Working on Labor Studies in Global Social History

Series Editor Marcel van der Linden International Institute of Social History, Amsterdam, The Netherlands

Editorial Board Sven Beckert Harvard University, Cambridge, MA, USA Philip Bonner University of the Witwatersrand, Johannesburg, South Africa Dirk Hoerder Arizona State University, Phoenix, AZ, USA Chitra Joshi Indraprastha College, Delhi University, India Amarjit Kaur University of New England, Armidale, Australia Barbara Weinstein New York University, New York, NY, USA

VOLUME 9

The titles published in this series are listed at brill.nl/sgsh

Working on Labor

Essays in Honor of Jan Lucassen

Edited by Marcel van der Linden and Leo Lucassen

LEIDEN • BOSTON 2012 Cover illustration: Cover of a bound copy of the forty weekly issues of Exposition de Paris de 1889 published at the occasion of the World’s Fair. Courtesy International Institute of Social History, Amsterdam [IISG ZF 10372]

Frontispiece: Jan Lucassen (Photograph: Viem Tummers).

Library of Congress Control Number: 2012943853

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.nl/brill-typeface.

ISSN 1874–6705 ISBN 978 90 04 22952 5 (hardback) ISBN 978 90 04 23144 3 (e-book)

Copyright 2012 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

List of Figures and Tables ...... xi Acknowledgments ...... xv

Introduction ...... 1 Marcel van der Linden and Leo Lucassen

Work Processes

Working for Diamonds from the 16th to the 20th Century ...... 19 Karin Hofmeester

Green Plants into Blue Cakes: Working for Wages in Colonial Bengal’s Indigo Industry ...... 47 Willem van Schendel

Dak Roads, Dak Runners and the Re-ordering of Communication Networks ...... 75 Chitra Joshi

Production Systems and Forms of Labor Control in the Javanese Cigarette Industry; 1920s–1930s ...... 99 Ratna Saptari

Occupations

Selling in the Shadows: Peddlers and Hawkers in Early Modern Europe ...... 125 Danielle van den Heuvel

The Worst Class of Workers: Migration, Labor Relations and Living Strategies of Prostitutes around 1900 ...... 153 Lex Heerma van Voss

Fighting for a Living in Europe and Asia ...... 171 Erik-Jan Zürcher viii contents

White Collar Workers of the VOC in Amsterdam, 1602–1795 ...... 191 Karel Davids

Mobilities

The Career Ladder to the Top of the Dutch East India Company: Could Foreigners also Become Commanders and Junior Merchants? ...... 215 Jaap R. Bruijn and Femme S. Gaastra

Demographic Change and Migration Flows in Holland between 1500 and 1800 ...... 237 Jan Luiten van Zanden and Maarten Prak

The Economic Contribution of Labor Migrants in the European Maritime Labor Market of the Long Eighteenth Century ...... 247 Jelle van Lottum

Income Differentials, Institutions, and Religion: Working in the Rhineland or Pennsylvania in the Eighteenth Century ...... 269 Richard W. Unger

Contextualizations

Labor Laws in Western Europe, 13th–16th Centuries: Patterns of Political and Socio-economic Rationality ...... 299 Catharina Lis and Hugo Soly

The First “Male Breadwinner Economy”? Dutch Married Women’s and Children’s Paid and Unpaid Work in Western Europe Perspective, c. 1600–1900 ...... 323 Elise van Nederveen Meerkerk

Wage Labor and the Household Economy: A Russian Perspective, 1600–2000 ...... 353 Gijs Kessler contents ix

Cane Sugar and Unlimited Supplies of Labor in the 1930s: New Thinking and the Origin of Development Economics ...... 371 Ulbe Bosma

Sources

Unwritten Autobiography: Labor History Libraries before World War i ...... 395 Jaap Kloosterman

Notes on Contributors ...... 417 Tabula Gratulatoria ...... 423 Index ...... 425

List of Figures and Tables

Figures

Hofmeester 1. Diamond fields in India ...... 22 2. Natural octahedron whose natural planes were polished; Table cut ...... 32 3. Rose cut, top view and side view ...... 32

Van Schendel 1. Laborers beating the indigo solution ...... 55 2. Laborers removing used indigo plants from steeping vats, 1850s ...... 58 3. Laborers working a Chinese pump at an indigo factory, 1850s ...... 61 4. A Buna factory laborer, ready to beat indigo, 1850s ...... 65

Joshi 1. Mail runners ...... 77

Davids 1. Shipping and sales of the Dutch East-India Company, 1602–1795 ...... 194

Van Zanden and Prak 1. The population curve of Holland, 1510–1800 ...... 243

Van Lottum 1. The share of migrants per labor market ...... 253

Unger 1. Number of ‘German’ immigrants entering at Philadelphia ...... 275 2. Average number of persons per ship ...... 276 3. Ports of embarkation ...... 278 xii list of figures and tables

4. Number of ships entering Philadelphia with ’German’ immigrants ...... 283 5. Average tonnage of ships carrying immigrants ...... 284

Van Nederveen Meerkerk 1. A basic taxonomy of work and labor relations in Western Europe, 1600–1900 ...... 325 2. Married women’s work per sector in the official census statistics, Netherlands 1899 ...... 331

Kessler 1. Labor relations Russia 1678—total population ...... 361 2. Labor relations Russia 1795—total population ...... 363

Kloosterman 1. The cyrillic alphabetical catalogue of the Lavrov-Goc library used by the Russian Party of Socialists-Revolutionaries ...... 412 2. The catalogue of the Twickenham Museum library, designed as a guide for workingman families; from a collection brought together by Leon Kashnor ...... 415 3. Pamphlet on ‘social museums’ by Leopold Katscher (1853–1939), in a series opened by Werner Sombart in 1904 ...... 416

Tables

Van Schendel 1. Colgong indigo factory: Employment of day laborers in 1827–1828 (in workdays/month) ...... 56

Heerma van Voss 1. Reasons given by former prostitutes for their entry in the profession 1948 and 1951 ...... 164

Davids 1. Number of bookkeepers, cashiers and clerks at the Chambers Amsterdam, Zeeland and Delft, 1630–1790 ...... 194 2. Statutory pay levels for white collar workers at the Chamber Amsterdam, c.1630–1780, by category and rank ...... 198 list of figures and tables xiii

3. Statutory remuneration and estimated actual levels of income of white collar workers of the Chamber Amsterdam, c. 1742 ...... 204 4. Estimated incomes of white collar workers of the VOC, compared with other white collar workers in Amsterdam in 1742 ...... 206 5. Careers of clerks of the Chamber Amsterdam, by cohort, 1640–1769 ...... 209

Bruijn and Gaastra 1. Provisional numbers of Dutch, foreign and unknown commanders ...... 222 2. Number of junior merchants sent to Asia (1700–1795) ...... 230

Van Zanden and Prak 1. Estimates of the crude death rate and the crude birth rate in Holland, 1510–1800 (in ‰) ...... 241 2. Sources of population growth and immigration in Holland, 1510–1800 (in thousands) ...... 244

Van Lottum 1. Origins of workers in the Dutch maritime labor market ...... 259 2. Average age of workers in the Dutch maritime labor market ..... 261 3. Numeracy levels among crews in Scandinavia, Germany and the Dutch Republic ...... 263

Unger 1. Estimated Population of Pennsylvania ...... 271 2. Subsistence Ratios: Laborers ...... 279

Van Nederveen Meerkerk 1. Self-employed women according to marital status and sector, Netherlands 1899 ...... 337 2. Rough estimates of assisting labor on family farms, the Netherlands 1500–1900 ...... 339 3. Married women’s and children’s housework and subsistence activities according to court records in Württemburg, c. 1650–1800 ...... 348 4. Estimates of married women and children in various types of labor relations, the Netherlands 1899 ...... 352

Acknowledgments

The timely production of this volume by an international team would have been impossible without the support of Marti Huetink, our publishing agent at Brill; Jurriaan Bendien, who translated the introduction, copy-edited most of the contributions at short notice, and offered some helpful suggestions; and Nancy Peiffer, who copy-edited the contributions of Danielle van den Heuvel and Jelle van Lottum in due time. We are also grateful to Gerben van der Meulen and Viem Tummers, who supplied the portrait photo of Jan Lucassen—taken without Jan realizing why it was done.

Introduction

Marcel van der Linden and Leo Lucassen

This collection of essays takes its inspiration from the scholarly achievements of the Dutch historian Jan Lucassen. They were written on the occasion of his 65th birthday, and reflect a central theme in his research: the history of labor. With his numerous studies of labor migration, labor markets and remuneration systems in Europe and Asia, Lucassen made a name for himself—at first in the Netherlands, but gradually and increasingly clearly also at the international level—as an unorthodox historian who dared to pursue new topics and try out new methods. As the readers of this festschrift travel through histories of labor across many different places and continents, they will discover that each of the contributing scholars has taken up the challenge of this innovative endeavor.

Jan Lucassen: champion of labor history on a grand scale

Jan Lucassen was born on 7 July 1947 in the township Meijel (Limburg province), as the oldest son of schoolmaster Leo J. Lucassen and his wife Maria Crijns. Maria was the daughter of the headmaster at the primary school in the nearby town of Helden, and until her marriage she worked as a nurse. From his father, himself the son of a cigar manufacturer and also born in Helden (though raised in Maasbree), he inherited a love of history; indeed they have jointly published about the local history of their home region.1 Growing up in a strongly Catholic milieu, Jan initially wanted to become a missionary in Africa. He therefore attended seminary school as a twelve year-old—the Congregation of the Holy Spirit, in Weert. It was a strictly regimented boarding school, where Jan made the discovery that a celibate life was not for him. He therefore chose not to attend the Great Seminary, where he would have been educated for the

1 Jan Lucassen and L.J. Lucassen, “De arbeiders aan het Grand Canal du Nord, het tra- ject Weert-Meijel-Venlo 1809–1810,” Medelo, 2 (1983), pp. 41–9; “Marskramers uit Meijel en omgeving: de Teuten, 1730–1830,” Medelo, 2 (1983), pp. 50–9. 2 introduction priesthood, and instead set his sights on an academic study of archaeology in Leiden, where he became a student in 1966. Already as a first-year student he met his wife to be, Lieske Vorst. Although he kept an abiding interest in the distant past, he quickly traded archaeology for history after only a few months study. His most important history teacher, aside from Peter Klein, was Dik van Arkel who, as a follower of Karl Popper’s critical rationalism, consistently argued for a social-scientific approach to history. Jan made a great impression on Van Arkel with his Master’s thesis about the earliest contacts between Dutch colonists of the East India Company and the indigenous inhabitants of South Africa at the Cape. With a careful analysis of letters and reports written at the time, Jan demonstrated that although prejudices had existed from the start, to a great extent there was also a relatively “open” interaction between Europeans and Africans.2 Jan obtained his Master’s degree in 1973, majoring in social and economic history. After a brief intermezzo as teacher in a teacher’s training course, he was appointed as scientific researcher in Utrecht, where he eventually gained his Phd in 1984. His supervisor was the wellknown Dutch Marxist Theo van Tijn, who just like Van Arkel had a social-scientific orientation. In 1988, Jan joined the International Institute of Social (IISH) in Amster- dam, where his Utrecht colleague Eric Fischer had meantime become director. Here Jan felt truly in his element, and he was to stay on the staff for the rest of his professional career. In 1990, however, he was also appointed part-time professor of international and comparative social history at the protestant Free University of Amsterdam. With great personal effort and dedication he built the research division at the IISH which, from 1993, became an independent unit; its emphasis was always on internationally comparative historiography. But services and infrastructure also captured Jan’s attention. Jan set a precedent when he accommodated the builders of historical data sets, and he initiated the very successful bi- annual European Social Science History Conferences, held since 1996. When he resigned his position as Research Director in 2000, ‘his’ research division had become what is probably the largest research group on social history in the world.

2 This insight was later used by Van Arkel in Dik van Arkel, Chris Quispel and Robert Ross, De wijngaard des Heeren? Een onderzoek naar de wortels van “die blanke baasskap” in Zuid-Afrika (Leiden, 1983). marcel van der linden and leo lucassen 3

In the last twelve years of his academic career, Jan was associated with the IISH as Senior Research Fellow. Because of his great achievements as researcher and science manager, he was made a member of the Royal Netherlands Academy of Arts and Sciences in 2004.

Labor migration

Labor migration was the central focus of Jan Lucassen’s intellectual production, the theme to which he always returned after sometimes lengthy excursions into other fields of research. This interest was not purely academic, but also politically motivated. As a student in the late 1960s, Jan belonged to a group in Leiden concerned with the predicament of local “guestworkers” (gastarbeiders). They campaigned against the bad housing conditions suffered by guestworkers, and provided education to the newcomers. Together with other members of this group—including the anthropologist Rinus Penninx, who became a friend for life already at the seminary in Weert—Jan published his first texts. They were writings which actually had little to do with history, but dealt more with contemporary phenomena.3 As Phd student in Utrecht, Jan originally wanted to write a dissertation on the labor market during the Dutch Republic, but that led to unanticipated consequences, as he described later: My idea in this was, that you had to know how many people there were, how much work there was, then you had a model. Then however I found out that there was a small deviation in the model, because there were also immigrants. In other words, the demographic history was not complete. So I told Van Tijn that there should be a paragraph about immigration and emigration. But as I studied the matter further, I hit upon the so-called “migratory labor”—temporary seasonal migration. That turned out to be such a vast topic, that I got bogged down in it. I never got as far as ‘migration in general’. I had to limit myself to a subtopic of a subtopic, and then I decided to expand that to a comparison between different European countries.4

3 For example, Jan Lucassen, Rinus Penninx, Leo van Velzen, and Ab Zwinkels, Trekar beid van de Middellandse Zeegebieden naar West-Europa. Een bibliografisch overzicht (Nijmegen, 1974). 4 Leo Noordegraaf, “In gesprek met Jan Lucassen,” in Noordegraaf, Waarover spraken zij? Economische geschiedbeoefening in Nederland omstreeks het jaar 2000 (Amsterdam, 2006), pp. 213–21, at 213–4. 4 introduction

Jan’s dissertation of 1984—of which a translation was published in 1987 under the title Migrant Labour in Europe 1600–1900—was a path-breaking study which gained wide acclaim.5 Its point of departure was an examination of the survey records on migration temporaire in the Napoleonic era. The survey was undertaken by decree, issued by the authorities in the French-occupied territories in 1808 (and in German territories in 1811). Each département was required not only to document the destination of residents who left their area, but also to identify the country of origin for migrant laborers. The intention of this Napoleonic survey was to acquire a better understanding of the causes of the low number of recruits for the army. It provided an historical source which enabled Jan Lucassen to chart with great precision the foundations and dynamics of the so-called “North Sea system”—both at the start of the 19th century, but also, retro- actively, for the whole early modern period. Crucial in his approach was, first of all, the concept of a “labor cycle.” As most migrant laborers finding work in the economically highly developed coastal provinces of the Neth- erlands and Flanders typically ran their own small business as well, seasonal labor provided only part of the total family income, and combined with the work of women and children in self-employment. The North Sea system involved some 20,000 to 25,000 men each year, who labored year after year in the fields as well as building infrastructure (digging canals, constructing polders etc.). However, Jan Lucassen was not satisfied just to investigate the North Sea system at the beginning of the 19th century. He also took two additional steps in his dissertation. Firstly, he combined the survey records with further historical research, which enabled him to dig out the roots of the system as far back as the 16th century. In this way, he proved empiri- cally the crucial importance of seasonal labor in Holland throughout the whole early modern period. In addition, he studied survey data from the years 1808–1813 for the rest of Napoleonic Europe, which enabled him to view the North Sea system from a West and Southern European perspective. In this way, he clarified at one stroke how dynamic early-modern societies had been in reality, and the large impact which migration has had on European countries. Subsequently, his findings were elaborated on

5 Jan Lucassen, Naar de kusten van de Noordzee. Trekarbeid in Europees Perspektief, 1600–1900 (Gouda, 1984); English edition, slightly abridged: Migrant Labour in Europe 1600–1900. The Drift to the North Sea (London, 1987). marcel van der linden and leo lucassen 5 by other leading historians in the area of labor migration research, which flourished in the 1980s.6 The comparative perspective which Jan adopted in his dissertation stayed with him throughout his career. But while in his Phd research he had concentrated on a comparison of different regions within Europe, his historical vision broadened greatly in the course of the 1990s—taking him far beyond Europe. Jan visited India for the first time in 1995, and Indian history has never ceased to fascinate him since then. In his discussions with IISH colleagues, Jan began to realize that labor history ought not to be just a European affair, but demanded a global approach. Together with Marcel van der Linden he therefore published a Prolegomena for a Global Labor History in 1999, a brochure in which the case was made for enlarg- ing the traditional research area very substantially.7 This text was also the basis of a change in direction for IISH research activities, which has been consolidated and made definitive since the turn of the 21st century.8 The change to “globalization” became visible in all of Jan’s research pursuits, but first and foremost in his migration research. A landmark was a congress which he organized together with his brother Leo Lucassen at the NIAS (Netherlands Institute for Advanced Study in the Humanities and Social Sciences) in Wassenaar in 1993. Its aim was to integrate the sub- specialism of migration history more structurally in general history, and involve other continents in this as well. The 1997 book Migration, Migra- tion History, History9 consequently had two objectives: firstly, to redefine the terrain of migration history and abandon unfruitful dichotomies (such as “labor versus refugee migration,” or “free versus unfree migration”) and, through systematic comparisons of data on all kinds of geographical mobility, identify differences and common characteristics. Secondly, what stands out in his 1997 book—recognized among professionals as a seminal publication—is that not just leading migration historians of the Atlan- tic region contributed to it (such as Leslie Moch, Dirk Hoerder, Nancy L.

6 Such as Leslie Page Moch, Moving Europeans. Migration in Western Europe since 1650 (Bloomington, 1992). 7 Available at the IISH website: . 8 See Jan’s overview of the developments in International Institute of Social History: Annual Report 2000 (Amsterdam, 2001), pp. 51–9, available at . To celebrate the 65th anniversary of the IISH, Jan Lucassen organized a Congress in 2000 about Global Labor History, resulting in a massive volume of research studies: Jan Lucassen (ed.), Global Labour History. A State of the Art (Berne, 2006). 9 Jan Lucassen and Leo Lucassen (eds), Migration, Migration History, History. Old Para- digms and New Perspectives (Bern, 1997; third revised edition 2005). 6 introduction

Green, Robin Cohen, Aristide Zolberg, en Donna Gabaccia), but also specialists from other parts of the world, so that the Pacific, India and Latin America also entered the picture. Jan’s research about global migrations expanded in de 21th century, in close collaboration with Ulbe Bosma, Gijs Kessler, Leo Lucassen, David Feldman, and Nancy Green. Jan aspired to put migration history on the map by means of smaller-scale conferences (“small” in the sense of being relatively brief, covering limited areas of research, and particular regions). He was inspired especially by the work of the historians Patrick Manning and Adam McKeown, with whom he began to work together closely.10 An important result of this activity was the collection of essays Migration His- tory in World History. Multidisciplinary Approaches, which featured contributions from prominent geneticists, linguists and paleo-archeologists.11 Jan’s global focus was also shown by his attempt to measure migration rates for the whole of Europe during the period 1500–1900. Inspired by Manning’s cross-community approach, he designed a new typology for cross-cultural migration (together with Leo Lucassen) which enabled the calculation of the probability that an individual European citizen experienced minimally one cross-cultural migration step during his lifetime.12 In 2011, a discussion dossier (with contributions from Josef Ehmer, Leslie Moch, Jelle van Lottum and Adam McKeown) was published in the Jour- nal of Global History, in which Adam McKeown applied this new method to China, thus opening up important systematic global comparisons.13 The next phase in this project is a conference which aims to map out migrations in Africa and the Middle East. It will be staged in Rabat (Morocco)

10 Pat Manning, Migration in World History (New York and London, 2005); Adam McKeown, “Global Migration 1846–1940,” Journal of World History, 15, 2 (2004), pp. 155–89. 11 Jan Lucassen, Leo Lucassen and Patrick Manning, Migration History in World History. Multidisciplinary approaches (Leiden, 2010). The second volume, after a congress in Min- neapolis on settlement regimes (edited by Ulbe Bosma, Gijs Kessler en Leo Lucassen) will appear in 2012, as well as a third: Globalising Migration History. The Eurasian experience (16th–21st centuries), edited by Jan Lucassen and Leo Lucassen. 12 Jan Lucassen and Leo Lucassen, “The Mobility Transition Revisited, 1500–1900: What the Case of Europe Can Offer to Global History,” Journal of Global History, 4, 4 (2009), pp. 347–77. The source data for this exercise was published a year later: Jan Lucassen and Leo Lucassen, The Mobility Transition in Europe Revisited, 1500–1900: Sources and Methods (IISH Research Paper 46; Amsterdam, 2010), at . 13 Adam McKeown, “Different Transitions: Comparing China and Europe, 1600–1900,” Journal of Global History, 6, 2 (2011), pp. 309–19; Jan Lucassen and Leo Lucassen (2011). “From Mobility Transition to Comparative Global Migration History,” Journal of Global His- tory, 6, 2 (2011), pp. 299–307. marcel van der linden and leo lucassen 7 in the Spring of 2012, in association with Patrick Manning and his World History Center at the University of Pittsburgh.

Guilds and brickmakers

A second theme to which Jan paid much attention as historian were corporate organizations, in particular the guilds. Often people think that the guilds were exclusivist professional organizations, aiming mainly at mutual security and the restriction of competition, and thus, that they were typically “European-feudal” institutions, incompatible with capitalism. Partly because of this interpretation, and because of their undeniable striving to maintain exclusive monopolies, it seemed to follow that the guilds must have resisted immigration. But Jan and his close colleague Piet Lourens convincingly overturned this traditional interpretation. Cre- ating an extensive database, they inventorized all the Dutch guilds since 1500, and proved that the guilds were still alive in the capitalist Northern Netherlands even in the 19th century, although their religious character had gradually vanished. Moreover, the guilds reached their maximum number of members only in the year 1770.14 Finally, they showed that the attitude of the guilds toward immigrants was not at all hostile: “Instead of the exclusion of foreigners [. . .] they opted for an integration as quickly as possible. Apparently the gildenbroeders (guild brethren) believed, that they could control this uncertain element better in this way.”15 As Jan has often argued since, the guilds in reality represented important social institutions with which, among other things, the early capitalist labor markets could indeed be regulated. After he had refuted the idea that the guilds were “typically feudal-European organizations,” Jan subsequently also published comparative studies with other scholars, first for the Low Countries, and later for Europe, Asia and Africa.16

14 Piet Lourens and Jan Lucassen, “Ambachtsgilden in Nederland: een eerste inventa- risatie,” NEHA-Jaarboek voor economische, bedrijfs- en techniekgeschiedenis, 57 (1994), pp. 34–62. The database will soon be available online via de website van het IISH. 15 Piet Lourens and Jan Lucassen, “De oprichting en ontwikkeling van ambachtsgilden in Nederland (13de–19de eeuw)”, in Catharina Lis and Hugo Soly (eds.), Werelden van ver- schil. Ambachtsgilden in de Lage Landen (Brussels, 1997), pp. 43–77, at 53. 16 Maarten Prak, Catharina Lis, Jan Lucassen, and Hugo Soly (eds.), Craft Guilds in the Early Modern Low Countries: Work, Power and Representation (Aldershot, 2005); Jan Lucas- sen, Tine de Moor and Jan Luiten van Zanden (eds.), The Return of the Guilds (Cambridge, 2008). 8 introduction

Closely connected to his fascination with migration history is a third area of research which has preoccupied Jan over the years: the history of brickmaking. In his Phd research, Jan had discovered that from the 18th to the early 20th century, large numbers of male seasonal workers, organized in hierarchical groups, came to the Netherlands from the German Lippe- Detmoldt region, to produce bricks. Over a number of years, Jan reconstructed the activities and experiences of this special migrant group with Piet Lourens—the workers’ backgrounds in Lippe-Detmoldt; the composition of the workteams; how they retained contact in the Netherlands with the families they left behind; what salary differences existed within the group; and even the family ties there were among them.17 Connected to this research was another project: an analysis of messages written by protestant church ministers, who accompanied the tichelaars (brickmakers) on their travels and provided spiritual edification; these texts contained a treasure of information about the migrant laborers, their culture and their work.18 When he “discovered” India in the second half of the 1990s, Jan noticed that local brickmakers still used the same techniques as the German Lippers had before 1900, even although the Indian workers’ activities of course took place in the completely different context of a caste society. In the course of the last fifteen years, Jan has spent many weeks, if not months, in North India pursuing anthropological field research to “study history via the present” in a comparative way. Needless to say, he has subsequently also placed the Indian brickmakers in historical perspective with the aid of extensive archival research, carried out especially in New Delhi. En passant, he discovered that already in the middle of the 19th century the Indian brickmakers had staged very large strikes—decades earlier than the Indian working class had supposedly first begun to assert itself, according to mainstream Indian historiography.19

17 Piet Lourens and Jan Lucassen, Lipsker op Groninger tichelwerken. Een geschiedenis van de Groningse steenindustrie met bijzondere nadruk op de Lipper trekarbeiders 1700–1900, (Groningen, 1987); Piet Lourens and Jan Lucassen, Arbeitswanderung und berufliche Spe- zialisierung. Die lippischen Ziegler im 18. und 19. Jahrhundert (Osnabrück, 1999); Lourens and Lucassen, “Karrieren lippischer Ziegler: Das Beispiel Delfzijl,” Lippische Mitteilungen aus Geschichte und Landeskunde, 76 (2007), pp. 63–80. See also the website based on the work of Piet Lourens en Jan Lucassen . 18 Albin Gladen, Antje Kraus, Piet Lourens, Jan Lucassen, Peter Schram, Helmut Talako, Gerda van Asselt (eds.), Hollandgang im Spiegel der Reiseberichte evangelischer Geistlicher. Quellen zur saisonalen Arbeitswanderung in der zweiten Hälfte des 19. Jahrhunderts (Mün- ster, 2007), 2 vols. 19 Jan Lucassen, “La fabricación de ladrillos en Europa Occidental y la India: un intento de historia social comparada del trabajo,” Historia Social, 45 (2003), pp. 3–33; idem, “Brick- marcel van der linden and leo lucassen 9

Brickmaking is traditionally an activity which does not take place in cities, but in rural areas, along rivers or at other places where suitable clay is found. It was among other things this observation which led Jan to the insight that urban workers have had privileged attention from historians, and still do. In 1994, he published a path-breaking essay titled “The Other Proletarians,” in which he argued that non-urban proletarianization processes deserve much more attention: “These in fact encompass a wide variety of labor processes such as mining, infrastructure construction, work in the transport sector—mainly at sea—and even work in the armed forces.”20 One of these groups had already claimed Jan’s attention much earlier: the seamen. In 1980, he prepared a source publication with a Leiden lecturer in maritime history, Jaap Bruijn, about the labor and living conditions of seamen aboard the ships of the Dutch East India Company (the V.O.C.).21 Later Jan would concentrate more on the early modern labor market aspects, both in Europe and in Asia.22

An unended quest

In Jan’s eyes, labor history is an infinitely variegated research terrain—a terrain much broader than the traditional labor and working-class historians previously imagined. He demonstrated convincingly, that even an innocent personal hobby of his—coin collecting—could be fruitful

makers in Western Europe (1700–1900) and Northern India (1800–2000): Some Compari- sons,” in idem, Global Labour History, pp. 513–71; idem, “The Brickmakers’ Strikes on the Ganges Canal in 1848–1849,” in Rana P. Behal and Marcel van der Linden (eds), India‘s Labouring Poor: Historical Studies c. 1600–2000 (New Delhi, 2007), pp. 47–83. 20 Jan Lucassen, “The Other Proletarians: Seasonal Labourers, Mercenaries and Miners,” in Catharina Lis, Jan Lucassen, and Hugo Soly (eds.), Before the Unions: Wage Earners and Collective Action in Europe, 1300–1850, Supplement 2 to International Review of Social His- tory, 39 (1994), pp. 171–94, at 171. 21 Jaap R. Bruijn and Jan Lucassen (eds.), Op de schepen der Oost-Indische Compagnie. Vijf artikelen van J. de Hullu, ingeleid, bewerkt en voorzien van een studie over de werkgele- genheid bij de VOC (Groningen, 1980). 22 Paul van Royen, Jaap R. Bruijn, and Jan Lucassen (eds.), “Those Emblems of Hell”? European Sailors and the Maritime Labour Market, 1570–1870 = Research in Maritime His- tory, 13 (1997); Jan Lucassen, “A Multinational and its Labor Force: The Dutch East India Company, 1595–1795,” International Labor and Working-Class History, 66 (Fall 2004), pp. 12–39; Jelle van Lottum and Jan Lucassen, “Six Cross-Sections of the Dutch Maritime Labour Market: A Preliminary Reconstruction and its Implications (1610–1850),” in Richard Gorski (ed.), Maritime labour in the Northern Hemisphere c. 1750–1950 (Amsterdam, 2007), pp. 13–42; Mathias van Rossum, Lex Heerma van Voss, Jelle van Lottum and Jan Lucassen, “National and International Labour Markets for Sailors in European, Atlantic and Asian Waters, 1600–1850,” in Maria Fusaro and Amélia Polónia (eds.), Maritime History as Global History (St. John’s, 2010), pp. 47–72. 10 introduction in the pursuit of labor history. In earlier centuries, Jan observed, wages were typically paid in relatively small coins, while big coins were used for the transactions of the rich. This fact led him to hypothesize in the 1990s that the relative volume of circulation of small coins could be used as an indicator of the amount of wage labor which must have existed in a given region. After testing this remarkable hypothesis with a pilot study for the Netherlands,23 Jan organized an international conference on the topic in 2002. On the strength of the papers presented there, with case studies of Spanish America, Europe, China and India, Jan could formulate his hypothesis as follows: “the greater the quantity of copper and small silver coins produced and the more the denominations that circulated simultaneously, the greater the chance that a substantial part of the population depended on free labor, remunerated by regular wage payments.”24 Later he would elaborate this “rule” further.25 While Jan Lucassen continually pursued new research themes in his career, he also retained the social engagement he had in his student days. In particular, the increasingly heated discussions about Islam, and the supposed “threats” for Western society by immigrants, have been of great concern to him. In 1985, he therefore published a popular-scientific book with his old companion Rinus Penninx, titled Newcomers, which went through several reprints and editions; and, more or less as a sequel to this book, he published a second book in the same vein with his younger brother Leo Lucassen, titled Winners and Losers, which upon its publication in the Netherlands in 2011 gained considerable media attention.26 Jan will remain associated with the International Institute of Social History also after his retirement, as honorary fellow. Indeed, his research is by no means at an end. His plans for the near future are ambitious—

23 Jan Lucassen, “Loonbetaling en muntcirculatie in Nederland (1200–2000),” Jaarboek voor Munt- en Penningkunde, 86 (1999), pp. 1–69. 24 Jan Lucassen, “Introduction: Wages and Currency, 500 bce–2000 ce,” in Lucassen (ed.), Wages and Currency. Global Comparisons from Antiquity to the Twentieth Century (Bern, 2007), pp. 9–58, at 55. 25 See especially Jan Lucassen, “Coin Production, Coin Circulation, and the Payment of Wages in Europe and China 1200–1900,” in Christine Moll-Murata, Song Jianze, and Hans-Ulrich Vogel (eds.), Chinese Handicraft Regulations of the Qing Dynasty: Theory and Application (Munich, 2005), pp. 423–46; Jan Lingen and Jan Lucassen: “The ‘Mansúri’ or ‘Munsooree Paisa’ and its Use: Combining Numismatic and Social History of India, c. 1830–1900,” Numismatic Digest, 31 (2007), pp. 187–220. 26 Jan Lucassen and Rinus Penninx, Nieuwkomers (Amsterdam, 1985, 1994, 1995, 1999); English translation: Newcomers. Immigrants and their Descendants in the Netherlands 1550–1995 (Amsterdam, 1997); Jan Lucassen and Leo Lucassen, Winnaars en verliezers. Een nuchtere balans van vijfhonderd jaar immigratie (Amsterdam, 2011). marcel van der linden and leo lucassen 11 he intends among other things to write a comprehensive treatise with the deceptively modest working title “Work: A Concise History”. In addition, he is collaborating with his Indian friend and colleague Sabyasachi Bhattacharya to produce a book of photographs about labor in India. The essays in the present festschrift, written with appreciation and respect for a truly impressive historian, intend to encourage and inspire Jan in his future endeavors. In each of the essays, the influence he has had on his fellow historians is evident—an influence which, we do not doubt, will endure.

Contributions to this volume

In the first essay of this volume, Karin Hofmeester describes how the refractory quality of a diamond is enhanced by the addition of numerous facets. This observation is in our view, also an excellent metaphor to sum up the contributions to our festschrift. The different contributors have— each according to their own talent and field of expertise—all elaborated on facets of Jan Lucassen’s oeuvre; together, they provide a worthy, color- ful and, we think, brilliant reflection of its light. That is why we selected Hofmeester’s work as the opening essay for this book. In her story—the first in a group of four about the production of specific commodities or services—she surveys the lengthy history of globalization which the diamond business experienced, with its continually shifting international centers of production, trade and consumption. In these shifts, many different factors have played a role, ranging from geological discoveries to labor costs and transport costs, and changing consumer preferences. Hofmeester reveals the nature of the labor relations which were involved in these relocations, and how they kept changing from the 17th century onward. Willem van Schendel, too, focuses on the production of a “global commodity”: indigo, the blue dye which was an extraordinarily profitable product until its replacement from the end of the 19th century by a synthetic alternative. In his essay, Van Schendel provides a vivid description of the harsh working conditions involved in idigo production—both among the peasants who cultivated the plants and among the factory laborers who subsequently manufactured the dye. He concentrates on Bengal, where indigo production flourished in the 19th century, after much of the industry had shifted—partly in response to the slave revolt in Saint Domingue—from the Caribbean to Asia. 12 introduction

A very different form of labor is analyzed by Chitra Joshi in her study of the dauriyas (road runners) who, in 19th century British India, carried public and private mail on foot across large distances. In lively detail, Joshi narrates the recruitment and working conditions of the runners, the dangers on the road, and especially how the employers tried to rationalize the labor process by means of the clocking and recording of postal speed, and by marking the stages where runners were changed; she shows how a new, “modern” time regime was introduced in this way, and a set of new ideals was constituted: “principles of speed and regularity, of the idea of public commitment and public good.” Ratna Saptari takes us to Central and East Java, and examines de nature of work and labor relations in the cigarette industry there during the first years of the Great Depression. She analyzes the shifting parameters of competition between factories, workshops and home-working; the multi- level sub-contracting arrangements; the changes in the work sites; and the modes of resistance of the cigarette workers, in the context of the changing policies and ideological notions created by the colonial state during the crisis years. In the second cluster of contributions, labor relations are also the focus, but here, labor categories are examined from the perspective of specific occupational groups. Danielle van den Heuvel devotes her contribution to a very diverse set of people which until now have been largely ignored by historians: informal street vendors in pre-industrial Europe and elsewhere. After an overview of the sparse historiographical literature on this topic, and a discussion of insights which can be gained from the literature about street selling in the contemporary Global South, Van den Heuvel probes the identities, activities and backgrounds of early-modern European street traders, opening up a new field of research. Likewise, Lex Heerma van Voss heeds Jan Lucassen’s plea to broaden labor history, and endeavors to study an occupational group which is usually not considered part of the working class: prostitutes in Buenos Aires, Bombay, Mexico City, Nairobi and San Francisco around 1900. Focusing on their working conditions and survival strategies, he shows that both male and female “demographic surpluses,” respectively increasing demand and supply, were conducive to large numbers of prostitutes. Heerma van Voss compares the diverse forms of trafficking, as well as variations in labor and kinship relations, and in this way is able to reveal a number of clear patterns. A third occupation group which was traditionally never studied by labor historians is featured in Erik-Jan Zürcher’s contribution. His contention is marcel van der linden and leo lucassen 13 that army activities are also a type of work, and therefore that the military ought to be regarded as workers—albeit a special category of workers. Basing himself on the research project “Fighting for a Living” which he coordinates, Zürcher tries to discover regularities in the different systems of recruiting and employing soldiers which were practised across the last five centuries by state-directed land armies in Europe, the Middle East, India and China. After sketching the broad spectrum of possible labor relations, Zürcher explores the most important determinants of military employment, in order to support his conclusion that it is possible to formulate an explanation—at this stage still somewhat provisional—of the various historical developments. White-collar workers are the topic of concern to Karel Davids. He raises the question of whether the Dutch East India Company (the V.O.C., 1602– 1795) just like the British East India Company (E.I.C., 1600–1874) developed an age-related salary structure; and whether the white collar workers of the V.O.C. enjoyed similar opportunities to make a career, earning more with age like their British colleagues. He concludes that the salaried employees of the V.O.C. “were not just doing well, they were doing extremely well.” Yet, his analysis shows that the V.O.C. salary structure was never as much age- or experience-related as it was in the E.I.C., partly because perquisites played a larger role in remuneration. The third cluster of essays is concerned with geographical and social mobility. Nearly half of all the V.O.C. employees were, in fact, non-Dutch. Jaap Bruijn and Femme Gaastra investigate the career opportunities of these foreigners, concentrating on two groups: commanders of the trading vessels and junior merchants in the overseas commercial networks. At the hand of a database still under construction, they reach the conclusion that the opportunity of foreigners to become commanders was quite great, but foreigners were comparatively rare among junior merchants. Factors which played a role in this difference included, among other things, the relative size of the supply of Dutch competitors, the number of available positions, and integration in informal networks. In their brief but significant contribution, Jan Luiten van Zanden en Maarten Prak try to improve the estimates of immigration into Holland from the 16th to the 18th century originally made by Jan Lucassen. Using basic demographic data they obtain the tentative result, that the urban natural decrease was lower than previously assumed. With his analysis of European sailors in the long 18th century, Jelle van Lottum embarks on an exploration of the effects of population movements on pre-industrial economic development. Using a dataset derived 14 introduction from the “Prize Papers” in the London National Archives (consisting of letters confiscated when Dutch ships were pirated by the English navy), Van Lottum shows that it is important in this inquiry to distinguish between sedentary and non-sedentary migrants. It is probable that their influences on labor productivity differed, but it was—sometimes more and sometimes less—mainly positive. Richard Unger in his reconstruction of the migration of German Rhinelanders to Pennsylvania, through the port of Rotterdam, in the eighteenth century contributes in various ways to our understanding the nexus work and migration. While criticizing simplistic economic assumptions about income differentials as the prime cause for people to migrate, he argues that rational choices by migrants were impossible. At the same time he stresses the crucial role of British and Dutch merchants, based in Rotterdam, in the transport business of transmigrants, linking the German hinterland of this Dutch port to North America. The work and involvement of these merchants (often migrants themselves) largely determined the volume, direction and business cycle of German workers who wanted to settle in North America, thus exerting a big influence on the shaping of the labor market in the North American colony. In the fourth cluster of essays, labor relations are contextualized in different ways. Catharina Lis and Hugo Soly discuss the labor laws which public authorities formulated in Europe from the 13th century: their defi- nitions of the “labor problem” as well as the social relations and balances of power operating in the background. Because Lis and Soly take a comparative perspective, they are able to refute a well-known interpretation, i.e. that the labor shortage, due to the effects of the Black Death, was the most important cause of late-medieval labor statutes. Countrywide labor legislation favoring employers was enacted already before the epidemic occurred, and was promoted by diverse factors such as the strength of states, the need for tax income, the striving to limit intra-rural and rural- urban mobility, the relative significance of wage-labor, and the convergence of agricultural interests. Elise van Nederveen Meerkerk queries the relationship between subsistence labor and paid labor performed by women. She counters the often- heard thesis that female labor force participation was low in the Dutch Republic because economic progress made women’s work less necessary, and she makes a credible case for the view that from the 17th century until the mid-19th century the percentage of married women only performing household labor was in fact very low; child labor also frequently occurred. marcel van der linden and leo lucassen 15

In other words, the “bourgeois ideal of domesticity” had, in this period, hardly reached the lower social classes yet. Gijs Kessler takes up a theme to which Jan Lucassen already devoted much attention in his dissertation: the fact that, in pre-industrial and early- modern households, wage labor was usually combined with other survival methods such as agriculture or handicrafts. Focusing on Russian peasant migrant laborers (otkhodniki), Kessler shows how, since the 17th century, the “modernization” of the economy combined with an increase of wage labor as an element within composite household economic strategies. He concludes that this trend is perhaps a better indicator of economic modernization than outright household specialization in wage labor as such. Ulbe Bosma offers a comparative analysis of the sugar-cane industries in Cuba, the French and British Caribbean, and the Netherlands Indies during the 1930s. In doing so, he makes the connection between different labor market relationships, the intensity of labor protest and the discovery of the discipline of development economics. He distinguishes three political trajectories, which were to an important extent determined by the relative abundance of labor and by the ethnic composition of the rank and file of sugar workers; these factors help us not only to understand why the chances for collective action in de French and British Caribbean were smaller than in Cuba, but also significantly larger than in Java; they also clarify why early development economists such as J.H. Boeke and Arthur Lewis followed divergent paths. Somewhat distinct from the earlier essays there is, last but not least, a very interesting contribution by Jaap Kloosterman. He concludes our com- memorative volume with an exploration of a subject which has always been close to Jan Lucassen’s heart: the origins of libraries and archives in the field of social and economic history. Kloosterman details the emergence of the first collections of documents about labor relations, and in particular about workplace safety; the pioneering role of the Parisian Musée Social, founded in 1894, which internationally served as an inspiration for other similar initiatives; and the formation of the first economic/ historical libraries and labor movement collections in Europe and the Americas. Faced with these seventeen historical investigations, we imagine a layman might well be daunted by the diversity of details mustered, or puzzled by the highly specialized research questions being pursued. We should therefore emphasize, in all modesty, that these essays do indeed represent only facets of work in progress; they are part of much larger 16 introduction research programs which will no doubt result in more definitive publications. Jan Lucassen has never pretended to reach “all the answers to all of the questions,” and neither do we. But by probing the historical evidence from many different vantage points and sharing our results, wrong inter- pretations can nevertheless be disarmed; a broad consensus can emerge about what happened in the global history of labor. Jan Lucassen endeav- ored to lead the way by personal example. Above all, he showed us the astonishing amount of new knowledge we can obtain about the past, if indeed we are prepared to research our questions conscientiously. work processes

Working for Diamonds from the 16th to the 20th Century1

Karin Hofmeester

Diamonds have a long history of globalization—a history characterized by only a few (though often changing) world centers of production and trade, strongly interconnected and linked to consumer markets. India has played a major role in this history. The first diamonds were mined and finished there, and nowadays more than 90% of all diamonds are cut and polished in this subcontinent. In between—from the 16th century until the second half of the 20th century—various European cities in succession became major diamond finishing centers, obtaining their rough diamonds from various parts of the world. Diamonds are not only a global commodity par excellence; they are also a luxury commodity pur sang. Diamond research has therefore focused mainly on the diamond trade, consumption and the associated consumer culture. Diamond production and the effects this had on labor are often neglected. Since a lot of work is involved along the route “from mine to finger,” and since this commodity chain has an age-old global character, diamond production offers us a perfect exemplar for “how to do” global labor history. My main aim in what follows is to show the global intercon- nectedness of diamond production, trade and consumption and its effects on labor relations world wide. The focus will be on India and Europe, although I will also touch on other parts of the world.

Diamond trajectories

Already in antiquity, Indians traded diamonds with the Romans, whose trading routes exported the stones as far away as China, where they were mainly used as tools to drill holes in jade.2 In the late Middle Ages, we

1 This article is part of my research project Luxury and Labour: A Global Trajectory of Diamond Consumption and Production, 16th to 19th Century, funded by the Fritz Thyssen Stiftung. 2 Also later on, no interest in diamonds for jewellery seemed to have developed. Craig Clunas, Superfluous Things: Material Culture and Social Status in Early Modern China (reprint, Honolulu, 2004) mentions mainly jewellery made of jade, pearls, ivory and silver. 20 karin hofmeester see a reintroduction of diamond consumption in Europe. The first diamonds reaching Europe from India in the early Renaissance were bought and sold by individual merchants, who transported the stones via cara- van routes (using parts of the silk route). They sold their merchandise in Venice, where most of the stones were finished. In the 16th century, the Portuguese discovered shorter sea routes to India, and they became the most important diamond dealers, rerouting the trade from India (often Goa) to Lisbon and from there to Antwerp, where a polishing industry started to flourish. At that time, Antwerp was already the principal market where the Portuguese sold their spices and bought copper, timber, grain and other essentials for the Asian trade and for the Iberian economies.3 The Dutch East India Company (VOC) entered the diamond trade at the beginning of the 17th century, buying stones from local merchants, and sending them off from their factory on the Coromandel Coast. Although the VOC for some time made large profits in the diamond trade—by forbidding its employees to engage in the much more effective private trade—it lost its position to the English traders who soon overtook both the Portuguese and the Dutch. Amsterdam nevertheless became an important diamond trading and finishing center in the 17th century. At first the East India Company (EIC) had a monopoly on the trade of diamonds (and many other commodities). However, after the 1660s it not only allowed company servants to buy small amounts of stones, but also permitted private traders to import Indian diamonds via EIC officers.4 The major “export hub” for diamonds in India at that time was Madras, and London became the major market for rough diamonds, although an extensive diamond finishing industry never developed in England. When large diamond deposits were found in Brazil in the late 1720s, the Portuguese—after a short “free-for-all” period—claimed a monopoly both on the exploitation of the mines, and on the transport and sale of rough diamonds. This made Lisbon the port where the rough diamonds first arrived. Since however most large merchants were British, and a lot of stones reached the British capital illegally, London remained the center

See also http://www.nytimes.com/2005/05/31/business/worldbusiness/31diamonds.html published in 2005, last assessed on 12 April 2011, where a young Chinese woman states that jade (‘popular for centuries in China’) is outdated and diamonds are more convenient and lightweight to wear. “They make you look sharp.” 3 James C. Boyajian, Portuguese Trade in Asia under the Habsburgs, 1580–1640 (Balti- more, 1993), pp. 135–6. 4 Søren Mentz, The English Gentleman Merchant at work. Madras and the City of London 1660–1740 (Copenhagen, 2005), pp. 116–7. working for diamonds from the 16th to the 20th century 21 of the rough trade. The indirect trading route between Rio de Janeiro and London via Lisbon eventually ended the Indo-European trade in rough diamonds, although the mining and finishing of diamonds in India continued, as we shall see. Globalization increased further after 1870, when huge diamond deposits were discovered in South-Africa. The De Beers Company soon established a mining and rough trade cartel, settling its Central Selling Organisation in London. Prices fell, the diamond democratized, and a large demand led to a large finishing industry which started to sell its products also in the United States. In the course of the 20th century, new deposits were discovered in Africa, Russia, Australia and Canada, and new consumer markets were explored. Yet the abundance of relatively cheap rough diamonds became one of the factors that shifted the finishing industry back to its original cradle, India.

Labor in the Indian diamond mines

India’s Golconda mines were world-famous in the early modern world. According to the Oxford dictionary, Golconda is synonymous for “a source of wealth, advantages, or happiness.” The mines were actually not located in Golconda itself—the diamonds were only sold there—but situated further East in and around Kollur, in the basin between the Godavari and Kristna river. Other fields on the Deccan plateau, in what is now Andhra Pradesh, could be found around Kurnool, between the Kristna and Penner river, and more South on the Penner River in and around Cuddapah. A second group of mines were dug in and around Sambalpur on the Maha- nadi river, in present day Orissa. The third group was located in Panna, in present-day Madhya Pradesh. There are several detailed 17th century European descriptions about the Indian mines. The earliest description is by Jaques de Coutre, a Bruges- born diamond merchant who went to Goa with his brother Joseph. Goa was ruled by the Portuguese since 1510, and functioned as a true global “trade hub.” The De Coutres settled there as merchants in precious stones.5

5 De Coutre’s son wrote down his fathers memoirs in 1640. The original manuscript Vida de Jaques de Coutre, natural de la ciudad de Brugas, condado de Flandes. Puesto en la forma que está por su hijo, Don Estevan de Coutre, Madrid 1640 is kept in the National Library in Madrid. For a translation see Johan Verberckmoes and Eddy Stols (eds), Aziatische omzwervingen. Het levensverhaal van Jaques de Coutre, een Brugs diamanthande- laar 1591–1627 (Berchem, 1988). 22 karin hofmeester

Figure 1. Diamond fields in India, from Precious Stones, Volume I by Max Bauer, published by Dover Publications in 1968 as re-issue of an original from 1904, p. 144. Courtesy of Dover Publications Inc. working for diamonds from the 16th to the 20th century 23

In the period from 1611 to 1618, Jaques visited several mines in the Deccan plateau, writing the most detailed report on the mine of Ramallakota, near Kurnool situated in the Golconda Sultanate, which was then ruled by the Qutb Shahi dynasty. A second description is from William Meth- wold, an English merchant in the service of the EIC who was engaged in buying diamonds.6 In his ethnographic sketch of several kingdoms along the East coast, he included an account of a mine very near the Kollur mine which he visited sometime in the 1620s.7 This mine was situated in the Bijapur sultanate, which at the time of Methwold’s visit was ruled by sultan Ibrahim Adil Shah II. A third and perhaps most well-known description is from Jean-Baptiste Tavernier, a French jeweler who traveled many times to India in the 1640s, 1650s and 1660s and bought and sold diamonds and other gemstones, amongst others to Louis XIV.8 He provides very detailed descriptions of the Ramallakota mine which he visited in the 1640s. A fourth sketch is from Pieter de Lange, a Dutch doctor and chief factor (merchant) in the service of the VOC, who later became governor of Masulipatnam, one of the factories of the Company on the Coromandel Coast. In 1663, the VOC sent him to the Kollur mine to buy diamonds. Though he did not succeed in making good deals, he did provide a lot of information about the exploitation of the mines.9 A fifth report is from Henry Howard, Earl of Marshall and 6th Duke of Norfolk who presented his findings about 23 mines in the Golconda sultanate and 15 mines in the Bijapur sultanate—including Kollur—to the Royal Society of England in 1677. Finally, there is a description of the diamond mines of Kollur and nearby Gollapilly, written in 1679 by Streynsham Master, an EIC official who was at that time an agent of Madras.10 From the descriptions of these

6 Mentz, English Gentleman Merchant, p. 116. 7 Methwold’s Relation was re-issued in 1931 as part of the volume of W.H. Moreland (ed.), Relations of Golconda in the Early Seventeenth Century (London, 1931). On page 33 of this edition, Methwold states that the mine was closed in 1622. His Relation was first published in 1626. 8 Tavernier also travelled trough the Levant and Persia. His original accounts were published in French: Le Six Voyages de J. B. Tavernier en Turquie, en Perse et aux Indes (Paris, 1676). 9 According to Tavernier, Pieter de Lange first worked as a surgeon at the court of the king of Golconda till 1656. J-B. Tavernier, Travels in India. Translated from the original French Edition of 1676 by V. Ball (reprint, New Delhi, 1989), 2 vols, I, pp. 240–3. For the report, see Pieter van Dam, Beschrijvinge van de Oostindische Compagnie, book 2, vol. I, (The Hague, 1932), pp. 176–81. 10 “A Description of the Diamond-Mines, as it was Presented by the Right Honour- able, the Earl Marshal of England,” Philosophical Transactions of the Royal Society, 12, 136 (25 June 1677), pp. 887–917, at 908. [i.e. Henry Howard] R.C. Temple (ed.), The Diaries of 24 karin hofmeester merchants and Company officials we can obtain a good picture of the activities in and around the mines. Usually the emperor, king or sultan who had the mines on his territory also owned the mines. He could choose whether or not he wanted to farm out the mine, and if so to whom.11 The ruler leased the mine to the highest bidder, who could be a native entrepreneur (a member of the goldsmith caste—kamsali—for example) but also could be a foreigner, like the Portuguese Albaro Mendez, a member of a wealthy Sephardic family from Lisbon who was trained as a goldsmith in Antwerp and then sent to India in 1545 to acquire diamonds.12 Governors—revenue farmers—acted as intermediaries between the king on the one hand, and the merchants (“adventurers”) who actually commissioned miners to dig for diamonds on the other.13 These merchants were often Banians from Gujarat, who maintained a tight grip on the diamond trade and a wide trade network.14 According to Henry Howard: The Merchants are the Banians of Guzzarat, who for some Generations have forsaken their own Country to take up the Trade, in which they have had such success, that’ tis now solely engros’d by them; who corresponding with their Country-men in Surrat, Goa, Colconda, Visiapore, Agra and Dillee, and other places in India, furnish them all with Diamonds.15 We know that the early modern European travellers tended to apply the term “Banian” to all Hindu merchants and traders, using it as an occupational category, rather than to signal caste affiliation. Today’s scholars however also explicitly mention the Gujarati Bania diamond traders as

Streynsham Master 1675–1680 and Other Contemporary Papers Relating Hitherto, vol. II: The First and Second “Memorialls” 1679–1680 (London, 1911), pp. 113–4 and pp. 172–5. 11 According to Henry Howard, several Raja’s and sultans in South India had (some of) their mines dug out only privately, “A description of the Diamond-Mines,” pp. 907–9. 12 For the highest bidder, see Mentz, English Gentleman Merchant, p. 111; for the goldsmith, see Methwold in Moreland (ed.), Relations of Golconda in the Early Seventeenth Cen- tury, p. 31, as well as Narahari Gopalakristnamah Chetty, Manual of the Kurnool District in the Presidency of Madras (Madras, 1886), p. 94; for the Portuguese, see Verberckmoes and Stols (eds), Aziatische omzwervingen, p. 177 and Lucien Wolf, “Jews in Elizabethan Eng- land,” Transactions of the Jewish Historical Society of England, 11 (1928), pp. 1–91, at 24–5. 13 Kanakalatha Mukund, “Mining in South India in the 17th and 18th Centuries,” Indica, 52–53 (1991), pp. 13–52, 17. 14 Tavernier, Travels in India, II, p. 47, Report by Pieter de Lange as included in van Dam, Beschrijvinge van de Oostindische Compagnie, p. 179, “A Description of the Diamond- Mines,” p. 915. 15 “A Description of the Diamond-Mines,” p. 915. working for diamonds from the 16th to the 20th century 25 having an important role in the trade, amongst others as middle men between the mines and the European traders.16 In the Ramallakota mine, merchants had to pay the king a fee of two pagodas a day per fifty miners; this money was collected by the governor of the mine.17 In the mine of Kollur, the governor had more responsibilities than that: here, the governor provided the merchants with workers, and made sure they were equipped with tools. In this mine, the merchants had to pay the governor a fee per worker, part of which went to the king and part to the miners.18 Rent was only one part of the king’s revenues from the mine; he also received 2% of all diamond purchases and sales.19 If the miners found a diamond weighing more than ten carats, they had to hand it over to the governor of the mine, who in turn had to transfer it to the king. 20 Notwith- standing the presence of overseers—hired by merchant commissioners— who supervised the mines, big diamonds were often excavated without this being reported to the governor, and directly sold on to “foreigners” for half the price the merchant would ask. If this was discovered, both the miner and the merchant were put to death, though De Coutre escaped this fate because he was a “Frank” (a European).21 De Lange reported about the Kollur mine that this rule “was not applied too strictly.” If the miners found a big stone and reported it, they received a small bonus.22 Though the miners worked in small units, the total number of miners in a field could be enormous. De Coutre counted 50,000 men, women and children in Ramallakota, while Methwold counted some 30,000 “souls” working in the diamond fields near the Kollur mine. About twenty years

16 Makrand Mehta, Indian Merchants and Entrepreneurs in Historical Perspective (Delhi, 1991), p. 35. For the present day observance, see R.J. Barendse, Arabian Seas 1700–1763, vol. II: Kings, Gangsters and Companies (Leiden, 2009), p. 690. Also see M.N. Pearson, “Ban- yas and Brahmins: Their Role in the Portuguese Indian Economy,” in Idem, Coastal West- ern India: Studies from the Portuguese Records (New Delhi, 1972), pp. 93–115, 104. 17 Tavernier, Travels in India, II, p. 46. 18 Report by Pieter de Lange as included in van Dam, Beschrijvinge van de Oostindische Compagnie, p. 180. 19 Tavernier, Travels in India, II, p. 46. 20 Different authors mention different numbers, see Verberckmoes and Stols, Azi- atische omzwervingen, p. 173; Methwold in Moreland, Relations of Golconda in the Early Seventeenth Century, p. 33 Tavernier, Travels in India, II, p. 47, Report by Pieter de Lange as included in van Dam, Beschrijvinge van de Oostindische Compagnie, p. 181. 21 Verberckmoes and Stols, Aziatische omzwervingen, p. 174. 22 Report by Pieter de Lange as included in van Dam, Beschrijvinge van de Oostindische Compagnie, p. 181. For the bonus see Tavernier, Travels in India, II, p. 47. 26 karin hofmeester later, Tavernier signalled no less than 60,000 men, women and children at work in the Kollur mine.23 These numbers are astonishingly large, and can be explained by the very labor-intensive production process. Mining methods differed somewhat from mine to mine, depending on the type of soil that covered the diamantiferous stratum, as well as on the composition of the stratum itself. Men dug the pits and took out the earth (and superfluous water) which was carried away in baskets by women and children. Indian miners could not dig below the water table, pits therefore varied from 4 to 14 feet deep.24 These pits were not supported with timber, as in Europe Methwold noted, and De Coutre personally witnessed the collapse of a mine after heavy rain falls, taking the lives of at least 150 people.25 The miners used no “pul- lies and such like devices” but sat on top of each other, and passed on the baskets.26 The women and children carried the soil to a piece of ground, surrounded with a wall, pierced with holes. They soaked the soil with water, often brought in from quite far away, which would leave the area through the holes in the wall; subsequently the soil would be laid to dry in the sun on a flattened piece of ground, to “loosen” the diamonds from the surrounding earth. Finally, the earth would be sieved and searched to discover the diamonds. If the soil was too hard to extract the diamonds, miners would smash the lumps of earth to reveal the gemstones.27 There was not only a division of labor by gender and age; there was also a labor hierarchy amongst the miners. The most experienced of them had a very important role in the mining process as “searchers,” since they could indicate from experience the places where diamonds could most likely be found. Though their status was high, they do not seem to have been paid extra for their skills.28 According to De Coutre, the miners were the poorest of the poor. Tavernier thought they were peasants, who went

23 Verberckmoes and Stols, Aziatische omzwervingen, p. 172, Methwold in Moreland, Relations of Golconda in the Early Seventeenth Century, p. 31, Tavernier, Travels in India, II, p. 59. 24 “A Description of the Diamond-Mines,” p. 910; Temple, Diaries of Streynsham Master, p. 174, Tavernier, Travels in India, II, p. 60. 25 Verberckmoes and Stols, Aziatische omzwervingen, p. 174. 26 Methwold, Relations of Golconda, p. 32. 27 Tavernier, Travels in India, II, pp. 60–1. 28 For the payment see Tavernier, Travels in India, II, p. 46, the expertise of the experienced miners is also signalled by Methwold in Moreland, Relations of Golconda in the Early Seventeenth Century, p. 31 and in the Report by Pieter de Lange as included in van Dam, Beschrijvinge van de Oostindische Compagnie, p. 180. working for diamonds from the 16th to the 20th century 27 back to tilling the soil when the mines were exhausted.29 The scholar Ish- rat Aslam categorizes the miners as “possibly pauperised peasants and landless workers from the villages”.30 We may therefore suppose that at least a large number of miners were labor migrants who moved from agricultural areas to the mines and vice versa. Miners were contract laborers who worked for wages in cash, and who were sometimes also partly compensated in food.31 Miners earned only three pagodas a year according to Tavernier, who may have mistaken the pay per month for a pay per year, as other observers noticed monthly wages from 0.5 to 1.5 pagodas (a pagoda roughly had the value of eight British shillings). According to Ravi Ahuja, a male “general worker” in Madras would have earned a monthly money wage of one pagoda around 1760.32 This estimate would imply that the monthly earnings of the mine workers were more or less an average wage. However, compared to the other production costs, wages were very low. These low labor costs were also signaled by De Coutre, who stated that they would be much higher in Spain, leading to a higher price per carat.33 Food and other necessities of life (such a tobacco and betel leaves) had to be brought in from other territories, and the local rulers imposed heavy taxes—sometimes as high as 50%—on the already expensive commodities. According to Streynsham Master, miners and traders—except for privileged foreigners—were obliged to live in the town near the mines where these taxes were levied.34 Methwold described the region of the Kollur mines as a place so barren, that before the discovery of diamonds it was hardly inhabited. But now it was “peopled with a hundred thousand souls, consisting of miners, merchants and such others as live by following such concourses.”35 Around the mine, a complete local economy developed, leaving traces in the landscape which were still visible long

29 Verberckmoes and Stols, Aziatische omzwervingen, p. 172, Tavernier, Travels in India, I, p. 230. 30 Ishrat Alam, “Diamond Mining and Trade in South India in the Seventeenth Cen- tury,” The Medieval History Journal, 3 (2001), pp. 291–310, 300. 31 Alam, “Diamond Mining and Trade in South India,” p. 300 and Report by Pieter de Lange as included in van Dam, Beschrijvinge van de Oostindische Compagnie, p. 180. 32 For the other observations on mine workers wages, see Mukund, “Mining in South India,” p. 18. For the wages of a general worker in Madras see Ravi Ahuja, Die Erzeugung kolonialer Staatlichkeit und das Problem der Arbeit: eine Studie zur Sozialgeschichte der Stadt Madras und ihres Hinterlandes zwischen 1750 und 1800 (Stuttgart, 1999), Appendix 8.1. 33 Verberckmoes and Stols, Aziatische omzwervingen, p. 174. 34 Temple, Diaries of Streynsham Master, p. 173. 35 Methwold in Moreland, Relations of Golconda in the Early Seventeenth Century, p. 33. 28 karin hofmeester after the mines had closed down. In 1908, an observer noticed “ruins of extensive habitations which are still to be seen on what is now a most desolate spot.”36 From De Coutre’s descriptions, we learn that the miners were sometimes “paid” per stone by merchants, who in return supplied them with food. As the miners sometimes would not find a stone for two or three months, they could easily end up in a position of debt bondage.37 Accord- ing to De Coutre, the miners—in this case in Ramallakota—barely had enough to eat. They lived in huts covered with straw, slept on mats, were covered with dirt of the mines, and wore no other clothes than a loin- cloth (which was obligatory; this was to prevent the miners from stealing). Unfortunately he gives no description of the other workers around the mine, so we have no idea whether they were better off. Miners who were too poor to support themselves sometimes organized in communities he calls compagnies.38 Streynsham Master reports a different situation at the Mallavilli mine near Kollur. He claimed “people are well favoured, well clothed, and looke as though they fed well to undergoe their great and hott labor.”39 How the miners perceived their position themselves, might be judged by the number of cases where a miner found a big stone, and took off with his wife and children.40 Though the miners were often exploited by the merchants, they were worse off if there were no “adventurers” to commission mining, because then they had to work for the king, in exchange for boarding only. De Lange described these miners as “slave like objects.”41 Howard noticed that the governor of the Mallavilly mine in the 1670’s had shut the mine down, and ordered all the miners to repair his residence. The miners had to obey, or else had to flee the area.42 Walking out seems to have been the only effective form of labor protest which miners had at their disposal. In 1655, both merchants and miners left the Kollur mine because of the bad governance by Jamal Beg. Only when a Brahman named Bhimanji was appointed, did the merchants and

36 The Imperial Gazetteer of India vol. XV: Karachi to Kotayam (Oxford, 1908), p. 328. 37 Verberckmoes and Stols, Aziatische omzwervingen, pp. 172–4. 38 Ibidem, p. 172. 39 Temple, Diaries of Streynsham Master, p. 175. 40 “A Description of the Diamond-Mines,” p. 916. 41 Report by Pieter de Lange, as included in van Dam, Beschrijvinge van de Oostindische Compagnie, pp. 179–81. 42 “A Description of the Diamond-Mines,” p. 909. working for diamonds from the 16th to the 20th century 29 the miners return.43 In this phase, labor relations in the Indian diamond mines seem to have been determined primarily by private decisions of local rulers. The Bijapur sultan appears to have shut down a number of mines in 1622, just when the VOC had started trading the stones very prof- itably, in order “to keepe the commoditie in request”. Indirectly, the Dutch demand thus made the miners lose their jobs, although only for a year; the same mines reopened in 1623.44 More often labor in the mines was disturbed by wars between rulers. The Mughal Emperors tried to conquer the Golconda and Bijapur sultanates, the mines being one of the reasons for their eagerness. In 1686 and 1687, they succeeded in their endeavours; mining operations remained disturbed till 1692 as a consequence of the wars. When the Mughals finally started farming out the mines, procedures were more or less the same as during the preceding sultanates.45

European demand and Indian production

The possible influence of European demand on labor in the Indian mines should be viewed in a broader context. As the eagerness of the Indian rulers for the bigger stones shows us, there was a large internal consumer market for diamonds in India. Rulers were not the only Indians who owned and cherished diamonds. De Coutre described how each morning a procession of lords went to the court of Ibrahim Adil Shah II, the ruler of the Bijapur sultanate—mounted elephants, decorated with coloured gemstones came in front, followed by horses that wore gold and silver chains with plumes, set off with jewels containing diamonds, rubies and emeralds.46 Stressing the difference in wealth between these rich lords and the ordinary population, De Coutre stated that if the members of the latter group possessed any wealth, the men would wear a brooch of gold and filigree and golden earrings with emeralds and rubies. He described the women as “very beautiful in their own way” with ribbons in their hair; a jewel on the front head set off with diamonds, rubies or emeralds; a

43 Report by Pieter de Lange as included in van Dam, Beschrijvinge van de Oostindische Compagnie, pp. 178–179, Alam, “Diamond Mining and Trade,” p. 296 and Mukund, “Mining in South India,” p. 16. 44 Tapan Raychaudhuri, Jan Company in Coromandel 1605–1690 A Study in the Interre- lations of European Commerce and Traditional Economies (The Hague, 1962), p. 171, and Methwold, Relations of Golconda, p. 33. 45 Omar Khalidi, Romance of the Golconda Diamonds (Middletown, 1999), pp. 34–5. 46 Verberckmoes and Stols, Aziatische omzwervingen, p. 119. 30 karin hofmeester pearl or emerald nose pin; earrings as big as hand palms and necklaces of heavy pearls, emeralds and rubies.47 For later periods, we know from the work of R.J. Barendse that the possession of diamonds was not restricted to emperors or kings: he describes the probate inventory of a subehdar (governor) of Bengal in 1728 who owned some fifty-two diamond rings. When the wealthy business man Mohan Das Seth in Bombay died, he left his wife a fortune in diamonds, and when the Bombay devadasi (temple dancer) Moti died in 1752, she left a considerable fortune in diamonds and joys. Not only members of the urban population owned diamonds, farmers in 18th-century Kerala possessed golden belts, inserted with pearls and diamonds48 A second point that has to be taken into consideration is that European traders were not the first, and certainly not the only merchants interested in diamonds. The gemstones formed an important part of India’s export across the Indian Ocean seaboard (including South East Asia, the Persian Gulf and the Southern part of the Arabic Empire) where, according to Janet Abu-Lughod and Andre Gunder Frank, a global economy existed since or even before the fifteenth century.49 According to Holden Furber, the European trade in diamonds formed but a small part of the whole, “as Golconda diamonds found their way to all parts of Asia.”50 European demand expanded in the late 17th century when the use of precious stones in jewellery increased, and diamonds were being worn in bourgeois circles. This might have been an effect of the larger supply of gemstones—not only diamonds but also emeralds from Colombia, for example—but it could also be explained by the increasing economic prosperity of the bourgeoisie. Probably it was a combination of both factors.51 In the 18th century, we can witness an increase in the use of diamonds in precious stone jewellery. This can be explained by the lower prices,

47 Ibidem, pp. 120–1. 48 Barendse, Arabian Seas 1700–1763, II, pp. 838–9 and 711, and Idem, Arabian Seas 1700–1763, Volume III: Men and Merchandise (Leiden, 2009), pp. 919 and 922. 49 Janet L. Abu-Lughod, Before European Hegemony: The World System A.D. 1250–1350 (New York, 1991); Andre Gunder Frank, ReOrient: Global Economy in the Asian Age (Berke- ley etc., 1998), pp. 86–96. For the early trade relations in the Indian Ocean region see K.N. Chaudhuri, Trade and Civilisation in the Indian Ocean. An Economic History from the Rise of Islam to 1750 (Cambridge, 1985), for his remarks on diamonds, see pp. 20 and 53. 50 Holden Furber, Rival Empires of Trade in the Orient 1600–1800 (Minneapolis, 1976), p. 260. 51 Gedalia Yogev, Diamonds and Coral. Anglo-Dutch Jews and Eighteenth-Century Trade (Leicester, 1978), p. 89. working for diamonds from the 16th to the 20th century 31 caused by the increased supply since the discovery of diamonds in Brazil in the late 1720’s, but also by a new fashion in diamond-cutting.

