fol. 74v–75v. the system of autonomous muslim and christian communities 383 the khan to nominate him.17 The third most important dignitary of the khanate, the nureddin, was always selected by the khans and was never confirmed by the sultan.18 The Porte could also attach a name-i hümayun that discussed practical questions.19

Şirvan In the eighteenth century, due to the Ottoman expansion, the Caucasian country of Şirvan—following sixteenth-century precedents—acknowledged the Ottomans as its sovereign. The sultan’s court discussed the person of the new vassal state’s head with the military leader of the main Ottoman army, stationed in Erzurum.20 The appointment procedure of the new khan of Şirvan and the relationship between the khanate and the Porte was based on the Crimean example. The conditions for assigning power were elaborated in a solemn name-i hümayun, and, as an inauguration document, a berat-i şerif-i ‛alişan (“inauguration letter granted by the Great Lord”) was used. The new khan of Şirvan, El-Hacc Davud, came under the power of the padishah without paying tribute (as was also customary in the case of Crimea), and could regard his country as his liege.21 He was nevertheless dethroned by the Ottomans shortly afterward and the Kumuk prince (hakim) was put in his place.22 El-Hacc Davud was exiled to the island of Rhodes and later to Gallipolli; this paralleled the experiences of the Crimean khans, many of whom were also transported to Rhodes for a certain period of time.

17 Feridun, Mecmu‛a, 135. (under the date evail-i şevval 1037 / 4–13 July 1628); Feridun, Mecmu‛a, 147–148. (11–20 Şaban 1051 / 15–24 November 1641); NSNB 4o Cod. MS. Turcica 30. fol. 79r–v (1–10 Receb 1095 / 24 June–7 July 1684); Topkapı Sarayi Müzesi Kütüphanesi (henceforth: TSMK) R 1954 fol. 5v. (Şevval 1120 / 14 November–13 December 1708). 18 Başbakanlık Osmanlı Arşivi [henceforth: BOA] Nâme-i hümâyûn defteri No. 6. 98–100 (1–10 Ramazan 1116 / 28 December 1704–6 January 1705). 19 BOA Nâme-i hümâyûn defteri No. 6. 149–150 (11–20 Zilkade 1118/14–23 February 1707). 20 İsmail Hakkı Uzunçarşılı, Osmanlı Tarihi [Ottoman history], vol. 4, part 1 (Ankara, 1988), 175. 21 “(. . .) ber-vech-i yurtluk ve ocaklık” BOA Nâme-i hümâyûn defterleri 7. fol. 65–66 (evahir Rebi’ülevvel 1135/ 29 December 1722–8 January 1723); İsmet Binark, Necati Gül- tepe, and Necati Aktaş, Osmanli devleti ile Azerbaycan Türk Hanlıkları arasındaki münase- betlere dâir arşiv belgeleri: Karabağ-Şuşa, Nahçivan, Bakü, Gence, Şirvan, Şeki, Revan, Kuba, Hoy, vol. 2, 1575–1918 [Archival documents concerning the contacts between the Ottoman Empire and the Turkish khanates of Azerbaijan] (Ankara, 1993), 43–44. 22 Uzunçarşılı, Osmanlı Tarihi, vol. 4/1, 196n2. 384 sándor papp

Mecca This is also the place to mention the emirs of Mecca. An embassy, headed by the twelve-year-old son of the emir, paid homage to the sultan for the first time and the country acknowledged Ottoman sovereignty in August 1517 during the Egyptian campaign of Selim I. The emirs of Mecca were inaugurated in the sixteenth and seventeenth centuries through short simple documents of the berat-i şerif or menşur-i emaret types, which became much more elaborate and pompous by the eighteenth and nineteenth centuries. When the position of the emir became vacant, a four-member committee—the kadi of Mecca, the beylerbeys of Egypt and Syria, and the sancakbey of Jedda—could propose the person of the new emir, who was nevertheless always elected from the same family. The structure of the berats sent to Mecca was largely similar to the inauguration documents given to other vassal leaders. A good indicator of the emir’s place in the hierarchy can be seen in the fact that his name was mentioned in the Friday prayer (hutbe) after that of the sultan.23

Tunis, Algeria, and Tripolitania Following the Ottoman practice of government, it was also possible for a local dynasty to establish itself in a territory that was conquered and integrated into the vilayet system, become hereditary rulers, and thus turn their province into a tributary state. This happened in the cases of Tunis, Algeria, and Tripolitania (Trablosgarb), and also Egypt in the period of Muhammad Ali in the early nineteenth century. This development was furthered by the specific treatment of these vilayets; after conquest their territory was not divided among the soldiers and officeholders as lands for service, rather, they remained in one piece, the incomes of which financed the yearly revenue (saliyane) of the head of the province and the wages (ulûfe) of the administration and the military.24 Among the three sovereign vilayets Algeria was the most important, the rulers of which tried several times to extend their powers over the entire southwestern Mediterranean coast.25 Documents show that the beys and pashas here

23 Feridun, Mecmu‛a, 6–7. İsmail Hakkı Uzunçarşılı, Mekke-i mükerreme emîrleri [The emîrs of noble Mecca], (Ankara, 1984), 19. 24 Hezafern Hüseyin Efendi, Telhîsü’l-beyân fî Kavânîn-i Âl-i Osmân [A clear description of the secular laws of the Ottoman dynasty], ed. Sevim İlgürel (Ankara, 1998), 114. 25 Robert Mantran, “L’évolution des relations entre la Tunisie et l’Empire Ottoman, du XVIe au XIXe siècle,” Cahiers de Tunisie 26–27, nos. 2–3 (1959): 319–333; Mme la Comtesse Drohojowska [née Symon Latreiche], L’Algérie Française (Paris, 1864). the system of autonomous muslim and christian communities 385

(called deys) also demanded the practice of the sultan’s confirmation of their power. Hüseyin bin Ali At-Türkî, of Turkish origin, the founder the dynasty that ruled for a long time in Tunis, was appointed as a sancakbey by the sultan with a berat in 1709.26 The process of separation from the Porte took centuries. As early as 1666 the Habsburg ambassador in Constantinople, Johann Baptist Casa- nova, doubted whether the governors of Algeria, Tunis, and Tripoli would follow the sultan’s orders and send the fleet requested from them, “as they have been obeying the Ottoman Porte out of their own free will for a long time.”27 Although Tunis refused to continue paying tribute in 1871, even after the establishment of the French protectorate in 1881 the Sublime Porte demanded that the sultan’s nominal sovereignty be acknowledged.28 Actual dependence and nominal sovereignty could well exist in parallel, as is clear from the example of Joseph von Hammer-Purgstall’s description of the Ottoman Empire in 1815, in which he speaks of these African provinces as independent entities.29

Further Hereditary Sancaks As can be seen in the example of Tunis, vassals were not only appointed by giving them the rank of beylerbey, but also by bestowing the lesser rank of sancakbey. In the eastern territories (Diyarbekir, Şehrizor, Van, Çıldır, Gürcistan, Bitlis, and elsewhere) there are also examples of the custom of local Armenian or Kurd notables paying homage to the sultan and staying in power as hereditary sancakbeys (and ruling their countries as yurts, ocaklıks, and mülks).30 This territory was a buffer zone between Safavid Iran and the Ottoman Empire, and according to his own account,

26 “Hüseyin dame izzuhuya tevcih.” I found this hitherto unknown berat during my research at TSMK (call number: Revan 1954. 6b–7a). 27 Extracts from Casanova’s reports (November–December 1666), and his report from 7 December 1666 (Österreichisches Staatsarchiv Haus-, Hof- und Staatsarchiv [henceforth HHStA] Türkei I. (Turcica) Kt. 139. Konv. A fol. 45–46, resp. 47. 28 Birken, Die Provinzen, 274. On paying the tribute earlier, I found the following sources (concerning the year 1254/1838–39): Les Archives Nationales de Tunisie (Tunis), Série Historique III. Empire Ottoman. Carton 220. Dossier 340. Pièce 68. 29 Joseph von Hammer, Des osmanischen Reichs Staatsverfassung und Staatsverwaltung (Vienna, 1815), vol. 2, 248. 30 Mehmet İnbaşi, “XVIII. Yüzyılda Bitlis sancağı ve İdareciler/Bitlis Province and Its Governors in 18th Century,” A. Ü. Türkiyat Araştırmaları Enstitüsü Dergisi 33 (2007): 243– 261. The Kurdish Hakkari was such a territory, governed until 1876 from the vilayet of Erzurum. Uzunçarşılı, Osmanlı tarihi, vol. 2, 572–573; Bernard Lewis, et al., “Hakkārī,” in EI2 (Leiden, 1986), vol. 3, 83. 386 sándor papp

İdris-i Bitlisi, the renowned fifteenth-century politician and chronicler, also worked on making the heads of the local Sunnite tribes accept the sovereignty of the sultan. Later these clans functioned for centuries as hereditary sancaks.31 Because of the war between Iran and the Ottomans, local notables were promised hereditary sancaks on the borderlands to secure their allegiance to the Porte.32 The most important document on the integration of the Kurdish territories into the Ottoman Empire was issued by Sultan Süleyman I, probably after the end of the eastern campaign of 1533–1535. The sultan included the conditions provided for the Kurdish dignitaries in a “treaty letter of the Great Lord” (mu‛ahede-i hümayun). As they had been serving the dynasty since the rule of the sultan’s father, Selim I, Süleyman granted them power over the territories, castles, and cultivated lands that they had ruled earlier. The Kurdish beys were allowed to hand down their lands as their own to their successors and could also divide them; if a family died out, the local lords could decide to whom the estates should be bequeathed. According to the letter of treaty, each Kurdish bey was supposed to receive a sultanic berat to reinforce his rule over his property. As in the case of Western European treaties, the sultan took an oath to keep the arrangements and also specified that he would deal with anyone who did not keep the agreement. In an extraordinary fashion, the sultan not only spoke in his own name, but also decreed the agreement valid for his succesors.33 So far, among these berats I have only found one given by the sultan to Muhammad, the father of the Kurd Şerefhân, probably assigned during the rule of Murad III.34

31 İdris-i Bitlisî, Heşt Bihişt [Eight paradises], ed. Mehmet Karataş, Selim Kaya, and Uzman Yaşar Baş (Ankara, 2008), vol. 1, 15. 32 Bekir Kütükoğlu, “Şeref Han,” in İslâm Ansiklopedisi (Istanbul, 1979), vol. 11, 427; Orhan Kılıç, “Yurtluk-Ocaklık ve Hükümet Sancaklar Üzerine Bazı Tespitler” [Some conclusions about the yurtluk-ocaklık and hükümet sancaks], Ankara Üniversitesi Osmanlı Tarihi Araştırma ve Uygulama Merkezi Dergisi 10 (1999): 119–139. 33 Topkapı Sarayı Müzesi Arşivi [henceforth: TSMA] E. 11969; Nazmi Sevgen, “Kürtler” [Kurds], Belgelerde Türk Tarihi Dergisi 2 (1968) Sayı: 9. 70–72. 34 Sarı Abdullah Efendi, Desturü l-inşa [The rules of writing letters], Österreichische Nationalbibliothek, Handschriftensammlung, Ho. 167. 255v–258r. The son is probably identical with Şeref Han, the hâkim of Mazo (?), who participated at the siege of Baghdad under the rule of Murad III; see Na‘ima Mustafa Efendi, Târih-i Nâ‘îmâ (Razvatü’l-hüseyn fî hulâsati ahbâri’l-hâfikayn) [The chronicle of Na’ima: The garden of Hüseyn in the summary of the chronicles of east and west], ed. Mehmet İpşirli (Ankara, 2007), vol. 2, 521. The question mark above is necessary, as the literature knows no sancak under this name; the name indicated in the source probably refers to the sancak of Hazo (Sason), which belonged to the vilayet of Diyarbakır and was a hereditary possession of a Kurdish ruling the system of autonomous muslim and christian communities 387

This system was also used in other territories of the empire. When the governor of Yemen, Şeyh Mutahhir, died, his son, Abdurrahmân, was appointed to be the head of his sancak on 31 July 1575. His berat states that the territory could be bequeathed from father to son.35 The ocaklık system also appeared in Bosnia, where the military elite received hereditary lands for their service and some units were financed through collective estates given to them, also called ocaklık. In the eighteenth century, this Bosnian military system built upon hereditary titles and clan (ocak) estates served well in defending the country against the Habsburg Empire.36

Egypt Finally, the example of nineteenth-century Egypt provides insight into the endurance of the Ottoman conception of the state. Ever since its integration into the empire, Egypt enjoyed a “special” status which was also used by the Ottomans as a pattern elsewhere. In the Long Turkish War of the late sixteenth century, when Koca Sinan occupied Wallachia and also wanted to conquer Moldavia, he planned to integrate both of them into the empire “in the way of the vilayet of Egypt” (“vilayet-i Mısriyye gibi”). This meant that the earlier internal structure of the countries would remain untouched, but beylerbeys and Ottoman state officials would be appointed in the place of the Christian voievods.37 This system was maintained for centuries along the banks of the Nile, even after the end of the Napole- onic wars, when, exploiting the turmoil and using the armed forces under his command, Kavalalı Mehmed ‛Ali seized power and was appointed vali of Egypt in place of the Ottoman governor, whom he overthrew

dynasty. See Birken, Die Provinzen, 189. It was probably this Şeref Khan whose inauguration was mentioned in the following orders: BOA 32. Mühimme defteri No. 506, 514, 543; Sevgen, “Kürtler,” 74–76. 35 “(. . .) ber-vech-i ocaklık” Bayezit Kütüphanesi (Istanbul), Elyazmalar, No. 1970. 166v–168r. 36 Nenad Moačanin, “Some Observations on the Kapudans in the Ottoman Northwest- ern Frontier Area 16–18 c.,” in Acta Viennensia Ottomanica: Akten des 13. CIEPO Sympo- siums, ed. Markus Köhbach, et al. (Vienna, 1999), 241–246; Nedim Filipović, “Ocaklık timars in Bosnia and Herzegovina,” Prilozi za orijentalnu filologiju 36 (1986): 149–172. Part of the literature questions whether the ranks were hereditary in kindreds, see Michael Robert Hickok, Ottoman Military Administration in Eigteenth-Century Bosnia (Leiden and New York, 1997). 37 Mihai Maxim, “Voyvodalık ou beğlerbeğilik? La politique ottomane envers la Mol- davie et la Valachie (novembre 1594–février 1596) à la lumière des nouveaux documents turcs,” in idem, Romano-Ottomanica: Essays and Documents from the Turkish Archives (Istanbul, 2001), 163–172. 388 sándor papp

(8 June 1804).38 In his book, written eleven years after the event, Joseph Hammer described the system of saliyane in Egypt: “there is neither ziamet nor timar, only twenty-four beys with saliyane, who, together with their 12,000 men strong regular militia receive a pre-defined sum of gold.” The central revenues covered the expenses of the local government, and even provided a yearly income of 600,000 ducats for the sultan’s treasury.39 Mehmed Ali Pasha led an armed revolt against the Ottoman Empire in the 1830s. After concluding peace, on 24 May 1841, under pressure from England, the sultan also granted him the titles of hereditary vali and honorary grand vizier. Although I could not find the original ferman of inauguration, the French version, published by Gabriel Noradungiyan Efendi (also used in the renowned Turkish chronology of Danişmend), deserves attention. According to the agreement, the title of vali would be hereditary from father to son, but the right of the appointment remained that of the sultan. Should the male line die out, the female line was excluded from the right to power. The vali of Egypt had the title of vizier, corresponding to similar dignitaries of the empire. Every measure introduced during the Tanzimat was to remain, as well as every Ottoman law. The tribute of Egypt—the basis of which did not change—was to be delivered at a given time of year; the sum was set by another document at 80,000 pouches of gold. The name of the padishah should appear on the money minted in Egypt and no qualitative difference was allowed between them and Ottoman coins. Every year Egypt had to send a contingent of 400 men to Istanbul, and the vali could only appoint his officers up to the rank of colonel; higher ranks were the prerogative of the sultan.40 Whether the agreement was observed and how its conditions were modified due to the advance of English power does not concern us here. At least one of the points above was still being observed when Egypt was a colony, that is, that money could only be coined in the sultan’s name. Even as late as 1909, silver coins made in Egypt bore the tuğra of Sultan Mehmed IV Reşad.41 For Islamic political theory, this was an important development, as coinage and mentioning the name of the ruler in the hutbe were considered important symbolic signs of sovereignty, expressed

38 İsmail Hâmi Danişmend, İzahlı Osmanlı Tarihi Kronolojisi [A detailed Ottoman chronology] (Istanbul, 1972), vol. 4, 81; Cevdet Paşa, Târîh-i Cevdet [The chronicle of Cevdet] (Istanbul, 1309), vol. 8, 24–28. 39 Hammer, Des osmanischen Reichs, vol. 2, 270–271. 40 Danişmend, İzahlı Osmanlı, vol. 4, 130. 41 I personally own one of these coins. the system of autonomous muslim and christian communities 389 by the regularly used term “sikke ve hutbe sahıbı” (the owner of the coinage and the Friday prayer).

The Foundations of the Christian Countries’ Vassal Relationship to the Porte

Everyday realities forced Muslim states to conclude political alliances with communities with whom they shared aims (at least for a while), but who were not ready to accept Islam. The first such treaties were concluded as early as during the life of the Prophet Muhammad. Not long after the hijra, he made an agreement with the ‛Aws and Khazraj tribes from Yathrib; these tribes also included the Jewish communities of Medina. Recently, eight different copies of this agreement were published by Robert B. Serjeant, with the Arab text, an English translation, and commentaries that identified the true character of the source, long considered a fundamental treaty between the Jewish tribes of Yathrib and the Muslims. The most important conclusion was that at this time the prophet had still not restricted the use of term umma (Turkish: ‛ümmet, designating the entire community of the believers) to Muslims, thus acknowledging the right of people from different religions to follow their own rites. Muhammad also convinced the opposing parties to accept blood money for members of their community killed during conflicts and give up the law of talion.42 Later, the prophet concluded another treaty with the Kurayiš tribe of Mecca, securing the possibility of pilgrimage for his new community, the believers of Islam (the treaty of Hudaybiya, 628). This agreement, concluded for ten years, provided the pattern for later interstate treaties of Islamic powers.43 For the new Islamic community, it might seem that its victorious advancement left no other choice for those of different religions than to convert or perish. By the time good relations with the Jews ceased, another community had appeared on the horizon: the Christians. The

42 At the same time, it also became clear that surviving texts include two successive treaties and that there are references to this agreement in the Qur’an (3:101–104), see Rob- ert Bertram Serjeant, “The Sunnah Jāmi‘ah, Pacts with the Yathrib Jews, and the Taḥrīm of Yathrib: Analysis and Translation of the Documents Comprised in the So-Called ‘Constitu- tion of Medina,’ ” Bulletin of the School of Oriental and African Studies 41, no. 1 (1978): 1–42; Majid Khadduri, War and Peace in the Law of Islam (Baltimore, 1955), 206–209; Tilman Nagel, Das Islamische Recht: Eine Einführung (Westhofen, 2001), 94–95. 43 Khadduri, War and Peace, 210–213. 390 sándor papp representatives of the Taghlib tribe came before the prophet, who concluded an agreement with them according to which, although the adults could keep their faith, the children were no longer to be baptised or to follow Christian religious customs.44 During the expedition to Tabûk the idea arose for the first time that the Islamic army could take up arms against the people of the book (ahl al-kitâb; that is, Christians, Jews and Zoroastrians), if they did not accept paying a tax (cizye).45 Muslim religious law sees the passage of the Qur’an (9:29) narrating this episode as the basis of the custom that people of the book can only live in Muslim territories in exchange for paying this tax. According to Róbert Simon’s exhaustive historical and linguistic analysis, however, the term used there (ğizya) was not understood under the later specific meaning of poll tax— distinguished from the land tax (harâğ)—only as a compensation for the benevolence shown toward non-Muslim communities.46 In any case, the followers of non-Muslim religions of the book not only had to abandon spreading their faiths, but also were obliged to pay a certain sum to the Muslim community. The prophet also concluded agreements with some Arab Christian communities who were not willing to give up their faith; Majid Khadduri noted the existence of seven such treaties and published three of them in English translation.47 The most detailed of these was that concluded with the town of Nağrân in North Yemen. According to the traditions of Arab historiography, Muhammad tried to convince them to accept Islam, but they refused; similarly, they refused to delegate judgment about their faith to divine signs, as was also suggested by the prophet. Nevertheless, they obliged themselves to pay a certain sum to the Muslims twice a year and to support them with equipment and draft animals should they go to

44 Nagel, Das Islamische Recht, 96–97. 45 When the news came that the Byzantine Empire was preparing to attack the Mus- lims at the borders of Syria, the prophet mustered a significant army. After suffering great hunger and thirst they found no enemy when they arrived at the border. The expedition of Tabûk took place in October and November 630. See Róbert Simon, A Korán világa [The world of Qur’an] (Budapest, 1987), 535; Salih Suruç, Kâinatın Efendisi: Peygamberimizin Hayatı [The master of the creatures: The life of our prophet], vol. 1 (Istanbul, 1984), 599. The expedition was nevertheless used to force several communities at the Syrian border to pay cizye, among them the Christians of Ayla. On the conclusion of the treaty, see Suruç, Kâinatın Efendisi, vol. 1, 622; for the English translation of the text: Khadduri, War and Peace, 181. 46 Simon, A Korán világa, 201–203. 47 Khadduri, War and Peace, 179–181. the system of autonomous muslim and christian communities 391 war.48 The bases of the later legal interpretation were developed in this early period, that is, if the followers of the religions of the book accepted the superiority of Islam and were ready to pay for their security they received the protection of the prophet and later of Muslim rulers. This is called dhimmî, or in the Turkish form, zimmi, status. The next stage of the system’s formation is related to the name of the second caliph, ‛Umar (‛Ömer), who gave a detailed form to the rather strict conditions according to which non-Muslims were allowed to live in the lands of the Muslims. These covered a ban on carrying weapons, restrictions on religious practices, clothing, the manner of speech, an obligatory humble attitude toward Muslims, and the obligation of three days of hospitality.49 The fully developed system is usually described by major historical overviews using three Arabic terms: dâra l-islâm (the Muslim lands),