Labor in the cutting and polishing industry

One of the earliest sources informing us about the cutting techniques known in Europe is a description of a number of gemstones, including a price list, made in 1403 by a Jewish jeweller in Venice, who knew about diamond-cutting as well as the polishing process. From his list we learn that the heavier the diamond, the bigger the relative price difference between the rough and the finished stone.52 This can be explained by the fact that a finished stone reveals its qualities immediately, whereas one cannot always tell how much of a heavy uncut stone is usable after cleaving. In the 15th century, the so-called table cut developed, which meant that the top of the octahedron was flattened, giving the diamond a flat “table.” The technique to cut and polish diamonds was most probably developed in India, and spread from there to Venice.53 In the 16th century, the technique of polishing facets developed. For this, one needs a polishing wheel and bort (diamond dust). In Europe, the first description of this technique was written by Benvenuto Cellini in 1568.54 At the same time, the first Indian Mogul Emperor Sultan Babur (1463–1530) described how he received a diamond (probably the Koh-i-noor) from his son.55 This was covered with triangular facets, arranged in a symmetrical radiating pattern, with the bottom of the stone left flat: a so-called “rose cut”. We do not know—yet—where this faceting technique originated, but we do have several descriptions of the work of Indian diamond cutters.56

52 Colette Sirat, “Les pierres précieuses au XVe siècle,” Annales. Économies Sociétés Civilisations, 23 (1968), pp. 1067–85, 1078. 53 Godehard Lenzen, The History of Diamond Production and the Diamond Trade (Lon- don, 1970), p. 72 and Alois Haas, Ludwig Hödel and Horst Scheider, Diamant. Zauber und Geschichte eines Wunders der Natur (Berlin, 2004), p. 231. 54 Benvenuto Cellini, Abhandlungen über die Goldschmiedekunst und die Bildhauerei [translation of Trattato dell’oreficeria from 1568 by R. and M. Fröhlich], (Basel, 1974), pp. 33–6. 55 The Baburnama: Memoirs of Babur, Prince and Emperor. Translated, edited, Cour- tesy Random House and annotated by Wheeler M. Thackston (reprint, New York, 2002), p. 328. 56 The discussion on the origin of the faceting technique is analysed in my : “Diamonds as a Global Luxury Commodity” in Bernd-Stefan Grewe and Karin Hofmeester (eds), Lux- ury in Global Perspective: Commodities and Practices, c. 1600–2000 (forthcoming). 32 karin hofmeester

Figure 2. Top: Natural octahedron whose natural planes were polished. Bottom: Table cut (from Godehard Lenzen, The History of Diamond Production and the Diamond Trade, published by Barrie and Jenkins Ltd in 1970, courtesy of Random House, p. 78)

Figure 3. Top: Rose cut, top view. Bottom: Rose cut, side view (from Godehard Lenzen, The History of Diamond Production and Trade, published by Barrie and Jenkins Ltd in 1970, courtesy of Random House, p. 79) working for diamonds from the 16th to the 20th century 33

Jean de Thévenot, a Frenchmen who traveled to India in 1666, describes the famous castle of Golconda, and how sultan Abdullah Qutb-Shah housed his favorite workmen there: The King will have the good Workmen to live there, and therefore appoints them lodgings, for which they pay nothing: He makes even Jewellers lodge in his Palace, and to these only he trusts Stones of consequence, strictly charg- ing them not to tell any what they work about, least if Aran-Zeb57 should come to know that his workmen are employed about Stones of great value, he might demand them of him. The Workmen of the Castle are taken up about the Kings common Stones, of which he hath so many that these Men can hardly work for any body else.58 The detailed description of Mughal Emperor Akbar’s administration as given in the Aín-i- Akbari also mentions “lapidaries, metal casters and other artificers” who were “constantly employed at the Imperial Court where their work is subjected to the test of criticism”.59 Whereas de Thévenot seems to imply that artisans in the royal workshops (kharkanas) could not work for anybody else, under Mughal rule they often would not even be allowed to work for other employers and—according to some authors—performed forced labor.60 Diamonds were not only cut and polished by artisans in the royal workshops, they were also worked near the mines. Tavernier’s description of the Ramallakota mine includes a report on the cutting and polishing activities he saw there. The extraction methods sometimes damaged the diamonds: when the miners smashed the lumps of earth to reveal the diamonds, they could cause fractures and flaws to the stones. If the miners noticed the flaws, they immediately cleaved it—“at which they are much more accomplished than we are”—according to Tavernier.61 After the cleaving and cutting, the stone was covered with facets “in order that its defects may not be seen”. If the diamonds had no flaws, “they do not more than just touch it with the wheel above and below, and do not venture it

57 He is referring to Mughal Emperor Aurangzeb to whom Abdullah Qutb-Shah had to pay tribute. 58 Surendranath Sen (ed.), Indian Travels of Thevenot and Careri (New Delhi, 1949), p. 138. 59 The Ain-I Akbari by Abu’l-Fazl Allami, translated from the original Persian by H. Blochmann and H.S. Jarrett (Calcutta, 1873–1907), III, pp. 312–3. 60 See, for this discussion, a.o. Tripta Verma, Karkhanas under the Mughal. From Akbar to Aurangzeb. A Study in Economic Development (Delhi, 1994). 61 Tavernier, Travels in India, II, p. 44. 34 karin hofmeester to give it any form, for fear of reducing weight”.62 In Ramallakota, there were numerous diamond-cutters according to Tavernier (it is remarkable that De Coutre, who visited the same mine some twenty to thirty years earlier did not mention these cutters at all). Each of them had a steel wheel “about the size of our plate”. To find the grain of the diamond, the cutters put water on it; to be able to polish the diamond, they poured on oil and ample diamond dust “although it is expensive,” to make the stone run faster. “The mill was like ours, the large wheel of which was turned by four blacks,” according to Tavernier, but it turned less fast than the European ones because “the wooden wheel which causes the steel one to revolve is seldom more than 3 feet in diameter.”63 Tavernier felt that “the Indians were unable to give the stones such a lively polish as we give them in Europe; this I believe, is due to the fact that their wheels do not run as smoothly as ours.”64 There was also another difference. In order to polish the stones, they were pressed against a revolving disk. As these disks would become dull after a while, they had to be ground regularly. As the Indian disks were made of steel (whereas the European disks were made of iron) they had to be taken off the wheel to be ground on emery, whereas the iron ones could stay in place, and be ground by a file. As a consequence, the Indian disks were ground less often, but when they were ground and the disks were put back on the wheel, they could run less smoothly. Pieter de Lange has described—though in less detail—the same process in the Kollur mine: damaged stones would be polished on disks and sold as laskes.65 Unfortunately neither Tavernier nor De Lange mentions for whom, and under which conditions, the cutters near the mine worked. We may conclude that the technique of facet polishing was well known in the 16th and 17th century, both in India as well as in Europe, and that more or less the same technique was used. What we know for sure, is that in Europe the facet polishing technique was developed further in the late 17th century, when the so called “brilliant cut” was invented. This would become the most popular cut in Europe in the 18th century.66

62 Ibidem. 63 Ibidem, p. 45. 64 Ibidem. 65 Report by Pieter de Lange as included in van Dam, Beschrijvinge van de Oostindische Compagnie, p. 177. 66 M.H. Gans, Juwelen en mensen. De geschiedenis van het bijou van 1400 tot 1900, voornamelijk naar Nederlandse bronnen (Schiedam, 1979), p. 173. working for diamonds from the 16th to the 20th century 35

The brilliant cut implied, that the polished diamond not only had a very symmetrically cut and multi-faceted top, but also a pointed bottom, the so called pavilion. This increased the refractory quality of the diamond, enhancing its brilliance, but also reduced its weight, often up to 50%. The latter was anathema to the Indian lapidaries. This is often explained by the fact that one classical Indian text states that the diamond loses its virtues if you cut it.67 However, the same text also mentions cutting and polishing as normal procedures, so maybe we should not attach too much value to that statement. It is however evident that the taste for stones which were left as big as possible was strongly developed in India. The number of carats was very important in the valuation, and therefore also in the pric- ing of diamonds in India. 68 In the 17th and 18th centuries, the Indian and European tastes for finished diamonds clearly began to diverge. Already in the 1670s, John Fryer (a scientifically trained servant of the EIC) wrote that the Indian cut and polished diamonds were mostly sold in the country, whereas the rough stones were sent to Europe—“they coming short of the Fringies in Fancy”. In Europe, they were “both set and cut to more advantage”.69 Elsewhere, he advised potential buyers in India that “Rough, brute or uncut stones, are in value half the price of cut or polished stones”.70 The same price difference, even increasing with more than 50% for stones of one carat and more, was also mentioned by the 15th century Venetian-Jewish merchant quoted earlier. The sources give no information about the reason for this price difference; of course, there was the direct visibility of the quality of a cut stone, but maybe the wages of Indian cutters and polishers also added value to it, even though the result was not fitting European tastes. As taste and fashion are important factors in the production and consumption of luxury commodities, one might plausibly assume that the European preference for brilliants prompted the relocation of the diamond finishing industry from India to Europe. But there were also more prosaic reasons for the relocation: more value could be added, if finishing

67 Louis Finot, Les lapidiaires indiens (Paris, 1896), pp. xxx–xxxi, he quotes the Agas- timata, a tenth century manuscript on gemstones with a lot of later date additions. In an appendix to this manuscript, it says that a diamond cut with a blade or worn out by repeated rubbing, becomes useless and looses its benevolent virtue. However, in the manuscript itself, cutting and polishing are described as normal, permitted procedures. 68 Oppi Untracht, Traditional Jewelry of India (New York, 2008), pp. 317–8. 69 John Fryer, A New Account of East-India and Persia in Eight Letters being Nine Years Travels, Begun 1672 and Finished 1681 (London, 1698), p. 113. 70 Ibidem, p. 213. 36 karin hofmeester and cutting could be done in Europe. If a merchant had enough skills to value a rough diamond, his profits would be bigger if he could avoid the re-cutting process that often had to be done in Europe to adapt the stone to local tastes. The relocation also fits the general tendency to start producing “Oriental” commodities in Europe.71 Since more money could be made by buying rough diamonds in India, and having them cut and polished in Europe according to the latest brilliant fashion, the discovery of new diamond fields in Brazil seemed to set off a true rearrangement of the diamond commodity chain with large consequences for the labor relations in both Brazil, India and Europe.

The discovery of diamond mines in Brazil and its consequences

In the late 1720s, the discovery of large deposits of alluvial diamonds in the Brazilian Minas Gerais district north of Rio de Janeiro seemed to her- ald a new era in the diamond trajectory history.72 No longer was India the sole supplier of rough diamonds (apart from the very small stream of diamonds that reached Europe from Borneo from the late 17th century onward) and no longer did local kings decide how and by whom the mines would be exploited.73 The diamond fields in Brazil were exploited by the colonial powers in Lisbon: a Crown that was not in the first place interested in diamonds for its own adornment, but in making as much profit out of them as possible. Initially, the new exploiter—happy to have a new source of diamonds after being outpaced and outplaced by the Dutch and the British in India—welcomed all merchants and miners to work the mines, as long as they paid a tax per miner to the Portuguese treasury. These miners were in fact mostly slaves, imported on a large scale from Africa especially for this purpose. From a labor relations perspective, the biggest impact of the discovery of diamonds in Brazil was of course to be found in Africa and Minas Gerais itself, not only for the massive increase of explicit unfree labor, but also for the development of a local economy

71 For this, see Maxine Berg, Luxury and Pleasure in Eighteenth-Century Britain (Oxford, 2005) and for a focus on the textile industry: Giorgio Riello, “The World of South Asian Textiles, 1500–1850,” in Giorgio Riello and Tirthankar Roy (eds.), How India Clothed the World: The World of South Asian Textiles, 1500–1850 (Leiden, 2009), pp. 1–27. 72 Later in the 1740s, diamonds were found in the Mato Grosso district in the East, and finally, in the 1840s, new field were discovered in Bahia. 73 Though some publications mention the Borneo mines, no serious attempts have been made to include Borneo in the global diamond commodity chain, which I hope to do in a later stage of research. working for diamonds from the 16th to the 20th century 37 around the mines, including the start of agricultural production and even imports from Europe.74 However, because this article focuses on India and Europe, I will skip this important interconnection here. The uncontrolled mining operations caused an enormous flow of rough diamonds in Europe, more than five times the value that usually came from India, which led to a price drop to half the previous value, and in some cases even to a third of the usual price.75 The Indian trade came to a complete standstill. Startled by the lowered prices and the responses of the European traders—who feared that in Brazil “diamonds were as plenty as transparent pebbles”—the Portuguese Crown shut the Minas Gerais mines down in 1734.76 When the mines reopened in 1739, they established a mining monopoly, with the actual mining entrusted to one single contractor or consortium (in practice it was usually a Brazilian merchant of Portuguese origin). The contractor had to pay rent per slave, and was not allowed to employ more than 600 slaves, to avoid overproduction.77 The trade in rough diamonds was linked up with this mining monopoly, so that representatives of the contractor could only sell their products in Lisbon, where trading procedures were state-controlled, and officials of the king had the first choice of stones though—unlike the Indian kings— they paid for the diamonds. Only after this procedure, the representatives could sell the remaining diamonds to other European merchants.78 In 1753, the Crown—in an attempt to stop the ongoing illegal mining and smuggling—decided to separate the two parts of the diamond commodity chain, and established a true trading monopoly, next to the mining monopoly. The Dutch consul in Lisbon, Daniël Gildemeester, obtained this extremely expensive trading monopoly in 1761, and held it for several decades.79 Concluding that it was impossible to combat corruption in the mining business, the Portuguese king in 1771 decided that the Crown

74 See for the increase of slave labor: Laird W. Bergad, Slavery and the Demographic and Economic History of Minas Gerais, Brazil, 1720–1888 (Cambridge, 1999); for the development of the local economy see Donald Ramos, “Slavery in Brazil: A Case Study of Dia- mantina, Minas Gerais,” Americas: A Quarterly Review of Inter-American Cultural History, 45 (1988), pp. 47–59. For the imports, see Tijl Vanneste, Commercial Culture and Merchant Networks: Eighteenth-Century Diamond Traders in Global History (PhD Thesis European University Institute, Florence, 2009), p. 293. 75 For the amounts see Yogev, Diamonds and Coral, p. 116. 76 David Jeffries, A Treatise on Diamonds and Pearls (London, 1751), p. 66. 77 Vanneste, Commercial Culture and Merchant Networks, pp. 231 and 281, Ramos, “Slavery in Brazil,” p. 48. 78 Vanneste, Commercial Culture and Merchant Networks, pp. 231 and 281. 79 Ibidem, pp. 231–5 and Yogev, Diamonds and Coral, p. 122. 38 karin hofmeester would be the sole mine exploiter (the ‘Royal Extraction’). This situation would last till Brazil’s independence in 1822, when the concession system was reintroduced. What were the consequences of the discovery of this new diamond mining region for production and labor, in India and Europe? Unlike India, a local consumer market for diamonds was nonexistent in Minas Gerais, all diamonds were mined to be sold directly in Europe.80 To earn as much as possible from their sale, the Portuguese Crown tried hard to concentrate the trade in rough in Lisbon. However, an enormous amount of smuggled diamonds directly found their way from Brazil to London—a city which kept its status as important market in rough diamonds.81 The large role of the Dutch trade monopolist Gildemeester (as well as the subsequent role of the Amsterdam based banking firm Hope & Co) in lending money to the Portuguese Crown in exchange for diamonds meant that Amsterdam had a constant and direct inflow of rough diamonds from Brazil.82 This had two important consequences: London had to share its position as rough market with Amsterdam, and, more importantly, the position of Amsterdam as the finishing center of the world was consolidated. It was hard to compete with Amsterdam’s supply of capital; its trade connections that guaranteed a constant influx of rough diamonds and its large number of finishers offering craftsmanship in the various, specialized branches of the industry, for relatively low wages.83 As a consequence, the trade in finished diamonds flourished in Amsterdam. After 1740, the trade with India surprisingly revived. The measures of the Portuguese Crown to regulate the supply seemed to be effective and the growing demand—spurred by the lowered prices and the invention

80 When the Portuguese Crown settled in Rio de Janeiro in 1808, taking refuge for the French troops that had occupied Portugal, they took their (taste for) diamonds with them, as well a the diamond finishers that had served the court. See Harry Bernstein, The Brazil- ian Diamond in Contracts, Contraband and Capital (Lanham, 1986), p. 57. For more consequences of the move from Lisbon to Rio de Janeiro on diamond production in Minas Gerais, see Bergad, Slavery and the Demographic and Economic History of Minas Gerais, pp. xviii, 93–94; 128. 81 Yogev, Diamonds and Coral, p. 122. 82 In exchange for loans to the King of Portugal, Hope & Co received an exclusive concession to sell diamonds originating in the Portuguese colony of Brazil. The Hopes would accept the diamonds and sell them on the Amsterdam market; then they used the proceeds to defray the interest and principal of the loans they had made to Portugal. For the Diamond Loan see Marten G. Buist, At Spes non fracta. Hope & Co., 1770–1815 (The Hague, 1974), p. 383 ff, for the technical explanation, pp. 386–7. 83 Yogev, Diamonds and Coral, p. 142; Jeffries, A Treatise on Diamonds and Pearls, p. 101. working for diamonds from the 16th to the 20th century 39 of the brilliant cut—balanced the supply. There is therefore no reason to assume that the discovery of Brazilian diamonds had any effect on the production in the Indian mines; the demand was big enough and for a very long time the Indian stones were preferred above the Brazilian ones.84 It might however have affected the cutting industry in India, as far as the Indian cutters worked for the European market. Only thorough research in available archives of European merchants and jewellers could statis- tically prove an increased demand for brilliants, and as a consequence extra demand for uncut stones from India. A quick comparison of the account books of Joseph Cope, a famous diamond polisher in London, for the years 1693–1710 shows us that he seems to have sold as many rose cut diamonds as brilliant cut diamonds, whereas the 1771–1776 accounts of EIC agent Harry Verelst with diamond merchant George Robertson show many more brilliants than rose cuts. Robertson also ‘made over’ an ‘India cut brilliant’. Still, Lyon Prager, a member of the famous London based diamond Jewish diamond merchant family, settled in Calcutta and bought diamonds polished in India as late as 1791, though they only formed a small part of his trade.85

Labor in the Indian diamond industry from the late 18th to the late 20th centuries

By the end of the 18th century, the production of the Indian mines seems to have slowed down, though the sources that inform us about this might be biased. There are quite a number of reports from British officials who explored the possible revenues of diamond mines in the recently acquired ‘Ceded districts’.86 These included the Bellary district (where several mines were located), as well as Cuddapah en Kurnool. These sources show the

84 Lenzen, The History of Diamond Production, pp. 129–130. 85 The National Archives of the United Kingdom, Kew, (hereafter TNA) Chancery Records, C 104/197 Cope vs Cope, (1693–1710) a.o. overview of diamonds bought (not dated), printed list of his jewel stock, to be sold in public auction after his death in 1710. British Library, London (hereafter BL), India Office Records (hereafter IOR), Mss Eur F 218/56 accounts of Harry Verelst with George Robertson (1771–76); TNA, Chancery Records, C 111/146 Elliot vs Willis, Sale book of diamonds and pearls from Lyon Prager (Calcutta 1787–1796). 86 In 1796 AD, the Deccan Nizam Asaf Jah II, harassed by the Marathas and Tipu Sultan, opted to get British military protection. In return, the Nizam ceded a large portion of the acquired territory to the British, to be added to the Madras Presidency. This area was also known as the Ceded Districts. 40 karin hofmeester direct influence of British demand on the situation in the Indian mines, especially now they had become in ‘their possession’. At the same time, they tell us how the mine was worked when the reporters came to visit, and how the British felt the mines should be worked. I will focus on the extensive reports on the mines near Cuddapah, and contrast these to the reports on the Panna mines that were not part of the Ceded districts. Starting in 1796, Benjamin Heyne wrote several of reports on a number of mines (he was a surgeon, naturalist and botanist, working for the EIC). Though Heyne was primarily interested in minerals, he also had an eye for the miners and their labor—and wrote about them. The exploitation of the mines near Cuddapah he described was more or less the same as in the 16th and 17th century: the mines were governed by a headman who paid rent—this time to the EIC—and who worked some mines himself, farming out the rest. The headman paid a yearly rent of 130 pagodas to the Company; for diamonds weighing more than 12 carats, he had to pay one third of its value to the Company. Miners were hired by the headman, and received one pagoda per month. Heyne noticed men, women and children working in the mines, about sixteen people per mine. As in previous centuries, the actual owner of the mine farmed it out, received rent for it, and a percentage of the larger stones, but the owner did not carry the financial risks; these were borne by the renters. The renter made 5,000 pagodas profit per year, against 2,000 pagodas in expenses.87 Labor costs still seemed to have been only a small part of the production costs. The mining procedures are much the same as in the earlier descriptions, though the number of miners was much smaller than in the earlier days, which could point in the direction of exhaustion of the mines, or at least the layers that could be worked on with the simple methods the miners had at their disposal. A report written in 1814 by C. Ross, collector to the Board of Revenue in Fort St George for Kurpah, near Cuddapah, gave a detailed overview of the revenue of the Company from the mines, showing that it earned more from the rent than from the share of the bigger diamonds. The bigger stones were sold in public auction; the profit was divided between renters and sub-contractors. According to Ross, most miners owned shares in the produce, only some miners were ‘mere laborers for hire’. Given the

87 Benjamin Heyne, Tracts, historical and statistical on India, with several tours through various parts of the peninsula: also an account of Sumatra, in a series of letters (London, 1814), pp. 101–2 and BL, IOR F/4/275/6149 and P/243/35. working for diamonds from the 16th to the 20th century 41 current output of the mines, he did not expect cultivators to go working in the mines, this might change though if mining would become more profitable. Ross was convinced that in reverse, miners would never engage in cultivation as their job was hereditary (suggesting a specific caste background). If they could find no work near the Penner river, they would go to the Kristna. One might conclude from Ross’s report that labor migration still existed, although not from agricultural areas to mining areas, but from one mining area to another. All in all, the labor relations between the owner of the mine and the main renter on the one hand, and the renter and the miners on the other, seem to be less unfree than in earlier periods. To make the exploitation of the mines more profitable, Ross suggested that smaller plots of the mine should be rented out. An advertisement was placed in the District Gazette, and some Indian entrepreneurs responded.88 In 1821, the Board of Revenue found that the rent had yielded no more than 200 pagodas a year, and therefore stopped the exploitation.89 In the meantime, a Mr Christy (a surgeon in service of the EIC), had suggested a more profitable way of exploiting the mine, using a very specific form of unfree labor. He suggested that the convicts of the Cuddapah district could be put to work in the mines nearby. They cost the Company a fortune each year, and although they worked to build water tanks, roads and bridges, they could be used for much more profitable work. According to Christy, nobody wanted to rent the plots of mine, as ‘coolly hire’ was expensive, and the rent was too high in relation to the profits. If convicts could be put to work in the mines, with the help of real “searchers,” there would be no expenses on wages. He attached a detailed account of the profit rates he expected for the next ten years, and also added an extra encouragement: “the government of Brazil employs all its convicts and many slaves in digging and searching for diamonds.”90 The Board of Governors of the EIC however did not like the plan. The convicts, they felt, did useful work already; it was not wise to put people who were con- victed for theft to work in a diamond mine, where they had to be guarded, and peons had to be paid, but most of all: the capital for the exploitation should be provided by private individuals and not by the government.91

88 BL, IOR F/4/540/13001, letter by C. Ross dated 24th of December 1814. 89 BL, IOR F/4/676/18769: extract revenue letter form Fort St George 6th of July 1821. 90 BL, IOR F/4/676/18769 letter of Mr Christy from 9th of January 1817. 91 BL, IOR F/4/676/18769 letter from England dated 22nd of May 1819. 42 karin hofmeester

In 1820, EIC Captain Buckley reported that the diamond mines in and around Panna in the central provinces of India yielded quite some diamonds of good quality. People thought only table cut diamonds came from these mines, but this was just a matter of Indian taste, they could be turned into brilliants for the Europeans. Buckley suggested a considerable investment of capital, knowledge and technique by the EIC and by private investors, but the Company turned this plan down as well, and decided not to work these mines as they were still ‘in the possession of a foreign power’.92 In 1877, the Frenchman Louis Rousselet described the diamond mines near Panna, which were quite deep but still worked with age-old devices such as candles (to heat and soften the hard layers of rock) and the Persian wheel (to transport water and earth to the surface). According to Rousselet, the mines yielded 1.5 million of francs per year, but very few diamonds reached Europe, as they were very much in demand in India itself. The Raja who owned the mines sold the rough stones to Allahabad and Benares, but lately had started a polishing workshop in Panna. Rous- selet felt their products could not compete with the Dutch finished stones, but he certainly liked the rose cuts and large facetted brilliants made in Panna.93 Diamonds were at that time not only cut and polished in Panna, Alla- habad and Benares. In 1842, EIC captain Newbold visited several diamond mines, and wrote about Munimadugu that here, some of the cutters and polishers ‘famous for their skill as lapidaries’ could be found.94 They used to process the stones found in the Ramallakota mines, where the cutters and polishers who had lived there for centuries, had fled the attacks of the Maratha troops and settled in Munimadugu. The Kurnool District Manual gives us a glimpse of the labor relations in and around the mines: the cutters were hired by Gujarati merchants, who fled with ‘their’ kamsalis from one mine to another.95 Most of the cutters and polishers would escape from violence again, and leave Munimadugu for Madras or Hyderabad.96

92 BL, IOR F/4/661/18326 Extract Bengal Public Consultation 15th of September 1820. 93 Louis Rousselet, L’Inde des Rajahs. Voyage dans l’Inde centrale er danse les présidences de Bombay et du Bengale (Paris, 1877), pp. 442–3. 94 Lieut. Newbold, “Mineral Resources of Southern India No 8. Diamond Tracts,” The Journal of the Royal Asiatic Society of Great Britain and Ireland vol. 7, no 8 (1843), pp. 226–40, 230–1. 95 Chetty, Manual of the Kurnool District, p. 95. 96 BL, IOR F/4/540/13001 Extract proceedings of the Board of Revenue at Fort St George of 2nd of January 1815. working for diamonds from the 16th to the 20th century 43

The Panna mines remained in the minds of the British who, in 1904, asked E. Vredenburg of the Geological Survey of India97 to write a report on the mines. He concluded that the mines could still be worked prof- itably, as long as modern science and technique were used. Of course, Indian laborers would still have to do the real mining work, and Vreden- burg noticed, this should be family work: men, women and children working together, as the Indians clearly preferred this way of working. Strict supervision of the workers was needed, but since there were women and children involved ‘such severe measures as are practiced in the South Afri- can mines’ were of course impossible.98 These reports, letters and books all show us that diamonds were still found in India, even in the 20th century, though not so much in the areas which the British had under their control. A publication of K.P. Sinor from 1930 on the Panna mines shows us that diamonds were still mined there, and also cut and polished, though on a small scale. It is therefore not correct to speak of a ‘return’ of the diamond polishing industry in the 20th century; it just never disappeared. When the Indian mines did not produce enough stones to satisfy the Indian appetite for diamonds, they were imported from South Africa.99 Indian merchants also started to buy rough and polished stones from Antwerp traders, who had local agents in Bombay. The fact that in the 20th century the Indian merchants turned to Antwerp, rather than to Amsterdam, proves that the latter city had lost its position to the first, which is another important shift in the diamond trajectory, already discussed elsewhere.100 To circumvent these middlemen merchants, several jeweller-merchants from Palanpur in Gujarat—all Jains—started to travel to Antwerp themselves to import polished stones.101 Some twenty Palanpuri dealers

97 The Geological Survey of India was established in 1851, as a follow-up of the EIC commission on Coal which aimed to study and explore availability of coal in the eastern parts of India. Eventually it became a government organization controlled by the Union Ministry of Mines for conducting geological surveys and studies. 98 BL, IOR/R/2/449/4 E. Vredenburg, Geology of the State of Panna, principally with reference to the Diamond bearing deposits (1904). 99 In 1867, India started importing rough diamonds from South Africa immediately after their discovery, see V. Ball, A Manual of the Geology of India. Part III Economic Geology (London, 1881), appendix A, pp. 576–9. 100 See for example Salvador Bloemgarten, Henri Polak, social democrat 1868–1943 (The Hague, 1993), chapter seven. 101 Jainism is a religion that prescribes pacifism and non-violence towards all living beings. It is a minority religion whose adherents form quite successful immigrant communities in North America, Western Europe and the Far East. For their activities in the diamond trade, see Sebastian Henn, “Transnational Communities and Regional Cluster 44 karin hofmeester travelled up and down to Antwerp in the 1920s.102 These imports stopped abruptly in 1947: after its independence, India’s new government established import regulations that prohibited the import of polished diamonds. Rather than spending money abroad on “luxury,” the government wanted to stimulate investments in the economic developments of India. In 1952, the Indian entrepreneurs were allowed to import diamonds again, on the condition that only 10% of their purchases were polished; the rest should be rough stones that had to be polished in India. This way, the Indian industry would be stimulated. Some Palinpuri merchants invited Antwerp cutters and polishers to India to teach them the modern techniques. These techniques were transferred to the people from the villages in Gujarat.103 The Indian diamond industry was further stimulated by the 1962 Replenishment Scheme. From then on, all import restrictions were repealed, as long as the finished goods were exported at a higher price.104 This led to a growth of the diamond cutting and polishing industry in India. From 1964, Indian merchants were welcomed as sight holders— authorized purchasers of rough diamonds—at the Central Selling Orga- nization in London. The discovery of diamond mines in Australia in 1985, which produced mainly small, low quality gems, really kicked off the Indian diamond industry, when Indian entrepreneurs decided to polish very small stones which previously used to be considered suitable for industrial use only. The price of large, and therefore rare, rough diamonds is very high and labor costs form a relative small percentage of the total production costs. However, with small stones, the value added by cutting and polishing is greater in proportion to its total price. This explains the interest of the Indian entrepreneurs: they combined a relatively skilled labor force with low labor costs, that could be achieved, amongst other things, by a high

Dynamics. The Case of the Palanpuris in the Antwerp Diamond District,” Die Erde, 141 (2010), pp. 127–47; 132–3. 102 Ibidem, 134. 103 Bernard Imhasly, “Schleifen am Familientisch. Über Indiens wichtigste internationale Industrie,” NZZ Folio. Die Zeitschrift der Neuen Zürcher Zeitung, 12/93, [last accessed on 11 August 2011]. 104 Henn, “Transnational Communities and Regional Cluster Dynamics,” p. 136 and Menahem Sevdermish, Alan R. Miciak, and Alfred A. Levinson, “The Rise to Prominence of the Modern Diamond Cutting Industry in India,” Gems and Gemology, 34 (1998), pp. 4–23, 6. working for diamonds from the 16th to the 20th century 45 input of child labor.105 The tiny stones were very fashionable in the USA, where a fast expanding market developed when the American department store chain Walmart started selling jewellery set with small stones.106 To illustrate the growth of the industry in India: in 1966, 6% of the world’s diamonds were polished in India, and in 1996, 92%.107 As a country with a long mining, finishing and consumption tradition, India is nowadays a finishing center for all types of diamonds, from small to large, and is also developing as a diamond consumption center.

Conclusion

The worldwide production and trade of rough and finished diamonds involved many different forms of labor, often interconnected. Much of this labor was migrant labor: Gujarati merchants moved from the West coast to the mines in South and Central India, forming a true trading Diaspora, employing local miners who went to and fro between agricultural and mining areas in the early modern period, and from one mine to the other in the 19th century. European merchants also migrated, to pick up their part in the commodity chain, leaving the European capitals in order to settle in Indian coastal trading hubs. As a consequence of the growing European demand for diamonds, the wish of the Portuguese Crown to earn as much as possible from diamond production in its colony, and the lack of a local labor force, a large scale forced migration took place from Africa to the Brazilian diamond mines from the late 1720’s onwards. Work in the cutting and polishing sector depended too on circulation, not only of knowledge and technology, but also of skilled workers. In India, Gujarati merchants took ‘their’ kamsalis from mine to mine, and the shift of the finishing industry from Antwerp to Amsterdam was spurred by the migration of skilled workers. In the 20th century, the travels of merchants and finishers between Antwerp and Bombay stimulated the large scale growth of the Indian cutting industry. Next to migration, ethnicity, religion and caste played a role in labor relations in the global diamond production. Gujarati merchants were

105 For a very critical review of the diamond sector, including child labor in the Indian sweat shops, see Janine Roberts, Glitter & Greed. The Secret World of the Diamond Cartel (New York, 2003), chapter 2. 106 Tom Zoellner, The Heartless Stone. A Journey through the World of Diamonds, Deceit and Desire (New York, 2006), p. 199. 107 Sevdermish, Misiak and Levinson, “Rise to Prominence,” p. 8. 46 karin hofmeester often Banias not seldom Jains, whereas European merchants were often— though certainly not always—Jewish. Both traditions can still be seen in the 21st century trading and finishing hubs Antwerp and Indian cutters and polishers all seemed to have been members of the goldsmiths caste, whereas the Amsterdam diamond industry for a long time was a mainly— though again certainly not completely—Jewish industry. These ties facilitated contacts with trade—and often family—relations in other parts of the world, sharing capital on the one hand and knowledge, skills, contacts and trust on the other. These forms of human capital are very important in the sector and travel easily, if needed globally. Next to the characteristics of people in the diamond production, there is also the geographical and more importantly the political context that shaped labor relations. There are striking similarities in the way mine owners exploited ‘their’ mines. Most of them farmed out the mines, leaving the financial risks to the governors, merchants or renters of the mines. The owners profited from the mine’s produce by demanding (a share of) the larger diamonds, mine rent as well as taxes. Whether the mine owner was a Deccan sultan, Mughal Emperor, the EIC or the Portuguese Crown, they more or less all followed the same exploitation methods, striving for a monopoly on the mines and the control of supply. Political and more specifically colonial contexts could also determine large differences between labor relations in different countries. Putting convict laborers to work in the diamond mines in the Indian districts ‘ceded’ to the British was unacceptable, while it was a daily routine in the Portuguese colony Brazil. Apart from these differences, colonial powers could also determine which mines were exploited, how and by whom. Depending on the nature of the ‘government’ of the mine, miners could be subject to slavery or harsh labor conditions such as low wages, debt bondage and corvee labor. Finally, there are number of very visible global-local interconnections, such as the local economies that developed around the diamond mines in India, but more especially in Brazil, as a consequence of the production of diamonds for the European market. Perhaps the most striking interconnection is the 18th century relocation of part of the finishing industry from India to Europe, as a consequence of a changing consumer taste there and the discovery of diamonds mines in Brazil, a development that was reversed in the 20th century after the discovery of mines in Australia and the start of a new fashion for small stone jewellery in the US. Green Plants into Blue Cakes: Working for Wages in Colonial Bengal’s Indigo Industry

Willem van Schendel

Throughout the nineteenth century, hundreds of thousands of laborers could be seen working for wages in thousands of factories in rural India. They produced a prime commodity destined for European markets: indigo, a blue dye in high demand. It has been said that, in rural colonial India, indigo companies were the first large organizations to operate that were not military or religious in nature.1 Most indigo factories were located in the huge province of Bengal, where they dotted the countryside.2 For about a century the indigo processing industry was the most important rural manufacture in Bengal. Astonishingly, we know very little about indigo factories and the laborers who worked in them. Labor historians of Bengal have largely ignored this early industry, concentrating instead on later ones such as jute, tea and steel. No doubt one of the reasons for this neglect is the way indigo came to be perceived in Bengal: as a peasant issue. Introduced in the late eighteenth century, indigo soon turned into a hated crop. It became the focus of an intense struggle between European entrepreneurs and peasant producers who resented indigo because of its forced nature and low remuneration compared to competing crops. There was no dovetailing of seasons between subsistence agriculture and indigo (as had been the case in the U.S. South, formerly an important indigo-producing area). On the contrary, indigo competed directly with both subsistence and market crops such as rice, vegetables and lentils. Campaigners, politicians, bureaucrats and journalists reported widely on the conditions under which indigo plants were being grown, especially in

1 Raaj Sah, “Features of British Indigo in India,” Social Scientist, 9, 2/3 (1980), p. 69. 2 Today “Bengal” refers to Bangladesh (“East Bengal”) and the state of West Bengal in India. In colonial times this region was sometimes known as “Bengal proper” and it underwent many administrative changes. For most of the period it was part of the province of Bengal, which also included present-day Bihar and Orissa. From the 1780s to the 1860s, indigo production was concentrated in Bengal proper and Bihar. After 1860, however, indigo decreased sharply in Bengal proper. As a result, Bihar now became the center of production till the industry ended in the 1920s. 48 willem van schendel the wake of peasant revolts against forced indigo production in the 1850s. Their writings turned the crop into a byword for the inhumanity of Brit- ish colonial rule and the imperial exploitation of India’s peasant producers. This trend, initiated by activists such as Mitra, Long and Day,3 soon asserted itself in official reports4 and academic history writing. And it has never left it: for historians, indigo labor continues to equal exploited field labor. They have used the source material thrown up by the controversy to study labor relations in indigo production and they have accepted the bias of their sources by equating indigo labor with indigo field labor.5 There is a second trend in the historiography of indigo. Economic historians have studied the financial, managerial and trade aspects of the rise of this major industry. They have concentrated on shifts in colonial policy, streams of investments and remittances, entrepreneurship, the rise and fall of “agency houses”, shipping and marketing.6 These studies have provided us with a much better understanding of the institutional context and vicissitudes of indigo production—the processes that made it possible for indigo factories to be set up and run successfully. But they do not advance our understanding of the factories themselves. As a result, in the extensive historical literature on indigo, the factory that turned green plants into cakes of blue dye remains uncharted territory. This chapter intends to begin redressing the balance by looking at labor in indigo factories. Who worked there? Where did they come from?

3 Dinabandhu Mitra, Nil Darpan (Dhaka, 1860, in Bengali); translated as: A Native, Nil Darpan; or, The Indigo Planting Mirror: A Drama (Calcutta, 1861); James Long, Strike, But Hear! Evidence Explanatory of the Indigo System in Lower Bengal (Calcutta, 1861); Lal Behari Day, Govinda Sámanta, or The History of A Bengal Ráiyat (London, 1874 (title later changed to Bengal Peasant Life). 4 A Ryot, Selections from Papers on Indigo Cultivation in Bengal, with an Introduction and a Few Notes (Calcutta, 1858); A Ryot, Selections from Papers on Indigo Cultivation in Lower Bengal with an Introduction and a Few Notes, No. II (Calcutta, 1860); Report of the Indigo Commission, 1860: Report, Minutes of Evidence and Appendix (Calcutta, 1860–1). 5 Much historical work on indigo relies heavily on the report of the Indigo Commis- sion, 1860. The brief of this commission was to “inquire into and report on the system and practice of indigo planting, and the relations between indigo planters [factory managers] and the ryots [peasants] and holders of land in Bengal.” It did not cover labor relations in the indigo factories. Report of the Indigo Commission, 1860, p. iii. 6 Benoy Chowdhury, The Growth of Commercial Agriculture in Bengal (1757–1900), vol. 1 (Calcutta, 1964); H.R. Ghosal, The Economic Transition in the Bengal Presidency (Calcutta, 1966); S.B. Singh, European Agency Houses in Bengal, 1783–1833 (Calcutta, 1966); Blair B. Kling, Partner in Empire: Dwarkanath Tagore and the Age of Enterprise in Eastern India (Berkeley, 1976 / Calcutta, 1981); Amales Tripathi, Trade and Finance in the Bengal Presi- dency, 1793–1833 (Calcutta, 1979); Tony Webster, The Richest East India Merchant: The Life and Business of John Palmer of Calcutta, 1767–1836 (Woodbridge, 2007). working for wages in colonial bengal’s indigo industry 49

What is known about their social and economic backgrounds? What was life in and around the factories like? How was the labor process organized? What did laborers earn? These are some of the empirical questions we will be concerned with in this chapter.

Field labor and factory labor

In order to understand how this agro-industry operated, it is important briefly to compare labor relations during its two successive phases: agricultural (growing indigo plants from seeds) and industrial (producing dye from plants). Field labor and factory labor remained quite distinct throughout the heyday of indigo production. Field labor. After initial and unsuccessful experimentation with other forms of labor management in the late eighteenth century, indigo production settled into a fixed pattern. Field labor, or the growing of indigo plants, was largely a smallholder affair based on household labor, sometimes with some additional, seasonal wage labor.7 Household heads managed this part of the indigo labor system but they did not have the luxury of deploying their household’s labor freely. Their freedom was restricted in two ways. They were answerable to landlords/tax collectors (zamindars), who owned the land and were in charge of collecting the colonial state’s taxes from them. Household heads needed to be able to pay those taxes and therefore needed cash. In addition, they frequently took advances on the indigo crop from (mostly European) factory managers and thus became indebted to them and forced to continue planting indigo. Indigo marked the moment when, for the first time, the choice of cropping pattern was taken away from cultivators in rural Bengal on a large scale.8 Factory labor. Indigo factories (nil kuthi) were located near the fields on which indigo was grown. There was a good reason for this: after harvesting, the indigo content of the leaves and stems diminishes rapidly and therefore processing must commence within hours. Given the slow transport of the period—boats, oxcarts or men—, this meant that factories were scattered all over the countryside. The importance of water transport was obvious to a visitor in 1838: ‘Jessore is a major indigo district; factories

7 In Bengal proper, where the government barred indigo entrepreneurs from owning large tracts of land, cultivation of indigo on their own land (known as nij cultivation) was of minor importance. 8 Sah, “Features of British Indigo,” p. 70. 50 willem van schendel proudly display themselves all along the riverbanks.’9 Despite this physical proximity of field and factory, however, their labor systems were entirely dissimilar. Factory labor was under the control of capitalist enterprises. It was wage labor, differentiated into a hierarchy of different tasks.

The Bengal indigo enterprise

Investors provided the capital to set up indigo factories and pay the wages of factory laborers. There were hundreds of shareholder companies in Bengal, most of them foreign-owned and carrying names such as Messrs Mackintosh & Co, or Bengal Indigo Company. Some of these companies focused entirely on indigo, owning between one and dozens of factories, others had interests in many different cash crops and commercial activities.10 Indigo companies typically had a tiered structure. A director in Kol- kata (Calcutta) was in charge of all activities; below him were managers of regional divisions (‘concerns’).11 In 1829 the first official survey of the indigo industry in British India found that there were 662 indigo concerns in the colony—three quarters in Bengal (72 percent) and the rest ‘Upcountry.’12 Below each manager were assistant managers in charge of local subdivisions (often confusingly named ‘factories’ but actually clusters of factories in different villages). All these employees were Europeans, or, occasionally, Armenians or Anglo-Indians;13 often they were referred to

9 P. Arriëns, Dagboek eener reis door Bengalen: in 1837 en 1838, met eenige vrijmoedige opmerkingen betreffende beginselen van koloniaal bestuur (The Hague, 1853), p. 217. 10 There were also indigo factories owned by Indian or Armenian entrepreneurs, most of them small. Some employed European (assistant) managers (Thomas Machell, “Journals of Thomas Machell” (1840–56) (British Library, India Office Private Papers, Mss Eur B369), vol. 3, 340–1/50; vol. 4, 232–3/126; vol. 4, 301/161). A notable exception to the pattern was the very prominent entrepreneur Dwarkanath Tagore (Kling, Partner in Empire). See also “Rights of Indigo Planters,” West Bengal State Archives, Kolkata, India (Judicial Depart- ment (Criminal) Proceedings, 9 June 1830, Nos. 1–25). 11 For a description of the manager’s work, see Rural Life in Bengal; Illustrative of Anglo- Indian Suburban Life [. . .] Letters from an Artist in India to his Sisters in England [attributed to Colesworthey Grant] (London, 1860); and W.M. Reid, The Culture and Manufacture of Indigo; with a Description of a Planter’s Life [. . .] (Calcutta, 1887), 21–33. 12 Fifty-seven per cent were in Bengal proper (24 per cent in East Bengal, 33 per cent in West Bengal), 15 per cent in Bihar and 28 per cent Upcountry (United Provinces). The largest concentrations of indigo concerns were in the districts of Dhaka (122), Murshidabad (51), Jessore (32), Bhagalpur (29) and Purnea (28). Calculated from “Indigo Planters “(1832), British Library (Asia, Pacific and Africa Collections; F/4/1277; Judicial B; LP; 51242). 13 The 312 names mentioned with country of origin in the 1829 survey show the following breakdown: England & Wales: 144, Scotland: 66, India (including entries such as Hin- dostan, Country-born, Indo-Britons, Chinsura, Madras, East India, Native of this Country): working for wages in colonial bengal’s indigo industry 51 as the shaheb lok (gentlemen).14 Finally, each village factory was managed by a factor or supervisor (gomashta), who was an Indian and the foremost of the amlah (‘native staff’).15 In the day-to-day running of an indigo factory, the assistant manager and the supervisor were key (see Box ‘Doing the Rounds’ for an impression of the assistant manager’s lifestyle).

Doing the Rounds

An assistant manager would constantly travel on horseback between his head factory and a handful of smaller village factories (‘out-factories’), which together made up the ‘factory’ in his charge. This is how Thomas Machell described the scene for his father, back in Britain, in 1851: in moving from factory to factory I am attended by all my servants, who have, as you may imagine, a pleasant time of it, trudging from one factory to another, seeing that the nearest of them are at least eight miles apart. It would amuse you to see the procession they make: two coolies carrying a bed, one carrying a table, another a chair, a third carrying a box of clothes, a fourth braving the hookah box, a fifth carrying a gun, three or four carrying the cooking apparatus, plates, dishes, &c. Then follows the Khansamah [house steward] with his umbrella, not that he feels the heat of the sun, in fact at this time of the year [January] he would be more comfortable without a shade but there’s dignity in a cotton umbrella. The bearer follows with a large bunch of keys hanging at his girdle, where he got them I don’t know for I have not a single key belonging to me except that of my writing desk. He also carries his master’s walking stick. Then follows the Cook with his iron spit for roasting fowls and lastly comes the Mhetur [sweeper] bringing up the rear with a broom under his arm and my solitary dog Dessie which he leads with a string. But I am a very moderate man in the way of servants and necessaries. [Charles] Sage used to have a train of twenty or thirty coolies, at least two Khidmutgars [table servants], two bearers, a shikaree [game servant], and half a dozen hangers on.16

40, Ireland: 29, France: 17, other European countries: 10, other Asian countries: 5, America: 1. Calculated from “Indigo Planters” ((1832); British Library). 14 Machell, “Journals,” vol. 2, 172/109. 15 For a description of the amlah, see also Reid, Culture and Manufacture of Indigo, pp. 34–41. 16 Machell, “Journals,“ vol. 4, 1–2/6–7. For a description of the assistant manager’s life, see also Reid, Culture and Manufacture of Indigo, pp. 3–20. An account of the houses of different levels of managers can be found in “Indigo Planters and Plantations,” The Penny Magazine of the Society for the Diffusion of Useful Knowledge (1843), pp. 178–80. 52 willem van schendel

The relationship between the European manager and his Indian second- in-command (gomashta) was fraught. The manager had to leave the day-to-day running of out-factories to gomashtas but he was constantly nagged by suspicions of being cheated. Their rocky relationship typified the indigo industry but also presented a classic colonial interface of Euro- pean and Indian styles of exerting power (see Box ‘The Gomashta (as Seen by a Manager)’ for one side of the story).

The Gomashta (as seen by a manager)

‘Ye Ghomastah, or as the poor people call him, Dewanjee massi, is the head amlah in every factory, who superintendeth everything. He looketh after ye Cultivation, reviseth accounts, payeth the wages and maketh out estimates for the monthly expenses. He is much feared and looked up to by all the people in ye country, for, in addition to his extensive authority, he is generally a high caste Brahmin. Nevertheless, he scrupleth not to accept bribes which come in the form of Salami, or offerings from an inferior, Nussurs, or presents from a servant to his master, &c &c, for the titles of these things are, like the ladies’ names for rouge, various but all meaning actually the same thing. He therefore requireth sharp looking after, although in spite of all you can do he will not the less avail himself of his perquisites, which in a large factory he generally contriveth shall far exceed his pay. He is often a small land owner himself and by means [of] the various pickings he generally succeeds in the course of a few years to amass a very pretty little sum of money and property. He also manages to provide for a few poor relations, nephews and cousins by obtaining inferior employment for them about the factory, for which they pay him according to their abilities and aid him in forward such pieces of roguery as he might not care that others should know of. In short, he is as full of slippery tricks as an eel and as difficult to be caught out but when he is first caught then the Sahib if he be knowing faileth not to squeeze him well, seeing that for one trick you have found out he hath played fifty of which you hear nothing. His pay at head factories is about twenty-five rupees, at out-factories less in proportion, nevertheless he realizes a hundred or thinketh he hath not done justice to himself.’17

17 Machell, “Journals,” vol. 2, 173/110. working for wages in colonial bengal’s indigo industry 53

The factory

In Bengal, indigo companies set up their factories according to a model developed in the Americas in the eighteenth century—and known as the ‘West Indian’ or ‘Caribbean’ model. The production process needed only minor technical adaptations for the climatic conditions of Bengal but, by contrast, labor relations were very different. We will first look briefly at the production system.18 It was based on relatively simple technology and little technological change over time. Indigo dye was the outcome of successive manipulations of freshly cut indigo plants. The leaves and stems would arrive at the factory in boats or by oxcart and immediately be submerged in a steeping vat (see Box “Steeping the Indigo”).

Steeping the Indigo

“Well, it’s a jolly time,” wrote a British factory manager in his journal in 1852, “this opening of the vats, the creaking of the hackeries or bullock carts loaded with Indigo, the cries of the drivers, the restless clacking of the pumps. A busy time it is and pleasant to behold in the calm clear evening the busy scene about the long deserted vats. Here are men staggering along with heavy beams used for pressing, there are women and boys, bringing in sheafs of green plant and fling them into vats where men are stowing these closely. Now the sun goes down and still the busy scene goes on; the red glow of twilight and the white cloth’s of the jocks in office, who make the most of their brief authority, show pleasantly among the dark skins and naked bodies of the laborers. Now with much clamour and shouting they bring the great levers and amidst shouting and singing and various strange cries they heave down the huge beams and secure them with Iron pins. Now they open the sluices of the great reservoir and the gleam of torches flash on the troubled water. It rushes forth in its narrow channels and pours into the plant fill’d vats. Hurrah the work is done, the torches flicker for a while, the rushing water is stayed, the vats are fill’d, the torches are extinguished, gradually all is silent and the fire flies twinkle amidst the dark foliage of the trees. Rejoicing in the stillness and beauty of the night, from time to time a white figure may be seen like a ghost flitting about the vats like the unquiet spirit of a deceased Indigo planter.”19

18 For more detailed descriptions, see Pierre-Paul Darrac and Willem van Schendel, Global Blue: Indigo and Espionage in Colonial Bengal (Dhaka, 2006); and John Phipps, A Series of Treatises on the Principal Products of Bengal, No. 1: Indigo (n.p. [Calcutta], 1832), 119–25, 141–53. 19 Machell, “Journals,” vol. 4, 203–4/111–2. 54 willem van schendel

Some twelve to sixteen hours later, the indigo-laden water was drained off to a beating vat.20 Laborers would enter the vat and stir and beat the mixture till it turned blue as a result of oxidation (See Box “Beating the Indigo” and Figure 1).

Beating the Indigo

“There is a sound of wooden mallets hammering at huge plugs and then a rush of orange coloured water, rushing and roaring in [the] night whilst the group of white robes are seen watching the rush of the golden tide. It is the planter and his men watching the letting off of that fermented juice of the Indigo from which the dye is to be made [. . .]. But now see a string of dark forms emerging from the shadow of the trees, each armed with staves, looking like a file of red Indians in the red glare of the torch light. However, they are only peaceably disposed coolies just turned out to beat the vat. First they fling out the plant from the steeping vats and then one after another leap into the beating vats, in which the water lies which was let off from the steeping vats, and now commences the work of the day, for to and fro, round and round, those men up to their waists in the dark green liquid stir the purls with their sticks. Now they work it into waves and now it foams like a huge washing tub full of lather. Again you look and blue streaks are seen in the lather and so they work on, now in silence, now breaking out in wild chants, keeping time with the working of the water, until the water turns to a dark inky blue.”21

The blue fluid would then be left so that the dye particles could precipi- tate. The excess water was drained off and the remaining blue slush transported to boilers.22 After boiling and straining, the indigo was put into large wooden presses to squeeze out the last water. Then it was cut in cubes (“cakes”) and left to dry for weeks. Once this was done, the cakes were packed in chests and sent off to markets overseas.

20 “Note sur la Culture et la Fabrication de l’Indigo au Bengale, dans les districts de Jessore, de Kichenagor et de Dacca” (1824), (Archives Nationales d’Outre-Mer, Aix-en- Provence, France (Dossier 527—“Affaires Générales—Indigo 1820–1863”; GEN/55, FM, Généralités, Affaires Économiques, Produits et Exploitations Industriels; Les Séries Docu- mentaires; Ministère des Colonies)). 21 Machell, “Journals,” vol. 4, 204–5/112. 22 Boiling was not part of the Caribbean or American way of processing indigo. In Brit- ish India it was widely used to prevent further fermentation and to purify the indigo. W. Roxburgh, “Account of a New Species of Nerium, the Leaves of Which Yield Indigo . . .,” Transactions of the Society . . . of Arts, Manufactures, and Commerce (London), Vol. 28 (1811), pp. 251–307; cf. W.L. de Sturler, Handboek voor den Landbouw in Nederlandsch Oost- Indië . . . (Leiden, 1863), pp. 851–8. working for wages in colonial bengal’s indigo industry 55

Figure 1. Laborers beating the indigo solution23

Indigo factory labor

The production of indigo dye was labor-intensive. One of the characteristics of the indigo industries that had previously flourished in the Carib- bean and the U.S. South had been their dependence on slave labor. In Bengal entrepreneurs could not make use of slaves and they had to recruit and manage factory labor differently. This made the indigo factory much less of a “capitalist enclave” than the American plantation had been. It was more dependent on the rural social context and had to develop strategies to attract, retain and remunerate labor that fitted the social realities of Bengal. A major difference was the use of both permanent staff and seasonal labor. Indigo processing is a seasonal activity, following the harvest. In the Americas, plantations had produced diverse commodities and slaves had been put to work to grow or manufacture these in indigo’s off-season. In Bengal, this was not the case. Indigo factories were exclusively in the business of dye production and they needed only a little labor, mainly for

23 Rural Life in Bengal, p. 127. 56 willem van schendel

Table 1. Colgong indigo factory: Employment of day laborers in 1827–1828 (in workdays/month)24 20000 18000 16000 14000 12000 10000 8000

Days of employment 6000 4000 2000 0 123456789101112

Months (November 1827–October 1828) maintenance and repair, during the off-season. As a result, most laborers were dismissed after the production season. Take the Colgong factory. In 1827–28 it reported that it had a permanent staff of 53 and that it employed hundreds of additional day laborers during the production season, together accounting for 92,000 workdays a year (see Table 1).25 This labor force was anything but an undifferentiated mass. Each village factory was a complex microcosm of well-defined labor tasks. The Colgong factory’s permanent staff was employed in 13 distinct jobs and its seasonal laborers were of 14 different designations.26 Colgong was a large

24 “An Abstract Statement,” p. 290. 25 “An Abstract Statement of the number of daily Coolies or Labourers, Boats, Carts, &ca &ca employed by the daily at Colgong during 1827–28 shewing the number of each in each Month”; in “Indigo Planters” ((1832); British Library), pp. 290–1. 26 Colgong (located in Bhagalpur, Bihar), was the head factory of a “concern,” so its establishment was larger than most. – The following were on monthly wages: 1 Superintendent European, 1 Asstt. Ditto Indo-Briton, 1 Gomashta, 1 Asstt. Gomashta, 1 English Writer, 3 Native Writers, 5 Ameens to measure cultivations, 11 Zillahdars to look after ditto, 11 Hurkurahs to assist ditto, 2 Chokeedars, 2 Peons, 10 Syces or Grooms, 4 Grass Cutters. – Day laborers included: Coolies or Laborers, Boys, Women, Carpenters, Blacksmiths, Bricklayers, Tilers, Tailors, Sawyers, People employed in factory boats, Hire boats bringing pants, Hire carts ditto, Bullocks ditto, other. “An Abstract Statement,” pp. 290–1. working for wages in colonial bengal’s indigo industry 57 factory but labor roles in smaller factories were equally complex. Thus a minor factory in Nadia district (Bengal) employed 14 different types of monthly laborers and 17 types of day laborers,27 in addition to some 7 to 18 types of household servants.28 Four broad types of labor could be distinguished: productive, managerial, coercive and domestic. First, the labor required to produce indigo was provided by hundreds of people usually described as “coolies.” These were men, women and children who took on many different tasks, from unloading indigo plants from boats or oxcarts to pumping up water, from beating indigo water to cleaning out vats, and from operating the indigo presses to cutting up and drying indigo cakes (see Figure 2). Most of them were seasonal laborers, employed during the dye production season. In addition to these all-purpose laborers, there were specialized jobs for sawyers, carpenters, blacksmiths, bricklayers and tile-makers, whose task it was to keep the factory in working order. Most of these were permanent staff on monthly salaries. Second, supervising these producers was an array of foremen, office workers and managers. The labor contractor (sardar) managed the laborers and the colour master (rang mistri) and his assistant superintended the beating of the fermented indigo water. Office staff included bookkeepers and peons. Finally, the gomashta and the (assistant) manager oversaw the entire process. Third, indigo factories needed to protect themselves from attack by local landlords, rival factories and itinerant robbers. They also needed a coercive force to ensure that peasants sent a steady stream of raw material to the factory. To this end, each factory employed men armed with clubs,

27 On monthly wages were Ghomastah, Naib, Piada, Sudder Ameen (or Mot Sircar), Mhorie, Under-Mhorie, Kranee, Ameen, Inkidgur (or Kelassie), Latteal, Soorkie Wallah, Chowkedar, Burkandauz, Dak wallah. Day laborers included Rung Mistree (or Colour Master), Rung Mistree’s mate, Pumper, Sirdar (or Foreman), Vat Beater, Sweeper, Weeder, Porter, Gardener, Grass Cutter, Moor- gee wallah, Beari wallah, Women and Children for light work around the factory, Herds- man, Pig Keeper, Carpenter and Blacksmith. (Machell, “Journal,” vol. 2, 171–88/109–20). 28 This factory manager distinguished between “standard servants,” whom you would find in every planter’s household, and “extras.” The former were the Khansamah [house steward], Khidmutgar [table servant], Bearer, Bawarchee [Cook], Doby [Washerman], Beesty [Water carrier], Mhetor or Mhetranee [Sweeper]; the latter were the Hookah- burdar [waterpipe servant], Khansamah’s boy, Ayah, Abdar [water cooler], Mussalchi [kitchen boy], Sirdar, Durzee [Tailor], Coachman, Shikaree wallah [game servant], Mahout [Elephant servant] and Punkah-wallah [hanging fan mover] “&c &c &c &c ad infinitum.” (Machell, “Journal,” vol. 2, 149–69/98–108). 58 willem van schendel

Figure 2. Laborers removing used indigo plants from steeping vats, 1850s29 spears, or guns (lathial, surkiwallah, barkandaz), who acted as a police force and night watchmen and, at times, as mercenaries selling their mar- tial services to the highest bidder.30

29 Rural Life in Bengal, p. 126. 30 In 1830 the Magistrate of Dhaka (Dacca), a major indigo-producing district, summed up the challenges that such groups of armed men, and the indigo factories that employed them, posed to state law enforcement: 1st. The first [evil] is the custom of entertaining at European and native Indigo Fac- tories a number of Khlasseys, Tacazageers, and persons under other denominations, inhabitants generally of other Zillahs [districts], and sometimes proclaimed offenders, escaped Convicts, or persons discharged from the Jails. These people find a secure Assylum in Indigo Factories, from whence they issue, under cover of the night, and are guilty of heinous offences—Dacoities [robberies], Thefts, and even Murders have been brought home to such Factory retainers. Their real characters are seldom, it is to be presumed, known to the Planters; and their desperate proceedings, of course never—Police officers are afraid to enter Indigo Factories, and as Planters in general, are not extremely ready to afford the Police assistance, especially when suspicion falls on their own Factory servants, these offenders too frequently escape detection, and so go on with impunity, in defiance of the best efforts of the Magistrate. 2nd. Organized gangs of Brigands, denominated “Lateals,” consisting of from 1 to 200 men under regular Sirdars, live by hiring themselves to fight the battles of Indigo Planters, it is to be hoped, are not now as commonly had recourse to, as formerly in the East of Bengal; nevertheless, some provisions appear to be necessary, the more working for wages in colonial bengal’s indigo industry 59

Finally, there was always a bevy of servants to make life comfortable for the factory manager. Some of these worked in his household— e.g. the house steward (khansama), bearer, cook and ceiling-fan mover (pankhawallah)—while others worked in the stables, garden, or factory grounds.