48 Nagel, Das Islamische Recht, 98; Werner Schmucker, “Die christliche Minderheit von Nağrân und die Problematik ihrer Beziehung zum frühen Islam,” in Studien zum Minder- heitsproblem im Islam (Bonn, 1973), vol. 1, 183–281. The edition of the Arabic text is available in Balâdurî, Al-imâm Abu al-‘Abbâs Aḥmad ibn Yaḥya ibn Ğâbir, Kitâb Futûh al-Buldân: Liber Expugnationis Regionum, auctore Imámo Ahmed ibn Jahja ibn Djábir al-Beládsorí (. . .) e codice Leidensi et codice Musei Brittanici, ed. M.J. de Goeje (Leiden, 1866), 65; English translations: Abu al-‘Abbâs Abu al-‘Abbâs Aḥmad ibn Yaḥya ibn Ğâbir al-Bâladhurî, The Origin of the Islamic State (Kitâb Futûḥ al-Buldân) (New York, 1916), 100–101; Khadduri, War and Peace, 179–180. I found no reference in the literature about early Islam to a valuable document preserved in Arabic and also in Ottoman translation in the document collection of Feridun bey. In contrast to the other letters known to me from the time of the prophet, it has a precise date; it was written in the prayer room (mescid) of the prophet in Medina in year 2 according to the hicra, on 3 Muharrem (1 July 623) and addressed to Ali ibn Tâlib. According to the translation, it was valid for all Christians (“Muhammed cümle al-i millet- i nasariyeye müntesib olanlara ‛ahd içün verdiler kim ki ol ‛ahdın boza Hakk-i te‛alanın ‛ahdını bozmuş olur”), who were given the protection of the prophet wherever they were. In times of war, no obligatory service was requested from them; they had to give whatever they were ready to give of their own free will. In times of war, Christians were free of personal service and did not have to pay more than two dirhams yearly; the document also secured full protection for the churches of the Christians. See Feridun, Mecmu‘a, 30–31; 32–34. As the text differs both in attitude and content from the treaties that are known and accepted by the literature to be preserved from the time of the prophet, it cannot be excluded that this one is apocryphal from a later period. It must be noted, however, that a seventeenth-century copy (of 1048/1623) was known and accepted as authentic by Muslim authorities in Istanbul at the turn of the nineteenth century, see Joseph von Hammer, “Sened, das ist Vertragsurkunde von Omar Ibn al Chattab, dem zweiten Chalifen, dem Patriarchen von Jerusalem unter seinem Siegel gegeben,” Fundgruben des Orients 5 (1816): 67. A historical analysis of the document is still a task for future research; it cannot have been a forgery done during the creation of the Feridun collection in the mid-nineteenth century. 49 Nagel, Das Islamische Recht, 99–100. Texts of several treaties from the time of ‛Umar are also known from Arab chronicles, among them those concluded with the Christian population in the towns of Hîra and Damascus, see Khadduri, War and Peace, 183–184. 392 sándor papp dâra l-harb (the land of the enemies), and dâra l-‛ahd (the land of treaty). Christian states and communities, as noted above, could also inhabit a territory under Islamic jurisdiction, and thus receive the protection of Islam, if they accepted the rule of Islam and kept the laws concerning those of different religions. On this basis, they could conclude a dhimmî (zimmi) treaty that would be valid for an undefined time and would include the payment of a poll tax (cizye) or a land tax (harac) by the state. In the case of dâra l-harb, the non-Muslim party was considered hârbî (people from enemy territory/an enemy). Diplomatic relations and peace treaties for defined periods could also be maintained with enemy territories; what is more, merchants and pilgrims coming from these lands could enjoy full security of life and property in Islamic lands, in which case these hârbîs were called musta‛min. The condition for such treaties—the terminology of which derived in each case from the Arabic concepts of ‛ahd (treaty) or sulh (peace)—was that the Muslim state would necessarily sus- pend permanent war with the enemy. European Christian powers, among them the Kingdom of Hungary until 1519, concluded such musta‛min treaties with the Ottoman Empire. In the political practice of the Ottoman Empire, the third term, dâra l-‛ahd, did not play a very important part. The question of whether the dhimma (zimma) treaty automatically meant the introduction of Islamic jurisdiction (šari‛ya) or not was debated among generations of Muslim legal scholars. The Ottoman state followed the Hanafi school, and, in a pragmatic manner, accepted the Christian vassal states and communities as Islamic lands even without the introduction of Muslim jurisdiction.50

50 Ahmed Özel, İslâm hukukunda ülke kavramı, dârulislâm, dârulharb (Istanbul, 1991), 214–227. Recently, Viorel Panaite studied the Hanafi legal school and its Ottoman application. In two books and a large number of studies, he has analyzed Kitâb as-Siyâr, the book on “international relations” of the most influential legal scholar of the Hanafi school, the tenth-century Šaybânî; Panaite examines the Ottoman questions primarily related to Moldavia and Wallachia. Viorel Panaite, The Ottoman Law of War and Peace: The Ottoman Empire and Tribute Payers (New York, 2000); idem, Pace, război şi comerţ în Islam: Ţările române şi dreptul otoman al popoarelor (secolele XV–XVII) [Peace, war and commerce in Islam: The Romanian countries and the Ottoman law of nations: fifteenth to seventeenth century] (Bucharest, 1997). A widely used edition of Šaybânî’s text is Muhammed Hasan Esh-Shaybani, The Islamic Law of Nations: Shaybani’s Siyar, trans. Majid Khadduri (Balti- more, 1966). There is also a shorter version: Muḥammad ibn al-Ḥasan al Shaybānī, The Shorter Book on Muslim International Law: Kitāb al-siyar al ṣaghīr, ed. Mahmood Ahmad Ghazi (Islamabad, 1998). the system of autonomous muslim and christian communities 393

Dhimmî Agreements Concluded with Non-Muslim States

Murcia The first document known to me which describes the status of a Chris- tian kingdom accepting the rule of Islam comes from the Iberian Penin- sula. During the Arab conquest, the province Murcia (named after King Theodemar) paid homage to ‛Abd al-‛Azîz bin Mûsâ bin Nuṣair, who concluded an agreement (kitâb ṣulh) with the Christian population of the territory. The Arab conqueror gave the population of the province his protection (dhimma), and the text of the treaty also included the term ‛ahd. Besides securing the life of Christian men and their relatives, the document also acknowledged the safety of their churches. In exchange, they had to concede seven towns and pay a precisely defined yearly poll tax consisting of money and agricultural products.51 The pattern of the document was fully identical with the Arabic practice of earlier years.

Galata The Genoese-ruled territory of Galata accepted Ottoman rule at the time of the fall of Byzantine Constantinople. The magistrates of the community turned to Sultan Mehmed II, who secured their privileges in a letter of treaty (‛ahdname). Although it contained some restrictions, such as a ban on ringing bells and building further churches, in this document the sultan promised that no one would disturb the population’s domiciles and houses, they could not be converted to Islam against their wishes, and they would not be enlisted as janissaries. They were also free from any kind of forced labor. From the formal point of view, this document was a treaty in which the inhabitants of Galata subjected themselves to the sultan (“ita‛at u inkiyad göstermişler”) and the privileges of the town were enforced by the sultan with an oath formula later used in international agreements. Mehmed II promised to consider the people of Galata on the same level as his subjects in other parts of his empire, thus, it was not

51 The treaty was drawn up in the month of Rağab of the year 94 (April 713), and the text even lists those who testified to the agreement, see Wilhelm Hoenerbach, Islami sche Geschichte Spaniens: Übersetzung der A‛mâl al-a‛lâm und ergänzender Texte (Zurich and Stuttgart, 1970), 54–55. According to this edition, the document is unquestionably authentic. 394 sándor papp their extraterritorial status but rather their rights under Ottoman rule that were specified here and reconfirmed in 1610.52

Naxos A somewhat different situation can be observed in the case of the Duchy of Naxos (or Duchy of the Archipelago, Nakşe Dukalığı). The duchy was founded in early thirteenth century by the Venetians, who deprived Byz- antium of the Aegean archipelago. The inauguration letter (a berat-i hüma yun) known to me was drawn up for Ciacomo IV Crispo on 20 December 1564, redrafted, and sent to the duke on 29 April 1565. According to the document, it was Ciacomo IV Crispo who, after the death of his father, asked Sultan Süleyman I to confirm him on the throne. In exchange, he paid a poll tax (cizye) of 162,000 akçes yearly. The berat then also elaborated on the rights and obligations of the duke, for example, that he not allow anyone other than the subjects of the Ottoman Empire to settle on the islands, and was also prohibited from welcoming enemies or pirates in his harbor.53 After Selim II took the throne, he gave the Duchy of Naxos to his Jewish confidant, Don Josef Nassi, who probably had the same privileges as the Crispo family.54 For his services related to the ratification of the Habsburg–Ottoman peace treaty, Gaspar Graziani also received the title of duke on 2 November 1616, and held it until his appointment as the voievod of Moldavia in 1619.55

52 İskender Yanko Hoci, “Galata’nın Osmanlılara Teslimi” [The surrender of Galata to the Ottomans], Tarih-i Osmani Encümeni Mecmu‛uası 25 (1330 AH): 52–53; Halil İnalcık, “Ottoman Galata, 1453–1553,” in idem, Essays in Ottoman History (Istanbul, 1998), 271–376. (on further editions of the text in Greek and Turkish versions, see ibid., notes 1, 3, 4, 5, and 6); Mahmut Şakiroğlu, “Fatih Sultan Mehmet’in Galatalılara verdiği Fermanın Türkçe Metinleri” [The Turkish text of the ferman of Sultan Fatih Mehmet to the people of Galata], Tarih Araştırmaları Dergisi 21 (1983): 211–216. 53 Saffet, “Nakşe (Naksos) Dukalığı, Kiklad adaları” [The islands of the Cyclades in the voivodate Naxos], Tarih-i Osmani Encümeni Mecmu‛uası, vol. 4/23, 1444–1457; 6 Numaralı Mühimme Defteri (972 / 1564–1565), ed. Hacı Osman Yıldırım (Ankara, 1995), vol. 2, 291–292 (No. 536). 54 Heinrich Graetz, Geschichte der Juden, vol. 9, Von der Verbannung der Juden aus Spa nien und Portugal bis zur dauernden Ansiedelung der Marranen in Holland (1618) (Leipzig, 1866), 398–459; Yasemin Demircan, “1720 Tarihli Tahrir Defterine Göre Nakşa Adası’nda Yapılan Düzenlemeler ve Reayanın Durumu” [The ordinances issued at Naxos on the basis of the 1720 tahrir defter and the situation of the re‘ayas], Belleten 73 (2009): 672–673. 55 Manfred Stoy, “Das Wirken Gaspar Gracianis (Graţianis) bis zu seiner Ernennung zum Fürsten der Moldau am 4. Februar 1619,” Südost-Forschungen 43 (1984): 102. the system of autonomous muslim and christian communities 395

Chios The situation of Chios (Sakız) offers yet another example. The island had earlier been Genoese property that paid tribute to the Ottomans even before 1454. After 1475, Chios remained the only Genoese colony and they continued to pay the tribute until 1566, when the Ottomans occupied the island.56 Sultan Süleyman confirmed the privileges of the rulers of Chios in an ‛ahdname-i hümayun at the end of October 1559, in which he referred to the peaceful and friendly relationship created after the wars of his predecessors, peaceful relations that he wanted to maintain: Before this, my forefathers and my father concluded a peace with (. . .) the beys of Chios. I accepted this, made peace and a contract with them (. . .) that in the form as friendship was kept with his spahis [i.e., knights] and men and the people belonging to them, it shall remain so, and from the part of the sancakbeys, kadis, subaşıs, kapudans and captains of my high empire (. . .) and from all people belonging to me, nothing shall be started against them, their people and lands should not be harmed; and would it happen, I shall amend it and punish those who caused it.57 The merchants of the island were assured that they could continue paying the customs’ fees in the old manner and they would be protected against damages done by Ottoman subjects. Their yearly tax, which was 4,000 gold coins in 1415, was at this time raised to 10,000 golden coins (450,000 akces).58 The sultan confirmed the conditions of the ‛ahdname with the usual oath.

Ragusa The archives of Dubrovnik (Ragusa) have a rich and valuable collection of early Ottoman sources, but many of the original copies of early ‛ahdnames are lost.59 The earliest letter of treaty is known from 1430, but the

56 Pitcher, Osmanlı İmparatorluğu’nun, 127. 57 “Sakiz beğlerine Sultan Süleyman eyyamında verilen ‛ahdnamedür (. . .) Bundan evvel benüm dedelerüm babam (. . .) Sakiz beğleriyle dostluk eylemişler. Ben-dahi kabul etdüm ve musalaha ve mu‛ahede eyledüm (. . .) anlarun sipahileri ve adamlarıyla ve anlara müte‛alik olan kimesneleriyle şimdiyedek ne-vechle dostluk olıgelmiş ise girü ol-vech üzre min ba‛d memalik-i mahrusemde olan sancakbeğlerimden kadılardan subaşılardan kapudanlardan ve re’islerden (. . .) fi-l-cümle bana müte‛alik kimesnelerden anlara ve anlara müte‛alik olan adamlara ve yerlerine bir ziyan yetişmeğe eğer yetişecek olursa yerine koyum ve edenlerün hakkından gelem,” Bayezit Kütüphanesi, Elyazmaları, Veliyüddin 1970, fol. 148v–149v. 58 “(. . .) her yıl derhâh-i mu‛allama haraclarıyçün on bin frengi filorı teslim edeler veya- hud filorı bahası içün dörtyüz elli bin akçe vereler.” Ibid., 148v–149v. 59 Franz Kraelitz, Osmanische Urkunden in türkischer Sprache aus der zweiten Hälfte des 15. Jahrhunderts: Ein Beitrag zur osmanischen Diplomatik (Vienna, 1921); Vesna Miović and 396 sándor papp document of the 1442 agreement is also kept here; both are written in South Slavic. The first treaty in Turkish was issued by Sultan Selim I in 1513 and its text was preserved in the kadi register (sicill) of Bursa.60 The eighteenth-century copy registers in the Dubrovnik archives contain the treaties and the orders of the sultans from the 1458 capitulation of Mehmed II on.61 The first letter of treaty that was preserved in the Turk- ish original dates from 1575. The points granted by Sultan Murad III— published in English translation by Nicolaas Biegman—were nevertheless built upon the ‛ahdnames of his predecessors.62 The first contacts of Dubrovnik with the Ottomans date to 1365.63 On the intervention of King Sigismund of Luxemburg—who had been their guest after the defeat at Nikopolis—the Council of Basle, summoned for the preparation of unifying the Catholic and Orthodox churches, granted a diploma to the Ragusans on 22 November 1433 that allowed them to

Nikša Selmani “Turska kancelarija i Acta Turcarum od vremena Dubrovačke Republike do danas” [Turkish chancellary and the Acta Turcarum from the times of the Ragusan Repub- lic until the present day], Anali Zavoda za povijesne znanosti Hrvatske akademije znanosti i umjetnosti u Dubrovniku 45 (2007): 235–284. The most important recent publications of Ottoman documents from Ragusan archives are Vesna Miović, Dubrovačka Republika u spisima osmanskih sultana [The Ragusan republic in the documents of Ottoman sultans] (Dubrovnik, 2005); eadem, Dubrovačka Republika u spisima namjesnika Bosanskog ejaleta i Hercegovačkog sandžaka: s analitičkim inventarom bujuruldija (1643–1807) serije Acta Tur- carum Državnog arhiva u Dubrovniku [The republic of Dubrovnik in the documents of the governors of the Bosnian eyalet and Herzegovinian sancak: With an analytical inventory of the buyuruldus (1643–1807) in the Series Acta Turcarum at the state archives in Dubrovnik] (Dubrovnik, 2008); Boško I. Bojović, Raguse (Dubrovnik) et l’Empire ottoman (1430–1520) (Paris, 1998). 60 Fahri Dalsar, “Bursa Şer’i Mahkeme Sicillerine Göre: Selim I’in Dubrovnik Cumhu- riyeti İle Yaptığı Muahede,” [According to the şeri‘at register of Bursa: The agreement of Selim I with the Republic of Dubrovnik], Tarih Vesikaları 2 (1943): 410–414. 61 Državni Arhiv u Dubrovniku [henceforth, DAD] 7 DA 15 st bv. 19: Traduzioni die Capitulazio e Fermani di Sultan Mehmed II, Baiezetto II, Selimo I, Solimano II, Selimo II, Amurate III, Mahomete IV, Ahmed I, Mustafa I, Osmano I, Mustafa I, Amurad IV, Ibrahim I, Maomete IV, Solimano III, Ahmed II, Mustafa II, Ahmete III, Mahmud I, Osmano II. See the editions of Mehmed II’s capitulation from October 1458 in Branislav M. Nedeljković, “Dubrovačko–turski ugovor od 23. oktobra 1458. godine” [Ragusan–Ottoman treaty of 23 October 1458], Zbornik Filozofskog Fakulteta 9, No. 1 (1970): 390–393; Bojović, Raguse, 21–30. 62 Nicolaas H. Biegman, The Turco–Ragusan Relationship According to the Firmāns of Murād III (1575–1595) Extant in the State Archives of Dubrovnik (The Hague and Paris, 1967), 55–59. 63 Franz Babinger, “Ragusa,” in İslâm Ansiklopedisi, vol. 9 (Eskişehir, 1997), 598–601; Şerafettin Turan, “Dubrovnik,” in Türkiye Diyanet Vakfı İslâm Ansiklopedisi (Istanbul, 1994), vol. 9, 542–544; Johann Christian von Engel, Geschichte des Freystaates Ragusa (Vienna, 1807), 137; Saffet, “Raguza [Dubrovnik] cumhurlığı” [The republic of Ragusa], Tarih-i Osmani Encümeni Mecmu‛uası, vol. 3/17, 1061–1070. the system of autonomous muslim and christian communities 397

maintain contacts with Muslims when organizing pilgrimages to the Holy Land or securing the safety of goods transported there. They were allowed to trade with the Turks, open consulates, and build churches in the Levant.64 The eastern trade of the town dates back to even earlier periods, as they also stayed in a contractual relationship with pre-Ottoman Muslim states such as Egypt.65 After 1440 they paid a regular yearly tribute for the privileges granted to them by the Ottoman state. The sum was first 500, later 1,000, gold coins; this was raised to 1,500 in 1459, 5,000 in 1469, 10,000 in 1472, 12,500 in 1478, and 15,000 in 1480. Sultan Bayezid II lowered the tribute to 12,500 gold coins, and this remained the practice ever after.66 Apart from the yearly tribute, the ‛ahdnames also regulated the question of customs to be paid in Ottoman harbors.67 According to the ‛ahdname of 1458, Ragusans had to pay a 2 percent customs duty. In the time of Mehmed II the tribute and the customs were merged into a single payment, but later this was abandoned: from 1521 on, they had to pay 5 percent customs’ duty in Istanbul, 3 percent in Bursa and Edirne, and 2 percent in other places, which, however, had to be paid in one sum of 50,000 akces every six months in a form resembling the system of tax- farming (iltizam).68 The same method was still in practice in the following century: the ‛ahdnames of both 1604 and 1649 prescribed that Ragusans pay 2 percent customs’ duty for their trade in Rumelia—except in Istan- bul, Edirne, and Bursa—which they covered by paying an overhead

64 The text of the 1433 privilege was published in Dubrovačka akta i povelje [Ragusan acts and privileges], vol. 1/1, ed. Jovan Radonić (Dubrovnik, 1934), 430. The original treaty is found in DAD 7 DA 15 st bv 19. For the historical context of this important document see Ivan Božić, Dubrovnik i Turska u XIV i XV veku [Ragusa and Turkey in the fourteenth and fifteenth centuries] (Belgrade, 1952), 57–60; Robin Harris, Dubrovnik: A History (London, 2003), 98–100. 65 Engel, Geschichte des Freystaates, 138. 66 Ibid., 141; Halil İnalcık, “The Ottoman State: Economy and Society, 1300–1600,” in An Economic and Social History of the Ottoman Empire, ed. Halil İnalcık and Donald Qua- taert (Cambridge, 1994), 261. The stipulations concerning the payment of the tribute were formulated in the following manner in the ‛ahdname-i hümayun of 1649: “they should send 12,500 florins each year according to the old customs through their envoys and there should be no harac demanded from them apart from these 12,500 florins.” DAD Acta Tur- carum. The ‛ahdname uses literally the same formula, cf. Mladen Glavina, “An Overview of the Formation and Functioning of the Institute of Capitulations in the Ottoman Empire and the 1604 Dubrovnik Capitulation,” Prilozi za orientalnu filologiju 58 (2008) 155. A good survey of the steady rise of Ragusan tribute is found in Harris, Dubrovnik, 95–96. 67 Boško Bojović, “Dubrovnik (Raguse) et les Ottomans II: Onze actes de Mehmed II en vieux-Serbe 1473–1476,” Turcica 24 (1992): 153–182; Turan, “Dubrovnik,” 543. 68 Turan, “Dubrovnik,” 543. 398 sándor papp

(mukata‛a) of 300,000 akces every three years, delivered to the treasury in even installments every six months.69 According to the important and long-lasting stipulations of the ‛ahdnames given by Murad III, the wealth of a Ragusan merchant who died in Muslim lands could not be confiscated for the treasury by the beytü l-malcı, the officer regularly concerned with the canonical distribution of inheritances; nevertheless, debates between Muslims and Christians had to be judged by the kadi.70 These conditions continued fundamen- tally unchanged into the seventeenth century, and the town maintained its vassal status with the Porte until the early nineteenth century.71 Wars could sometimes tear apart century-long connections, but trade relations were regularly re-established in more peaceful periods. Such was the case in 1756, when the Porte excused the payment of the last few years’ unpaid tribute and thus resumed the trade and legal relationship between the Ottoman Empire and Ragusa based on the earlier tribute and services.72

Corfu The relationship between Dubrovnik and the Sublime Porte also served as a model for other arrangements. In the late eighteenth century, the Ottoman Empire made an alliance with England and Russia to prevent France from conquering Egypt. As one of the common actions of the unified fleet, Corfu and the smaller islands belonging to it were captured, together with some harbors on the Albanian coast near the entrance to the Adriatic. These territories had belonged to Venice, but were occupied by French forces. The conquered territories were turned into a new vassal state called the Republic of the Seven Islands (Cezayir-i Seba’-i Müctemia Cümhuru). The Ottoman–Russian treaty regulated the relationship of the islands to the Sublime Porte, which followed the Ragusan example,

69 “(. . .) altı ayda bir elli bin akçe hazine-i ‛amireye teslim edelim,” DAD Acta Turcarum (‛ahdname-i hümayun 1649); Glavina, “An Overview,” 156. 70 Biegman, The Turco–Ragusan Relationship, 55–59. 71 On the seventeenth century, see Glavina, “An Overview,” 139–166; DAD Acta Turcarum (‛ahdname-i hümayun 1649). The envoys bringing the harac continued to travel to Istanbul as late as 1804, see Vesna Miović, Dubrovačka diplomacija u Istambulu [Ragusan diplomacy in Istanbul] (Zagreb and Dubrovnik, 2003), 253–257. Only one year after the peace treaty, Ragusa had already paid the tribute, as is shown by a document confirming this act, see Friedrich Giese, “Die osmanisch-türkischen Urkunden im Archive des Rektorenpalastes in Dubrovnik (Raguza),” in Festschrift Georg Jacob zum siebzigsten Geburtstag 26. Mai 1932 gewidmet von Freunden und Schülern, ed. Theodor Menzel (Leipzig, 1932), 45. 72 DAD Acta Turcarum (‛ahdname-i hümayun 1756). Similarly, the payment was excused in 1703, see Harris, Dubrovnik, 345–346. the system of autonomous muslim and christian communities 399 but with a much smaller tribute to pay as cizye (75,000 guruş). The new state—together with Dubrovnik—was ceded to the French in the peace of Tilsit (1807) and the Porte lost its short-lived vassal.73

Moldavia and Wallachia Moldavia and Wallachia can be seen as the best examples of Christian vassal states in the Balkans; unfortunately, few documents have been preserved concerning the inauguration of the voievods. Romanian historiography has paid significant attention to the question of the tribute, and there are a huge number of sources on the harac, the cizye, and the pişkeş (gift).74 The various lists show diverse data, a result of the typological and linguistic variety of sources that give the sums in different currencies and also show the fluctuating exchange rate between the gold florin and the akçe.75 The best data are to be expected from sources of the Ottoman financial administration, a number of which Mihai Maxim found recently during systematic archival research in Istanbul.76 Wallachia paid tribute from 1394; this can be seen as the beginning of its vassal relationship with the sultans. Although there are references to a payment of Moldavian tribute from as early as 1377, the acknowledgment of permanent payment and Ottoman sovereignty can be dated to 1455. The tribute of Wallachia was 10,000 florins in 1474 and 14,000 in 1480. Recently discovered sources show that it was raised to 1,100,000 akçes (approximately 20,000 Hungarian florins) in 1541/42.77 Earlier it was assumed that the tribute was doubled from

73 İsmail Hakkı Uzunçarşılıoğlu, “Arşiv vesikalarına göre Yedi Ada Cumhuriyeti” [The republic of the seven islands according to archival documents], Belleten 1, no. 3–4 (1937): 627–639. 74 Mihai Berza, “Haraciul Moldovei şi al Ţării Româneşti în sec. XV–XIX” [The harac of Moldavia and Wallachia from the fifteenth to the nineteenth century], Studii şi Materi- ale de Istorie Medie 2 (1957): 7–47; Tahsin Gemil, Românii şi Otomanii în secolele XIV–XVI [Romanians and Ottomans from the fourteenth to sixteenth century] (Bucharest, 1991), 211–219. 75 The basis of the exchange was Venetian, Egyptian or Hungarian gold florins in the fifteenth and sixteenth centuries; although the difference among them was not really huge it still results in a divergence of several percent. 76 Mihai Maxim, “Circonstances de la majoration du kharadj payé par la Valachie à l’Empire Ottoman durant la période 1540–1575,” and idem, “Recherches sur les circonstances de la majoration du kharadj de la Moldavie entre les années 1538–1574,” both in idem, L’Empire Ottoman au nord du Danube et l’autonomie des Principautés Roumaines au XVIe siècle: Études et documents (Istanbul, 1999), 215–230 and 186–214, respectively. 77 Maxim, “Circonstances,” 227. According to the Romanian literature, this sum was paid as early as the 1530s, see ibid., 217; İnalcık, “The Ottoman State,” 67. 400 sándor papp