Wages

Reliable information on what employees in indigo factories earned is hard to come by. The following figures are based on entries in Thomas Machell’s diary for factories in western Bengal in 1850–2.31 The lowest-paid were the productive laborers, the “general coolies” in charge of various tasks such as steeping and beating the indigo. They were employed for the season and then earned three Rupees (Rs.3) per month. Skilled workers such as carpenters were employed throughout the year and earned at least Rs.6.

effectually to put them down; and any persons, native or European, proved to have employed such persons, on any trial before a Magistrate or Commissioner, should be severely dealt with. (Letter from H. Walters, Magistrate of Dacca, to H. Shakespear, Secretary to the Govern- ment in the Judicial Department, Fort William, 12 February 1830, in “Indigo Planters” ((1832); British Library), pp. 816–7. 31 Machell, “Journals,” vol. 2, 173–211/110–32; 151–60/99–103, 163–83/105–14; vol. 4, 83/47, 162–3/90, 282–3/153, 304–5/164. John Phipps, who reported on an earlier period, gave slightly lower wage figures. Phipps, A Series of Treatises, 155, 159, 162. Colesworthey Grant (Rural Life in Bengal, 138–140) gave the following wages in the Mulnath concern for the late 1850s. He distinguished three categories: Bhoona (immigrant from Chhota Nagpur), Midnapore (immigrants from the southernmost district of Bengal) and Dasee (“belonging to this part of Bengal” [= deshi]). Bhoonas were paid less, and Midnapore immigrants more, than locals for the same tasks. Rs. 2/month: Godown coolie boy (Bhoona), Ghât-motiya woman for carrying the plant (Bhoona), Sirdar of Coolies (Bhoona), Plant Measurer (Dasee). Rs. 3–3.75/month: Vat coolie or beater (Bhoona), Boiling-house man (Dasee), Pressing man (Dasee), Stoker or Furnace man for the boilers (Dasee) and Chinese pump worker (Dasee). Rs. 4/month: Vat coolie or beater (Dasee), Plant Measurer (Midnapore), Carpenter (Dasee), Cake cutter (Dasee), Ghât-manjee (Dasee), Pin coolie—attendant on the plugs etc. (Dasee), Sirdar, or superintendent of hackries [carts]. Rs. 5/month Vat coolie or beater (Midnapore). Rs. 7,50/month Hackrey man—for service of self and cart for bringing in the indigo (Dasee). Rs. 9/month Boat man—for service of self and boat for bringing in the indigo (Dasee). Rs.5–30/month Krânee or writer, according to duties. Rs. 25–50/month Gomastah. Rs.50/month Naib. 60 willem van schendel

Managerial and office jobs fetched better wages. Bookkeepers earned Rs.8, clerks Rs.10 and the factory supervisor (gomashta) earned Rs.25 or more. Coercive labor was also better paid. Armed retainers earned Rs.7.5 “when not fighting” and their wages were doubled on the days when they were “at work.” Renumeration for domestic work varied considerably. The lowest-paid house servants, e.g. the ceiling-fan mover (pankhawallah) or the torch- bearer (masalchi), were paid Rs.3 a month but most servants earned Rs.4 to Rs.5. The butler or house-steward (khansama) earned Rs.10 or more and the coachman could earn about Rs.15 in the best-appointed factories. The Europeans earned their salaries in sterling. Their income fluc- tuated with the success of the season and the size and productivity of the factory. An assistant manager’s salary could be around £60 per year (which equalled Rs.50 per month) or more. After a fair season Machell, then manager of the Rudrapur [Rooderpore] factory in Jessore district, earned £240 (or Rs. 200 per month), half of what he would have earned in a really good year. It was the cheapness of labor in Bengal that made indigo such a profitable crop.32 At the same time it hampered technological innovation. For example, factories used the so-called Chinese pump to raise water from the water reservoir to the level of the steeping vats. The pump consisted of a revolving spindle which worked a chain of wooden buckets travelling up over a roller, releasing the water, and down again (see Figure 3).33 This cumbersome contraption needed six men treading constantly to keep it in motion: “Well, this is the way we raise our water and as we have to raise a great quantity and to a considerable height, we have six of these pumps at work and six times six = 36 men employed at it. Labor is so cheap here that we find this the most effectual and cheapest plan.”34 Chinese pumps remained in use till the end of indigo production in Bengal.

32 Indigo entrepreneurs regularly referred to this. Thus, in a letter of 1829 to the Gov- ernor-General, a number of leaders of the industry ascribed the “successful prosecution of [the indigo] industry” (despite government not allowing indigo entrepreneurs to hold land in Bengal) to the “extraordinary fertility of soil and cheapness of labour”. Letter of Palmer & Co. et al. to W.C. Bentinck, Governor General in Council, Calcutta, 28 Janu- ary 1829. West Bengal State Archives, Kolkata (Proceedings of the Revenue Department, 17 February 1829, Nos. 2–4). 33 For a detailed description dating from 1823, see Darrac and Van Schendel, Global Blue, pp. 95–7. 34 Machell, “Journals,” vol. 2, 182/114. working for wages in colonial bengal’s indigo industry 61

Figure 3. Laborers working a Chinese pump at an indigo factory, 1850s35

Recruitment

Supply of labor does not figure as an issue in the accounts of entrepreneurs, managers and observers of indigo factories in Bengal, leaving the impression that labor was easy to recruit.36 But where did the laborers come from? There are reasons to suppose that the proportion of local laborers in the factories was low. Many nineteenth-century reports remarked on the fact that in most parts of Bengal the proportion of households dependent on labor as an occupation was much lower than in other regions of British India. And yet, the first comprehensive investigation, based on about 100 village surveys, concluded that by the late nineteenth century, this picture no longer fitted realities: “[the] statement frequently made in official reports [. . .] that there are very few landless workers in Bengal generally or

35 Rural Life in Bengal, p. 124. 36 It is important to realize that it is hard to generalize because of large differences in the supply of labor between different regions, from one village to another, from season to season, and over time, as argued (for North Bihar) by Jacques Pouchepadass, “The Mar- ket for Agricultural Labour in Colonial North Bihar, 1860–1920,” in Mark Holmström (ed.), Work for Wages in South Asia (New Delhi, 1990), pp. 11–27. 62 willem van schendel in particular districts, appears to rest on no sufficient foundation.”37 This “Dufferin Report” dates from 1888, however, well after indigo production had declined in Bengal; it is unclear how many rural households depended on labor during the heyday of indigo production (1800–1860). In any case, laboring households in rural Bengal seem to have had a preference for agricultural work, possibly because of marginally better wages (including wages in kind) and long-term security of employment. Muslims also may have had specific problems with factory work. Maibulla, a villager from Jhenaidah district, told the Indigo Commission in 1860: “The Sahib insists on our working in the [indigo beating] vats; we are Mussulmans; and if we do this we cannot get our daughters married.”38 Another reason to suppose that indigo factories did not employ many laborers from the direct vicinity is that indigo laborers lived on the premises rather than in surrounding villages.39 There seems to have been only one category of laborers who were drawn from the surrounding areas of the factory: owners of ox-carts and boats who delivered the indigo harvest at the beginning of the processing season. These were often indigo cultivators themselves. They seem to have been under the system of coercion characteristic of the agricultural side of indigo production: although they were paid for their services—“hire of carts” and “hire of boats” often appear as items in indigo factory accounts—there is also evidence that they were often coerced.40 Most factory employees were from further afield, however. The permanent staff consisted of a mix of Bengalis from different parts of the province and people from “upcountry” (regions to the northwest). Their many different roles reflected the social differentiation of rural society. Certain types of labor were the preserve of (or forbidden to) specific social groups,

37 Report on the Conditions of the Lower Classes of Population in Bengal (Calcutta, 1888), p. 12. For an analysis of this report, also known as the Dufferin Report, see Willem van Schendel and Aminul Haque Faraizi, Rural Labourers in Bengal, 1880 to 1980 (Rotterdam, 1984). 38 Evidence given by Maibulla, inhabitant of Lakhandi (Jenaida Police Office, Jessore District (now: Jhenaidah District), called in; and examined on oath. Report of the Indigo Commission, 1860, p. 207. 39 An account of laborers employed in an indigo factory near Krishnagar during ten days in 1825 makes a distinction between “dasee labourers” (deshi = Bengali) and “Boona labourers, namely coolys, women and chokras [boys]”. The Boonas, or immigrants from Chhota Nagpur, made up almost 70 percent of this labor force. A. Abraham, “Report 3d, 21 October 1825, Augurdeep division”; British Library, London (Verlée Collection, Mss Eur F193/94, India Office Select Materials). 40 For example, evidence given by Ram Chunder Biswas of Baroda village, Nuddeah district before the Indigo Commission. Report of the Indigo Commission, 1860, pp. 151–2. working for wages in colonial bengal’s indigo industry 63 leading to a highly fragmented labor market. Manual labor was considered beneath the dignity of high-caste Hindus, well born Muslims (ashraf ) and anyone with a formal education. It was not always easy for members of these groups to find suitable work, however, leading to scenes such as the one described in Box “In Search of Work.” Certain domestic servants were traditionally Muslims. Other jobs around the factory, such as sweeping, were the preserve of members of particular low-caste Hindu groups.

In search of work

“A young Brahmin came to me to day,” wrote Thomas Machell in his diary in 1850, “with evidently a very good opinion of himself and made his salaam.” “Well, Sir, who are you?” said I. “Amaschunder Mookerjee is your servant’s name.” “Well, what do you want?” “Your slave has come to make his salaam.” “Very good, anything else?” “Light of the World, your slave has been long in search of employment.” “Can you read and write?” “Your slave reads and writes Bengali.” “Do you speak English?” “No, Your Highness, but you are the Prince, you are the Father and Mother of the country, and your servant seeks employment.” “Well, well, what can you do?” “Your slave can do everything. Every business your servant comprehends.” Well, this is Bengali modesty with a vengeance, so thought I, and told him that really I thought I had no situation in my service worthy of such a genius. Now what do you think he wanted? The Head Ghomastah’s place? No such thing. Not even a Mhorie’s [clerk’s] berth. He merely came to ask for an ameen’s berth (a man who looks after the cultivation of a division of a factory) and this is the style in which they generally prefer their requests for even so small a salary as ten shillings a month. How are the mighty fallen. A Brahmin of the great family of Mookerjee pre- ferring a position for a small appointment of ten shillings a month to a man of no caste at all. The poor descendent of the proud line of powerful Indian priests petitioning for an ameen’s berth and refused. But I give the above as a specimen of what occurs so often that I have, I believe, forgotten to register it before.”41

41 Machell, “Journals,” vol. 3, 352. 64 willem van schendel

Enforcement tasks were the domain of Hindi-speaking immigrants from upcountry. Groups of them, armed with clubs, daggers and spears, moved about the Bengal countryside, offering their services to landlords and indigo factories, or acting independently as brigands. Considered dangerous but indispensable, they were the sinews of the indigo industry—keeping cultivators in line and warding off attacks by landlords, competing indigo entrepreneurs and roving bands of raiders.42 Most general indigo laborers (“coolies”) were also immigrants. They came from central India, especially the region of Chhota Nagpur. Although they belonged to various ethnic and linguistic groups (today known as Santhal, Munda, Bhuiya, Oraon, and many more), in the indigo factories of nineteenth-century Bengal they were collectively known simply as Buna (or Boonooah, Bonwallah, etc.)43 (see Figure 4). Physically distinct and cul- turally despised, they lived in separate hamlets around the factory.

42 “|Ye Latteal and Soorkie Wallah are two gentlemen of the Piada genus who are employed in times of riots to keep refractory Ryots in order or to bully the Saheb for they will fight indifferently on either side and it is a common thing when one side has discharged them to go and tender their services to the opposite party in order to loot (i.e. plunder) the village of their former employers. Living in gangs they are a formidable set of men and their lawless life induces lawless habits. When they have our honest employment as the[y] call fighting for or against a factory they do a little in the Dacoity line (i.e. Highway robbery and Burglary). They do not however often try a European house for they well know the telling effect of a rifle ball. The Lattial is so called because his weapon is the Latty or Quarterstaff with which he is pretty expert. The Soorkie Wallah or Spearman is a more dangerous enemy wielding an iron javeling some six feet long. He throws it with great force and precision insomuch that within fair range he will spit his man clean through the midriff [. . .]. Of course one never sees anything of this sort. It is the Planter’s business to be out of the way on these occasions [. . .]. One does hear of factories occasionally paying four or five hundred lattials but it always happened a long time ago: never nowadays although there are curious entries in the books of so many hundred rupees for law business when there happens not to have been any cases in court”. Machell, “Journals,” vol. 2, 189–90/113; cf. vol. 4, 4–5/8. 43 “The Boonooahs are the inhabitants of the Boons, or Jungles. They are a lively race, fond of music, dancing and drinking, eat swine’s flesh and have very loose notions about religion or morality. They are very ingenious in making mats, nets, baskets, &c., and are supposed by some to be the aborigines of Hindoostan” (Machell, “Journals,” vol. 4, 141). Another common term for them was Dangar (or Dangur). See e.g. W.W. Hunter, Statisti- cal Account of Bengal (London, 1868), vol. 1, pp. 226–8; “Maori” (James Inglis), Sport and Work on the Nepaul Border, or Twelve Years Sporting Reminiscences of an Indigo Planter (London, 1878), pp. 14–6. working for wages in colonial bengal’s indigo industry 65

Figure 4. A Buna factory laborer, ready to beat indigo, 1850s44

A complex of economic and social factors forced inhabitants of Chhota Nagpur to look for seasonal work outside their region, mostly as agricultural laborers and forest clearers in Bengal.45 They were quick to take advantage of the emerging indigo industry.46 By 1827 the pattern was well established:

44 Rural Life in Bengal, p. 128. 45 Detlev Schwerin, Von Armut zu Elend: Kolonialherrschaft und Agrarverfassung in Chota Nagpur, 1858–1908 (Wiesbaden, 1977); Biplab Dasgupta, “Agricultural Labour under Colonial, Semi-Capitalist and Capitalist Conditions: A Case Study of West Bengal,” Eco- nomic and Political Weekly, 19, 39 (29 September 1984), pp. A129–33 and A136–48. 46 “In every part of Nuddea little communities of Santals, Dangars, or other hill-men, may be found living apart from the Hindus, and preserving their national customs in the middle of the lowland population. Many indigo factories in the eastern districts have villages of these western highlanders. A family of them makes its appearance wherever manual labour is wanted, builds its leaf huts in a few days, and before the end of the month feels as much at home as if it were still among the mountains. Patient of labour, at home with nature, able to live on a penny a day, contented with roots when better food is not to be had, dark-skinned, a hearty but not habitually excessive toper, given to pig-hunting on holidays, despised by the Hindus, and heartily repaying their contempt, the hill-men of the west furnish the sinews by which English enterprise is carried on in Eastern Bengal. 66 willem van schendel

The[y] emigrate in great numbers annually in search of employment, and are entertained by indigo planters and others. They are generally preferred to the laborers of other parts of the country, on account of their performing more work and at a lower rate. In a family consisting of four or five persons, two are left at home to take care of the family affairs, and cultivation, &c, the rest go abroad to seek service.47 Although mass migration is a prominent theme in the historiography of Chhota Nagpur, this early industrial work in the indigo factories of Bengal has not received much attention. Labor migration from Chhota Nagpur is usually analysed in relation to an industry that developed later: tea. The tea plantations in Assam drew large numbers of labor migrants from the region, but only after 1860. This was the time when indigo production in Bengal declined (although it continued vigorously in Bihar). To what extent the rise and fall of these industries was linked to changes in the supply of labor from Chhota Nagpur (and to the development of infrastructure, notably railways) remains to be explored, but it is a safe bet that without the influx of cheap labor migrants from the late eighteenth century the indigo industry could not have flourished as it did.

Making indigo laborers visible

Most accounts of the indigo industry render factory laborers almost completely faceless—mere “living machinery” in the process of dye production.48 Being largely illiterate, laborers themselves have left no accounts

Many of them come from the central highlands, where the population is permanently just one degree above absolute starvation, where the extension of tillage is only possible after a considerable outlay of capital in dicing tanks, where the winters are severe, where cutaneous diseases and every infirmity common to half-starved hunting communities are rampant, and where the political disaffection which springs from a chronically hungry stomach is never unknown. They settle in a land where Nature has done her utmost to render unnecessary the toil of man, where good wages are always to be had in ready money, and where the very jungle produces as ample a subsistence as their little cultivated patches at home. Every winter, after the indigo is packed, numbers of the labourers visit their native villages, and seldom return unaccompanied with a train of poor relations, who look forward to the wages of the spring sowing season as the soldiers of Alaric contem- plated the spoils of Lombardy. The law of supply and demand operates in the long-run as effectively, although more tardily, in the valley of the Ganges as on the banks of the Mersey or the Clyde [. . .] the hill-races, uncivilised though they be, are sagacious enough to find out and frequent the districts where they can get the highest price for the one marketable article that Providence has given them—the work of their hands”. W.W. Hunter, Statistical Account of Bengal (London, 1868), vol. 1, pp. 226–8. Cf. vol. 7, pp. 40, 167–8, 172. 47 S.T. Cuthbert, “Extracts from a Report on Chota Nagpore,” Journal of the Royal Asiatic Society of Great Britain and Ireland, 8 (1846), p. 413. 48 The term is Colesworthey Grant’s in Rural Life in Bengal, p. 138. working for wages in colonial bengal’s indigo industry 67 of their lives and they are often overlooked in contemporaneous portray- als of indigo production. In a well-known book, W.M. Reid provides a detailed description of indigo manufacture in which he presents the factory manager as the sole actor: “During the day, as the plant comes in, measure and stack it up in the shade that you may be able to load your steeper all at once in the afternoon [. . .]. I repeat, stow the vats so as to let the impregnated water run off freely, keep the plant horizontally square as you fill up.” Tellingly for a man who spent many years surrounded by indigo laborers, he disregards the fact that it is they, and not the manager, who carry out all these tasks.49 If we go by the accounts that he and other indigo managers have left behind, most appear to have considered laborers not worthy of much thought. As a result, we know hardly anything about how indigo laborers lived or what they thought. There is just one indigo manager’s unpublished diary that provides some information. Thomas Machell—who kept his detailed illustrated account for his father, a Yorkshire clergyman—showed a keen interest in the lives of the migrants from Chhota Nagpur who worked in his factories; he regretted not being able to communicate better with them. His description of the beginning of the manufacturing cycle is quite different from Reid’s: The night scene was truly beautiful, the dark foliage of the trees in the background, the dusky figures of the laborers and torch bearers, the group of women enveloped from head to foot in their white clothes looking like groups of spectres gliding amongst the trees whilst here and there dotted about the tall figures of the Burkandauses [armed guards] with their crimson Turbans and sashes, their swords under their arms and their shields on their backs, gave a military appearance to the scene whilst the carts and oxen with their vociferous drivers kept up a constant bustle. Scattered about in every direction were numerous little fires with dusky figures moving about them, preparing the evening meal and at the end of the vats with a large fan of palm leaves contemplating the busy scene with all the placid con- tentment of a Hindoo stood the colossal figure of old Mooty the Elephant,

49 “During the day, as the plant comes in, measure and stack it up in the shade that you may be able to load your steeper all at once in the afternoon, so as to have all done, in this way, before dark, taking care in filling the vats, that the water will run freely out of the steepers into the beaters when the pins are taken out, by forming a channel with the plant at the bottom next to the wall [. . .] I repeat, stow the vats so as to let the impregnated water run off freely, keep the plant horizontally square as you fill up”. Reid, Culture and Manufacture of Indigo, p. 66. For a similarly manager-centric account of indigo processing, see “Extract from a letter written in October 1814, from a gentleman in the superinten- dence of some extensive Indigo Factories, on the Ganges, near Boglipore, giving a concise account of the process of Manufacture, &c.,” Southern Agriculturist and Register of Rural Affairs, 3, 8 (August 1830). 68 willem van schendel

quietly fanning himself and looking as if he considered the whole scene as got up for his special amusement.50 He also gives us some insight into the sacred significance that laborers attached to factory work when he describes the beginning of the manufacturing cycle: Labourers (or Coolies as we call them) came trooping into the factory, carts and boats were being looked up in all quarters and properly put in order for the business of the coming month, the Vats were washed out, the Chinese pumps were erected, the pressing frames repaired and finally all the workmen having arrived a couple of goats were given to them with which they performed the ceremony of striking off their heads and sprinkling the blood on the Vats which were about to be opened. They then sat down and soon had the animals cut up, dressed and eaten.51 Machell also visited his laborers in their hamlets and witnessed their social life. He was especially fascinated by their dances and music (see Box “A Wedding Feast”).

A Wedding Feast

“Hearing a quiet drumming and fifing in the Boona parrah [living quarters of the labor migrants from Chhota Nagpur], I sallied out to see the Tumashah [entertainment] and found it was the celebration of a Boonoah’s wedding. You must know that the Boonoahs [. . .] work at anything, do anything, eat anything and in short are regular junglees [wild, uncultivated people], the merri- est, blackest and most hard working of all the inhabitants of this country. Give them their grog and a pig or a goat and they get out their rude instruments and they’ll drum and sing and dance and drink as long as they have a pice [cent] in their possession and then go to work again. We generally have a number of them in every factory and they erect most primitive little huts and

50 Machell, “Journals,” vol. 1, 145–6/27–8. 51 Machell, “Journals,” vol. 1, 144–5/26–7; cf. vol. 4, 203/111. Elsewhere he observed: “July 18th. On arriving at the factory [Rooderpore], I found they had just slain the goat in honour of Kalli and sprinkled its blood on the vats preparatory to opening them for the season. The Pundit still mouthed his mantras. The incense still smoked and victim’s limbs still quivered whilst the sacrificial sword was covered with blood. The pundit had done his office well: one blow had taken the head off and doubtless that knife would have taken off a man’s head just as expertly to the honour and glory of the horrid toothed goddess”. (Machell, “Journals,” vol. 4, 203/111). See also Pemberton on sacrificial offerings in Java sugar mills (John Pemberton, “The Ghost in the Machine,” in Rosalind C. Morris (ed.), Photographies East: The Camera and Its Histories in East and Southeast Asia (Durham & London, 2009), pp. 29–56. working for wages in colonial bengal’s indigo industry 69

form a separate village living entirely apart from all others and leading a life of alternate hardship and reckless gaiety, the happiest race of outcasts that ever existed [. . .]. I have [. . .] done my sketch of the wedding party but alas I can convey no idea of the comicality of the whole scene [. . .]. The long thin legged drummer alone was a character you might have watched for hours, his looks of extasy at the excellence of his own performance, his fantastic attitudes and the extreme affectation with which he pirouetted round and round and frisked about dis- playing a pair of legs no thicker than an ordinary drumstick and then to see him bend his head, shut his eyes and tap a difficult passage was excruciat- ing. The gentlemen with swords worked themselves [. . .] into a furious state of excitement performing pass seuls in which they slashed about in a most heroic manner [. . .]. Laloo my old sweeper, now Charley’s dog feeder, a half crazy fellow played the fool with a horse’s tail and a big blue stick and he assured me privately that none of them knew how to dance except himself. Certainly he had a very eccentric style of his own. There was a little by play going on amongst the youngsters that tickled me very much. The little rascals had also had their share of the shrob, which is Asiatic for grog, and imitated their seniors’ capers. But where is the bride and bridegroom? Look to the right and you will see a yellow cloth which conceals the blushes of the lovely angel of darkness— really however she was a very pretty working girl for on my enquiring after her an old lady led her out and after the Oriental fashion she bowed herself with her face to the earth before the Saheb. In this part of the world a woman does not bow or salaam like a man but kneeling down she touches the dust at your feet with her forehead. I must say it goes against the grain with me to see a pretty girl especially making so abject a prostration. If ever I have it in my power as a Zemindar in this country I will endeavour to raise the servile condition of the poorer class of females and by establishing a silk manufactory or employing them in picking cotton and having them taught to sow I think much might be done to better their condition. As it is they are regular hard worked slaves. The happy man you will see in the background on the left clothed also in yellow garments smelling of nine alehouses [is the groom] [. . .] but why do they wear yellow on these festive occasions? I must find that out. The natives themselves always reply to questions of this sort “dustoor assee” “it is our custom” and no further information can you get untill you search their musty black letter records or meet with some knowing fellow who if he does not actually know will loose no time in inventing an answer which he thinks will please or satisfy the Saheb.”

Apart from Machell’s depictions and some scattered references and drawings from other sources, the experience of being an indigo factory laborer in Bengal still remains largely beyond our grasp. But we have considerable evidence about who worked in the factories and under what conditions. 70 willem van schendel

The most striking characteristic of factory labor was that it was so highly differentiated. Linguistically, indigo laborers represented many different communities, among them Bengali, “Hindustani” (Bhojpuri, Hindi, Urdu), Kurukh (Oraon), Santali, Mundari and Bhumij. In the factory, the link languages were Hindustani, Bengali and English. Socially, factory labor was equally varied, representing a range of different caste groups, religions and educational achievements. Both women and men worked in the factories; in some factories women made up 30 percent of the labor force.52 Boys and adults worked side by side, although boys sometimes had specific tasks, such as cleaning vats or working in the house in which the indigo cakes were drying.53 In terms of skills, productive laborers were differentiated into those without specifically defined skills (“coolies”), those with specific indigo skills (e.g. colour master), those with caste-bound skills (e.g. sweepers) and craftsmen (e.g. carpenters). In addition, as we have seen, there were laborers with managerial, coercive and domestic skills. And finally there was the distinction between permanent and seasonal employment. In other words, the people who received a wage for turning indigo plants into indigo dye did not form a homogeneous category of “labor” at all, but rather an intricate amalgam of astonishingly different interests, capabilities and opportunities. What is most striking is the fact that they managed to cooperate despite their differences. In an industry riddled with confrontations and oppression, factory laborers appear to have been the least disruptive—a major reason why they are relatively invisible in the historical record. Indigo conflicts cen- tered on possession of the crop and involved indigo managers, zamindars and peasant producers. Once the crop reached the factory, it entered a much more controlled zone. There is very little evidence of labor unrest in indigo factories. This may be due to a number of factors. Indigo managers in Bengal rarely expressed concern about the supply of labor to their factories. Clearly laborers were attracted by the wages that the factories offered and therefore came of their own volition—often from hundreds of kilometres away and returning annually for the season. Within the factory, the spectacular fragmentation of the labor force—in terms of tasks, wages, contracts and social categories—militated against self-organization to demand better working conditions.

52 For example, see the figures for nine factories in Bhagalpur in 1827–28 in “Indigo Planters” ((1832), British Library), pp. 280–1, 286–7. 53 Phipps, Series of Treatises, p. 160. working for wages in colonial bengal’s indigo industry 71

But the main factor appears to have been a harsh disciplining of factory labor. The power of the managers was awesome. It was demonstrated largely outside the factories, in pitched battles with landlords and managers of neighboring factories and in oppression of cultivators who had no recourse to an independent judiciary. As one cultivator, who was forced to sow his fields with indigo, reported: “I had nobody to complain to. Mr Tripp [the indigo manager] is judge, magistrate, collector, and everything.”54 European indigo managers produced intense fear in Bengali villagers55— and they could turn equally violent inside the factory: As I dismounted at the factory, a man to whom I had given an awful licking yesterday came running up to take my horse. “Well, my man,” I said, “What made you say such foolish things as you did yesterday?” “Oh,” he replied, “I was mad; I was foolish.” “Well, so I think and faith you’ve got what’ll last you many a long day if I may guess by the marks on your hide.” The man did not seem at all put out about it. He seemed to be fully convinced that he had deserved it and did not appear to retain one spark of indignation against his flogger; and I can assure you a smart cowhiding on the bare back is not to be laugh’d at.56 Apparently, such power, combined with the presence of armed retainers in the factory grounds, was sufficient to sort out any labor unrest that might occur.

54 Evidence given by Sri Bhobonath Joardar (inhabitant of Tarai, Hardi Police Office, Nuddea District (now: Chuadanga District)); examined on oath. Report of the Indigo Com- mission, 1860, p. 178. For indigo entrepreneurs “oppressions” as represented to the British Parliament in 1831–2, see House of Commons, Reports from Committees, East India Com- pany’s Affairs, Report and General Appendix Vol. 1, Session 6 December 1831–16 August 1832 (London, 1831–2), vol. 8, Appendix V: Settlement of Europeans in India, pp. 253–380. 55 “Lost our way and rode into a village to ask some one to point out the road. The people all bolted, one or two men scrambled up trees and in spite of our solicitations would neither come down nor tell us which way to go, but looked at us like a set of skinned baboons. As they are my people, I dismounted determined to bring them to reason. I took up an axe and commenced work. “Show me the road,” said I. No answer. Smash went the axe and shivered a large earthen jar. “Now will you show me the road? “No answer. Smash went another jar. “Now, my lads, will you show us the road? “One of them pointed north, another pointed south. “Come down and show us the road.“ No answer. Smash went the axe again. Every blow of the axe was accompanied by a similar request but it was not until I had done a pretty good morning’s work that we could persuade one of these fools to come down and show us the road”. Machell, “Journals,” vol. 4, 228–9/124. 56 Machell, “Journals,” vol. 4, 230/125. 72 willem van schendel

Indigo labor in Bengal

In the Americas, indigo production had been sustained by slave labor. This was not possible in Bengal but it was not necessary either. Here indigo production was embedded in a rural society that had long known seasonal labor migration and in which colonial transformations created new streams of rural labor migrants willing to work for extremely low pay. In this system it was the cultivation of indigo, not its processing, that formed a major bottleneck. Indigo entrepreneurs had insufficient access to land to grow their own indigo and therefore had to persuade, entice, trick or coerce peasant producers into providing their factories with freshly har- vested indigo plants, even if peasants preferred to grow other crops. The intense frictions and violent confrontations that inevitably resulted from this were a world apart from labor relations inside the factories. Waged factory workers appear never to have made common cause with rebel- lious peasants. Ultimately, their interests were closer to those of the indigo entrepreneur than to those of the indigo cultivator. As industrial workers, their jobs and incomes depended on the timely arrival of raw material—fresh indigo leaves. Highly exploited workers in colonial enterprises producing commodities and profit for the metropolis, they never found champions for their cause. For over 150 years the literature on indigo has been remarkably vocal about the injustice meted out to cultivators and remarkably silent about the fate of indigo factory workers. It is instructive to compare this with another major agro-industry in colonial Bengal. Just as indigo was an indigenous natural dye that colonial enterprise developed very successfully for European markets, so jute was an indigenous natural fibre that gave rise to a booming industry. Jute was cultivated mostly in eastern Bengal. At first it was processed in factories in Scotland but from the 1850s jute factories thrived around Kolkata. Like indigo, jute was produced by peasant producers who were indebted to agents of the processing industry.57 Unlike indigo, however, jute cultivation never gave rise to large-scale peasant uprisings and never became a nationalist icon of colonial evil.58 Moreover, jute fibre is non-perishable

57 Sugata Bose, Agrarian Bengal: Economy, Social Structure and Politics, 1919–1947 (Cam- bridge, 1986). 58 In the case of indigo, this happened twice. First, peasant resistance in Bengal in the late 1850s fired the imagination of educated Bengalis, mainly through Dinabandhu Mitra’s play Neel Darpan, James Long’s treatise Strike But Hear! and Lal Behari Day’s novel Govinda Sámanta. Second, in the 1910s peasant resistance in North Bihar drew national working for wages in colonial bengal’s indigo industry 73 and jute factories did not need to be close to the fields; jute factories were located in urban surroundings. These differences led to distinct historiographical traditions. The indigo industry became the domain of nationalist and agrarian historians who focused entirely on the cultivation phase of the production process. By contrast, the jute industry became the domain of labor and urban historians who focused predominantly on the processing phase of the production process. Thus we have a very rich industrial labor history for jute but nothing comparable for indigo.59 This is a striking disparity in view of the indigo industry’s size, economic importance and longevity—and the hundreds of thousands of laborers who worked in it.

attention when Mahatma Gandhi organized a mass civil disobedience campaign known as the Champaran satyagraha. See Jacques Pouchepadass, Champaran and Gandhi: Planters, Peasants and Gandhian Politics (Delhi, 1999). 59 See Dipesh Chakrabarty, Rethinking Working Class History: Bengal 1890–1940 (Delhi, 1989); Arjan de Haan, Unsettled Settlers: Migrant Workers and Industrial Capitalism in Cal- cutta (Hilversum, 1994); Samita Sen, Women and Labour in Late Colonial India: The Case of the Jute Industry, 1890–1940 (Cambridge, 1999); Arjan de Haan and Samita Sen (eds.), A Case for Labour History: The Jute Industry in Eastern India (Calcutta, 1999); and Subho Basu, Does Class Matter? Colonial Capital and Workers’ Resistance in Bengal (1890–1937) (New Delhi, 2004).

Dak Roads, Dak Runners and the Re-Ordering of Communication Networks*

Chitra Joshi

The early history of postal communications in India is intimately connected with the figure of the runner, the courier who carried mail from place to place. The dauriya, or the runner, is in many ways a familiar figure in 19th-century travel writing and literature. Kipling’s lines about the imperial mail being carried dutifully through rain and flood resonate in many stories about runners in the 19th century.1 A wellknown story dramatically recounted in histories of the post office in colonial India, concerns a tragic encounter between a dak (mail) runner and a tiger. The runner, we are told, knew there was a man-eating tiger on the prowl, yet he braved his way and set out with his mailbag. He was killed just two miles away from his last halt. The story does not end at this point, however—it goes on to recount how the village chowkidar (watchman) was able to successfully retrieve the mailbag . . . and deliver it safely to its destination.2 What we see in stories like this, is the connected history of communications and labor. Cast in a heroic mode, Kipling’s poem celebrates the heroism and commitment of the runner, the problems and dangers he encountered in fulfilling his duties. The romanticization of the dauriya becomes at the same time an assertion of confidence in the “imperial” dak. There is a certainty that the mail will reach its destination . . . whatever the obstructions that may come its way.

* Earlier versions of this paper were presented at the International Conference of the Association of Indian Labour Historians, 18–20 March 2010 at the VV Giri National Labour Institute Noida and at the International Research Centre on Work and Human Lifecycle in Global History, Humboldt University, Berlin, 29 June 2010. 1 “Is the torrent in spate? He must ford it or swim. Has the rain wrecked the road? He must climb by the cliff. Does the tempest cry “Halt”? What are tempests to him? The Service admits not a “but” or and “if.” While the breath’s in his mouth, he must bear without fail, In the Name of the Empress, the Overland Mail’. Rudyard Kipling “The Overland Mail,” Kipling’s Poems: Plain Tales from the Hills and Others (1899; 2004), p. 120. 2 Geoffrey Clarke, The Post Office of India and Its Story (London, 1921), pp. 4–5. 76 chitra joshi

In what follows, I will examine other narratives behind these stories, to understand the system of communications that was so crucially dependent on road runners. In the early 19th century, when the East India Com- pany was expanding the frontiers of its territory, the widening of postal networks and speeding up of communications became important topics of concern. The mechanisms through which official and private agents tried to regulate the movement of runners provide important insights into the processes through which communication networks were established, and how these impinged on the lives of dauriyas.

Who were the dauriyas?

Runners were quite literally people “who ran for their living,” carrying public and private mail to distant places. Traditions of running as couriers existed among castes like the Kahars in North and East India, Pattamars in the South and Mahars in the West. Runners who were drawn from such groups had precise knowledge of routes, and an ability to negotiate difficult terrain. The Pattamars in South India were known for preserving their monopoly over routes through closed shop practices, denying interlopers access to the knowledge necessary to enter the field.3 Dauriyas from certain regions, like Oudh and Mewat, had a reputation for speed and endurance.4 The ability of dauriyas to run long distances acquired legendary status, even as officials charged them for neglect of duty, and attempts to shortchange the Postal Department. The body of the dauriya became an object of interest for medical researchers investigating questions of health and life. An important tract on Code of Health and Longev- ity by John Sinclair Bart, for example, carried the results of an enquiry conducted among “a class of men who perform long journeys on foot”.5 In popular magazines, the endurance and speed of the Indian runners and palanquin bearers was the subject of discussion. In one such essay,

3 On the skills and traditional networks of Pattamars in South India, see: Gagan D.S. Sood, “The Informational Fabric of Eighteenth Century India and the Middle East, Inter- mediaries and Postal Communication,” Modern Asian Studies, 43, 5 (2009), pp. 1099–100. 4 Taylor to Postmaster General, Home Public, A. 30 January 1839, no. 35. See also, C.A. Bayly, Empire and Information: Intelligence Gathering and Social Communication in India, 1780–1870 (Cambridge, 1996), pp. 63–4. 5 John Sinclair Bart, Code of Health and Longevity or Concise View of the Principles Cal- culated for the Preservation of Health and the Attainment of Long Life Being an Attempt to Prove the Practibility, of Condensing, within a Narrow Compass the Most Material Informa- tion Hitherto Accumulated Regarding the Benefits of Arts Sciences or any Particular Branch Thereof, volume II (Edinburgh, 1807), p. 107. dak roads, dak runners and the communication networks 77

Figure 1. The almost legendary status of runners as a people with remarkable endurance and speed is graphically reaffirmed in representations like this. The strong, muscular body of the runners is foregrounded, conveying both a sense of power and dedication to their mission. Source Ilustrated London News (February 20, 1858; pg. 196; Issue 904) their constitution was celebrated as infinitely superior to that of educated young men, who were unused to exercise and labor.6 Stereotypes drawn from ethnic categories, characterizing certain groups as “good” or “bad” runners were important to the self-representations of runners. Profes- sional runners themselves, it seems, acquired a pride in their ability, saw running as an art, a skill that was learnt over generations.7 The skill and training necessary for long-distance running was underlined to define a sense of self, and to negotiate possible competition when the market of the job expanded.

6 In a study published in 1910 in The Popular Science Monthly, the dauriya was an object of scientific investigation, focusing on the impact of long distance running on physical longevity and vitality. The study valorized the physical attributes of the kahar dauriya, con- trasting his physical stamina with that of young college students engaged in cross-country running. C.E. Hammett, “Middle and Distance Running,” The Popular Science Monthly, 27 (July–December 1910), p. 40. 7 Stories about them described how they were initiated into their profession by doing stationary running in shoes of lead (!). See Irfan Habib, “Postal Communications in Mughal India,” Proceedings of the Indian History Congress (Amritsar, 1985). 78 chitra joshi

In the early 19th century, the increasing demand for courier services threatened the monopoly of specialized dauriyas over routes and services. In the late-18th and early-19th centuries, reports of postmasters and magistrates “impressing” peasants to provide services as runners were common. They were usually mobilized from social groups at the bottom of the caste hierarchy in the village, those classified in official records as “village servants” and “menials.” Government officials advocating liberal ideas of rule in the 1820s and 1830s tried to distance themselves from such forms of coercion;8 informally however, forcible methods to mobilize runners and palanquin bearers for the expanding needs of the East India Company continued to be used.9 The emphasis in official writings in the 1840s was on the need for “able- bodied” runners: postal authorities were advised “to retain in the service only strong and active runners”.10 Plagued with competition from other daks, officials saw the efficiency of Awadh dak, in terms of its method of selecting runners. In Awadh, in contrast to Company territories, they pointed out: “Great care has been taken . . . to select proper runners. The situation is sought after and it is always given to the best runner, who has to run a long race, in good time, and to beat all his competitors”.11 A system which had its origins in selecting skilled runners for the King’s dak was seen, in this discourse, as characteristic of the region as a whole. What did “working for the company” mean for dauriyas? Carrying mail in the name of the company was important to the sense of identity of dauriyas. To be seen and recognized as a “company ka dauriya” was empowering, especially in regions outside company control. Dauriyas demanded various privileges: they expected food, fuel and other provisions gratis, in their position as runners for the company. In a petition addressed to the British Resident in Indore, Shahamut Ali, a village head-

8 “The Post Office is a profitable establishment and ought in my view of the subject pay fairly for any service required in aid of it. “W.L. Mcville, Magistrate to Bayley, Chief Secretary to Government, 18 July 1821, Home Public, 17 August 1821, no. 18. 9 On this, see also my paper on “Palanquin bearers, Travellers and the Rules of the Road,” presented at the International Workshop on “Space, Capital and Social History in South Asia,” Göttingen, 24–6 June, 2010. 10 Bushby, Secretary, Government of India to Oldfield, Officiating. Postmaster General, 23 Sept 1840, Home Public, A, 23 September 1840, no. 41. 11 Taylor, who was surveying dak routes in Awadh proposed longer stages for Awadh, than in British territories: “I do this, because it has been already proved that in that province the men can run and run well too, such long stages.” T.I. Taylor, Office of Agent for conducting Post Office Enquiries to Postmaster General, Home Public, A. 30 January 1839, no. 35. dak roads, dak runners and the communication networks 79 man pointed out: “It is a notorious fact that the hurkarra speaking generally do receive fuels etc. . . . without payment in almost every village within the limits of native states.”12 Using the name of the company was also a way of clearing the passage, and forcing compliance from groups hostile to the entry of dauriyas. In April 1832, dauriyas travelling through Sum- bulpur tried to ward off attacks by Kol tribal people, announcing their identity as “Company ka dauriya.”13 On important routes, like the Calcutta—Bombay route, especially in the forest tracks, it was difficult for new entrants to displace the virtual monopoly of established runners, even when they demanded almost double the prevailing rate of wages.14 The expanding network of “company dauriyas” was dependent on the services of runners who were familiar with routes that ran through the dense forest tracts bordering Bengal and the Central Provinces region. Strikes by “jungle runners” paralyzed the movement of mail on the route.15 Runners played on the fear of the jungle, terrorizing new dauriyas with stories of tigers and other predatory beasts. In jungle tracts in particular, the ability to negotiate dangers was important to the self-representation of dauriyas—a representation that became part of folklore, and proved useful in the market for jobs.

Speed, time, and the mail contract

The consolidation of East India Company power in the early 19th century was marked by an assertion of control over networks of communication. This also involved attempts to subordinate local posts to its authority. These efforts were continuously challenged by rival agencies carrying dak for rulers in neighboring territories. Merchants and bankers argued that mail delivered by private establishments could ensure greater speed and safety than services provided by the company.16 Seasoned runners were

12 A letter signed by a local patel (headman) Shahamut Ali to C. Hamilton, Resident, 9 June 1853, Home Public, 22 July, 1853, no. 68. Hurkarra=courier. 13 Copy of Petition from: Bishonauth writer of Chokey Seersah, 10 April, 1832, to C.L. Babington, Deputy Postmaster, Sambhulpoor District, Home Public, 24 April, 1832, no. 41. 14 Babington, Postmaster, Sambulpore, to Oldfield, Postmaster General, 19 Aug 1840, Home Public, B., 23 September 1840, no. 31. 15 From: Babington, Postmaster, Sambulpore to Oldfield, Postmaster General, 19 August 1840, Home Public, B., 23 September 1840, no. 31. 16 See, for instance, Cox to Minto, Governor General, 7 November 1807, Home Public, 1 January 1808, no. 30; Postmaster General to Thomas Brown, Chief Secretary, 17 June 1807, Home Public, 2 July 1807, no. 34. 80 chitra joshi valued for their speed and knowledge of routes. Often, private agencies poached on runners working for the company to conduct their own business.17 These transactions were done through the agency of dak moon- shies (clerks) employed by the company, who accumulated private profits through such deals. Efforts to control and regulate the speed of runners have to be understood against this background. Within the new regime of governance, efficiency was seen as intimately linked to speed. Commercial information, revenue and police dispatches, all required speedy services. In the 1830s, demands for a speedy post were also related to the new overland route opened between India and Europe via Egypt.18 Mail leaving by the “overland” route had to be on time for the steamers departing for the Suez from Bombay.19 Company officials were anxious that there should be no delays in transporting the “overland” mail. Kittoe, superintendent of the Raipur post road, suggested that contractors should be warned to keep runners along the route ready for the overland mail: as people could not be expected to read or write, they might be made to understand when the mails were expected, by a small piece of red cloth fastened to a thin stick about a foot long, to be passed from stage to stage by the express runners.20

The claims to superior efficiency of company post were asserted by incorporating many signs of modernity. Regulations were written up and elaborated. Terms under which contractors conveying mail were to function were specified and fixed.21 A format was devised for formal contracts. Con- tracts for the East India Company mail were held by postmasters, doctors,

17 See, for instance, correspondence around “Bustee Ram, an overseer of the Govt runners on the Kurnaul road, having been accused of being the principal partner in a firm of native dak masters carrying on business between Delhi and Jeypore’. James Ranken, Agent Post Office Enquiries, to Secretary to Governor General North Western Provinces, 16 March 1839, Home Public, C., 24 April 1839, no. 41. 18 On the overland route see: Daniel Thorner, Investment in Empire: British Railway and Steam Shipping Enterprises in India 1825–1849 (Philadelphia, 1950), pp. 22–43; see also: John K. Sidebottom, The Overland Mail: A Postal Historical Study of the Mail Route to India (London, 1948), pp. 52–72. 19 The date for the steamer’s departure was advertised in advance and mail had to reach Bombay in time for that. G. Alexander, Postmaster General to H.T. Prinsep, Secretary Government of India, 5 September 1839, Home Public B., 18 September, 1839, no. 6. 20 Kittoe, June 1840, Home Public, 6 January 1841, no. 26. 21 A notice circulated in 1810 inviting contracts for mail between Bombay and Monghyr, for instance, specified the number of chowkies (posts), distances and terms for contractors. Postmaster General, 29 January, 1810, Home Public, 9 February 1810, no. 31. dak roads, dak runners and the communication networks 81 army officers, private merchants and others.22 The period for which dak contracts were held could vary from twenty months to three years.23 One crucial term of the contract was the speed of mail: contractors had to guarantee that runners employed by them maintained the required speed through the period of the contract. Non-compliance with the terms of agreement, by the contractors meant the payment of fines and ultimately a cancellation24 or non-renewal of the contract. How was speed calculated? Speed was initially a nebulous concept. Officials, contractors and dauriyas had different notions of what could be legitimately defined as the expected speed of postal delivery. A variety of opinions were expressed, and conflicting views persisted even when certain terms became part of the official axiom, codified and specified through authorized regulations and instructions. The renewal of each contract involved negotiation over the basis on which calculations were made. The continuous discussions over the terms of contract reflect official efforts to evolve standardized rules according to the new demands for precision and uniformity.25 Through the debates and conflicts that ensued, we see the notion of an “average speed” being gradually defined. Attempts to introduce new standards of measurement in the late 1830s were resented by contractors. Earlier, the custom in most regions was to calculate the speed of mail on the basis of a monthly average rate. Contrac- tors complained against the imposition of new rules by which they were accountable for the speed of each individual mail transaction.26 There was

22 Sometimes postmasters who started with dak contracts for the company, left company service and set up their own independent courier services. For contractors, working for the Company meant many restrictions; yet carrying official dak also gave them access to routes and special privileges. On this see: Michael H. Fisher, “The East India Com- pany’s Suppression of Native Dak,” Indian Economic and Social History Review, 31 (1994), pp. 311–48. 23 Contractors were usually given a fixed monthly remuneration calculated according to the distance they covered. For example, Luckeenarain (sic) Chatterjee was given a fixed monthly sum of Rs 280 for carrying mail between Calcutta and Jessore in both directions. 24 To cancel a contract before the stipulated term, the government was required to give a notice and pay an agreed amount as compensation. See for instance contract with Moorlee Dhur for the Benares–Delhi road, Home Public, 11 August 1841, no. 3. 25 For a useful discussion on quantification and questions of governmentality see: Ian Hacking, “How We Should Do a History of Statistics,” in Graham Burchell, Colin Gordon and Peter Miller (eds), The Foucault Effect: Studies in Governmentality (Chicago, 1991), pp. 181–96; see also: Theodore M. Porter, Trust in Numbers: the Pursuit of Objectivity in Science and Public Life (Princeton, 1996), pp. 17–45. 26 Luckeenarain Chatterjee who signed a contract for mail between Calcutta and Jes- sore promised to the following terms: “I agree to convey the mails not exceeding in weight 82 chitra joshi also an effort to remove factors that could introduce uncertainty in the calculation. Earlier, contractors were allowed remissions for delays in mail for rain, flood, and poor roads on the production of certificates from village headmen establishing the reasons for detention.27 Standardized rules of contract enforced in the 1830s gave no remission for rain or flood, or any other obstacle. Contractors pleaded for allowances to be made for crossing flooded rivers, and for poor roads. In many regions dak contractors resented the attempt to impose a common standard of a daily average rate of traveling. Contractors in Central India for instance, argued for a continuation of the monthly system, which was the “usual custom” in the region.28 On the Calcutta-Rungpore road, an agreement for horse dak with Bashford, a mail contractor, was cancelled . . . because he did not provide a daily average speed of 5–6 miles per hour. Bashford argued for a lenient view. The Hooghly-Rungpore stretch, he explained, was intersected by rivers at several points. It was impossible, he stated, for horses or even runners to ford the rivers on this route. The mail had to be ferried by boat. The terms of his contract however, did not take into account the time taken in traversing such difficult routes. Bashford pressed for a remission in fines, and more flexible rules that calculated the speed of dak on the basis of a monthly average rate of traveling and not on an average daily rate. Postal authorities considered such demands not only unreasonable but “an abrogation of one of the fundamental rules under which contracts are now allowed to be accepted.”29

1 maund and 10 seers by means of foot runners at the rate of 5 mile per hour, during 8 months in the year, i.e. from October to June, and at the rate of 4 miles per hours, during the remaining 4 months”—document signed by Luckeenarain Chatterjee, 31 July 1842, Home Public, 7 September 1842, no. 12. 27 Contractors had to give documentary proof from heads of villages if mail was delayed, because of bad weather or bad roads, Rungaswamy Moodaliar, late mail contractor, Home Public, B., 3 November 1841, no. 30, from: Rungaswamy Mudaliar, Late mail contractor, Nagpore, 11 August 1841, to Secretary, Government of India, General Department. 28 R. Moodaliar, contractor for the Nagpore to Ryepore road, drew attention to the fact that” [. . .] in all Provinces the rates of traveling are allowed to be calculated monthly [. . .].” A monthly rate, Moodaliar pointed out, was also the practice, till recently, in the Nagpur post office. From: R. Moodalliar, to Secretary, Government of India, 2 June, 1841, Home Public, B., 3 November, 1841, no. 26. 29 G. Alexander, Postmaster General to Prinsep, Secretary Government of India, Home Public, 23 October 1839, no. 31. In other regions, the system of calculating speed on the basis of monthly average was more common. See, for instance, correspondence on Rungas- wamy Mudaliar, dak contractor on the Nagpore Raepore road, Home Public, 3 November, no. 26. dak roads, dak runners and the communication networks 83

Conflicts over determining the average rate reveal some of the problems of fixing universal standards of speed at which mail runners had to travel. The search for averages was a perpetual problem: the assumption of average conditions was difficult when runners had to traverse varied landscapes, and continuously faced unanticipated hurdles. Quite simply, conditions often did not exist for the anticipation of “average” conditions. It was commonly acknowledged that an average daily rate of five miles an hour over a long stretch was difficult for a runner to maintain. Postal officials themselves recognized this problem. Babington, a post master, who also held contracts for mail in the Midnapur and Sambulpur region, pointed out that an average rate of five miles an hour over the whole distance between the two posts was a “fair” rate, when the weather was fine. But he feared even this speed was not possible to maintain with heavy loads over a continuous period.30 The speed of the mail was intimately connected to the volume of mail the runners had to carry. To maintain speed, contractors had to ensure that the runners were not weighed down, carrying too heavy a load. This was difficult to accomplish when there was a sudden increase in the volume of mail, or a squeeze on the contractors’ income. To maintain the margin of profit, contractors sought to cut costs, employing fewer runners, forcing them to carry an increasing volume of mail. Reports noted how the weight of the dispatches made it difficult for runners to maintain a good speed.31

Negotiating the contract

What did the new terms of contract and regulation mean for the dauriyas? How did the demands for increased speed and the expansion of mail networks impact on them? From the late 1790s onwards, there were frequent instances of protests by dauriyas who demanded higher wages and timely payment. There were apprehensions that desertion by runners would lead “to the entire stoppage of the Dak.”32 In districts where

30 From: Babington, Postmaster, Sambulpore to Oldfield, Postmaster General, dated 19 August 1840, Home Public, B., 23 September 1840, no. 31. 31 With the opening of the “overland” route to Europe, the volume of mail on the Calcutta—Bombay route saw a phenomenal increase, from 450 packets to 5260 by 1839. Home Public, 18 September 1839, no. 6. 32 From: Vanderheyden, Postmaster, Benaras, to Lawrance, late acting Postmaster, Benares, 4 June 1799, Home Public, 28 June 1799, no. 22. 84 chitra joshi the rates of pay were lower than other neighboring territories, runners made representations demanding similar terms. In the Chupra district of Bihar, for instance, they submitted a petition that their wages of Rs 2/8 per month were much lower than in the neighboring districts of Arrah, Patna and Tirhoot, where they were around Rs 4 per month.33 Scared of runners stopping work, many officials felt that district collectors and magistrates should be given greater control over the movement of post.34 In contrast to postmasters who lacked effective control over runners, magistrates, it was argued, could “inspire dread”: they could, with the help of the police, seize, dismiss and replace recalcitrant runners.35 Magistrates were seen as officials who could, in fact, help to ”materially reduce the time [. . .] taken to convey the mails from station to station”.36 Police officials demanded new rules giving additional penal powers to magistrates, sanctioning corporal punishment to runners who neglected their duties.37 These strategies of control, however, did not become the general practice. One petition, submitted by mail runners against Madobchunder Sir- car and Russiklal Bose in July 1838, provides important insights into the ways in which relationships between dak contractors and their employees worked. Mail runners employed by Madobechunder Sircar and his partner Russiklal Bose complained to the Postmaster of Rungpore that they had not been paid for their services. Around the same time, a similar complaint was filed against Goopeemohun Burnall to the Postmaster of Bhagulpore.38 Sircar and Burnall denied these allegations. The runners, they claimed, had been instigated to make false complaints. Justifying their claims,

33 Petition submitted to W.H. Traut, Acting Postmaster of Sarun, 22 April 1814, Home Public, 27 May 1814, no. 51. 34 Bushby, Secretary Board of Revenue was categorical that “the influence of magiste- rial authority will doubtless have a wholesome effect upon the behaviour of the servants and runners belonging to the different establishments so transferred [. . .].” Bushby to Lushington, dated 10 July 1827, Home Public, 1 May 1828, no. 19. 35 From: Ewer, Superintendent daks to Shakespear, Secretary Judiciary Department, Home Public, 1 May, no. 18; see also: Home Public, 22 September 1829, no. 12. 36 Extract from: letter signed Halhed, Magistrate, Moradabad, Home Public, 1 May 1828, no. 19. 37 Barwell, Police Superintendent, argued that Magistrates should have the sole power to nominate, runners and peons, “without reference to any Court or Board except in cases of dismissal [. . .].” Home Public, 1 May 1828, no. 21. Extract from: Proceedings of Governor General in Council, Judicial Department 24 April 1828. 38 From: Alexander, Officiating Postmaster General to H.I. Prinsep, Secretary, General Department, 2, September 1838, Home Public, 12 September 1838, no. 27. dak roads, dak runners and the communication networks 85

Sircar and Burnall stated that they had receipts from surbardars, showing that payments had been made. Senior postal officials admitted their helplessness, arguing that it was not possible for the government to intervene in such situations. One of them declared: “as the servants of contractors are not the servants of government it is totally out of my power to assist them . . .”39 (emphasis added). Others asserted that the contractors had claimed that deductions were made only when runners failed to perform their duty. Having denied the government’s right to intervene in the operation of a private contract, postal officials reaffirmed the rights of the contractor and the obligations of the dauriya. A contractor engages his runners upon the same principle as he undertakes his contract. He fixes a rate of pay, to which they agree, for a certain rate of traveling with the stipulation that they are to be rewarded in proportion to the rate of celerity acquired above, and to be amerced rateably, for a falling short of the fixed rate.40 A familiar method deployed by contractors to exercise control over runners was to withhold wages and make deductions from them. The monthly earnings of runners in the 1820s and 1830s remained around Rs. 2.8, rising to Rs.4.00 in some regions.41 But this was not always what runners actually received. The amount paid to them was linked to the question of speed. Charges of delay—whether they had any basis or not—became an argument for deduction, and a basis of control. Fines, in fact, became a common strategy of controlling the runners’ speed. Underlining the discussion around the petitions against Sircar and Bur- nall is also the importance of the written “receipt” in the system of dak contracts. The negotiations show how the language of law was acquiring

39 From: G. Alexander, Officiating Postmaster General to Thomas Leckie, Postmaster Bhaugulpore, 28 July 1838, Home Public, 12 September 1838, no. 27. He pointed to instances when government employees holding dak contracts were dismissed. Bose the contractor for the Jessore road for instance was deprived of his contract before his term. 40 Alexander, Officiating Postmaster General to H.I. Prinsep, Secretary, General Depart- ment, 2, September 1838, Home Public, 12 September 1838, no. 27 (emphasis added). 41 In the Allygurh (sic) division for instance runners received Rs 4.00 a month in contrast to Mynepoorie where they received Rs. 3.8 a month. From: Shakespear, Postmaster General to E. Malony Acting Secretary Board of Revenue, dated 2 Oct. 1823, Home Public, 16 October 1823, no. 27. In 1838, Alexander, the Postmaster General reports: “Runners pay in the Mirzapore division to be increased from Rs 3.8 to Rs 4 to bring them on par with runners elsewhere”. From: G. Alexander to H.T. Prinsep, 16 November 1838, Home Public, 28 November 1838, no. 31. 86 chitra joshi significance, and writing was becoming important in defining relations not formally regulated by a contract. Receipts, I will suggest, became a site on which the unequal negotiation between the contractor and the dauriya was played out. Senior postal officials expressed their helplessness when contractors produced receipts as proof of payments disbursed. Who signed the receipts? The record of correspondence about cases like this points quite clearly to the role of subordinates employed by contractors/postmasters. Postmasters like Sircar, Bose and Burnall did not personally superintend their contracts. The task of overseeing the runners was sub-contracted to subordinates, at very low rates of remuneration. As the levels of sub-contracting became increasingly complex, and hierarchies of controllers emerged, the politics of payment became trickier. Both Sircar and Rasiklal Bose argued that the “full amount (due to the runners) to the last pice” had been handed over to surbardars (sic) who were in charge of disbursing payments. They claimed that they had receipts for the payment to the runners from the surbardars.42 Quite clearly the receipts were not documents through which amounts disbursed to individual runners could be verified. These were records of the total amount distributed by post masters to contractors, and by contractors to sub-contractors. So, the petition of the runners about non-payment of dues could be as true as the claims of Burnall and Sircar that they had in fact made the contracted payment to the last piece. New forms of control required written evidence, receipts and proof. At every level, written evidence of payment and proof of work had to be collected. Detailed schedules charting the speed of mail on different routes also detailed the performance of runners. Mootsuddies (accountants) were given instructions to inscribe the time at which runners arrived on their backs.43 In Bengal Presidency, the Postmaster General, Elliot proposed the introduction of watches to keep time:

42 Note for instance, a statement of Alexander, Postmaster General “[. . .] they have uniformly declared that the full amount to the last pice was remitted by them to their surberadars, who have furnished them with receipts stating to be for the due payments of their full wages to the runners”—Alexander, Officiating Postmaster General to H.I. Prin- sep, Secretary, General Department, dated 2 September 1838, Home Public, 12 September 1838, no. 27. 43 “Abstract Schedule of Routes and Rate of Travelling in Nov 1825,” Home Public, 2 March 1826, no. 13. dak roads, dak runners and the communication networks 87

It is my intention in the course of a few days to dispatch a watch with the mail from hence towards the westward as an experiment in order to see whether some more correct check cannot be obtained over the traveling of the mails than now exists.44 Measures like these served as ways to control and penalize “defaulting” runners, standardizing time, marking pace, and producing an idea of the “average”. Using watches, recording time with a new precision, also had a symbolic significance. It gave the imperial dak a stamp of modernity, dif- ferentiating it from other regional and zamindari daks.

Dangers on the road

The description of “rain and torrent,” of “ravines and tigers” in the poem by Kipling, cited at the beginning of this essay, are part of a story of the dangers and obstacles that had to be confronted to make modern communications possible. It is also about the runner’s commitment to public service. Kipling’s “Overland Mail” is about the heroic runner who is integrated into the imperial civilizational project. However, we need to go beyond an understanding of this familiar imperial trope, to see what such representations tell us about the runner’s journey, their negotiations with dangers on the roads. In official accounts, forest roads appear as spaces of danger. Forests had to be cut to make roads safe for runners.45 Thick forests were seen as “interrupting” and “blocking” the field of vision from the roads; they were “infested” with wild beasts.46 But clearing the jungles was almost an impossible task in many areas. Officials repeatedly complained how it was difficult to persuade locals inhabiting the region to cut the jungles. In the Nagpur Raepore region, there were reports that “the dislike of the country

44 Elliot, Postmaster General to Deputy Postmaster Bancoorah, Sherghatee, Benares Allahabad, Futtehpore, Cawnpore, Home Public, 12 April 1831, no. 76. Steps had to be taken to train supervisors to read the time. Elliot circulated a lithographic impression of a clock to teach supervisors to use the clock. 45 Bushby to Prinsep, dated 15 April 1828, Home Public, 17 April 1828, no. 27. Similar statements about the need to cut forests to make safe tracks for runners are made in a later period. Officiating Postmaster General, to H.T. Prinsep, Secretary Government of India, 9 November 1839, Home Public, B., 13 November 1839, no. 19. 46 In the Mirzapore Saugor route, from: Smith, Agent Governor General, Saugor Nerbudda Territory, to G.A. Bushby, Officiating Secretary General Department, dated 25 March 1833, Home Public, 19 April 1833, no. 27. 88 chitra joshi people to engage in the work” made the task of clearing particularly difficult.47 Besides, in the forested regions of Central India, the job of clearing was never over: new growth came up rapidly, and covered the tracks that had been cleared. Thick forests and treacherous terrain created spaces that were difficult for the state to access. In many such regions, dauriya tracks provided the links, creating territories of control and empire, penetrating spaces that were seemingly beyond the reach of the state. But there were real dangers runners had to face. Reports came in from many areas, in the 1830s and 1840s, of runners being attacked by tigers. From the Bagh Nudee region (a spot that became famous for such incidents), eyewitness accounts recorded: From about 12 o’clock at night of the 29th of Oct. 1831 the Calcutta mail arrived. I accompanied by Mohun Singh Rajpoot aged 45 years together with two sepoys was carrying the dak from Baugh Nuddee to Moondipore distance 5 cos about one coss from Baugh Nuddee there is [. . .] a very thick jungle. From this jungle a tiger came about midnight and carried away Mohun Singh.48 On the 26th September, as one of the runners was conveying Calcutta dak from the Bauge Nuddee towards the Moondepore Tuppa, accompanied by a Burkundauze, a tiger sprung on the Burkundaz and carried him off; the Hurkurrah shouted, and ran after the Tiger a few yards, but he disappeared in the long grass.49 From the Kunjipur and Haldia route, there were reports that runners had threatened to stop traveling at night, because of a fear of tigers.50 In many of the border districts of Orissa and Bengal—Sumbulpur, Lorapalli (in Ganjam), Haldia—runners refused to carry mail. In the Haldia region, runners insisted that they would not travel at night unless shikaries

47 G.I. Fraser, Nagpore Residency to Major Wilkinson, British Resident, Nagpore, 31 December 1839, Home Public, 22 January 1840, no. 12. Ten months later, Fraser writes again: “The clearing of the jungle to the distance of 400 yds on each side of the road an operation which has been twice performed, is in course of performance for the third time and which must be repeated annually”—20 October 1840, Home Public, 4 November 1840, no. 22B. 48 Deposition of Turwar Sah Goud, aged 40 years, stationed at Chokee of Bhaug Nuddee Home Public, 17 January 1832, no. 38. Bagh=tiger; coss= a measure of distance, approximately two to three kilometers. 49 Translation of the duffadar’s report, Gordon Postmaster Nagpor Residency, to Post- master General Calcutta dated 2 October 1831, Home Public, 17 January 1832, no. 38. 50 From: Alexander to Prinsep, Secretary, Government of India, Secretary Government of India, Home Public, 28 August 1839, no. 43. dak roads, dak runners and the communication networks 89 were sent out to protect them.51 In Lorapalli, it was reported: “seven out of twelve [. . .] deserted immediately after receiving their pay which has caused a delay in the receipt of the Bombay mail [. . .] of 24 hours”.52 Postal officials responded to such reports with a sense of alarm and dis- belief. In official discourse, the dauriya’s refusal to run because of a fear of tigers appeared almost incredulous. In these official writings, dauriyas are seen as people who inhabit the jungles: their bodies are inscribed with the marks of the wilderness they traverse. A history of the Post Office (1921) by Geoffrey Clarke, a civil servant, described thus: Postal runners are largely drawn from the less civilized races of India, many of whom are animists by religion. They will face wild beasts and wandering criminals, but will go miles to avoid an evil spirit in a tree.53 Post officials attached little credence to stories of encounters between runners and tigers. When a runner named Mohan Singh was killed by a tiger on the Bagh Nuddee, the Postmaster insisted that the Resident at Nagpur establish clearly that he was taken away by tigers: it is unlikely that a tiger which had possession of a body for so long a time as is described should have only taken a little of the breast and have carried off the head of his unfortunate victim, that being the last part of a human body I should suppose a hungry animal would carry away.54 On the Mohan Singh case, for instance, officials attributed no credibility to the statements of witnesses. They insisted that there was no verifiable proof that Mohan Singh had been killed by a tiger. In an official discourse that placed increasing emphasis on the need for “proofs” and “evidence,” even fatal encounters with tigers had to be proven. And, as mentioned, the idea that dauriyas should stop running for fear of predatory beasts seemed incredulous to postal officials. To develop modern communication, all hurdles had to be confronted and overcome. Mail networks could not be disrupted by the forces of nature. Officials typically suggested that reports on tigers and other dangers were attempts to extort compensation or ways of avoiding running at night. Officials

51 L.L. Rousseau, Postmaster Kedgeree (Khejuri) to W. Moore, Deputy Postmaster, dated 25 Aug, 1839, Home Public, 28 August 1839, no. 43. 52 Babington called for urgent measures: “I fear there will be no possibility of replac- ing the men who have deserted especially as I have no power either to punish them or compel others to serve”—Babington to Moore, 6 October 1842, Home Public, 12 October 1842, no. 15. 53 Clarke, Post Office of India, p. 5. 54 From: Elliot Postmaster General to G.A. Bushby, dated 7 January 1832. 90 chitra joshi mistrusted the claims of runners, demanded “proof” of tiger attacks, and feared that the dauriyas were using tiger stories to bargain for better wages, and negotiate the pace of running. Highlighting the dangers of the road, it appeared, was a strategy to negotiate the terms of labor. Forest areas like Midnapur and Sambulpur, to give some specific examples, were areas with a history of resistance to outside intrusion and to the opening of a dak route through the region. Officials in charge of postal communications in this stretch were convinced that the dangers of tigers were greatly exaggerated. Stories about predatory animals, officials believed, were circulated because of local opposition to the opening up of dak routes.55 In other areas, the Singhbhum region, for instance, the jungles were seen as hideouts for dacoits. Reports about tigers, officials warned were used by locals as a ploy and given the “specious appearance of being forwarded in truth.”56 Yet delays in mail and the refusal of dauriyas to run created anxieties in official circles. Thus, the Postmaster at Sambulpur wrote: “I have no power to punish the men and if I dismiss them it will be utterly impossible to replace them; of the latter circumstance they are well aware.”57 On the Bombay and Calcutta route, where the runners had to be in time for the arrival and departure of ships carrying the overland mail, reports of attacks by tigers caused serious concern. In March 1828, when the Calcutta mail was over 15 hours late, Babington, the official in charge, sent out a party of armed peons, offering “a reward of twenty five rupees, on event of their bringing the head of the tiger”.58 It is clear that postal officials both suspected and reaffirmed stories of runners and tigers on the road. They could not get away from the evidence of death in the jungles. Yet the networks of Imperial mail could not be sustained through a politics of suspicion. Officials had to ensure that runners did not desert. Running had to continue, and the fears of runners had to be allayed. A sense of security had to be created, and compensation and rewards had to be granted in case of death and injury. In agree- ing to give a lump sum to the families of runners who were killed, the state informally validated the principle of giving compensation to workers

55 On local opposition to the intrusion of dak runners, see also: Michael Fisher, “East India Company’s Suppression of Native Dak,” p. 325. 56 Major N. Jackson, to Prinsep, Home Public, 13 April 1830, no. 56. 57 Postmaster Sumbulpore to W. Moore Deputy Postmaster, 3 October 1842, Home Pub- lic, 12 October 1842, no. 15. 58 Babington to G. Stockwell, Postmaster General, dated 26 March 1828, Home Public, 17 April 1828, no. 27. dak roads, dak runners and the communication networks 91 killed during work.59 Stockwell recommends “a boon” of rupees Rs100 for the widow of the deceased who fell a victim to the tiger while on duty as a “public servant.”60 The fear of tigers played a part in the calculations of contractors mobilizing runners. They demanded concessions from the state: extra allowances for employing burkandauzes (guards), for the supply of pistols and firearms, and for carrying mussals (torches) to accompany runners. Postal officials conceded many of these demands, arguing that this would allow runners to travel fearlessly at night.61 The state thus had to make concessions, especially if the terrain was particularly inaccessible, and if dauriyas and their contractors could bargain from a position of strength. But were dangers of the jungles only a bargaining ploy? Runners also had real fears of predatory beasts which official representations tended to misrecognize. They feared running alone, especially at night. Lieutenant Kittoe, in charge of constructing a dak road, described: The runners generally travel in pairs both night and day and at night, sometimes three together, one carrying the wallet, and the other two, hoolahs and fire brands made of split saul poles 10 to 12 feet long, tied with grass which at once give a vivid light and frightens wild animals particularly bears. Single men sometimes venture in the daytime where the country is comparatively plain.62 Traveling through regions like Nagpur- Seonie, Sambulpur, Ramgurh or the Bagh Nuddee area was always precarious. Dauriyas, like others who inhabited the region, had practices and rituals to propitiate deities who, it was believed, could protect them from wild beasts. In the Sunderbans area, there were many stories of Dakhin Rai, the vengeful tiger god and the nurturing figure of Bonbibi who protected village folk and others against attacks by predatory beasts.63 Those entering jungle areas sought the help of fakirs and others known to possess special charms which could keep away predators, and protect travellers in the forests. Humans themselves, it seemed, could not acquire the power to confront the dangers;

59 From: Elliot, Postmaster General to Saunders, Officiating Secretary, General Department. 60 From: Stockwell to Bushby, dated 23 January 1828, Home Public, 7 February 1828, no. 22. 61 From: Captain Charles Johnson, Postmaster Khandesh, to Elliot, Postmaster General, Bombay, 17 April 1839, Home Public, 17 July 1839, no. 17. 62 Home Public, 6 January 1841, no. 25. 63 On this see, Annu Jalais, Forest of Tigers: People, Politics and Environment in the Sun- darbans (London, 2010), pp. 70–4, 157–64. 92 chitra joshi divine mediation was necessary. With regard to central India, Sleeman recounted fabulous stories about tigers with magical properties or “men turned into tigers” who targeted human prey with a vengeance—stories that seem to reaffirm the idea of human vulnerability and helplessness.64 Dauriya demands for an allowance to make offerings to “an old woman” (believed to have special powers against wild jungle beasts) should be understood against this background. Widows, as we know, occupy a limi- nal space, in between the normative and the non-normative, the human and the extra-human. Undisturbed by the mythic world that such stories open up, postal officials sanctioned the sum demanded, but recorded it as an amount given to an “old woman.” The Postmaster regarded this practice as an important concession to runners: “operating as such belief does, as advantageously in preventing delays in the transit of the mails, by inducing the runners to travel in confidence at times when they might otherwise be afraid to proceed [. . .].”65

What happened to horse daks?