1543 on, but a fiscal record from 1549 proved this wrong—there was only a small addition to the earlier payments, in the form of salt deliveries in the value of 150,000 akces to the kitchen of the sultan.78 By the 1550s the tribute had probably reached 50,000 florins yearly.79 Due to a devastating predatory raid from Vidin, the divan decided to lower the tribute by 400,000 akçes (6,800 florins), but this was abrogated after a month and the arrears had to be paid retroactively.80 Four months later, the Porte raised the Wallachian tribute by 8,000 florins and the Moldavian tribute by 5,000 florins.81 This rise in the tribute is discussed in the literature as if the payment expected from Wallachia was to be doubled, but actually the addition did not amount to more than 16 percent. In parallel it should be noted that István Báthory, prince of Transylvania, was forced by the Porte to raise the tribute by 5,000 florins in 1575, 50 percent of the sum paid until that point.82 Based on a teşrifat defter, a register including the appointments of Otto- man dignitaries, Mihai Maxim notes that the tribute was further raised to 5,850,000 akçes (97,500 florins) in 1568/69.83 Until the seventeenth century the sum remained the same; the increases derived from the inflation of the akçe and the corrections were in accordance with that: in 1574/75, Wallachia had to pay 6,150,000 akçe (102,500 florins), whereas between 1575 and 1584 its tribute was 104,237 florins.84 The tribute paid by Mihnea II (who later turned renegade and was known as “Turcitul”)—according to his berat (the first to be preserved among those inaugurating voievods) was 7,000,000 akçes (117,000 florins) at the time of his appointment in 1577, and also at his second appointment in 1583.85 Although the nominal value

78 Maxim, “Circonstances,” 228. 79 This is also confirmed by a record from the Ottoman financial administration from 1564, which mentions 2,850,000 akçes, that is, 48,305 florins. Ibid., 217; İnalcık, “The Otto- man State,” 67. 80 7 Numaralı Mühimme Defteri (975–976 / 1567–1569), ed. Murat Şener, Nurullah İşler, and Hacı Osman Yıldırım (Ankara, 1999). 267, No. 547; Maxim, “Circonstances,” 220. Cf. 7 Numaralı Mühimme Defteri, 367, No. 755. 81 7 Numaralı Mühimme Defteri 59, No. 1306; Maxim, “Circonstances,” 219. The source refers only to the added sum of the Moldavian tribute, but Mihai Maxim is surely also right about the Wallachian tribute. 82 Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 84–88. 83 Maxim, “Circonstances,” 220. 84 İnalcık, “The Ottoman State,” 67; Maxim, “Recherches,” 205. 85 The berat of 1577: Bayezit Kütüphanesi, Elyazmaları No. 1970, fol. 7v–8r, published in Romanian translation with a photocopy, but without indicating where it has been preserved, by Tahsin Gemil, “Documente turceşti inedite (sfîrşitul sec. XVI şi XVII)” [Unpub- lished Turkish documents (late sixteenth and seventeenth centuries)], Revista Arhivelor 3 the system of autonomous muslim and christian communities 401 of the akçe grew during the time of Radu Şerban (1602–1611), the actual value of the tribute almost halved because of inflation, as it was 72 yük (7,200,000 akçes), that is, exactly 60,000 florins.86 Besides the tribute, a predefined sum of gifts had to be presented to the Sublime Porte. The tribute of Moldavia was 2,000 florins in 1455, 6,000 from the end of the 1470s, 10,000 from 1527/28 (and still the same in 1538), and 15,000 in 1541. This sum then remained stable for years, as is shown by an Ottoman fiscal document from 1542/43.87 In the next decade, it was raised again: in 1551, the tribute was 20,000 and in 1553 it was 30,000 florins. It rose again in April 1568 with an additional 5,000 florins; in 1574, it was 50,000, from 1591/92 on, 60,000 florins per year.88 This sum was then lowered in the early seventeenth century, as in 1620 Moldavia had to pay 5,600,000 akces, that is, approximately 38,000 florins.89

(1981): 351–361. The berat of 1583: BOA Maliyeden Müdevver defteri No. 17932. 11–12, published with Romanian and French translations by Mihai Maxim, Culegere de texte otomane [Collection of Ottoman texts], vol. 1, Izvoare documentare şi juridice (sec. XV–XX) [Docu- mentary and legal sources, fifteenth to twentieth century] (Bucharest, 1974), 62–68 (no. 14); idem, “Sur la question des ‘ahidnâme octroyés par les sultans Ottomans aux princes de la Moldavie et de la Valachie,” Transylvanian Review 3, no. 1 (1994): 3–12; idem, “L’autonomie de la Moldavie et de la Valachie dans les actes officiels de la Porte, au cours de la seconde moitié du XVIe siècles,” in idem, L’Empire Ottoman au nord du Danube et l’autonomie des Principautés Roumaines au XVIe siècle: Études et documents (Istanbul, 1999), 11–82. 86 BOA KK DH 71. 42/113 (10 Şaban 1017 / 9 November 1608). 87 1455: Archivum Główne Akt Dawnych (Warszawa), Kart. 66. Teczka 1. no. 1 [The letter of Sultan Mehmed II to Petru III Aron, voievod of Moldavia], Mustafa A. Mehmet, Documente turceşti privind istoria României [Turkish documents concerning the history of Romania], vol. 1, 1455–1774 (Bucharest, 1976), No. 1 (a Romanian translation with a list of earlier editions). 1470s: Süleymaniye Kütüphanesi, Esad Efendi No. 3369. Istanbul; Aurel Decei, “Tratatul de pace—sulhname—încheiat între sultanul Mehmed al II-lea şi Ştefan cel Mare la 1479” [Peace treaty/sulhname concluded between Sultan Mehmed II and Ştefan cel Mare in 1479], Revista Istorică Română 15, no. 4 (1945): 465–494; idem, Relaţii Romano- Orientale [Romanian–eastern relations) (Bucharest, 1978), 118–139; Sándor Papp, “Ştefan cel Mare, le roi Mattias et l’Empire ottoman,” in Enjeux politiques, économiques et militaires en Mer Noire (XIVe–XXIe siècles): Études à la mémoire de Mihail Guboglu, ed. Faruk Bilici, Ionel Cândea, and Anna Popescu (Brăila, 2007), 363–390; idem, “Keresztény vazallusok az Oszmán Birodalom észak-nyugati határainál (Diplomatikai vizsgálat a román vajdák szultáni ‘ahdnâméi körül)” [Christian vassals at the northwestern borders of the Otto- man Empire: Diplomatic analysis of the ‘ahdnames for the Romanian voievods], Aetas 17, no. 1 (2002): 84–85 (a Hungarian translation of the text); 1527–1528: İnalcık, “The Ottoman State,” 67; 1541 and 1542–1543: Maxim, “Recherches,” 186–214. 88 1551–1553: İnalcık, “The Ottoman State,” 67. 1568: 7 Numaralı Mühimme Defteri 59, No. 1306; Maxim, “Recherches,” 192. 1574: İnalcık, “The Ottoman State,” 67. 1591–1592: Sán- dor Papp: “Moldavia,” in Encyclopedia of the Ottoman Empire, ed. Gábor Ágoston and Bruce Masters (New York, 2009), 389–391. 89 “(. . .) meziyet-i ‘atifet-i mülukanem zuhura getürüb vilayet-i Boğdan voyvodalığı isbu sene [missing] güninden vilayet-i mezburenün mu‘ayyen olan elli altı kere yüz bin akçe haracı sal be-sal südde-i refi’ü l-bünyânuma irsal etmek üzre hatt-i hümayun-i 402 sándor papp

From the documents regulating the contacts of the Romanian principalities with the Sublime Porte, the first one preserved—a trade agreement between Moldavia and the Ottoman Empire—dates to the summer of 1456, the period just before Mehmed II began the siege of Belgrade. In exchange for concluding peace with the sultan, Voievod Petru Aron secured the right for merchants from Akkerman to trade in the Ottoman cities of Edirne, Bursa, and Istanbul.90 The only actual peace treaty the text of which is preserved is available in a copy without a date; according to the Romanian literature, it was drafted in 1479–80, but it is more likely that the real date is 1477.91 The text is, beyond doubt, a treaty (sulhname/‛ahdname) between the sultan and a vassal ruler swearing loyalty after a revolt and uses a language far more elaborate and sophisticated than most documents known from the period of Mehmed II. The relationship between the Sublime Porte and the rulers of Moldavia and Wallachia changed in the first half of the sixteenth century. Instead of the treaties described above—which in any case did not function as documents of appointment or confirmation—berats were used to transfer the sultan’s power over the voievodates. Two charters of this type survive from the late sixteenth century; both were given to Mihnea Turcitul, voievod of Wallachia, in 1577 and 1586, and both clearly define the sum of the tribute and the gifts to be delivered each year.92 The insignia of inauguration—similar to the practice in Transylvania—were granted to the voievods in several steps: thus, apart from the berat, each appointment was documented with two letters of order (of the type hükm-i hümayun),

se‘adet-makrunumla tevcih u ‘inayet idüb,” Feridun, Mecmu‛a, 488–489; Sándor Papp, “Christian Vassals on the Northwest Border of the Ottoman Empire,” in The Turks, vol. 3, Otto- mans, ed. Hasan Celal Güzel, C. Cem Oguz, and Osman Karatay (Ankara, 2002), 719–730. 90 Editions of the text: Kraelitz, Osmanische Urkunden, 44–46; Mihail Guboglu, Paleografia şi diplomatica turco-osmană: Studiu şi album [Ottoman Turkish palaeography and diplomatics: A study and album] (Bucharest, 1958), 131–132; Papp, “Keresztény vazallusok,” 84 (no. 1). 91 Maxim, “Recherches.” On the dating, see Decei, “Tratatul de pace”; Mihai Maxim, “Ştefan cel Mare şi Înalta Poartă: noi documente turceşti” [Stephen the Great and the Sublime Porte: New Turkish documents], Revista de istorie militară 3 (2004): 17–24; Mihail Guboğlu, “Fâtih’in Ştefan çel Mare üzerine iki Boğdan seferi (1474–1476)” [The two Mol- davian campaigns of Fatih against Stephen the Great], Belleten 57 (1983): 138–194; Panaite, The Ottoman Law, 267. Cf. Papp, “Ştefan cel Mare.” Şerban Papacostea’s analysis of the developments in foreign policy—although he does not elaborate on the treaty between Ştefan cel Mare and Mehmed II—also confirms my conclusions concerning the dating; see his “Caffa et la Moldavie face a l’expansion ottomane (1453–1475),” in Colocviul româno- italian “Genovezii la Marea Neagră în secolele XIII–XV,” 27–28 martie 1975, ed. Ştefan Paşcu (Bucharest, 1977), 131–153. 92 See the editions in note 86. the system of autonomous muslim and christian communities 403 one as a notification of the sultan’s decision and one accompanying the transfer of banners as symbols of power.93 Only one berat from the seventeenth century has survived in the full original text—the one deposing Gaspar Graziani and appointing Alexandru Iliaş—but extracts and copies from the eighteenth and nineteenth centuries show that this order of inauguration survived until the end of the Ottoman vassalage.94 On the method of selecting the future voievods, we have some information from the second half of the eighteenth century: in 1777, the grand vizier (probably Derviş Mehmed Pasha) presented Sultan Abdülhamid I with a list of the names and short descriptions of seven Greek dignitaries from Fener; and the one which the sultan pointed at—in this case, Constantin Moruzi—was appointed voievod of Moldavia.95 According to the Descriptio Moldaviae of Dimitrie Cantemir, an early eighteenth-century Romanian scholar and himself the voievod of Moldavia between November 1710 and July 1711, the voievods’ appointment was limited to three years and then had to be renewed.96 A large number of primary sources also document this custom from the earlier century. Here, let me refer to only two examples. In October 1607, a ferman of the sultan informed Radu Şerban—who suffered many interruptions during his rule, but between 1602 and 1610 was continuously the voievod of Wallachia— that if he did not pay the tribute before a given point in time the voievodate would be given to someone else; which, being the only stipulation in

93 Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 135–140. See also the description of Dimitrie Cantemir based on his own experiences: Moldva leírása [Descrip- tion of Moldavia] (Bucharest, 1973), 80–108. A similar process of employing a berat and hükm was also used in the case of Catherine of Brandenburg when she was appointed princess of Transylvania in 1627 (while her husband was still alive); a hükm reinforced her rule after his death in 1629; see the corresponding section. 94 See, for example, the extracts from the inauguration letter for Ştefan Cantacuzino of Wallachia (14 June 1714), Valeriu Veliman, Relaţiile româno–otomane (1711–1821): Documente turceşti [Romanian–Ottoman contacts 1711–1821: Turkish documents) (Bucharest, 1984), 90–91 (no. 13); Papp, “Ştefan cel Mare,” 363–390. 95 İsmail Hakkı Uzunçarşılıoğlu, “On sekizinci asırda Buğdana voyvoda tayini,” [Appointment of the voyevod to Moldavia in the eighteenth century], Tarih Semineri Der- gisi/Travaux du Séminaire d’Histoire 1 (1937): 32–37. 96 On the first occasion, the letter of inauguration included the following section: “Voyvodalığı tevcih olunmak ihsan-i hümayun olmışdur” [I give the voievodship of my grand-lordly benevolence]; when it was renewed, the text changed to “Voyvodalığı tecdid u mukarrer olunmak ihsan-i hümayun olmışdur” [I renewed and reinforced the voievodship of my grand-lordly benevolence]. Cantemir, Moldva, 107. The Turkish phrases, given in the original in the Latin version, are only reproduced in translation in the eighteenth- century German edition: Dimitrie Cantemir, Beschreibung der Moldau: Faksimiledruck der Originalausgabe von 1771 (Bucharest, 1973), 169. 404 sándor papp the document, points to the fact that this was far from obvious.97 More straightforward is the report of the imperial ambassador Casanova from November 1667, which notes that after having paid the money requested from them, the Porte reinforced the voievods of Wallachia and Moldavia in their positions, the former of which also went to Constantinople for this purpose.98 He refers to Alexandru Iliaş (voievod of Moldavia between May 1666 and November 1668) and Radu Leon (voievod of Wallachia between December 1664 and March 1669), which means that the reinforcement of the voievods’ rule may have been necessary at even shorter intervals than the three years Cantemir suggests.99

Hungary and Transylvania A large number of sources are available for the study of the vassalage of Hungary and later Transylvania, a principality that was established in the sixteenth century during a long process ending in 1570, based partly on the medieval Transylvania—a province of the Kingdom of Hungary with some limited autonomy—and partly on the eastern territories of the kingdom. The early contacts, when the Sublime Porte and the Kingdom of Hungary still treated each other as equals, are represented by two peace treaties preserved in the archives of the Topkapı Sarayı Museum (Istanbul). The

97 “(. . .) voyvodalığın ahere verilmek ihtimali yokdur” BOA Mühimme defteri 76. 49/123 (10 Cemâziy ül-ahir 1016 / 2 October 1607); Tahsin Gemil, Relaţiile Ţărilor Române cu Poarta Otomană în documente turceşti (1601–1712) [Contacts of the Romanian countries with the Ottoman Porte in Turkish documents, 1601–1712] (Bucharest, 1984), 116–117, No. 25. 98 “Die wallachische und Moldauische Fürsten seindt durch erlegung der von ihnen begehrten Summen von der Porten in ihren Fürstenthumben bestätigt worden, der wallachische befindt sich alhier, sein theil geldt darzu zuversamblen, hat sich dem Sultan, alß selbiger von der iagt käme, in hieherweg mit grossem pracht praesentiert, dessen gehorsamb der Sultan gelobet, ihn ermahnet alßo weiter fortzufahren vnd ein Caftan verehret, mit disen so machenden verenderungen der fürsten, vnd deßwegen so offt- mahligen geld forderungen, werden diese beyde fürstenthumber nothwendig gar ruiniert werden.” Casanova’s report from Constantinople (6 November 1667) HHStA Türkei I. (Tur- cica) Kt. 139. Konv. C. fol. 140. 99 Further examples could also be quoted, such as that of Gheorghe Duca, who— after having been deposed twice already—was the voievod of Wallachia between 1678 and 1683. His rule was reinforced in July 1681; see the report of the imperial ambassador, Kaunitz (Constantinople, 31 July 1681) HHStA Türkei I. (Turcica) Kt. 151. Konv. A fol. 82–83: “Den 28 dito hat der Duca Begh aus Moldau beym Groß vezier offene audienz gehabt, dieser ist von Neuem Zum fürsten in Moldau confirmirert, vnd zugleich an statt des Inzko Chmilinsky über die Casaken als Hatman erklärt worden, von selbiger Provinz begehrt de facto die Porten kein tribut, sodern Er, Duca Begh solle sich befleissen, das nur das Landt wiederumb populiert werde, damit die Porten Künfftig zu ihrer disposition /83v/ allezeit etlich Tausendt Cosaken derselbst aufbringen möge.” the system of autonomous muslim and christian communities 405 earlier one dates to 1488 and is probably a draft version of a peace treaty between Sultan Bayezid II and Matthias Corvinus of Hungary.100 The second one was drawn up in the name of Bayezid II when he confirmed the peace with Wladyslaw II, king of Hungary, on 5 November 1503.101 The Hungarian State Archives also preserves the text of two treaties, but as Latin ratification copies, the originals of which were sent to the Porte with ceremonial envoys. One is the ratification of the above-mentioned treaty of 1503, drawn up on the order of Wladyslaw II; the other was produced by the chancellery of Louis II in 1519. According to Hungarian historiography, this was the last Hungarian–Ottoman treaty concluded before the battle of Mohács.102 The two versions of the 1503 treaty that were ratified by the different rulers do not show any divergence in the contents, even if the introductory formula and the concluding oath were transformed according to the expectations of their own religion and chancellery traditions. It is thanks to the research of Dariusz Kołodziejczyk that we also know the earliest preserved treaty: he identified the texts, published more than fifty years later, as versions of the peace treaty concluded between Sultan Murad II and Wladyslaw I, king of Hungary and Poland, in Edirne, on 12 June 1444.103 During the fifteenth and sixteenth centuries, treaties

100 TSMA E 5861. György Hazai, “A Topkapu Szeráj Múzeumának magyar vonatkozású iratai” [The documents of the archives of Topkapı Sarayı Müzeşı relevant for Hungary], Levéltári Közlemények 26 (1955): 294–295; idem, “Urkunde des Friedensvertrages zwischen König Matthias Corvinus und dem türkischen Sultan 1488,” in Beiträge zur Sprachwissen- schaft, Volkskunde und Literaturforschung: Wolfgang Steinitz zum 60. Geburtstag am 28. Febr. 1965, ed. Aleksandr V. Isačenko, Wilhelm Wissmann, and H. Strobach (Berlin, 1965), 141–145. 101 TSMA E 7675. See in facsimile in Tayyib Gökbilgin, “Korvin Mathias (Mátyás)ın Bayezid II.e mektupları ve 1503 (909) osmanlı—macar muahedesinin türkçe metni/La traduction des letters de Korvin Mathias á Bayezid II. et le texte turc du traité Hungaro- Ottomans de 1503 (909),” Belleten 87 (1958) 369–390, nos. III–XXII. An Ottoman envoy brought this document to Buda, where the king took an oath to keep it. The fact that it can be found in a Turkish archive means that it must have been brought back to Istanbul after the conquest of Buda, that is, after 1526, 1529 or 1541. 102 The treaty of 1503: Magyar Országos Levéltár (Hungarian State Archives) [henceforth: MOL] DL 30498. The copy preserved in the archives is fragmentary, as is its edition here: Lajos Thallóczy and Sándor Horváth, Jajcza (bánság, vár és város) története: Codex Diplomaticus Partium Regno Hungariae Adnexarum (Banatus, Castrum et Oppidum Jajcza) [The history of the banate, castle, and town of Jajce] (Budapest, 1915), 167–170. For the full text, see Joseph Hammer-Purgstall, Geschichte des Osmanischen Reiches, vol. 2 (Pest, 1828), 616–620. The treaty of 1519: MOL DL 24393. For an interpretation, see Domokos Kosáry, Magyar külpolitika Mohács előtt [Hungarian foreign policy before Mohács] (Buda- pest, 1978), 146–147. 103 Dariusz Kołodziejczyk, “Der Vertrag von Szegedin 1444: Ein Beitrag zur Geschichte der frühosmanischen diplomatischen Beziehungen mit den christlichen Staaten,” in Turkologie heute: Tradition und Perspektive: Materialien der dritten Deutschen Turkologen-Konferenz, 406 sándor papp were always ratified through ceremonial embassies sent by both rulers, a tradition which was later kept in Habsburg–Ottoman contacts.104 From the Ottoman period of Hungary (1526–1699) we have thirty-one documents that serve to define the character of vassalage.105 The document in which the vassal relationship was established is unknown, in spite of the fact that the “‛ahdname of sultan Süleyman” was an important reference point in early modern legal debates as the basis of any further letters of inauguration that the Ottoman Empire gave to the princes of Transyl- vania. This document, presumably drawn up for János Szapolyai, the first king of Hungary to accept Ottoman sovereignty, is actually a later forgery, created using parts of the Polish–Ottoman peace treaty of 1528.106 Never- theless, a contemporary Hungarian translation of the agreement of 1540 is preserved in which the sultan acknowledged the son of the deceased Szapolyai as king of Hungary and demanded the payment of 50,000 gold coins as tribute, which had never been transferred to the Porte before.107 A typological survey of the surviving documents shows interesting results. From the time of Süleyman I, a new ruler of Hungary (and later Transylvania) received an ‛ahdname-i hümayun from the sultan that was quite similar in content to the berats given to Moldavian and Wallachian rulers, but clearly different from them in structure and how they labeled themselves. These documents were drawn up in the form of sultan’s letters (name-i hümayun), which were also used for correspondence with the kings of Poland, France, and the emperor of the Holy Roman Empire. Thirteen such documents have survived in the original, a later copy, or in a contemporary Hungarian or Latin translation. The introduction of this type begins by naming the sultan and then enumerating the territories

Leipzig, 4–7. Oktober 1994, ed. Nurettin Demir and Erika Taube (Wiesbaden, 1998), 175–183. See the text of the treaty also in Oskar Halecki, The Crusade of Varna: A Discussion of Controversial Problems (New York, 1943), 88–90; Kołodziejczyk, Ottoman–Polish Diplomatic Relations, 99–109. 104 Sándor Papp, “Hungary and the Ottoman Empire (From the Beginning to 1540),” in Fight against the Turk in Central-Europe in the First Half of the 16th Century, ed. István Zombori (Budapest, 2004), 37–90. 105 I have published the documents up to 1606 in my dissertation (see Papp, Die Verlei- hungs-, Bekräftigungs- und Vertragsurkunden), and a follow-up volume for the seventeenth century is currently in the making. A list of the relevant documents appears in the appen- dix of the Hungarian version of this chapter in Keletkutatás (2011) tavasz, 67–71. 106 Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 27–51; Papp, Hungary and the Ottoman Empire, 70–76; idem, “II. Rákóczi György és a Porta” [György Rákóczi II and the Porte], in Szerencsének elegyes forgása: II. Rákóczi György és kora, ed. Gábor Kármán and András Péter Szabó (Budapest, 2009), 110–118. 107 Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 43. the system of autonomous muslim and christian communities 407 in his power, followed by stating the reason for drafting the document.108 This pattern was generally used in the ‛ahdnames given to the rulers of Hungary and Transylvania, even if sometimes the introduction characteristic for berats preceded the name-type enumeration.109 Between 1608 and 1648, five ‛ahdnames were drawn up for Transylvanian rulers, which— although they included the word berat as a self-reference—featured more and more prominent characteristics of the name type.110 At the same time, the documents given to the voievods of Moldavia and Wallachia show the formal elements of the berat/nişan/menşur used for appointing dignitaries in the Ottoman Empire.111

108 See, for instance, the introduction to the first ‘ahdname given to István Báthory in 1572: “Erdel begleri tarafına isdar buyurılan ‘ahd-name-i hümayun suretidür: Tevki’-i refi’-i hümayun vasıl olıcak ma’lum ola ki benki sultanü s-selatin burhanü l-havakin tac- bahş-i husrevan-i ruy-i zemin Ak denizün ve Kara denizün ve Rumili ve Anatolınun ve Erz-i Rum ve Diyar-i Bekr ve Kürdistan ve Mısr ve Haleb ve Bagdad ve ‘Aden vilayetler- inün ve Deşt-i Kıpçak iklimlerinüñ ve Cezayir ve vilayet-i Vân ve Gürçistan ve Cezire-i Kıbrıs ve dahı şemşir-i zafer-te’sirimüz ile feth u teshir olunan vilayet u memalikün hazret-i Hakkun ‘inayeti ve iki cihan fahri peygamberimüz salla llahu te’ala ‘aleyhi ve- sellem efendimizün mu’cizat-i kesirü l-berekatiyle padişahi Sultan Selim han bin Sultan Süleyman hanam vilayet-i Erdel voyvodası Batori İştvan ve sa’ir begler südde-i seniye-i se’adet-medar ve ‘atebe-i ‘aliye-i gerdun-iktidarımuza ki melaz-i selatin-i cihan ve melca-i havakin-i deverandur ademler ile mektublar gönderüb,” Feridun, Mecmu‛a, 461; Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 214–219. 109 “The order of the signature of Grand Lord is the following: I, who am sultan Murad Khan” (in the original: “Nişân-i hümâyûn hükmi oldur ki ben ki ‘lh [that is, ‘lh: ilâ âhiri (Arabic) “etc.”] Sultân Murâd hânam,” see the ‛ahdname to István Báthory in Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 223–227. 110 See, for instance, the ‛ahdname given to Gábor Báthory in 1608: “ben-ki sultanu selatin-i şark u garb sahib-kiran-i memalik-i Rum ve ‛Acem ve ‛Arab tac-bahş-i husrevan-i cihan zıllu llahi l-meliki l-mennan sani İskender zu l-karneyn hadimü l-harameyni ş-şerifeyn hafız-şi‛arü kıbleteyn eşrefi l-medayin ve-l emsar ve akdasu memalik ve-l aktar,” Österreichische Nationalbibliothek Handschriftensammlung [henceforth: ÖNB Hs.]; Mixt, 1598. 111 See, for instance, the berat given to Alexander Iliaş, voievod of Moldavia: “çûn hazret-i kâdir u hâb ve cenâb-i râzikü müsebbibi l-esbâb celle şânuhu ve te‛âlâ ve ‛amma ihsânuhu ve tevâlî-i kemâl-i kerem ve ‛inâyet ve fazl-i vücûd-i ‛adîmü n-nihâyetden ‛atebe-i ‛alîye-i ‛izzet-medâr ve südde-i se‛âdet-i karârumı maksim-i irzâk-i kâffe-i ümem ve âsitâne-i devlet-âşyân-i rif‛at-bünyânumi melâz-i ‛âmme-i hükkâm-i sümmüvvü l-hemmüm eyledi lâ-cerem-i tevâlî-yi elâ-i sübhânî ve nu‛mâ-i bî -müntehâ-yi rebbânînün edâ-i merâsim-i şükri ve îfâ-i levâzim-i hamdi içün zimmet-i himmet-i ‛alîye-i şâhâne ve ‛izzet-i nişân-i pâdişâhâneme vâcib u lâzim olmışdur-ki âsitâne-i müşeyyidü l-erkânuma hulûs-i fu’âd ile itâ‛at u inkiyâd edüb hidmet-i ‛ubûdîyetümle iftihâr-i aydin-i ‛ibâd-i sadâkat-i nihâdum hakklarında hemvâre-i ebvâb-i ‛inâyet u ihsânum meftûh u küşâde ve esbâb-i mevhebet bî-imtinânum hemîşe-i âmâde olub merâm u mukâsıdları husûli ile şâdigâm olalar binâen ‛alâ zâlike boğdân voyvodası olan Gaşpâr voyvodanun ba‛z-i nâ-hemvâr-i evzâ‛ u atvâri zuhûr edüb,” Papp, “Keresztény vazallusok,” 92; Feridun, Mecmu‛a, 488–489. 408 sándor papp

Apart from the ‛ahdnames, there are also inauguration documents from this period which were drawn up in berat form, but only in cases when the princes of Transylvania needed a quick acknowledgment of their power from the Sublime Porte. Such was the case when Grand Vizier Lala Mehmed appointed István Bocskai king of Hungary and prince of Transylvania in 1604; at the time, Bocskai actually held power over the non-Ottoman part of Hungary and was in dire need of support for his anti- Habsburg movement. In a similar vein, in early 1607 Kuyucu Murad Pasha also sent a berat to Bálint Homonnai Drugeth, appointing him king of Hungary and prince of Transylvania, in order to secure the succession of Bocskai, who had died shortly before.112 The Hungarian politicians were, however, never content with receiving a berat only, and supplicated—for the most part successfully—for the issuing of an ‛ahdname, as happened in the case of Bocskai in 1605.113 Berats were also issued for princes on the occasion of their inaugura- tions. As early as the sixteenth century this was a four-step process; after the Transylvanian estates informed the Sublime Porte about the election of their new ruler, at least two sultan’s orders ( fermans) were sent to the country—one for the estates and one for the new prince—confirming the acknowledgment of his rule by the sultan. Later, a ceremonial embassy delivered the insignia of inauguration: the banner (sancak), the cap (üsküf, börk), the mace (topuz), the panache (sorguç), the saber (tiğ, seyf, kılıç), and at least two kaftans (hit‛at).114 This embassy also brought an elaborate sultanic letter of order, but the ‛ahdname-i hümayun was only drafted after the princes had paid a certain berat fee through their envoys.115 From the seventeenth century on, the system was simplified. It became custom for the ‛ahdname (only) to show the characteristics of the name type

112 Both are preserved in the original; berat-i hümayuns given to István Bocskai (5–14 November 1604), Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 258–261 and to Bálint Homonnai Drugeth (whose name was later changed to Zsigmond Rákóczi); 20–29 January 1607, Arhivele Naţionale României [henceforth, ANR], Direcţia Braşov, Arh. Bisericii Negre E. IV, 150. 113 ‛Ahdname-i hümayun given to István Bocskai from Sultan Ahmed I (20 April–18 May 1605), the text of the contemporary Ottoman translation and Hungarian translation are edited in Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 267–286. In contrast, Homonnai Drugeth never actually assumed power in Hungary and Transylvania. 114 On the insignia of inauguration, see János B. Szabó and Péter Erdősi, “Ceremonies Marking the Transfer of Power in the Principality of Transylvania in East European Con- text,” Majestas 11 (2003): 111–160. 115 Strangely enough, none of these berats are preserved from this period; the data may refer to the ‛ahdname, see Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 136. the system of autonomous muslim and christian communities 409

(contrary to, for instance, the nişan type used for Dubrovnik), use a mixture of self-identificatory terms, and also fulfill the function of the inau- gurational berat. Such a document, referring to itself both as ‛ahdname-i hümayun and berat-i şerif, was sent, for example, to Gábor Bethlen in 1614; to Catherine of Brandenburg, his wife and appointed successor, in 1627; to György Rákóczi I in 1631; and to György Rákóczi II in 1649.116 In three cases during the seventeenth century Transylvanian princes managed, during their lifetimes, to have their sons appointed as successors; the documents used on these occasions were berats similar to those granted to the voievods of Moldavia and Wallachia. The documents given to György Rákóczi II in 1642, to Ferenc Rákóczi I in 1652, and to Mihály Apafi II in 1684 did not, however, actually secure their rule, but only confirmed that they were the successors of their fathers; when they assumed actual power their rule had to be confirmed again by the Otto- man administration.117 A turning point in the history of Transylvania came in 1657, when Prince György Rákóczi II invaded Poland in alliance with Sweden without the consent of the Sublime Porte, and was deposed as a consequence. The inauguration documents (of which no copy is preserved) for the new prince were already on their way to the principality when Rákóczi assumed power again. In 1658, shortly after conquering the castle of Borosjenő, Grand Vizier Köprülü Mehmed, who led the punitive campaign personally, gave the princely title to Ákos Barcsai (the head of the Transylvanian delegation that begged for his mercy). The conditions of his rule were detailed in a hüccet, a form of order of the kadiasker from

116 Gábor Bethlen’s (1 July 1614) is found in Feridun, Mecmu‛a, 443–446. Catherine of Brandenburg’s (6–16 April 1627) is available in a contemporary Ottoman copy, together with German and Hungarian translations at the Geheimes Staatsarchiv Preußischer Kul- turbesitz (Berlin-Dahlem), I. Hauptabteilung, Geheimer Rat, Rep. 11, Auswärtige Beziehun- gen, 255a Siebenbürgen No. 3. vol. 3. fol. 339–344. György Rákóczi I’s (3–12 April 1631) is in the original: MOL Mikrofilmtár 21050. doboz, published by Adem Handžić, “Diploma sultana Murada IV erdeljskom knezu Đorđu Rakociju” [The charter of sultan Murad IV to György Rákóczi I, prince of Transylvania], Prilozi za orijentalnu filologiju (Sarajevo) 6–7 (1956–1957): 175–191. The original of György Rákóczi II’s (12–21 June 1649) was kept in Karlsruhe, but was destroyed in World War II. The manuscript was published by Babinger, “Zwei türkische Schutzbriefe.” There is a contemporary official copy at NSNB 4o Cod. MS. Turcica 29. fol. 62v–64r. 117 The berat given to György Rákóczi, Junior (1642, in fragment): Feridun, Mecmu‛a, 470–471. That for Ferenc Rákóczi I (2–13 December 1652) in a contemporary official copy: at NSNB 4o Cod. MS. Turcica 29. fol. 96r–96v. The one given to Mihály Apafi II (19–28 November 1684), ibid., fol. 77r–v. Ferenc Rákóczi I never became a ruling prince. 410 sándor papp a lower administration.118 Barcsai seems to have been the first ruler of Transylvania to be granted only a berat by the Porte for his inauguration, and even that document is unknown. Mihály Apafi I, who ruled for thirty years after 1661, also did not receive an ‛ahdname, only a berat, which, however, included the points of the old ‛ahdnames.119 After Apafi’s death, another Transylvanian ruler, Imre Thököly, received an ‛ahdname again (drawn up in the form of a name-i hümayun); however, his appointment was made under specific circumstances; in order to understand them, the question of the ‛ahdnames given to the estates of Transylvania or Hungary must be addressed. Some ‛ahdnames from the years 1604, 1621–1622, 1629, 1681 and 1682 were given by the sultan not to the ruler of Transylvania, but to the estates of the country; and, in a similar manner, to the estates of Upper Hungary.120

118 The original is supposed to be in the collections of the BOA, but I could not trace it. It was edited by İsmail Hakkı Uzunçarşılı, “Barcsay Akos’un Erdel kırallığına ait bazı original vesikalar [Some documents concerning the rule of Ákos Barcsai in Transylvania],” Tarih Dergisi 7 (1953): 51–64. 119 The document, dated 5–14 October 1661, is available in ANR Direcţia Generală (Bucharest; henceforth ANR DG), Doc. turc. suluri 137. The changed political situation was described by Walter Leslie, Habsburg ambassador in Istanbul: “Turks do not make the smallest difference between them [that is, the Transylvanians] and their other subjects; and should it be more useful for the Turks and the Grand Lord would find it better to place there a vizier or a pasha instead of the prince, I do not think that they would hesitate much to do it. They can get more profit from this country—just like from Wallachia and Moldavia—in this way than if there would be a pasha there. Apafi does not have the slight- est credit or trust among them; he could be deposed for the smallest reason if they can find a person in the country better fitting their interests.” ÖStA HHStA Türkei I. (Turcica) Kt. 138. Konv. B. fol. 112. 120 ‛Ahdname to the Hungarian lords from Ahmed I (30 May–28 June 1604), in a contemporary Hungarian translation: Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 258–261. ‛Ahdname for the estates of Upper Hungary from Osman II (turn of 1621–1622) Feridun, Mecmu‛a, 446–448. ‛Ahdname for the inhabitants of Kassa, the Hajdúság region and seven Upper Hungarian counties from Murad IV (end of 1629) in a contemporary Hun- garian translation: MOL R 298 Erdélyi iratok, I. tétel, fol. 79; another copy is in Rozsnyay Dávid, az utolsó török deák történeti maradványai [The historical monuments of the last Turkish scribe, Dávid Rozsnyai], ed. Sándor Szilágyi (Budapest, 1867), 77–79. ‛Ahdname to the estates of Upper Hungary from Mehmed IV (18 April 1681) in a contemporary Hungar- ian translation: Farkas Deák, A bujdosók levéltára [The archive of the exiled] (Budapest, 1888), 288–291; another copy from the eighteenth century in Tiszántúli Református Egy- házkerület és Kollégium Levéltára (Debrecen), Sinai Miklós hagyatéka, Z 38. ‛Ahdname to idem by idem (5–14 August 1682) in a contemporary official copy: BOA Nâme-i hümâyûn defteri No. 5. p. 347–352. The ‛ahdname from the turn of 1621–1622 is preserved in the Feridun collection without dating, and was mistakenly interpreted by Tahsin Gemil as the inauguration document given to Prince Gábor Bethlen in 1614, also leading Viorel Panaite to the same mistake, cf. Gemil, Relaţiile Ţărilor Române, 163–166 (No. 61); Panaite, The Ottoman Law, 267. The actual inauguration document of Bethlen was issued between 29 June and 8 July 1614; see the fragmentary edition in Feridun, Mecmu‛a, 443–446; and the system of autonomous muslim and christian communities 411

These documents, securing among other things the free election of rulers and Ottoman protection over Hungarian territories, were issued by the Sublime Porte at the request of the estates; what is more, the leaders of anti-Habsburg movements after the second half of the seventeenth century themselves drafted the ‛ahdnames they wished to have from the Sublime Porte.121 The documents from 1620, 1629, and 1681 defined the Transylva- nian prince as the ruler of the Hungarian territories under Ottoman protection, however, they did not codify their unification with the principality. In contrast, in 1682, a new principality—Middle Hungary (Orta Macar)—was founded. A treaty letter of fourteen articles was sent to the estates of the north eastern counties of Hungary (the territory of present-day East Slova- kia) based on the draft they had submitted to the Sublime Porte. The document provided full security of property and the free exercise of religion for the inhabitants of the new vassal state and in exchange prescribed the payment of 40,000 gold coins.122 This sum was also included in the inauguration berat sent personally to Thököly, which was almost an exact copy of the one granted to the Cossack hetman, Petro Doroshenko, in 1669.123 Orta Macar did not exist for more than a few years, but Thököly was appointed as prince of Transylvania in 1690. His past, as well as the insecure relationship with Transylvania—which was already as good as lost to the Ottoman Empire by that time—explain why instead of a berat, a document with higher prestige was issued for him again. All in all, the treaty letters of the sultan for the Hungarian estates can be seen as attempts to give Ottoman backing to rebels against the Habs burg kings of Hungary and thus always promised wide inner autonomy

a contemporary Hungarian translation in Erdélyi Történelmi Adatok [Transylvanian historical data], ed. Imre Mikó (Kolozsvár, 1856), vol. 2, 336–342. 121 As happened in 1666, 1682, 1705, and 1718, of which only the second received a response of any value; see Sándor Papp, “Ottoman Accounts of the Hungarian Movements against the Habsburgs at the Turn of the 17th and the 18th Centuries,” in Frontiers of Otto- man Studies: State, Province, and the West, ed. Colin Imber, Keiko Kiyotaki, and Thoads Murphey (London and New York, 2005), vol. 2, 37–48. 122 Sándor Papp, “Thököly Imrének és Közép-Magyarország népének kiállított szultáni szerződéslevél / İmre Thököly ve Orta Macaristan halkına verilen ahidname-i hümayun” [Treaty letter granted by the sultan to Imre Thököly and the people of Middle Hungary], in István Seres, Thököly Imre és Törökország/Imre Thököly ve Türkiye (Budapest, 2006), 278– 289. From another copy, edited earlier by Zdenka Veselá-Přenosilová, “Quelques chartes turques concernant la correspondence de la Porte Sublime avec Imre Thököly,” Archiv Orientální 29 (1961): 553–556, with a French translation (565–568). 123 Berat-i hümayun for Imre Thököly from Mehmed IV (5–14 August 1682) HHStA Türkische Urkunden; the 1690 ‛ahdname: MOL G 15. 3. cs. Caps A/1 19. fasc. Literae Missilis Turcicae fol. 42. 412 sándor papp and religious freedoms. This practice was also continued in the early eighteenth century. When Ferenc Rákóczi II, then living in exile in Paris, was invited by the Sublime Porte to be the head of a newly formed anti- Habsburg movement, a new ‛ahdname was drafted for him on the basis of the one designed for Thököly in 1690, but it never reached the prince. The last prince of Hungary and Transylvania appointed by the sultan was his son, József Rákóczi, who was granted an ‛ahdname-i hümayun of eleven points in the last third of the Ramazan 1150 (20–31 December 1737). The Latin version was ratified by Rákóczi, which shows that the agreement was, after several centuries, again considered bilateral.124 The Ottoman defeat in their war against the Habsburgs and the early death of the prince in 1739, however, provided no other results than the official protection and pension given to the Hungarian emigrants in Tekirdağ until the early nineteenth century. The persons appointed as their leaders (baş/buğ) received sultanic orders about their titles and the supplies granted to them; although the accounts all refer to berats that defined their revenues, none of these documents are known today.125 As mentioned above, paying tribute was accepted by Hungary in 1529, but it seems that it did not take place before 1541. The first Transylvanian tribute (of 10,000 florins) was paid in 1543.126 It was raised to 15,000 in 1575, set back to the original sum in 1624, and raised again to 15,000 after 1649.127 After the revolt of György Rákóczi II it was further raised to 40,000 in 1659, a sum which was submitted for the last time to the treasury of the sultan in 1687.128

124 BOA Nâme-i hümâyûn defteri 7. 483–486. This is is the only case when the agreement with a head-to-be of a Christian vassal state was made in a bilateral manner. On the usual unilateral character of ‛ahdnames, see Panaite, The Ottoman Law, 241–242. 125 “(. . .) ba-berat-i ‘alişan mutasarrıf oldığı yevmi yedi yüz yiğirmi üç akçe nafaka bahamızdan.” The certificate of Kelemen Mikes that he received the monthly revenue from the custom of Tekirdağ in the months of Şaban 1172 (30 March–27 April 1759), BOA, Cevdet Hariciye 808/2. 1172. 126 Gábor Ágoston, A hódolt Magyarország [Hungary under Ottoman rule] (Budapest, 1992), 51; Cristina Feneşan, “Der Harac Siebenbürgens in der ersten Hälfte des 17. Jahrhun- derts,” Revue des Études Sud-Est Européennes 34, nos. 1–2 (1996): 97–106. 127 On 1575, see Papp, Die Verleihungs-, Bekräftigungs- und Vertragsurkunden, 87. On 1624: Feneşan, “Der Harac Siebenbürgens.” On the 1640s: Sándor Papp, “II. Rákóczi György,” 124–134; also the report of the envoy István Rácz to Prince György Rákóczi I (Con- stantinople, 15 May 1642), ed. Áron Szilády and Sándor Szilágyi, Török-magyarkori állam- okmánytár [A collection of state papers from the Turkish Hungarian period] (Pest, 1870), vol. 3, 105–107; as well as the manuscript ÖNB Hs. Mixt 174, fol. 54v–55v. 128 On 1659, see the conditions for peace drawn up by Köprülü Mehmed Pasha for Prince Ákos Barcsai (MOL R 315 Török iratok No. 89). On 1687, see the document ANR DG Doc. turc. XXIX/2362; Gemil, Relaţiile Ţărilor Române, 374–380. the system of autonomous muslim and christian communities 413

Cossack Ukraine The relationship of the Cossacks, who lived in the territory of present-day Ukraine, and the Sublime Porte show the characteristics of vassal status after the revolt of Bohdan Khmelnytsky against Poland, which began in 1648. The Cossacks had held certain privileges in the Ukrainian region since the rule of István Báthory as king of Poland; this was based on their military service, primarily against the Tatars. Khmelnytsky planned to defend these liberties and sought alliances against Poland with the Crimean Khanate, Muscovy, Transylvania, Sweden, and the Ottoman Empire. As early as the spring of 1650, he sent a letter and an envoy to the Sublime Porte asking for military support and expressing his loyalty to the sultan.129 It is still an open question whether—parallel to the written alliance with the Tatars—a document existed that gave a detailed description of the relationship between the sultan and his new vassal in this period. Omeljan Pritsak has published, from a Polish translation, a treaty of trade and defense that dates to 1648, but this has generally been regarded as a forgery by recent scholarship.130 It is clear, however, from a letter of the sultan dated to February–March 1651, that the Cossack hetman asked specifically for an ‛ahdname in which the conditions of the treaty (‛uhud u şurutı) would be listed.131 Another hetman, Petro Doroshenko, later took further steps to receive help from the sultan.132 His inauguration letter (“berat-i se‛adet-feza-yi husrevani ve bu menşuru l-lazımi l-imtisal-i hakani”), dated 1–10 Muhar- rem 1080 (1–10 June 1669), is preserved both in the original and a contemporary official copy. In this, he was appointed hetman of the three tribes of the Cossacks (“Sarı kamış ve Barabaş ve Potkal üç kavim kazağı

129 Jan Rypka, “Aus der Korrespondenz der Hohen Pforte mit Bogdan Chmelnicki,” in Z dějin východni Evropy a slovanstva: Sbornik věnovaný Jaroslavu Bidlovi, profesoru Karlovy university k sědesatým narozeninam, ed. Miloš Weingart, Josef Dobiaš, and Milada Paulova (Prague, 1928), 346–350. 130 Omeljan Pritsak, “Das erste türkisch-ukrainische Bündnis (1648),” Oriens 6 (1953): 266–298, and the Turkish translation: “İlk Türk-Ukrayna ittifakı (1648),” İlmi Araştırmalar 7 (1999): 255–284. The source was first published in Moscow in 1822 as part of a collection of sources. For doubts concerning its authenticity, see Edgar Hösch, “Der türkisch-kosakische Vertrag von 1648,” Forschungen zur osteuropäischen Geschichte 27 (1980): 233–248. 131 The Sublime Porte indicated that the precondition would be receiving a Cossack embassy and a detailed exposition of the political situation, see Jan Rypka, “Weitere Beiträge zur Korrespondenz der Hohen Pforte mit Bogdan Chmel’nyćkyj,” Archiv Orien- tální 2 (1930): 269–273. 132 Beáta Varga, Önállóság, autonómia vagy alávetettség: Ukrajna 1648–1709 között [Inde- pendence, autonomy or submission: Ukraine between 1648 and 1709] (Szeged, 2008). 414 sándor papp hatmanı”) and the territory under his rule was given to him permanently as a sancak. It is clear from the document that the protection of the sultan was requested by the Cossacks with reference to the protection given to Moldavia and Wallachia. There was, however, an important difference between the Cossacks and other Christian vassals: they did not have to pay tribute in exchange.133 As noted above, the berat of Doroshenko is almost identical to that given to Imre Thököly as king of Orta Macar in 1682. These berats, however, do not elaborate on the details of the relationship between the padishah and his vassal. In the Hungarian case, it was done in an ‛ahdname accompanying the berat, and although in the Cossack case no Ottoman Turkish ‛ahdname was preserved, we know its content from a document in Russian, which regulated the details of the relationship in seventeen points.134