A question often raised is why horse daks were not used in the first half of the 19th century. In principle, contractors were free to use any mode of transport, provided they delivered mail at the required speed.66 There were sporadic attempts to introduce horse daks in the 1830s and 1840s. Yet these experiments remained localized, and never completely displaced the existing system of runners. What happened to the horse daks?67 In official discussions, there was a constant attempt to rationalize the continued use of foot runners, as against horses drawn vehicles for carrying mail. In negotiations between mail contractors and postal authorities, the relative speed of dak and the costs of a horse establishment were

64 Sleeman, for instance, refers to tigers with magical properties in the Jabalpur- Mirzapur region: W.H. Sleeman, Rambles and Recollections of an Indian Official (London, 1844), pp. 161–7. 65 Officiating Postmaster Gen Nagpur Div to G.A. Bushby, Secretary Government of India, 23 June 1842, Home Public, 6 July 1842, no. 5. 66 Advertisements inviting tenders for mail contracts did not specify the mode of transport to be used. Contractors were free to use any mode, provided they delivered mail at the required speed. See, for instance, the copy of the contract signed with Moorlee Dhur: I Rankin, Post Master General, North West Provinces to I Thomason, Secretary to Govern- ment, North West Provinces, 12 July, 1841, Home Public, 11 August, 1841, no 3. 67 See also Bayly, Empire and Information, p. 65. dak roads, dak runners and the communication networks 93 crucial. In taking tenders for horse daks, contractors had to guarantee a minimum average speed of 5–6 miles per hour—i.e. a speed higher than the average of 4–5 miles per hour for runners—and they had to agree to pay fines for any delays.68 In official calculations, however, this increase in speed had to be weighed against the additional costs of a horse establishment. Did the gain in average speed justify the additional costs involved? This question was repeatedly raised. Moorleedhur, a jeweler, was given a contract for horse dak on the Benares—Delhi route.69 On the Cawnpore—Allahabad stretch of the route (a distance of 129 miles) the average speed of his horse dak was between 8–9 miles per hour; giving a total gain in speed of 10 hours over the 129 miles stretch.70 The gain in speed, in official calculations, was undermined by the increased cost of the establishment. The total expense on horse daks over the 477 miles between Benares and Delhi, at Rs 12000, was estimated to be three times more than the expense on the existing establishment of runners. The Governor General’s office was unwilling to sanction the increased expense. The postal receipts on the route, they argued, did not compensate for the expense.71 Contracts for horse dak were often discontinued, because of a failure to maintain the terms of speed specified in their contract. Why did horse daks fail to keep up speed? To maintain good speed, especially at night, horses, required tracks that were clear of trees, stumps, and other obstacles. The post horse system in Tokugawa Japan, or in Europe in the 17th and 18th century, functioned efficiently in regions where an elaborate network of roads and bridges had been built.72 In India, by contrast, dauriyas were preferred in most regions, because they could move fast through forests and difficult terrain even in the absence of proper tracks. Horses,

68 The fines were on a graduated scale according to the time lost. See for instance contract with Bishumbar Dey for horse dak between Calcutta and Midnapore, dated 15 June 1842, Home Public, B., 13 July 1842, no. 25 & KW. 69 Moorleedhur, like many other Indian merchants managed the contract for a Euro- pean speculator, who was a Company employee. 70 The extra expense involved in the horse dak establishment for the stretch of 129 miles between Cawnpore and Allahabad was Rs 1700 a month. 71 From: Bushby, General Department to Hamilton, Secretary, North Western Prov- inces, dated 23 March 1842, Home Public, A, 23 March 1842, no. 7. Similar considerations were involved in overruling horse daks on the Calcutta Meerut route in the 1830s. See Home Public, 1 June 1830, no. 52. 72 See for instance Robert Hall, “Tokkaido Road and Region,” Geographical Journal, 27, 3 (July 1937), pp. 353–77; Constantine Vaporis, “Corvée Labour and Peasant Contention,” Monumenta Nipponica, 41, 4 (Winter 1986), pp. 377–414. 94 chitra joshi particularly when they had to pull carts and ekkas (one-horse carriages), averaged a slower rate than foot runners.73 For contractors, investing in horse daks was a risky proposition. Unlike a runner establishment, investment in horse daks involved a larger outlay of capital and greater possibility of loss. When contracts were cancelled it meant a double loss: a loss of capital invested in horses, carts and other equipment and on their future earnings. Moodalliar, who introduced a horse dak on the Nagpore to Raepore route, purchased 58 saddle horses, 34 pairs of bullocks and 8 carts to transport the “overland mail”. Muda- liar’s contract was terminated shortly before his twenty-month term, from May 1839 to December 1840, was over. In addition to loosing the contract, he had to pay a fine of Rs. 9,000 (a third of the total amount due to him for 20 months). Investment in horses was also risky, since horses were frequently seriously hurt in negotiating difficult terrains. Babington, who had a contract for horse dak in Orissa, complained that seven of his horses became lame whilst negotiating the difficult tracks in the Phooljur area. Another reason why contractors were reluctant to continue with horse daks was the system of payments in arrears. The practice of “payment in arrears,” they complained, forced them “to borrow to provide for their support.” Beauchamp, the deputy postmaster of Gaya, who managed the contract for the ekka dak in Bancoorah saw this as an important reason for the collapse of the ekka dak in Bancoorah.74 The practice of withholding wages and deducting fines—commonly used as disciplinary methods— made investment in horse daks unattractive for dak contractors.75 Finally, in looking for the missing horse daks, we need to recognize also the resistance of local communities to new modes of transport. Dauri- yas with long experience in a region were familiar with the routes and with people in neighboring villages. Networks connected with dauriyas were entrenched in local communities, and these were usually opposed to the entry of horse daks.76 Postal establishments often acted in collu-

73 The fact of mails traveling no faster on horses than one would suppose would be attainable by foot runners naturally appears strange to those who are unacquainted with present state of the roads [. . .]. “Postmaster General to Prinsep, Home Public, 1 June 1830, no. 52. 74 Like many other dak contracts, Beauchamp managed a contract which was officially run by a local, Mirza Sarafraz Allie. See: Home Public, 12 May 1829, no. 27. 75 For horse daks to work efficiently, Stockwell, the Postmaster General, recommended that special measures be taken for contractors to be paid “at the close of every month.” Stockwell to Bushby, 16 April 1829, Home Public, 12 May 1829, no. 27. 76 On practices through which traditional runners, deterred the entry of outsiders into their profession, the pattamars in South India for instance see: Gagan D.S. Sood, “The dak roads, dak runners and the communication networks 95 sion with these interests. Dak contractors in Bengal pointed to the enormous problems they faced in running horse daks, against which, “the natives particularly the whole body of the Post office subordinates are greatly prejudiced.”77 In Bancoorah, reports described how a contractor who introduced an ekka dak, had to soon abandon the project because of intense local opposition.78 The ekka drivers who were brought in from Patna were terrified of traveling through the jungles of Ramgarh. Local runners displaced by the new ekka service, worked on these fears, terrorizing the ekka drivers with exaggerated stories of tigers and wild beasts. Many ekka drivers, we are told, fled in alarm.79 Such opposition to new technology and resistance by local interests, is known to have been common in many parts of India. However, among postal officials such stories were woven into a narrative that vindicated the postal establishment. Failed horse daks were attributed to local opposition, and not to non-payment of contract dues.

In conclusion

Clocking the runner’s speed was, as I have pointed out, part of the process of creating a public post marked by its difference from existing postal systems. Calculating speed, ensuring regularity, and projecting estimated time of travel was part of building a “modern” postal system, even when it was sustained by traditional dauriyas. Within the new system, the dauriya operated with the signs of colonial modernity; they were subjected to new rules and regulations, new systems of written contracts, new documents, and receipts. Speed was clocked and recorded. The number of stages, and the rate of travelling were noted down, and the runner’s body was marked with the time taken to reach a particular destination. The point is not to see whether these efforts produced the intended results but to understand the concerns that underlined the framing of such regulations. Tracking the runner’s speed, marking the stages where runners and palanquin bearers were changed (dak chowkis), was also part of a new politics of spatialization. The term dak, in fact, is derived from the term

Informational Fabric of Eighteenth Century India and the Middle East: Intermediaries and Postal Communication,” Modern Asian Studies, 43, 5 (2009), p. 1099. 77 To G. Alexander, Postmaster General Calcutta, from: F. Bashford, dated 9 October 1839, Home Public, 23 October 1839, no. 31. 78 Ekka = a carriage drawn by one horse. 79 Postmaster General to Bushby, Home Public, 12 May 1829, no. 27. 96 chitra joshi

“post”—that is, the stages at which relays of couriers or other modes of communication were stationed.80 Establishing stages and dak chowkis became a way of charting territory, and marking the boundaries of colonial rule. Runners could travel through all kinds of terrain, through deep impenetrable forest where the state could not easily enter. Dauriya tracks were important in connecting and creating the territories of control and empire, penetrating spaces that were seemingly beyond the reach of the state.

And finally the question: where were the dak roads? New road networks moved away from the more familiar routes, and cut through forest, often taking steep ascents to cut down distance. However, these roads could not reach all places, or penetrate all forests. Large stretches of roads were literally created by the footfalls of the dauriya. Continuous running created a path, and charted a route.81 In many places, however, even these mud tracks were not present. In jungle terrain in particular, the official concern was more with removing the stumps of trees, to make smooth running possible.82 Dak roads, in a sense, were present in their absence. In many places, the runner was the road. The modern networks of postal service were built on their backs. I began my story with Kipling’s image of the runner. Let me end with another, popularized by the Bengali poet Sukanto Bhattacharya, in a poem called the “Runner.”83 Written with a deep empathy for the runner, it seeks to capture the heroic and tragic, the moments of joy and pain, the

80 “it is interesting to notice that the vernacular words (dak in Northern India and tap- pal in South and West) are derived like the English word from the stages at which relays or other methods of conveyance, were stationed.” The Imperial Gazetteer of India vol III, Economic (Oxford, 1908), p. 418. 81 Statements like the following were often made: “On stretches of road like the Rae- pore road, where the rains fell at intervals the route of the mail runners by running along the same path created a route which even if not good was passable.” Home Public, 14 October 1840, no. 6 A. 82 Home Public, B., 6 January 1841, no. 28. It was important that routes like the Cal- cutta–Bombay route remained operational all round the year, yet the expense on the road were calculated in terms of their usefulness to the carriage of mail. “[. . .] it is very questionable whether the heavy expense of forming such a road would ever be compensated either by the gain of lines in transmission of public mails or by any increase to traffic towards Sumbhulpore and Nagpore [. . .]. The advantage gained (above alluded to) would be the transmission of the mails at the same a very nearly the same rate of speed the whole year round.”—Fraser, Postmaster to British Resident, Nagpore, Home Public, 4 November 1840, no. 22 B. 83 Sukanto Bhattacharya, Sukanto-Samagra (Kolkata 1970) pp. 77–9. dak roads, dak runners and the communication networks 97 proximity and distance form the burden the runner carries on his back. Beyond these usual tropes of radical poetry, we see other images. What is particularly striking about the runner in this poem is the continuous motion of the runner: villages slip by rapidly, as he speeds towards his destination . . . Tired and hungry, worn out with fatigue, the runner cannot rest at night; he has a long distance to go before the break of dawn. The story of the runner is linked to the oppression of a new time regime, and the constitution of a set of new ideals: principles of speed and regularity, of the idea of public commitment and public good. What I am doing in this paper is to explore the small stories behind such popular images, to understand the system within which the runners worked and lived.

Production Systems and Forms of Labor Control in the Javanese Cigarette Industry, 1920s–1930s

Ratna Saptari

Introduction

It is often claimed that multi-layered production arrangements emerge because of a need to reduce production costs; yet, with the accumulation of knowledge on hierarchical and vertical labor processes, scholars have become increasingly aware of the varying circumstances in which the form, development and linkages of sub-contract relations are shaped, and of the way workers have responded to working under such conditions.1 Such variations underline the need to avoid simple generaliza- tions about sub-contracted work. It is a discussion which to a certain extent recalls Jan Lucassen’s work on artisanal producers and workers, and his earlier work on guilds in 16th and 17th century Holland, in which variations in production relations according to geographical location are demonstrated.2 Although my case study involves an entirely different historical context, it is in some respects in line with Jan Lucassen’s argument against simplistic categorizations of workers and labor relations. The starting point of my discussion is the debate about labor process and labor control, which was stimulated especially by Harry Braverman’s influential analysis of the impact of monopoly capitalism on the structure of the workplace in the United States.3 Braverman tried to show how larger economic structures bring about changes in workplace structures.

1 See, for instance: Rosalinda Pineda-Ofreneo, “Philippine Domestic Outwork: Sub- contracting for Export Oriented Industries,” Journal of Contemporary Asia, 12 (1982), pp. 281–93; Lourdes Benería and Martha Roldan, The Crossroads of Class and Gender: Industrial Homework, Subcontracting and Household Dynamics in Mexico City (Chicago, 1987); I.S.A. Baud, Forms of Production and Women’s Labour: Gender Aspects of Industriali- sation in India and Mexico (London, 1992); Eileen Boris, Home to Work: Motherhood and the Politics of Industrial Homework in the United States (New York, 1996). 2 Jan Lucassen “The Brickmakers’ Strikes on the Ganges Canal in 1848–1849,” Interna- tional Review of Social History, 51 (2006), pp. 47–83. See also: Maarten Prak, et al. (eds), Craft Guilds in the Early Modern Low Countries: Work, Power, and Tepresentation (Ashgate, 2006). 3 Harry Braverman, Labor and Monopoly Capital. The Degradation of Work in the Twen- tieth Century (New York, 1974). 100 ratna saptari

Many scholars have criticized the linear trend and mechanical approach which framed his empirical findings, sometimes at the hand of case studies from Western Europe and North America. A number of points can be discerned in their critique. First of all, the coexistence of different forms of production is not just a “transitional” phase, where gradually the simple forms of production characteristic of small-scale industries make way for more mechanized forms of production within large-scale industries. Quite often, these two types of production may coexist not just as a transitional phase, but in a more or less stable fashion. And when small-scale industries transform themselves into large-scale ones, “traditional” forms of production relations are often carried over to more “modern” forms. At issue here is the historical timing of these changes, and the type of state regime which stimulated particular forms of labor relations. Another point of criticism concerns the labor process itself, the kinds of work created by different types of capitalism (i.e. competitive or monopoly capitalism) and the nature of labor control systems existing within given production regimes. While Braverman pointed towards deskilling in the labor process, where any possibility of control by the worker over the production process is removed through technology and the fragmentation of work, others have identified greater variety in labor control mechanisms. For example, Littler shows that even within “traditional” forms of production three types of control exist, namely familial forms of control, craft control and labor control on a gang basis. These forms may overlap with the more “modern” types, where the piece-work system develops together with new payment systems, i.e. the premium bonus system linked to the recording of job times.4 Braverman’s conception of “skill” and “deskilling” was also criticized by feminists, who argued these categories should not be regarded as descriptions of a mere “technical” process, but as social constructions. A similar criticism was made by Burawoy, who emphasizes that in the workplace not just “things” are produced, but also relations between people. Braverman makes no reference whatever to the processes through which workers comply with the dictates of capital. Yet,

4 “Familial control” is obtained when skilled workers are put in charge to pay and recruit their own child assistants; “craft control” means that the skilled worker or craftsman recruits his own men, determines the hours of work and discipline, and even the organization of production; work on a “gang” basis implies that the gang boss may not only be chosen because of his skill, but because of his relationship with the employer, or with the potential workers. For a discussion about this issue, see: Craig Littler, The Development of the Labour Process in Capitalist Societies (London, 1982). labor control in the javanese cigarette industry 101 quite often, it is through the way control is exercised—whether through coercive or through persuasive means—that workers resist or comply.5 Taking into account these earlier debates, the more contemporary literature has tried to understand the multiple and often contradictory influences determined by competing interests of the state apparatus, management and (both organized and unorganized) workers, which all contribute to the shaping of production regimes.6 Therefore, particular sets of relationships may co-exist and intertwine, and the nature of such combinations cannot be assumed in advance. These discussions are especially important to understand the mechanisms of labor control, the underlying reasons for the emergence of particular modes of labor control, and how they are interlinked. Grounded in this theoretical debate, I focus on one particular historical conjuncture: the cigarette industry in Java and its multi-level sub-contracting arrangements during 1920s and 1930s. It was a period of expansion and contraction, when industrialization was being pushed by the colonial state, but then confronted the Depression years.7 In many respects, this industry shows different forms of what Littler has called “traditional” modes of labor control. However, in a highly competitive, inward-looking historical conjuncture, the observed internal variations within this industry invite us to understand the circumstances under which these variations emerge. First of all, I will examine the changing policies by the colonial state on industrial development and the ideological notions maintained by state officials. I will then discuss the transformations in production units during the crisis years. To finish, I will focus on the diverse forms of labor relations and problems of labor control.

5 For instance, Friedman introduced the term “responsible autonomy”—arguing that, in particular conditions, employers try out forms of control where considerable autonomy is left in workers hands. See: A. Friedman, “Responsible autonomy versus direct control over the labour process,” Capital and Class, 1 (1977), p. 17. 6 This had already been argued by Littler, Development of the Labour Process; Michael Burawoy, The Politics of Production (London, 1985); and Friedman, “Responsible autonomy.” However, their discussions were also framed by attempts to classify particular modes of labor regimes, focusing on management strategies. 7 These periods of contraction and expansion continue to be present even today, yet the kinds of transformations taking place as a consequence have varied. 102 ratna saptari

Colonial perceptions and policies

Industrial development in Netherlands Indies only became a concern of the Dutch colonial government in the early 20th century. The colonial regime became increasingly aware of the limited capacity of the agricultural and extractive sectors to be an “engine of economic growth and welfare.” In 1901, the Dutch Parliament emphasized that prosperity in the Indies could only be increased if a secondary industry could be developed.8 Despite the general commitment to promoting industrialization, it took some time before steps were taken to initiate interventions into the industrial development of the Netherlands Indies. Different arguments and views were raised about the type of industrialization deemed necessary. Some authorities were of the opinion that the standard of living of the indigenous population could be improved only by promoting large-scale industrialization; small industries were considered too weak to face foreign competition. The indigenous sector was considered unable to compete, due to various circumstances—among others, because of the lack of capital and knowledge that was required to improve quality and to become competitive with imported goods. Furthermore there were also very strong (stereotypical) pronounce- ments on the character of indigenous entrepreneurs and laborers. Pleyte for instance mentions the lack of initiative ( gebrek aan initiatief ) and persistence ( gemis aan doorzettingsvermogen) which according to him were characteristic of the Javanese worker.9 Versluys mentioned the “indolence” of the indigenous laborer; among the Javanese workers, he believed, there were no other stimuli than the very modest demand of food and clothing— and this made them “carefree” (zorgeloos). According to Versluys, this meant: a) irregular attendance of the Javanese worker, shown by the fact that absenteeism was higher than in Western countries; b) a tendency to give up easily when he was given work that displeased him; and c) the absence of any initiative to produce more each day.10

8 Peter H.W. Sitsen, Industrial Development of the Netherlands Indies (s.l., 1942), p. 59. 9 He attributed these characteristics to the mild climate in the Netherlands-Indies (as opposed to the difficult conditions in Holland for instance): “[. . .] after all, a very mild, hardly changeable climate . . . does not compel a steady effort, nor saving for wintry seasons with partial cessation of labor and a nature that slumbers.” C.M. Pleyte, De inlandsche nijverheid in West Java: sociaal-ethnologisch verschijnsel (Batavia, 1911), p. 20. 10 J. Versluys, Fabrieksnijverheid in Nederlandsch-Indie (Batavia, 1917), p. 46. labor control in the javanese cigarette industry 103

In the two decades that followed, little was done to put into practice the general views which circulated in the earlier part of the century. In the meantime, Sitsen, the director of the Industrial Division of the Department of Economic Affairs in Batavia, argued that too little attention was given to the social and economic conditions in the Netherlands Indies itself.11 According to him, small-scale industry was more conducive to the general character of the indigenous population, because it was “completely suitable to the mentality and nature of the population.” He referred to “the sense of obligation to give mutual assistance” (italics in the original) and secondly, to the “yearly recurrences of slack periods in the cycle of consumption connected with the harvest periods.”12 During the period between World War I and the Depression, particularly the import of textiles from the Netherlands and the United Kingdom suffered; each country contributed only 7% to total textile imports, while imports of Japanese textiles reached more than 75% of the total in 1933.13 In 1933, the Crisis Import Ordinance was enacted in Java. This included the introduction of an import quota system and a licensed import system.14 At the same time, attempts were made to protect domestic industries, and prevent destructive competition between firms of different sizes. In 1935, the Ordinance for Regulation of Enterprises (Bedrijfsreglementerings Ordonnantie) was introduced. This regulation was basically an attempt by the government to prevent the establishment of “undesirable” enterprises, and to protect the weak ones. Among other things, it included restrictions on the introduction of large machinery in the factories. During the depression years and the period thereafter, local manufacturers found a good market, because imports were stopped. Indeed, it was between 1930 and 1940 that manufactures grew rapidly. The proportion of the workforce in this sector increased from 10.6% to 12.8%; there was a growth of

11 Sitsen, Industrial Development, p. 60. 12 Sitsen, Industrial Development, pp. 16–17. Sitsen’s idea about the nature of industrialization in the Netherlands-Indies reflects to a large extent the general outlook of the colonial government during and after the Depression of the ’30s. Two three-volume reports of the Labor Office (kantoor van arbeid) focused on the clove cigarette industry and the batik industry, describing the production systems, wages, working conditions, observance of the Factory Act, and prohibitions on night work and child labor. 13 G. Prince, “Dutch Economic Policy in Indonesia, 1870–1942,” in A. Maddison and G. Prince (eds), Economic Growth in Indonesia, 1820–1940 (Dordrecht, 1989), pp. 203–26, at 216. 14 H. Matsuo, The Development of Javanese Cotton Industry (Tokyo, 1980), p. 22. The first was meant to balance imports and exports of certain goods, and the last was aimed at controlling imports through allotment of licenses to importers. 104 ratna saptari manufactures which were less dependent on government protection, and both modern factories as well as small non-mechanized firms increased their share of total output.15 Within this context, a series of reports were prepared on the condition of “native” industries in the Netherlands Indies and the social welfare of the workers. The Labor Office initiated a number of comprehensive surveys, which focused among others on the metal, batik and cigarette industry in the Netherlands Indies. These surveys were done between 1924 and 1934. Their findings reflect the colonial government’s ambivalent attitude to non-European industrialization: on the one hand it attempted to protect domestic enterprises, but on the other it regarded them as backward and lethargic.16 The cigarette industry had already developed prior to any intervention by the colonial government to stimulate industrialization in the colonies. From the early moments of its development, small producers were mainly part of a large network of putting-out merchant capitalists. Indeed, accounts of the early “indigenous” (i.e. Javanese) entrepreneurs almost never referred to the artisanal klobot 17 producers, who gradually accumulated enough capital to establish their workshops or factories.18 Although records are scarce for these early periods, it seems that the more successful indigenous entrepreneurs had backgrounds either in trade (usually textiles and tobacco), or were the sons of village heads. Thus, they had enough capital to try out diverse trading activities before they entered the

15 George Missen, Viewpoint on Indonesia: A Geographical Study (Melbourne, 1972), pp. 169–71. As Missen himself acknowledges, the extent of this growth is nevertheless subject to controversy, and some argue that the high rate of growth should not be explained so much as a result of direct government intervention, but, more significantly, as a result of changes in the export sector, in the pattern of imports and the overseas trade balance (Ibid., p. 169). 16 The reports on the strootjes and cigarette industry were undertaken by van der Reijden, and they were published between 1934–1936. See the impressive volumes by van der Reijden, Rapport betreffende eene gehouden enquête naar de arbeidstoestanden in de industrie van strootjes en inheemsche sigaretten op Java. Publicatie van het Kantoor van Arbeid, no. 9, Bandoeng [West Java], 1934; Publicatie van het Kantoor van Arbeid no. 10, Bandoeng [Central Java] 1935 and Publicatie van het Kantoor van Arbeid no. 11, Bandoeng [East Java] 1936. 17 Cigarettes containing a mixture of tobacco and cloves wrapped in maize sheath. 18 Independent home production did exist; the exact development of this category is quite difficult to obtain, since the strootjes rapport did not include the cottage industry as a separate recording category. Van der Reijden classified the cigarette industry with three categories. “Small enterprises” were those producing below 10 million cigarettes a year; “medium enterprises” produced between 10–50 million cigarettes a year; and “large enterprises” produced more than 50 million cigarettes. Van der Reijden mentioned the cottage industry, but considered it not sufficiently significant to merit further investigation. labor control in the javanese cigarette industry 105 klobot industry.19 Chinese producers appeared later on, and gradually they dominated the industry.20 Partly as a result of the discriminatory policies of the colonial government which privileged Chinese traders, Chinese tobacco trading became a stepping stone for the manufacture of tobacco products. Particularly with the lifting of travel and residence restrictions for the Chinese between 1910 and 1926, access to labor and markets at the village level became easier.21 It was during this period that tensions between Chinese and indigenous capitalists reached a climax. In 1918, a number of Chinese enterprises were burnt down by indigenous traders and kretek producers in Kudus (the stronghold of the Javanese kretek manufacturers).22 Given the arrest of these indigenous producers, and given their absence from any economic activity in the Jepara-Rembang residency of Central Java, the Chinese producers managed to gain a stronger foothold in this area.23 When the industry spread to East Java in the early 1920s, Chinese entrepreneurs were dominant at almost all levels of production, while the indigenous entrepreneurs of East Java continued to operate mainly on a small scale. Meanwhile other changes were taking place. With the colonial government’s attempt to implement an import substitution policy in 1924 and 1926, import duties on tobacco and taxes on paper (particularly important for rolled tobacco and cigarette-wrapping) were enacted. These interventions were meant to save foreign exchange, and stimulate the growth of domestically-produced foreign cigarettes.24 Indeed kretek cigarette

19 Van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kan- toor van Arbeid no. 10, 1935, p. 25; A. Budiman and Onghokham, Rokok Kretek: Lintasan Sejarah dan Artinya Bagi Pembangunan Bangsa dan Negara (Kudus, 1987), pp. 92–106; See also: Lance Castles, Religion, Politics and Economic Behavior in Java. The Kudus Cigarette Industry (New Haven, 1967), p. 50. 20 According to Sitsen, Chinese penetration in the secondary industries was not yet threatening the position of the Javanese entrepreneurs. In textiles, for instance, although a number of Chinese entrepreneurs bought up looms from Indonesians between 1935 and 1940, indigenous entrepreneurs managed to set up thousands of modern weaving establishments (Sitsen, Industrial Development, p. 20). 21 For a concise analysis, see: Richard Robison, The Rise of Capital (London, 1986), p. 301. 22 Castles states that about fifty houses were destroyed, and 8 Chinese killed; and about 2,000 Chinese fled to Semarang. On the side of the rioters, two were killed, 60 were wounded and 75 were arrested and tried, receiving sentences ranging from 9 months to 15 years. See: Castles, Religion, Politics and Economic Behavior, p. 63. 23 The riot was primarily anti-Chinese, rather than anti-capitalist. The Dutch “Office of Chinese Affairs” at that time registered that the majority of those arrested for inciting the riot, were hajis, and that not anti-capitalist, but religious slogans were shouted out (Mailrapport 471x/1918, p. 5). 24 W.A.I.M. Segers, Changing Economy in Indonesia. Vol. 8: Manufacturing industry, 1870–1942 (Amsterdam, 1988), p. 171. 106 ratna saptari production increased rapidly in the years that followed, the main producing area being the Northern part of Central Java and especially the town of Kudus. The dominant form of production in the main producing area of Kudus was putting-out work done by home-based workers (kornet).25 Until 1929, the home-workers (not including the abon26) in this town still comprised 82% of the labor force.27 As one report mentioned, one of the largest factories in Kudus (the largest producer in the residency) in 1924 had only 20 permanent workers sorting tobacco and putting cigarettes into packs. The rolling of klobot cigarettes, which occupied the majority of the workforce, was done at home by family labor, often involving very small children.28 That cigarette workers were mainly part-time workers is also apparent from the fact that during agricultural peak seasons, many cigarette workers were absent from the workplace.29 Why was home-working the most predominant form of employment in the Kudus area? Mangoenkoesoemo attributes the incidence of putting- out to the fact that workers were still reluctant to travel long distances and stay overnight.30 Castles argues that “the largest factories are located several miles out of town, and even those within the town, a high proportion of the workers come in each day from the surrounding villages.”31 Even in the 1950s (the period on which his fieldwork was based) a homogeneous industrial proletariat had, according to Castles, still not emerged from the cigarette industry, even though the total number of workers then amounted to 20,000 workers.32 He explains this situation by pointing to a combination of (social) factors existing in rural society at the time. He

25 Kornets refer to the home-based rollers only. Van der Reijden distinguishes them from the Strootjesrollers (cigarette rollers) who roll the cigarettes in the factories; siga- rettenmakers are those who roll the cylindrical cigarettes. 26 The middle-men, working independently for several factories or working for one specific factory, were those who consigned work to the home-workers. 27 Van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kan- toor van Arbeid no. 10, 1935, p. 17. Van der Reijden started his research on Java in 1929, and ended it in 1934. For this reason, the figures cover this period. 28 Dienst der Belastingen, 1925, p. 198. The report mentioned that sometimes the children were as young as three years old. Their task was to tie up the cigarettes, and cut the uneven tobacco at the ends (ibid.). 29 This situation was usually already anticipated by producers, since a few months before the harvest and planting seasons, production was increased above the amount required. D. Mangoenkoesoemo, Bijdrage tot de kennis van de kretekstrootjes industrie in het Regentschap Koedoes (Batavia, 1929), p. 11. 30 Mangoenkoesoemo, Bijdrage tot de kennis, p. 6. 31 Castles, Religion, Politics and Economic Behavior, p. 53. 32 Ibid., p. 53. labor control in the javanese cigarette industry 107 argues that because the workers were mainly women, and were “not likely to be the principal breadwinners of their households,” and because most of them supplemented their husband’s or father’s earnings (as farmers, laborers or craftsmen) by rolling cigarettes, they did not become commit- ted to “a way of life molded largely by factory employment.”33 Although (lack of ) orientation to factory employment might emerge out of the unwillingness to look farther afield for jobs (or might be attributable to ideological notions about the desirability to seek employment out of one’s hometown), one should also examine the rationale of workshop owners for opening establishments outside the towns. Many of these abons were factory workers who were retrenched between 1930 and 1932, and many of them did not rely only on cigarette production for their income. Van der Reijden notes that out of 64 workshops established in 1933, only three took up cigarette production as a main occupation, while the rest were involved in other occupations. The cigarette sector for them was therefore also a sideline activity. Around 40% of these workshop owners were also waroeng owners, another 40% were village officials, and the rest were farmers or in the trade sector.34 This probably explains why cigarette workshops were set up in areas outside the towns. I should add that not all work conducted outside the factory consisted of cigarette-rolling by women. There were other types of work, such as box-making and tobacco processing, which were also done outside the factories.35 It is therefore important to examine the circumstances under which putting-out occurred; a distinction should be drawn between “labor-abundant” and “labor-scarce” situations. These variants refer not only to demographic conditions, but also to social and political conditions. In labor-abundant situations, the bargaining position of workers may be weaker; their labor can be easily replaced by others. In labor- scarce situations, employers became more dependent on the willingness of workers to work for them; they had to search for workers willing to provide their labor for the production of cigarettes, which strengthened the latter’s position somewhat. Three further points should be made when analyzing the predominance of putting-out and home-based work in the cigarette industry.

33 Ibid., 53–4. 34 Van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kan- toor van Arbeid no. 10, 1935, p. 9. 35 In practice, certain forms of sub-contracting were arrangements with specialized small units which provided processed tobacco, cloves or maize sheaths. 108 ratna saptari

Firstly, we should note the timing of its boom period. Kudus, being the major center since the early 1920s, had to look beyond the town borders to the adjacent villages. Although extra labor was needed, large investments for the construction of larger workshops were avoided because of the instability of markets and capital.36 Secondly, an important factor determining the structure of the labor force is, of course, the nature of capital. Particularly in the Japara-Rembang residency, the majority of the kretek producers were small-scale indigenous producers. Often they could not afford to build separate structures, or to acquire additional plots of land to house the workers. Thirdly, this industry was marked by a very irregular consumer market, which involved fluctuations in consumption in several periods during the year. Van der Reijden mentions for instance the paceklik37 period (November, December and January) at the beginning of the rainy season, when the majority of the rural population were working in the fields. At that time, not much money was available to purchase klobot. Also, since the milling period of the sugar factories began in May, and finished at the latest in September, less money was available in the period between October and May. In addition, various taxes had to be paid during this time. Because of this irregularity in (consumer) demand, retaining a large labor force as well as large stocks of raw material, even if affordable, would be disadvantageous. This type of production, where a few factories sub-contracted to numerous home-workers, was however not universal across the whole of Java. Other firms in Central Java already had a large entourage of factory workers, and only recruited a very small number of home-workers.38 For instance, in the Principalities (Solo and Jogja), the main categories of workers mentioned in survey reports were the factory-based klobot rollers, cigarette rollers, piece-rate workers and daily wage workers. This finding can be explained by the smaller role of cigarette production in this area.39 Also, in Pekalongan, where a relatively established Javanese middle class

36 Van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kan- toor van Arbeid no. 10, 1935, pp. 7–10. 37 When people were out of money. 38 Solo and Jogja, where the earliest firms were found in 1897 and 1914 respectively, were also primarily dependent on home-based workers. Van der Reijden mentions that it was only in 1933 (after the tobacco ordinance) that a larger percentage of factory workers were recruited (van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kantoor van Arbeid no. 10, 1935, p. 157. 39 In 1929, the Principalities (Vorstenlanden) constituted 4% of total production and 1.7% of the total labor force in Central Java (Ibid., pp. 156–8). labor control in the javanese cigarette industry 109 had developed (primarily out of the batik industry), almost all the workers were concentrated in small factories or workshops, and only those making the cigarette packs worked at home.40 In East Java, during the early stages of the industry’s development, labor in factories was the primary form, with labor drawn from the kampungs of the urban centers, or from the villages in the immediate surrounds of the town. This structure of production changed quickly when the Netherlands Indies revenues plummeted as a result of the Depression. Production did indeed fall sharply between 1930 and 1932; in Central Java, output slumped from 4 billion cigarettes in 1929 to 2.6 billion cigarettes in 1932. In contrast, however, East Java never experienced a slump, and in 1932 it was producing more cigarettes than Central Java.41 Another important turning point was reached when the government enacted an excise on tobacco in 1932, known as the “Ordonnantie Bia Tembakau”. This Ordinance stipulated a 20% excise on the retail price for klobot cigarettes, without dis- criminating between different-sized enterprises. This dealt a heavy blow to the smaller industries, particularly the relatively independent cottage industries owned by indigenous producers. The excise tax was accompanied by a regulation which required a separation of the place of sale from the place of production; small workshops using family labor and less than two wage laborers working in the corner of their tobacco shops either had to obtain a separate building, or were forced to close down.42 Thus, either the home industries went out of business, or small workshops were established instead. Soeara Oemoem, a leading newspaper at that time, calculated that more than 100,000 workers were unemployed in Kudus, Blitar and Tulungagung, the main klobot centers. Larger factories were also hit by the new excise. Nitisemito’s factory, the largest Javanese enterprise at the time, collapsed. In Ponorogo, a factory with 1,000 workers closed

40 And this category constituted only a fraction of the labor force: 1.6% in 1929, and 2.8% in 1933 (van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kantoor van Arbeid no. 10, 1935, p. 112). 41 It was producing more than 2.8 billion cigarettes. This difference between Central and East Java is also due to the fact that prior to the economic crisis, Central Java was producing double the output of East Java (4 billion compared with 2.2 billion). However, it is interesting that production in East Java never declined after that time (van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kantoor van Arbeid no. 11, 1936, p. 141). 42 van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kantoor van Arbeid, no. 11, 1936, p. 120. 110 ratna saptari down. In Jogja and Solo, many indigenous factories also closed down. In Tulungagung all factories stopped producing temporarily.43 Ironically enough, consumer demand increased after 1932, as can be inferred from the fact that output doubled in comparison to 1929. In Central Java, production increased from 4 billion cigarettes in 1929 to 6 billion in 1934, while in East Java, production increased from 2.3 billion in 1929 to 4.4 billion in 1934. To a certain extent, this explains the increase in the number of small enterprises during the depression years. In 1929, the ratio between small and large enterprises was 9 : 1 in Java as a whole, but by 1934 this had changed to 32 : 1.44 Although the number of large- scale factories in Java (those producing over 50 million kretek annually) remained at 32, the number of medium scale factories increased from 61 to 87, and the small scale plants from 295 to 1,046. The medium and small scale plants employed 80,133 in 1934.45 At the same time, and despite the fact that small firms were becoming more dominant, these firms were operating in a more centralized fashion. Thus, in terms of the proportion of home workers to factory workers, the gap between firms in Central Java and East Java also started to decrease. The proportion of home workers in Central Java was halved to 46% by 1934, while that of East Java remained constant (but with a lower base). Not only were Central Javanese firms using more factory-rollers, but they were also recruiting more factory coolies and monthly workers.46 Central- ization was certainly a means by which firms could retain close control of production and quality of the cigarettes,47 but it did not necessarily mean a shift towards a more permanent and costly workforce. These transformations taking place in the 1920s and 1930s show how the expansion and contraction of an industry produce divergent consequences for different types of enterprises. How this played out in the workplace and in the different strategies of labor control will be detailed next. I will examine labor relations at different levels: the factories, the workshops and home-working.

43 Soera Oemoem, 5 February 1933. From the collection of Sneevliet, B-15. 44 van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kan- toor van Arbeid no. 11, 1936, p. 145. 45 See: John Orval Sutter, Indonesianisasi: Politics in a Changing Economy, 1940–1955. Data Paper no. 36, Department of Far Eastern Studies, Southeast Asia Program, Cornell University (Ithaca, 1959). 46 van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kan- toor van Arbeid no. 11, 1936, p. 145. 47 See also Castles 1967, p. 34. labor control in the javanese cigarette industry 111

Production in the factories

The basic phases of work involved in clove cigarette production are tobacco mixing, clove-grinding, the rolling of the cigarettes (including cutting), packaging, and the making of boxes and packs. How these different tasks were done, what type of workers were assigned which tasks, and the wage categories they were allotted to depended quite strongly on the local exigencies and the particular economic conjuncture. Workers in the factories in this period can be divided into: monthly workers (mandor, scribes, drivers); daily workers (sorters, mixers); piece-rate workers (sorters, mixers, clove grinders, packagers and rollers) and casual workers (klobot ironers). Those who were not working in the factories, were called kornet or lemburan.48 Work in the factory mainly involved the task of rolling, ironing of klobot, cutting and packing. Sometimes the preparation of the tobacco mix was done in the factories, but often enterprises would order ready-made mixes.49 The impact of the crisis in the period between 1929 to 1933, was highly uneven for these different categories of workers. In total, the cigarette sector in Central Java experienced a decrease in its total workforce from 45,869 workers to 31,117 workers. In contrast East Java experienced an increase even though the total was still much less than in Central Java, namely from 12,329 in 1929 to 18,867 workers in 1934.50 Employment strategies were designed both to create a flexible work force and to ensure efficient systems of payment. Thus, quite often the same kind of job categories could be based on different forms of payment, with the consequence that even within the same sections, workers could be differentiated. For example, as mentioned above, sorters and mixers could be paid according to piece-rates or time-rates; supervisory work could also be at piece-rates, or it could be time-rate work. Furthermore,

48 The difference in terminology relates to the difference between the two major cigarette-producing areas, namely Kudus in Central Java and Kediri in East Java. Both refer to “home-workers,” but kornet is used in Kudus and lemboeran is the term used in Kediri. There are also some differences in the nature of the putting out system, which will be discussed further below. 49 The use of apprentices became more common in the earlier years, when the work of rolling was divided with another worker, usually children; afterwards, it was somewhat reduced because of the arbeidsinspectie (labor inspectorate) and the economic crisis, given that large numbers of adult workers became unemployed (van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kantoor van Arbeid no. 10, 1935, p. 55). 50 Van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kan- toor van Arbeid no. 11, 1936, p. 101. 112 ratna saptari certain tasks (e.g. packaging, etc.) could be done by both factory labor and home-workers. However, it is not clear how decisions were made about who should take up piece-rate work, and who should work on a time-rate basis. In examining the modes of wage payment in 19th century Britain, Littler argues that, up to the 1880s, time-wage payments were the norm, and that they occurred within the internal contract system.51 How- ever, this system quite often failed to induce more effort intensification; subsequently the piece-rate system was therefore introduced to enhance control over labor. In the case of the cigarette industry, cigarette-rolling had always been recorded as piece-rate work; workers were paid for every 1,000 cigarettes, whether the work was done within the factories, or outside them. But other types of work such as clove-grinding, which was initially paid by day, became piece-rate work after 1931. This practice is explained not so much by workers’ resistance to effort intensification, but by the need of employers to reduce capital risk.52 Group control already existed through the wage payment system: although rollers in the factories usually trimmed the cigarettes themselves, they would incorporate other workers (often family members) to trim the rolled cigarettes, in order to increase efficiency, and these trimmers would be given a small part of the wage by the rollers. The recruitment of trimmers was not considered to be the responsibility of the firm. The fact that factory workers utilized their family labor to roll cigarettes at home also meant that the factory workers themselves were the ones responsible for supervising production. In Pati, the ironing of the maize sheaths, trim- ming/cutting and packing were done by knechtjes—assistants or apprentices (either children, youth or adult women) who might, or might not be related to the rollers. Besides group control, the supervision of the production process was done by the supervisors and mandor. “Simple and hierarchical” control was operated, where more often than not arbitrary forms of punishment and criteria for punishment were left to the mandor or depot owner.53

51 The “internal contract system” referred to a situation where a factory worker would either develop a team-based work relationship based on internal agreements, or where workers asked others to work with him. Littler distinguishes three types of systems, operated through familial relations, through craft control, and on a gang basis, with the gang boss in charge of the work. See: Littler 1982. 52 Van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kan- toor van Arbeid no. 11, 1936, p. 54. 53 See: R. Edwards, Contested Terrains. The Transformation of the Workplace in the Twen- tieth Century (London, 1979). labor control in the javanese cigarette industry 113

Factories differed in their worker composition, depending on the extent to which they put-out their work. Box-making in Madiun (East Java) was at first done in the factories, but after 1930 boxes were also made outside the factories. In Pati, Central Java, box making was initially done by factory workers without pay; but later it was done outside the factories—to a large extent by women and children at home (most of them lived in the urban kampungs with no intervention from middle-men). This division of labor by gender also showed variations within and between regions. Cigarette rollers could involve an unequal proportion of men and women. In Pekalongan, Central Java, factory rollers were exclusively men, whereas in Ponorogo, East Java, they were exclusively women. In other regencies, there was usually a mix of men and women rollers within the factories. In Kediri (East Java), women constituted 65% of the factory rollers. However after the depression years (1931–1932), this situation changed. In Pekalongan, women started to enter the factories; and in Ponorogo men began to enter factories where previously only women were the rollers. Child labor also competed with adult labor in periods of crisis. For instance, in Semarang prior to the depression, 40% of factory workers were young workers (halfwassenen), but after 1931 they were replaced by adults, of which one-third were adult women. In Pati, Central Java, the ironing of maize sheath for the klobot and the cutting and packing were previously done by children and youth, and sometimes adult women. However, this situation changed as the crisis intensified, adult unemployment grew, and the age of children became a criteria for Labor inspection approval.54 It is also in this area that factory apprentices were initially considered to be employees of the factory, and obtained wages from the factory. However, subsequently they became employed by the rollers. In Semarang (Central Java) until 1931, some 40% of factory workers were young workers. After 1931, there were more adults, and one-third of these were women (p. 100). However in Pekalongan, there were only men in the factories until 1932, when women daily workers started to be recruited with wages lower than men. In Ponorogo (East Java), the rollers were exclusively women, while in other regencies of Madioen, both men and women were involved in rolling. Grinding of cloves, the packing of

54 Van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kan- toor van Arbeid no. 11, 1936, p. 54. 114 ratna saptari cigarettes and box-making were often done outside the factories, by male workers at home who were employed directly by the factories.55 Fines and deductions were often introduced, and these had a number of overlapping purposes. They were primarily mechanisms to ensure the quality of the product, and to place part of the cost of production on the shoulders of the workers. However, quite often measures targeted at production activity might also have disciplinary implications.

Production in workshops

In the workshop system, a crucial role was played by the abon or depot, a man or woman who was responsible for the production and delivery of the cigarettes. This responsibility included also the recruitment or selection of the workers, and the purchase of some of the components of production. Since this role was crucial for the enterprise owner, a number of selection techniques were used to choose a worthy abon or depot. In Kudus, it involved a guarantee from the village head stating that a candidate showed good behavior (bewijs van goed gedrag); this was followed by a verification of the financial background of the designated abon by factory staff; on top of this, other abon who live in the same village were asked to monitor the behavior of the abon. In their capacity as sub-contractors, these abon or depot owners were given some power in the production process, particularly through their position in distributing the tobacco-clove mixtures, or other ingredients; through their roles in disbursing payments; and through their power to select the good cigarettes from the bad ones. However part of the production costs were also in their hands, as they had to provide the transportation of the raw material to the villages, or the products to the factories. Although most of the work put out to the sub-contractors was cigarette-rolling, after the crisis other tasks were also handed over to these sub-contractors— such as the making of cigarette-packs and boxes, the grinding of cloves or the mixing of tobacco with cloves. The recruitment of labor was a task of these sub-contractors. The skill- level needed or the willingness of workers to provide their labor was considered unproblematic since, as van der Reijden states, “elke Inlander

55 In the end, the putting-out system specific to Kudus became too cumbersome, because the factories were concentrated in Kudus town, while the home-workers lived as far as 12 to 24 kilometres away (Mangoenkoesoemo, Bijdrage tot de kennis, pp. 10–1). labor control in the javanese cigarette industry 115 verstand heeft van het rollen van strootjes en dus de keus ruim genoeg is” (“every person knows how to role cigarettes, so that there is enough choice of workers”). But the control of the quality and speed of their work was usually considered a problem. The idea that Javanese workers were simple, careless folk, lacking a sense of responsibility, is voiced by van der Reijden in his report: In determining the method of working, [the employer] takes into account the less developed sense of responsibility of simple desa people and their notorious carelessness.56 To deal with this problem of control, the attempt was made to set up workplaces in the sub-contractors’ own homes. This approach intended to prevent theft, and also to have more control over the work, so that fewer cigarettes would be disqualified from the “sortiran”. However the “willingness” of workers to come to the middle-man’s home varied from place to place, possibly depending on labor market conditions and the degree to which workers had other means of income at their disposal. In the Kudus (Central Java) area, workers were less “willing” than the workers in Kediri (East Java), for instance. Van der Reijden speculated that this difference might be because workers did not like the tight control or the work schedule imposed by the abon, if the work was to be done at the abon’s place. In East Java, where depot workers normally worked in the depot- owners’ house or in a rented workplace, and workplaces were often just covered squatting places in the backyard, workers sometimes had to bring their own mats to sit on. Since the workforce might total to 20–40 rollers and workplaces could become larger-scale affairs, the distinction between factory-workers and home-workers became rather blurred. Whereas the abon system in Kudus was a regular phenomenon throughout the year, in Kediri the depot system was used especially during periods of high demand. In this respect, the difference between the Kudus area in Central Java and Kediri in East Java is difficult to explain. One possible explanation is that because the batik industry also predominated in Central Java, the labor supply available for cigarette production was reduced. This was not the case in Kediri, possibly a location where the cigarette industry was one of the few attractive options for the workers in the area, encouraging them to go to the workshops.

56 Van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kantoor van Arbeid no. 10, 1935, p. 48. 116 ratna saptari

Some exceptions aside, not so many workers stayed very long in the workplace (as in the batik industry). A few coolies with monthly status worked in the smaller enterprises, and also did household chores for the owners; this also applied to some who had supervisory roles, and originated from other towns in East Java. When long working hours were involved, workers might have no energy left to go home, and some workplaces indeed provided sleeping quarters for the piece-rate workers. The impact of the crisis was also shown by the total number of home- workers whom abons or depots could hire. Survey data suggest that each sub-contractor worked with a reduced number of home-workers during the crisis years. In Pati between 1929–1930, for example, each abon recruited approximately seven home-workers; in 1932, they were only able to recruit two or three home-workers. The distribution of raw material to the home-workers could occur on a daily basis; alternatively, they could be given the whole amount of the material to be worked on. In the latter case, the rolled cigarettes were supplied to the sub-contractor one day before the sub-contractor had to deliver to the manufacturer. The amount of tobacco mixture not only depended on the time available, but also on the number of people who were involved. In some cases, a family with two to three other persons were involved. In other cases, only the housewife was involved, or women with children (p. 48).

Home-working

Home-workers were an integral part of factory work, and they were recruited particularly during periods of expansion or during seasonal peaks in the annual production scheme. Although between 1929 and 1934 the proportion of factory rollers increased two and half times (from 18,499 workers to 47,149 workers) in Java as a whole, and the total number of home-workers decreased by 31% (from 46,670 to 32,362 workers), it should be noted that 40% of the cigarette rollers produced at home. In Kudus, until 1933, each abon would have about ten home-workers (kornets) under his or her wing, and each kornet would in turn involve also family labor. In Malang, most small enterprises initially employed wage workers only for the task of rolling; the other tasks were done by their own family labor. As mentioned earlier, the raw material given to the home-worker not only depended on the time available, but also on the number of people who would be involved. Some abons or depots would give material on a daily basis (to be distributed in the morning, and col- labor control in the javanese cigarette industry 117 lected in the evenings) and some were supplied the whole amount, the cigarettes being delivered to the middle-men one day before they delivered to the factories. In Kudus, the number of home-workers decreased since 1929; but in East Java, where the industry grew, the number of home-workers also increased. Prior to the implementation of the Factory Act, factories would recruit migrant workers from outlying districts especially to cope with peak periods. During periods of long working hours, workers preferred to stay overnight, rather than to go back home. They slept either in the work location itself, or in make-shift huts quickly constructed by enterprise owners. These sleeping quarters were usually adjacent to the workplace, or to the owners’ homes.

Workers, vulnerabilities and modes of resistance

According to van der Reijden, incidences leading to abuse were not usual in Central Java; in East Java, however, they were more the norm rather than the exception, particularly in the Blitar-Tulungagung area. For instance, usurous trucking (winkelnering) was particularly rampant after the enact- ment of the tobacco excise in 1928. It took the form of paying wages partly or fully in natura, instead of cash; or of forcing workers to buy goods from the owner, which would then be deducted from their wages. One example from Blitar, mentioned by van der Reijden, involved an assistant accoun- tant who sold various articles such as rice, bicycles, furniture, as well as cheaper merchandise such as cloth and traditional blouses (kebaya), sandals, trousers etc. These items were given to the factory personnel (such as the supervisors of the rolling, packing and sorting sections) to sell to the workers at much higher rates than the market rates. In one enterprise, the compulsion to buy tea from the owner at the rate of 1 cent per pack was a weekly occurrence. In such cases, workers could not refuse, for fear of reprisals, either through more stringent selection, or even total rejection of the cigarettes they produced. Physical abuse such as beating and kicking occurred, and public shame and humiliation was inflicted on workers, both on men and on women.57 The cases which van der Reijden

57 For instance, a man had to stand in the gudang (godown) with a large pack on his head, while other workers were forced to spit on him. His wage for the day was withheld from him. Another man had to stand for two hours in front of the employer, and others 118 ratna saptari cited were more illustrative of this particular area than of Central Java. His own response was to seek explanations for this behavior in the cultural background of the predominantly Chinese factory owners: It suffices to point out, that because of the cigarette industry—which is wholly in the hands of the Chinese population—the desa population of Blitar has a very bad moral reputation. Not only is this city teeming with prostitutes and brothels, but several Chinese have even established a European house as a rendez-vous outside the city perimeter, where many employees and mandoers play the “big gentleman” on Saturday and Sunday evenings.58 Coercion and punishment, as Edwards and Littler have argued, are forms of labor control in the face of an absence of other means of control which are built into the bureaucratic or technological structures in the workplace.59 However it is not clear in this case whether employers were reacting to worker’s resistance, or just to worker’s inefficiency and failure to meet targets. Fines and deductions were not phenomena specifically linked to the large factories. Such practices were also found in the workshops. In fact, many of the workshop owners whose workplaces were an extension of their homes had more opportunity to enforce their power and authority over the workers, especially when, during peak periods, workers were sometimes forced to stay overnight. A second aspect one should also consider is that of the ethnic background of employers. Van der Reijden suggested that the kinds of violations mentioned would have occurred mainly in the Chinese factories. Accounts of employers’ violations in the batik industry, for instance, have also pointed to the Chinese rather than to the Javanese employers as the usual culprits.60 Workers’ attitudes about the Chinese in such cases were often laden with racial stereotypes about their meanness and sever- ity. However, it is not clear whether the lack of similar incidents in the Javanese workshops and factories are due to the adherence by Javanese civil servants conducting surveys to the prevailing stereotypes, or whether such occurrences really did not occur there.

were also forced to spit on him, but they all refused. Two women were forced to conduct knee-bending exercises, and their wages were also withheld. Van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kantoor van Arbeid no. 11, 1936, pp. 69–70. 58 Van der Reijden, Rapport betreffende eene gehouden enquête, Publicatie van het Kantoor van Arbeid no. 11, 1936, p. 70. 59 Edwards, Contested Terrain; Littler, Development of the Labour Process. 60 Incidents reported included workers being locked up or starved of food, while they were forced to keep working. labor control in the javanese cigarette industry 119

Modes of resistance are difficult to identify, particularly for the workshops. If, in the batik industry, flight and high labor-turnover are interpreted as ways in which workers tried to avoid too heavy exactions and pressure from employers,61 the reports on the cigarette industry do not seem to record such actions by the workers. In the factories, the possibility for collective action, although not often reported, seems to have been more viable. During the crisis years, workers’ wages in one enterprise in Kudus were reduced, with the result that 123 workers went on strike. The employer did not want to give in, and decided to take on other employees instead. But the workers had saved up money with the employer, and wanted their money back. The employer, however, was only willing to grant the money in December of the following year. The workers then went to the assistant wedana in Kudus. It is reported that this government official then “managed to convince the employer,” but whether this convincing also meant that an actual payment to the workers was made, remains unclear. In another case, it was reported that as a result of the shift towards more centralized production (and a reduction of home- working), many of these workers were instructed to work in the factories or workshops. However around 8,000 to 12,000 home-workers (kornets) did not have a workplace to go to, and were housed in very cramped quarters (pp. 80–81). Workers’ organizations do not seem to have existed in this period.62

Conclusion

When the manufacturing industry in Java was experiencing a rapid growth throughout the 1920s and 1930s, the cigarette industry in Java also faced major transformations. The streamlining of the industry which manifested itself in the decline of the large enterprises and of home-working on the one hand, and the dramatic growth of small-scale industries on the other, brought with it shifts in industrial linkages and production relations. This

61 Often reported was that employers used the services of village officials to track down their workers, who were given supplies of cloth to batik. See: P. de Kat Angelino, Batikrap- port Vol. III, Oost Java (Batavia, 1931), pp. 30–2. 62 The first trade union recorded in the Javanese cigarette industry was the Kudus Cigarette Workers’ Union (Serikat Buruh Rokok Kudus), established in 1946, but even this organization was ineffective. Firstly, the war and revolution cut off most of the cloves supply, which meant a drop in production; secondly, with the suppression of the Madiun communist revolt in 1948, the SBRK disappeared. 120 ratna saptari change did not in the least mean the decline of sub-contracting arrangements, since equally complex links of putting-out and home-working were retained. Factories themselves were recruiting labor to work not only in the factories, but also in the homes and workshops. With the need to tighten production costs, many of them also bought prepared tobacco and clove mixtures, rather than produce this material themselves. Boxes and packages were in many cases also produced in the small workshops. The brief period examined here, i.e. the 1920s and 1930s, is significant because we can observe a period of great fluctuation—both a peak and an ebb—in the cigarette industry. This period differs from the period of the 1970s onwards, when machine-made cigarettes started to be introduced; nevertheless centralization of production and work fragmentation were to a certain degree already realized. Although the van der Reijden report shows that different forms of production emerged at the same time, from cottage industry to medium and large enterprises, the writing of cigarette history has been associated with the history of the great cigarette pioneers, who had to a certain degree accumulated some capital, either through tobacco trading, or through other forms of trade. The general shift from factory to workshop occurred with the economic crisis and heavy excise taxes; former employees of these factories would set up the workshops. This change differs greatly from the pattern of industrial development in advanced industrial countries. A general pattern of industrial production of the cigarette industry is also quite difficult to find, considering the regional variation within Java itself. This variation is to a large extent also due to the nature of competition, a competition not only created by class differences, but also ethnic ones. The later development of cigarette production in East Java was possibly the reason for the different strategies of cigarette producers. It was here that workshops were able to obtain the labor-time of their workers and concentrate them in the workshop sites, reflecting both workers’ choices (or lack thereof) and employer’s strategies. In Central Java, workshop owners had more difficulty forcing workers to work in their workshops. The variations in the labor process and the physical separation between different groups of workers with the enterprise owners brings to the surface the problem of control—which involves controlling the hours of work, the quality of work, and the level of wages. Entrepreneurs competed with each other in controlling the labor power of the workers, which also implied that employers tried to prevent workers from running away to labor control in the javanese cigarette industry 121 other employers, reducing their own labor supply. Multi-tiered control strategies were developed, using both direct and indirect methods. To what extent can control in the workshop be regarded as familial and benevolent, as compared to the factories? Examining the changes taking place in the cigarette industry, one finds a highly segregated labor force based on different production sites, clear wage differentials, and different places of origin, which all had gender and class dimensions. Even within each production unit, worker differentiation also existed, so that one cannot make categorizations according to the place of production. What Edwards defined as “simple control,” can be identified in Java at all levels of production; yet, different labor market conditions shaped how eventually production was carried out. Cultural and political constraints for women and children to enter the production process seem to have operated differently, resulting also in a diversity of ways in which workers entered this process. Women’s participation in factory work varied between one place in another. In some areas, factory work was done entirely by men; in others, entirely by women; but in most factories, it was done by a mixture of men and women workers. In some places, children or young workers were frequently used, but in others they are absent. Dur- ing the crisis years, children were known to have been substituted for unemployed adult male workers in many places. This means also that the issue of deskilling, and the meaning of skill reflected by the existence of wage differentials, is relevant, although that issue is not acknowledged in the government reports.

OCCUPATIONS

Selling in the Shadows: Peddlers and Hawkers in Early Modern Europe

Danielle van den Heuvel

Introduction

Street sellers were ubiquitous in early modern Europe. They sold a large array of products, from fresh foodstuffs to fashion accessories, and catered to a great variety of customers, serving not only the urban poor and middling sorts, but also the wealthy in their country houses.1 With shopkeepers and stallholders, street sellers formed the principal actors in pre-industrial retailing. In most towns, the shopkeepers and stallholders formed guilds, which governed large sections of the retail trade. These guilds generally were not open to those who solely sold from the streets and they prohibited street vending to take place without their authorization.2 This resulted in constrained selling opportunities for those involved in ambulant trading, such as hawkers, peddlers and chapmen, who therefore often resorted to operating illegally, or semi-illegally. Strikingly, even though many peddlers and hawkers operated in the margins of legality, little attention has been paid to how this affected them and the economy more widely. Whereas there is an intense and ongoing debate on how regulations and the attitude of local and national governments have had an impact on the position of street vendors in present-day developing economies, historians have largely ignored this issue.3 This is

1 Fernand Braudel, The Wheels of Commerce. Civilization and Capitalism 15th–18th Cen- tury, vol. 2 (2002; original edition 1982), pp. 75–9. 2 Vanessa Harding, “Shops, markets and retailers in London’s Cheapside, c. 1500–1700,” in Bruno Blondé, Peter Stabel, Jon Stobart and Ilya van Damme (eds), Buyers and Sellers. Retail Circuits and Practices in Medieval and Early Modern Europe (Turnhout, 2006), pp. 155–70, Peter Stabel, “From the Market to the Shop. Retail and Urban Space in Late Medieval Bruges,” in Bruno Blondé et al., Buyers and Sellers, pp. 79–108, Danielle van den Heuvel, “Partners in Marriage and in Business? Guilds and the Family Economy in the Urban Food Markets in the Dutch Republic,” Continuity and Change, 23, 2 (2008), pp. 217–36. 3 An exception is Anne Montenach, “ ‘Schattenarbeiterinnen’. Frauen im Lebensmit- telkleinhandel im Lyon des 17. Jahrhunderts: Ressourcen und Strategien,” L’Homme. Euro- pean Journal of Feminist History, 2 (2006), pp. 15–36, and Anne Montenach, Espaces et pratiques du commerce alimentaire à Lyon au XVIIe siècle. L’économie du quotidien (Grenoble, 126 danielle van den heuvel especially surprizing, as there seem to be great similarities between people who sell in the street in current-day Africa, Latin America and Asia, and those who peddled their wares in Europe some 300 years ago.4 Like today, many of the street vendors in early modern Europe were people, such as migrants, women, and the poor, who struggled to make a living. Moreover, both in contemporary societies and in pre-industrial Europe, street vendors are subject to intense scrutiny from the authorities and are constantly targeted by the governing bodies of stationary retailers. This chapter addresses the questions of why relatively little attention has been paid to informal street vending in pre-industrial Europe and what we could gain from greater insight into this issue. I will identify the main issues in debates on peddling and hawking in contemporary contexts, and then move on to address the similarities and differences between informal street trading in the pre-industrial and contemporary world. I will do so by combining new evidence from the Northern Netherlands with findings previously reported by historians of peddling, retailing, the urban poor, migration and women’s work. The goal of this chapter is not to give an extensive summary of all that has been uncovered on street vending by historians in the past, but rather to give a comprehensive overview that provides us with direction on how to proceed in studying street vending in a wider context of the position of street vendors within the early modern economy. However, before investigating possible overlaps between street vending in an early modern and in a 21st-century context, I will start by giving an overview of how historical street vending has been studied in the past.