Privileges to Religious Communities

Data about granting the privileges of religious communities date back to the time of the prophet. In the following centuries this tradition was followed by Muslim rulers such as the Persian shahs, who granted tax exemptions and property rights to Armenian monasteries; the Persian correspondence of the Carmelite fathers living in Iran also shows the existence of similar rights in the seventeenth century under the rule of Shah ‛Abbâs the Great.135 Caliph ‛Umar (seventh century) granted the patriarch of Jeru- salem a treaty which—besides elaborating on the contents of the zimmi status—defined the rights and duties of the town’s inhabitants according to the traditional pattern.136 Another type of agreement for the patriarch of Jerusalem secured the exemption of taxes and services for the Christian community of the Holy Temple at the Mount of Olives and envisaged the collection of three dirhams from each pilgrim.137 These documents are

133 BOA, İbnülemin Hâricîye 52; Uzunçarşılı, Osmanlı tarihi, vol. 3/2, 113n1. 134 Akty, otnosiashchiesia k istorii iuzhnoi i zapadnoi Rossii [Documents concerning the history of southern and western Russia] (Saint Petersburg, 1878), vol. 9, 218–220 (No. 73). Let me express my thanks to Beáta Varga for the data and photocopy of this source. 135 Bert G. Fragner, Repertorium Persischer Herrscherurkunden: Publizierte Origi- nalurkunden (bis 1848) (Freiburg im Breisgau, 1980), 70–71 (No. 153); Manūchir Sutūdih and Iraj Afshār, Asnād-e Pādimīryān-e Karmilī bāzmāndih ez ‘aṣr-e šāh ‘Abbās Safāvī: Remaining Documents of Carmelite Padres since Shah Abbas Era (Tehran, 2004). I would like to express my thanks to Miklós Sárközy, from whom I received this volume. 136 Nagel, Das Islamische Recht, 98–100. 137 Hammer, “Sened,” 67–69. It must be noted that, apart from this text edition of Ham- mer, the other works on the history of Islamic law do not discuss this document. the system of autonomous muslim and christian communities 415 evidence that even during this early period of Islam documents securing protection and reducing duties were not only given to Christian inhabitants of certain settlements, but also to Christian institutions. These and similar documents served as examples for later Muslim dynasties, which gave letters of protection to instutionalized communities of the Christian church. Such treaties are known, for instance, in the case of the Orthodox monastery of Mount Tur in the Sinai from the Mamluk period, which was also confirmed by Sultan Selim I and his successors.138 It is also traditionally assumed—also in a recent study by Halil İnalcık—that after the fall of Constantinople, Patriarch Gennadios Scholarios II was confirmed with a berat by Sultan Mehmed II.139 In the Ottoman state it was a general custom to appoint or confirm the leaders of Christian religious communities, that is, the sultans practiced the right of investiture. Original documents are preserved primarily concerning the confirmation of Orthodox metropolitans. Although he did not edit any berats, the existence and function of these documents is discussed extensively by Aleksandar Fotić, and Josef Kabrda published a list of twenty-one berats given by the sultan to the metropolitans of the East- ern church between 1633 and 1848.140 The last one is adorned with the signature of Sultan ‛Abdulmecid (1839–1861) and grants Kalogeros A gapios, the new metropolitan of a monastery called “Old John” (Eski Yogan), the same privileges enjoyed by his predecessor, Miletos.141

138 Hans Ernst, Die mamelukische Sultansurkunde des Sinai-Klosters (Wiesbaden, 1960); Klaus Schwarz, Osmanische Sultansurkunden des Sinai-Klosters in türkischer Sprache (Freiburg im Breisgau, 1970). 139 Halil İnalcık, “The Status of the Greek Orthodox Patriarch,” in idem, Essays in Otto- man History (Istanbul, 1998), 195–223. Cf. Hammer-Purgstall, Geschichte des Osmanischen Reiches, vol. 2, 1–3; Kritobulos von Imbros, Mehmet II. erobert Konstantinopel: Die ersten Regierungsjahre des Sultans Mehmet Fatih, des Eroberers von Konstantinopel 1453, ed. Diether Roderich Reinsch (Graz, 1986), 144–145. 140 Josef Kabrda, Le système fiscal de l’Eglise orthodoxe dans l’Empire ottoman (d’après les documents turcs) (Brno, 1969), 33; Aleksandar Fotić, Sveta Gora i Hilandar u Osmanskom Carstvu XV–XVII. vek. [Mount Athos and Hilandar in the Ottoman Empire (fifteenth to seventeenth centuries)] (Belgrade, 2000), 95. See also the berats in Ottoman, with German translations: Helmuth Scheel, Die staatsrechtliche Stellung der ökumenischen Kirchenfürsten in der alten Türkei: Ein Beitrag zur Geschichte der türkischen Verfassung und Verwaltung (Berlin 1943). An interesting case study is Josef Kabrda, Berât vidinského metropolity Josefa z r. 1763: Příspevek k dĕjinám pravoslavné církve v říši osmanské [The berat of Joseph, metropolitan of Vidin, from 1763: Contribution to the history of the Orthodox church in the Ottoman Empire], Věstnik Královské české Společnosti nauk: Třída pro filosofii, historii a filologii 1937/1 (Prague 1937), 76. 141 Kabrda, Le système fiscal, 33. (original call number ANR-DG Sul. No. 76), see also Guboglu, Paleografia şi diplomatica, 160–161 (No. 192); facsimile 192. The publication does not attempt to identify the monastery. 416 sándor papp

The Ottoman state incorporated into the empire every group that could be conceptualized as a distinct entity along the same lines. One specific form was used when the community asked collectively for the confirmation of its inner autonomy and privileges. It was in this vein that Sultan Mehmed II secured the rights of the Bosnian Franciscan monks. This well- known document was drawn up, according to tradition, on 28 May 1563 on the fields of Milodraž and one copy is still kept at the Franciscan monastery of Fojnica (Bosnia). While its authenticity was debated for a long time in the literature, it is now clear that this manuscript is a copy made between 1645 and 1669, but the existence of the ‛ahdname is documented in Franciscan sources from 1486 on. Thus, current literature regards the text as unquestionably authentic.142 This document was then also used in the activities of the Franciscan monastery at Szeged; the fathers used a copy of the Fojnica ‛ahdname—authenticated by the kadi of Lippa— to counteract harassment by the Orthodox metropolitan, as an official acknowledgment that the rites of the Catholic and Orthodox churches differed and thus their coreligionists should be exempt from the church tax and services due to the representatives of the Eastern church.143

Conclusions

In conclusion, let me stress again some elements of my analysis. First, I have to emphasize that the examples above do not cover all the autono-

142 The most recent and comprehensive study on the document is Srećko M. Džaja, “Fojnička Ahdnama u zrcalu paleografije, pravne povijesti i politike: Kontekstualizacija Ahdname bosanskih franjevaca” [The ‛ahdname of Fojnica in the mirror of paleography, legal history, and politics: Contextualization of the ‛ahdname given to the Bosnian Francis- cans], Bosna Franciskana 31 (2009): 103–128. The manuscript was first described by János Asbóth, Bosnien und Herzegowina: Reisebilder und Studien (Vienna, 1888), 441. It was only after the World War II that Hazim Šabanović proved that the manuscript is a copy, see his “Turski dokumenti u Bosni iz druge polovine XV stoljeća” [Turkish documents from Bosnia from the second half of the seventeenth century], Istorisko-pravni zbornik Pravnog fakulteta Sarajevo 1–2 (1949): 201–202. See also Dominik Mandić, “Autentičnost ahd-name Mehmeda II. B. H. franjevcima” [The authenticity of the ‛ahdname of Mehmed II to the Franciscans of Bosnia-Herzogovina], Radovi Hrvatskog povijesnog instituta u Rimu 3–4 (1971): 61–90; Vančo Boškov, “Pitanje autentičnosti Fojničke ahd-nama Mehmeda II iz 1463. godine” [The question of the authenticity of the Fojnica ‛ahdname of Mehmed II from the year 1463] and “Ahd-nama Murata III stanovnicima Bara iz 1575. godine” [The ‛ahdname of Murad III to the inhabitants of Bar from the year 1575], both in Godišnjak Društva istoričara Bosne i Hercegovine 28–30 (1977–1979): 87–105, respectively 279–285. 143 Sándor Papp, “Gesetzliche Garantien für die christlichen Gemeinden im Osmanis- chen Reich: Zur Diskussion des Millet-Systems und der Vertragsurkunde der Franziskaner in Bosnien,” in Osmanische Orient und Ostmitteleuropa, ed. Robert Born and Andreas Puth (Stuttgart, 2013), forthcoming. the system of autonomous muslim and christian communities 417 mous communities of the Ottoman Empire, chronologically or spatially. As the archival research into the specific fields leaves much to be desired, apart from those published earlier I could only use the few documents I found. Although the system of vassal states and privileged communities in the Ottoman Empire was quite complicated and finely tuned, the Ottoman state had an administrative method for integrating various Mus- lim or non-Muslim autonomous communities without forcing uniformity upon them. This system also granted privileges to merchant communities coming from the outside, from Venice, Genova, France, England, the Netherlands, and so on, as well as their trade depositories, and included them in the system of the empire with trade treaties (capitulations), which were also written in the pattern of berats. On the basis of the above discussions it can be stated that the Ottoman Empire, which gives the impression of having been a society built in a strictly uniform manner, was actually a conglomerate of local communal autonomies, especially near the borders. As is clear from the data presented above, apart from the autonomous Christian vassal states, a form of autonomous “statehood” was also provided for entities on the level of countries (vilayet) or provinces (sancak) with Muslim leaders. Generally, the literature does not differentiate between the origins of the non-Muslim and the Muslim autonomous entities. In the case of the Christian states, for a long time the historiography of Islamic law referred to the examples from the time of the prophet, when communities of growing sizes were included in the Islamic state by concluding treaties and acknowledging their second-class subject (zimmi) position under the protection of the Muslim ruler. A hierarchical system among the Muslim states is a later development, and only came into being after the dissolution of the caliphate. From the examples I have quoted, the cases of Caramania and Crimea show a transformation of an equal neighborly status into a subordinate one; however, most of the autonomous entities were created in another manner. Halil İnalcık was the first to attempt a systematic description of developments in the relationships between autonomous Muslim tribal or clan- based societies ruled by dynasty or kin groups and the Islamic dynasties replacing each other through the centuries.144 According to his study, although it was also customary during the rule of the Abbasid and Selcuk

144 Halil İnalcık, “Autonomous Enclaves in Islamic States: Temlîks, Soyurghals, Yurdluḳ- Ocalḳlıḳs, Mâlikâne-Muḳâta’as and Awqâf,” in History and Historiography of Post-Mongol Central Asia and the Middle East: Studies in Honor of John E. Woods, ed. Judith Pfeiffer and Sholeh A. Quinn (Wiesbaden, 2006), 113–134. 418 sándor papp dynasties for large territories to be privately owned, the autonomy typical of these Muslim entities originated in the soyurgal system under the Ilhanid dynasty.145 Through the soyurgal diploma (also called yarlıg) the owner was granted full property rights and became exempt from every state tax and service (mu‛af u musallam). In the early Ottoman state (until 1453) the system of free property was maintained, and donations of estates were made with so-called temik-names, in the form of a berat/nişan. Many examples of the same type of document in Persian are known from Timu- rid Iran, where governors were autonomous rulers of the provinces they had under their power. As the terms soyurgal and yarlıg also show, this system has Mongol origins: the princes were autonomous rulers over their grazing areas on the steppe.146 The results of my ongoing research, not yet conclusive, can be summarized in the statement that among the vassal states of the Ottoman Empire the origins of the relationships with Christian and Muslim entities must be treated separately. While in the case of the Christian vassals the zimma system of Islamic protection provided the framework for the relationship, in the case of Muslims it was the Asian Mongol–Turkish tradition. By the sixteenth century, the differences stemming from the divergent backgrounds had disappeared and the appointment letters given to Christian vassals include the same terminology as those granted to Mus- lims. According to the analysis of the chancellery practice, the Ottoman power saw, in the person of a vassal ruler or the leader of an autonomous community, the same type of official as any other from the group of classic dignitaries of the empire. The discussion of the system’s modification in Egypt also identifies the regular method used even as Ottoman power decreased in the nineteenth century. In the provinces that the Sublime Porte could no longer govern through the earlier centralized method, it attempted an alternative solution of keeping them in the empire—often only nominally— in the framework of a revitalized vassal network. This idea of a certain

145 In the territory of today’s Azerbaijan, the Safavid Iranian shahs appointed governors called malek/melik, that is, king, who received their title with the right to make it inherit- able in their families. See, for instance, the ferman of Shah Hüseyin from August 1701: T.M. Musavi, Orta asr Azërbajcan tarihinë dair farsdilli sënëdlër [Persian documents from the medieval history of Azerbaijan] (Bakı, 1977), 74–76 (No. 10). 146 The Turkish term for it was yurd, the Mongol nutug. The word yarlıg in the sense of “letter of appointment” is also found in the terminology of Ottoman inauguration letters given to their Muslim vassals, as well as the yurt (in the form of yurtluk, completed with ocaklık). the system of autonomous muslim and christian communities 419

“Ottoman Commonwealth” can be seen behind the dependence of Serbia, Bulgaria, and Romania on the Sublime Porte in the nineteenth century.147 The Balkan Christian countries did everything to rid themselves of even this level of subordination, although it remained a slow process; in the case of Bulgaria, for instance, the final cancellation of Ottoman dependent status only took place in 1908. The Mediterranean provinces facing Eng- lish and French colonization, however, held on to it, because it secured their contacts with the Islamic world and some religious protection under Christian rule.

147 Eyalet-i mümtaze, that is, privileged vilayets; see Panaite, The Ottoman Law, 467–468.

What is Inside and What is Outside? Tributary States in Ottoman Politics

Dariusz Kołodziejczyk

A few years ago, a group of Crimean Tatar PhD students from Simferopol attended the East-Central European School in the Humanities held at the University of Warsaw. One of them announced that the aim of her future PhD thesis was to prove that the Crimean Khanate had been a sovereign state, independent from the Ottoman Empire. She had not started the research yet, but her final conclusion was already there. I was not her supervisor, so I could only express my reservations regarding a methodol- ogy in which conclusions preceded the actual research. At the same time, inwardly, I thought I was simply witnessing a natural process in which one scholarly abuse triggers an abuse in the opposite direction. If I were a Marxist, I could console myself that sooner or later such extreme views would result in a more cautious Hegelian synthesis. Over a century ago, when Vasilii Dmitrievich Smirnov published his seminal monograph on the history of the Crimean Khanate, he gave it the title The Crimean Khanate under the Suzerainty of the Ottoman Porte.1 The title was anything but innocent, as Smirnov was neither a dispassionate antiquarian nor a passive witness of historical events. He did great service for the Russian Ministry of Foreign Affairs and even catalogued Oriental manuscripts arriving as spoils in St. Petersburg after the Russian–Otto- man War of 1877–1878. In fact, Smirnov perfectly fit the Saidian notion of a European nineteenth-century Orientalist, though Edward Said did not include Russia in his critique of Western academia. By stressing the Crimean Khanate’s lack of sovereignty and its primitive economy, Smirnov consciously exposes its “immature” status and thus convinces the reader that the Russian annexation of the Crimea in the late eighteenth century was a blessing both to its inhabitants and to human civilization. It is no

1 Vasilii Smirnov, Krymskoe khanstvo pod verkhovenstvom Otomanskoi porty do nachala XVIII veka [The Crimean Khanate under the suzerainty of the Ottoman Porte until the beginning of the eighteenth century] (St. Petersburg, 1887); recently republished in Mos- cow in 2005. 422 dariusz kołodziejczyk wonder that the vision of the khanate as a parasite entity, totally dependent on the Porte, proposed by some—but certainly not all—Russian historians, provoked some Tatar historians to idealize its past and stress its sovereignty. The correspondence between the Polish and Ottoman courts on the aspect of the khan’s sovereignty demonstrates that the matter was equally confusing in the sixteenth century. In 1531, Süleyman the Magnificent notified the Polish court that the Poles did not need to negotiate separately with Sa‘adet Giray, since the khan was not a sovereign ruler, he owed his position to the sultan, and was obliged to obey him.2 The same question arose with the accession of a new khan, Sahib Giray. In January 1533, Süleyman issued an ‘ahdname confirming the first “permanent” treaty between the Ottoman Empire and Poland–Lithuania. Since the Pol- ish court pressed the Porte to include the khan in the treaty and oblige him to obey the peace, the sultan’s instrument stipulated that Sahib Giray should behave in a friendly manner toward King Sigismund, but the latter should withdraw his assistance toward the Crimean malcontents in return (a clear allusion to İslam Giray, Sahib’s nephew and contender for the throne, who enjoyed support at Sigismund’s court and among Polish– Lithuanian border commanders).3 The Polish king was still unsatisfied and asked İbrahim Pasha, Süley- man’s powerful grand vizier, to strongly order the khan to keep peace

2 “(. . .) calchularete che talle han non sia principe da per si ne in dyparte”; see Warsaw, Archiwum Główne Akt Dawnych [hereafter, AGAD], Archiwum Koronne Warszawskie [hereafter, AKW], Dział turecki, karton 66, teczka 29, no. 66; cf. Zygmunt Abrahamo wicz, Katalog dokumentów tureckich: Dokumenty do dziejów Polski i krajów ościennych w latach 1455–1672 [Catalogue of Turkish documents: Documents concerning the history of Poland and its neighboring countries in the years 1455–1672] (Warsaw, 1959), 41. The letter is preserved only in Italian and Latin translations; for the Latin, see Acta Tomiciana, vol. 13 (Poznań, 1915), 150–153. On the Polish embassy to Istanbul, headed by Jan Ocieski in 1531, see Andrzej Dziubiński, Stosunki dyplomatyczne polsko-tureckie w latach 1500–1572 w kontekście międzynarodowym [Polish–Turkish diplomatic contacts in the years 1500–1572 in an international context] (Wrocław, 2005), 82–85. 3 “Al prezente lo inperator di Tartaria, potentisimo, nominato Sachip Cheraichan, el qual è cresuto a la nostra felicisima Porta e ne le nostre inpreze ett operation suefato tanto, che li è nostro fiol, ett esendo tra noi l’amicitia, li avemo dito che ancor lui vi sia amicho, bisognia che voi eciam teniati con lui bona amicitia ett operar di sorte, che Tartari over parenti del dito inperator che non li siano obedienti ett scanpando da lui abino da voi recapito e cercando aiuto, non li ascoltareti anci schazarli del vostro paeze e perseguitarli.” The document, preserved only in Italian translation, is published in Dariusz Kołodziejczyk, Ottoman–Polish Diplomatic Relations (15th–18th Century): An Annotated Edition of ‘Ahd- names and Other Documents (Leiden, 2000), 230–231. An analogous clause was entered in the royal document, issued in Cracow, in May 1533; see ibid., 232–233. tributary states in ottoman politics 423 with Poland–Lithuania. İbrahim’s letter, written in July 1533, contained a Pythian answer. Admitting that Sahib Giray was like a son to Süleyman and owed his throne to the Porte, the vizier yet stressed that the khan was a sovereign ruler and it was not fitting to include him in the instruments exchanged between the sultan and the king.4 Nevertheless, he promised to write to the khan, expecting that the latter would maintain a friendship with Sigismund. Studying the aforementioned correspondence, covering just three years, one gets the impression that the Ottomans were true postmodernists, as they were unable to answer the simple question whether the Crimean khan was a sovereign ruler or not. My second example comes from Yemen. Conquered between 1538 and 1547, and again in 1569–1570, it remained under loose Ottoman control until the early seventeenth century, to be lost and annexed once more in the nineteenth century.5 Nevertheless, the maps enclosed in textbooks of Ottoman history teach us that—unlike Moldavia or Dubrovnik—Yemen was a regular Ottoman province and remained as such until World War I. In fact, while the control of lowland Tihama, inhabited by Sunni Shafi‘i Muslims, was relatively easy, in the mountainous north the Ottomans encountered the Zaydi imams, who claimed their own leadership over the people of Islam, and sometimes even the caliphate. To quote a modern French scholar, it was la rencontre de deux légitimités.6 A prominent scholar of Yemen, Robert Serjeant, made a capital remark regarding the troubles experienced in Yemen by all foreign invaders, beginning with the Abbasids and ending with Egyptian president Nasser in the 1960s: “one broad pattern easily discerned is the entry of foreign conquerors from the lowlands, their initial success but ultimate inability

4 “(. . .) lakin müstakil selatin ‘ıdadından olub başka memlekete vali ve hakimdır bunların ‘ahdnamelerinde anun ahvali bile zikr olunub mukayyed olması münasib görül- meyüb zikr olunmadı”; see AGAD, AKW, Dział turecki, karton 67, teczka 35, no. 78; cf. Abrahamowicz, Katalog dokumentów tureckich, 45–46; a contemporary Polish translation is published in Acta Tomiciana, vol. 15 (Wrocław and Cracow, 1957), 68–69, and Kazimierz Rymut, ed., Listy polskie XVI wieku, vol. 1 [Polish letters from the sixteenth century] (Cra- cow, 1998), 39–40. 5 Cf. Caesar E. Farah, The Sultan’s Yemen: Nineteenth-Century Challenges to Ottoman Rule (London, 2002). 6 François Blukacz, “Le Yémen sous l’autorité des imams zaidites au XVIIe siècle: une éphémère unité,” in Yémen, passé et présent de l’unité, ed. Michel Tuchscherer (Aix-en- Provence, 2004), 39–51, esp. 39. On the Zaydi claim to the caliphate, expressed by the imam’s titles, cf. the seventeenth-century report of the Yemeni ambassador to Ethiopia, al-Haymi, in Emeri Johannes van Donzel, ed., A Yemenite Embassy to Ethiopia 1647–1649 (Stuttgart, 1986), 84–85. 424 dariusz kołodziejczyk to conquer the northern highlands and their eventual retreat.”7 An early seventeenth-century British traveler claimed that the Ottomans kept 30,000 soldiers in Yemen fighting “continuously in the field against the Arab king in the mountains.”8 Ottoman Turkish as well as pro-Ottoman Arab chroniclers employed rich vocabulary to discredit and dehumanize the Zaydi enemies by referring to them and their troops as “bastards” (haramzadeler), “the enemies of religion” (a‘da’ al-din), “heretics” (ahl al-ilhad), “idle troops of the devilish army” (djaysh al-batil min al-djundi`l-shaytani), and—perhaps most typically—as “insects” (hasharat).9 Yet, sooner or later, subsequent Otto- man commanders learned that Yemen was unconquerable. Thus, another policy and language had to be adopted. Qutb al-Din al-Nahrawali, a pro- Ottoman Arab chronicler, related the negotiations that took place in 1570 following the seven-month siege of the Yemeni fortress of Kawkaban by the Ottoman army. Seeing the futility of his efforts, Sinan Pasha, the Otto- man commander-in-chief, entrusted his local Arabic adviser with writing a letter to Muhammad bin Shams al-Din, the Zaydi commander of the fortress. In the letter, the addressee was titled “my brother” (ya akhi) and was assured that the Ottoman sultan “was not in need of these lands” since he “ruled most of the inhabited world,” yet his will must not be defied. Therefore, assured the author on behalf of the Ottoman commander, the pasha “will not reject your request but will [only] stipulate that the sermon and currency (al-khutba wa al-sikka) be in the name of the sultan10 [. . .] and [after you fulfill these conditions] he will issue authority

7 Robert B. Serjeant and Ronald Lewcock, eds., San‘ā’: An Arabian Islamic City (Lon- don, 1983), 77. 8 Ibid., 109 (in the chapter on the “Western Accounts of San‘ā’ 1510–1962” by R.L. Bidwell); the passage refers to Joseph Salbank from the East India Company. 9 The epithet haramzadeler was constantly used by Rumuzi, a sixteenth-century Ottoman chronicler; see Jane Hathaway, A Tale of Two Factions: Myth, Memory, and Iden- tity in Ottoman Egypt and Yemen (Albany, NY, 2003), 64–65. All the remaining epithets can be found in the Arabic chronicle of Rumuzi’s contemporary, Qutb al-Din Muham- mad al-Nahrawali al-Makki, a pro-Ottoman Muslim scholar from Hejaz; for the English translation, see Qutb al-Dīn al-Nahrawālī al-Makkī, Lightning over Yemen: A History of the Ottoman Campaign (1569–71), trans. Clive K. Smith (London and New York, 2002), 41, 59, 61 (Smith’s translation reads “devilish troops of the false army”) and 126; for the Arabic text, see Qutb al-Din Muhammad bin Ahmad al-Nahrawali al-Makki, al-Barq al-yamani fi’l-fathi’l-‘uthmani, ed. Hamd al-Jasir (Riyadh, 1387 AH/1967 CE), 248, 274, 277, and 372; on the term hasharat, cf. also Farah, The Sultan’s Yemen, 31. 10 Pronouncing the ruler’s name during the Muslim Friday prayer (Ar. khutba; Tur. hutbe) and engraving his name on the coins (Ar. sikka; Tur. sikke) were the traditional attributes of sovereignty in the Muslim world. By consenting to the Ottoman request, Muham- mad bin Shams al-Din would formally acknowledge the authority of the Ottoman sultan. tributary states in ottoman politics 425 for you to have the standard your father had in the past.”11 As a result of the ensuing negotiations, Muhammad bin Shams al-Din was granted the attributes of power: a standard and an imperial diploma (berat), according to whose tenor he had admitted his error and prostrated before the Ottoman might, whereas Sinan Pasha “made him subject to pacts and agreements and imposed on him conditions and limits.”12 In short, he was granted his own lands by the Ottoman sultan on the sole condition that he would formally accept Ottoman suzerainty. The above narration is quite striking in its resemblance to official Otto- man reports on peace negotiations with Christian European rulers. And Muhammad bin Shams al-Din was not even a Zaydi imam! His uncle, Imam al-Mutahhar, repeatedly challenged Ottoman sovereignty and negotiated several peace treaties with the Porte, first in 1552,13 and then—after his rebellion—in 1570.14 The peace in Yemen did not last long and soon another Zaydi pretender, Imam al-Qasim, raised the standard of rebellion. He concluded peace treaties with successive Ottoman pashas in 1602, 1608, and 1619.15 His territories first encompassed only the mountain fortress of Shahara, but then also Sa‘ada and al-Hayma. His son Imam al-Mu‘ayyad continued the struggle against the Ottomans, and the last Ottoman garrisons evacuated Yemen by 1635, to return only in the nineteenth century. Paradoxically, as a modern expert remarks, “at no time did [the Ottomans] abdicate their sovereign rights over the land, for Yemenis still looked to them and their agents for protection” against foreign—especially European—danger. “Ironically,” concludes Caesar Farah, “the imam sought autonomy only.”16 And Yemen was not alone. Baki Tezcan quotes a document of Sultan Süleyman from around 1536, granting hereditary rule, land property rights, tax exemptions, and wide autonomy to the Kurdish emirs in return for

11 Lightning over Yemen, 161; al-Barq al-yamani, 419. 12 “(. . .) akhadha ‘alayhi al-muwathiq wa’l-‘uhud wa ashtarata ‘alayhi al-shurut wa hadda ‘alayhi al-hudud”; see Lightning over Yemen, 165–167; al-Barq al-yamani, 423–426. 13 John R. Blackburn, “The Documents on the Division of Ottoman Yemen into two Beglerbegiliks (973/1565),” Turcica: Revue d’études turques 27 (1995): 223–236, esp. 230. 14 Lightning over Yemen, 170–171; al-Barq al-yamani, 430–431. 15 Frédérique Soudan, Le Yémen ottoman d’après la chronique d’al-Mawza‘ī (Cairo, 1999), 270; Yahya bin al-Husayn, the seventeenth-century Yemeni chronicler and grandson of Imam al-Qasim, referred to all these treaties as al-sulh, a term equally fitting the “domestic” or “international” peace; see idem, Gayat al-amani fi akhbari’l-qutri’l-yamani [The ultimate desire to acquire the knowledge of the country of Yemen] (Cairo, 1388 AH/1968 CE), pt. 2, 783, 794, and 811. 16 Farah, The Sultan’s Yemen, 1 and 272. 426 dariusz kołodziejczyk their loyalty to the Ottomans. The document is referred to as a mu‛ahede (a term identical with an ‘ahdname, granted also to foreign rulers) and is corroborated by the sultan’s solemn oath (yemin . . . iderim ki).17 Some of the Kurdish territories, which retained their autonomy until the nineteenth century, were situated deep inside the Ottoman domains, hence they could not even be labeled as frontier or borderlands. There were many more local Muslim rulers whose position was analogous, to mention only the sharifs of Mekka, Daghestani shamkhals, or even some provincial notables, like the Lebanese Ma‘ns who enjoyed local autonomy long before the era of the ayans. These minor rulers were often accused of duplicity as the Ottomans never accepted their full sovereignty.18 The term “mountain bandits” (eşkiya-i cebeliyye), by which Ottoman authors referred to the Christian Montenegrins—known as the notorious “rebels” (or “freedom fighters”)—could equally apply to the Yemeni Muslim Arabs who inhabited the opposite corner of the empire.19 At first glance, Poland and Yemen do not seem to have much in common. And yet, in the seventeenth century, their history was interrelated. Hacı ‘Ali, a seventeenth-century Ottoman chronicler of the last decades of Ottoman rule in Yemen, describes how, in 1622, the Ottoman gover- nor Fazli Pasha waited desperately for reinforcements from Istanbul. The chronicler explains that no help could arrive due to the janissary rebellion and the murder of Osman II, both resulting from the sultan’s ill-fated campaign against Poland, effected in 1621 (the so-called Hotin seferi).20 When assistance finally arrived, it was probably too late; the Ottomans had

17 Baki Tezcan, “The Development of the Use of ‘Kurdistan’ as a Geographical Descrip- tion and the Incorporation of this Region into the Ottoman Empire in the 16th Century,” in The Great Ottoman-Turkish Civilisation, ed. Kemal Çiçek (Ankara, 2000), vol. 3, 540–553, esp. 547 and 553. 18 Rudi Matthee invokes the opinion about “the Kurdish duplicity” which was shared by both the Safavid and Ottoman chroniclers; see idem, “The Safavid-Ottoman Frontier: Iraq-i Arab as Seen by the Safavids,” in Ottoman Borderlands: Issues, Personalities and Polit- ical Changes, ed. Kemal Karpat and Robert Zens (Madison, WI, 2003), 157–173, esp. 168– 169. Analogously, also the Tatars, Moldavians, and Ukrainian Cossacks who strove toward independence or greater autonomy from the neighboring powers centered in Istanbul, Warsaw, and Moscow, were typically labeled as “treacherous” by Ottoman, Polish, and Russian authors. 19 For the term eşkiya-i cebeliyye, almost synonymous with the Montenegrins in the Ottoman bureaucratic jargon, see Maurus Reinkowski, “Double Struggle, No Income: Otto- man Borderlands in Northern Albania,” in Ottoman Borderlands, 239–253, esp. 242n10. 20 All these events are recorded on the same page of the chronicle; see Haji ‘Ali, “Ahbarü’l-yamani” [Information about Yemen], Istanbul, Süleymaniye Library, ms. Hamidiye 886, fol. 205a; cf. Hathaway, A Tale of Two Factions, 86–87 and 221n50. On the importance of the Polish campaign of 1621 for Ottoman domestic politics, see Baki Tezcan, tributary states in ottoman politics 427 pulled out of Yemen by 1635. In 1672, the Porte contemplated a reconquest of the province.21 Again, these plans were frustrated by a more urgent campaign against Poland. The latter resulted in the Ottoman annexation of the southeastern Polish provinces (Podolia and the Cossack Ukraine which gained autonomy under Ottoman patronage), while the Polish king consented to pay a tribute and was formally reduced (albeit temporarily) to the status of an Ottoman tributary. There was another element that linked Yemen and Poland: their rulers concluded peace treaties with the Porte and obtained ‘ahdnames from the Ottoman sultans. One is tempted to ask, somewhat provocatively: what makes a modern historian resolve that Yemen lay inside the Ottoman Empire, while Poland, Venice, and a number of other countries, lay outside? If judged according to the intentions of Ottoman propaganda, rulers such as the Yemeni imam, the Venetian doge, the Habsburg emperor (referred to as Beç kıralı, i.e., the “king of Vienna”), the Polish king, and— last but not least—the Russian tsar were all Ottoman vassals, at least for a time.22 In reality, all these countries preserved sovereignty although none of them could entirely ignore Ottoman diplomatic pressure or open military threats.23 Ottoman influence was at best disputable, often negligible,

“Khotin 1621, or how the Poles Changed the Course of Ottoman History,” Acta Orientalia Academiae Scientiarum Hungaricae 62 (2009): 185–198. 21 Hathaway, A Tale of Two Factions, 91. 22 Even though the Polish king engaged to pay a tribute only once (in 1672; in fact, this tribute was not paid), the Porte could always argue that the Polish king was a tributary of the Crimean khan who, in turn, was a vassal of the sultan; on this issue, see my article “La Res Publica polono-lituanienne était-elle le vassal de l’Empire ottoman?” in Studies in Oriental Art and Culture in Honour of Professor Tadeusz Majda, ed. Anna Parzymies (War- saw, 2006), 125–136. Likewise, the tsars also sent annual “gifts” to the khans until 1685 (the custom was formally abolished in 1700). Interestingly, the alleged tributary status of the tsar compromised Moscow’s negotiating position in its early encounters with Beijing. A. Ruslanov, a baptized Crimean Tatar turned Russian diplomat in the service of the Yakutsk governor, deserted to China and spent over two decades in the Manchu dynasty’s service. By presenting the Russian tsar as a Crimean tributary, Ruslanov weakened Russia’s prestige and position in the Russian–Chinese negotiations that took place in the 1670s; see Vladimir Miasnikov, The Ch’ing Empire and the Russian State in the 17th Century (Moscow, 1985), 86 and 159–160. 23 For instance, in both Venice and Poland the idea of a crusade was quite unpopular and commonly dismissed as suicidal. Infrequent cases when these two states entered the Holy Leagues can be explained by earlier Ottoman invasions (e.g., of Cyprus in 1570), which left Venetians and Poles little choice but to join the Habsburgs, with whom they were usually at odds. Ottoman political pressure also played a role in elections to the Polish throne in 1573, 1575, and 1587, all won by anti-Habsburg candidates (Henry Valois, Stephan Báthory, and Sigismund Vasa, respectively). Yet, the catchy title “Lehistan’da türk hakimiyeti” [Turkish rule in Poland], given by Ahmed Refik to his classical article devoted to this subject, does not do justice to the compact political situation and entirely ignores 428 dariusz kołodziejczyk but at times the rulers concerned consciously invoked the sultan’s authority in their disputes with other courts. While the Yemeni imams could use the Ottoman “umbrella” against the Portuguese, Dutch, or English invaders, the Polish kings found it equally useful in their disputes with Vienna, Stockholm, and St. Petersburg in the sixteenth, seventeenth, and eighteenth centuries respectively. By analogy, one could ask why modern historical maps depict the Kurd- ish principalities, such as Bitlis, as “regular” Ottoman lands, while the Crimean Khanate, Moldavia or Cossack Ukraine (in the years 1672–1699)24 are depicted by stripes which refer to their tributary status. In the seventeenth century, the rulers of all the aforementioned countries received Ottoman berats, which embodied their dependence on Istanbul but also granted them substantial autonomy. It seems that the more we study the Ottoman realities, the more “stripes” we can see. Like any other early modern ruler, the Ottoman sultan faced constant challenges to his legitimacy with regard to his expected role as distributor of justice, provider of security, and Muslim gazi. Negotiation with his own subjects, not just external monarchs, was part of the sultan’s métier. In fact, the difference between the “domestic” and “external” realms was often blurred.25 domestic factors that influenced the results of the Polish elections; cf. idem, “Lehistan’da türk hakimiyeti,” Türk Tarih Encümeni Mecmu‘ası 14 (1340 AH/1924 CE): 227–243; for more balanced views of Polish historians who admit the importance of Ottoman pressure in the Polish elections but do not treat it as the most decisive factor, see Wojciech Hensel, “Uwagi o stosunkach polsko-tureckich w XVI wieku do panowania Stefana Batorego” [Remarks on the Polish–Turkish relations in the sixteenth century until the reign of Stephan Báthory], in Stosunki polsko-tureckie: Materiały z sesji naukowej zorganizowanej przez Instytut Orien talistyczny i Towarzystwo Polska Turcja w 1988 roku, ed. Tadeusz Majda (Warsaw, 1995), 19–29, esp. 23–26; Dziubiński, Stosunki dyplomatyczne, 261–262, 269, 273; and my review of the latter book in Kwartalnik Historyczny 114, no. 4 (2007): 101–110, esp. 109. 24 As early as 1669, Petro Doroshenko received attributes of power, including a standard and an imperial diploma (berat), from Sultan Mehmed IV. According to the imperial diploma, the status of the Cossack hetman was to be equal to the status of the rulers of Wallachia and Moldavia, although the hetman was not obliged to pay tribute; see Dariusz Kołodziejczyk, ‘ “Tertium non datur? Turets’ka al’ternatyva v zovnishnii politytsi Kozats’koii derzhavy” [Tertium non datur? Turkish alternative in the foreign policy of the Cossack state], in Hadjats’ka unija 1658 roku, ed. Pavlo S. Sochan’ (Kiev, 2008), 67–80, esp. 72–74. 25 By analogy, Polish kings used to send envoys to foreign courts, such as Vienna, Paris, and Moscow, but also to Königsberg and Mitau, the capitals of the duchies of Prussia and Courland, whose rulers were Polish tributaries, and even to Danzig and Riga, the royal cities that acknowledged the royal suzerainty and were de iure situated within Poland– Lithuania, but nevertheless enjoyed large autonomy; cf. Historia dyplomacji polskiej [His- tory of the Polish diplomacy], vol. 2, 1572–1795, ed. Zbigniew Wójcik (Warsaw, 1982), 88–89 and 129. Moreover, the kings often negotiated and exchanged envoys with the hetmans of the Ukrainian Cossacks, the leaders of formalized noble rebellions (called rokosz or kon- federacja, i.e., confederation, and claimed to be legal by their participants, who depicted tributary states in ottoman politics 429

To further emphasize my point, I propose to compare the status of several political entities vis-à-vis the Porte, measured merely by such formal criteria as the obligation: (a) to pay tribute, (b) to provide military assistance, and (c) to pronounce the sultan’s name during the Friday prayer (Tur. hutbe; Ar. khutba):

Table 1. Ottoman “vassals” and their obligations rulers of tribute military hutbe assistance Moldavia x x — Wallachia x x — Transylvania x x — Georgian kingdoms and principalities x x — Dubrovnik x — — Venice x — — Austria (or perhaps the whole Habsburg x26 — — Empire?) Poland-Lithuania x27 — — Muscovy/Russia x28 — — Cossack Ukraine29 — x — Crimean Khanate — x x Bitlis and other Kurdish beyliks — x x some Arab tribes — x x Hejaz — — x

themselves as the defenders of the Commonwealth’s constitution), and even with the representatives of rioting soldiers who collectively negotiated the payment of overdue salaries. 26 The tribute, formally—at least in Habsburg eyes—paid from Royal Hungary, was discontinued after 1606, although, contrary to long-lasting belief, it was not formally abolished by the Treaty of Zsitvatorok; cf. Gustav Bayerle, “The Compromise of Zsitvatorok,” Archivum Ottomanicum 6 (1980): 5–53, esp. 27. 27 Stipulated by the Treaty of Buczacz (1672), but not paid. Nonetheless, according to the Polish–Ottoman treaties, the kings were obliged to send annual “gifts” to the Crimean khans, who were Ottoman vassals. This obligation was formally abolished by the Treaty of Karlowitz in 1699 (in fact the practice was discontinued already in 1683). 28 Like the Polish kings, the tsars were obliged to send annual “gifts” to the Crimean khans, who were Ottoman vassals. This obligation was formally abolished in 1700 (in fact the practice was discontinued already in 1686). 29 Formally—in Ottoman eyes—in the years 1669–1699 and 1711–1714; on the first period cf. note 24 above; on the second period, see Omeljan Pritsak, “Odyn chy dva doho- vory Pylypa Orlyka z Turechchynoiu na pochatku druhoho desiatylittia visimnadtsiatoho stolittia?” [One or two treaties of Pylyp Orlyk with Turkey at the beginning of the second decade of the eighteenth century?], Ukraiins’kyi Arkheohrafichnyi Shchorichnyk n.s. 1 (1992): 307–320. In both cases, actual Ottoman control over Cossack Ukraine was much shorter. The Cossack hetman was the sultan’s only Christian vassal who was not obliged (at least for the time being) to pay tribute, either directly to the Porte or to the Crimean khan. 430 dariusz kołodziejczyk

Table 1 (cont.) rulers of tribute military hutbe assistance northern Yemen — — x Aceh (northern Sumatra) and Maldives30 — — x Kashgar31 — — x etc.

The composition of the above table might cause the reader to doubt its author’s sanity. Yet, by drawing this seemingly absurd list I want to draw the reader’s attention to the viability of such “objective” criteria by which modern historians like to depict and classify early modern political organizations. The fetishization of state sovereignty, still apparent in modern historical writing (especially, but not exclusively, in regard to post-Westphalian Europe), is hardly useful if one aims to describe the more nuanced political mosaic that was typical for the early modern world and which did not disappear in 1648.32 In the seventeenth century, Russia extended to the Pacific Ocean, but continued to send tribute to the Crimean khan, who was in turn an Ottoman vassal. Even though common sense refuses to regard Russia as well as Venice, the Habsburg Empire, or Poland–Lithuania as Ottoman vassals, a historian must deal with the fact that their consent to send tribute to their Muslim neighbors did affect their sovereignty—at least de iure. By comparing Moldavia with Poland on the one hand and with Yemen on the other, I would argue that the status of a so-called Ottoman tributary state was less unique and unusual than is often argued. Judging by the

30 See Giancarlo Casale, The Ottoman Age of Exploration (Oxford, 2010), 124–130 and 149 (the map entitled: “Sokollu Mehmed Pasha’s Soft Empire 1565–1579”). 31 In the 1870s, Yakub Beg, the ruler of Kashgar, sent two successive embassies to Istan- bul and agreed to pronounce the name of Sultan Abdülaziz in the hutbe in the hope of receiving Ottoman assistance against China; see Roderic H. Davison, Reform in the Otto- man Empire 1856–1876 (New York, 1973), 273; Mehmet Alpargu, “Khanates of Turkestan,” in The Turks, vol. 2, Middle Ages, ed. Hasan Celâl Güzel, C. Cem Oğuz, and Osman Karatay (Ankara, 2002), 899–922, esp. 914–915 and 921. 32 To quote Andreas Osiander: “Westphalia—shorthand for a narrative purportedly about the seventeenth century—is really a product of the nineteenth- and twentieth- century fixation on the concept of sovereignty”; see idem, “Sovereignty, International Rela- tions, and the Westphalian Myth,” International Organization 55 (2001): 251–287, esp. 251; for another revisionist approach toward the perception of the year 1648 as a turning point in the history of international relations, see Benno Teschke, The Myth of 1648: Class, Geo- politics and the Making of Modern International Relations (London and New York, 2003). tributary states in ottoman politics 431 nature of their relations vis-à-vis the Porte, these three countries would all roughly fit in the vague category “between annexation and mere alliance,” described as typical for imperial structures by Garry Runciman.33 Instead of asking whether such political entities as Moldavia, Dubrovnik, the Cos- sack Ukraine, the Crimean Khanate, Yemen, or Bitlis were sovereign or not, it seems more reasonable to discuss the degree of their sovereignty in a given sphere and in a given period by examining not merely formal, legal criteria, but also such aspects as their political and military ability to defy the sultan’s will, the strength of their economic links with the imperial center, and, last but not least, their participation in a shared imperial culture.34 In fact, there was nothing abnormal in the limited sovereignty enjoyed by the Ottoman tributary states. Both in the early modern era and in the twentieth century, a large number of Europeans, not to mention the inhabitants of other continents, lived in political entities that were not fully sovereign or not sovereign at all. Also today, unless one is a citizen of the United States, or perhaps China or North Korea, one cannot claim that their government is fully sovereign (and perhaps it is for the better).35

33 W.G. Runciman, “Empire as a Topic in Comparative Sociology,” in Tributary Empires in Global History, ed. Peter Fibiger Bang and C.A. Bayly (Basingstoke, 2011), 99–107, esp. 99. 34 Ottoman court ceremonial, architecture, fashion, music, and cuisine were adopted by provincial elites, Muslim as well as non-Muslim; for the impact of the Ottoman ceremonial blueprint at the seventeenth-century Moldavian court in Iaşi, see Michał Wasiu- cionek, “Diplomacy, Power and Ceremonial Entry: Polish-Lithuanian Grand Embassies in Moldavia in the Seventeenth Century,” Acta Poloniae Historica 105 (2012): 55–83, esp. 81. Attraction to Ottoman culture is also visible in the writings of Demetrius Cantemir (1673–1723), a future prince of Moldavia educated in Istanbul, notwithstanding the fact that he ended life as a refugee in Russia; cf. the subchapter “Ottoman or Not? An Educated Non-Muslim,” devoted to Cantemir in Suraiya Faroqui, Subjects of the Sultan: Culture and Daily Life in the Ottoman Empire (London and New York, 2005), 81–85. Numerous recent studies within the so-called new cultural history trend aim to rehabilitate political culture and ceremonial as not merely void ornaments but rather the factors that have continued to influence “real” politics well into the modern era; in the context of the Habsburg Holy Roman Empire, see Barbara Stollberg-Rilinger, Des Kaisers alte Kleider: Verfassungsge- schichte und Symbolsprache des Alten Reiches (Munich, 2008); in the context of the British Empire, see David Cannadine, Ornamentalism: How the British Saw Their Empire (Oxford and New York, 2001); in the context of Byzantium, the Arab Caliphate, and their successor states in the Orthodox and Islamic worlds, see Garth Fowden, Empire to Commonwealth: Consequences of Monotheism in Late Antiquity (Princeton, NJ, 1993). For more general thoughts, see Barbara Stollberg-Rilinger, “Symbolische Kommunikation in der Vormo derne: Begriffe—Thesen—Forschungsperspektiven,” Zeitschrift für historische Forschung 31 (2004): 489–527. 35 According to Osiander: “the global system today in certain respects bears more resemblance to the type of system exemplified by the Holy Roman Empire than to the 432 dariusz kołodziejczyk

Being much more heterogeneous than it has been admitted in earlier historiography, the Ottoman Empire owed its longevity to its composite character and its functioning was conditioned by the cooperation of provincial elites—Muslim and Christian alike.

so-called Westphalian model. There is a clear de facto trend in international politics away from classical sovereignty and toward something closer to Landeshoheit, territorial jurisdiction under an external legal regime shared by the actors. Like the estates of the empire, modern states are also tied into a complex structure of governance that creates a network both of cooperation and of mutual restraint”; see Osiander, “Sovereignty, International Relations, and the Westphalian Myth,” 283. Parallels between the Holy Roman Empire and the contemporary European Union, already suggested by Osiander, have been further explored by Peter Haldén, who looks for analogies between the medieval/early modern universal order(s) and the globalized, interdependent world system of the early twenty- first century; see his Stability without Statehood: Lessons from Europe’s History before the Sovereign State (Basingstoke, 2011). Notes on Contributors

Ovidiu Cristea (PhD in Medieval Studies, 2003, at the Institute of History “Nicolae Iorga,” Bucharest) has been researcher at the Institute of History “Nicolae Iorga” of the Romanian Academy since 1993. He is the author of the monographs Acest Domn de la Miazănoapte: Ştefan cel Mare în documente inedite veneţiene [A lord from the distant north: Stephen the Great of Moldavia in Venetian documents] and Veneţia şi Marea Neagră în secolele XIII–XIV [Venice and the Black Sea thirteenth to fourteenth centuries], as well as a number of studies concerning the later Crusades, the “Long Turkish War” (1593–1606), and the medieval history of Walla- chia and Moldavia.