The historiography of early modern peddlers and hawkers

Although there has been remarkably little attention in historical studies on the question of how informality impacted upon street vendors in pre- industrial Europe, street selling has been studied by a number of promi-

2010). See also the remarks by Sheilagh Ogilvie on informal street selling in pre-industrial Germany, Sheilagh Ogilvie, A Bitter Living. Women, Markets, and Social Capital in Early Modern Germany (Oxford, 2003), pp. 302–3. 4 Hernando de Soto earlier pointed to similarities between informal workers in early modern Europe and twentieth-century Peru. Hernando de Soto, The Other Path. The Invis- ible Revolution in the Third World (New York, 1989), pp. 212–3. An inspiring work in which contemporary and early modern practices of selling are compared is Laurence Fontaine (ed.) Second-hand Circulations from the Sixteenth Century to the Present (Oxford, 2008). peddlers and hawkers in early modern europe 127 nent scholars since the last decades of the twentieth century. In the 1970s it was Fernand Braudel who recognized the importance of street vendors in early modern distribution networks. In his seminal work The Wheels of Commerce Braudel acknowledged that whilst peddlers were usually poor itinerant merchants, the sheer numbers of peddlers and the areas they covered meant that they “stimulated and maintained trade, and spread it over a distance.”5 Braudel also pointed out the crucial influence peddlers had on the distribution of certain consumer goods, such as Bohemian glassware, almanacs and popular literature.6 Perhaps not surprizingly, it is precisely in the area of book history that we find the importance of itinerant vending most readily reflected. A pioneering work in this respect is Margaret Spufford’s Small Books and Pleasant Histories, in which Spufford devoted a full chapter to the chapmen who peddled popular literature.7 The exploration of the activities of petty chapmen in books and pamphlets who travelled around England was extended in 1984 when Spufford published The Great Reclothing of Rural England.8 In this work, Spufford took her investigation of the identities and activities of itinerant traders in pre-industrial England even further. Central to this work was the question of how novel consumer goods, especially textiles, clothing and accessories, had spread through the country, and how they had ended up in the houses of ordinary workers in particular. By analysing the movements and activities of itinerant traders in rural England over a longer period of time, Spufford convincingly showed that it was thanks to various peddlers and hawkers that cheap goods were distributed to rural areas. Whilst Braudel’s assumptions on the importance of itinerant small-scale trading had been largely based on scattered evidence, Spufford’s work for the first time showed that it was indeed possible to systematically study the people whose existence had been largely on the margins of the economy, and thereby offered an inspiration for several historians who would follow in her footsteps.9

5 Braudel, Wheels of Commerce, p. 76. 6 Ibidem, pp. 76–7. 7 Margaret Spufford, Small Books and Pleasant Histories: Popular Fiction and its Reader- ship in Seventeenth-century England (Cambridge, 1981), pp. 111–28. 8 Margaret Spufford, The Great Reclothing of Rural England: Petty Chapmen and their Wares in the Seventeenth Century (London, 1984). 9 Spufford’s work on chapmen in books was for instance followed by Margaret Hunt, “Hawkers, Bawlers and Mercuries: Women and the London Press in the Early Enlightenment,” in Margaret Hunt et al. (eds), Women and the Enlightenment (New York, 1984), pp. 48–68, by Laurence Fontaine’s History of Pedlars in Europe (Cambridge, 1996) and Jeroen Salman, “Peddling in the past. Dutch itinerant bookselling in a European perspective,” 128 danielle van den heuvel

A characteristic of peddlers that Braudel signalled in his work on early modern commerce additional to their decisive role in the distribution of specific consumer goods to rural areas was that many of them were seasonal migrants.10 Indeed the second strand of history that has devoted substantial attention to peddlers and other itinerant traders is that of migration history.11 In 1984, the year that Spufford’s Great Reclothing appeared, Jan Lucassen’s book on seasonal migrants in early modern Northwest Europe Naar de kusten van de Noordzee was published.12 Lucassen was the first to systematically analyse one of the most important migration systems of early modern Europe: that of migrant laborers who moved in large numbers to the coasts of the North Sea, and especially to the Dutch provinces of Holland and Friesland. Among these migrant workers were also itinerant traders. In his book, Lucassen exposes the character and functioning of the networks of itinerant traders who visited the North Sea region, and compares these migrants to other migrant laborers operating in the same area, such as grass mowers and peat diggers. His findings show that whilst the migrant laborers who were active in commerce did share some characteristics with the other migrant laborers, such as their agrarian background, their regions of origin, and, to some extent, the seasonal- ity of their activities, there were also some striking differences. Lucassen found that in contrast to migrant laborers working in other occupations, those active as itinerant traders often belonged to family networks in the same occupation, had greater financial means, and, while they occasionally visited the coastal regions of the Dutch Republic, most travelled to other areas, such as Denmark, the eastern and central parts of present-day Germany, and France.13

Publishing history, 53 (2003), pp. 5–21. Spufford’s work on sellers of textiles found followers in Beverly Lemire, Dress, Culture and Commerce: The English Clothing Trade before the Factory 1660–1800 (Oxford, 1991), and Harald Deceulaer, “Second-hand Dealers in the Early Modern Low Countries: Institutions, Markets and Practices,” in Laurence Fontaine Second-hand Circulations, pp. 13–42. 10 Braudel, Wheels of Commerce, pp. 78–9. 11 Despite the attention of historians of migration for peddling they have largely ignored the (sales) activities of travelling groups. See: Leo Lucassen, “A Blind Spot: Migratory and Travelling Groups in Western European Historiography,” International Review of Social History, 38 (1993), pp. 209–35. 12 Jan Lucassen, Naar de kusten van de Noordzee. Trekarbeid in Europees perspectief, 1600–1800 (Gouda, 1984). 13 Lucassen, Naar de kusten, pp. 111–7. His work inspired (amongst others) Marlou Schrover, “Women and long-distance trade migration in the nineteenth-century Netherlands,” in Pamela Sharpe (ed.), Women, Gender and Labour Migration (London, 2001), pp. 85–107; Hannelore Oberpfennig, Migration und Fernhandel im Tödden-System. Wanderhändler peddlers and hawkers in early modern europe 129

A final characteristic of early modern itinerant traders that was high- lighted by Braudel was their marginal position in society, expressed both in their income levels and in their activities, which were often on the fringes of legality. Interestingly, whilst most scholars who have written about peddlers and hawkers in pre-industrial Europe address issues regarding licensing, and the often strenuous relationship between peddlers and stationary retailers, for a long time their inferior status was left relatively unexplored. In the mid-1990s, this changed with Laurence Fontaine’s History of Pedlars in Europe.14 Although Fontaine was not the first to question the supposedly marginal position of itinerant traders, her work was the first that focussed on testing that assumption.15 Fontaine studied the role of itinerant traders, especially those selling books, in Central Europe. As such, her work overlapped with both Spufford’s and Lucassen’s work on this topic, and indeed Fontaine’s very rich findings largely reflect the conclusions for Northwestern Europe as put forward by the previously mentioned authors. According to Fontaine, the itinerant trade of early modern Europe was not marginal but a “multifaceted activity and a vital phenomenon in past communities.”16 Fontaine showed that, as was demonstrated earlier for Northern Europe by Lucassen, the itinerant trade was also highly organized in Central Europe, with long-established relationships between traders, customers, and suppliers, as well as between itinerant traders and their landlords.17 In addition, Fontaine made the

aus dem nördlichen Münsterland in mittleren und nördlichen Europa (Osnabrück, 1996); Wilfried Reininghaus, Wanderhandel in Europa. Beiträge zur wissenschaftlichen Tagung in Ibbenbüren, Mettingen, Recke und Hopsten van 9–11 Oktober 1992 (Dortmund 1993). 14 An earlier version of this book was published in French in 1993: Histoire du colportage in Europe (Paris, 1993). See Spufford, Reclothing, pp. 6–13. 15 Olwen Hufton had earlier identified this issue. Olwen Hufton, The Poor of Eighteenth- century France (Oxford, 1974). 16 Fontaine, Pedlars, p. 202. Interestingly, the issue of economic and social marginality is much less of an issue in the work of Lucassen. Instead he addresses the perception of itinerant traders as reflected in the Dutch historiography of the early 20th century, which assumes that the itinerant traders, especially the Germans, were highly successful businessmen, as (some of) their offspring were the founders of large department stores that sprung up in the nineteenth century. Later, the view that especially the German peddlers were such successful entrepreneurs was disputed by Marlou Schrover and Hannelore Ober- pfennig, who both showed that the majority of the German peddlers were of more modest means and that only a small group founded successful retailers “dynasties.” Lucassen, Naar de kusten, pp. 116–7; Marlou Schrover, Een kolonie van Duitsers. Groepsvorming onder Duitse immigranten in Utrecht in de negentiende eeuw (Amsterdam, 2002); Oberpfennig, Migration. 17 Fontaine’s work was followed by, amongst others, Anne Montenach: Montenach, Espaces. 130 danielle van den heuvel point that those sources that seem to contain most information about peddlers, such as legal and police records, only revealed the exceptional and marginal, and in order not be misled on the nature of ambulant trading, thorough and imaginative methods are required to reveal its true character.18 Typical for the three principal strands of history that have dealt with itinerant traders in the past—represented by their pioneers Spufford, Lucassen and Fontaine—is that they looked mostly at rural ambulant traders, thereby largely ignoring the role of itinerant vending in urban areas. As such, they reflect the assumption expressed by Braudel that it was precisely in the rural areas of early modern Europe that street vendors flourished.19 The role of ambulant traders in urban contexts, therefore, remains largely obscure.20 Coincidentally, whilst the history of urban retailing underwent an impressive boom in the last two decades, most of these works focus on shops, thus in this branch of history urban street trade remains in the shadows. There are a few notable exceptions in which both stationary and ambulatory retailing are studied, such as the 1989 work of Hoh-Cheung and Lorna Mui on shopkeeping in England, and a number of case studies of book selling and the second-hand clothes trade.21 Never- theless, recent work suggests that the eyes of retail historians seem to be slowly opening to the issue of ambulant trading in early modern distribution networks, and various studies that have appeared over the last years integrated urban stationary and ambulant trade in their analysis.22

18 Fontaine, Pedlars, p. 203. 19 Braudel, Wheels of Commerce, pp. 75–80. 20 An exception is the attention for the depiction of urban street vendors in prints. For example: Sean Shesgreen, The Criers and Hawkers of London: Engravings and Drawings by Marcellus Laroon (Stanford, 1990). Vincent Milliot, Les Cris de Paris ou Le peuple travesti: Les représentations des petits metiers parisiens (XVIe–XVIIIe siècles) (Paris, 1995). 21 Hoh-Cheung Mui and Lorna H. Mui, Shops and Shopkeeping in Eighteenth-Century England; Spufford, Reclothing; Deceulaer, “Second-hand;” Salman, “Peddling;” Bibi Panhuy- sen, Maatwerk. Kleermakers, naaisters, oudkleerkopers en de gilden (Amsterdam, 2000). 22 For example Salman, “Peddling;” Ilja Van Damme, Verleiden en verkopen. Antwerpse kleinhandelaars en hun klanten in tijden van crisis (ca. 1648–ca. 1748) (Amsterdam, 2007), Danielle van den Heuvel, Women and Entrepreneurship. Female Traders in the Northern Netherlands, c. 1580–1815 (Amsterdam, 2007), Sheilagh Ogilvie, Janine Maegraith and Markus Küpker, “Krämer und ihre Waren im ländlichen Württemberg zwischen 1600 und 1740,” Zeitschrift für Agrargeschichte und Agrarsoziologie, 59, 2 (2011), and various chapters in Bruno Blondé et al., Buyers and Sellers. peddlers and hawkers in early modern europe 131

One area, which does have a long tradition of studying aspects to street vending in pre-industrial European towns, is women’s history.23 As small- scale retailing was very often practised by women, ambulant trading in studies of early modern women’s work traditionally plays an important role. In contrast to the works that specifically focus on long-distance itinerant trading, these works mostly study street vending in urban areas, practised by women who were residents in the town or had travelled to the town from the surrounding rural areas.24 Similarly, we find hawkers appearing in studies on early modern Europe’s urban poor, many of which devote at least a small section to street vending.25 In contrast to the studies on long-distance itinerant vendors, ambulant traders are generally not the focus of the works on women and the poor, but rather an outcome of an identification of what economic activities these groups engaged in. Interestingly, the studies on women and the poor that include examples of street vending raise issues that, as we will also see later in this chapter, show considerable overlap with those raised by development economists: the problem of entry barriers, illegality and the consequent sufferings of those involved in informal street trade. A further striking difference with the historiography of long-distance peddlers is that, especially in the history of women, the position of street vendors in the economy is seen as rather marginal. It is generally claimed that female hawkers were suppressed individuals, working in trades that were hardly profitable, as they had no other option to earn a living.26

23 This started as early as 1919 with Alice Clark’s seminal Working Life in which she dedicated a special section to street and market trading. Alice Clark, Working Life of Women in the Seventeenth Century (London, 1919). 24 For example: Maryanne Kowaleski, “Women’s Work in a Market Town. Exeter in the Late Fourteenth Century,” in B.A. Hanawalt (ed.), Women and Work in Preindustrial Europe (Bloomington, 1986), pp. 145–66, at 156, Merry Wiesner Wood, “Paltry Peddlers or Essential Merchants? Women in the Distributive Trades in Early Modern Germany,” Sixteenth Cen- tury Journal, 12 (1981), pp. 3–14. 25 Hufton, Poor; Ingrid van der Vlis, Leven in armoede. Delftse bedeelden in de zeventiende eeuw (Amsterdam, 2001), Hilde van Wijngaarden, Zorg voor de kost. Armenzorg, arbeid en onderlinge hulp in Zwolle 1650–1700 (Amsterdam, 2000); also studies on crime often include female street vendors: Paul Griffiths, Lost Londons. Change, crime and control in the capital city 1550–1660 (Cambridge, 2008), pp. 123–34; Lotte van de Pol, Het Amsterdams hoerdom. Prostitutie in de zeventiende en achttiende eeuw (Amsterdam, 1996), pp. 103–4. See also the remarks by Lucassen, “Blind spot,” 209–10. 26 Andrzej Karpinski, “The Woman on the Marketplace. The Scale of Feminization of Retail Trade in Polish Towns in the Second Half of the 16th and in the 17th century,” in Simonetta Chiaviocchi (ed.), La donna nell’economia, secc XIII–XVIII (Florence, 1990), pp. 283–92, at 292; Sheilagh Ogilvie, A Bitter Living. Women, Markets, and Social Capital in Early Modern Germany (Oxford, 2003), p. 169; Jannie Stegeman, “Scheveningse visverkoopsters 132 danielle van den heuvel

As we have seen, the consensus on long-distance itinerant trading is rather different and many scholars follow Fontaine in her judgment that peddlers in early modern Europe were not the marginal figures they were long claimed to be.27 This difference in the appreciation of the status of street vendors can be ascribed to the different contexts the historians of (rural) peddlers and (urban) hawkers were studying, but it is also possible that it may partly be explained by the specific focus of the historians. Whilst Fontaine aimed to expose the activities of rural itinerant vendors and prove that their existence was far from marginal, many historians of (urban) women aimed to find out why women often ended up in lesser- paid jobs with low status. In addition, we must also recognise that the lack of systematic analyses of especially the position of urban street vendors makes that we currently have too little information to accurately assess how shadow work affected people in urban and rural contexts. Despite the shortfalls in the historiography of pre-industrial street vending as pointed out here, we must commend the work of the historians who have in the past worked on this topic. Perhaps even more than for the present, identifying street traders and their activities is enormously difficult for the pre-industrial period. As in the case of scholars who study modern developing economies, we often lack official records and statistics which record informal street traders. In addition, we are further hampered by the fact that the historical figures cannot be cross-examined like the subjects of anthropologists and development economists.28 The question, however, is whether we can use the insights from studies on present-day ambulant trading to enhance our understanding of street trading in historical societies such as that of early modern Europe. In order to answer this question I will now move to identifying the key issues in the debates on contemporary peddlers and hawkers.

ca. 1600–1900,” Holland, 21 (1991), pp. 38–53. An exception is Merry Wiesner, who argued that despite the small volume of their individual trades they were crucial, resembling the argument that was made by Braudel and by Fontaine regarding rural itinerant traders. Wiesner Wood, “Paltry Peddlers.” 27 Van Damme, Verleiden, p. 99. 28 See earlier remarks on the invisibility of street vending in the historiography by Susanne Schötz, Handelsfrauen in Leipzig. Zur Geschichte van Arbeit und Geslecht in der Neuzeit (Cologne, 2004), pp. 195–6; Van Damme, Verleiden, pp. 98–9. peddlers and hawkers in early modern europe 133

Street selling in modern developing economies

In contrast to historical studies, in contemporary contexts street selling is very widely studied. Ever since the 1970s when Keith Hart published his seminal article on informal income opportunities in Ghana, the role of street vending in the overall economy of developing countries has been intensely debated by academics from different backgrounds, ranging from social anthropology to development economics and economic geography.29 Large institutions such as the World Bank and the International Labor Organization (ILO) have also been studying street sellers as part of their programmes to improve economic conditions in developing countries since the 1970s. Despite their origination in different academic disciplines and their focus on different geographical areas, most studies on street vendors deal with one central issue: the role of street vending in the enhancement of economic conditions for the poorer inhabitants of large metropolis and in economic development more widely. The early social-anthropological studies on street vendors in particular originated from studies that were aimed at increasing the understanding of the economic situation of rural migrants who moved to the cities of developing countries, such as Accra in Ghana, and Lima in Peru, in the 1960s and 1970s.30 As the numbers of rural-urban migrants were exceptionally high, most of these (often) colonial cities were unable to absorb them into their infrastructures, forcing these migrants to live in shanty towns on the outskirts of the city.31 Not only was there a shortage of housing, also the local economy had difficulties absorbing these people, resulting in a lack of jobs for those who had newly arrived from the countryside. One of the most common ways of earning a living for the new inhabitants of the ever-growing mega-cities was street vending. With other “invented” jobs, such as working in waste collection and in transport, street selling was part of the makeshift economy that characterised most rural-urban migrants’ struggle for survival. The early studies of anthropologists working on the explosive growth of

29 Earlier, Clifford Geertz had acknowledged the importance of understanding the position of small traders in an economy for economic development. His work however, does not so much focus on itinerant traders such as hawkers and peddlers, but rather on the overall system of bazaar trade. Clifford Geertz, Peddlers and Princes (Chicago, 1963). 30 Keith Hart, “Informal Income Opportunities and Urban Employment in Ghana,” Journal of modern African studies, 11, 1 (1973), pp. 61–89; De Soto, Other Path. 31 According to De Soto this was caused by hostility of urban dwellers against rural migrants. De Soto, Other Path, pp. 10–1. 134 danielle van den heuvel these cities confirmed that similar patterns could be found in developing countries across the globe from Africa, to Latin America and Asia.32 Over the last four decades the debates on the informal economy, and the role of street vending within that, have gradually changed. To a certain extent, this is simply due to a greater availability of data and an increasing understanding of the phenomenon, but it also has to do with the changing world in which we live and the coinciding rise of different world views.33 This is also apparent in the way scholars view street vending. Based on the extensive body of literature on ambulant trading in the informal economy spanning almost half a century, what can we say about the characteristics of street vending and the actors involved in the contemporary contexts of developing economies? Firstly, if we look at the people involved in street vending, we find that most scholars would identify street vendors as relatively poor urban dwellers who are often, but not always, new to the cities they live in.34 A further characteristic of the street vendors is their gender. Whilst we find both men and women engaged in street vending, in many cases the majority of the street traders are indeed women. For example, in 1999 in Benin, the Philippines, Burkina Faso and Mali, the share of women among informal traders ranged from 66% to over 90%.35 Secondly, when we look at what their activities entail, we discover that enterprises of street vendors are often small family-based operations.36 Both in the products street vendors sell and in the way they

32 ILO, Incomes, Employment and Equality in Kenya (Geneva, 1972); De Soto, Other Path; T.G. McGee and Y.M. Yeung, Hawkers in Southeast Asian Cities. Planning for the Bazaar Economy (Ottawa, 1977). 33 An interesting exploration of these changes and how they came about is Keith Hart’s paper “Bureaucratic Form and the Informal Economy,” in Basudeb Guha-Khasnobis, Ravi Kanbur, and Elinor Ostrom (eds), Linking the Formal and informal Economy. Concepts and Policies (Oxford, 2006), pp. 21–35. For recent and more positive views on rural-urban migration in developing economies see: E.L. Glaeser, Triumph of the City: How Our Great- est Invention Makes Us Richer, Smarter, Greener, Healthier and Happier (London, 2011), and D. Saunders, Arrival City. How the Largest Migration in History is Reshaping Our World (London, 2011). 34 See McGee and Yeung, Hawkers, pp. 16–7; Narumol Nirathon, Fighting Poverty from the Street. A Survey of Street Vendors in Bangkok (Geneva, 2006), p. 17; Sally Christine Roever, “Negotiating Formality: Informal Sector, Market, and State in Peru,” (unpublished PhD dissertation, University of California at Berkeley, 2005), pp. 37–9. 35 Jacques Charmes, Gender and Informal Sector. Contribution to the World’s Women 2000, Trends and Statistics (New York, 1999). In some countries where cultural norms restrict women’s role in public life their shares are lower than 10 percent. Website WIEGO (Women in Informal Employment Globalizing and Organizing): [accessed 30 October 2011]. 36 McGee and Yeung, Hawking, p. 17; Roever, “Negotiating,” p. 35. peddlers and hawkers in early modern europe 135 operate, interesting gender differences can be observed. Although men are more likely to sell from push-carts and bicycles, women are more likely to only have a basket or a piece of cloth which they use to carry and display their products. In addition, men are more likely to sell durable products, and women are more likely to sell perishable goods.37 Finally, most street traders operate in the shadows of the official economy, are unlicensed and pay no, or very little taxes.38 In sum, the typical street vendor in a modern developing country is a woman of limited means, often a new- comer to the city, who sells perishable goods from street corners, or on the edges of formal markets, sometimes assisted by family members such as her children. When we then move on to the question of how social scientists have explained the rise and persistence of informal street selling, we find they put forward a number of explanations that are strongly linked. As has already been stated, the growth of street vending in modern developing countries was initially very much linked to the process of rapid urbanization and large waves of internal migration. In addition, according to many scholars, it was also caused by processes of modernization, industrialization and bureaucratization. Whereas bureaucratization and subsequent increased regulation are mostly seen as a negative feature forcing people out of formal jobs and enterprises and into activities such as street vending, in the discussions on the informal economy, modernization has a dual face.39 On the one hand, it is argued that modernization led to increased regulation and therefore to more limited opportunities for the urban poor, hence resulting in high levels of informal workers. On the other hand, the process of modernization in itself was long regarded positively, as it would eventually lead to resolving the problem of informal workers as it was assumed that the stream of new migrants arriving to the city would simply dry up.40 However, forty years on it has turned out that the phenomenon

37 Monique Cohen, Women Street Vendors: The Road to Recognition (New York, 2000), p. 3; Martha Chen, “Rethinking the Informal Economy: Linkages with the Formal Economy and the Formal Regulatory Environment,” in Guha-Khasnobis, Kanbur, and Ostrom, Linking the Formal and Informal Economy, pp. 75–92, at 79. 38 De Soto, Other Path; Chen, “Rethinking,” p. 81, Ray Bromley, “Street Vending and Public Policy: A Global Review,” International Journal of Sociology and Social Policy, 20, 1/2 (2002), pp. 1–28, at 8–9. 39 The strongest advocate of the idea that regulation had a negative effect on the opportunities of the poor and indeed created an informal economy is Hernando de Soto. De Soto, Other Path. 40 L. Benería and M. Floro, “Distribution, Gender, and Labour Market Informalization: A Conceptual Framework with a Focus on Homeworkers,” in N. Kudva and L. Benería 136 danielle van den heuvel of informal street trading has not disappeared, and that it is likely to remain a feature of most countries in some form or other, despite changes in regulatory regimes, and the decline in rural-urban migration.41 Indeed it is increasingly recognized that the large numbers of street vendors in, for instance Asian cities, are (at least partly) caused by the great demand for prepared food to be readily available on the streets as many people lack the time or space to prepare a meal at home. Thus, it is not solely due to an overabundant supply of new informal workers.42 It is furthermore also acknowledged that the extent of informal street trading depends on the functioning of the labor market as a whole and that in times of economic decline, and related increased pressures on the labor market, the number of street traders increases.43 Apart from the fact that scholars are gradually recognising that changes in informal street trading are not the result of a linear process, they have also been acknowledging the role of micro-economic factors in explaining why street vending exists and per- sists in so many developing economies. In addition to the impact of regulation, and limited opportunities in other segments of the economy, the street sellers themselves often stress that street vending is an occupation that suits them and their families well. A commonly mentioned reason for being a street vendor is the flexibility it provides. Many female street vendors indeed claim that being a street trader allows them to care for their families and to earn an income at the same time.44 A question that has proved even more difficult to answer than why street selling is such an important feature in the cities of many developing economies, is the question what effect informal street vending has on the people involved and on the wider economy. As one can imagine, the debates on this question are very intense and views on this issue range

(eds), Rethinking Informalization: Poverty, Precarious Jobs and Social Protection (New York, 2005), pp. 9–27. 41 Chen, “Rethinking.” John C. Cross, “Street Vendors, Modernity and Postmodernity: Conflict and Compromise in the Global Economy,” International Journal of Sociology and Social Policy, 20, 1/2 (2000), pp. 29–51. See also Leo Lucassen, “To Move Or Not to Move. A Global Review of Migration to the City Since the 18th century,” unpublished paper, p. 7, for the the newly developed argument that in parts of Sub-Saharan Africa and Asia rural- urban migration was not one-directional but rather a “rural-urban continuum”. 42 T.T.-F. Lam, “Food for the City: The Role of the Informal Sector,” Geojournal, Supple- mentary Issue, 4 (1982), pp. 49–59, at 57–8. 43 Nirathon, Fighting Poverty, pp. 15–6; Zoe Elena Horn, “No Cushion to Fall Back on. The Global Economic Crisis and Informal workers,” at , pp. 12–14. 44 Cohen, Women Street Vendors, p. 2; Nirathon, Fighting Poverty, p. 17. peddlers and hawkers in early modern europe 137 from very positive (regarding it as a way for marginal groups to make a living and eventually to enter the formal sector) to very negative (seeing it as a dead end, jobs with no protection, very little income and opportunities). Over the last few years, scholars have started to recognize that much of the rather black-and-white thinking is not very helpful in understanding the phenomenon and acknowledge that despite the difficulties informality can result in for certain groups, working as an informal trader can mean some form of economic enhancement. Nevertheless, it is clear that this is a question that social scientists are increasingly struggling to answer, as both optimistic and pessimistic views on informality as well as its effects have not come true.45 Indeed, we find that development economists more and more stress the historical nature and therewith the persistence of the phenomenon, especially in urban contexts. However, the question is to what extent are the modern street trader and the historical street seller similar? Let us investigate this by looking at early modern Europe.

Identities of early modern street traders

As already became clear from the overview of the historiography on early modern peddlers and hawkers, the identities of ambulant traders in pre- industrial Europe differ according to the discipline of history. The impression from the most dominant line of works on ambulant trading is that most itinerant traders were male migrants with a rural background. These male peddlers migrated to places other than their home regions to sell their wares, but they would generally not settle in those localities. For large parts of the year, they were more or less continuously on the move. If women were involved in ambulant trading, they would do so as family members of male peddlers, such as Westerwald pottery vendors who travelled throughout Northwestern Europe, as part of Gypsy groups, or as hawkers who lived and operated locally.46 The latter we can derive mostly from other fields of history such as women’s history, which points to the significant levels of activity of women in local hawking. Interestingly, there is generally no mention of the origins of female hawkers, which probably

45 Chen, “Rethinking,” p. 75. 46 On Westerwalder pottery vendors in the Netherlands see Schrover, Kolonie, 264. On female gypsies see: Leo Lucassen, Wim Willems and Annemarie Cottaar (eds), Gypsies and Other Itinerant Groups. A Socio-historical Approach (Basingstoke, 1998), especially the chapter by Cottaar. On local women as hawkers in early modern cities see for instance: Merry E. Wiesner, Working Women in Renaissance Germany (New Jersey, 1986), p. 119. 138 danielle van den heuvel indicates that these women were local (or at least that that was what the historians who wrote about them assumed).47 This dichotomy between male long-distance ambulant traders and female local ambulant traders is very striking as it not only seems to bear very little resemblance to the world of the street vendor in the 20th- and 21st-century developing world, but also because it presents an image that is rather neatly organized and static. I am not claiming here that these two very distinct types of ambulant traders are not accurate reflections of the world of early modern European street vending. However, since they originated from two different fields in history, and because there has not been a holistic study into street vending in the pre-industrial past, it is rather risky to assume that these two types fully capture peddling and hawking in 17th- and 18th-century Europe. Indeed, when one looks more closely at studies on the urban economies of early modern Europe, one notices much more variation in the identities of street vendors than a first glance at the historiography would suggest. It is true that many of the accounts of street sellers operating in the streets of towns and cities concern women, but we find women from a great variety of backgrounds, different marital status, and age. In sixteenth-century Munich, we encounter widows of day laborers, wives of semi-skilled laborers, and women married to men employed in major crafts as street vendors.48 Many of the male, and some of the female street vendors were of Jewish origin, especially in the urban rag trades, but also among those who peddled through town and country.49 In addition, we also find children hawking goods in the street.50 For instance, in France,

47 Also, it is not always made explicit what the gender of migrants is, and the historians seem to be assuming that it is often male. For instance: Laurence Fontaine, “The Exchange of Second-hand Goods between Survival and ‘Business’ in Eighteenth-century Paris,” in Laurence Fontaine (ed.), Second-hand Circulations from the Sixteenth Century to the Present (Oxford, 2008), pp. 97–114, at 103. Rommes does provide some insight into the origin of newcomers who settled in the Dutch city of Utrecht and set up a retail trade in the eighteenth century. Although not always explicitly stated it seems, however, that this mostly concerned stationary retailers. Ronald Rommes, Oost, west, thuis best? Driehonderd jaar migratie en migranten in de stad Utrecht (begin 16e-begin 19e eeuw) (Amsterdam, 1998), pp. 161–2. 48 Wiesner, Renaissance, p. 138. 49 Panhuysen, Maatwerk, pp. 249–54, 268–72; Cf. the identities of those arrested for street selling without permission in the Dutch towns of Arnhem and The Hague in the eighteenth century: Gelders Archief, 2001, inv. nos. 1457 and 1464, Gemeentearchief Den Haag, Archief Ambachtsgilden en Bussen, inv. nos. 133–8. 50 Hufton, Poor, pp. 110–1; Wiesner, Renaissance, pp. 113–4, 121, 138; J.C. Streng, Vrijheid, gelijkheid, broederschap en gezelligheid. Het Zwolse Sint Nicolaasgilde (Hilversum, 2001), pp. 75, 76, 89; Van Damme, Verleiden, p. 85. peddlers and hawkers in early modern europe 139 children from poor rural farmers were sent out as peddlers of pins and needles, and in Strasbourg, urban children also hawked in the streets.51 Moreover, it appears that greater overlaps between the past and the present migrant peddler can be established than initially assumed. Firstly, as Olwen Hufton stressed in her work on the poor in eighteenth-century France, seasonal migrants could turn into permanent migrants when not enough work was available and in those cases, they would look for work in agriculture, building, or, indeed, as a trader.52 Moreover, in Southern Europe we find non-seasonal migrants moving from regions in France to Spain, where they would settle for periods between two and nine years. Many of those non-seasonal migrants worked in informal jobs in retailing, as lemonade sellers, vendors of fancy breads, and as peddlers of haberdashery.53 According to Hufton, these permanent displacements from home communities became more common during the eighteenth century.54 Such examples closely correspond to the experiences of informal peddlers in the modern developing world. Finally, evidence from Northern Europe shows that all these different groups of ambulant traders sometimes came together and shared one specific trade or marketplace. In 1749, the guild of old clothes sellers in the medium-sized Dutch town of Den Bosch complained about the foreigners and other needy people at the market selling old clothes, as well as about the increasing numbers of door-to-door sellers, which were both locals and people coming from elsewhere.55 Similarly, in the urban fish trade we find that whilst fish was sold in the official fish markets by citizens, locals, or both, female hawkers from surrounding coastal villages offered fish from door to door, a practise that is also reflected in the trade in produce.56 Rather than being destitute inhabitants of large metropolis (either male or female), it is likely that the latter were generally better off than ambulant traders who lived in the cities, or on its outskirts. In sum, mapping the identities of early modern peddlers and hawkers shows that very different groups were active as ambulant traders, each with different patterns of movement. We find both those who were on the move

51 Hufton, Poor, pp. 110–1, Wiesner, Renaissance, p. 125. 52 Hufton, Poor, pp. 73–4. 53 Hufton, Poor, pp. 89–90. 54 Hufton, Poor, pp. 70–1. 55 Panhuysen, Maatwerk, p. 267. 56 Ian Archer, Caroline Barron, and Vanessa Harding (eds) Hugh Alley’s Caveat: The Markets of London in 1598 (London, 1988); Ariadne Schmidt, Overleven na de dood. Wedu- wen in Leiden in de Gouden Eeuw (Amsterdam, 2001), p. 131; Van den Heuvel, Women, p. 113. 140 danielle van den heuvel permanently and those more incidentally, and that despite seeming to be occupying completely separated worlds, all these traders regularly shared the same markets.

Activities of early modern street traders

When establishing the activities of peddlers and hawkers in what goods they sold and how they operated, we find a wide array of undertakings. Because of the great diversity of goods and services offered by street vendors, it is helpful to systemize their activities, which I have tried to do below. This is not a scheme without any fluidity, as that would simply not do justice to the historical complexity of street vending. Firstly, there were the predominantly male itinerant traders who were highly specialized and offered unique products to their customers. Many, but not all, of the seasonal migrant peddlers sold wares that were specific to the regions they originated from. Examples are peddlers from Sauer- land who would sell wooden utensils including wooden toys, the West- erwald traders who sold earthenware and the sellers of straw hats from the Jeker valley.57 These ambulant traders would sometimes also include other goods in their stock, for instance, in times when supply of their principal product was limited, but overall we can regard these traders as specialized dealers offering unique products which could not easily be obtained in the areas they peddled their wares. Equally specialized, but in a slightly different manner, were the chapmen of books, who travelled throughout Europe selling popular literature from door-to-door. Like the Westerwalders and the traders from the Sauerland and Jeker valley, they too provided their customers with highly specialised wares that were difficult to come by via other channels. However, rather than producing the books themselves, they would obtain them from publishers in the cities.58 These specialized traders would travel long-distances, serving urban, but mostly rural areas, and as such seem to bear little resemblance to the rural-urban migrants turned street sellers in current-day developing countries.59

57 Lucassen, Naar de kusten, p. 111; Schrover, Kolonie, pp. 219–35, 255–6. Annemarie Cottaar and Leo Lucassen, “Naar de laatste Parijse mode. Strohoedenmakers uit het Jeker- dal in Nederland 1750–1900,” Studies over de sociaal-economische geschiedenis van Limburg/ Jaarboek van het Sociaal Historisch Centrum voor Limburg, 46 (2001), pp. 45–82. 58 Spufford, Reclothing, pp. 69–83; Salman, “Peddling,” pp. 13–5. 59 Although in the academic literature on ambulant trading in modern developing countries this type of peddling is largely ignored it does exist in a similar form to the early peddlers and hawkers in early modern europe 141

Secondly, other groups of male migrant peddlers sold a large variety of goods, and combined selling with offering a wide range of services.60 Migrant peddlers from the Auvergne for instance, “peddled anything that could possibly be sold, such as pins, needles, laces, combs, lace-bobbins, rabbit skins and mole pelts.”61 Other groups, such as the Teuten and Töd- den also sold a wide variety of products, but in their cases, the choice of products on offer was less random. Most of the items they sold can be classified as homewares: household textiles, such as table linens and napkins, and housewares, such as sifts, pots and pans, formed the principal products in their packs.62 Contrary to the migrant peddlers referred to above, the majority of the products peddled by Teuten, Tödden and Auvergnats were sourced from urban wholesalers and rural producers, rather than home-made.63 The goods and services these traders offered came as a pack- age, and their services were sometimes even more important trademarks than their goods. The Savoyard chimney sweeps are good examples, as their main livelihood consisted of sweeping chimneys, but they additionally also sold small wares such as precious stones and spectacles.64 Other types of “services” offered by these peddlers are various forms of entertainment. Indeed, Hufton claims that many itinerant vendors from the Auvergne would turn to storytelling when trade was bad.65 A third category of early modern street vendors consists of those who sold on the streets on a daily basis, and, if they moved at all, only moved very short distances, generally within one town. These were mostly women who very often sold from more or less designated areas, and tended to sell items that were not for sale through any of the regular channels of stationary retailers, for example exotic fruits, pamphlets, or, at the other

modern European cases described here. In the late 1990s I visited Guatemala where rural men from the highlands would travel in kinship groups to Costa Rica for several months a year to sell their home-produced woodwork to wealthy Costa Ricans and foreign tourists. 60 These groups of street traders seem to resemble those peddlers who work in the streets of European cities after migrating from Asia and Africa, such as sellers of sun- glasses and DVDs in current-day Barcelona, see: Uma Kothari, “Global Peddlers and Local Networks: Migrant Cosmopolitanisms,” Environment and Planning D: Society and Space, 26 (2008), pp. 500–16. 61 Hufton, Poor, p. 83. 62 Cf. the traders from these groups who were fined in The Hague in the eighteenth century. Gemeentearchief Den Haag, Archief Ambachtsgilden en Bussen, inv. nos. 133–8. 63 Oberpfennig, Migration, p. 68. 64 Van Damme, Verleiden, p. 83. Other examples are knife grinders and tinkers, see for example: Moch, Moving, p. 80. 65 Hufton, Poor, p. 84; see also Van Damme, Verleiden, pp. 83–4. 142 danielle van den heuvel end of the spectrum, home-produced goods such as bread and cakes.66 The reason for their choice of products probably lies in the fact that they were forbidden by the local councils, guilds, or both, to sell items that were sold by local shopkeepers.67 For example, in early modern Germany, women who sold imported Mediterranean fruits were specifically forbidden to sell fruit that was grown closer to home.68 When locally operating street vendors would sell products that were also on offer by local stationary retailers, such as vegetables and fish, they were generally obliged to sell at different, and mostly unfavourable, times.69 Lastly, we find people who only hawked occasionally, and when they did, they generally sold in small quantities. Housewives sold small pieces of cloth, small home-produced goods, or cooked foods.70 Young men and women who were arrested for vagrancy were active as hawkers, and others were arrested for stealing and reselling items of clothing.71 Overall, the impression one gets is that in early modern European economies, perhaps even more than in current developing economies, street vendors filled gaps in the market. They sold very specific items, new and exotic wares, or very cheap goods. Although this is sometimes explained by underde- veloped distribution systems, I think it is fair to argue that their choice of products was probably partly due to restrictions on which goods they were allowed to sell, and partly due to their entrepreneurial spirit.

66 Streng, Vrijheid, pp. 75, 95; Wiesner, Renaissance, p. 118; Salman, “Peddling.” 67 Streng, Vrijheid, p. 75. 68 Wiesner, Renaissance, p. 120. 69 Wiesner, Renaissance, pp. 117, 121; Archer, Hugh Alley’s Caveat, p. 6; Betty R. Masters, The Public Markets of the City of London Surveyed by William Leybourn in 1677 (London, 1974), p. 13; Van den Heuvel, Women and Entrepreneurship, pp. 103–4; Regionaal Archief Leiden, Archief van de Gilden, Beurzen en de Rederijkerskamers, inv. no. 1215. 70 Wiesner, Renaissance, p. 177; Streng, Vrijheid, p. 67; Griffiths, Lost Londons, pp. 123–34; Annette de Wit, “Zeemansvrouwen aan het werk. De arbeidsmarktpositie van vrouwen in Maassluis, Schiedam en Ter Heijde (1600–1700),” Tijdschrift voor Sociale en Economische Geschiedenis, 3 (2005), pp. 60–80; 74, Ogilvie, Bitter living, pp. 302–3. 71 Anne Winter, “ ‘Vagrancy’ as an Adaptive Strategy: The Duchy of Brabant, 1767–1776,” International Review of Social History, 49 (2004), pp. 249–77, at 263–4; Lemire, Dress, pp. 145–6; Thera Wijsenbeek-Olthuis, “Gestolen goed. Diefstal van textiel in Den Haag 1600–1800,” Textielhistorische bijdragen, 47 (2007), pp. 42–63, at 47–8. In the case of reselling of stolen goods we also find people who were part of professional gangs who stole textiles and clothing for resale on a more permanent basis. peddlers and hawkers in early modern europe 143

Reasons for entering into street selling in early modern Europe

We have seen that in modern developing economies, entry into street vending depends on a combination of macro- and micro-economical factors. On the one hand, it is caused by the combination of large-scale rural-urban migrations and the existence of dual labor markets. On the other hand, we find that of the various jobs available to newcomers to the mega-cities as well as others who failed to gain access to formal types of work, street selling is mostly favoured by those who appreciate its flexibility. To what extent can such mechanisms (or parts of them) explain the extent and different types of street vending we have observed in early modern Europe? Let us start with the factor that is most commonly mentioned in the academic literature on developing economies: the impact of rural-urban migration. At first sight, the situation in pre-industrial Europe seems similar to that of the present, as many ambulant traders were indeed rural migrants. However, we have also seen that most of the rural peddlers, even when they visited cities, generally did not settle there, and moreover, many worked as ambulant traders only part of the year. Indeed, a closer look reveals that the situation for many of these rural migrant- peddlers seems very different from many street vendors in the 20th- and 21st-century. Although most of them needed the income from peddling and hawking, not all of them did it out of sheer economic desperation. For the German Tödden, peddling was part of their annual cycle of work, as was the case with the itinerant traders from the Auvergne, although the economic circumstances of the latter group were clearly direr.72 This sometimes forced them to abandon their plans to return to their home regions, and become someone who may indeed had closely resembled the street sellers of present-day Africa or Latin America.73 Nevertheless, we can conclude that for substantial groups of long-distance itinerant traders, migration was not the cause but rather a consequence or, perhaps even more accurate, a necessary tool to be able to engage in street vending.74 Does this mean that mass migration to the early modern cities did not lead to higher numbers of street vendors and that there were no rural-urban migrants among the urban street vendors? Interestingly, the

72 Lucassen, Naar de kusten; Fontaine, Pedlars. 73 See Hufton, Poor, pp. 70–1. 74 Lucassen, “Blind Spot,” p. 219. 144 danielle van den heuvel available evidence only sheds oblique light on these issues. Earlier in this chapter, I pointed to the fact that local hawkers were mostly female, and that hardly any information is available on their origins, except for women who came into the cities from the surrounding countryside to sell fresh produce or fish. However, overall we are left in the dark on whether these women were (relative) newcomers or not. From what these women reported to the local authorities when requesting permission to sell, the majority seemed to live in the cities they operated in, and in terms of affluence, ranged from not the very wealthy to the very poor.75 Moreover, whilst the available evidence on urban hawkers gives us very little clues on the role of migration in their economic activities, the historiography of early modern European urban migration also leaves the issue largely unexplored. Street selling is regularly mentioned as an employment opportunity for migrants who had recently arrived in the city, but such activities are hardly investigated in these contexts.76 The question is whether this is simply a blind spot or whether there was no need for migrants to work in street vending and the phenomenon of ambulant trading in early modern cities should be explained differently. In the context of this chapter, it is not possible to answer this question, but it certainly deserves further attention.77 Another very prominent feature in the debates on street vending in modern times is the impact of regulatory regimes, with some authors stressing restrictive policies, forcing especially the poor into informal trading, and others pointing to informal trade being an outcome of weighing up costs and benefits of formal versus informal work. It is clear that also in early modern European street vending regulation mattered, both in restricting access to the labor market, and therefore pushing people towards street selling, and in the decision to operate as an informal rather than a formal street vendor. We see that entry costs into formal street

75 There is a danger to relying on such evidence to establish a social profile of street hawkers, especially regarding their levels of wealth as one can expect a greater willingness from urban authorities to let the very poor peddle some wares in order to make a living. 76 For example Catharina Lis, Social Change and the Laboring Poor. Antwerp, 1770–1860 (New Haven, 1986), p. 60; Jan Lucassen and Rinus Penninx, Nieuwkomers, nakomelingen, Nederlanders. Immigranten in Nederland 1550–1993 (Amsterdam, 1994), p. 55; Leslie Page Moch, Moving Europeans. Migration in Western Europe since 1650 (second edition, Bloom- ington, 2003), p. 98; Erika Kuijpers, Migrantenstad. Immigratie en sociale verhoudingen in 17e-eeuws Amsterdam (Hilversum, 2005), p. 247. See also Lucassen, “Blind Spot.” 77 Issues that require further attention and that might shed more light on the character and dynamics of early modern street vending are migration of the middle-aged and the elderly, and of the existence and functioning of shanty towns in pre-industrial Europe. peddlers and hawkers in early modern europe 145 selling, imposed by guilds and local governments through requiring guild memberships and sales licenses, were a significant factor in the choice to be an informal street vendor.78 For example, those who only engaged in street selling occasionally or part-time, such as a housewife who was caught in the Dutch town of Zwolle selling a small piece of linen without a sales license, the costs of formality simply did not outweigh the benefits.79 Indeed, when the price of a guild membership of one of the Amsterdam retailers’ guilds was drastically lowered after great concerns over the high numbers of informal traders operating in the city in the mid-eighteenth- century, membership numbers soared.80 The evidence on the impact of regulations on general access to work and the impact that they had on the character and existence of street vending is, however, less straightforward. On the one hand, historians are adamant that many rural-urban migrants and other often poor inhabitants of Europe’s cities were forced into small-scale peddling because institutional restrictions meant that access to other occupations were closed to them.81 Indeed, we find that the groups whose entry into crafts and trades were hampered or sometimes fully blocked by citizenship requirements, hefty guild entrance fees and the straightforward exclusion from guilds, such as migrants, Jews and women, seemed active in street selling in disproportionate numbers.82 At the same time, one can find the argument put forward that precisely when institutional barriers imposed by guilds disappeared at the end of the 18th century, peddling would have increased substantially. Indeed, as was earlier stressed by De Soto, some historians of early modern European retailing claim that the end of mercantilism and the introduction of “liberal” 19th-century regulation resulted in extraordinary increases in street vending.83 This assumption is confirmed in an example presented by Braudel in which he cites a Metz magistrate who in 1813 wrote to the general council on trade in Paris that because of the modest price of the Patent license, which replaced

78 Van den Heuvel, Women and Entrepreneurship, p. 166. 79 Streng, Vrijheid, p. 76. 80 Clé Lesger, “De locatie van het Amsterdamse winkelbedrijf in de achttiende eeuw,” Tijdschrift voor Sociale en Economische Geschiedenis, 4 (2007), pp. 35–70, at 60. 81 Fontaine, “Exchange of Second-hand Goods,” p. 103; Panhuysen, Maatwerk, pp. 267, 271; Ogilvie, Bitter Living. 82 Salman, “Peddling”; Ogilvie, Bitter Living, p. 169; Montenach, “Schattenarbeiterin- nen”; Van den Heuvel, Women and Entrepreneurship, pp. 214–5. 83 Van Damme, Verleiden, pp. 97–100; Deceulaer, “Dealing,” p. 192. Compare J.D.H. Harten, “De verzorging van het platteland van de Zeeuwse eilanden in de Franse tijd,” Bulletin historische geografie, III/1 (October 1971), pp. 31–73, at 60. 146 danielle van den heuvel obligatory guild membership, anybody could set himself up as a trader, and that the only way to put a stop to this was to re-establish the guilds.84 The question is, whether these historians in addressing the impact of regulation are dealing with the same mechanisms. If the historians who argue that corporate restrictions were forcing people into street trading are right, then it is rather surprising that the abolishment of such restrictions would have resulted in a sharp increase in the phenomenon. One would instead expect that while some people would have perhaps entered into retailing as it became more accessible, many others would have chosen to work in perhaps more profitable occupations that had been previously closed to them. It is therefore more likely that it was not solely the abolishment of guild regimes that caused the increase, but that at the same time, the lowering of thresholds to formal forms of street selling encouraged many of those who were already active as hawkers on an informal basis to obtain official permission. Nevertheless, and here we arrive at a third factor, it is clear from most of the available evidence that the eighteenth century, and then especially after 1750, was a time in which street vending became much more perceived as a problem throughout Europe, precisely when large parts of Europe were experiencing worsening economic conditions.85 We see complaints about street sellers increasing, as well the numbers of arrests of street vendors. It is also in this period that in many countries, licensing systems for peddlers were introduced and we discover a concentration of changes in policies regarding street vending. Although the impact of economic downturn on the size of the informal sector is still heavily debated, the early modern European evidence seems to suggest that the need for people to engage in small-scale peddling was higher during times of economic difficulty. However, at the same time we need to be aware that economic downturn could have also lead to a greater visibility of peddlers and hawkers in primary sources as a greater urgency to protect sedentary retailers and their businesses would have led to increased legislation, and greater intensities of policing.86 This issue clearly needs further

84 Braudel, Wheels of Commerce, p. 80. 85 Moch, Moving Europeans, p. 98; Hufton, Poor, pp. 70–1; Mui and Mui, Shops, pp. 135–220; Van Damme, Verleiden, p. 55. An exception seems to be early seventeenth-century London where complaints also soared. Griffiths, Lost Londons. 86 James E. Shaw, “Retail, Monopoly, and Privilege: The Dissolution of the Fishmongers’ Guild of Venice, 1599,” Journal of Early Modern History, 6, 4 (2002), pp. 396–427, at 423; Archer, Hugh Alley’s Caveat, p. 35. peddlers and hawkers in early modern europe 147 exploration, but it is nevertheless very clear that for many street vendors, poverty was a reason to engage in hawking.87 Indeed, in their appeals to the authorities many stress how their lives had radically degraded in quality through sickness of household members, absence of a male head of household through war and widowhood, and the like, and that they therefore needed the income from a small trade.88 Moreover, various examples indicate that during tough economic times, and in years of war and general unrest, the numbers of street sellers went up.89 For instance, during the Thirty Years’ War, Strasbourg saw a sharp increase in the number of brandy stands in the city, as well as the number of women requesting to operate them.90 Whilst claims of poverty can be seen as problematic evidence to fully explain which people entered into street vending, and they probably are only one side of the story, the evidence on the increase in street vending during tough times, does suggest the two were linked.91 A fourth factor proposed by social scientists to explain why people end up in street-selling is that it is an attractive alternative to other makeshift jobs. Due to the flexible hours, married women who also have the responsibility of caring for their family, would particularly find street selling appealing.92 The choice for street vending thus is also determined by the life cycle. Recent research on present-day Malaysia shows that many women who work as cleaners in hotels or as factory workers choose to change their job to street selling after marriage. Interestingly, this is not only because they themselves feel that street vending is easier to combine with motherhood, but also as a result of their husbands considering their previously held jobs as not suitable anymore.93 Although we still have very little information on the precise identities of early modern street vendors who operated in urban contexts, the accounts we find on those who were

87 Laurence Fontaine, “Märkte als Chance für die Armen in der Frühen Neuzeit,” Zeitschrift für Agrargeschichte und Agrarsoziologie, 59, 2 (2011), pp. 37–53. See also: Melissa Calaresu, “From the Street to the Stereotype. Urban Space, Travel and the Picturesque in Late Eighteenth-century Naples,” Italian Studies, 62, 2 (2007), pp. 189–203, at 197–201. 88 Shaw, “Retail,” pp. 400–1; Streng, Vrijheid, p. 89; Wiesner, Renaissance, pp. 113–4, 117, 120, 130, 138. 89 Van Damme, Verleiden, p. 55; Mui and Mui, Shops, pp. 135–220; Wiesner, Renaissance, p. 117. 90 Wiesner, Renaissance, p. 130. 91 Compare the remark by Catharina Lis who states that among migrant women arriving in Antwerp at the run of the nineteenth century only the “happy few” would be able to find work as a retailer. Lis, Social Change, p. 60. 92 Nirathon, Fighting Poverty, p. 17. 93 Anja Franck, “ ‘I Am Too Old! Who is Going to Give me a Job.’ Women Hawkers in Teluk Bahang, Penang, Malaysia,” Journal of Workplace Rights, 15, 1 (2011), pp. 111–32. 148 danielle van den heuvel vending in Europe’s urban streets seems to confirm that also in the seventeenth and eighteenth centuries one’s phase in the life cycle mattered in the choice for this economic activity.94 Although we do discover evidence for younger women operating as hawkers, most cases one comes across are of housewives and widows, who often had young children to take care of.95 Moreover, for eighteenth-century Holland it has recently been established that opening a small retail outlet was something that was precisely done by married women with a small but growing family.96 Interest- ingly, as Merry Wiesner pointed out, whilst urban authorities saw street vending as an appropriate job for older women, they were rather negative about young women hawking in the streets as they were seen as “persons who could easily do some other kind of work, like being a maid, but are causing great disorder by selling fruit.”97 The latter is a rather striking resemblance of how cultural norms on what types of jobs were suitable for women in different phases of their lives affected the choice to become a street vendor, in both present-day Asia and early modern Europe.

Concluding remarks

This chapter aimed to provide insight into the question why historians have largely ignored informal retail practices and what could be gained from studying those street vendors that operated in the shadows of the official economy. The results of looking at early modern street vending through the prism provided by insights on street selling in developing economies shows that we cannot explain this by large dissimilarities between contemporary and early modern street vendors and their position in society. Whilst at first sight the historiography seems to reveal that early modern ambulant trading was mostly a seasonal phenomenon, undertaken by rural, mostly male, migrants who peddled their wares through town and

94 Van Damme, Verleiden, p. 83, stresses that many street vendors were involved on incidental and temporary basis rather than on a permanent basis. 95 Streng, Vrijheid, p. 89; Wiesner, Renaissance, pp. 138, 113–4, 130, 117, 120; Rudolf M. Dekker, “Women in Revolt. Popular Protest and its Social Basis in Holland in the Seven- teenth and Eighteenth centuries,” Theory and Society, 16 (1987) pp. 337–62, at 352. 96 Danielle van den Heuvel and Elise van Nederveen Meerkerk, “Huishoudens, werk en consumptieveranderingen in vroegmodern Holland. Het voorbeeld van de koffie-en theeverkopers in achttiende-eeuws Leiden,” Historisch Tijdschrift Holland, 42, 2 (2010), pp. 102–24. 97 Wiesner, Renaissance, p. 120. peddlers and hawkers in early modern europe 149 country, a more diversified image appears when the various strands of history that have been studying ambulant trading in rather isolated contexts are brought together. Resulting from this enlarged view, there emerges at least a partial overlap with contemporary developing economies. We found that no such thing as “the early modern street vendor” existed, and that a great variety of people was involved in ambulant trading, some on a permanent basis, others more occasionally. It also appeared that the various groups of ambulant traders shared selling spaces and commercial circuits and their worlds were hence probably not as separated as one might assume on the basis of the historiography. However, in classifying itinerant vendors according to their identities and activities, it became clear that interesting differences seemed to have existed between male and female itinerant sellers, with the former generally active in long- distance itinerant trading, often on a seasonal basis, and the latter covering shorter distances all year round. The question is whether this should solely be ascribed to their gender, or rather to the type of ambulant trading they were involved in (long vs. short distance, urban vs. rural), as the latter had important consequences for which people could engage in the different types of ambulant trade. Such consequences were mostly related to age and gender, and varied from practical issues such as the weight of the merchandise and the dangers on the road, combined with family care needs, as well as cultural and institutional restrictions. These issues need further investigation. In this context, it also remains to be seen whether we have fully captured the identity of early modern street vendors: many locally operating males may have simply been hidden under a different classification. We have furthermore found that two of the most commonly cited causes for informal street vending in the context of modern developing economies, migration and regulation, also played a large role in the early modern period, but that their precise influence is as yet unclear. Whilst many long-distance peddlers were clearly migrants, it was not necessarily their migratory behavior that caused them to become a street trader. At the same time, for most of the urban hawkers, information on their origins is (still) lacking. With regard to the impact of regulation, the dynamics of restrictive regulatory regimes and the intensity of policing remain unclear, as well as how these impacted the extent of street selling and the choice for those engaging in peddling and hawking to opt for informality. Although most, if not all, historians who have studied early modern ambulant trading in the past seem to recognize the restrictions on street 150 danielle van den heuvel traders by regulatory regimes, and the constraints that resulted from that, only a handful have addressed their implications. From my analysis, it appeared that the lack of attention for informal street selling in an early modern context can be explained by two reasons. The first is that peddling and hawking are often immediately linked to marginality. The efforts to prove that ambulant traders were not marginal but instead crucial to the pre-industrial European economy have resulted in ignoring what the precise effects of operating in the shadows were, for the people involved and the economy at large. Whilst marginality is clearly an issue that is associated to street vending, both in the experiences of street traders and perhaps even more so in the perception of ambulant trading by social scientists and historians alike, it is, however, also a very problematic notion. Marginality holds multifaceted meanings, which likewise seem to be continuously disputed, and as such distracts from what we really want to know: what role street vending, with all its limitations and opportunities, plays in a society, either historical or modern.98 Perhaps, it is therefore best, before we return to the issue of marginality, if at all, to start by properly describing and analysing street vending in its multiple forms. A second reason for why we have such limited knowledge of the phenomenon of early modern informal street selling needs to be sought in the flexible character of street vending, especially in urban contexts, and the fact that (as a result) it was often performed by married women. The ad-hoc nature means that many people who would have sold in the street only part-time or occasionally may not have considered it as a “job,” and the very fact that many engaged in the trade without permission resulted in hiding it from authorities who tried to register their activities, for instance when compiling tax and population registers.99 From a recent investigation of eighteenth-century vagrancy it appeared that whereas in criminal court cases many were simply registered as vagrants, several actually engaged in some form of economic activity, including hawking.100 Additionally, whilst over the last decades important progress has been made in our understanding of women’s work in Europe’s pre-industrial

98 See the comments by Leslie Moch, Moving Europeans, p. 144 and by Lucassen, “Blind Spot,” pp. 216–8. 99 winter, “Adaptive Strategy,” pp. 263, 268. In contrast to this assumption Hufton stresses that the authorities in 18th-century France did see colportage as a “definitive job.” Hufton, Poor, p. 121. 100 Lucassen, “Blind Spot,” pp. 230–3; Winter, “Adaptive Strategy.” peddlers and hawkers in early modern europe 151 past, it is exactly on the occupational activities of married women that we still lack information.101 Understanding why people were involved in informal street vending, and who precisely these people were, will therefore require not only a more scrupulous analysis of records that are all too familiar to us, such as petitions and court records, but also a greater attention to the household economies of especially poorer households and the roles that wives, as well as perhaps also sons and daughters, played in this.102 In the light of fundamentally changing roles of production and consumption within households in the run-up to the Industrial Revolution, as the hypothesis of the Industrious Revolution recently put forward by Jan de Vries suggests, fully grasping the workings of this segment of the pre-industrial economy has become ever so important.103

101 Cf. Danielle van den Heuvel and Elise van Nederveen Meerkerk, “Partners in Busi- ness: Spousal Cooperation in Trades in Early Modern England and the Dutch Republic,” Continuity and Change, 23, 2 (2008), pp. 209–16. 102 Anne Winter’s inspirational methodology to uncover more information about those arrested as vagrants in 18th-century Brabant is a good example of how one could go about finding out more on street vendors, and in a recent paper with Elise van Ned- erveen Meerkerk we showed that by combining wide ranges of sources it is possible to expose activities of married women, their husbands, and changes in those according to life cycle. Winter, “Adaptive Strategy.” Van den Heuvel and Van Nederveen Meerkerk, “Huishoudens.” 103 Jan de Vries, The Industrious Revolution. Consumer Behavior and the Household Econ- omy, 1650–Present (Cambridge, 2008).

The Worst Class of Workers: Migration, Labor Relations and Living Strategies of Prostitutes around 1900

Lex Heerma van Voss

Only a couple of decades ago, labor history had a clear subject: male industrial workers in developed capitalist countries, who were fully engaged in wage work. Labor history was the story of their struggle through unions and political parties to improve their working and living conditions. This interpretation was criticized first by feminist historians, who pointed out that women should be brought into this story, and that bringing in women points to different occupations, different struggles, and other forms of organization.1 Over the past two decades, labor history has been broadened even further, as labor historians realized that several other parts of the traditional definition of their subject and their traditional story were an indefensible limitation. Not only were workers not exclusively males but also females; they were also not as individualistic as they had been perceived—they belonged to families, were in need of reproductive labor to survive, and often were employed not as individuals but in groups. They were not only or even typically employed as industrial wage workers, but also in services and in agriculture, and, for instance, as unfree or semi-free workers. They may prefer other forms of organization than trade unions or political parties, and their struggle may take other forms than striking and voting. In short, they may have living strategies which differ from the traditional script that labor historians were looking for. These may include setting up private businesses, illegal activities, or migrating to places that offer better working and living conditions, or seem to do so. And the majority of them had to live, work and struggle for change outside of the developed capitalist countries.2 In what follows, I want to consider whether this approach is helpful in analyzing the working conditions and living

1 To quote just one example from an immense body of literature: Alice Kessler-Harris, Gendering Labor History (Urbana and Chicago, 2007). 2 Jan Lucassen (ed.) Global Labour History. A State of the Art (Bern etc., 2006); Marcel van der Linden and Jan Lucassen, Prolegomena for a Global Labour History (Amsterdam 1999) (also at ). 154 lex heerma van voss strategies of prostitutes in six non-European towns, in the period between roughly 1875 and 1940. The towns selected are Nairobi, Buenos Aires, Shanghai, Mexico City, Bombay and San Francisco. None of these was part of the core area of global capitalism at the time under consideration, but all of them were in intense contact with this core, and the labor conditions and labor relations in prostitution were influenced by this global entanglement. These six towns were also selected because monographs on the local history of prostitution were available which supplied the necessary data.3 The comparison deals with labor relations within prostitution, but also with the position of women more generally: in migration, on the labor market outside of prostitution, and in the family. Each of the six towns started the period with tolerating prostitution. In all towns, different forms of prostitution were found, but the forms locally available differed from town to town. Usually a hierarchy was felt between different forms of prostitution. In some towns, the top of the hierarchy was formed by a more or less respected sort of prostitution, like courtesans. Types often found were prostitutes working in brothels, from their private rooms and streetwalkers. Of these forms, brothels had the clearest location. They were sometimes licensed, therefore subject to registration by the authorities, and retrievable in historical records. This is the form of prostitution that is therefore most studied by historians.4 The top layer of courtesans and well paid prostitutes had considerable control over working conditions, and was often able to end their career with a good marriage. If they played their cards well, they could thus achieve an amount of agency that was large for women in their societies and enjoy a comfortable standard of living.5 In Shanghai, traditionally,

3 Luise White, The Comforts of Home. Prostitution in Colonial Nairobi (Chicago and London, 1990); Donna J. Guy, Sex and Danger in Buenos Aires. Prostitution, Family and Nation in Argentina (Lincoln and London, 1991); Gail Hershatter, Dangerous Pleasures. Prostitution and Modernity in Twentieth-Century Shanghai (Berkeley, Los Angeles and London, 1999); Katherine Elaine Bliss, Compromised Positions. Prostitution, Public Health and Gender Politics in Revolutionary Mexico City (University Park (Pennsylvania), 2001); Ashwini Tambe, Codes of Misconduct. Regulating Prostitution in Late Colonial Bombay (New Delhi, 2009) and Benson Tong, Unsubmissive Women. Chinese Prostitutes in Nineteenth- Century San Francisco (Norman, 1994). Other relevant literature on prostitution in these towns includes: Christian Henriot, Belles de Shanghai. Prostitution et sexualité en Chine aux XIXe–XXe siècles (Paris, 1997) and Jacqueline Baker Barnhart, The Fair but Frail: Prostitution in San Francisco, 1849–1900 (Reno, 1986). 4 Guy, Sex and Danger, pp. 90, 118. 5 Hershatter, Dangerous Pleasures, p. 64. the worst class of workers 155 courtesan houses were very visible. The existence of courtesans was “not furtive or stigmatized. When a new brothel opened or a courtesan changed houses or professional names (. . .) it was often announced in the tabloid press (. . .) Brothels also advertised in the newspaper”.6 In Bombay, the top layer of courtesans had a similar position.7 The top segments of brothels often also offered other services beyond sex, like food, music or educated company. Below this top group, the hierarchy was often less clear. Streetwalk- ing is often seen as the most vulnerable form of prostitution. However, whether prostitutes working in brothels were really better off than those working the street depended on circumstances that differed from place to place and from time to time. Streetwalkers were often harassed by police officers, but were in a better position to refuse going with clients whom they did not trust. Prostitutes working in brothels were more vulnerable to pressure from brothel owners to accept customers whom they would have preferred to turn down. Women working from their homes or in brothels were sometimes also in a weaker position vis-à-vis customers who refused to pay for services rendered. To avoid confronting neighbors with their activities, prostitutes might prefer not to quarrel with customers. Fauzia Abdullah, a prostitute working in the Pumwani neighborhood of Nairobi from the mid-1920s, explained this problem as follows: “I mean that my people, the Muslim people here in Nairobi, often knew that a woman was a prostitute but she will have their respect as long as she doesn’t bring this to their attention.”8 Besides women working regularly as prostitutes, some women would supplement their main individual or family income with casual prostitution. Fulltime prostitutes with a certain amount of craft pride would look down on these occasional prostitutes, but others might well have thought them morally slightly above their professional sisters. In the period under consideration, government attitudes towards prostitution changed. At the start of our period, in the middle of the 19th century, the relevant governments either condoned prostitution, or regulated it with the intention of curbing sexually transmitted diseases. Where prostitution was regulated, women were licensed to work as prostitutes, and they were often required to undergo regular medical inspections.