Mária Ivanics (DSc in History, 2009), is head of the Department of Altaic Studies and of the Turcology Research Group of the Hungarian Academy of Sciences at the University of Szeged. She is the author of the monograph A Krími Kánság a tizenötéves háborúban (1591–1606) [The Crimean khanate in the long Turkish war] (1994) and publisher of a seventeenth-century Tatar chronicle Daftar-i Cingis-name / Das Buch der Dschingis-Legende (2002). She has also published a variety of studies and source editions, concerning the history of the Golden Horde and its successor states and also their relationship to the Ottoman Empire.

Gábor Kármán, PhD in Early Modern Hungarian History (2009, Eötvös Loránd University, Budapest) and in the Comparative History of Central, Southeastern and Eastern Europe (2010, Central European University, Budapest) is currently a research fellow at the Geisteswissenschaftliches Zentrum Geschichte und Kultur Ostmitteleuropas an der Universität Leipzig. His main research topics focus on the foreign policy of the Prin- cipality of Transylvania, with special emphasis on related questions from social and cultural history. He is the author of Erdélyi külpolitika a veszt fáliai béke után [Transylvanian foreign policy after the Peace of West- phalia] (2011), and another monograph, the biography of Jakab Harsányi Nagy, a seventeenth-century diplomat and intellectual from Transylvania, is forthcoming. 434 notes on contributors

Dariusz Kołodziejczyk (PhD in history at the University of Warsaw, 2000), is professor of early modern history at the University of Warsaw and at the Polish Academy of Sciences and currently vice president of the Comité International des Études Pre-ottomanes et Ottomanes. He has been visiting professor at the University of Notre Dame (2004), Hokkaido University (2009), and Collège de France (2011). His publications include Ottoman– Polish Diplomatic Relations (15th–18th Century): An Annotated Edition of ‘Ahdnames and Other Documents (Brill, 2000), The Ottoman Survey Reg- ister of Podolia (ca. 1681): Defter-i Mufassal-i Eyalet-i Kamaniçe (Harvard University Press, 2004), The Crimean Khanate and Poland-Lithuania: International Diplomacy on the European Periphery (15th–18th Century): A Study of Peace Treaties Followed by Annotated Documents (Brill, 2011), and Universal Empire: A Comparative Approach to Imperial Culture and Representation in Eurasian History (Cambridge University Press, 2012; co-edited with Peter Fibiger Bang).

Natalia Królikowska (PhD in history, 2010, at the University of Warsaw) is currently assistant professor at the University of Warsaw. Her research focuses on the Crimean, Ottoman and Caucasian history in the early modern period. Her publications include “Crimean Crime Stories: Cases of Homicide and Bodily Harm during the Reign of Murad Giray (1678–1683)” published in The Crimean Khanate between East and West (15th–18th Cen- tury), edited by Denise Klein (Wiesbaden, 2012) and “The Law Factor in Ottoman-Crimean Tatar Relations in Early Modern Period” published in Law and Empire: Ideas, Practices, Actors, edited by Jeroen Duindam, et alia (Leiden and Boston: Brill, 2013).

Lovro Kunčević (PhD in Medieval Studies, 2012, from Central European University in Budapest) is presently higher assistant at the Institute for Historical Sciences of the Croatian Academy of Sciences and Arts (HAZU) in Ragusa (Dubrovnik). He has published on republican ideology, diplomatic rhetoric, image of the other, and political institutions of medieval and early modern Ragusa. He is currently preparing a book on the discourses on collective identity in Renaissance Ragusa.

Domagoj Madunić (PhD in history, 2012, at the Central European Univer- sity, Budapest), is an associate member of the “Croatian Eastern Adriatic Space and the Republic of Venice in Early Modern Period” project at the Croatian Institute for History (ISP) in Zagreb. He is also visiting lecturer at Zagreb University and Central European University. His articles focus notes on contributors 435 on the administrative and institutional context of the Venetian defensive system in the Adriatic during the War for Crete, as well as early modern demography and problems of Croatian proto-national discourse.

Vesna Miović (PhD in historical studies in 1996 at the University of Zagreb) is currently scientific adviser at the Institute for Historical Sciences of the Croatian Academy of Sciences and Arts (HAZU) in Dubrovnik and associate professor at the University of Zagreb and University of Dubrovnik. She is the author of the monographs Dubrovačka diplomacija u Istanbulu [Ragusan diplomacy in Istanbul] (2003), Dubrovačka Republika u spisima osmanskih sultana [Dubrovnik Republic in the documents of the Otto- man sultans] (2005), The Jewish Ghetto in the Dubrovnik Republic 1546–1808 (2005), as well as four other books and numerous papers related to the Ottoman-Ragusan relationship and the history of Ragusan Jews during the sixteenth, seventeenth, and eighteenth centuries.

Teréz Oborni, PhD, is senior fellow at the Institute of History of the Research Centre for the Humanities of the Hungarian Academy of Sci- ences and associate professor at Eötvös Loránd University (Budapest). She earned her PhD from the Hungarian Academy of Sciences (1998) and habilitation at Eötvös Loránd University (2012). Her research focuses on the political, economic, and constitutional history of the Principality of Transylvania. Apart from her dissertation, published as Erdély pénzügyei I. Ferdinánd uralma alatt 1552–1556 [The financial administration of Transyl- vania under the Rule of Ferdinand I (1552–1556)] (2002), she is also author and co-author of four books and numerous articles.

Victor Ostapchuk (PhD, Inner Asian and Altaic Studies, Harvard Univer- sity, 1989) is associate professor, Department of Near and Middle Eastern Civilizations, University of Toronto. He is editor-in-chief of the Harvard Ukrainian Research Institute’s series “Studies in Ottoman Documents pertaining to Ukraine and the Black Sea Countries,” and co-director of the Akkerman Fortress Project. He has published on history and historical archaeology of the Ottoman Black Sea. Recent publications include “The Ottoman Northern Black Sea Frontier at Akkerman: The View from a Historical and Archaeological Project” in The Frontiers of the Ottoman World, edited by Andrew Peacock (Oxford and New York: British Institute at Ankara and Oxford University Press, 2009); “Long-Range Campaigns of the Crimean Khanate in the Mid-Sixteenth Century” in Warfare in Eastern Europe, 1550–1800, edited by Brian Davies (Leiden and Boston: Brill, 2011); 436 notes on contributors

“Political-Personal Intrigue on the Ottoman Frontier in Hetman Bohdan Khmelnytsky’s Relations with the Porte: The Case of Ramażān Beg vs. Velī Beg,” Journal of Ukrainian Studies 33–34 (2008–2009) [2012].

Viorel Panaite (PhD in History (Turkology) at the University of Bucharest 1995, habilitation at the University of Bucharest, 2007) is currently professor in Islamic-Ottoman history at the University of Bucharest and senior researcher at the Institute of South-East European Studies of the Roma- nian Academy. He is the author of the monographs The Ottoman Law of War and Peace: The Ottoman Empire and Tribute Payers (2000), Diplomaţie occidentală, comerţ şi drept otoman. Secolele XV–XVII [Western diplomacy, commerce and Ottoman law: Fifteenth to seventeenth century] (2004), and Război, pace şi comerţ în Islam. Ţările române şi dreptul otoman al popoarelor [War, peace and trade in Islam: The Romanian principalities and the Ottoman law of nations], in a revised and augmented second edition (2013), as well as numerous studies including the political and legal relationship between the Ottoman Empire and its tributaries, and the status of Western merchants in the sixteenth and seventeenth-century Ottoman Mediterranean.

Sándor Papp (PhD in Oriental studies 1998 at the University of Vienna, habilitation at the University of Szeged, 2011) is currently associate professor in history at the University of Szeged and recurring visiting professor at the Károli Gáspár University of the Reformed Church (Budapest). He is the author of the monograph Die Verleihungs-, Bekräftigungs- und Ver- tragsurkunden der Osmanen für Ungarn und Siebenbürgen: Eine quellenkritische Untersuchung (2003) as well as numerous studies and source editions on early modern Transylvanian foreign policy, and the legal relationship between the Ottoman Empire and its tributaries.

Radu G. Păun (PhD in History at École des Hautes Études en Sciences Sociales, Paris) is currently chargé de recherche at the Centre National de la Recherche Scientifique in Paris, as a member of the Centre d’Études des Mondes Russe, Caucasien et Centre-Européen. His fields of research cover the history of the political elites in the Greco-Romanian world in the sixteenth to eighteenth centuries, political theology and power representations in the post-Byzantine era, and the relations between the Otto- man Empire and the tributary principalities of Wallachia and Moldavia. He is the editor of the volume The Biblical Models of Power and Law / Les modèles bibliques du pouvoir et du droit (Peter Lang 2008, co-edited with Ivan Biliarsky). notes on contributors 437

János B. Szabó is currently project manager at the Budapest History Museum. He has published widely on questions of medieval and early modern Hungarian and Transylvanian military history, as well as various aspects of the relationship between Transylvania and the Ottoman Empire. His latest monographs include A mohácsi csata (2006) [The battle of Mohács, also in French with Ferenc Tóth as co-author, as Mohács, 1526: Soliman le Magnifique prend pied en Europe centrale (2009)], and A honfoglalóktól a huszárokig: A középkori magyar könnyűlovasságról [From Magyar conquerors to the hussars: On medieval light cavalry in Hungary] (2010).

personal names

Abaza Hasan Pasha 323 Arslanay Bey 289 ‛Abbâs the Great, shah of Persia 414 as-Sarakhsi 30 Abeatar, Daniel 192 ash-Shaybanî, Muhammad 23–24, 30 Abeatar, Solomon 192 Axtelmeyer, Stanislaw Reinhard 244 ‛Abdulmecid, Ottoman sultan 415 Abdullah, son of Rizvan Pasha 47–48, 61 Baba Bey 281n Abdulgaffar el-Kırımi 48 Bakšić, Peter 248n Abdülhamid I, Ottoman sultan 403 Balassa, János 330n Abu Hanifa 29 Ballarino, Giovanni Battista 223–225 Abdurrahmân, şeyh of Yemen 387 Baltacı Mehmed Pasha 44, 219 Adil Giray, Crimean kalga 284 Bánffy, Boldizsár 268 Agapios, metropolite of Eski Yogan 415 Barca, Luka 196 Ágoston, Gábor 306 Barcsai, Ákos, prince of Ahmed III, Ottoman sultan 219 Transylvania 304n, 309, 409–410 Ahmed Pasha, Gedik, see Gedik Ahmed Barkey, Karen 212 Pasha Barnovschi, Miron, voievod of Ahmed Pasha, Köprülü, see Köprülü Moldavia 272 Ahmed Pasha Barta, Gábor 305n Ahmed Giray 286 Bartl, Peter 215 Ahmed Reşid 33n Barton, Edward 229–230, 233, 241n al-Maverdi 29 Basarab Ţepeluş, voievod of al-Mutahhar, Imam of Yemen 425 Wallachia 266 al-Mu‘ayyad, Imam of Yemen 425 Basta, Giorgio 79, 210 Alexander Aldea, voievod of Báthory, András, prince of Wallachia 258, 265, 270 Transylvania 79, 313 Alexander Movilă, see Movilă, Alexander Báthory, Boldizsár 79, 229 Alexandru Iliaş, see Iliaş, Alexandru Báthory, Gábor, prince of Alexandru Beldiman, see Beldiman, Transylvania 81–84, 330–331 Alexandru Báthory, István, king of Poland and prince Alexandru Mircea, voievod of of Transylvania 71, 76–77, 83, 228, 266, Wallachia 248n 307–310, 312, 316–317, 319, 330, 400, 413 Alexei Mikhailovich, Russian Tsar 50, Báthory, István, Jr., nephew of István, king 242 of Poland 317n Ali Efendi Akkermanlı 27 Báthory, István, voievod of Transylvania Ali Pasha, Grand Vizier in 1866 20n (fifteenth century) 266 al-Qasim, Imam of Yemen 425 Báthory, Kristóf, voievod of Althan, Adolf count of 221n Transylvania 84n, 312 Anastasios Mihail of Naousa 244 Báthory, Zsigmond, prince of Andrew of Caesarea 241 Transylvania 78–79, 229–230, 233, 289, Angiollelo, Giovanni Maria 262, 266 298, 313, 319, 321, 326, 337 Apafi, Mihály, I, prince of Battitore, Gabriel 192 Transylvania 72, 181n, 292, 308–309, Bayezid I, Ottoman sultan 11, 256–257 324, 326–329, 336–338, 410 Bayezid II, Ottoman sultan 11, 15, 125, 188, Apafi, Mihály, II, prince of 259, 267, 271, 279, 397, 405 Transylvania 409 Beauplan, Guillaume le Vasseur de 288 Apkerim Efendi 53 Ben Castiel, Mattatia 192 Aron the Tyrant (Tiranul), voievod of Benessa, Šimun 207 Moldavia 210, 214–215, 228–236, Bekes, Gáspár 330 248–249 Beldiman, Alexandru 35 440 personal names

Beneveni, Ivo 196 Chalkokondyles, Laonikos 258, 267 Bethlen, Gábor, prince of Charles I, [of Hohenzollern], king of Transylvania 72, 81, 83–88, 120, 160, Romania 20n 177, 222–223, 248n, 285, 293n, 296, Charles XII, king of Sweden 219 303–304, 307, 309, 311n, 313–314, Charles V, Holy Roman Emperor 270 320–324, 331–339, 409–410 Chirico, Đuro 196–197 Bethlen, István, prince of Chirico, Frederik 196 Transylvania 157n Chirico, Luka 196 Biegman, Nicolaas H. 18, 93, 95, 396 Chukhlib, Taras 151 Bielski, Polish chronicler 260 Ciocîltan, Virgil 253 Birken, Andreas 377 Claparède, Arthur de 19 Bíró, Vencel 155–156 Codignac, Michel de 24n Bocskai, István, prince of Comuleo, Alessandro 246n, 249 Transylvania 79–80, 303, 309, 408 Constantin Brâncoveanu, see Brâncoveanu, Bodin, Jean 112–113 Constantin Bogdan the Blind, voievod of Constantin Şerban, voievod of Moldavia 34 Wallachia 210, 214, 226n, 242n, 248n, Bogdan Lăpuşneanu, see Lăpuşneanu, 320 Bogdan Cosimo II de’ Medici, see Medici, Cosimo Borisi, Bernardo 221 Costin, Miron 274 Borisi, Marc’ Antonio 221 Costin, Nicolae 34–35 Bornemisza, Anna 271 Crasso, Frano 192 Borsos, Tamás 331 Crispo, Ciacomo, IV, duke of Naxos 394 Božović, Miho 193 Curić, Đuro 196 Brâncoveanu, Constantin 40n, 257 Čingrija, Pero 196–197 Brătianu, Gheorghe 125 Bratis, Mato 196 Daničić, Juraj 345 Brukhovetsky, Ivan, hetman 140 Daniel, prophet 238–241 Bull, Hedley 117 Danişmend, İsmail Hâmi 388 Dávid, Géza 96 Caboga, Pavao 199 Dávid, János 183 Canibek Giray II, khan of Crimea 291n, Davul İsmail Effendi 219 332, 382 de la Croix, Pétis 244 Cantacuzenus, Andronicus 232–235 Decei, Aurel 256, 267–268 Cantacuzenus, Antonius 240 Derviş Mehmed Pasha 403 Cantacuzenus, Demetrius 247n Despot-vodă, see Ioan Iacob Heraclid Cantacuzenus, Georgios 240–241 Devlet II Giray, Crimean khan 44, 63 Cantacuzenus, John, see John VI Doroshenko, Dmytro 143 Kantakuzenos Doroshenko, Petro, Cossack hetman Cantacuzenus “Shaytanoglu”, 139–151 411, 413–414 Mihail 226n, 230–232 Dousa, Georgios 241n Cantemir, Antiochus, voievod of Duca, Gheorghe, voievod of Moldavia 219 Moldavia 262, 404n Cantemir, Dimitrie, voievod of Ducas 40n Moldavia 12, 25, 34, 39n, 179–180, 211, Dupront, Alphonse 236 214, 218–219, 225–226. 243–248, 403–404, 431n Ebu Bekr-i Tihrani 262 Cappello, Girolamo 236 Ebussuud Efendi 23 Carile, Alfonso 241–242 El-Hacc Davud, khan of Şirvan 383 Casanova, Johann Baptist 385, 404 Eminek Bey 279 Casimir IV, king of Poland 258 Eton, William 31, 35 Catherine of Brandenburg, princess of Eugenikos, Manuel 241 Transylvania 403n, 409 Evliya Çelebi 47, 223n, 276–277, 281 personal names 441

Farah, Caesar 425 Heraclid, Ioan Iacob, voievod of Fazli Pasha 426 Moldavia 314 Ferdinand I, Holy Roman Homonnai Drugeth, Bálint 330, 408 Emperor 70–71, 74, 78, 268, 270, 310 Homonnai Drugeth, György 322 Ferdinand II, Holy Roman Emperor 81, Hrizea 320 220, 228, 332, 334 Hrushevsky, Mykhailo 128 Feth Giray 283 Hübsch, Frederich 196–197 Finkel, Caroline 53, 58n Hüseyin bin Ali At-Türkî 385 Fisher, Alan 43–44, 50 Hüseyin Efendi 286 Fodor, Pál 306 Fotić, Aleksandar 415 İbrahim I, Ottoman sultan 135, 192, 202 Ibrahim al-Halebî 22 Gazi Giray II, Crimean khan 283–284, İbrahim Kefeli 61 286, 289, 291n, 294, 297–298 İbrahim Peçevi 16, 283 Gedik Ahmed Pasha 276 İbrahim Pasha, beylerbey of Buda in 1682 Gemil, Tahsin 410n 328–329 Genghis Khan 46, 57, 60, 294 İbrahim Pasha, grand vizier 272 Gennadios Scholarios II, patriarch of İbrahim Pasha, Deli, see Sarhoş Ibrahim Constantinople 415 Pasha Georgewitz, Bartholomew 238, 240 İbrahim Pasha, Sarhoş, see Sarhoş Ibrahim Gheorghe Duca, see Duca, Gheorghe Pasha Gheorghe Ştefan, voievod of İdris Bitlisi 25, 386 Moldavia 226n, 242, 245n, 320 Iliaş, Alexandru, voievod of Ghica, Gheorghe, voievod of Moldavia 403–404 Moldavia 226n Imber, Colin 46–47 Ghica, Grigore, voievod of Wallachia 270 İnalcık, Halil 43, 50, 125–126, 276–279, Giese, Friedrich 256 415, 417 Giorgi, Bernard 192 Ioan Iacob Heraclid, see Heraclid, Ioan Giurescu, Constantin 36 Iacob Gonzague, Charles, duke of Nevers Ioan the Terrible (Ioan cel Cumplit), 221–222, 241 voievod of Moldavia 210, 214–215, Gothofredus, Johannes 113 226–228, 247, 249n, 265, 317 Gozze, Sekundo 206 Iskender Pasha 217, 321–322 Graziani, Gaspar, voievod of İslam Bey Cegielski 50 Moldavia 210, 214–215, 217, 220–226, Islam II Giray, Crimean khan 47–48 246–249, 322, 394, 403 Ismail, shah of Persia 280 Grigore Ghica, see Ghica, Grigore Ivan IV the Terrible, Russian tsar 227n Grigorij of Kiprovac 248n Ivan Kalita, grand duke of Muscovy 60 Gritti, Alvise 267–268, 272 Jan Sobieski III, king of Poland 143, 329 Hacı ‘Ali 426 János Zsigmond, see John Sigismund Haci Giray Sultan 290 Jeremias II Tranos, patriarch of Hacı Receb 54 Constantinople 230 Haga, Cornelis 176–177 John I Szapolyai, king of Hungary 70, Hammer-Purgstall, Joseph von 134n, 277n, 261, 268, 272, 305n, 312–315, 406 385, 388, 414n John III, king of Poland, see Jan Sobieski Halim Giray 48, 61 III Harlay, Achille de 24n, 31 John VI Kantakuzenos, Byzantine Harsányi Nagy, Jakab 223n emperor 240–241 Hasan Pasha, Abaza, see Abaza Hasan John Sigismund, elected king of Pasha Hungary 71–72, 74–76, 87, 305n, Hasan-ı Rumlu 262 314–315, 337 Heissler, Donat 257 Joseph, Père, see Père Joseph 442 personal names

Kabrda, Josef 415 Ligarides, Paisius 237–238, 242, 248n Kádár, János 304n de Linda, Luca 113 Kara Mustafa Pasha 147, 328–329 Louis I the Great, king of Hungary 191 Karácson, Imre 168 Louis II, king of Hungary 191, 405 Karakaş Mehmed Pasha 331 Louis XV, king of France 31 Karamanoğlu II İbrahim Bey 380–381 Lucinge, René de 244 Karaş Bey 292 Lučić, Luka 193 Karwowski, Stanisław 63–64 Lybyer, Albert Howe 21 Katib Çelebi 335n Lupu, Vasile 133–134 Kavalalı Mehmed ‛Ali, vali of Egypt 387 Kemalpaşazade 23, 25 Macarius Za’im, patriarch of Kemény, János, prince of Antioch 242n Transylvania 292 Magrini, Andrija 193, 196 Kenan Pasha 218 Malkoçoğlu Bali Bey 260 Khadduri, Majid 390 Mansfeld, Ernst count of 334 Khanenko, Mykhajlo 143 Manuel II Paleologos, Byzantine Khlesl, Melchior 83–84 emperor 256n Khmelnytsky, Bohdan, hetman 123–124, Marcello, Girolamo 260 128–139, 150, 413 Marcello, Ieronimo 271 Khmelnytsky, Iurii, hetman 143, 148 Marini Poli, Pascal de’ 231 Khmelnytsky, Tymish 134 Matei Basarab, voievod of Wallachia 320 Kinizsi, Pál 266 Matthias, Holy Roman Emperor 80–86, Kıvami 11–12 289 Klontzas, Georgios 239n Matthias I Corvinus, king of Hungary Knolles, Richard 14n 405 Koca Sinan Pasha 286, 294, 424–425 Mavrocordatos, Nicolaos, voievod of Kołodziejczyk, Dariusz 146, 405 Moldavia 214 Komninos, Yoannis 244 Maxim, Mihai 253, 400 Komulović, Aleksandar, see Comuleo, Maximilian II, Holy Roman Alessandro Emperor 75–76, 227n Köprülü Ahmed Pasha 144, 326 Medici, Cosimo II de’, grand duke of Köprülü Mehmed Pasha 135–136, 182, Tuscany 220 218n, 225, 409 Mehmed I, Ottoman sultan 11, 14n, 40, Kortepeter, Carl Max 44 256 Križanić, Juraj 237–238 Mehmed II the Conqueror, Ottoman Kutlu Giray Bey 289 sultan 11, 14–15, 41, 43, 61, 125, 187–188, Kuyucu Murad Pasha 217, 408 194n, 258–259, 262–263, 266–267, 279, Kurat, Akmes Nimet 44 381, 393, 396–397, 402, 415–416 Mehmed III, Ottoman sultan 16, 236, 240 Laiotă Basarab, voievod of Wallachia 34, Mehmed IV, Ottoman sultan 27, 144, 148, 36, 266 238, 324, 388, 428n Lala Mehmed Pasha 408 Mehmed Giray I, Crimean khan 295 Lambeccius, Petrus 240 Mehmed Giray II, Crimean khan 287n Lăpuşneanu, Bogdan, voievod of Mehmed Giray III, Crimean khan 49, 58, Moldavia 214 63–64, 287n Latif Çelebi 219 Mehmed Giray IV, Crimean khan 382 Lazarević, Stefan, see Stefan Lazarević Mehmed Pasha, Baltaci, see Baltaci Le Maire, André Alexandre 31, 197 Mehmed Pasha Leo the Wise, Byzantine emperor Mehmed Pasha, Karakaş, see Karakaş 237–239 Mehmed Pasha Leon, Radu, voievod of Wallachia 180, Mehmed Pasha, Köprülü, see Köprülü 226n, 404 Mehmed Pasha Leopold I, Holy Roman Emperor 115n, Mehmed Pasha, Rami, see Rami Mehmed 308, 326–329 Pasha personal names 443