6 Hershatter, Dangerous Pleasures, pp. 72–3. 7 Tambe, Codes of Misconduct, pp. xxii–xxiii. 8 White, Comforts of Home, p. 59. 156 lex heerma van voss

In Britain, the Contagious Diseases Acts of the 1860s allowed the police in towns with army or navy establishments to arrest prostitutes, and have them checked for venereal diseases. Arrangements for health checks were also introduced in many British colonies. In metropolitan Britain, public opinion came to consider such medical checks distasteful and objected to the double standard they embodied, as male clients were left untouched. The Contagious Diseases Acts were repealed in 1886, but licensed brothels and health checks remained common practice in many colonies.9 The campaign to repeal the Contagious Diseases Acts broadened into other aspects of prostitution. From the 1870s, it embodied the idea that girls and young women were kidnapped and lured into prostitution against their will. Although reports of white slavery were much exaggerated, the abolitionists succeeded in getting the age of consent for girls raised from thirteen to sixteen. From 1885, brothels were prosecuted, and the police cracked down on streetwalkers soliciting clients.10 This campaign reverberated internationally: the licensing of brothels came under attack, and over the next decades, many countries repealed licensing systems, and prohibited procuring, pimping and soliciting. The campaign against white slavery culminated in three international treaties (in 1904, 1910 and 1921) and an immense investigation by a special Expert Committee of the League of Nations in 1924–1927. The Committee interviewed 6,500 individuals in 28 countries. It failed to prove the existence of a large network of procurers luring European girls into prostitution. Where it established the existence of international migration of European women to work as prostitutes elsewhere, this concerned women who had already worked as prostitutes in their home countries and who expected better earnings in a foreign town.11 Even if in reality hardly any trafficking in white slaves could be proven, the international campaign against it still contributed to measures repressing prostitution.

9 Philippa Levine, Prostitution, Race & Politics. Policing Venereal Disease in the British Empire (London, 2003); Philip Howell, Geographies of Regulation. Policing Prostitution in Nineteenth-Century Britain and the Empire (Cambridge, 2009). 10 Judith R. Walkowitz, Prostitution and Victorian Society. Women, Class and the State (Cambridge, 1980). 11 Jean-Michel Chaumont, Le mythe de la traite des blanches. Enquête sur la fabrication d’un fléau (Paris, 2009); Thomas Fischer, “Frauenhandel und Prostitution—das Beispiel Buenos Aires in den 1920er Jahren,” in Petra Bendel (ed.), Menschen- und Bürgerrechte: Ideengeschichte und Internationale Beziehungen (Erlangen, 2004), pp. 229–258 (consulted at on 4 November 2011). the worst class of workers 157

Prostitution in the six towns

Nairobi was a recently founded colonial administrative and military town. It drew immigrants from the surrounding areas, where native tribes found a livelihood in agriculture, especially raising cattle. This hinterland had been plagued by rinderpest, and social relations had been changed by increasing monetization, induced by taxes introduced by the British. Pros- titution in Nairobi took three forms: malaya (prostitutes who received in their own rooms), watembezi (streetwalkers) and wazi-wazi (working from a door, a window or a porch). In the wazi-wazi form, women had somewhat more control over who entered as a customer. It was typically used by daughters helping their parents out. The house where several malaya prostitutes had their room might be owned by a former prostitute, but it was not a brothel, and the owner was not a madam. She received rent, not a share in the prostitute’s income. In the period under consideration, Kenya followed the Indian Penal Code. Prostitution was not a crime, but soliciting was. However, prostitution was not very accepted, and all types of prostitutes tried to be inconspicuous. Black inhabitants of Nairobi were confined to two neighborhoods: Punwami and Pangani, where a curfew was in place. Buenos Aires drew a large number of European immigrants, and would be a magnet for any international network of prostitute migration. In the town, prostitution was practiced in brothels and by streetwalkers. Licensed brothels were legal businesses from 1875. Later laws regulated where in town prostitution could be practiced. In 1934 Buenos Aires abolished municipally regulated prostitution. In Shanghai, the foreign concessions comprised a large part of the surface of the inner town, but the number of resident Westerners was limited: about 1% of the population. Between 1870 and 1930 Shanghai became a major economic, political, and cultural center and prostitution changed from a luxury service by courtesans into a varied market primarily geared to commercial and working-class immigrants. Prostitutes were found in different classes of brothels and on the streets. Qing and Republican legislation allowed prostitution, but it forbade procuring and trafficking. In Shanghai, brothels were licensed and taxed. In 1920–1924, a successful prohibition movement led to the gradual closure of brothels in the Inter- national Settlement, but this only drove prostitution to the Chinese part of town and to the French Concession, where prostitution remained legal and licensed. 158 lex heerma van voss

Under the 1872 Reglamento in Mexico City, prostitution was licensed and prostitutes were registered in an attempt to limit the spread of venereal diseases. In 1898, the minimum age was raised from 14 to 16 years. Brothels of different classes were distinguished, and rules were laid down for their location. Aisladas, independent women not working in brothels, could meet their clients wherever they liked. Among its 200,000 female inhabitants in 1908, it was estimated that Mexico City counted 10,000 registered prostitutes and as many unregistered clandestinas. Licensing was suspended in 1940. In Bombay, prostitution as such was not against the law. Brothels and streetwalkers operated in the town. From 1867, brothels were licensed and inspected to curb venereal diseases. From that date, prostitution by unregistered women became illegal. In 1870, prostitutes were segregated by neighborhood. Over time, rules on holding women against their will and against trafficking were tightened, and from 1926 soliciting was prohibited. As regards San Francisco, I do not consider all forms of prostitution, but only Chinese prostitutes. Both male and female Chinese migrants to San Francisco predominantly came from peasant families in the Kwantung area in South China. Chinese males migrated to work on the railroads, in mines or lumber camps and in various other occupations. If they married a Chinese woman, they usually did so in China, and their family remained there. Among the Chinese women who migrated to California, by far the largest group worked as prostitutes. The first women arrived in the early 1850s and soon set up a number of brothels. From 1854, Chinese prostitution was run by gangs, the Tong, which bought and sold the women it brought over. From that moment, Chinese prostitutes were seldom self- employed, as Euro-American prostitutes in San Francisco were. Chinese prostitutes worked in the red light district around Dupont Street, and led a secluded existence, segregated from other Americans. Streetwalking was rare: the skewed sex ratio in California did not force prostitutes to solicit clients. Prostitution was outlawed by a 1854 City Ordinance, but this rule was not enforced. After the Chinese Exclusion Act of 1882 Chinese migration to the United States became insignificant, and Chinese prostitution in San Francisco declined. the worst class of workers 159

Other opportunities for women to earn their livelihood

In the historiography of prostitution in countries in the capitalist core, industrialization has often been regarded as a cause of a rising number of prostitutes. With industrialization, young working class men and women left the protection of their rural family homes and migrated towards towns. There, young women often did not find the waged labor they had hoped for, or the wage was insufficient. Girls born to urban working class families, too, could not find proper paid employment. These women found in prostitution a solution to make ends meet.12 An example is Shao Meiting, a woman from the countryside to the South of Shanghai. She found a job in a cigarette factory in Shanghai, but when the factory folded, she told the police: “I had no way to support myself, and had to sink to becoming an unlicensed prostitute. Originally I planned to do it for several months and then return to the countryside. My parents don’t know I am doing this.” She declined help to leave prostitution: “I can’t go, because my parents are old and my sisters are still young, and there is no one at home to take care of them.”13 Indeed, all six towns surveyed went through a period of fast urban growth, with either increasing government and military activity, or growing industrial employment. This meant that there were large numbers of young men and women migrating to these towns, with the men usually outnumbering the women. Job opportunities for women were limited. The lists of waged occupations open to women in the different towns are remarkably similar. Nairobi was founded in 1899 by the English coloniz- ers. It had a surplus of white male settlers, and no industrial base, so the permanent African population, male and female, worked in the service sector. By 1938, it had a total population of about 65,000, of whom 40,000 were Africans. In that year, there were 8 African men for every African woman. Of the latter, only 237 were legally employed: 164 in child care, 58 as hospital ward attendant and 15 brewing beer.14 The pattern of

12 Ruth Rosen, The Lost Sisterhood: Prostitution in America, 1900–1918 (Baltimore and London, 1982), pp. 3–4; Victoria Harris, Selling Sex in the Reich. Prostitutes in German Society, 1914–1945 (Oxford, 2010), pp. 46–55. 13 Quoted in Hershatter, Dangerous Pleasures, p. 195. 14 White, Comforts of Home, pp. 80, 98. A similar pattern existed in Johannesburg, where prostitution, laundry work and liquor selling were the main female occupations in 1927 (Kathy Eales, “Popular representations of Black Women on the Rand and their Impact 160 lex heerma van voss women working in domestic care is repeated in the other towns, with the addition of working in the textile industry. Buenos Aires grew from 180,000 inhabitants in 1869 to 1,4 million in 1914, mainly through international immigration. In 1869 16% of the female population worked in textiles, sewing or cigarette manufacturing. Over the following decades employment in domestic work and the needle trades increased, but not enough to absorb the influx of women.15 The population of Shanghai grew from one to three million between 1910 en 1930, mainly through immigration from the Chinese countryside. Men outnumbered women about 4 to 3. The immigrant women worked in manufacturing, especially cotton textiles, as household servants, wet-nurses, itinerant peddlers, and entertainers.16 Mexico City grew from about 500,000 inhabitants in 1900 to 1.6 million in 1940, mainly through immigration from the Mexican countryside. Young women outnumbered young men significantly, and besides in prostitution, they found jobs in laundries, sewing, restaurants and light manufacturing, but these rarely provided female workers with a living wage.17 Bombay counted 650,000 inhabitants in 1872, 980,000 in 1906 and 1,5 million by 1941, due to migration from the Indian countryside. Between 1870 and 1910, men outnumbered women about 5 to 3.18 The Chinese migration to San Francisco was almost totally male. In 1870, the 2,794 Chinese prostitutes formed 77% of the Chinese women in California. The other occupations were in declining order of magnitude: laundresses, miners, servants, seamstresses, cooks and lodging house operators. The 1880 census counted 1,726 Chinese women with a gainful employment in California: 44% were prostitutes, the other recorded occupations were: seamstress, servant, laundress, cook, entertainer, laborer, miner and lodging house operator.19 Male or female surpluses both seem conducive to large numbers of prostitutes, increasing respectively demand and supply. What all these situations have in common, is that large numbers of women could not find other employment that paid a living wage. Not only did prostitution offer work, it was even often well-paid work, unless middlemen like the Tong gangs in San Francisco took a large cut.

on the Development of Influx Controls, 1924–1937,” University of Witwatersrand History Workshop, Structure and Experience in the Making of Apartheid, 6–10 February 1990, p. 8). 15 Guy, Sex and Danger, pp. 41, 45. 16 Hershatter, Dangerous Pleasures, p. 40. 17 Bliss, Compromised Positions, pp. 25–26, 66, 46, 151, 33. 18 Tambe, Codes of Misconduct, pp. xx, xxiv. 19 Tong, Unsubmissive Women, p. 30. the worst class of workers 161

Even in rich and developed countries, prostitution offers an income that lies well above what women working as a prostitute can earn elsewhere. As Havelock Ellis observed at the beginning of the 20th century: “no prac- ticable rise in the rate of wages paid to women in ordinary industries can possibly compete with the wages which fairly attractive women of quite ordinary ability can earn by prostitution”.20 Luise White’s study of prostitution in Nairobi offers us an interesting view on the origins of the oldest profession in a town which was only founded in 1899, and where prostitution as an occupation only became established in the 1910s. She quotes Amina Hali, who was born around 1895, and who described her first contacts as a prostitute: When we went to pick beans, we sometimes found these Kibura [white] men, so it was extra money, we went to pick beans and had a man in secret. Sometimes a woman would go there just for the men, she would take a gunnia [gunny sack] so that no one would be suspicious, it looked like she was going to pick beans but she would use the gunnia as a blanket . . . When they saw a woman lying on her gunnia they would take out their money, and she would motion for him to lie down with her. They paid us and sometimes they gave us babies, so we were rich, we had money and babies that way.21 That the women earned a good income, is confirmed by a male observer: I remember seeing Nandi [a Kenyan tribe] women before the German War [World War I]. They were cutting wood over near Muthiaga and they took it to Mambase Village to sell. Some of these girls would go . . . with anybody, even white men and Indians. You could give her anything you wanted—half a rupee, one rupee—but some men gave a cow or some goats. You know, in those days a very important African only made Rs. 4 or Rs. 5 a month, and workers got even less. So these prostitutes really made a lot of money— more than most men—and they even raised their prices after the German War. That’s how they came to have so many houses in Pumwani.22 So even here, from the start of the establishment of prostitution and when it was only a secondary activity besides bean-picking, prostitution paid relatively well.

20 Quoted in Lena Edlund and Evelyn Korn, “A Theory of Prostitution,” February 16, 2001, consulted at [4 November 2011]. 21 White dates this recollection to the years 1909–16, Comforts of Home, p. 41. 22 Elderly Kikuyu man, interviewed in 1966, quoted in Kenneth G. McVicar, “Twilight of an East African Slum: Pumwani and the Evolution of African Settlement in Nairobi,” PhD thesis, University of California, Los Angeles, 1968, pp. 240–1. Here quoted after White, Comforts of Home, p. 41, who doubts whether ever a cow was traded for intercourse. 162 lex heerma van voss

Trafficking and other forms of migration

In all six cases, recent immigrants were an important recruiting ground for prostitution. These were internal migrants in Nairobi, Mexico City, Shang- hai and Bombay.23 In Buenos Aires, they were mainly foreign immigrants, but that held true for the population of the town as a whole.24 The Chi- nese women who migrated to San Francisco were brought to that town to work as prostitutes. Two types of trafficking demand our attention here. The first one is “white” slavery, an idea which led to a moral panic in the West lasting from the 1870s to the 1920s. This idea was of course inherently racist: the concern of the activists was not that any girl might be abducted and forced to work as a prostitute, but that this might happen to Caucasian girls. In line with this concern, when international concern about white slavery led to pressure on Bombay police commissioner Stephen M. Edwardes to act against trafficking in European prostitutes, he declared the adjective “white” incorrect for Bombay, as the women “were chiefly of Eastern Euro- pean origin”. Catering to anti-Semitic sentiments, Edwardes also reported repeatedly that many “white” prostitutes were actually Jewish. Clearly, Mr. Edwardes considered Jews and East Europeans less white than other whites and trafficking in these cases consequently a less urgent problem.25 A global migration network of white prostitutes certainly existed. Euro- pean prostitutes were available in Mexico City, Shanghai, Bombay and Buenos Aires.26 In Mexico City, the arrival of French and Jewish pimps and prostitutes was noted in the 1920s, but by the end of the decade only 1–2% of the registered prostitutes was foreign. In Shanghai, some 8,000 Russian were to be found, their number swollen by refugees from revolutionary Russia, and 2,000 West-European prostitutes. This would mean that 10–15% of all prostitutes in Shanghai were foreigners. In Bombay, the percentage declined from about 5% to about 1%. In Buenos Aires, the number was considerable, but this was in a town were foreign immi-

23 White, Comforts of Home, pp. 37–9; Bliss, Compromised Positions, pp. 16, 26; Hershat- ter, Dangerous Pleasures, p. 40; Tambe, Codes of Misconduct, pp. xx, xxiv. 24 Guy, Sex and Danger, pp. 41–2. 25 Tambe, Codes of Misconduct, pp. 57–8. 26 Bliss, Compromised Positions, pp. 91–2, 138–45; Hershatter, Dangerous Pleasures, pp. 50–3; Tambe, Codes of Misconduct, pp. 57–64, 72; Harald Fischer-Tiné, “ ‘White Women Degrading Themselves to the Lowest Depths’: European Networks of Prostitution and Colonial Anxieties in British India and Ceylon ca. 1880–1914,” Indian Economic Social History Review, 40 (2003), pp. 163–90. the worst class of workers 163 grants constituted a majority of the population as a whole. Until the 1870s, both upper- and lower class brothels in Buenos Aires featured native-born women, but from the 1880s French women, or women who posed as such, set the tone in upper class brothels. The women came from Southern, Middle and Eastern Europe, partly via Uruguay. An extensive network of intermediaries was involved in recruiting and shipping these women. However, they were not the innocent young victims of seducers and kid- nappers, but seasoned prostitutes who migrated to a promising market, where their ethnic background sold at a premium. In 1912, the women who arrived were asked about their previous status, and 70% declared that they had been prostitutes before. All in all, the number of immigrant women that might have been misled or forced to come to work as a prostitute in Buenos Aires, was extremely small.27 Real trafficking existed, with a considerable percentage of women being sold into prostitution against their will and against their expectations. But it was not “white,” but Asian. In Shanghai, in Chinese prostitution in San Francisco and in Bombay, women were led to brothels by false promises of marriage or employment as mill worker or nanny. In 1928, the Bombay Vigilance Association intercepted 54 trafficked Indian girls, of whom four had been promised marriage and seven a job. Thirteen were seduced or kidnapped, en the rest were dedicated by their parents to temples or were considered a prostitute on the basis of their caste.28 Similar data exist for Shanghai, were just before and after the Chinese Revolution prostitutes were asked how they had become prostitutes. Her- shatter reports on two such surveys, one in 1948 and one in 1951. They give different categories of answers, but both give low figures for being lured into prostitution. For most Shanghai prostitutes, the decision to enter prostitution was not made under circumstances completely of their own choosing; but it was a decision made by their families and themselves. Prostitutes often introduced other members of their family or native place network into their brothel. Compared to the alternative, that could be a positive choice. One social worker who interviewed prostitutes before 1949 summarized their feelings: “Other work was rather tiring. This kind of work made more

27 Guy, Sex and Danger, pp. 16, 33–4, 46–7, 63, 73, 112–6, 120, 126–8; Fischer, “Frauenhan- del und Prostitution—das Beispiel Buenos Aires.” 28 Tambe, Codes of Misconduct, p. 74. Similar cases are mentioned in Tong, Unsubmis- sive Women, pp. 34–53, 193. 164 lex heerma van voss

Table 1. Reasons given by former prostitutes for their entry in the profession 1948 1951 tricked or seduced 4% kidnapped 9% family pressure 5% sold or pawned by family 11% poverty 60% death or unemployed earner 43% unemployment/bankruptcy 18% family circumstances/divorce 27% affinity for the work 13% total 100% 90% N 500 501 Source: Hershatter, Dangerous Pleasures, p. 194. The author does not explain why the 1951 survey does not add up to 100%. income, and one ate well, so they did it.”29 But quite often the choice was not as positive as that, but induced by debts the family had incurred, or the inability to provide an income by other means. Women could either be sold or pawned to brothels. Hershatter concludes that pawning a woman into prostitution usually was a family decision to deploy an economic resource—a woman’s body—without relinquishing it. As with many family strategies, it remains unclear how much say the family member most involved, the woman, had in this decision. But an example suggests that women not necessarily resisted being pawned. In a 1929 court case, a father charged his son with selling his wife to a brothel. At the hearing, both the woman and the madam testified that she had not been sold, but pawned. The woman did not use the suit to try and win her freedom. As her father in law was willing to file a suit against his son over his daughter in law, and as the son himself worked in a clothing store just outside the brothel to which he had pawned his wife, the woman cannot be seen as cut off from her family network.30 Pawning remained common until the 1930s, when protests against the practice grew. Some sources suggest that half of the women that entered prostitution were pawned. In pawning, the woman acted as collateral to a loan made to the family. The sum loaned was half the value of the woman. She lost her freedom, and all her income as a prostitute went to the brothel owner. The family received money, sometimes also some

29 Hershatter, Dangerous Pleasures, p. 195. 30 Ibidem, p. 199. the worst class of workers 165 wages, and had a mouth less to feed. Pawning allowed the brothel owner to buy a prostitute’s most productive years.31

Labor relations in prostitution

With the pawning and selling of Chinese women, we have already introduced one form of labor relation in prostitution. In these cases a Chi- nese woman would lose her freedom, permanently or temporarily, and this was of course only possible in legal systems that allowed for people losing their freedom. The Tong gangs exported these practices—under Chinese law—to the United States. In Bombay too, it was possible to own women. Here, in 1917, the prostitute Akootai was tortured to death by her owners, which led to a criminal court case which sheds some light on the relations between brothel owners and prostitutes. The brothel owner Mirza was a Pathan, an ethnic group which was “commonly portrayed as greedy moneylenders who forced the wives of debtors into prostitution”. Akootai’s fellow prostitutes, who were heard at the trail as witnesses, were in Mirza’s power, because they respectively had borrowed twenty rupees from him, because Mirza paid off a debt the woman owed (she had been enticed away from another brothel by Mirza’s brother, who promised he would keep her as his mistress), and because the third women was sold for fifty rupees to Mirza by a former paramour.32 In other instances, free Bombay women worked in a brothel and shared their income with the madam according to set rules. Pimps figure not very prominently in these cases. In Mexico City, the brothels were run by madams. Pimps start being mentioned from the 1920s, specifically foreign pimps in connection with foreign prostitutes. However, they apparently do not become very important.33 In Shanghai, pimps became important even later.34 In Bombay pimps and male owners could be brutal, as Akootai’s case showed. Matters improved when a 1930 law prohibited brothels and prostitutes started working from individual rooms.35 In Buenos Aires, pimps cornered a large part of the prostitution market in the 1920s, but even here their grip was less strong than it was

31 Ibidem, pp. 196–8. 32 Tambe, Codes of Misconduct, pp. 79–99. 33 Bliss, Compromised Positions, pp. 91–2. 34 Hershatter, Dangerous Pleasures, pp. 341–2. 35 Tambe, Codes of Misconduct, p. 112; Santosh Kumar Mukherji, Prostitution in India (New Delhi, 1986) [Reprint 1934 edition], pp. 244–45, 465. 166 lex heerma van voss reputed to be.36 Perhaps the influence of pimps and their cut of prostitutes’ income were in general not so large as reputation would have it.37 Like pirates, pimps trade in fear, and they had certainly every reason to magnify their bad reputation, as it was an important tool to extort money from the prostitution branch. In Nairobi, prostitutes were generally self-employed. This meant that they did not have to share their income—which, as we saw, was considered high—with pimps, madams or brothel owners. That they were self- employed prostitutes, and thus tried to avoid wage labor, for themselves and their dependents, did not mean that they had no relation with wage labor. Working women combined different strategies and types of labor relations to generate income. In Nairobi, living from wage labor alone was a feasible strategy for only a very limited number of women. From the historical literature on prostitution in the West, we are familiar with the idea that women occasionally work as prostitutes to supplement a low-wage income. In Nairobi, it makes just as much sense to conclude that some women used wage labor as a way to increase income from prostitution. White gives several examples.38 Hadija binti Nasolo came in 1938 from Uganda to live in Pumwani. She was employed as a part-time sweeper in a Nairobi police station. She used this job to meet well paying European men, one of whom took her to a hotel outside Nairobi. This was what made her poorly paid wage labor worthwhile. Wanjira Ng’ang’a decided to leave her agricultural village, and seek her fortune in Nairobi in 1937/38, together with a friend. The reason to do so was that the friend had already “gone with European men and they paid a lot of money”. As it turned out, her friend landed a job as a domestic servant for an Asian family in the Westlands neighborhood of Nairobi. In colonial Nairobi, prostitution in itself was not a crime, but soliciting was. In the black neighborhoods Pumwani and Pangani strict curfews applied. Inhabitants of these neighborhoods could not stay outside them in the evening; whites were not allowed in Pumwani and Pangani between seven o’clock at night and six o’clock in the morning. As rents were high, domestic servants often shared rooms supplied by their employers with friends

36 Guy, Sex and Danger, pp. 120–9. 37 Other indications for this in Harris, Selling Sex, pp. 78–86; Edlund and Korn, “A The- ory of Prostitution,” p. 8; and Ben L. Reitman, The Second Oldest Profession (New York and London, 1987), pp. 85–94. 38 White, Comforts of Home, pp. 65–6, 98–101. the worst class of workers 167 or co-workers. When her friend was housed on her employers’ premises, Wanjira Ng’ang’a shared her room, and was thus allowed to reside at a strategic place: “so we both lived there and every morning she would wake up early and go to work, while I would go out and look for men.” Since she had to safeguard the respectability of her friend and her friend’s employers, Ng’ang’a could not opt for the malaya form, but became a watembezi prostitute: I would walk up and down the streets, or I would stand at a corner, and if a man asked me to go with him I would tell him that I had no place to go, and could we go to his house. Sometimes, they were African men, and they were somebody’s servants we could go to their rooms, but sometimes we couldn’t and then we would go to the bushes, just for short-time. But if it was an Asian man or a white man who wanted to go with me, then there was no problem, we could go directly to his house. In Buenos Aires, the chance that a woman could land a paid job deteriorated in the 1930s. This led to protests, in the wake of which a vocal minority of prostitutes demanded respect as women and as humans working for honorable reasons in a shameful industry. One prostitute told a reporter “we have not all become what we wanted to become, but the fact is that we are workers, the worst class of workers, but we have a right to live as gente decente.”39 Another source of respect for prostitutes was the income that they could generate, which was the basis of their material wealth or of their remittances home. In Nairobi, malaya prostitutes earned enough to buy houses, and rent out rooms to other Africans. Sometimes their lodgers would also be prostitutes, but the house owners would not act as madams. In other parts of the world, prostitutes used their savings to buy a brothel and become a madam. In San Francisco, Ah Toy was one of the first Chinese prostitutes in business. She procured other Chinese women and owned two brothels, which she lost when, from 1854, the Tongs took over the prostitution business in San Francisco. After that date, women seldom more managed to develop themselves to independent entrepreneurs.40

39 “La vida miserable y trágica de las carabeteras revelanda ante varios funcionarios oficiales,” El Gráfico, October 19, 1937, p. 12, quoted in Guy, Sex and Danger, p. 200. 40 Tong, Unsubmissive Women, pp. 6–11. 168 lex heerma van voss

Relatives

Respect as workers or business success were not the only possible source of pride for prostitutes. As we saw, many women entered prostitution to deal with a family crisis. By enabling their family to settle as debt, or to survive a drought, prostitutes contributed to family survival.41 In Kenya, around 1900, increased wage labor by young men led to higher prices for the livestock they had to acquire as bridewealth, and to higher bridewealth payments. Livestock also became scarce because of war and rinderpest. Prostitutes sent home remittances that were used to replenish stock.42 In Shanghai, working on behalf of family, to clear a debt or pay for a parent’s funeral, even could be construed as a form of filial piety.43 Sometimes real family relations existed within a brothel, for instance when a prostitute raised her children there, or when a prostitute went back to her native village to fetch her younger sisters to work in the brothel.44 Often, within brothels, surrogate family relations were enacted.45 Madams and prostitutes addressed each other as mother and (adopted) daughter. Little girls were raised in Shanghai brothels as apprentices: they received no salary, but were taught a trade. Eventually, the defloration fee would defray the costs of raising a girl. Their position resembled that of “little daughters in law” who were adopted by the family of their future husband, ensuring the husband’s family of their labor, and save the future cost of purchasing a grown bride.46 Even in the Bombay brothel where Akootai was tortured to death, family-like relations between non-kin were adopted.47

The indecent way to live as gente decente

I posed the question whether the toolbox of enhanced labor history, developed over the last two decades, is useful in analyzing the working condi-

41 Guy, Sex and Danger, p. 45; White, Comforts of Home, p. 20. 42 White, Comforts of Home, pp. 34–9. 43 Hershatter, Dangerous Pleasures, pp. 194, 199. 44 Tambe, Codes of Misconduct, p. 95; Hershatter, Dangerous Pleasures, p. 75; White, Comforts of Home, p. 42. 45 Hershatter, Dangerous Pleasures, p. 311. 46 Hershatter, Dangerous Pleasures, pp. 75–6. 47 Tambe, Codes of Misconduct, pp. 86–7. the worst class of workers 169 tions and living strategies of prostitutes in six non-European towns in the period around 1900. The answer, I would like to suggest, is affirmative. In three of the six cases discussed, women could be sold, also to a brothel. Not by coincidence, these were also the three cases in which women actually were kidnapped or lured into prostitution, while they were led to believe that they were being married off or migrating to urban wage employment. Even so, in these three cases, the huge majority of women sold or pawned to a brothel knew what awaited them, and accepted this as a survival strategy to help carry their family through troubled times. How reluctant this acceptance was, is hard to fathom. Also not coincidentally, in these three cases, the position of women in the sex trade was on average clearly weaker than in the three other cases. The women were physically mistreated more often, had less say over their daily lives and earned less. In the other three cases, the legal position of women was somewhat stronger. Still, in Kenya women went into sex work both because they thought prostitution less burdensome than agricultural work in the countryside, and because this was a way to accumulate capital or to help their families through times of hardship. In the Argentine countryside, abduc- tion of women was a prelude to consensual marriage. Wives were abused sexually by their husbands’ employers. This may have made prostitution more palatable, but in all six cases women experienced the transition to commercial sex probably as less traumatic than middle- and upper-class— reformers—both in Europe and in the six towns—could imagine.48 Labor relations in prostitution were often harsh. They were also diverse, ranging from self-employment in Nairobi, via sharing revenues, wage labor and debt bondage to slavery. When women were working in a brothel, this list is probably ranged in order of declining agency and worsening working conditions. It would be less clear where streetwalking would fit in this range, which was usually self-employed, but had a lower status than working in a brothel. Pimps were only of some importance in Buenos Aires, but madams were very central in running brothels. They ran from abusive to motherly. The conclusion that prostitutes were worst off in slavery does not sound very surprising. However, we should keep in mind that the reverse relation was not automatically true. The predominance of slavery in a society needs not necessarily lead to a high incidence of prostitution, nor to the

48 Guy, Sex and Danger, pp. 43–5. 170 lex heerma van voss exploitation of female slaves as prostitutes. In the Southern, slaveholding states of the United States, blacks were underrepresented among prostitutes. Fogel and Engerman reported on a count of prostitutes in the town of Nashville, where, in 1860, 4.3% of the prostitutes were black, compared to a fifth of the population of the town. All black prostitutes where free and light-skinned. They concluded that white men who desired illicit sex had a strong preference for white women. The growth of black prostitution in the United States was a 20th-century phenomenon.49 This tallies well with some of the characteristics that our cases had in common. All towns experienced rapid urbanization, with an influx of both large numbers of males and females from the countryside (and, in the Argentine case, also from abroad). Most towns also were developing an industrial sector. Jobs that offered women a living wage were scarce. The jobs to be had were typically in domestic service, in textiles and the needle trades. Still, the situation on the countryside was so precarious, that migration to the town seemed the better option, even to women who knew full well that they would be working as prostitutes. In the Asian cases, trafficking was only supplementary to the much larger migration flows of women who knew where they were headed. European women had usually worked in prostitution before, and migrated in search for better prostitution markets. This too, was a reality that was hard to stomach for the reform movement. For most women, working in commercial sex was the only viable way to earn a good income. They had to accept to be the worst class of workers, to live to some extent as gente decente.

49 Robert W. Fogel and Stanley Engerman, Time on the Cross. The Economics of Ameri- can Negro Slavery (London, 1976) pp. 134–6; Eugene D. Genovese, Roll, Jordan, Roll. The World the Slaves Made (New York, 1974), pp. 460–1; Herbert G. Gutman, The Black Family in Slavery and Freedom 1750–1925 (New York, 1977), p. 613, n. 9; Jacqueline Jones, Labor of Love, Labor of Sorrow. Black Women and the Family, from Slavery to the Present (New York, 1985), pp. 181–2; Kevin J. Mumford, Interzones. Black and White Sex Districts in Chicago and New York in the Early Twentieth Century (New York 1997). Fighting for a Living in Europe and Asia

Erik-Jan Zürcher

Introduction

For a long time, labor historians did not regard the activities of soldiers as “work.” “Work” was essentially defined as “an activity yielding surplus value,” and the efforts of soldiers were considered to be essentially destructive, rather than productive. This assumption, that military work is necessarily destructive, and does not produce a surplus value, is debatable. There are at least two reasons. The first is that soldiers everywhere spend far more time in barracks than on a campaign, and while they are garrisoned, they have very often been employed as “cheap labor”—in agriculture, or in building works and road repair. In countries as far apart as France and China, many of the greatest infrastructural works would never have been realized except for the massive use of military manpower. The second, more profound reason is that, as Peter Way has argued, the end result of warfare, if successful, is that an additional surplus value is created for states and their elites through territorial gain, or economic advantage.1 Whatever the merits of this argument may be, the effect of the view that “soldiers do not work” has been that military labor has not become an object of research, in the same way as the labor of (for instance) dock- workers, textile workers, miners or agricultural workers has.2 One of the very first people to resist this approach was Jan Lucassen. As early as 1994, he devoted attention to the “proletarian experience” of mercenaries

1 Following Marx, Peter Way closely identifies the growth of capitalism and the modern state with warfare, particularly, colonial warfare. See: Peter Way, “Klassenkrieg: Die ursprüngliche Akkumulation, die militärische Revolution und der britische Kriegsarbei- ter,” in Marcel van der Linden and Karl Heinz Roth (eds), Über Marx hinaus. Arbeitsge- schichte und Arbeitsbegriff in der Konfronation mit den globalen Arbeitsverhältnissen des 21. Jahrhunderts (Hamburg, 2009), pp. 85–114. 2 Chris Tilly and Charles Tilly express this point of view in their book Work Under Capitalism (Boulder, CO, 1998), p. 23: “To be sure, not all efforts qualify as work; purely destructive, expressive, or consumptive acts lie outside the bound; in so far as they reduce transferable use value, we might think of them as antiwork.” 172 erik-jan zürcher in early modern Europe.3 I myself became more and more aware of the degree to which a soldier’s life can be understood in terms of work, when I did empirical research in the 1990s on the everyday realities of Otto- man soldier’s lives during World War I.4 Our paths converged, and during pleasant lunch breaks at the IISH, Jan Lucassen and I talked about the issue many times. In 1999, we published “Conscription as Military Labor: The Historical Context.”5 When I started a collaborative project on the history of military labor at the IISH in 2009, entitled Fighting for a Living, it was therefore no surprise that Jan immediately volunteered to attend the workshops of the research team, to listen, learn and debate our findings. His interventions were particularly valuable because they enabled the members of the team—mostly specialists in military history, rather than in labor history—to compare their own findings to those in the wider field of labor history, in which Jan Lucassen is such a prominent figure. The project brings together twenty specialists from nine countries for three workshops in the period 2009–2012; its results will be published in two volumes in Amsterdam in 2013, under the title War, Work and the Soldier. The aim of the project is to investigate the circumstances which have produced starkly different systems of recruiting and employing soldiers, in different parts of the globe and across a five hundred year period, as well as to analyze the social and political implications that the different systems have had in a number of states and societies. The project concentrates on land armies in Europe, the Middle East, India and China in the period 1500–2000. That means that some important continents are not included: Latin America, Africa and Australasia. Categories like non-state forces (guerrilla movements, slave- or peasant armies) are also not included. The project explicitly concentrates on state armies in the

3 Jan Lucassen, “The Other Proletarians: Seasonal Laborers, Mercenaries and Miners,” in Catharina Lis, Jan Lucassen, and Hugo Soly (eds), Before the Unions: Wage Earners and Collective Action in Europe, 1300–1850, supplement 2 to International Review of Social His- tory, 39 (1994), pp. 171–94, at 185. 4 “Between Death and Desertion. The Experience of Ottoman Soldier in World War I,” Turcica, 28 (1996), pp. 235–58; “The Ottoman Conscription System in Theory and Prac- tice, 1844–1918,” International Review of Social History, 43, 3 (December 1998), pp. 437–49; “Ottoman Labor Battalions in World War I,” in Hans-Lukas Kieser and Dominik J. Schaller (eds), Der Völkermord an den Armeniern und die Shoah. The Armenian Genocide and the Shoah (Zürich, 2002), pp. 187–96; “Hizmet Etmeyi Başka Biçimlerle Reddetmek: Osmanlı İmparatorluğunun Son Dönemlerinde Asker Kaçaklığı” [Refusing to Serve by Other Means: Desertion in the Late Ottoman Empire], in Özgür Heval Çınar and Coşkun Üsterci (eds.), Çarklardaki Kum: Vicdani Red (Istanbul, 2008), pp. 59–68. 5 International Review of Social History, 43, 3 (December 1998), pp. 405–20. fighting for a living in europe and asia 173 context of advanced state formation. That we also excluded the navy could be regarded as a serious omission. But precisely because we recognize that navies (which in many respects are very different from land armies in their skill levels, traditions and recruitment) offer a hugely interesting field for comparative research on military labor in their own right, we decided to leave naval activities to a possible separate project. Fighting for a Living is still very much work in progress. Based on the twenty draft chapters that the members of the research group have pro- duced6 and the many thought-provoking discussions we have had, I will however attempt a preliminary taxonomy here of military labor relations in Europe and Asia during the last five hundred years, and moot some suggestions about what kind of determinants influence the prevalence or demise of particular types of labor relation.

Huge variations

At a phenomenological level, even when we limit ourselves to land armies in the service of the state, the variety of forms of military labor is almost endless. To make meaningful comparisons possible, a basic classification nevertheless has to be applied. The search for such a classification was high on the agenda of the research group of Fighting for a Living.

6 Given that the project papers have not been published yet, it is unfortunately impossible to give page references for the data from the different contributions that I have discussed in this essay. The following manuscripts have been used: David M. Robinson, “Military labor in Ming China”; Kaushik Roy, “A social history of military service in South Asia, 1500–1650”; Gilles Veinstein, “On the Ottoman Janissaries”; Frank Tallett, “Soldiers in Western Europe, 1500–1790”; James Miller, “The Scottish mercenary as migrant laborer in Europe, 1550–1650”; Michael Sikora, “Transitions of mercenary armies, Central Europe 1650–1750”; Dirk Kolff, “Peasants fighting for a living in North India”; Robert Johnson, “Recruiting and managing military labor in the Army of the East India Company, 1746–1763”; Peter Way, “Recruiting the British Army in the Eighteenth Century”; Virginia Aksan, “Mobile warriors in the Ottoman context, 1750–1850”; Christine Moll-Murata and Ulrich Theobald, “Military employment in Ch’ing dynasty China”; Elise Kimerling Wirtschafter, “Military service and the Russian social order, 1649–1861”; Thomas Hippler, “The French Army, 1789–1914: pressed soldiers and conscripts”; Herman Amersfoort, “The Dutch army in transition, 1795–1830”; Marco Rovinello, “Drafts and draftees in Italy, 1861–1914”; Khaled Fahmy, “The army of Mehmed Ali Pasha”; Jörn Leonhard, “Nation-building and war experiences: the British case”; Mehmet Beşikçi, “Ottoman conscription before and during World War I”; Beth Bailey, “The All-Volunteer Force in the United States of America”; Yelda Kaya, “Private contractors in war, 1990–2010”. 174 erik-jan zürcher

One method of grouping the different phenomena is that of John Lynn, in his seminal work on the developments of European armies.7 Lynn distinguished four basic “army styles”: the “feudal army,” the “aggregate contract army,” the “state commission army,” and the “conscript army.” Central to his thesis is the notion that, around 1650, the aggregate contract army—which Lynn describes as “a force cobbled together from a small number of state troops, the hiring of mercenary bands, and the incorpo- ration of private armies raised by major aristocrats and put at the ruler’s service”8—gave way to a recognizably different style, that of the “state commission army”. The state-commissioned armies which dominated in Europe after 1650 were both more national in composition and more uniform, as well as much bigger than the mercenary armies had been. How- ever, in his recent work, Lynn has recognized that in some countries (like the Dutch United Provinces with their small population but well-filled coffers), mercenaries remained very important after that date. In fact, the 18th century was the heyday of the system in which smaller German states hired out their regiments to richer, more powerful, states in exchange for “subsidies.” As is well known, the British fought their wars in North Amer- ica partly with Hessian regiments, acquired in exchange for subsidies. Although the dividing line of 1650 has kept its validity, the exceptions show that in fact army styles rarely occur in a pure form. Like Max Weber’s concept of bureaucracy, they are “ideal types.” In the real world, armies were composite bodies with different army styles coexisting at the same time. Mercenaries continued to play a role in the state commission armies of the 17th and 18th centuries, even if they no longer dominated. And, as Hippler shows, the army of the ancient regime also included conscripted soldiers, long before the formal introduction of conscription during the French revolution. Lynn’s classification is both convincing and useful as an analytical tool, but it should be recognized that it is based on European history only. Several “army styles” that have been extremely important in Asia and the Middle East in the early modern period therefore are not included. The first that comes to mind is that of the “slave army”. For a thousand years, from the early 9th century to the early 19th, mamluks—soldiers bought

7 John A. Lynn, “The Evolution of Army Style in the Modern West, 800–2000,” Interna- tional History Review, 18, 3 (August 1996), pp. 505–45. 8 Lynn, in his commentary on the Fighting for a Living project, p. 2. fighting for a living in europe and asia 175 as slaves by rulers outside their own realm, and regarded as their private possessions—were a prominent feature in the military from Algiers to India. The janissaries of the Ottoman Empire clearly belonged to the same category of slave troops, although they were levied within the Ottoman domains, and not purchased abroad. The second style is that of the tribal forces, which were no longer an important feature of warfare in early modern Europe, but remained important to the Ottoman, Mughal and Chinese empires until the end. Naturally, a classification of army styles pertains to forms of military organization, and not to labor relations. The temptation is great to assume that the different army styles coincided with different forms of labor relation; but, as the case studies in our project have shown, the two do not necessarily coincide. While one type of labor tributary relation seems to dominate in what we may perhaps broadly call “feudal” armies, the studies show us that in a state commission army (or a conscript army) different labor relations can very well co-exist within one single “army style.” A single type of army may be home to very different types of labor relation. On paper, the French army of Third Republic was a conscript army, in which citizens exercized their right and duty to defend their fatherland; but like all 19th century conscription systems, the French one enabled its more affluent citizens to pay for replacements, an opportunity they took up on a very large scale. As most of the people who were available as replacements were soldiers who had served their turn (but who, because of their long service in the army, had little chance of a job in civil society), the conscript army in fact consisted for a considerable part of veteran professionals. The Italian case, studied by Marco Rovinello, reflects the same reality: the state was quite happy that the extra income from bourgeois “liberations” (exemptions from military service) allowed it to recruit veterans to beef up the army.

Everywhere and always, the officer corps was treated very differently from the rank and file, and had its own set of labor relations. So, we are faced with different army styles that succeed each other, but in part also overlap, and we also note that within a single identifiable army style, a variety of labor relations is possible. The twenty cases studied in the context of the project in total, yield about one hundred different forms of labor relation. As I will discuss below, there are several reasons for this. One of them is that in many places smaller forces of “experts” coexisted with the mass of the main army. From the European Landsknechte and the Albanian 176 erik-jan zürcher cavalry to the Ottoman and later Portuguese artillery, and from experts in the Moghul army to the French officers of Mehmed Ali Pasha’s new Egyptian army—everywhere armies have felt the need to employ high- skilled specialists for specific tasks. The 17th century Swedish army of Gustavus Adolphus II, offers a very good example of the coexistence of different army styles and labor relations within a single institution. In many ways the most modern army of its day, the Swedish force rested on Europe’s oldest conscription system; yet, at the same time, the Swedish king was one of the biggest employers of mercenaries in Europe. Another complicating factor is, that different types of labor relationship sometimes figure on different levels of a single system. In analogy to food chains or commodity chains, one could perhaps speak about “recruitment chains.” An early modern European state may contract a mercenary colonel, who will then conclude contracts with officers, often from the nobility, who will bring peasants with a tributary relationship to their lord into the army from feudal estates. In the Ottoman army of the 19th century and right until World War I, Kurdish tribal chiefs were given officer rank and placed in the army hierarchy; but it proved impossible to impose regular army discipline on the Kurdish units commanded by these officers, because the rank and file only recognized tribal allegiance, not the hierarchy of the army. Theirs was a reciprocal mini-system within a tributary (because conscription-based) whole, with free commodified labor (the officers) at the top. In order to create a basis for comparative analysis of all these forms of military labor relations, we need to find a common language to describe the phenomena; one that is not bound up exclusively in the historical development of one of the regions studied. In other words: rather than busy ourselves overly with the question whether the Timar-system of the Ottoman Empire or the Mansabdari-system in Mughal India is a type of “feudalism” (which is, after all, a term from European social history), we should simply recognize that, for hundreds of years, states have been in need of a form of military service where soldiers (mostly, relatively expensive mounted warriors) were remunerated with land (or the usufruct of land) in exchange for exclusive service to one court or state. Once we have learned to look at the different forms of military labor in terms of their common features rather than in terms of all the differences, we then need to establish a taxonomy in which all the different forms of military employment that have occurred in the different areas over a period of five hundred years can find a place. For this, we can have recourse to the basic threefold division of labor relations that has been fighting for a living in europe and asia 177 developed earlier in the IISH’s Global Collaboratory on Labor Relations 1500–2000: reciprocal labor, tributary labor and commodified labor.9 Using this taxonomy, the two questions our research group has tried to answer are:

1. how can we explain the predominance of certain types of labor relations, and combinations of labor relations, in certain circumstances? and: 2. how can we explain the replacement of one dominant system by another?

The very empirical richness which makes military labor such an attractive subject to a social historian means that it is as yet too early to give a definitive answer. In this essay, I would nevertheless like to present some preliminary findings that have resulted from the project. My aim here is not to give a complete overview of the cases and their relevance, but to illustrate the main findings of the project with examples taken from a number of studies, in order to give a sense of what is possible with this kind of comparative approach—both in terms of testing the usefulness of the basic taxonomy (reciprocal/tributary/commodified) and in terms of finding explanatory factors for the dominance of a particular system of army recruitment and army employment, or the change from one system to another.

Reciprocal, tributary and commodified labor

To determine where a particular form of military labor should be placed in this taxonomy, we consider the following variables as criteria:

9 Of course, one might argue that in addition to the three broad categories outlined above (reciprocal, tributary and commodified), volunteerism should be included as a fourth variant. There are however, several reasons why we prefer not to do so. First of all, and except for individual cases, it is almost impossible to obtain accurate information about people’s motivations to join the military. Representatives of the state and commanding officers may grossly misrepresent people’s mindset. Even if the soldiers themselves are literate, and write about it in ego documents, there is no telling how accurate this information is. It may be a rationalization or self-justification. Tradition, economic need or social pressure may force people to volunteer, and in some cases (as during the Vietnam War) volunteering may even be a stratagem to avoid being drafted, and so acquire a better, privileged position with the army. It is not the voluntary character of the “all-volunteer force” of the United States introduced in 1973 that is relevant for us, but its all-professional nature, and the fact that it can be seen as a form of free and commodified labor. 178 erik-jan zürcher

• income (wages or fees, high or low, coin or kind, regular/irregular) • duration of service (short-term contracts to life-long employment) • Legal constraints (freedom to enter or leave the system, to change employers) • social and cultural constraints (inclusion or exclusion on the basis of ethnicity, gender or identity)

The reciprocal form of labor relation perhaps figures least in our studies. Nevertheless, we see references to the use of tribal forces by the Ming emperors in China and by the Ottoman sultans. The states of Hindustan often had recourse to Afghan tribal warriors. Whether the Eight Banners of Ch’ing China, the original Manchu tribal forces, represented reciprocal or tributary labor seems debatable. Perhaps the one was succeeded by the other, as the Manchu tribal chiefs acquired their new status of Chinese emperor, and old tribal allegiances were given a place in the Chinese imperial order. Local militias very often were also based on reciprocity: there was a pre-existing mutual obligation within closely knit communities to share the burden of defence. But when state, or “national” armies were built by incorporating these militias into centralized structures commanded by professional officers—as we see in Ming China or Ancien Regime France, but also during the American War of Independence— militias evolve into a kind of primitive conscription system. We can also argue that, at the lowest level of early conscription systems like those of 17th century Sweden or 18th century Russia, a degree of reciprocity, an equal sharing of burdens and benefits within the community, was involved. The local village community, charged with delivering recruits under the supervision of the landed aristocracy, spread out the burden of conscription in much the same way it shared out the use of common lands, or the obligation of agricultural labor. We can also note that followers of military leaders who themselves had a contractual relationship with the court or the state, were tied to these leaders through bonds of kinship or patronage, and that, because of this, their labor relations with their commanders were of a reciprocal nature, even if this relationship was itself part of a larger system in which other types of labor relation (the free commodified labor of the mercenary) dominated. The Scottish mercenaries quite often seem to fall into this category, but, as Herman Amersfoort notes, Swiss mercenaries in early modern Europe often had kinship ties with their recruiters as well. The large majority of military labor relations and recruitment practices surveyed in our project fall into one of the other two categories of the IISH fighting for a living in europe and asia 179 collaboratory on labor relations: tributary or commodified. Tributary labor occurs when the official position of the state is that military service is an obligation that can be legally imposed, and that is essentially interchange- able with the categories of tax and corvée—other obligations imposed by the state. The precise form that the tributary labor relationship takes on can vary from legal enslavement (as in the Ottoman devşirme) to levies for specific campaigns, and early and modern forms of conscription. In levies and early forms of conscription, the obligation is typically imposed on a community (the military households of the Ming or the Russian mir); in modern conscription systems, it is essentially an individual duty incum- bent on the citizen. In some of the state commissioned armies of 18th century Europe, like the Hessians, which at first sight seem to be composed of volunteers, an important part of the soldiery was in fact indentured, serving in the army to pay off the family’s tax arrears. Tributary and reciprocal forms intermingle in the case of tribes that have a tributary relationship with, for instance, the Mughal or Ming empire, but which mobilize their own tribal warriors on the basis of reciprocity.

The second, quite common category is that of commodified labor. This seems to be the category into which both the aggregate contract army and the state commission army of Lynn’s classification fall. Typical for these categories is a contractual relationship for a limited time between the court (or state) on the one hand, and the military on the other. Both the modern volunteer army (like the all-volunteer force of the United States studied by Beth Bailey) and the contractors operating in Iraq or Afghani- stan, on whom Yelda Kaya reports, fall into this category as well. The question of pay does not in itself determine in which category (tributary or commodified) the different forms of military labor should be placed. It is true that defining military service as a “duty” analogous to the payment of taxes allows the state to escape from the need to compete in the labor market (and therefore to offer competitive wages). But the Chinese empire, for example, did pay its garrison troops during the Ming era, even if these troops were recruited from members of hereditary military households obliged to produce soldiers; and the Ottomans paid their janissary troops handsomely, even if legally they were the Sultan’s slaves, and the members of the corps had originally been levied as a form of tax- in-kind in Christian Balkan villages. On the other hand, mercenaries and state-commissioned armies were often paid badly. Mercenaries could be compensated by giving them the right to pillage (about which I will say more later). Once armies grew in size and permanence (something that 180 erik-jan zürcher seems to have happened in China five hundred years before it happened in 17th century Europe), however, states were forced to allow soldiers to pay their way (and earn a living). The soldiers did so either by doing non- military labor for the state (road repair being a popular option, all over the world) or by producing goods for the market. Otherwise, these mass armies would simply have been unaffordable. In this respect, the Russian example is very clear: the soldiers of the Russian army nominally were full-time soldiers, but they were allowed to do productive work, and even exploit workshops and farms of their own while they were garrisoned. Standing armies like the state-commissioned armies in early modern Europe, the Ottoman janissary garrisons or the military households of the Ming, could (and in fact had to) reduce their costs by allowing soldiers to become part-time producers. The garrison troops of the Ming military households spent most of their time in agricultural labor, not on military duties. In cities like Damascus, Aleppo or Cairo, janissaries very often became co-owners of shops in the bazaar. When discussing the remuneration of soldiers, it is important to include the long term effects as well as the immediate reward. Some of the most valuable elements of remuneration may be in the shape of future rewards like land, pensions or (in the modern state) insurance, educational opportunities for the soldiers themselves or their children. This is true of civilian labor as well, of course, but armies have often pioneered this kind of remuneration scheme. Basically, the state has three options in the way it remunerates its soldiers: through the apportioning of land, or the usufruct of land; through cash payment or payment in moveable goods; or through granting rights, notably the right to pillage. The granting of land or usufruct always was a popular option for cash-strapped states. It had clear advantages for societies with low levels of monetization, and it seems to have been the preferred option for relatively expensive cavalry forces and commanding officers in most societies. Generally speaking, the rank and file were paid in cash or kind. Only in India do we see remuneration with land also for the lowly soldier. According to Kaushik Roy, this had to do with cultural prejudice, which saw “fighting for money” as something essentially dishonorable. Cash payment became more widespread after the flow of silver from the Spanish Americas started, but it was primarily an attractive option for states which achieved a high degree of centralization and huge powers of extraction (the Chinese Empire being in a class of its own in this respect), or states with highly developed credit- and banking systems, like the Dutch Republic or Britain. For most early modern European fighting for a living in europe and asia 181 states, but also for the Ottoman Empire, raising the cash for the expanding armies of the 17th and 18th Centuries was, of course, notoriously difficult. One way of overcoming the problem was by allowing soldiers to raise their income, by granting them a “right to pillage.” As both John Lynn and Frank Tallett show, for the mercenaries in the armies of the Thirty Years War this source of income was far more important than their nominal wage. In addition to regular pay, there are many examples of “bonus and incentive systems” in the form of rewards for valour in battle, for enemies killed, for signing up or extending one’s service. Virginia Aksan quotes the memoirs of an Ottoman soldier of the early 19th century, who confesses to cutting off the heads of unsuspecting Christian villagers with the intention of handing them in as proof of the number of enemies he killed. When subsequently he encounters a company of janissaries, the first thing they do is rob him of his heads . . . The incentive, both for him and for the janissaries, is not “bloodthirstiness” or “fanaticism,” but simple material gain— the heads represent a source of extra income in the form of a bonus. Perhaps a distinction should be made between the right granted to soldiers to live off the land and exact “contributions” from the population— especially in enemy territory, which can be regarded as a form of regular income—and the right to pillage, for instance after the taking of a town, which, because of its unpredictable nature, can more properly be regarded as a “bonus” or “incentive.” Throughout the period studied there seem to have been huge differences in remuneration between officers and troops both in Europe and Asia, but also between the well-trained professionals that were hired for their expertise (and who on the whole were much smaller in numbers) and large masses of peasant soldiers with only basic skills. With their strong corporate identity and hierarchy based on skill and experience, the professional mercenaries of early modern Europe, the Household Men of Ming China and the Ottoman janissaries can perhaps best be compared to guild members and artisans. The soldiers of the mass armies raised in 18th century France or Prussia, the levends of the Otto- man Empire, the garrison troops of the Ming, or the Green Standard forces of the Ch’ing can be more usefully compared with unskilled labor. The evidence seems to show that pay levels for this type of soldier were fairly consistent with wages being paid at the lower end of the civilian labor market, both in Europe and in Asia. Where soldiers were recruited in the labor market, the army generally seems to have been an employer of last resort, as shown by the fact that recruitment generally was easier in times of economic crisis, when it was hard to find other jobs. 182 erik-jan zürcher

The problem of determining whether soldiers in the different armies can be classified as “free” or “unfree” labor is complex. Very few soldiers in history have been legally completely free actors, in the sense that they could terminate or change their employment without being subject to persecution under criminal law. In almost every country, joining the army altered people’s legal status. In most cases, this restricted their freedom, but in the case of Russia the opposite was true: conscription turned serfs into free men, albeit free men subjected to military discipline. His- torically, the prototypical Marxian “free worker” seems to have been the exception in the world of the military, just as in many other fields. In his essay “Who are the workers?” Marcel van der Linden has argued that “there is an almost endless variety of producers in capitalism, and the intermediate forms between the different categories are fluid rather than sharply defined.”10 He provides examples of slaves working voluntarily for wages part of their time, and he also points out that “free” wage laborers have at times been locked up by their employers, a practice that in countries like China or India is still a regular occurrence. Our research seems to confirm the truth of this statement. Members of aggregate contract armies undoubtedly come closest to the status of free worker. In theory, they were free to choose their employer, which gave them some negotiating power, and their contracts were of limited duration. But even they were subject to articles of war, once they had signed up and received their bonus. Appearances can be deceptive: the Ottoman janissaries were techni- cally the possession of their sultan, but they had accumulated traditional rights which they guarded jealously, much like a guild. Many of the janissary mutinies that occurred from the 17th to the early 19th century started as “industrial actions” or pay disputes, when they interpreted government measures as unjustly transgressing on their acquired rights.11 On the other hand, the soldiers who signed up of their own accord, as free men, for an 18th century state commission army were faced with draconian regulations and frequent physical abuse (which, in armies like the Prussian one, could be just as bad as what plantation slaves had to face).

10 Marcel van der Linden, “Who are the Workers?,” in Marcel van der Linden, Workers of the World. Essays toward a Global Labor History (Leiden, 2008), pp. 17–38. 11 See Annemarike Stremmelaar, “Justice and Revenge in the Ottoman Rebellion of 1703,” Leiden: unpublished Phd dissertation, 2007. fighting for a living in europe and asia 183

When judging conditions of service, whether in terms of pay or in terms of the free/unfree polarity, we should always take into account contemporary conditions in society at large. Conditions of service that may seem “unfair” or even “atrocious” in our eyes may have looked very differently to a Scottish day laborer, a Russian serf or a Hindustani peasant. The status of free actor in the labor market, enjoyed by a European mercenary or a Rajput military leader, is historically the exception rather than the rule, and that is true for the military profession just as much as for society at large.

Determinants: general considerations

Having established the place of the different forms of military labor in the taxonomy on the basis of the variables of remuneration, term of service, legal status, and relationship with court or state, we can try to gauge which factors influence the choice of the state for a particular form of military employment. All forms of military recruitment and labor represent different solutions to shared problems. We therefore ought to look first at the basic problems and aspirations of the people and the state. As a rule, people like to be free from interference in their activities. Outside the ruling elite, they are fully occupied by their daily concerns to make a living, to preserve their health, to protect their children and, in the more dynamic societies, also to gain advancement or amass wealth. They are prepared to defend their homes and families, and throughout recorded history they have also shown themselves ready to defend the larger community of which they perceive themselves to be a part: the village, the town or the tribe. Of course, history is also riddled with instances in which people have united in much larger, more anonymous groups to fight in a “cause”: the crusades in mediaeval Europe, rebellions like those of the Celalis in the early 17th century Ottoman Empire, or the Tai Pings in 19th century China. Some- times hundreds of thousands, even millions, of people have taken part in these armed movements, but two characteristics distinguish these movements from the kind of organized violence we discuss in our project: they are generally short-lived (even the longest lasting only about fifteen years, most of them much shorter) and at least at the start they were spontane- ous. In our project, we deal with military systems established by states for the longer term. States therefore have a problem: It is very difficult to get people to devote themselves exclusively, predominantly or permanently to fighting, 184 erik-jan zürcher killing and dying in the service of distant entities like courts, or abstract notions like “the state” or “the nation”. Yet this is exactly what princes and states need. Faced with the need to raise soldiers, states have a basic choice between two options. To put it in Gramscian terms, they can either coerce people into serving, or persuade them to do so through the establishment of an “hegemonic cultural code,” in other words: to create a measure of consent. Chris and Charles Tilly make a threefold distinction between compulsion, reward and persuasion, but the demarcation line between the last two categories is vague.12 Both coercion and consent have obvious advantages and disadvantages, which are wellknown from the debates about “slavery versus wage labor.”13 At first sight, the hiring of professionals, both in the form of mercenaries and standing armies, may seem the more expensive option, because it makes high demands on the state’s ability to pay, and often forces the state to compete with other employers in the labor market. But coercion is also expensive because of the requirements for forcible recruitment and constant supervision after soldiers have been recruited. Like slaves, coerced soldiers may also be less well-motivated or “productive” than those who have joined the colors out of their own free will. On the other hand, coercion allows the state to escape the need to compete in the labor market. It does not have to entice people to become soldiers with “sign up bonuses,” nor does it have to pay wages in conformity with market rates. Ultimately, what is the decisive factor may not be cost, but “value for money,” or in other words: cost effectiveness. It is extremely difficult though to introduce the concept of “productivity” into discussions on military labor. After all: what is a soldier’s productivity when he is engaged in his core business of fighting and killing? Is it the measure of destruction he manages to inflict on the enemy? Or is it the degree to which his activities help to enlarge the tax base of the state through conquests, or advance the economic interests of the elites that control the state? Whatever the definition, it seems to be the case that courts and states historically are looking for the army that is most effective on the battlefield at the lowest possible cost (that even this “lowest possible cost” can still be crippling to state and society alike is another matter). However, there can be a huge difference between the immediate costs and the long-term financial burden. Both early

12 Tilly and Tilly, Work Under Capitalism, pp. 87, 134, 259. 13 Stefano Fenoaltea, “Slavery and Supervision in Comparative Perspective: A Model,” Journal of Economic History, 44, 3 (September 1984), pp. 635–68. fighting for a living in europe and asia 185

17th century mercenaries and early 21st century contractors have been expensive in the short run, but they were and are easily dismissed at the end of the conflict. By contrast, standing state commission armies were a continuous drain on the treasury, and the modern all-volunteer forces bring with them huge long-term obligations to the soldiers and their families. The choice made by different states at different times is influenced by many more factors than economic or financial ones alone, however. If maintaining a monopoly of violence (or, to put it more realistically, getting as close as it can to a monopoly of violence), is a central function of the state, states may face a very real dilemma: they may spawn a powerful military which they cannot keep under control themselves, and that could threaten the established order. This is just as true for the state that recruits highly specialized military experts (like the Mamluks of the Middle East or the Turks and Afghans of Hindustan) as for the state that recruits mass armies via conscription from a population which is denied any share of political power (as in the cases of Prussia and Russia). Apart from this kind of political consideration, ideological considerations or cultural prejudices may play a part. The Ottoman decision to exclude non-Muslim citizens from the conscription system (a decision that cost them up to 40% of their total pool of manpower) is a case in point, but so is the commitment to general conscription of the late 19th century French Republic and the Kingdom of Italy, which was informed by notions of patriotism and nation-building. As Jörn Leonhard shows, the rejection of conscription in Great Britain was influenced both by the Whig interpretation of history, which saw large standing armies as “instruments of tyranny” and essentially “un-British,” and by an idealized view of the army as representing “traditional country values,” with aristocratic officers and a sturdy peasantry for soldiers. When surveying the different case studies in our project, we are struck by a number of characteristics that seem to be almost universal. One is, as noted before, that we always see the coexistence of different types of army style, and different forms of recruitment and labor relations. Change may be sudden, but is rarely absolute (the transition of the U.S. army from conscription to an all-volunteer force in 1973, followed by similar transitions in most NATO countries, seems to be the exception that confirms the rule). While it is true that war nearly always brings with it some degree of change, innovations triggered by defeat in war are very often introduced side by side with the continued existence of older forces, which remain important even if they are obsolete and have lost their credibility on the battlefield. The Ottomans kept their Sipahi forces in being for 186 erik-jan zürcher at least two centuries after their military usefulness had ended, and the Chinese empire seems to have been equally conservative. Military corps were, after all, in an excellent position to defend their vested interests, especially when garrisoned in major cities or even the capital. This is one reason that both the Ming and the Ottomans, when they started hiring, or levying, mercenary troops, left their obsolete formations (garrison troops and household troops in the case of the Ming, Cipahis and Janissaries in that of the Ottomans) in place. Another, almost inverse reason that also seems evident in both these cases is that a military system, even when obsolete on the battlefield, can still be an important element of control inside the country, not just in “law and order” terms, but also in ideological terms. The concept of military households was important to the Ming as a vital element in its social order, just as the concept of a “military class” (askeri) was upheld by the Ottomans. The continued reliance of the French state on its nobility for the recruitment and officer material of its army even after that nobility had lost its autonomy, can be interpreted in the same sense. It is very clear that there is no teleological sequence; there is no single process of armies progressing from one “stage” to another, along some developmental or modernization path. There was strong ideological resistance in the Anglo-Saxon world against the idea of conscription, which was closely identified with tyranny. Therefore, although it became universal in 19th century Europe, this system was not introduced in Britain until a century later, and then only temporarily. A century later again, the reduction of the armed forces of industrialized countries after the end of the Cold War, combined with a glut of arms and officers (caused in part by the end of the Warsaw Pact, and in part by the ending of Apartheid in South Africa), prompted a resurgence of mercenary forces in the form of “contractors” like Blackwater as a major component in the military campaigns of NATO countries. A few decades earlier, when mercenaries played a role only in the post-colonial conflicts in Africa, the resurgence of a type of military labor declining since the 17th century was not foreseen by anyone. Nevertheless, in particular periods of history, certain systems definitely become dominant, while others fade. As John Lynn noted, the mercenary did not disappear after 1650, but in Europe the state commission army did become the norm. In the Middle East, the janissaries remained operative until 1826, but irregular levies had become the mainstay of the army. After 1815, many restoration regimes, like those in the Netherlands or Italy, rejected conscription as a revolutionary legacy, but in later decades fighting for a living in europe and asia 187 conscription became dominant throughout Europe and the Middle East. An important question is: what were the factors determining these changes?

Determinants: a preliminary survey

(a) Manpower and money. The availability of people and money seem to be the most important determinants. Let us first look at demographics. Both the Chinese and Indian experience is determined first and foremost by the availability of an enormous, and seemingly unlimited pool of manpower. This gave the Mughal Empire the opportunity to raise vast peasant armies, and the Chinese Empire to raise armies that were of a different order of magnitude altogether, when compared with their European, South Asian and Middle Eastern counterparts. The Chinese manpower pool was clearly unique, looked at from a global comparative perspective. As Beth Bailey shows, the transition from a conscript army to an all volunteer force in the United States was also very much the result of demographic development, i.e. the baby boom. 16th and 17th century Scotland had a very small population, but relative to its size it had plenty excess labor used to handling arms. When population growth stagnated in the 18th century, the recruitment of Scots by the British army became a problem. The manpower demands of the army in the 19th century meant that the British had to start recruiting in the urban centers of England, rather than in the countryside and among the Scots and Irish, in spite of strong objections to the enlisting of urban riff-raff in the army. The second factor is money. Where labor markets were tight, states essentially had only two ways to strengthen their armies: either by using more coercion (isolating groups of people from the labor market, which also carries a cost), or by improving the relative position of army careers in the labor market (with higher wages or other benefits). Coercion is much in evidence, and here too we see recurrent patterns in a number of cases. Although not used as frequently and brutally as in the case of the navy, the “press” or similar systems were used by British, German and Otto- man authorities to get rid of socially undesirable elements, which usually meant vagrants and beggars, and more generally, men without property, protection or regular work. It is hardly surprising that armies complained time and again about the quality of the personnel that was provided to them in this way. As Robert Johnson shows, it meant that well-trained native troops in the East India Company army, who were essentially 188 erik-jan zürcher volunteers, were considered much better than the soldiers shipped out from the mother country. Higher wages were a difficult option for the states. Financing the troops was a continuous problem for most states, certainly in Europe and the Middle East. This is true as much for the Habsburgs (who during the Thirty Years War became dependent on a new breed of general contractors that provided credit as well as an army), as it was for France in the late 17th century, or the Ottomans in the 19th. The Dutch Republic was on the opposite side of the spectrum. In spite of its small population, which was averse to military service because there were more profitable opportunities in the Dutch labor market, the Dutch managed to raise sizeable aggregate contract armies because of their financial strength and advanced banking system. United under the Ming and Ch’ing dynasties, the Chinese empire was able to extract such gigantic quantities of silver and grain, that it could provide for its armies despite their enormous size (i.e., between five and ten times that of the biggest European armies).

(b) Technology. As mentioned above, most states were constantly on the look-out for “the biggest army at the lowest cost” to the treasury. But the army had to be effective as well, which meant—and still means— technologically at the state-of-the-art and operationally reliable. Many of the most far-reaching changes in army recruitment were due to the desire to apply the harsh lessons learned in war (primarily through defeats) and to emulate more successful competitors. As Frank Tallett has shown, it was not necessarily a question of technological innovations (in the sense of hardware) and, more often, a matter of “social technologies”—things like new forms of discipline, training and institutional structures. This seems to have been a decisive factor in the lengthy Austro-Ottoman wars of the late 17th and early 18th century, as well as in the success of relatively small European colonial forces all over Asia. Ultimately, this led to the adoption of Western-style discipline, with uniforms and drill, in Egypt, the Ottoman Empire and China. Changes in military technology, financial constraints and the size of the available labor pool undoubtedly were the most important factors determining the choice for a specific form of recruitment and military employment, but other considerations also played a role.

(c) Politics. Political considerations were always important, because bal- ancing the need for a larger army against the need to maintain control over those who could provide it or finance it (in the early modern states), fighting for a living in europe and asia 189 or the need to manufacture or maintain consent among the public (in modern states) have always been high on the agenda of those in power. One of the major reasons behind the widespread use of (sub-)contractors in Iraq and Afghanistan was that it eases the state’s need to mobilize public support for its own actions.

(d) Ideological and cultural factors determining who should fight or should be excluded from the bearing of arms are also prominent. In the eyes of the restoration regimes after 1815, conscription was so much bound up with the revolutionary period, that they preferred to fall back on state- commissioned professional armies and militias (as Amersfoort shows for the Dutch case). For the French Third Republic, conscription as an expression of “citizenship,” and as the supposed legacy of the great revolution became an issue of almost mythical proportions, as Thomas Hippler demonstrates. As I noted earlier, the refusal of the Ottomans to conscript non- Muslims severely limited their manpower base until 1909.

(e) Popular cooperation and resistance. The analysis so far has concentrated almost exclusively on the needs and actions of the state. But we should not, of course, imagine that the people targeted by the state’s interventions were merely passive; they have always had agency as well. The people have a repertoire of options open to them just as much as the state does. Of course, they can “comply” with the demands of the state, but this “compliance” does not necessarily have to be “acquiescence”. People can see the army as an opportunity structure, offering them chances of social advancement, improving their living standards, or simply providing the possibility to travel and see more of the world beyond their own village or valley. Rovinello shows that this was a real factor among Italian recruits in the 19th century. The reality that the state is in need of manpower to fill the ranks of its army also enables people to instrumentalize military service for their own ends, or even to domesticate it. Whether in Asia, the Middle East or Europe, there is good evidence that the army was rarely a popular employer, at least among the rank and file. The army was often an “employer of last resort.” But even so, when work was scarce, when harvests failed, or when industries went through a slump, the army offered low but regular pay, food and lodging—in other words, security. On the other hand, people may also resist. But the repertoire of resistance is also varied. At first, there is the tendency to avoid service altogether. Conscription systems, old and new, just like enslavement, have generally been deeply unpopular. Privileged sections of society have generally been 190 erik-jan zürcher able to make use of exemptions, and both communities and local authorities seem to have done their best to make sure that “undesirables” (who were unproductive and might otherwise create “unrest” in society), were recruited into the army. Other forms of avoidance were: going into hiding or self-mutilation, which, according to Khaled Fahmy, was especially widespread in 19th century Egypt. Once in the army, both desertion and defection became options, even if sometimes highly dangerous ones. The ultimate form of resistance was the mutiny. “Industrial action” by its own armed forces was, of course, the most serious crisis which any ruling elite could face. There seems to be some evidence that groups with a strong corporate identity as “artisans of war” (like mercenaries or janissaries, and troops raised from an urban source) seem to be more prone to mutiny. Peasant armies seem to be more prone to desertion.

A final word

What the project has revealed is that there is, on the one hand—to para- phrase van der Linden—an almost “endless variety” of military workers in history, but also that we can develop a taxonomy grouping all these different forms from far-flung countries and periods in categories defining shared characteristics, and to do so in a meaningful way. When we combine the classification thus achieved with a set of the most important determinants, we can discern a number of patterns and reach tentative conclusions about the circumstances that influence the choice for a certain type of recruitment and a certain form of military employment. It is hoped that, alongside similar research conducted at the IISH about industries that offer opportunities for comparative research because of their global nature (textiles, docks, prostitution), this study of military labor will increase and enrich our understanding of labor relations, a topic so close to Jan’s heart. White Collar Workers of the VOC in Amsterdam, 1602–1795

Karel Davids

Introduction

Labor historians have generally been more concerned with the experience of manual workers than with salaried employees. Blue-collar workers have been better studied than white-collar workers. In their synthetic overview of the Dutch economy between 1500 and 1815, Jan de Vries and Ad van der Woude pointed out “daunting problems” in interpreting the evolution of earnings of salaried employees, as compared to those of wage laborers. The “heterogeneity of occupations,” “salary variations among individuals within an occupation” and the range of possible combinations between incomes from various sources—to mention but a few of these difficulties—made it hard to “interpret findings [about salaried workers] with confidence,” they declared.1 Nevertheless, they presented some intriguing evidence about the evolution of salaries and differentials among wage earners, to which I will return further on in this essay. The evolution of salaried incomes also remains an underresearched topic in the long-running debate about living standards in Britain during the Industrial Revolution. One of the most substantial and innovative contributions on the subject is H.M. Boot’s case-study of the earnings of clerks in the home service of the British East India Company (EIC) between 1760 and 1850. Boot observed that, except for a few relatively brief periods, salaried employees enjoyed a higher and more sustained rise in income than wage workers. The explanation he offered for this trend was, that “the growing demands of expanding trade and a newly acquired empire generated a burgeoning demand for administrative, management, and clerical skills of all kinds,” which exceeded the supply of trained people, leading to “salary gains far in excess of those in other parts of the economy.” In these conditions, the East India Company developed an age- (experience-)related salary structure, which offered white collar

1 Jan de Vries and Ad van der Woude, The First Modern Economy. Success, Failure, and Perseverance of the Dutch economy, 1500–1815 (Cambridge, 1997), pp. 617–9. 192 karel davids workers possibilities to make a career, and increase their earnings during their lifetime.2 Boot’s suggested explanation raises the question whether a similar trend in earnings can be found among salaried employees of the EIC’s counterpart in the Netherlands, the Dutch East India Company (VOC). After all, the Netherlands in the 17th and early 18th centuries, for similar reasons as Britain after 1760, must have seen a vigorous growth in demand for “administrative, management, and clerical skills of all kinds.” As a follow- up issue, the question arises whether the VOC just like the EIC developed an age-related salary structure, and whether white collar workers of the VOC enjoyed similar opportunities to make a career, and earn more in their lifetime like their colleagues in Britain. This essay therefore deals with the experience of a particular category of white collar workers in the early modern Netherlands, namely the administrative personnel of the VOC, and more specifically the bookkeepers, cashiers and clerks employed in the offices, storehouses and yards of the Amsterdam Chamber. Focussing on a single category of employees may help, I would suggest, to negotiate some of the “daunting problems” in studying the history of salaried workers in early modern times. To interpret the findings, I will draw, where appropriate, brief comparisons with other groups of workers. Apart from the clerks of the English East-India Company studied by Boot, and an amalgamated class of salaried workers discussed by De Vries and Van der Woude, I will also briefly look at other groups of clerks and bookkeepers in Amsterdam, at clerks in the state’s service in The Hague, and at other categories of personnel employed by the VOC.

Growth of a white collar workforce

The growth of a white collar workforce in the Dutch East-India Company was closely linked to the expansion of its maritime and commercial activities. The expansion of shipping and trade led to an increasing demand for administrative support; the growth of the administrative staff, in its turn, aided the further growth of the maritime and commercial activities. To understand the experience of the white collar workers of the VOC, we

2 H.M. Boot, “Real incomes of the British middle class, 1760–1850: the experience of clerks at the East India Company,” Economic History Review, 52 (1999), pp. 638–68, esp. 638, 643–7, 659–62. white collar workers of the voc in amsterdam 193 therefore first have to take a brief look at how the Company, in quantitative terms, developed as a shipping and trading concern.3 The graph presented below gives a succinct overview of the evolution of its maritime and commercial activities, with several indicators that indirectly also tell us something about the volume of work involved for the administrative staff at home: the number of ships sent to Asia, the number of people leaving for the East and the sales of imports from Asia in the Netherlands. The most striking phenomena in this graph are, of course, a (nearly) continuous rise in the number of ships sent to Asia until the 1730s, an (almost) continuous growth in the number of people on board until the 1760s and a huge expansion in sales, which, apart from a brief relapse in the thirties and forties, continued up to the 1760s as well. Thus, the size of maritime and commercial activities of the VOC did not really contract until the very last decades of the 18th century. The expansion of the maritime and commercial activities illustrated in figure 1 was, at a distance, followed by the build-up of an administrative staff, particularly in the Chamber of Amsterdam. As table 1 shows, the number of bookkeepers, cashiers and clerks in Amsterdam rapidly rose in the middle decades of the 17th century. The “steady growth of the workload,” which the directors acknowledged as early as 1629,4 first led to a revised division of labor among existing personnel and then, from the mid-fifties onwards, to a substantial increase of the number of employees, especially clerks. While there were only a few clerks around about 1630, their number (including the First Clerk) grew to nine in 1656, eleven in 1662 and fourteen in 1690 in the copy office (schrijfcomptoir) alone; another four were employed in the storehouse and on the shipyard.5 As the size of the staff increased, the division of labor became more and more elaborate. By the 1720s, the staff had branched out into a large number of offices and positions: apart from the office of the Chief Bookkeeper with his two assistants and the copy office directed by the First Clerk, where most of the clerks worked, there was a cashier’s office (ontvangkamer), a pay office (soldijcomptoir), a storehouse office (pakhuis), a shipyard’s office (timmerwerf ), a bookkeeper for mustering (monstering), a bookkeeper for

3 See Femme S. Gaastra, De geschiedenis van de VOC (Leiden, 2002). 4 Nationaal Archief The Hague (NA), Archief Verenigde Oostindische Compagnie (VOC) 230 res. Chamber Amsterdam 5 March 1629. 5 NA VOC 307 under “Clerks,” esp. 5 May 1656 and 16 February 1662, Pieter van Dam, Beschryvinge van de Oostindische Compagnie, Eerste boek, deel I (ed. F.W. Stapel, The Hague 1927), pp. 395–6. 194 karel davids

2500

2000

1500 Number of ships Number of people (x 100) 1000 Sales (in ., x 10,000)

500

0 0 710

00‒1 50‒176 1602‒1610 1650‒1660 17 17

Source: Gaastra, Geschiedenis van de VOC, 82, 115, 147 Figure 1. Shipping and sales of the Dutch East-India Company, 1602–1795

Table 1. Number of bookkeepers, cashiers and clerks at the Chambers Amsterdam, Zeeland and Delft, 1630–1790 Year Amsterdam Zeeland Delft 1630 11 1668 24 1690 31 3 1712 4 1726 31 1730 35 13 4 1737 16 1762 41 4 1790 44 21 Source: Eekhout, “Een bedrijf,” Enthoven, “Veel vertier,” Gaastra, “Arbeid op Oostenburg,” Van Dam, Beschrijvinge, NA VOC 4469. white collar workers of the voc in amsterdam 195 making settlements between accounts (liquidatiecomptoir) and a bookkeeper for the armoury.6 In the later 18th century, separate positions for bookkeepers and clerks at the slaughterhouse and a lottery office were created as well.7 A similar expansion and specialisation, though on a smaller scale, took place in the second largest Chamber of the VOC, Zeeland. The main difference with Amsterdam was that the Chamber Zeeland lacked a separate copy office and separate bookkeeper’s positions for mustering, and for the armoury and the slaughterhouse.8 The administrative staff of the other Chambers always remained quite small, as the data on Delft illustrate.9 The office personnel in Rotterdam in 1730 consisted of a First Bookkeeper, as cashier, a bookkeeper for payments (also Second Book- keeper) and a clerk,10 and the size of the staff in Hoorn and Enkhuizen was probably equally modest. The reason why the white collar workforce in Amsterdam, and to a lesser extent in Zeeland, was so much larger than in the other Chambers of the VOC was not only that these two chambers equipped many more ships, recruited many more people and sold many more goods than the rest, but also that they handled the bulk of the correspondence with Asia and the central administration of the Company in Holland, including the staff work for meetings of the Heren XVII and the regular consultations between directors of the six Chambers in The Hague (Haags Besogne). Clerks in Amsterdam were also producing lots of copies for Zeeland, Rotterdam, Delft, Hoorn and Enkhuizen.11 The Chamber Amsterdam over time assumed the other trappings of a real bureaucracy as well.12 Once the division of labor among the office

6 na VOC 253 res. Chamber Amsterdam 31 October 1726 Lijste van de beampten van de Oost. Ind. Comp ter kamer Amsterdam. 7 f.S. Gaastra, “Arbeid op Oostenburg. Het personeel van de kamer Amsterdam van de VOC,” in J.B. Kist (ed.), Van VOC tot Werkspoor. Het Amsterdamse industrieterrein Oosten- burg (Utrecht, 1986), pp. 63–80, at 80; NA VOC 7226. 8 Victor Enthoven, “Veel vertier.” De Verenigde Oostindische Compagnie in Zeeland, een economische reus op Walcheren,” Archief van het Zeeuws Genootschap 1987, pp. 49–127, appendix VI, Lijst van bedienden bij de Oostindische Compagnie ter kamer Zeeland. 9 l.L.M. Eekhout, “Een bedrijf in bedrijf: organisatie en werkverdeling van de kamer Delft,” in H.L. Houtzager et al. (eds), Delft en de Oostindische Compagnie (Amsterdam, 1987), pp. 13–36 esp. pp. 23–6. 10 na VOC 4469 Lijste van de bedienden . . . kamer Rotterdam 1 May 1731. 11 na VOC 7230, notes about the duties and schedules of the clerks of the copy office by Johannes Adami, late 18th century. 12 see the (partly Weberian) study of the VOC’s bureaucracy by M.A.P. Meilink-Roelofsz, “Hoe rationeel was de organisatie van de Nederlandse Oost-Indische Compagnie?,” Economisch- en Sociaal-Historisch Jaarboek, 44 (1982), pp. 170–90. 196 karel davids staff started to grow, the directors began to develop more and more elaborate regulations for different categories of administrative personnel. Book- keepers, cashiers and clerks were not only formally required to swear an oath that they would faithfully serve the VOC, closely guard all Company secrets and refrain from participating in any operations that would harm the interests of the organization, but they were also subjected to increasingly specific regulations concerning work procedures, standards of conduct and forms of remuneration.13 By far the most comprehensive rules were devised for clerks in the copy office. Work hours, duties and authority relations on the shop floor were spelt out in minute detail.14 These regulations were clearly intended to mould the scribes into a disciplined work force under the strict supervision of a dedicated floor manager (the First Clerk), ready to turn out any quantity of written paper at the directors’ wish, at a moment’s notice. A comparison of graph 1 and table 1 finally reveals two other interesting features of the evolution of the white collar force of the VOC. First, the size of the staff did not contract when the volume of the maritime and commercial activities at the end of the 18th century diminished. There were no cuts in the number of administrative personnel when the VOC run up against financial difficulties. Whether economies in personnel expenses were made in other ways, e.g. in earnings, remains to be seen. Secondly, the growth in staff size means that career opportunities for white collar workers at the VOC in the later 17th and 18th centuries must have increased as well. Let us now examine staff earnings, skills and careers more closely.

Earnings and skills

Despite the relatively high degree of bureaucratization of the Company’s organization, reconstructing earnings for the office staff of the VOC is a no less complex affair than is the case with other groups of white collar workers. To be sure, it is fairly easy to obtain data about official levels of pay in different ranks and categories of the Company’s administrative hierarchy. But it is not very simple to establish the actual levels of income,

13 Van Dam, Beschryvinge, Eerste boek, deel I, pp. 371–3, 377, 381–96, 412–3. 14 Ibidem, p. 395; NA VOC 231 res. Chamber Amsterdam 9 August 1632, VOC 307 f. 99v 5 March 1629, 4 March 1630, 28 June 1638, 22 December 1653, 29 July 1666, f.110 2 December 1717. white collar workers of the voc in amsterdam 197 because the earnings of many white collar workers of the VOC, like the clerks at the British East India Company before 1800, consisted both of a fixed salary and of variable perquisites and gratuities, in money or in kind. Bookkeepers, cashiers and clerks often earned more than a mere glance at the official salary scales would suggest. First, the easy part: indexes on the resolutions of the Chamber Amster- dam contain quite a lot of information on statutory pay levels for various categories and ranks of white collar workers.15 A general overview of these data is presented in table 2. Unless stated otherwise, these figures concern official levels of salary paid per year. In contrast with sailors and soldiers, white collar workers of the VOC earned payments expressed as an annual, instead of a monthly, sum. The figures cited in table 2 give an approximate indication of the starting salaries paid to employees in different categories and ranks; salary levels of individual white collar workers as a rule would rise over the years, as I will discuss shortly. Moreover, the archive of the Chamber of Amsterdam also allows us to obtain an almost complete picture of the official salaries paid to its employees at two given moments in time: 1726 and 1762;16 these data have been included in the columns for the years 1720 and 1750 in the table below. The most obvious conclusions to be drawn from this table are (1) that bookkeepers and cashiers of the VOC generally enjoyed higher wages than clerks, (2) that pay levels of clerks in the copy office showed a significant rise between about 1630 and the 1720s and (3) that the wage levels of some categories of bookkeepers (notably, the chief bookkeeper and bookkeepers in the pay office) remained remarkably stable. The first two findings can be more easily explained than the last one. The difference in wage levels can be taken to reflect the skill premium. The skills required from bookkeepers and cashiers in the VOC were more advanced, and in all probability more scarce, than those required from the majority of clerks. The most important ability that clerks in the Com- pany’s service needed to possess was the competence to read and write rapidly, clearly, and accurately. In contrast with clerks at the registry of Generality in The Hague, whose incomes in the 1740s were almost as high as the pay levels of most bookkeepers of the VOC in Amsterdam, clerks of the Dutch East India Company were not expected to be conversant in

15 NA VOC 307, 308, 309, 310. 16 NA VOC 253 res. Chamber Amsterdam 31 October 1726, 6846 Notitie van de tractementen . . . kamer Amsterdam 31 May 1762; cf. Gaastra, “Arbeid op Oostenburg,” p. 78. 198 karel davids

French, or to have a smattering of Latin or English.17 Bookkeepers and cashiers, by contrast, were not only required to be literate and punctual, but also to be highly numerate and to master all the intricacies of the art of bookkeeping. Just as in towns in Italy and Flanders earlier, bookkeeping in Amsterdam from the late 16th century developed into a kind of “separate craft,” complete with its special instructors and its own supply of text- books and how-to guides. The extent of specialization even went so far, that an Amsterdam teacher in 1696 advertised a special training course in the local newspaper on bookkeeping aboard East India vessels.18

Table 2. Statutory pay levels for white collar workers at the Chamber Amsterdam, c.1630–1780, by category and rank* c. 1630 c. 1660 c. 1690 c. 1720 c. 1750 c. 1780 Chief bookkeeper’s office Chief bookkeeper 1000 1000 1000 1000 1000 1000 First assistant 600 1200 1200 bookkeeper Second assistant 600 1000 1000 bookkeeper Settlement office First bookkeeper 1200 1000 1200 2000 Second 1600 bookkeeper Third bookkeeper 1050 Assistant/First 750 clerk Clerk 400/500 400/500 Pay office First bookkeeper 700 800 800 800 800 800 Assistant 600 600 600 600 600 600 bookkeepers

17 O. Vries, “Klerken ter Griffie van de Staten-Generaal in de achttiende eeuw. Een pros- opografisch onderzoek,” Bijdragen en Mededelingen betreffende de Geschiedenis der Neder- landen, 96 (1981), pp. 26–70, pp. 42, 61. 18 Karel Davids, “The bookkeeper’s tale. Learning merchant skill in the Northern Nether- lands in the sixteenth century,” in Koen Goudriaan, Jaap van Moolenbroek and Ad Ter- voort (eds), Education and learning in the Netherlands, 1400–1600 (Leiden, 2004), pp. 235–51, at 242–5. white collar workers of the voc in amsterdam 199

Table 2 (cont.) c. 1630 c. 1660 c. 1690 c. 1720 c. 1750 c. 1780 Yard First bookkeeper 1300 2000 2000 2000 Second 25 p/ 800 1200 bookkeeper week First clerk 1200 Second clerk 1000 Third clerk 700 [ Junior] clerks 350 200/500 200/350 400 400 Stores First bookkeeper 1250 1500 Second 1100 bookkeeper Third bookkeeper 600 600 Clerks 450/600 450/600 400/600 16/20 16/20 p/week p/week Mustering Bookkeeper 200 300 200 200 Cashier’s office Cashier 0,1% 0,125 % 5000 5000 monetary monetary transactions transactions Bookkeeper 1000 receipts Assistants 600 bookkeeper receipts Copy office Clerk, first rank 350 600 1000 1000–1300 1300 1300 Clerk, second 450 900/1100 800/1050 rank Clerk, third rank 300 600/725 650/750 Clerk, fourth rank 200/250 450/550 450/550 Clerk, fifth rank 50 50/150 150 350 350 350 * in guilders per year (unless otherwise stated) 200 karel davids

The rise of pay levels of clerks in the copy office of the VOC between about 1630 and 1720 is not a puzzling phenomenon either. It was more or less in line with the increase in annual salaries of other categories of employees in the Western Netherlands during in this period. Municipal clerks in Amsterdam in fact even enjoyed a nearly threefold increase in pay between c. 1620 and 1712.19 The VOC could no more afford to neglect the pull of market forces in its search for administrative personnel, than the British East India Company could 150 years later. As regards cashiers, market conditions were the very reason for a substantial increase in salary as early as 1616. When the directors of the Chamber Amsterdam observed that cashiers left the Company’s service because they considered the payment of one stuiver per hundred guilders of all monetary transactions too low, and replacements “for that small reward” proved to be impossible to find, they decided to raise the salary to one and a half stuiver per hundred guilders, rising to two stuivers in 1631, and two and a half in 1651.20 But why were the wage levels of some categories of bookkeepers remarkably stable? Why would VOC-clerks of the highest rank, earning a salary of 1300 guilders a year, in the middle of the 18th century sometimes accept position as bookkeeper in the pay office, even though the statutory pay levels in the latter category by then had fallen well behind those of the former? The answer to this particular riddle is, that salary tables tell only half the story of remuneration of white collar workers of the VOC—which brings us to the second, more difficult part of the analysis. Many of the white collar workers of the VOC derived the bulk of their income not from regular salaries, but, like clerks of the English East-India Company before 1800,21 from various sorts of extra benefits in money or in kind. Such extra benefits were often of a variable nature: they were linked to the volume of activities of the Company and its personnel or shareholders. For bookkeepers in the chief bookkeeper’s office and in the pay office of the Chamber Amsterdam, the most important source of income was their share in levies on the transfer of transport-letters, deeds, sentences and other documents relating to VOC-personnel overseas. Every transport- letter of a crewmember that was not left to immediate family, was subject to a levy of 5% for the benefit of the bookkeepers of the VOC. Likewise, the bookkeepers received 2 to 5% of payments of inheritances (other than

19 De Vries and Van der Woude, The First Modern Economy, pp. 617–9; J.L. van Zanden and L. Soltow, Income and Wealth inequality in the Netherlands (Amsterdam, 1998), p. 44. 20 Van Dam, Beschryvinge, Eerste boek, deel I, pp. 388–9. 21 Boot, “Real incomes,” p. 642. white collar workers of the voc in amsterdam 201 to next of kin), 5% of remittances of less than 150 guilders from the Indies (and ½% on larger sums) and a fixed sum (ranging between three and twelve stuivers) for a copy of a death certificate, an attestatio de vita, a sentence or a similar kind of official document.22 The income generated by these levies was distributed among the bookkeepers according to a specific code, which from time to time was subjected to minor revisions. The code laid down by the directors of the Chamber Amsterdam in 1741 stipulated that the chief bookkeeper would receive 19%, the first bookkeeper of the pay office 16%, the two senior bookkeepers in the chief bookkeeper’s office or the pay office 13% each, the next two 9.8% each and the two junior ones 9.4% each.23 Bookkeep- ers themselves thought that these variable benefits, and especially “the benefit of the transport-letters,” in fact formed the bulk of their earnings. When the Heren XVII in 1781 resolved to abolish transport-letters for soldiers, bookkeepers of all six chambers of the VOC drew up a joint memorandum, furnished with calculations based on figures relating to the period 1750–1778, to demonstrate that the new regulation would lead to an average loss of income of some four guilders and four or five stuivers per soldier, and they asked to be compensated accordingly.24 Chief bookkeepers were also allowed to charge buyers and sellers for the registration of transfer of VOC-shares; this charge was initially fixed at twelve stuivers per transaction, but it was later raised to twice that amount. In 1748, the right of collection was transferred from the chief bookkeeper to the first bookkeeper of the settlement office.25 Bookkeepers responsible for mustering people on East-Indiamen were permitted to collect a payment for every person registered. Until 1718, cashiers earned their income by collecting a levy for every transaction they conducted (fixed at two an a half stuiver per hundred guilders in 1751). Cashier Jan Bos, appointed in October 1718, was the first one to be put on a regular salary.26

22 NA VOC 6941 extr. res. Heren XVII May 5, 1669, October 1, 1682 and December 16, 1682, memorandum (s.d.) on perquisites enjoyed by bookkeepers in the pay office; Van Dam, Beschryvinge, Eerste boek, deel I, p. 386. 23 NA VOC 258 res. Chamber Amsterdam 26 January 1741. 24 NA VOC 6940 correspondence of J. Adami, bookkeeper of the pay office of the Cham- ber Amsterdam, with the chief bookkeepers of other chambers of the VOC, January 27– April 2, 1781. 25 Van Dam, Beschryvinge, Eerste boek, deel I, p. 383, NA VOC 308 f.387, see under “liquidatiecomptoir,” Nicolaas Wiltens. 26 NA VOC 307 f. 85, see under “cassiers,” Jan Bos. 202 karel davids

Unlike clerks at the Registrar’s office in The Hague,27 clerks of the Chamber Amsterdam of the VOC did not receive a given sum for every piece of written paper they produced but a fixed annual salary instead. While the clerks in The Hague enjoyed a remarkable rise in wealth in the second half of the 18th century, thanks to the sheer increase in the number of pages they managed to scribble,28 clerks of the VOC could not benefit from an expansion in the volume of business in a similar way. In this respect, clerks in The Hague had more in common with the bookkeepers at the offices of the Company in Amsterdam, than with their colleagues in clerical jobs, which may go some way to explain the fact that their earnings in the 1740s were higher than those of most clerks of the VOC. It may also help us to understand why even VOC-clerks of the highest rank in the middle of the 18th century were sometimes prepared to accepted a lower- salaried position as bookkeeper in the pay office. This finding does not exclude, however, that clerks of the Company posted at places such as the pay office or the storehouse (where—especially in the 18th century—a vast amount of business was going on) may have had more opportunities for earning extra money, officially or unofficially, than clerks employed at the humdrum business of copying letters and documents for meetings, although these, too, sometimes received gratuities for extra services.29 Other perquisites of white collar workers were paid out in kind. Accord- ing to a specification of 1730, two bookkeepers of the settlement office, three bookkeepers of the storehouse and the two senior clerks of the copy office in Amsterdam were entitled to receive portions of pepper, nutmeg, mace, cloves and cinnamon. In the Chamber Zeeland, similar payments in kind were handed out to the cashier and the first bookkeepers in the pay office, the storehouse and the settlement office.30 The chief bookkeeper of the Chamber Amsterdam received, on top of his annual salary, 300 guilders per year for rent. The bookkeeper of the shipyard enjoyed free rent; in addition, the first bookkeeper of the yard was entitled to free fuel and free beer.31

27 Vries, “Klerken ter Griffie,” p. 32. 28 Ibidem, p. 63. 29 See for example NA VOC 307 under “eerste clerquen” f. 110 4 April 1687, 6 June 1688, f.102 13 December 1708, 26 Februari 1714. 30 NA VOC 4469 Lijste van de portien specerijen . . . kamer Amsterdam August 1, 1730, Specificatie van de tractementen en daggelden . . . Kamer Zeeland. 31 Huibert Schijf and Bernadette van Woerkom, “Boekhouders en peperwerkers; werknemers van de VOC in Amsterdam tussen 1602 en 1680,” in F. Wieringa (ed.), De Verenigde Oostindische Compagnie in Amsterdam (Amsterdam, 1982), pp. 96–154, at 126–7. white collar workers of the voc in amsterdam 203

White collar workers of the Chamber Amsterdam, like clerks at the Registrar’s office in The Hague,32 moreover could earn extra income by holding more than one job at a time. Bookkeeper Pieter van Rhijn of the settlement office, for example, from 1699 onwards also acted as librarian of the VOC, which yielded an additional income of 200 guilders a yearz33 Pieter Vlaming, bookkeeper of mustering since 1719, combined his job at the VOC with a business in victuals.34 A clerk at the Company’s storehouse, Jacob van Hoogstraten, in the 1740s supplemented his earnings by working as a gravemaker on the side.35 Clerks at the copy-office presumably sometimes earned extra money by assisting at other offices of the Company,36 or by doing work commissioned by outside clients. Tellingly, a new regulation issued by the directors of the Chamber Amsterdam in April 1763 expressly forbade clerks at the copy office to accept any secondary employment inside or outside the Company, and told them to concentrate “entirely on the service of the schrijfcomptoir.”37 What were the total earnings of white collar workers of the VOC? Given the variety of income sources, exact figures are hard to come by, but it is nonetheless possible to give an approximate indication with the help of two sources from the second quarter of the 18th century. The first one is a survey made by the directors of the Chamber Amsterdam on the order of the States of Holland in 1726, which lists the salaries and perquisites of the white collar workers employed by the Chamber; however, the sums mentioned in this statement, as Femme Gaastra suspected, are most certainly too low, especially those of the non-salary incomes. The directors probably deliberately underestimated the earnings of their servants.38 The second source is a tax register drawn up in 1742, the Personeele Quotisatie, which gives an estimated annual income of all heads of family in Amster- dam (a few groups excepted), who showed specific marks of wealth such as a particular rental value of a house, the employment of servants or the possession of a countryhouse and/or a boat or carriage and horses. The

32 Vries, “Klerken ter griffie,” pp. 36–7. 33 NA VOC 307 f. 275, see under “liquidatiecomptoir” Pieter van Rhijn. 34 M.S.J. Cox-Andrau, De dichter Pieter Vlaming (1687–1734). Een studie over zijn werk met een levensbeschrijving (Bussum, 1976), pp. 280, 285. 35 NA VOC 4469 Notitie van de tractementen en dagloonen [1730]; W.F.H. Oldewelt (ed.), Kohier van de Personeele Quotisatie te Amsterdam over het jaar 1742 (Amsterdam 1945), 2 vols, II, 115 nr. 567. 36 NA VOC 307 f. 101 res. Chamber Amsterdam 3 January 1707. 37 NA VOC 7226 extr. res. Chamber Amsterdam 4 April 1763. 38 NA VOC 253 res. Chamber Amsterdam 31 October 1726; Gaastra, “Arbeid op Oost- enburg,” p. 70. 204 karel davids

estimates were based on assessments by tax farmers, combined with evidence from other fiscal documents such as registers of real estate.39 The annual incomes of white collar workers of the VOC listed in the Personeele Quotisatie should therefore be interpreted as approximate figures, which are possibly somewhat higher than the total earnings from salaries or extra benefits received in the Company’s service, because they may also have included income from non-Company sources, such as from rents or interests. All the data on white collar workers in the second quarter of the 18th century from the different sources discussed above, ordered by category and ranks, are listed in table 3.

Table 3. Statutory remuneration and estimated actual levels of income of white collar workers of the Chamber Amsterdam, c. 1742 Category/rank Extra benefits Regular Estimated Estimated annual salary perquisites actual annual 1726 income 1742 Chief Rent fl. 300 % of 1000 1000 14000 bookkeeper levy on transfer of transport letters etc.; levy on transfer of VOC-shares First assistant % of levy on transfer 600 400 2500 Bookkeeper of transport-letters etc. Bookkeeper Portion of spices; daily 1200 3500 settlement payments office Clerk 400–500 600–1200 settlement office Bookkeeper % of levy on transfer 800 1200 6000 pay office of transport-letters and copies of deeds, sentences etc. Assistant % of levy on transfer 600 400 2500 bookkeeper transport-letters pay office and copies of deeds, sentences etc.

39 Oldewelt, Kohier, I, pp. 3–10. white collar workers of the voc in amsterdam 205

Table 3 (cont.) Category/rank Extra benefits Regular Estimated Estimated annual salary perquisites actual annual 1726 income 1742 First Free rent/fuel/beer 2000 8000 bookkeeper yard Second Free rent 800 2500 bookkeeper yard Clerk yard 200–350 600–800 Bookkeeper Portion of spices 1250 3000 stores Clerk stores 1200 4500 Clerk stores 900 4000 Bookkeeper Payment per person 200 1300 5000 mustering mustered Cashier 5000 10000 Bookkeeper 1000 3000 receipts Assistant 600 1000 bookkeeper receipts Senior clerk Portion of spices 1300 600 2500 copy office Source: see text.

The figures presented in the right-hand column of table 3 probably represent the highpoint of the earnings realized by the various groups of white collar workers of the VOC in the 17th and 18th centuries. As the earnings of many white collar workers were directly linked to the volume of activities of the Company and its personnel or shareholders—as we have seen above—, they must have benefitted from the huge expansion in numbers of ships, personnel and all sorts of transactions between, especially c. 1690 and 1730, in a massive way. The opportunities to make a large income from the Company’s service must have been, for most clerks and bookkeepers, much greater around 1730 than, say, in the middle of the 17th century. The decrease, or stagnation, in the volume of activities which started in the 1730s and became more pronounced in the 1770s, must have 206 karel davids

Table 4. Estimated incomes of white collar workers of the VOC, compared with other white collar workers in Amsterdam in 1742 Classes of All taxable Taxable white Taxable white income persons collar workers, collar workers non-VOC* VOC* fl. 600–800 44% 43% 21% fl. 1000–2000 37 42 27 fl. 2500–9000 18 15 45 fl. 10000– 2 6 Total 100% = 12655 100% = 274 100% = 33 * white collars workers include bookkeepers, cashiers and clerks Source: Oldewelt, Kohier, I. implied that the palmy days for the office staff in Amsterdam were over as well. Levels of income of most white collar workers of the VOC in 1790 may well have been somewhat lower than about 1730. The sheer affluence of white collar workers of the VOC stands out clearly when their estimated incomes in 1742 are compared with those of bookkeepers, cashiers and clerks who were not employed by the VOC, and with those of the taxed population as a whole. Salaried employees of the VOC were not just doing well, they were doing extremely well. Not only was the proportion of taxable persons among all white collar workers of the VOC in Amsterdam much higher than the proportion of taxable persons among the heads of family in Amsterdam as a whole (80% against 30%),40 but the distribution of income was more skewed towards the higher brack- ets of income (table 4). The category of white collar workers of the VOC who were least well off, largely consisted of clerks in the copy office. Yet, even those employees were in a sense better placed than most workmen, because they enjoyed ample opportunities for getting ahead and improving their incomes during their lifetime. I turn now to a consideration of career opportunities.

Careers

In his wide-ranging survey on the history and organization of the VOC composed in the 1690s, the then highest-ranking official of the Company,

40 Cf. table 5 and table 1, and Oldewelt, Kohier, I, p. 10. white collar workers of the voc in amsterdam 207

Pieter van Dam, made the observation that “the salaries of employees were not always uniform, but varied according the level of abilities of the persons concerned,” and that “someone who first entered into a particular position normally would not immediately be given the highest salary, but would gradually earn more,” in order to stimulate employees to improve their performance.41 Van Dam’s remark captures some key features of the salary structure of the administrative staff of the VOC: this structure was already in the 17th century designed in such a way that (1) salaries were graded according to ability, (2) new entrants were placed in a lower scale than older hands, and (3) individual employees could earn more when their experience and achievement grew. Like the clerks of the British East India Company studied by H.M. Boot,42 salaried employees of the VOC thus could look forward to increase their earnings their during their lifetime, provided they performed well in their job. Yet, the salary structure at the VOC never developed to the same extent into an age- or experience- related arrangement as it did in the British East India Company after 1800. Perquisites, after all, continued to form an important part of the total earnings of many employees in the white collar workforce, in particular of bookkeepers and cashiers. The size of these kinds of earnings was much less related to age or experience than formally fixed salaries. The category of employees whose system of payment conformed most closely to an age- or experienced-related structure, was that of the clerks in the copy office. From the clerks’ point of view, this can hardly have been a demoral- izing state of affairs. Being clerk at the copy office of the VOC required a lot of hard, disciplined work, but it was not a dead-end job. A competent clerk could look forward to getting a higher salary as he became older and more experienced, and moreover, he could use his post as clerk as a springboard to attain a more lucrative position elsewhere in the VOC’s organization in Amsterdam, or in the East-Indies. Table 5 shows how seven cohorts of clerks, entering the copy office of the VOC in different decades, fared during their lifetime in terms of earnings and position in the Company’s organization. Column three illustrates the striking rise in starting salaries of clerks between the 1640s and 1720s, column four shows to what extent individual clerks at the copy office enjoyed a rise in salary during their career as clerk. Clerks who

41 Van Dam, Beschryvinge, Eerste boek, deel I, p. 370. 42 Boot, “Real incomes,” pp. 643–7. 208 karel davids entered the Company’s service before 1720 clearly could look forward to a substantial improvement in payment in the course of their career. Of the eleven clerks hired at a modal starting salary of 150 guilders in the 1680s, for example, eight earned 800 guilders or more at the end of their career as clerk fifteen to twenty years later. Once the modal starting salaries stabilized at a level of 350 guilders per year from the 1720s onwards, the chances of seeing one’s pay increased by a factor three or more visibly diminished. Yet, the rise in nominal salaries of clerks for a long time far outstripped the increase in the cost of living, which amounted to some 150 % between the 1640s and 1720s and then hovered at more or less the same level until the 1740s.43 It was not until after the middle of the 18th century that clerks at the copy office of the VOC must have seen their incomes increasingly threatened in real terms. While the cost of living began to show a sharp upward trend, starting salaries of clerks remained more or less unchanged and salary increases became more piecemeal and on average no longer attained the high rates achieved before the 1720s. Although I have not found the reasons for this salary policy spelled out in the directors’ own words, I suspect that the Chamber Amsterdam, faced with the expansion of the administrative staff (and, later on, the growing need to cut expenditures),44 tried to keep the personnel in the copy office increasingly on a tight rein. Clerks in the copy office were after all the easiest category of staff to control. The flipside was presumably, that clerks were more often tempted to seek additional income outside the Company, or to leave the Company’s service altogether. Not only did the directors find it imperative in 1763 to prohibit clerks explicitly from accepting any secondary employment inside or outside the Company,45 a remarkable proportion of clerks in that period left the staff on their own accord. Of the nine clerks entering the copy office in the 1760s, five handed in their resignation within a few years after being appointed. Such an exodus had never happened before.

43 De Vries and Van der Woude, First Modern Economy, p. 628, figure 12.4. 44 Ingrid Dillo, De nadagen van de Verenigde Oostindische Compagnie 1783–1795 (Amster- dam, 1992), pp. 28–33. 45 NA VOC 7226 extr. res. Chamber Amsterdam April 4, 1763. white collar workers of the voc in amsterdam 209

Table 5. Careers of clerks of the Chamber Amsterdam, by cohort, 1640–1769 Cohort Size Modal End salary as Position at end of career at VOC starting clerk, as% of salary starting salary 100–200 clerk first other in 201–300 copy clerk staff East- 301–500 office position Indies 501– 1640–1649 10 Fl. 50 3 1 5 6 1 2 1 1660–1669 10 100 1 1 3 5 4 5 1 1680–1689 11 150 1 2 8 6 1 3 1 1700–1709 4 200 2 1 1 2 1 1 1720–1729 8 350 4 1 3 3 3 2 1740–1749 9 350 1 8 2 6 1 1760–1769 9 350 5 1 3 7 2 Source: based on NA VOC 307 and 308.

Table 5 also shows that the prospects for clerks at the copy office to land a more lucrative job elsewhere in the VOC’s organization in Amsterdam or in the East-Indies were not at all bad before the 1760s. Many of them succeeded after a while to get a bookkeeper’s post in the pay office (or in the chief bookkeeper’s office) and a few even rose to become First Clerk. It was not unusual either for a clerk to leave his desk at the copy office in Amsterdam to pursue a career as a merchant or as a military officer in the Company’s service in the East Indies. The most remarkable career of a clerk was doubtless that of Joan Gideon Loten. Loten entered the VOC’s service as a clerk of the Cham- ber Amsterdam in March 1728, then left for the East Indies as a junior merchant in January 1732, rose through the ranks as prosecutor and merchant in Semarang, became first administrator of Onrust, then senior merchant in Batavia and governor/director of Macassar, and finally councillor extraordinary of the Supreme Government in Batavia. He ended his career in the Company’s service as governor and director of Ceylon, and councillor ordinary of the Supreme Government in Batavia. He returned to the Netherlands in 1757–1758 as Admiral of the Return Fleet and commis- sary at the Cape of Good Hope. But then, Loten was not a man with an ordinary background. He was born from a landowning family in Utrecht, had studied at Utrecht University and boasted an uncle who had made 210 karel davids a fortune in the Indies and, like Joan Gideon later on, had returned as Admiral of the Return Fleet.46 Were careers in the white collar workforce of the VOC equally open for everyone who commanded the necessary skills? More so perhaps than in the case of the Registrar’s Office in The Hague, I suspect, but not entirely. Appointments in the Registrar’s Office were much more a private affair than in the Chamber Amsterdam of the VOC. A large number of the clerical appointments were at the behest of the Registrar himself. Ordinary clerks, in fact, formed part of his domestic personnel. Anyone who wished to obtain a clerk’s position therefore had to be well-connected to the Reg- istrar, or at least to people close to him.47 Good connections, including family networks, were important for career opportunities in the Cham- ber Amsterdam as well, but in a slightly different way. Decisions about appointments and salaries of bookkeepers, cashiers and clerks were normally made by the twenty directors of the Chamber as a collective body, even though recommendations were drawn up by the different committees or “departments” (dealing with audit, receipt, merchandise and equipment) in which the directors had organized themselves. The selection of new entrants to the administrative staff was therefore carried out by a larger number of people than in the case of the Registrar’s Office in The Hague, and the number of “access points” for applicants was corre- spondingly greater. Money sometimes entered into the equation, too. There is some evidence from the 18th century, especially relating to the Chamber Zeeland, that candidates for positions in the staff of the VOC were paying slush money to get a job. The most notorious case is that of Pieter Vlaming, who is said to have landed the highly attractive post of “bookkeeper of mustering” at the Chamber Amsterdam in 1719, by secretly paying down 16,000 guilders into the private account of the Amsterdam regent and VOC-director Jeronimo de Haze de Georgio.48 For some positions, possession of wealth was even formally considered to be an necessary condition. Cashiers of the VOC were always required to deposit a huge security

46 A.J.P. Raat, The Life of Governor Joan Gideon Loten (1710–1789). A Personal History of a Dutch Virtuoso (Hilversum, 2010), pp. 31, 54–6. 47 Vries, “Klerken ter Griffie,” pp. 39–44. 48 Enthoven, “Veel vertier,” pp. 87–8; Gaastra, “Arbeid op Oostenburg,” p. 70; Meilink- Roelofsz, “Hoe rationeel,” p. 186, Cox-Andrau, Pieter Vlaming, pp. 279–85; J.E. Elias, De vroedschap van Amsterdam 1578–1795 (Amsterdam, 1963²), 2 vols, I, pp. clx–clxii, esp. footnote 2. white collar workers of the voc in amsterdam 211

(usually 50.000 guilders) before they received their official appointment from the Company.49

Conclusion

The experience of white workers of the Dutch East India Company in the 17th and 18th century in many respects prefigured that of clerks working for the British East India Company between about 1760 and 1830. Bookkeepers, cashiers and clerks of the VOC enjoyed a long-term rise in earnings in excess of most wage workers, which probably lasted at least until the 1760s. Like their colleagues at the British East India Company, white collar employees of the VOC also had the advantage over many other groups of workers of having good prospects of making a career, and increasing their earnings during their lifetime. The age- or experienced related salary structure found in the British East India Company was to some extent paralleled in the VOC, too. The category of employees whose system of payment conformed most closely this structure, was that of the clerks in the copy office. Yet, the similarities were not complete. The Dutch East India Company never entirely shifted to the age-related salary structure as the British East India Company did about 1800, as we have seen. In the VOC, perquisites continued to form an important part of the total earnings of many employees in the white collar workforce, in particular of bookkeepers and cashiers. The size of these sorts of earnings was much less related to age or experience than formally fixed salaries. Attempts to restrict the income from perks, such a levy on transport letters of soldiers collected by bookkeepers, seem to have been deftly defused by resistance from bookkeepers themselves.50 A comparable undertaking to reorganize the registry of the Generality in The Hague in 1792, including the reduction of income of clerks from payments per written piece of paper, proved to be much more successful.51 Was the inertia of the salary structure in the Chamber Amsterdam connected, perhaps, with the nature of the Amsterdam labor market for higher employees? If H.M. Boot is correct in his claim that the British

49 Van Dam, Beschryvinge, Eerste boek, deel I, pp. 388–90; NA VOC 307 under “cassiers” Adriaan Valkeniers and Jan Rijniers 25 August 1611. 50 See above, note 24, and NA VOC 308 f.323 res. Chamber Amsterdam 9 April 1781: Amsterdam endorses a proposal for compensation prepared by the bookkeepers themselves. 51 Vries, “Klerken ter griffie,” pp. 37–8. 212 karel davids

East India Company, in developing an age- (experience-)related salary structure, in fact followed the example of “employers of higher clerical and administrative skills elsewhere in the [British] economy,”52 the question arises whether other employers in late 18th-century Amsterdam, who competed with the VOC for the services of highly skilled personnel, used a reward system that was less age- (experience-)related and more based on perquisites than in the case of their colleagues in Britain. This may be an interesting hypothesis to test in further research. Under these conditions, changes in the salary system in the VOC may have been much more difficult to achieve than in a market structure for higher employees as existed at the time in Britain. The “bonus culture” among bookkeepers and cashiers in 18th century Amsterdam was perhaps (barring outside interference) as hard to eradicate as it is in the world of finance today.

52 Boot, “Real wages,” p. 662. mobilities

The Career Ladder to the Top of the Dutch East India Company: Could Foreigners also Become Commanders and Junior Merchants?

Jaap R. Bruijn and Femme S. Gaastra

In the 17th and 18th century, the Dutch East India Company employed hundreds of thousands of people. The size of its workforce at sea and overseas is known fairly precisely: between 1602 and 1700, some 317,800 men left the Republic on the vessels of the VOC, while 114,400 returned, and between 1700 and 1795 no less than 655,200 persons sailed out, and 252,500 repatriated.1 Many of those who left the Dutch Republic for Asia came from abroad. Jan Lucassen calls them “transmigrants.”2 These foreigners first migrated to the Netherlands, looking for employment. Com- pany service was one of the major sources of employment; it could last for a certain number of years, or for the rest of their life, when they stayed overseas. As a result of a large research project, started in 2002 and financially supported by a number of important institutions, information has become available about the individuals who were enlisted by the Company and about their careers in its service. 2,800 volumes of ship’s pay-ledgers from the 18th century have been computerized. All people who departed for The Cape and Batavia (Jakarta) are mentioned in these pay-ledgers, with the exception of senior functionaries, women and other passengers. Chamber by Chamber, step by step, this computerization has been done by Drs. A.J.M. van Velzen of the National Archive at The Hague and his team of collaborators. The project is now reaching its completion. One feature of the project has not yet been finalized. This is the identification of the thousands of geographical names, stating the places of origin of those enlisted. So far almost 78 per cent of the names has been sorted out. The rest concerns the unmentioned and unreadable names and, much more numerous, the places which were only a few times registered

1 Jaap R. Bruijn, Femme S. Gaastra and Ivo Schöffer (eds), Dutch-Asiatic Shipping in the 17th and 18th Centuries, 3 vols. (The Hague, 1979–87; further referred to as DAS), I, p. 170. 2 Jan Lucassen, Dutch Long Distance Migration. A Concise History (Research paper IISH no. 3; Amsterdam 1991), p. 27. 216 jaap r. bruijn and femme s. gaastra in the course of almost a whole century. An obvious distinction which an historian would like to make is that between “Dutchmen” and “foreigners.” At present, out of the total individual records, 45 percent are identified as Dutchmen, and 33 per cent as people from abroad. From the unexplored 22 percent, 8 per cent was very likely Dutch, and 14 per cent foreign. The overall implication is that a small majority (53 percent) of the Company’s personnel of seamen and soldiers in the 18th century was Dutch.3 It is noteworthy that Jan Lucassen published about this issue already more than thirty years ago. Based upon the data provided by Dutch- Asiatic Shipping in the 17th and 18th Centuries and research by two Leiden students, he and his co-author calculated the numbers of Dutch and foreign seamen and soldiers in four specific periods.4 Foreigners were found among all categories of the Company’s personnel, among sailors, soldiers as well as merchants. It has been argued by Roelof van Gelder, that for non-Dutchmen—he discussed the position of Germans in particular— difficulties were greater to be promoted to higher ranks, as being dis- criminated against Dutchmen.5 Foreigners, however, were present in all positions, also in the top ranks on board ships and in the Company’s overseas administration. The best proof is that, on three occasions, even the governor-general was a man born outside the Dutch Republic: Abraham Patras (1735–1737), Johannes Thedens (1741–1743) and Gustaaf Willem, baron Van Imhoff (1743–1750). Two groups of Company employees in the 18th century are explored in this essay with regard to their opportunities to reach the top in their profession: commanders of the East Indiamen (i.e. the trading vessels), and junior merchants in the overseas commercial networks.

Commanders

During the 18th century, the proportion of foreigners among the seafar- ers on board Dutch East Indiamen was far from stable. On the contrary,

3 See . See also: Ton van Velzen, “Uitgeva- ren voor de Kamers: 700.000 mensen overzee,” in Jan Parmentier (ed.), Uitgevaren voor de Kamer Zeeland (Zutphen, 2006), pp. 31–46. 4 Jaap R. Bruijn and Jan Lucassen (eds), Op de schepen der Oost-Indische Compagnie. Vijf artikelen van J. de Hullu (Groningen, 1980), pp. 139–40. 5 Roelof van Gelder, Het Oost-Indisch avontuur. Duitsers in dienst van de VOC (Nijmegen, 1997), pp. 184–6. the career ladder of the dutch east india company 217 it increased enormously from the 1720’s onwards. After 1750, there were regularly less Dutch seamen on board than foreign. On a Dutch East India- man, the top rank of commander (schipper in Dutch) was open to foreign seamen as well. As has been described elsewhere, the average commander of a Dutch East Indiaman had made his career in the Company’s service.6 There were not many commanders who had been active in the navy, merchant marine or whaling trade previously. This implies an upward mobility among the Company’s seafaring personnel, which is only self- evident aboard vessels with complements of two to three hundred men, on voyages lasting eight to nine months or more. Mortality could be high, and vacancies for petty officers, officers and commanders always had to be filled quickly from a lower rank. Both Dutchmen and foreigners qualified for promotion. Foreigners were not rated inferior; that, at least, is the impression historians have. This norm also applied to the position of commander; that was never a point of controversy. Almost any commander had completed a certain number of outward and homeward voyages before he was appointed, at least three. So he had survived the dangers of shipboard life, and of a shorter or longer stay in the tropics. He had proved to have a good constitution, next to qualities such as a substantial knowledge of navigation, and to have the attitude of a “person in charge.” The Company rarely appointed “good- for-nothings.” There were strict procedures and exams, from the rank of third mate upwards. By and large, the Company worked with good and qualified commanders. Dutchmen as well as foreigners were appointed. It is quite remarkable that there is no evidence of a language barrier. It means that a commander of foreign origin had mastered Dutch to to the point where he could read and write it. Life on board and ashore in the Republic had provided ample opportunities for this. Moreover, Dutch is said to have been the lingua franca in seafaring in those days. Aboard the ships, and on the boards of the six Chambers, a non-Dutch commander was unexceptional. In comparison with his Dutch colleague, however, a foreigner might have one handicap which he had to face. We return to this point further on.

6 Jaap R. Bruijn, Schippers van de VOC in de achttiende eeuw aan de wal en op zee (Amsterdam, 20083) and Idem, Commanders of Dutch East India Ships in the Eighteenth Century (Woodbridge, 2011). 218 jaap r. bruijn and femme s. gaastra

A dwindling supply of Dutchmen

This having been said, one finds only a few percent of foreign commanders during the first decades of the 18th century, while later on their numbers increased remarkably. Of course, the enlistment of foreigners in proportion to Dutchmen was low early in the century. For instance, 7,922 to 29,400 in 1715–25, as compared with 28,512 to 25,488 respectively in 1765–75. That is to say, the pool of foreigners as possible candidates for the position of commander varied very considerably across the century.7 However, this fact by itself cannot explain the large increase in foreign commanders. The composition of the Dutch supply of seamen also deserves attention. Early in the century, a solid proportion of Dutchmen originated from the local communities in the six Chamber cities. Lots of families were strongly connected with “their” Company. By tradition, the male family members generally served in all kinds of ranks on board ships of their local Chamber, and intermarriage was quite common. Many such a family delivered future mates and commanders. These local Company communities disappeared in the 1730’s and 1740’s. An important explanatory factor was a sudden high mortality rate among newly arrived seamen and soldiers at Batavia, caused by a severe (and endemic) malaria epidemic.8 The impact on families was often disastrous. Just to take one example from the city of Enkhuizen: Commander Ype Engelsman, who had four brothers-in-law who were commanders as well, died in 1730, followed by two brothers-in-law 1737 and a third one in 1741; only one survived to make more voyages. For reasons still unknown, married commanders and mates often had only one or two children. The disappearance of the Company’s seafaring communities coincided with a general economic decline in most of the Company’s Chamber cities, resulting in shrinking population figures and growing poverty. Later in time, recipients of poor relief were stimulated to go into Company service. Their physical and social condition often deteriorated over the time. The almost traditional local candidates for mate and commander on the ships, which were annually equipped, made way for other Dutchmen and a growing number of foreigners. One had, of course, to apply for these functions. In the course of the century, lists of applicants show a diminishing

7 See note 4. 8 P.H. van der Brug, Malaria en malaise. De VOC in Batavia in de achttiende eeuw (Amsterdam, 1994). the career ladder of the dutch east india company 219 number of locals, who previously had often been the favourite candidates of Chamber board members. They mention more and more foreign names. The Chamber of Delft is a good case in point. In 1740, a Frenchman from Rochefort was the first foreign commander since time immemorial. Next, no less than 21 mates from Sweden, Norway, Denmark and Schleswig-Hol- stein were appointed in the 1740s, followed by two foreign commanders in 1749 and 1751. In Hoorn, a Chamber lacking any tradition of foreign commanders all of a sudden appointed four foreigners in 1741–1748, and in the next decade even twelve!

The usual steps to a command

At this point, one has to consider the procedures governing appointments made by the directors (bewindhebbers) of a Company Chamber. The directors handled the appointments of higher personnel in accordance with a set procedure. Commanders, ship’s officers and petty officers were usually appointed during a meeting set for a specific date. Announcements were posted on placards, stating that candidates for specific positions on certain ships could apply. Those who applied were given the opportunity to present their qualifications in person during the meeting. The potential applicants were apparently supposed to be in the vicinity, and to make sure they did not stray too far from the neighborhood of their Chamber. In the 18th century, in any Dutch administrative system, the directors would take turns to nominate their own preferred candidates for vacant positions. In the Company it was not any different. At first glance, the system seems to have been very open and above board and that, indeed, a number of candidates were allowed the opportunity to offer their services. Applicants attended the meeting of the directors at the agreed time. They gave a description of their previous appointments and their voyages. Candidates would seek places on a ship which would set sail only some four or five months later. The date of the required exam was fixed, usually directly afterwards. A week after the presentations, the directors made their choices. That meant it was the choice of the director whose turn it was. His favorite candidate was appointed if the exam results were “competent” or “well answered.” A framework of examinations surrounded each promotion to officer and commander. It all began with third mate, followed by second and first mate (opperstuurman). At the time of his first appointment, a commander was between 30 and 40 years old, on average 34 to 35. There were 220 jaap r. bruijn and femme s. gaastra always exceptions, also among foreigners. Jens True, who later changed his name into Jan de Wit, left Aarhus in Denmark at the age of 14, and was only about 25 when he was made a commander on a return voyage in 1740. Diederik Morel from Hamburg claimed the position at the age of 30 in 1749, and Arnoldus Rogge from Bremen at the age of 33 in 1792, but Jakob Wiebe from Elbing was already forty-two when he was appointed in 1746. But what was the relation between the newly appointed commander and the director, whose choice he was? Did they know each other locally? What about the applicant from outside the city, or even from abroad? The so-called outsiders may have had a handicap in this respect. An answer may be found in the system of payment for a position, a thorny problem in the history of the ancien régime. Theoretically, there was no “sale of offices” and anyone who trespassed the prohibition could expect to be severely punished. The use of a go-between, an intermediary or agent, was therefore common practice. The agent “sold” the vacant position, which was in the gift of the director who commissioned him. With this system, a director of the Company did not violate the obligations of his own position. The money paid for his services reached him under another designation. The converse was equally true. The agent would present a candidate, and such an introduction was perfectly normal. Only incidental documentation reveals this system, and provides information about the sums paid by the applicants. The amounts of money required for a position of commander in reality ranged from f. 1,500 to f. 2,500. These sums required to purchase a command were relatively high. Prices could vary, but were generally the equivalent of two to three years’ stipend. This indicates that the position of commander, once achieved, opened the door to other opportunities to earn money, and considerably more than the wages paid by the Company. In a business like the VOC, which had so much continuity in its rules and customs over nearly two centuries, and whose directors held office for scores of years without any break, the opportunities pertaining to patronage in order to acquire the posts of commander or first mate probably never changed much. Foreigners who were keen to climb the Company’s ladder to the top had to become familiar with this whole system, a system which in one form or another also existed in their countries of origin. In order to be able to participate in the system of gifting places and payments for this purpose, it was crucial for them to have settled in a city in which the Company had its six Chambers, and to have learnt the local networks and customs. Next, they had to have some capital at their disposal and/or build up sufficient the career ladder of the dutch east india company 221 confidence and credit amongst lenders, who would loan the money for the job’s payment and the purchase of the great quantity of commodities which a commander was allowed to take on board for his private trade overseas. And above all, these foreigners had to be in town, when the announcements for new applications were made. This happened mostly well before the actual departure of the ship—as mentioned, mostly within four or five months. The best way, of course, to come up to the requirements just mentioned, was to live in the city as a married man with a family. And most commanders of foreign origin did.9

The Amsterdam Chamber

By tradition, Amsterdam had the reputation of being the most important city of migrants and foreigners who got naturalized. It continuously offered employment to numerous people, in particular to seamen, but also in the local industries. Golden chances grew on the trees, it was said, for those who could persevere and had good luck.10 This idea from the 17th century still took root in the next century, more than for any other Dutch city, apart from Rotterdam. The other four Chamber cities totally lacked this attraction and fame. The Chamber of Amsterdam offered by far the greatest opportunities to make a career in seafaring and to earn a lot of money. Half of all the East Indiamen were equipped by this Cham- ber, 10 and upwards per year. However, during the first decades of the 18th century, the command of an East-Indiaman by a foreigner was the exception rather than the rule, as shown in table 1. In 1700, Jan de Klunder from Hooksiel in East Frisia provided an early example. Such appointments indeed remained sporadic, despite the rapid increase of the numbers of ships equipped. Around 1740, various North Germans, Scandinavians and other North- ern and Eastern Europeans were appointed commander. Men from Amsterdam had always been the first preference, next officers from the province of Holland. Admittedly, mariners from the Danish Wadden Sea Islands and the adjacent continent had risen to the rank of commander before. These foreigners always adopted the Dutch version of their names, but when Johan de Klerk was appointed commander in 1775, he was given

9 This paragraph is mainly based upon Bruijn, Commanders, chapters 9 and 10. 10 Erika Kuijpers, Migrantenstad. Immigratie en sociale verhoudingen in 17–eeuws Amsterdam (Hilversum, 2005). 222 jaap r. bruijn and femme s. gaastra

Table 1. Provisional numbers of Dutch, foreign and unknown commanders: ( ) those of Amsterdam11 Dutchmen Non-Dutchmen Unknown Total 1700/09 209 (109) 8 (7) 25 (7) 242 (123) 1710/19 287 (138) 0 (0) 19 (7) 306 (145) 1720/29 341 (159) 8 (6) 14 (3) 363 (168) 1730/39 333 (160) 19 (17) 21 (4) 373 (181) 1740/49 231 (74) 50 (26) 14 (4) 286 (104) 1750/59 145 (59) 86 (44) 28 (10) 249 (113) 1760/69 175 (73) 87 (52) 30 (16) 293 (141) 1770/79 212 (95) 62 (41) 25 (20) 299 (156) 1780/89 127 (61) 42 (21) 30 (16) 199 (98) 1790/95 59 (27) 29 (11) 10 (2) 98 (40) Total 2119 (955) 391 (225) 216 (89) 2708 (1269) permission to resume his own name: Johann Alexander von Schkopp! For- eigners in command eventually became so normal, that they also kept their original Christian names: in the administration, Clemens and Ced- erborg were called Johann Carl and Swerus Magnus respectively. From about 1740, it had become normal in Amsterdam that 3 or 4 commanders out of 10 were foreigners. One can find similar developments in the other Chambers. Most commanders from foreign origins, ranked according to their numbers, came from Germany (160), Denmark (57), Sweden (48) and Poland (44). So far, there are only 22 names of men from Norway, and 6 from Belgium.