Mehmed Pasha, Tabanıyassı, see Nassi, Josef, duke of Naxos 394 Tabanıyassı Mehmed Pasha Navagero, Bernardo 268n, 270 Mehmed-Paşa Küçük Nişanci 15 Neculce, Ion 34 Mehmed Tahir b. Mehmed Râhîmî 22 Nefioğlu 219 Mehmed Senai 52 Nevers, see Gonzague, Charles Mehmet Neşri 11 Nikephorus “the Deacon” 230–232, 248 Meienberger, Peter 172n Nikousios, Panagiotis 225–226, 238 Mengli Giray, Crimean khan 43n, 62, Noradungiyan Efendi, Gabriel 388 276, 279 Mesgnien-Meninski, François 166 Oborni, Teréz 305n Mesih Pasha 259, 267 Ömer Pasha 331 Methodius of Patara 238n Orhan, Ottoman sultan 40n, 105–106 Mihai the Brave (Viteazul), voievod of Özbek, khan of the Golden Horde 60 Wallachia 210, 214–215, 226, 228–236, Osman II, Ottoman sultan 323, 426 246–248 Osuna, see Téllez-Gíron de Osuna, Pedro Mihail Radu, see Mihnea III Mihnea II (Turcitul), voievod of Païsios, patriarch of Jerusalem 242n Wallachia 16, 231n, 400, 402 Panaite, Viorel 53, 157n, 212, 392n, 410n Mihnea III, voievod of Wallachia 210, Papacostea, Şerban 402 214–215, 218, 223–226, 237, 240, 243, Papp, Sándor 157n, 306–307, 315n 245–249 Parrot, David 358, 362 Mikołaj Prażmowski, Polish Crown Parthenius, patriarch of Chancellor 50 Constantinople 242n Miletos, metropolite of Eski Yogan 415 Paul V, pope 249 Mircea the Elder (cel Bătrân), voievod of Pedani, Maria Pia 62, 178n Wallachia 14, 34, 36, 255–257, 265 Pemfflinger, Mark 268 Miron Barnovschi, see Barnovschi, Miron Père Joseph 222, 238n, 241n Molla Hasân El-Kâfi 375 Peysonnel, Charles 59 Montecuccoli, Raimondo 224 Peter I the Great, Russian tsar 211, 243, Moréri, Louis 113 248, 280 Moro, Giovanni 209–211, 217 Petriceicu, Ştefan 245n, 247, 262 Moruzi, Constantin, voievod of Petru Aron, voievod of Moldavia 15n, Moldavia 403 258, 402 Movilă, Alexander, voievod of Petru Cercel, see Cercel, Petru Moldavia 321 Petru Muşat, voievod of Moldavia 257 Muhammad, prophet 290, 379, 389–391 Petru Rareş, voievod of Moldavia 16, 21, Muhammad bin Shams al-Din 424–425 30, 34, 261, 268, 272 Murad I, Ottoman sultan 40 Petru the Lame (Şchiopul), voievod of Murad II, Ottoman sultan 187, 253, 265, Moldavia 210, 317, 319 380, 405 Pidkova, Ivan 317 Murad III, Ottoman sultan 38, 289, 309n, Popescu, Radu 14 319, 325–326, 386, 396, 398 Poumarède, Géraud 249n Murad Giray, Crimean khan 61n Pseudo-Sphrantzès 237 Murad Pasha, Kuyucu, see Kuyucu Murad Pritsak, Omeljan 132n, 164n, 413 Pasha Murteza Pasha 334 Querini, Vicko 192 Musa Çelebi 256, 265 Qutb al-Din al-Nahrawali 424 Mustafa Cenabi 277 Mustafa Pasha, Kara, see Kara Mustafa Rácz, István 180n Pasha Rácz, Mihály 317n Mutahhir, şeyh of Yemen 387 Radák, László 314–315 Radelja, Nikola 193 Nani, Bernardo 91 Radu of Afumaţi, voievod of Nagy, László 305 Wallachia 14 444 personal names

Radu the Handsome (cel Frumos), voievod Selim I Giray, Crimean khan 54, 63 of Wallachia 14, 262 Selim II Giray, Crimean khan 59n Radu Leon, see Leon, Radu Semion, knez of Rostov 227n Radu Mihnea, voievod of Moldavia and Serjeant, Robert B. 389, 423 Wallachia 218 Seyyid Muhammed Riza 47–48, 50, 61 Radu Şerban, voievod of Wallachia 221n, Shaytanoglu, see Cantacuzenus, Mihail 248n, 331, 401, 403 Sibrik, Gáspár 319 Rajić, Gavrilo, patriarch of Peć 242n Sigismund, Holy Roman Emperor 396 Rákóczi, Ferenc, I, elected prince of Sigismund I, king of Poland 21, 270, Transylvania 409 422–423 Rákóczi, Ferenc, II, duke of Hungary and Sigismund III, king of Poland 49n prince of Transylvania 311, 412 Sinan Pasha, Koca, see Koca Sinan Pasha Rákóczi, György, I, prince of Skilliter, Susan 38n Transylvania 131, 160, 173, 311n, 323, Skordylis, Zacharias 239 335, 337, 339, 409 Smirnov, Vasilii Dmitrievich 421 Rákóczi, György, II, prince of Sobieski, Jan, see John III, king of Poland Transylvania 131, 135, 174, 178, 182, 210, Sorgo, Trojan 192 214, 223, 248n, 292n, 296n, 320, 409, 412 Stefan Lazarević, Serbian prince 256 Rákóczi, József 412 Ştefan Petriceicu, see Petriceicu, Ştefan Rami Mehmed Pasha 219 Ştefan Lăcustă, voievod of Moldavia 15 Reinhard, Wolfgang 213 Ştefan Răzvan, voievod of Moldavia 234n Remmal Hoca 44–45, 47, 278, 292n Ştefan Tomşa, voievod of Moldavia 314, Reniger, Simon 223n 321 Rhalle, Dionysius 232, 248n Ştefan II, voievod of Moldavia, see Stephen Roe, Thomas 120 II Rozgonyi, István 258, 265 Stephanos of Alexandria 237 Rozsnyai, Dávid 176 Stephen, king of Poland, see Báthory, Rudolph II, Holy Roman Emperor 77–79, István 228, 233, 298 Stephen the Great (cel Mare), voievod Runciman, Garry 431 of Moldavia 15, 34, 41, 258–263, 267, Rycaut, Paul 163, 165–166 271–272, 433 Stephen the Young (cel Tânăr), voievod of Seydlitz, Melchior 112 Moldavia 260–261, 270–271 Sa‘adeddin Hoca 289 Stephen II, voievod of Moldavia 257–258 Sa‘adet I Giray, Crimean khan 45 Sukhovy (Sukhovienko), Petro, Sahib Giray, Crimean khan 45, 47, 52, hetman 141, 143 61–62, 278–279, 281n, 285, 292n, Şükrullah 40n, 257 422–423 Süleyman I the Magnificent, Ottoman Şahin Giray 296, 382 sultan 9–11, 15–17, 21, 24–30, 40n, Şahin Pasha 192 42, 45, 52, 71, 73–74, 203n, 239n, 256, Sansovino, Francesco 113 260–261, 266–267, 272–273, 280, 283, Sarhoş İbrahim Pasha 321, 333–334, 285n, 293, 314n, 386, 394–395, 406, 422–423 422–423, 425 Sárosi, János 183 Szalánczy, István 161n, 182 Scaglia, Marco Aurelio 241n Szapolyai, János, see John I Schmid, Johann Rudolf 164n, 178n Székely, Mózes, prince of Sefer Gazi Ağa 50 Transylvania 309, 313 Selamet Giray 57–58 Szepsi Lackó, Máté 322n Selânikî 283 Szilágyi, Sándor 302 Selim I, Ottoman sultan 22–23, 47, 94n, Szilvássy, Bálint 183 280, 384, 386, 396, 415 Selim II, Ottoman sultan 53, 239, 317n, Tabanıyassı Mehmed Pasha 160 394 Téllez-Gíron de Osuna, Pedro 220 personal names 445

Testa, Ignace de 38n Vlad II Dracul, voievod of Wallachia 253, Tetera, Pavlo 139 265 Tezcan, Baki 425 Vlad III the Impaler (Ţepeş), voievod of Theodemar, Suevic king of Galicia 393 Wallachia 258 Thököly, Imre 166, 303, 328, 336, 377, Vlad the Monk (Călugărul), voievod of 380n, 410–412, 414 Wallachia 267, 271 Tholdalagi, Mihály 178–179 Vlad Vintilă, voievod of Wallachia 272 Thomassy, Gabriel 248n Thornton, Thomas 31–32 Wallenstein, Albrecht von 334 Thurzó, György 83–84 Wheaton, Henry 19n Tiepolo, Stefano 239n Wiśniowiecki, Dimitry 316n Timoni, Angelo 194, 196 Wladyslaw I, king of Hungary 405 Tohta Bey 276 Wladyslaw II, king of Hungary 405 Tohtamış Giray 57–58 Tudisi, Augustin 192 Yani “the Ban” 229–232, 235n Tursun Bey 265, 267 Zaitsev, Iliia V. 277n ‛Umar, caliph 391, 414 Zamoyski, Jan 321n, 325 Uzun Hasan, Ak Koyunlu khan 262, Zane, Matteo 236 266 Zani, Giovanni Luigi 269–270 Uzunçarşılı, İ. Hakki 380 Zápolya, János, see John I Zarini, Miho 193 Vasilii Aitemiriev 53–54, 64 Zborowski, Samuel 319 Verancsics, Antal 73 Zenbilli Ali Efendi 23 Vernazza, Antun 193 Zeno, Caterino 262, 266 Vlachos, Gerassimos 242, 244 Zólyomi, Miklós 169 place names

Adrianople (Edirne) 19, 35, 54, 93n, 160, Busza 322 169, 183n, 324, 397, 402, 405 Byzantium 12, 41, 42n, 233, 236, 239n, Akkerman (Bilhorod-Dnistrovskyj) 58, 240n, 259n, 377, 394, 431n 125, 279, 402, 435 Alba Iulia, see Gyulafehérvár Cahul (Kagul) 265 Algeria 384–385 Caffa, see Kaffa Andrusovo 139–140, 148 Çaldiran, see Chaldiran Ankara 256, 256n, 262 Caramania 380, 417 Astrakhan 46n, 62, 123, 126, 297n Carei, see Nagykároly Azak, see Azov Castel Nuovo, see Herceg-Novi Azerbaijan 280, 383n, 418n Cătlăbuga 267 Azov (Azak) 125, 244, 284 Cattaro, see Kotor Cecora, see Ţuţora Bahçesaray (Bakhchesarai) 46n, 53–54, Cetatea de Baltă, see Küküllővár 55n, 63, 123, 137, 148, 176n, 294 Chaldiran 280 Bakhchesarai, see Bahçesaray Chehrin, see Chyhyryn Balat 168–169 Chilia, see Kilia Banská Bystrica, see Besztercebánya Chios 395 Başkent 262, 266, 273 Chyhyryn (Chehrin) 133, 135, 137, 141, Batum (Batumi) 285n 146–148, 148n, 149n Belgrade 16, 56n, 193, 202, 218, 284, 329, Chocim (Hotin, Khotin) 33n, 126, 271, 343, 347, 347n, 348, 402 323, 426, 427n Bender (Tighina) 33n Ciceu, see Csíkcsicsó Berestechko (Beresteczko) 131 Çıldır 385 Besztercebánya (Neusohl, Banská Constantinople, see Istanbul Bystrica) 286 Corfu 215, 224n, 398 Bilhorod-Dnistrovskyj, see Akkerman Courland 428n Bilohirsk, see Kara Su Crimea 1, 2, 43n, 44n, 45–48, 48n, 49, Bitlis 385, 385n, 428, 429, 431 49n, 53–54, 54n, 58, 59n, 63–65, 125–126, Boğdan, see Wallachia 128–131, 138–139, 141, 149, 274, 275–276, Borosjenő, see Jenő 276n, 278n, 279, 281, 281n, 282, 282n, Bosnia 10n, 12, 21, 41n, 113n, 156, 157n, 285n, 288–289, 292, 294, 295n, 297, 316, 165n, 179, 187, 187n, 202, 207n, 262, 321, 381, 381n, 382–383, 417, 421 333–334, 387, 387n, 416, 416n Csíkcsicsó (Ciceu) 261n Brăila (Ibrail) 33n, 285 Curzola, see Korčula Brassó (Kronstadt, Braşov) 42n, 285n, Cyprus 37n, 342n, 345, 359n, 427 319n, 323n, 331n Csík-, Gyergyó- and Kászonszék Braşov, see Brassó region 319n Bruck an der Leitha 329 Buczacz (Buchach) 145, 429n Davutpaşa 174, 174n Buda (Budun) 16, 71, 73, 86, 86n, 87, Derbent 278, 284 156, 157n, 163, 164, 179, 286n, 290n, 293, Déva (Deva) 330 311n, 314, 314n, 328, 329, 331, 334–335, Divény (Divín) 330n 405n Diyarbekir 262n, 385 Budun, see Buda Dobruja region (Dobruca) 284 Bulgaria 10n, 18n, 239n, 377, 419 Dubrovnik, see Ragusa Bursa 93n, 396, 396n, 397, 402 Dunaföldvár 289 place names 447

Edirne, see Adrianople Iaşi 210, 218n, 232, 317, 319, 431 Eflak, see Wallachia Ibrail, see Brăila Eger (Eğri, Erlau) 86–87, 285n, 332, 334, Imbros 12 335n, 375n Ineu, see Jenő Egypt 22, 23n, 31–32, 35, 263, 384, Istanbul (Constantinople) passim 387–388, 397–398, 418, 424n England 37n, 38, 40n, 161, 162n, 164, 194, Jelgava, see Mittau 220n, 227n, 229n, 233n, 325n, 388, 398, Jenő (Ineu, Yanova) 145, 285n 417 Erlau, see Eger Kaffa (Caffa, Kefe) 48, 56n, 125, 258, Érsekújvár (Neuhausel, Nové Zámky, 276, 279, 285, 285n, 287n, 291n, 381, 382, Uyvar) 286, 326–327 402n Erzurum 285n, 383, 385n Kagul, see Cahul Kálló 336 Fener 168–169, 171, 171n, 172, 194n, 403 Kamaniche, see Kamjanec’ Podolskij Fiľakovo, see Fülek Kamieniec Podolski, see Kamjanec’ Filibe, see Plovdiv Podolskij Florence 218 Kamjanec’ Podolskij (Kamieniec Podolski, Fojnica 416, 416n Kamaniche) 274, 324 France 24n, 37, 37n, 39n, 114, 115n, 117, Kanizsa 286n, 290, 334 161, 162n, 164, 194, 194n, 197, 203, 221n, Kara Su (Bilohirsk) 46n, 53, 53n, 55n, 222n, 227n, 362, 378, 398, 406, 417 56n Fülek (Fiľakovo) 328 Karlowitz (Sremski Karlovci) 39n, 146, 429n Galata 169, 193, 194, 194n, 200n, 393, Kartil 377 394n Kaschau, see Kassa Gallipolli 383 Kassa (Kaschau, Košice) 81–82, 119, 285n, Genoa 98, 194, 355n 325n, 410n Georgia 297, 377 Kawkaban 424 Gherla, see Szamosújvár Kazan 46n, 276n, 281n Giurgiu (Gyurgyevó) 311n Kefe, see Kaffa Gözleve (Yevpatoria) 46n, 285, 285n Kékkő (Modrý Kameň) 330 Großwardein, see Várad Kerelőszentpál (Sânpaul) 330 Güns, see Kőszeg Kiev (Kyiv) 49n, 123n, 126, 130, 139, 141, Gürcistan 385, 407n 148 Győr (Raab) 286, 289, 291n, 293, 329 Kilia (Chilia) 258, 258n, 259, 267, 267n, Gyulafehérvár (Alba Iulia) 72n, 174n, 279 177n, 178n, 179n, 229, 233n, 234n Kodak (Kudak) 141 Gyurgyevó, see Giurgiu Konavle region 107n, 347–348, 348n, 369 Habsburg Empire 16n, 37n, 38–39, 56n, Konya 380 57n, 120n, 168, 303n, 327–328, 336, 378, Korčula (Curzola) 347, 371 430 Košice, see Kassa Hakkari 385n Kotor (Cattaro) 370, 371 Hârlău 270 Königsberg (Królewiec, Háromszék region 319n Kaliningrad) 428n Hatvan 290n Kőszeg (Güns) 314n Herceg-Novi (Castel Nuovo) 347, 348n, Kronstadt, see Brassó 369 Królewiec, see Königsberg Hermannstadt, see Nagyszeben Küçükçekmece 174 Hodonín 334 Kudak, see Kodak Hotin, see Chocim Küküllővár (Cetatea de Baltă) 261n Hudaybiya 389 Kyiv, see Kiev 448 place names

Lagosta, see Lastovo Nitra, see Nyitra Lastovo (Lagosta) 103, 105, 107, 109, 342, Nógrád 334 342n, 348, 351–352, 354n Nové Zámky, see Érsekújvár Lemberg, (Lwów, L’viv) 228, 260, 317 Nyitra (Nitra) 327 Lemnos 12 Lesbos 12 Ochakiv (Ochakov, Özi, Vozia) 33n, 125 Linz 83, 83n, 84n, 218 Or, see Perekop Lippa (Lipova) 321–322, 416 Oradea, see Várad Lipova, see Lippa Orta Macar 166n, 328, 328n, 377, 411, 414 L’viv, see Lemberg Özi, see Ochakiv Lwów, see Lemberg Paris 116, 197n, 240, 293n, 412, 428 Madrid 116, 228n Parme 221 Malta 359n Pasarofça, see Passarowitz Mantua 221, 221n, 222n Passarowitz (Pasarofça, Požarevac) 39n Mecca (Mekka) 384, 384n, 389, 426 Pelješac region 349 Međimurje region (Muraköz) 286, 293 Pera 24n, 31n, 169, 171–172, 180n, 193–194, Medina 389, 389n, 391n 200n, 207 Mekka, see Mecca Pereiaslav 136–137, 140 Mesembria (Nesebar) 256 Perekop (Or) 43n Mezőkeresztes 290, 311n, 325n Peremyshl’, see Przemyśl Middle Hungary, see Orta Macar Persia 216n, 228, 262, 262n, 263, 272, 273, Milodraž 416 279, 280, 297, 323, 337 Mingrelia 377 Ploieşti 320 Mitau (Jelgava) 428n Plovdiv (Filibe) 193 Modrý Kameň, see Kékkő Podolia region 123, 123n, 144–146, 146n, Mohács 16, 21, 69n, 70, 191, 261, 305n, 147, 149, 274, 280n, 427 316n, 405, 405n Poland 21, 35n, 37n, 38, 41n, 75, 76n, 77, Mohyliv (Mohylów) 145 77n, 99n, 116, 117, 126, 128n, 129, 130–131, Moldavia passim 136, 138, 140, 145, 150, 156, 164, 171, 176, Morea 262 209, 227, 227n, 228n, 229, 229n, 235, 258, Moscow 63, 131, 132n, 136–142, 146, 148, 259, 260–262, 267, 270, 282n, 293, 295, 227, 228n, 242, 242n, 297, 413n, 426n, 295n, 296–297, 298, 306n, 309n, 310, 316, 427n, 428n 316n, 317, 317n, 333, 378, 380n, 405–406, Mount of Olives 414 409, 413, 422, 422n, 423, 426–427, 427n, Muraköz, see Međimurje 428n, 429, 430 Murcia 393 Polanów, see Polianowka Muscovy, see Russia Polianovka (Polanów) 132 Požarevac, see Passarowitz Nağrân 390, 391n Przemyśl (Peremyshl’) 141 Nagykároly (Carei) 336 Putivl’ 141 Nagyszeben (Hermannstadt, Sibiu) 15, 181n, 270n, 285n, 307n, 323n, 331 Raab, see Győr Nagyszombat (Tyrnau, Trnava) 87, 119, Ragusa (Dubrovnik) passim 305n Rhodes 227, 227n, 260n, 383 Naples 114, 116, 204, 216n, 220, 228n, Rome 116, 201, 215, 222, 225n, 235, 246n, 354n, 335n, 363, 367, 367n 248n, 249, 259 Naxos 12, 217, 217n, 394, 394n Ruschuk (Ruse) 193 Nemyriv (Nemirov) 148 Ruse, see Ruschuk Nesebar, see Mesembria Russia 19n, 35, 38, 38n, 39, 39n, 40n, 43n, Netherlands 161, 162n, 164, 176, 358, 417 44, 49n, 50, 115n, 136, 136n, 137n, 140n, Neuhausel, see Érsekújvár 194, 211, 219, 235, 241n, 243n, 244, 244n, Neusohl, see Besztercebánya 245n, 277, 282n, 295, 295n, 297, 398, Nicopolis 256 414n, 421, 427n, 429, 430, 431n place names 449

Sakız, see Chios Trebinje 346 Sambor 141 Tripolitania 384 Sânpaul, see Kerelőszentpál Trnava, see Nagyszombat Satu Mare, see Szatmár Trnovo 1, 232, 232n, 248n Savoy 221, 221n, 222n Tunis 37n, 346, 384–385 Schäßburg, see Segesvár Tuscany 220, 221, 221n Şcheia 267 Ţuţora, see Cecora Sebenico, see Šibenik Tyrnau, see Nagyszombat Segesvár (Schäßburg, Sighişoara) 329 Segna, see Senj Uyvar, see Érsekújvár Şehrizor 385 Senj (Segna) 345, 345n, 346n, 347n Vác (Waitzen) 326n, 331, 334 Serbia 1, 10n, 12, 19n, 41n, 187, 187n, 377, Valona 346, 369 419 Van 315 Sevsk 141 Várad (Großwardein, Oradea) 71, 163n, Šibenik (Sebenico) 359n 183, 289, 304, 315n, 327, 336 Sibiu, see Nagyszeben Vaslui 266 Sighişoara, see Segesvár Venice 37n, 91n, 99n, 112–114, 114n, Silistria (Silistra) 156, 157n, 164, 179, 320, 116–117, 129, 135, 140, 156, 162, 162n, 164, 322–323, 329, 334, 334n 167n, 189, 190, 193, 194–195, 220, 221n, Şirvan 383, 383n 223, 224n, 228n, 242, 242n, 243, 248n, Smolensk 323 337, 342, 342n, 348, 350, 363, 368–369, Sofia 193, 249n 370–372, 378, 378n, 398, 417, 427, 427n, Sokol 349, 356 429–430 Somoskő (Šomoška) 330n Vienna 19, 67, 74, 78–81, 83, 87–88, 116, Şoplea 320 149, 158, 165n, 178, 201, 218, 227, 228n, Spain 114, 189, 190, 190n, 201, 204, 221, 262, 271, 271n, 280, 290, 290n, 298, 314, 228, 337, 342, 362, 368 328, 328n, 329–330, 332, 375n, 427–428, Sremski Karlovci, see Karolowitz 428n Stagno, see Ston Vozia, see Ochakiv Stockholm 428 Ston (Stagno) 347, 349, 351n, 356, 361, Waitzen, see Vác 363 Wallachia passim St. Petersburg 421, 428 Warsaw 63, 138–139, 143, 228n, 297, 421, Syria 384, 390n 426n Szamosújvár (Gherla) 328n Szatmár (Satu Mare) 336, 336n Yanbolu (Yambol) 277 Szeged 416 Yanova, see Jenő Yathrib 389, 389n Tabriz, see Tebriz Yemen 387, 390, 423–428, 430–431 Tărăsăuşi 260 Yevpatoria, see Gözleve Tebriz (Tabriz) 280 Temesvár (Temeschwar, Timişoara) 156, Zadar (Zara) 372, 372n 157n, 163, 258, 289, 321, 325, 325n, 330n, Zara, see Zadar 331–332, 334 Zboriv (Zborów) 131–132 Temeschwar, see Temesvár Zărneşti, see Zernyest Tighina, see Bender Zernyest (Zărneşti) 257, 257n Timişoara, see Temesvár Zhvanec’ (Żwaniec) 131 Trablosgarb, see Tripolitania Zsitvatorok 79, 79n, 161, 298, 298n, 429n Transylvania passim Žuravno (Żurawno) 298