A few examples of non-Dutch commanders

A good example of an immigrant/commander who enjoyed success both professionally and socially is Frederik Schouten. At a very young age, he had left Stettin in Pomerania—a city from which twenty to thirty inhabitants enlisted annually with the Company—for Amsterdam, where he married in 1731. In 1740, at the age of thirty-five, he became a commander. Schouten was a Lutheran, and had three children of whom one son died in Company service as a first mate. He often spent long periods in the East, and earned a great deal of money. In 1757 he retired. As a result of his

11 The table is based upon preliminary results from the project referred to above. Mr. Van Velzen kindly provided these data, which do not pretend to be complete or exact. the career ladder of the dutch east india company 223 overseas commercial activities, he received several bills of exchange with amounts of f. 12,657, f. 10,000 and nearly f. 12,000. In 1762, his property was assessed at nearly f. 50,000. Jakob Wiebe, born circa 1704 at Elbing in Poland, is another example. He had a splendid career at sea, but was unable to enjoy his considerable earnings as a retired commander. He died in 1766, during his seventh return voyage as commander. Wiebe had become a specialist in the Chi- nese and Ceylonese trade. He sailed home four times from Ceylon, and three times from China. His private business affairs in the East flourished. Between 1746 and 1764, bills of exchange worth more than f. 44,000 were made out in his name, and, over and above his stipend, on five occasions after his arrival home he was paid the usual bonus of f. 2,000. At the age of 42, Wiebe had been fairly old when he was given his first command, but he had not delayed his marriage until this milestone. In 1731, he had already married a woman from North Frisia in Amsterdam. They had four children, of whom their son Jakob, just like his father, died in the service of the VOC. Eschel Juel from Hadersleben in South Jutland was unfortunate. His maiden command in 1764 was a voyage to The Cape, while his second ended in Ceylon in 1766. There he was dismissed from his command for misconduct. He spent years trying to restore his good name, claiming that he had been maltreated by his officers. He was never re-instated, but in 1775 he was finally given a weekly allowance of seven guilders.12 Axel Land from Egersund in Norway, sailing from 1778 onward as a commander for the Chamber of Amsterdam, had another kind of experience. Like four of his colleagues, his ship was captured by the British in Saldanha Bay, North of Cape Town. It happened in July 1781, during the Fourth Anglo-Dutch War. The five commanders were all suspended. Ten months later, Land repatriated on board a Portuguese vessel, first to Lis- bon and then to Amsterdam. His documents and personal papers were all taken by the enemy; even today they are kept at The National Archives in Kew (Surrey). It took some time before he was reappointed commander. He died in Asia, in 1790.13 There are many more examples available of foreign boys and men who reached the top of the seafaring ladder. Most of them came from Lutheran

12 For the careers of Schouten, Wiebe, and Eschel, see Bruijn, Commanders, pp. 115–6. 13 Peter de Bode, Erik van der Doe and Perry Moree, Dirck, Michiel en Benedictus en de slaven van Ardra en andere Sailing Letters. Bundel aangeboden aan Dirk J. Tang (Pijnacker, 2011), pp. 48–9; DAS, II, no. 4594 and III, nos. 8074 and 8083. 224 jaap r. bruijn and femme s. gaastra countries. There was an official rule which stated that Company personnel should belong to the Reformed religion, but in practice this was honored more often in the breach than in the observance. After the appointment of the Lutheran Gustaaf Willem, baron Van Imhoff as governor-general in 1742, even Lutheran churches were built in both Batavia and Cape Town. So Lutheranism did not form an obstacle for a foreigner to become commander. There could be problems with “Romish inclinations” of a candidate. As time passed, and less more preferred candidates were available in the last decades of the century, Roman Catholics were also found in Company service as commanders of East Indiamen.

Assimilation

Most commanders from abroad were completely assimilated in the city they lived in. Materially and socially, they did well. The ordinary fixed stipend of a VOC commander was much higher than that earned by anybody else who sailed for a living. Until 1742, on an annual basis, a commander earned f. 1,000 on average for a round-trip voyage, and hereafter this sum was doubled, and could sometimes be even higher. Because the ships spent less time in Batavia as the century progressed, which meant that the people on board returned home more quickly, a commander had far more chances than before to apply for a new ship. An extensive system of bonuses enlarged his chances of making extra money. Over and above these generous earnings, a commander enjoyed a real opportunity to make very large sums of money from his trade in private goods, as already referred to. There had always been legitimate possibilities to do this, and they were expanded in 1742.14 In general, the behavior of foreign commanders on board and overseas did not differ from that of his Dutch colleagues. They were not fined, punished or downgraded more often than their Dutch counterparts. An exception to this rule was Jochem Outjes from Stockholm, who left the service of the Company not out of his own free will. In 1755, he was given a dishonourable discharge; he had tried to smuggle various contraband goods on board, and was caught red-handed. Outjes later resumed his former career of ship’s master to Surinam and St Eustatius. Before he was appointed commander by the Company in 1747, he had already sailed to

14 For the financial side, see Bruijn. Commanders, chapters 11 and 12. the career ladder of the dutch east india company 225 the West Indies. At the time of his appointment, he had not fulfilled the usual requirements for being a commander. A good example of complete inburgeren (assimilation) is Steven Booms in Enkhuizen. He was a native of Mandal in southern Norway. In 1760, he married the daughter of an Enkhuizen coppersmith. He had been living in the town since his appointment as third mate in 1757. His climb up the ladder was swift, but, during his 4th voyage home as commander in 1771 he had to remain behind at the Cape because of a “severe indisposi- tion.” When he had recovered, he travelled home as a passenger. After this experience, Booms put an end to his sailing career, and stayed ashore in Enkhuizen, where he died in 1795 from a “debilitation in strength,” child- less but belonging to the richest group of citizens. Booms was a good busi- nessman, who exploited the chances for trade which came his way as a mariner to the best of his ability. In the time when he was no longer at sea, he regularly bought debentures and, in partnership with a successful merchant, provided all those who sailed with the Company with con- signments of various kinds of wares, which they could use for their own private trade. Jan de Wit, the man from Aarhus, serves as a last example illustrating the vast opportunities which Dutch society and its labor market offered to people from abroad. Having begun as a boy, by 1736 he had climbed to the rank of third mate and in 1740, as we have seen, he was made a commander on the homeward-bound voyage of a ship of the Rotterdam Chamber. At the very last minute, in 1745, he was appointed commander of the whole return fleet when the intended commander suddenly died. This voyage with a fleet of seven ships succeeded well and, in September 1746, De Wit was presented with the usual tokens of approval by the Gen- tlemen Seventeen. His third command for the Chamber of Amsterdam was from 1749 till 1753. After his return, De Wit did not lose much time before he departed for Dokkum in Friesland, where he married a wealthy widow. In 1758, he could afford to buy the estate Staniastate in Oenkerk, which he named after his first command, the Hofwegen. He died in 1781. In the house on his estate, De Wit kept something special, a finely carved model of a ship in ivory from Canton. It was not a junk, but what was known as a “flower boat” which sailed back and forth on the Pearl River. It was the sort of vessel on which Chinese prostitutes received their clients. They were absolutely forbidden territory for Europeans. Whether De Wit’s staff and visitors understood the true nature of this model, is not known for certain. 226 jaap r. bruijn and femme s. gaastra

In 1700, Jan de Klunder was one of the few foreign commanders in Company service. In the final days of the Company, between 1790 and 1795, exactly half of the commanders were foreigners. And they were no longer all “true” Company servants. Only some had made their careers with the Company, the rest had begun either with the merchant navy or the whaling fleet. In summary: one cannot deny foreigners could climb the ladder to the top in the VOC fleet.

Junior merchants: appointment

The junior merchant rank became the best start for an on-land career within the Company, which, in the end, could bring to a young man the fortune he was looking for. A junior merchant received a salary of 36 guilders per month, but it was not the salary but the possibilities to earn a considerable unofficial income that made this a coveted post. The administration of the warehouses in Batavia provided ample opportunities to make a small fortune, as the profits varied between 12,000 to 50,000 guilders per year. But there were even better positions: the two administrators on the small island of Onrust near Batavia, where the VOC had its repair facilities for ships and a number of warehouses as well, could, according to some sources, earn nearly 200,000 per year.15 It was not any different from the commanders’ posts: the directors made deals among themselves about these appointments, every director taking his turn in proposing his favorite or relative for the job. In 1771, it became an official rule that on every first-rate East Indiaman (i.e. the larger ships), one junior merchant was appointed. However, this rule implied that the smaller Chambers (Delft, Rotterdam, Hoorn and Enkhuizen) would not have the opportunity to hire a junior merchant, since they built and equipped only East Indiamen of the second charter. The rule was therefore adjusted in 1772. All of these four Chambers were now allowed to appoint a junior merchant once every two years: the first year Delft and Hoorn received the privilege, the next year Rotterdam and Enkhuizen, and so on. But more adjustments were necessary. Stadtholder William V, in his capacity as opperbewindhebber, wanted to have some junior mer-

15 Femme S. Gaastra, “Soldiers and Merchants. Aspects of Migration from Europe to Asia in the Dutch East India Company in the Eighteenth Century” in Wim Klooster (ed.), Migration, Trade, and Slavery in an Expanding World. Essays in Honor of Pieter Emmer (Leiden and Boston, 2009), pp. 99–116. See Table 3 on p. 115. the career ladder of the dutch east india company 227 chant positions at his disposal as well. The Gentlemen Seventeen gave their consent, and decided that Amsterdam had to cede two places and Zeeland one, which would provide the Stadtholder with three junior merchant places.16 It was also decided that promotions from lower ranks to junior merchant in Asia had to be deducted from the available positions in the Netherlands. Such promotions needed the consent of the directors in the Netherlands, who could choose to fill their assigned places in this way. Finally, in 1777, the Gentlemen Seventeen agreed that the minimum age for a junior merchant should be 20 years. It meant in practice that some ships sailed without a junior merchant on board, and that it became even more difficult for young men to get such a position in the Netherlands. These arrangements can also been seen as an attempt by the directors at home to gain tighter control over their personnel overseas. The appointments, not only those in the Netherlands, became more and more a matter of being the protégé of one of the directors, or from the Stadtholder. The available records of the autumn meeting of the Gentlemen Seventeen contain long lists of approval—and sometimes disapproval—of promotion of the staff in the Company’s factories in Asia; the Resolution Books of the directors of Zeeland precisely record whose turn it was to appointment someone in the rank of junior merchants.17 Despite the restrictions in the Netherlands, a fair number of junior merchants sailed out every year. It is not clear, if it was normal practice to pay for an appointment, but occasionally it may have helped. It is well-known that in 1721 the director Willem Sautijn of the Chamber Amsterdam asked 3,500 guilders for an appointment as junior merchant. In 1725, Balthasar Boreel, the director of the Chamber Amsterdam, paid ƒ 2,000 for the appointment of Gustaaf Willem, baron van Imhoff who later became governor-general, to the position of junior merchant.18 However, having the support from the directors at home for an appointment was itself not enough for a profitable career overseas. Letters of

16 National Archive, The Hague (NA), VOC archive no. 181, resolutions Gentlemen XVII, April 17 and October 9, 1772. 17 These are the so-called “Lijsten van recommendatie van officieren voor de equipage,” containing names of other officers as well, such as the masters and other officers of the ships. See Femme S. Gaastra, “Zeeuwen in de VOC,” in Jan Parmentier (ed.), Noord-Zuid in Oost-Indisch perspectief (Zutphen, 2005), pp. 99–118. 18 Femme S. Gaastra, “De VOC als werkgever,” in Jan Parmentier (ed.), Uitgevaren voor de Kamer Zeeland (Zutphen, 2006), pp. 31–46; A.J.P. Raat, The Life of Governor Joan Gideon Loten (1710–1789). A Personal History of a Dutch Virtuoso (Hilversum, 2010), p. 56. 228 jaap r. bruijn and femme s. gaastra recommendation to patrons overseas were also necessary to get ahead and gain access to the best jobs and positions abroad. The letters of the young August van Greven, who had entered the Company’s service in 1771 as a gunner’s mate, provide a vivid picture of what was in reality required. First, he tried to obtain a post as junior merchant. Once he had achieved it, he thought he had a position from which he could work his way up through the ranks on his own strength. But soon after his arrival in Bata- via, he realized that he needed more and enduring support to get anywhere. He therefore wrote home that he was in urgent need for more recommendations. “Without yearly recommendations from patria, one cannot make fortune or promotion.”19 Van Greven succeeded, because he could depend on a strong and influential network at home and in Bata- via. The influential politician Adriaan Bergsma at The Hague, who had close connections with Stadtholder William V; the governor of the Cape of Good Hope Joachim Plettenberg; and the member of the Council of the Indies and future governor-general Willem Arnold Alting—they all played a role in this network. The whole system of recommendations and patronage led to a situation in which, according to the resident Roelof Palm in Padang (Sumatra), the promotion went to the personal favorite rather than to most skilled person.20 There was some controversy about the value of the recommendations— were the recommendations given by the Stadtholder more weighty than those given by de advocaat (solicitor) of the Company, who had a central position in the VOC’s directorate? Were recommendations sent from the Netherlands strong enough in Batavia, where Indo-European families or clans had such a powerful position? Recommendations from patria were often expressed in general terms, requesting “advancement,” while the Council of the Indies could decide over the jobs that were actually allotted—and there was great variation in the profit and extra earnings connected with each position. The right of appointment to certain extraordinarily lucrative positions, such as resident of Cheribon (the coffee district near Batavia) was seen as a privilege for the governor-general. But

19 KITLV/ Royal Netherlands Institute of Southeast Asian and Caribbean Studies, Manuscripts and Archives, no. 169, Coll. De Blau, August van Greven to Quirijn de Blau in Leeuwarden, 19 februari 1772. (“Daar ik thans hier zijnde noodzakelijk recommandatie van doen heb, want zo men jaarlijks niet met recommandatie wordt onderhouden, die uit het vaderland komt, so is hier geen fortuin nog avancement voor ymant te wagten.”) 20 NA, Archive Hope, no. 36, Roelof Palm to Thomas Hope, Padang, May 9, 1769. Pro- motions in Asia were “niet altoos na de kunst, maar veeltijds na de gunst.” the career ladder of the dutch east india company 229 however mighty the clans in Batavia might have been, recommendations from patria remained necessary. Dirk van Hogendorp, who made a career in the VOC administration in Bengal and at Java, and later criticized the Company sharply for its conservative policy, noted in 1785 that the Company solicitor enjoyed an “unlimited credit”; his recommendations were conclusive. But that was during the years when William V had temporally lost his influence in the Republic and when his opponents, the Patriots, had become the most influential party in Holland. In 1789, the situation changed, and Van Hogendorp admitted that letters issued by William V carried more weight once again.21 Another employee explained, that the letters of recommendation should never include phrases such as “if the person deserves the advancement,” or comparable words that could be used as an excuse for the promotion of the person in question. The best letters were handwrit- ten by the Stadtholder himself, brief, and formulated as an order rather than as a request.22 Still, patience was necessary. Governor-general Jacob Mossel reported once that at least three letters were common practice. The first letter was regarded as an announcement, the second served to bring the person to attention, while only a third letter could lead to the promotion.23

Junior merchants: numbers

The database of the project referred to at the start of this article shows that 1,127 onderkooplieden (junior merchants) sailed on board of Dutch East Indiamen to Asia (see table 2). It is remarkable, though not surprising, that so many junior merchants originated from the city in the province from where they sailed. It is clear, that the directors often awarded this privileged position to fellow townsmen, to relatives and family members. Among the 117 junior merchants from Zeeland, 72 came from Middelburg

21 E. Du Perron-de Roos (ed.), “Correspondentie van Dirk van Hogendorp met zijn broe- der Gijsbert Karel,” Bijdragen tot de Taal-, Land- en Volkenkunde van Nederlandsch Indië 102, 1 (1943), pp. 134, 148. 22 L. Brummel, “Achttiende-eeuws kolonialisme in brieven,” Bijdragen en Nededelingen betreffende de Geschiedenis der Nederlanden, 87 (1972), pp. 171–204. 23 Ibidem, p. 101; Mossel remarked “dat de eerste brieven van aanbeveling dienen om in kennisse te brengen, de tweede, om d’attentie op te wekken en te fixeeren; maer dat de derde de saeckelijke begunstinge en bevordering daer stelden.” 230 jaap r. bruijn and femme s. gaastra

Table 2. Number of junior merchants sent to Asia (1700–1795) Total number Own province Foreigners VOC area (Zeeland) or city Zeeland 289 117 21 34 Delft 40 10 1 2 Rotterdam 45 9 3 8 Hoorn 41 15 0 5 Enkhuizen 21 15 1 0 Amsterdam 685 220 58 46 Total 1127 387 84 95 and 17 from Vlissingen. The Chamber Amsterdam allotted 220 places to Amsterdam residents, while other Chambers appointed another 30 Amsterdam citizens to junior merchant positions. Remarkable is the fact that many of this group of junior merchants were born in the VOC settlements overseas: no less than 95 of them, including 48 who stated Batavia as their place of birth. It fits into a pattern where Batavian families sent their sons to Europe to be educated. After their education, they would enter Company service and return to Asia, following in the footsteps of their father or family members to pursue a career in Asia. The proportion of foreigners hired was low, only 7.5 percent. That is no great surprise; it must have been difficult for outsiders to enter the group of privileged VOC employees. In fact, the proportion is even smaller than the figures indicate, because the figures are derived from the place of birth stated in the ship’s ledgers. The fact that someone was born some- where outside the Dutch Republic did not necessarily mean that he was not a Dutchman. No doubt Company clerks often wrote foreign names in a Dutch way, but in some cases we can only conclude that the young merchant entering the Company’s service was a Dutchman born in a city abroad. Thus for example, we can assume that “Cornelis Cruys from Peters- burg” and “Paul Nannings from Moscow” were Dutchmen, and not Rus- sians. “Carel, count of Gronsveld” who stated Berlin as his birthplace was Dutch as well. “Barend, baron van Neukirchen” (or Nieuwkerken) named as Nijvenheim came from Cleve and belonged to the German aristocracy (Uradel), although he had strong links with the Netherlands. He sailed as a junior merchant in 1772, His brother Evert Jan, born in Tiel, also served in the VOC. The brothers had been supported by William V; Evert Jan became resident of Cheribon in 1771, one of the most lucrative posts on the career ladder of the dutch east india company 231

Java, while Barend was appointed as the administrator of Onrust in 1770. Barend returned to Europe in 1777 and was knighted in Nijmegen.24 Boudewijn Versélewel Faure, who embarked in 1756 and eventually became director of the VOC factory of Bengal, was born in Dendermonde in the Southern Netherlands; his father was an officer in the Dutch army. Pieter van Teylingen, from London, must have been a member of a well- known VOC-family. George van Motman, who sailed in 1790 aboard the ship Gerechtigheid to Batavia, stated “Ginipperhaus in’t Cleefse” (nowadays Genneperhuis near Gennep, a place at the German border) as his birthplace. His father worked as rentmeester for the Prussian King as well as for William V. More of the men fell in this category, and sometimes it is difficult to distinguish clearly between “Dutchmen” and “non-Dutchmen.” Among the “foreigners”—or more neutrally, among those who were born outside the Dutch Republic—junior merchants from Germany had the biggest share with 53 towns of origin. Bremen and Frankfurt both occur 5 times in the list. Junior merchants not only originated from coastal areas or harbor cities, but also from cities and towns quite far from the seaboard. Listed are not only Dantzig and Koningsbergen (both three times) or Hamburg, but also places such as Hanau, Hannover, Berlin, Magdeburg and Lipstadt. Other junior merchants came from Denmark (Copenhagen), Sweden (Göteborg), France (Paris, Metz, Nantes), the Southern Nether- lands (Dendermonde), Switzerland (Bern) and even Russia (Petersburg and Moscow). The number of junior merchants on the outgoing East Indiamen reduced in the course of the 18th century. During the first quarter of the century, 407 junior merchants left patria for Asia, while during the 25 years from 1770 to 1795 only 174 sailed out. It is not clear whether this decline was a direct result of the hiring policies of the Gentlemen Seven- teen. The trend was the same for all the Chambers. The number of junior merchants hired by the Chamber Rotterdam, for instance, dropped from 31 during the period 1700–1724, to 6 in the years 1770–1795. The numbers for Delft were 28 and 2 respectively. The numbers of those from foreign origin likewise shows a decline. During the period 1700–1724, a total of 24 junior merchants originating from places outside the Dutch Republic

24 S.A.C. Dudok van Heel, “Familie van de gouverneurs-generaal. De afstammelingen van Dirck Comans (Hoorn 1648–Batavia 1708),” Jaarboek Centraal Bureau voor Genealogie, 56 (2002), p. 171. 232 jaap r. bruijn and femme s. gaastra left Holland, as against 12 in the period 1770–1795. Among these 12 junior merchants, 6 can be identified as Dutchmen rather than foreigners. Despite the decline at the end of the 18th century, there seems to have been a surplus rather than a shortage of junior merchants in Batavia. The phenomenon of a junior merchant without a job (“onderkoopman zonder emplooi”) was common. Junior merchants arriving from patria had sometimes to wait for months before they were placed in the administration of one of the warehouses or other offices, or before they were sent to the factories of the VOC elsewhere in Asia. Dirk van Hogendorp noted in 1791 that some 20 junior merchants had arrived in Batavia, all “with letters” (of recommendation).25 The young junior merchant Reynier Hoynck van Papendrecht, who arrived in Batavia in June 1778, and who was four months later still looking for a job, remarked sadly that he was not the only one without a proper job, since some twenty of his colleagues found themselves in the same predicament.26 If one was strongly supported by powerful patrons, that made all the difference. Johannes Bergsma, who arrived in 1778 at Batavia, wrote home cheerfully: “Luck follows me always. Within six weeks, I have obtained one of the most profitable administrations, after Onrust the best, from which I can live a pleasant life and put aside some pennies as well.” Johannes Bergsma could rely on the same network that supported August van Greven. Indeed, Bergsma even wrote that Willem Alting cared for him as if he was his son! The administration where he was placed was the provisioning warehouse, a function which— according to some sources—yielded a yearly income of 48,000 guilders.27 During the 18th century, the great majority of the young merchants never returned to Europe. In total, only some 23 per cent or 263 men ever went home again, in many cases with at least a moderate amount of capital, and in some cases as very rich persons. This percentage was almost the same for the group of foreigners, of whom 17 (20 percent), eventually sailed home. Many others lived overseas for many years and made their whole career there. On average, a junior merchants was in the Company’s service in Asia for 9.5 years, regardless of how they ended their career

25 Du Perron-de Roos, “Correspondentie van Dirk van Hogendorp,” p. 162. 26 P.C. Hoynck van Papendrecht (ed.), “Some Private Letters from the Cape, Batavia and Malacca (1778–1788),” Journal of the Malaysian Branch of the Royal Asiatic Society, 2, 1 (1924), pp. 16–7. 27 Coll. De Blau, J.C. Bergsma aan Q. de Blau, Batavia 17 mei 1774 (“Het fortuin loopt me altijd na! Na 6 weken al één van de vetste administratiën gekregen (buiten Onrust de beste), waarvan ik ruim in vrolijkheid kan leven en een klein stuivertje overhouden”). the career ladder of the dutch east india company 233

(by repatriating, by death or otherwise). There were some exceptionally long terms of service. Hendrik Scoffier from London was appointed as junior merchant in 1751, and served in the VOC for 34 years until 1785. The employment of Jean Rosier from La Rochelle’s lasted 33 years, from 1755 until 1788. Plenty evidence exists to show that many of these merchants had successful careers overseas, and became part of the overseas VOC-elite.

Junior merchants: the road to the top

The only one of the foreign junior merchants in the 18th century who went to the top of the Company hierarchy in Asia was Gustaaf Willem, baron van Imhoff. He was born in Leer, in Germany, but in reality not an such an outsider as his birthplace might suggest. His mother was Sophia van Boreel, and the Boreel family belonged to the highest circles in Amster- dam. His uncle Balthasar van Boreel was a bewindhebber (director) in the Chamber Amsterdam of the VOC, and had helped him to get his appointment. He sailed out in 1725, was already in 1726 promoted to merchant status, and contracted a very good marriage in 1727—all indications that he could rely on support from mighty patrons. His life has been described in many studies.28 It is noteworthy that his two brothers sailed out as “onderkooplieden” as well—Hieronimus in 1734 and Hendrik in 1741. Of interest is that Van Imhoff’s predecessor, Johannes Thedens (1679– 1748), was not such an “outsider” either. Thedens first embarked at the age of 18 with the low rank of a soldier. He originated from Friedrichstad, a town founded in 1621 by Dutch protestant dissenters (remonstranten) in Holstein. Many families from Friedrichstad had close connections with the Netherlands. Thedens’ father was married to Anna van Loo from Haar- lem. Her brother, Jacob van Loo, had been the head of VOC factory in Tonkin, and later worked as merchant in Batavia. He would have been a great help for Johannes Thedens, who abandoned his job as soldier immediately after his arrival in Batavia, in 1699, and acquired a place in the administration, from where he worked his way up. In 1731, he became a member of the Council of the Indies, and in November 1741 he was chosen as provisional governor-general.

28 See N.J. Krom, Gouverneur-generaal Gustaaf Willem van Imhoff (Amsterdam, 1941); Lodewijk Wagenaar et al. (eds), Gouverneur Van Imhoff op dienstreis in 1739 [Werken uit- gegeven door de Linschoten-Vereeniging, 106] (Zutphen, 2007). 234 jaap r. bruijn and femme s. gaastra

There were more successful young merchants. Elias Guillot from Bor- deaux had arrived in Asia in 1714. He served mainly on the Dutch settlements at the Coromandel coast and became in 1733 director of the VOC factory there. Five years later, he obtained a seat in the Council of the Indies in Batavia (as was usual, first in an extra-ordinary capacity, and then, in 1742, as ordinary councillor). He died in 1743.29 Johan Theling from Holstein, who served in the VOC from 1723 until his death in 1747, also became a councilor at the end of his career, in 1754. Another example of a successful career is that of Dithard van Rheden, from Bremen. Van Rheden, who arrived in Batavia in 1741 and died in 1759, became administrator of Onrust—as said, one of the most lucrative posts in Batavia. Later he became a member of the Council of the Indies. Van Rheden’s letters show, again, how patronage and protection functioned in Batavia. He was an ardent supporter of Lubbert Jan van Eck, who became governor of the VOC’s Coromandel factory in 1758 and finally, in 1762, governor of Ceylon. Van Rheden did his utmost to have Van Eck promoted—and sometimes was engaged in trade relations with his protégé as well. But in his letters he complained about the numerous requests for promotions, and about the number of recommendations he received. It was simply impossible to satisfy everyone. “You have what you have, but to get it, that is the trick.” The good relationship with Van Eck contributed to Van Rheden’s prosperity as well. He was engaged in several commercial enterprises with Van Eck. After his death in 1761, his heirs in Bremen received ƒ 332,139, money that was transferred via bills of exchange from Batavia to Amsterdam.30

Conclusion

Foreigners were able to make their way to the highest positions within the VOC. Yet that did not mean that they had the same chances to obtain these positions. Foreign commanders of East Indiamen were quite usual and their numbers increased in the course of the 18th century, partly because of a shortage of qualified Dutch sailors. Foreigners in the ranks of the junior merchants were an exception, however. While there was no formal obstacle for foreigners to obtain such positions, the practical

29 W. Wijnaendts van Resandt, De gezaghebbers der Oost-Indische Compagnie op hare buiten-comptoiren in Azië (Amsterdam, 1944), p. 110. 30 NA, Coll van Eck, no. 62, 64; NA, VOC Archive, no. 2971, f. 811–819, list of bills of exchange, 15 October 1760. the career ladder of the dutch east india company 235 obstacles were considerable. The number of places available was limited, and it was difficult to obtain the desired position without support from the directors, or from circles close to the directors. The same applied to the careers in Asia. Successful careers could be made, if the young VOC employees gained entry to local overseas networks, and had connections with the influential families or clans, in combination with continuous support from the Netherlands. A foreigner who already had connections with such networks, and who was less an “outsider,” was better placed to succeed overseas than those who did not have such contacts.

Demographic Change and Migration Flows in Holland between 1500 and 1800

Jan Luiten van Zanden and Maarten Prak

The long-term trajectory of Holland’s population in the early modern period is well-known: it grew from about 275,000 in 1514 to a peak of about 880,000 in the 1670s, after which it most likely reduced to about 780,000 in 1750, remaining fairly constant during the rest of the 18th century.1 But we know much less about the demographic factors explaining this curve. What was the balance between births and deaths? How much immigration occurred, and how many people left the region? The reason why we do not have a definite answers to such questions is that population reconstructions based on the vital demographic events have not previously been carried out for Holland. The Wageningen School, which from the 1960s to the 1980s provided most of the new research findings about the demographic history of the Netherlands, concluded that such a reconstruction was not a feasible option, because of the limitations of the available historical records (the registration of births and deaths). A synthesis of all the demographic data obtainable for the early modern period, such as produced by the Cambridge Group for England, has not yet been undertaken here. That means we still lack knowledge about the basics of demographic development before 1800. It is true that a study was undertaken for the city of Amsterdam only—more or less along the lines of the Cambridge Group—by Van Leeuwen and Oeppen. But their backward projection stops in the 1680s, again due to the limitations of the base data obtainable from historical records before that time.2 However, estimates of annual population growth were required for the project “Reconstructing National Accounts of Holland before 1800.” At first, we tried to obtain a series by means of interpolation, using benchmark years for which fairly reliable estimates are available: 1514 (275,000),

1 See the summary in Jan Lucassen, “Immigranten in Holland 1600–1800. Een kwanti- tatieve benadering,” CGM Working Papers, 3 (2002). 2 Marco H.D. van Leeuwen and James E. Oeppen, “Reconstructing the Demographic Regime of Amsterdam 1681–1920,” Economic and Social History in the Netherlands, 5 (1993), pp. 61–102. 238 jan luiten van zanden and maarten prak

1570 (c. 400,000), 1622 (672,000), 1670 (880,000), 1750 (783,000) and 1795 (783,000).3 But that procedure was not very satisfactory. In the first place, the literature tells us that there were years of civil war (1572–1580) and plague epidemics during the period of interest (1624–1627, 1635–1638, 1653–1657, and 1666–1667). At those times, the population may actually have declined, and the death rate would certainly have been much higher than the birth rate.4 But more importantly, it remains unclear if such interpolated estimates are consistent with the available evidence for other aspects of the demographic system: how do the estimates relate to the strong increase in the average age of marriage that occurred? And how did urbanization affect this pattern? Was the evidence consistent with what we know about migration flows? Jan Lucassen had already been grappling with some of these questions when he constructed the first systematic estimates of migration into Holland between 1580 and 1800.5 Is it possible to improve his estimates? We can try.

The strategy

We do not know much about Holland’s demographic system in the period before 1800. The little we do know, however, we can incorporate into one set of estimates of the main demographic parameters—and then see what results we get. So, what do we know? We already mentioned the benchmark estimates for 1514 until 1795; we see no reason to challenge them. The same sources also inform us about the share of the urban population in the total, which increased from 45% in 1514 and 1570 to 60% in 1670, and stabilized afterwards.6 We also accept Van Leeuwen and Oeppen’s reconstruction of the demographic development of Amsterdam; thus, we derive estimates of the birth rate and the death rate for the period 1680–1800 from that source (but we do

3 Jan de Vries, “The Decline and Rise of the Dutch Economy, 1675–1900,” in G. Saxon- house and G. Wright (eds.), Forms and Methods in Economic History: Essays in Honor of William N. Parker (Greenwich, 1984b), pp. 149–89; Jan de Vries and Ad van der Woude, The First Modern Economy. Success, Failure and Perseverance of the Dutch Economy, 1500–1815 (Cambridge, 1997); also Lucassen, “Immigranten in Holland”. 4 Leo Noordegraaf and Gerrit Valk, De Gave Gods. De pest in Holland vanaf de late mid- deleeuwen (Amsterdam, 19962); Ronald Rommes, “Pest in perspectief. Aspecten van een gevreesde ziekte in de vroegmoderne tijd,” Tijdschrift voor Sociale Geschiedenis, 16 (1990), pp. 244–66. 5 Lucassen, “Immigranten in Holland”. 6 Ibidem. demographic change and migration flows in holland 239 not use their estimates of the size of the Amsterdam population in our reconstruction).7 Moreover, we assume that these death rates and birth rates are representative for the urban population of Holland as a whole. This is, perhaps, a rather “pessimistic” assumption, because Amsterdam was the largest city, and may therefore have had the “worst” demographic structure (in particular a (very) high death rate compared with a (very) low birth rate). Other towns—such as The Hague—may have had a more favorable balance between births and deaths, but there is insufficient information to take this into account. In addition, we also assume that the urban death rate before 1680 did not change, except for the years of excess mortality which are well documented in the literature.8 Between 1680 and 1800, the Amsterdam death rate was also quite stable in the long run. To be more precise: there was a decline between the 1680s and the mid 18th century—from about 35‰ to 30‰—followed by an increase to about 38‰ in the 1790s, but the average for the second half of the 18th century was very close to the values of the 1680s and 1690s.9 To estimate the urban birth rate before 1680, we assume a link with the average age at marriage for women, which is known for Holland in the 16th century (about 20 years in 1500 and 1540), and for Amsterdam from the 1580s onward, where it increased from 24 years in the 1580s and 1590s to 25.7 in the 1620s, 27.7 in the 1680s and 28.7 in the 1780s.10 The assumption is that a delay of marriage leads to a decline in fertility, a link very well established in the literature about the premodern demographic system. The rise in the average age of marriage from 20 to almost 29 years must have depressed the average birth rate significantly—although it is difficult to establish by how much. Additional assumptions concern the demographics of the countryside: we assume that the rural birth rate was higher and the rural death rate lower, and that this difference did not change over time, and was therefore the same between 1514 and 1800. We now have a lot of speculations and a few unknown variables, in particular about the link between the age of marriage of women and the birth rate, and about the urban-rural gap in birth and death rates. And we do not know the size of the net migration that occurred in this period; estimates of this magnitude will be among the (highly speculative)

7 Van Leeuwen and Oeppen, “Reconstructing the Demographic Regime of Amsterdam.” 8 Taken from Noordegraaf and Valk, De Gave Gods and Rommes, “Pest in perspectief.” 9 Van Leeuwen and Oeppen, “Reconstructing the Demographic Regime of Amsterdam.” 10 Jan Luiten van Zanden, The Rise and Decline of Holland’s economy. Merchant Capital- ism and the Labour Market (Manchester, 1993). 240 jan luiten van zanden and maarten prak results of this experiment. We can calibrate these estimates, however, via one further assumption—that net migration for Holland before 1570 was zero. Some immigration would have occurred, of course (as is for example clear from the lists of new burghers of Amsterdam in this period), but emigration was probably almost equal in size and scope: merchants and some skilled craftsmen moved to the Baltic and to Flanders, but not on any large scale. If we assume that net immigration was zero between 1514 and 1570, it follows that the natural increase of the population—the difference between the birth rate and the death rate—must be equal to the total growth from 275,000 to 400,000 inhabitants. This finding allows us to experiment with various combinations of the three unknown variables (the urban-rural gap in natural population increase, and the link between the age of marriage and the birth rate). The system is clearly under-identified, in the sense that several different solutions of the three variables can lead to the observed population growth between 1514 and 1570, but the most plausible set of variables is as follows: the rural death rate is 5‰ lower than the urban death rate, the rural birth rate is 5‰ higher than the urban birth rate, and an increase in the age of marriage by one year leads to a decline in the (urban) birth rate of 1‰. An “urban graveyard” effect of in total 10‰ (or 1%) is likely, and consistent with data published by De Vries for the early 19th century.11 If we accept this 10‰ estimate, then the link between the birth rate and the age of marriage directly follows from the calibration. The result is a set of estimates of the natural increase of the urban and rural population of Holland between 1510 and 1800. The gap between actual population growth and natural increase is our estimate of net immigration—the final assumption we make is that immigrants settled in the cities, and became part of the urban population.12

The results

As explained above, we obtained very rough estimates of the basic demographic features of Holland between 1510 and 1800. We estimated the crude death rate and the crude birth rate for cities (on the basis of the Amster- dam data) and for the countryside. The results are presented in Table 1.

11 Jan de Vries, European Urbanization 1500–1800 (London, 1984), p. 194. 12 The assumption that part of them settled in the countryside has, however, only marginal effects on the overall estimates. demographic change and migration flows in holland 241

Table 1. Estimates of the crude death rate and the crude birth rate in Holland, 1510–1800 (in ‰) Cities Countryside Death rate Birth rate Natural Death rate Birth rate Natural increase increase 1510–1572 35,6 35,9 0,3 30,6 41,4 10,8 1572–1600 37,9 32,1 –5,8 32,9 37,1 4,2 1601–1700 37,2 30,9 –6,3 32,2 35,9 3,7 1701–1800 34,2 30,8 –3,4 29,2 35,8 6,6 1510–1800 35,9 32,1 –3,8 30,9 37,2 6,3 Sources: see the text.

In the 18th century, there is a relatively small gap of 3.4‰ between the birth and the death rates of Amsterdam, a result entirely based on the Van Leeuwen and Oeppen study.13 Jan and Leo Lucassen estimated that the “urban graveyard” effect for European cities in the 18th century was about 5‰,14 marginally higher than the Amsterdam estimates obtained here (the gap is far from constant, however; in 1811–1815 it had increased to almost 6‰, but after 1816 the number of births exceeded deaths for more than a decade).15 In the 17th century, the rate of natural decrease was higher—about 6‰—but before 1600 we find a small positive balance of births over deaths, mainly due to the effect that the low average age of marriage has on the birth rate. That such a scenario is not unlikely is shown by the demographic history of York in the same period, where Galley also found a positive natural increase in the 16th century, followed after 1603 by a negative balance between births and deaths.16 In view of the much smaller size of cities in 16th-century Holland, a positive rate of natural increase is also plausible. As mentioned, our estimates, which take the Amsterdam data as representative of the cities of Holland as a whole, tend to err on the pessimistic side, and therefore probably overestimate the urban death rate in this

13 Van Leeuwen and Oeppen, “Reconstructing the Demographic Regime of Amsterdam.” 14 Jan Lucassen and Leo Lucassen, “The Mobility Transition Revisited, 1500–1900: What the Case of Europe Can Offer to Global History,” Journal of Global History, 4, 3 (2009), pp. 347–77. 15 See van Leeuwen and Oeppen, “Reconstructing the Demographic Regime of Amsterdam.” 16 C. Galley, “A Model of Early Modern Urban Demography,” Economic History Review, 48, 3 (1995), pp. 448–69. 242 jan luiten van zanden and maarten prak period. Still, as regards the 16th and 17th centuries, the estimated rate of natural decrease is much lower than the rate calculated by Lucassen and Lucassen for a sample of European cities in this period (they estimated a rate of natural decrease of 1% on average).17 The countryside almost by definition shows a consistently positive net balance between births and deaths, which was quite high in the 16th century; a large number of net births was required to increase the population from 275,000 to 400,000 in this period, without migration. The 18th century saw a rather strong reduction in the death rate, which was caused by the disappearance of years of excess mortality due to large plague epidemics; the last major crisis was in the 1660s. This was followed by what was probably a general decline of mortality during the second half of the 18th century.18 Another way to present our results is to look at the difference between the actual population growth of Holland, and what it would have been without immigration—the result of the natural increase of the population. Figure 1 below presents these estimates: in reality, the population grew from 400,000 in 1570 to 880,000 in 1670, but after 1580 only a small part of this growth was due to natural increase. For Holland as a whole, natural increase became negative after the 1620s. The sharp decline in natural increase between the mid-16th century, when it was still positive at about 6‰, and the mid-17th century when it became negative, was caused by two changes: the rise of the female age of marriage from 20 to 26 years, depressing the birth rate by 6‰, and the increase in the share of cities in the total population from 45% to 60%, which brought about a reduction in the rate of natural increase of 1.5‰. Population growth became entirely dependent on the inflow of migrants; on average they may have numbered between 6,000 and 10,000 annually during the classic “Golden Age” between 1590 and 1670. After 1670, however, the level of net immigration declined sharply (Figure 1). Table 2 presents the same estimates in another way. Just as before, we assume there is no net immigration to Holland between 1510 and 1570, but the growth of city populations is made possible by a very small urban surplus of births over deaths (of less than 2,000), and the net migration of 125,000 people from countryside to towns. Between 1570 and 1700, by contrast, urban growth is entirely dominated by the large immigration

17 Lucassen and Lucassen, “The Mobility Transition Revisited.” 18 See in particular D.J. Noordam, Leven in Maasland (Hilversum, 1986), p. 93. demographic change and migration flows in holland 243

1000

900 90

800

70 700

600 50 population without 500 immigration Population with 400 immigration 30 net immigration 300

200 10

100

0 -10 0 1510 1530 1550 1570 1590 1610 1630 1650 1670 1690 171 1730 1750 1770 1790 Figure 1. The population curve of Holland, 1510–1800: benchmark estimates and simulated results without immigration (left-hand axis), and estimates of the level of net immigration (right-hand axis) movement of those days, and people from the countryside in Holland play only a secondary role in urban growth. After about 1670 (in this table, 1700), this changes again; immigration from outside the region drops sharply, and migrants from the countryside again supply a large part of the demographic gap of the cities. Our estimates of the urban deficit of Holland in this period are lower than those presented in the classic study on the Dutch economy in 1500–1815 by De Vries and Van der Woude; they arrived at a total deficit of 750,000 for the period 1580–1800, compared to a deficit of 800,000 for the whole 1500–1800 period estimated by Lucassen.19 These estimates are probably too high: 750,000 in 320 years would mean 2,300 per year, which, assuming a 5‰ net deficit,20 implies an average urban population of 469,000, which is too large: the average urban population between 1580 and 1800 is only 430,000. More importantly, in spite of the fact that we base our estimates

19 De Vries and Van der Woude, First Modern Economy, p. 98; Lucassen, “Immigranten in Holland,” p. 8. 20 De Vries and Van der Woude, First Modern Economy, p. 98. 244 jan luiten van zanden and maarten prak

Table 2. Sources of population growth and immigration in Holland, 1510–1800 (in thousands) Total From the Immigration Rural migration to countryside in from outside Urban surplus of Net surplus/ the cities Holland Holland deficit births deficit 1510–1572 125 125 0 2 125 126 1572–1600 129 22 107 –31 27 –4 1601–1700 615 95 520 –265 121 –144 1701–1800 94 47 47 –162 211 49 Total 962 288 674 –456 483 27 Sources: see the text.

on the Amsterdam figures (which are relatively pessimistic), the average urban deficit is at 3.8‰ significantly smaller than 5‰, and much smaller than the 10‰ assumed by Lucassen and Lucassen in 2009.

Conclusion

We presented estimates of the basic demographic data for Holland in the period 1500–1800 which can be derived from previous studies, in combination with a number of assumptions about, for example, the link between the average age of marriage of women and the birth rate. The result is a plausible scenario for the demographic development of Holland in this period. Perhaps the most interesting finding of our research is that cities in the 16th century may still have had a positive natural rate of increase, due to the low average age of marriage, their still limited size, and the lower frequency and virulence of the plague. This changed after 1570, however, and between 1630 and 1700 the natural increase of the entire population of Holland became negative—a change which was due to urbanization, plague epidemics and the rising age of marriage. After 1700, the death rate in cities and the countryside started to fall again, which was partly caused by the disappearance of the plague. But this progress was not maintained during the second half of the 18th century: high urban death rates returned at the end of the 18th century, perhaps caused by a decline of economic activity and real wages. The new series we have constructed also has implications for estimates for the level of immigration to (the cities of ) Holland during this period. Our results are somewhat lower than previous estimates, because the “urban graveyard” effect is probably smaller than assumed by De Vries and demographic change and migration flows in holland 245

Van der Woude, Lucassen, and Lucassen & Lucassen21—despite the likeli- hood that our estimates, by assuming that Amsterdam death rates were representative for all cities in Holland, are too pessimistic. This does not, however, alter the fact that from the 1620s onwards, population growth in Holland was entirely due to immigration.

21 De Vries and Van der Woude, First Modern Economy; Lucassen, “Immigranten in Hol- land”; Lucassen and Lucassen, “The Mobility Transition Revisited.”

The Economic Contribution of Labor Migrants in the European Maritime Labor Market of the Long Eighteenth Century

Jelle van Lottum

In their seminal paper “The mobility transition revisited,” which appeared in the Journal of Global History in 2009, Jan and Leo Lucassen provided for the first time a solid quantitative framework that convincingly refuted the persistently prevalent theory of the mobility transition.1 This thesis, formulated by Walter Zelinsky in the 1970s, proposed that large scale population mobility was mainly a product of modernization and industrialization and that as such, the (early to mid) nineteenth century formed a watershed between the sedentary age of the early modern period and the much more dynamic and mobile population of the modern age.2 Although several studies on early modern or early industrial migration in Europe had argued along similar lines,3 by providing a well-structured quantitative overview of migration rates between 1500 and 1900, Lucassen and Lucassen made for the first time truly explicit that in Europe such a watershed did in fact not exist, and that changes in migration rates actually developed much more gradually over time. Nevertheless, their analysis showed that despite the fact that a clear break in migration rates in Europe cannot be discerned, in the underlying characteristics of Europe’s moving population significant shifts took place in the transition from the pre-modern to the industrial era. The most important change the quantification of European migration reveals

1 Jan Lucassen and Leo Lucassen, “The Mobility Transition Revisited, 1500–1900: What the Case of Europe can Offer to Global History,” Journal of Global History 4 (2009), pp. 347–77. For the underlying estimates and historiography, see Jan Lucassen and Leo Lucas- sen, “The mobility transition in Europe revisited, 1500–1900: Sources and methods,” IISH Research paper, no. 46 (2010). 2 W. Zelinsky, “The Hypothesis of the Mobility Transition,” Geographical Review, 61 (1971), pp. 219–49. 3 Jan Lucassen, Naar de kusten van de Noordzee. Trekarbeid in Europees perspektief, 1600–1900 (Gouda 1984); Leslie Page Moch, Moving Europeans: Migration in Western Europe since 1650 (Bloomington, IN, 1992); Nicholas Canny (ed.), Europeans on the Move: Studies on European Migration, 1500–1800 (Oxford, 1994); Dirk Hoerder, Cultures in Contact: World Migrations in the Second Millennium (Durham, NC, 2002); Klaus J. Bade, Migration in Euro- pean History (Oxford, 2003). 248 jelle van lottum can be found in the relative share of what the two authors typified the category of “soldiers and sailors”—the latter of the two we will focus on in this paper. They showed that throughout the early modern period this category did in fact add most to overall migration rates.4 With the advent of industrialization, this category became much less important; urbanization became the relatively most important category in the period thereafter. Although the number of migrant soldiers was in fact much larger than that of migrant sailors,5 the figures shown in the aforementioned article, which was later complemented by more detailed estimates published in a research paper,6 demonstrate that the early modern European maritime labor market was of considerable size and was indeed an international labor market par excellence, showing on average much higher participation migration levels than most other sectors of the European economies. In fact, the early modern maritime sector stood out not only because of its high migration rates; but because it was a key sector in the early modern European economy. Not only did transport play a crucial role in bringing about economic growth in the pre-industrial period,7 it was also a highly dynamic sector in which significant technological advancement through the ages had led to strong increases in labor productivity.8 The estimates of migration rates provided in the article in the Journal of Global History are indeed a useful framework in diachronically assessing the relative importance of migration in the European economy and society at large. However, as a framework, it is relatively limited in determining the extent to which migration in general,9 and the different migrant categories in particular, contributed to relative economic performance.10

4 Lucassen and Lucassen, “Mobility Transition,” p. 370, Figure 11. 5 Ibidem, p. 369, Figure 9. 6 Lucassen and Lucassen, “Revisiting.” 7 Richard W. Unger (ed.), Shipping and Economic Growth 1350–1850 (Leiden and Boston, 2001); D. Acemoglu, S. Johnson, and J.A. Robinson, “The Rise of Europe: Atlantic Trade, Institutional Change and Economic Growth,” American Economic Review, 95 (2005), pp. 546–79; Robert C. Allen, The British Industrial Revolution in Global Perspective (Cambridge, 2009); E.A. Wrigley, “Urban Growth and Agricultural Change: England and the Continent in the Early Modern Period,” Journal of Interdisciplinary History, 15 (1985), pp. 683–728. 8 Jan Lucassen and Richard W. Unger, “Shipping, Productivity and Economic Growth,” in Unger, Shipping, pp. 3–46; Richard W. Unger, Dutch Shipbuilding before 1800: Ships and Guilds (Amsterdam, 1978). 9 This is by definition the case when studying migration rates in isolation, because the data is not qualified as such. 10 This was, in fact, one of the goals of the quantification exercise, see Lucassen and Lucassen, “Mobility Transition,” p. 351. For a more elaborate response to the applicability of their approach as a tool in understanding migration’s role in relative economic devel- the economic contribution of labor migrants 249

Because the contribution of particular migrant groups can indeed differ strongly, the size of these groups may be important, but is not necessarily correlated to its economic contribution. What this essay aspires is to use the quantitative framework of Jan and Leo Lucassen’s article as a starting point, but to then take an additional step by trying to qualify the actors in it—in this case, that of the maritime workers in the early modern period. In other words, this essay will try to establish to what extent the quality or skill of these migrants added to overall skill levels, and if changes over time can be discerned. The maritime labor market of the early modern period makes for a very useful “historical laboratory” for such an excercise. It was, as we have just learned, a sector characterized by high levels of labor migration, but also one which played a key role in the early modern European economy. Moreover, as we will see below, it is one of the best documented early modern labor markets. The theoretical framework from which we depart in this essay is that of so-called new growth theorists, which postulates that human capital formation is an important determinant of long-term economic growth, and can thus explain different economic trajectories.11 Similarly, as studies of contemporary labor mobility have pointed out, there is a potential positive effect of labor migration on economic performance, not only because labor is an essential production factor, but also because the influx of skilled workers into an economic sector will increase its human capital stock.12 Although it has been argued that human capital (or skills) of “common workers” did not play an important role in economic growth before the Industrial Revolution,13 recent research argues the contrary.14 It has been shown that in the European merchant marine of the eighteenth century the human capital level of ordinary sailors had a positive and significant effect on the fleet’s labor productivity. In particular, numeracy skills

opment, see Jelle van Lottum, “Some Considerations about Economic Development and Migration,” Journal of Global History, 6 (2011), pp. 339–44. For a general discussion of the article, see the discussion dossier in Journal of Global History, 6 (2011), pp. 299–344. 11 R. Lucas jr., Lectures on Economic Growth (Cambridge, 2002). 12 R.E. Lucas, International Migration and Economic Development. Lessons from Low- income Countries (Cheltenham, 2005). 13 Allen, Industrial Revolution; Joel Mokyr, The Gifts of Athena (Princeton, 2002); Joel Mokyr, The Enlightened Economy: An Economic History of Britain, 1700–1850 (Princeton, 2010). 14 Jelle van Lottum and Jan Luiten van Zanden, “Labour Productivity and Human Capital in the Maritime Sector of the North Atlantic, c. 1672–1815” (working paper), downloadable at: . 250 jelle van lottum of seamen, which can be calculated by using the method of “age heaping”,15 had a strong effect on labor productivity of the vessels they worked on. Therefore, during the eighteenth century not only did labor productivity make significant advances as a result of capital investments (or technological change), investments in human capital of maritime workers appeared to have mattered as well.16 The work of Jan Lucassen has shown that migrants were very aware that skills played an essential role in bringing about productivity increases, and that as such, having particular skills were acknowledged and rewarded. This was not only true for those migrants who operated individually, but especially for those migrants who (like sailors) operated in groups, such as grass mowers and brickmakers.17 This contribution takes a consecutive step by establishing the effect on the sector at large. Was it “simply” a matter for maritime employers or local communities to train one’s own people to bring about increases in skill levels, or did in fact migrants play a role in providing fleets with skilled workers? In other words, to what extent did employers increase overall skill levels aboard their ships endog- enously, by relying on their native labor supply, or externally, through attracting foreign laborers? If the latter is the case, that would of course mean that the answer to the general question just raised is that indeed migration contributed to economic performance. But, the opposite can be true as well of course; migration can also bring overall skill levels down. Is there any evidence that the latter was the case, and consequently that economies that relied on immigration were hampered by it? To answer these questions we will use a dataset derived from the so called interrogation sections of the Prize Paper Archive, which has been described extensively elsewhere.18 The dataset consists of all relevant

15 B.J. A’Hearn, J. Baten and D. Crayen, “Quantifying Quantitative Literacy: Age Heaping and the History of Human Capital,” Journal of Economic History, 69 (2009), pp. 783–808. 16 It is important to note that skill levels of maritime workers were relatively high compared to other occupational groups, see Jelle van Lottum and Bo Poulsen, “Estimating levels of numeracy and literacy in the maritime sector of the North Atlantic in the late eighteenth century,” Scandinavian Economic History Review, 59 (2011), pp. 56–81. 17 Lucassen, Naar de kusten; Piet Lourens and Jan Lucassen, Arbeitswanderung und berufliche Spezialiserung. Die lippischen Ziegler im 18. und 19. Jahrhundert (Osnabrück, 1999); for a more global comparison see Jan Lucassen, “Brickmakers in Western Europe (1700–1900) and Northern India (1800–2000): Some Comparisons,” in Jan Lucassen (ed.), Global Labour History. A State of the Art (Bern, 2006). 18 Jelle van Lottum, Jan Lucassen and Lex Heerma van Voss, “Sailors, National and International Labour Markets and National Identity, 1600–1800,” in Unger, Shipping, pp. 309–52. the economic contribution of labor migrants 251 variables of merchant ships and its crews, such as the country of origin of the ship, and the age, place of birth and place residence of its crew- members. As we will see below, this dataset not only makes it possible to estimate skill levels of maritime workers, it also allows for an analysis of international labor mobility in the sector. Furthermore, it consists of data samples for two periods, one at the turn of the seventeenth century (1672–1720), in the text this will be referred to as Period 1, and one a century later (1770–1813), or Period 2.19

The dynamics of the international labor market for sailors

Despite the fact that the early modern maritime labor market was one of the most international of all European labor markets, the extent to which maritime employers from different European nations relied on foreigners to man their ships could differ strongly. The literature on the Euro- pean maritime labor market suggests that in Southern Europe (including France) shipowners tended to rely largely on native workers, whereas the crews on the fleet belonging to the countries bordering the North Sea, those from the Dutch Republic in particular, recruited a more international crew.20 The variety in foreign labor participation across the Euro- pean merchant fleet reflected to a great extent on the different stance of (national) authorities towards the employment of foreigners. Indeed, regulations regarding hiring could vary from the very strict, for instance in the case of Spain and France,21 to virtually non-existing, like in the case of the Dutch Republic.22 Below we will see that the analysis here broadly

19 Period 1 deals with ships taken during the Third Anglo-Dutch War (1672–4) and the War of the Spanish Succession (1701–14); the second period covering the War of the Ameri- can Revolution (1776–83), the Fourth Anglo-Dutch War (1780–4) and the French Revolu- tionary and Napoleonic Wars (1793–1815). 20 Jan Lucassen, “The International Maritime Labour Market (Sixteenth-Nineteenth Centuries),” in Paul C. Royen, Jaap R. Bruijn and Jan Lucassen (eds), “Those Emblems of Hell”? European Sailors and the Maritime Labour Market, 1570–1870 (St. John’s, 1997), pp. 11–24; Van Lottum et al., “International Labour Markets.” 21 For Spain see Carla Rahn Phillips, “The Labor Market for. Sailors in Spain, 1570–1870,” in Van Royen et al., “Those Emblems”, pp. 329–48. For France: T.J.A. Le Goff, “The Labour Market for Sailors in France,” in Van Royen et al., “Those Emblems”, pp. 287–328. 22 Karel Davids, “Maritime Labour in The Netherlands, 1570–1870,” in Van Royen et al., “Those Emblems”, pp. 41–71; Jelle van Lottum, Across the North Sea. The Impact of the Dutch Republic on Labour Migration, c. 1550–1850 (Amsterdam, 2007), Chapters 2 and 4 in particular. For an overview of the formal regulations towards foreigners in the Dutch Republic, see Jan Lucassen and Rinus Penninx, Newcomers. Immigrants and Their Descendants in The Netherlands 1550–1995 (Amsterdam, 1995); Jan Lucassen and Leo Lucassen, Winnaars 252 jelle van lottum confirms the established view, but it also elucidates that the international market for sailors was much more dynamic over time than existing literature suggests, demonstrating not only considerable change in overall migrant participation in the sector, but notably with regard to its composition and characteristics. Before taking a closer look at comparative levels of foreign labor participation in Europe in the period under scrutiny, it is first important to define the term migrant or migrant worker more precisely. Pertaining to migrant workers in the maritime sector, two main groups of migrant workers may be discerned: sedentary and non-sedentary migrants.23 These two groups formed what has been labelled the dual layered immigrant labor market.24 The first layer consists of those migrants who not only participated in a foreign labor market, but settled there as well, indeed were sedentary. Many of these migrants founded a family in their new country of residence and often, though by no means in all cases, stayed there for the rest of their life.25 The group of non-sedentary migrants were made up of workers who participated in a foreign labor market, but did not take up residence there.26 Whereas the first group is relatively well-recorded, as they ended up in civic or church records such as marriage, baptism, burial and tax registers, the latter usually is not, making them much more difficult to trace. However, fortunately the dataset used for this contribution uniquely allows analyzing both groups.27 As we will discuss later, these

en verliezers. Een nuchtere balans van vijfhonderd jaar immigratie (Amsterdam, 2011); Jan Lucassen, “Holland, een open gewest. Immigratie en bevolkingsontwikkeling,” in Thimo de Nijs en Eelco Beukers (eds.), De Geschiedenis van Holland, Vol. II: 1572–1795 (Hilversum, 2002), pp. 181–215; Piet Lourens and Jan Lucassen‚” “Zünftlandschaften” in den Niederlan- den und im benachbarten Deutschland,” in Wilfried Reininghaus (ed.), Zunftlandschaften in Deutschland und den Niederlanden im Vergleich (Münster, 2000), pp. 11–43. 23 Van Lottum, Across the North Sea, pp. 139 ff; idem, “Labour Migration and Economic Performance: London and the Randstad, c. 1600–1800,” Economic History Review, 64, 2 (2011), pp. 531–70. 24 Van Lottum, Across the North Sea, p. 146. 25 Lucassen and Lucassen, Winnaars en verliezers, pp. 191–3. For sailors’ families in particular see Danielle van den Heuvel, “Bij uijtlandigheit van haar man”. Echtgenotes van VOC-zeelieden aangemonsterd voor de kamer Enkhuizen (1700–1750) (Amsterdam, 2005). 26 Apart from sailors this group also consisted of soldiers and domestic servants, the latter whom were mainly female, see Lucassen and Lucassen, Winnaars en verliezers, pp. 194–9. Seasonal migrants should be regarded as a separate category, mainly because of their particular cyclical migratory behavior. Nevertheless, the underlying mechanism (such as the pull and push factors), and the overall migration patterns (also part of the North Sea system), did not differ significantly from the latter group, see Lucassen, Naar de kusten. 27 The source provides information about the place of birth and residence, as well as the labor market in which an individual worked in. the economic contribution of labor migrants 253

two groups are thought to have different characteristics, making a distinction between the two imperative. In the next two sections we will take a closer look at the characteristics of these two migrant groups, particularly in relation to their skills and their economic contribution; let us for now, however, focus on their quantitative development over time. In Figure 1, we can see how the migrant composition aboard three merchant fleets developed between the two periods mentioned in the introduction. In the figure, the Scandinavian countries of Sweden, Nor- way and Denmark are clustered in one group with (northern) Germany as “Northern Europe”; the data for France, and the Netherlands was large enough to analyze as individual countries.28

Netherlands Northern Europe France 1 12% 15% 6% 11% 0.9 26% 22% 3% 14% 2% 0.8 9% 14% 0.7 16% 0.6 0.5 91% 87% 0.4 74% 76% 64% 0.3 58% 0.2 0.1 0 12 12 12 Period Native Sedentary migrants non-sedentary migrants Source: Prize Paper Dataset. Note: the share of migrants aboard Scandinavia and German ships is calculated by taking into account absolute numbers of foreigners on individual national fleets. This implies that migration between Scandinavian countries is also taken into account. Figure 1. The share of migrants per labor market

28 These two countries and one country-group broadly represent the different recruiting regimes in the European maritime labor market, see: Van Lottum et al., “Labour Markets,” pp. 337–8. 254 jelle van lottum

When we look at the overall shares of foreigners in Figure 1, we find that French merchant vessels did not have large numbers of foreigners amongst their crews, which confirms the findings of earlier studies. Although throughout the eighteenth century the share of foreigners aboard French ships increased slightly from 9 to 13 percent in Period 2, the figure shows that compared to the countries north of the border, this was in fact still quite low. The difference between Southern and Northern Europe becomes indeed immediately clear when we shift to the Dutch Republic and North- ern Europe, which both show migration levels more than twice that of France. What is remarkably striking about Period 1 is that the Northern European fleet exceeds the share of foreigners aboard Dutch ships—with a share of 36 percent foreigners aboard the former versus 26 percent on the latter. This was mainly the result of migration between Scandina- vian countries, which, in contrast to other studies, are here regarded as separate nations.29 In any case, it is clear that the tables were turned a near century later. While amongst the Scandinavian and German merchant vessels, overall foreign labor participation dropped by 12 percentage points to 24 percent—indeed about the same level the Dutch Republic had in Period 1; shares of foreigners in the Dutch Republic increased quite spectacularly by 16 percentage points to 42 percent. Again, this confirms earlier estimates of foreigners amongst Dutch merchant crews, which showed a strong increase in the eighteenth century.30 When we shift to the changes in the composition of the migrant workers aboard the three fleets, a point that has not been considered in earlier studies, interesting changes took place between the two periods, particularly in the Dutch labor market. Here we see that the significant rise of foreign labor participation was nearly the sole effect of rising levels of non-sedentary migrants. In contrast, in Northern Europe the decline of foreign participation in the merchant fleet was considerably more balanced between the two migrant groups. Again, among the French merchant fleet hardly any change took place, although it is interesting to note that that the increase of migrants was only due to the growth of non-sedentary workers. In the following sections, in which we focus more closely on the maritime labor market of the Dutch Republic,31 we will take a closer at the

29 Van Lottum et al., “Labour Markets,” pp. 337–8, in particular p. 338, note 69. 30 Van Lottum, Across the North Sea, pp. 136–7, Figures 4.3 and 4.2 and Appendix III. 31 For an overview of the functioning of the Dutch maritime labor market see amongst others Lucassen, “Maritime Labour Market”; Davids, “Maritime Labour”; Jaap R. Bruijn and the economic contribution of labor migrants 255 changes that took place in the migrant labor market and the potential consequences thereof for economic performance in this sector. However, for now it is important to point out that the data presented in Figure 1 provides for the first time quite robust evidence for the hypothesis advanced elsewhere about the way in which the two layers of the Dutch immigrant labor market had developed.32 This hypothesis proposed that the Dutch immigrant labor market underwent a process of transformation during the eighteenth century, changing from a largely sedentary immigrant labor market in the seventeenth century to a much more volatile one a century later. In this market, migrants increasingly participated without actually settling in the host nation. This development coincided with a strong increase in the number of seasonal migrants to the coastal provinces of the Dutch Republic, as shown by Jan Lucassen’s seminal account on seasonal migration in Europe.33 All in all, it is apparent that during the eighteenth century the Dutch immigrant labor market became much more volatile and this is indeed exactly what we can retrace in Figure 1. Around the turn of the seventeenth and eighteenth centuries, sedentary migrants were a majority amongst immigrant sailors working on Dutch ships; in other words, more than half of the foreign sailors serving the Dutch merchant marine were residents of the Dutch Republic. Although in the period thereafter their share slightly increased by 2 percent, the figure shows indeed that the overall rise in foreign participation was mainly the result of a surge in levels of non-sedentary migrants. The most mobile of the migrant workers more than doubled in the intermediate period and their share rose to a little more than a quarter of the entire labor force. In conclusion, the data presented in Figure 1 shows that the international labor market for sailors was highly dynamic and showed major changes over time as well as demonstrated strong regional differences. Whereas in Southern Europe, here represented by France, the recruitment of sailors was a largely national affair, in Scandinavia, Germany and the Dutch Republic, shipowners relied greatly on foreign workers. Interest- ingly, while the former underwent a process of “de-internationalization”,34

Jan Lucassen (eds), Op de schepen der Oost-Indische Compagnie. Vijf artikelen van J. de Hullu (Groningen, 1980); Van Lottum, Across the North Sea, esp. Chapter 4. 32 See Van Lottum, Across the North Sea, pp. 142–3; Jan de Vries, “How Did Pre-industrial Labour Markets Function?,” in George Grantham and Mary MacKinnon (eds), Labour Market Evolution. The Economic History of Market Integration, Wage Flexibility and the Employment Relation (London, 1994), pp. 39–63. 33 Lucassen, Naar de kusten. 34 Van Lottum, “Labour Migration,” pp. 554–6. 256 jelle van lottum the latter become more and more dependent on migrants in general and non-sedentary ones in particular. Before we can determine what exactly the changes signalled here meant for the overall economic contribution of migrants to the sector, which is after all the goal of this essay, we first need to get a better understanding of the character of the changes that occurred in the international maritime labor market. The question is whether the transformation we have seen in this section reflect mere changes in the migratory behavior, or if the character of the migrant population underwent modifications as well. To establish this, we now shift our focus to the labor market that underwent the most significant change, and was one of largest in Europe: the Dutch maritime labor market.

The Dutch maritime labor market in the long eighteenth century

In the previous section we saw that the during the eighteenth century the Dutch Republic witnessed a strong rise in the share of migrants aboard Dutch vessels, growing from 26 percent in Period 1 to 42 percent at the end of the century. This growth was significantly due to a growing demand for labor in the maritime sector, as well as a stagnating national supply. Indeed, during the eighteenth century the Dutch maritime sector witnessed a period of considerable expansion, not only in overall tonnage shipped, but also in terms of the number of men employed.35 The latter was particularly the case in the merchant marine. It is estimated that between 1694 and 1780, overall employment in the sector rose from a total workforce of 20,000 to 25,000, reaching levels that were even higher than those during the Dutch Golden Age.36 However, because the native population growth remained more or less flat during this period,37 and

35 Jelle van Lottum and Jan Lucassen, “Six Cross-sections of the Dutch Maritime Labour Market: A Preliminary Reconstruction and its Implications (1607–1850),” in Richard Gor- ski (ed.), Maritime Labour in the Northern Hemisphere c. 1600–1950 (Amsterdam, 2007), pp. 13–42. See also: Jan Luiten van Zanden and Milja van Tielhof, “Roots of Growth and Pro- ductivity Change in Dutch Shipping Industry, 1500–1800,” Explorations in Economic History, 46 (2009), pp. 389–403. For an overview of the size of the European maritime labor market see Lucassen and Lucassen, “Mobility Transition,” p. 365, Table 3. Their figures rely heavily on Lucassen, and Unger, “Labour Productivity” and Richard W. Unger, “The Tonnage of Europe’s Merchant Fleets 1300–1800,” American Neptune, 52 (1992), pp. 247–61. 36 In showing such growth, the maritime sector clearly was quite exceptional. When we look at the overall growth levels, studies have shown that from the mid-seventeenth century the economy of the Dutch Republic entered a period of persistent stagnation (combined with spells of decline) which lasted until the end of the century. 37 On the causes of the stagnating population growth see Jan de Vries, “The Popula- tion and Economy of the Pre-industrial Netherlands,” Journal of Interdisciplinary History, the economic contribution of labor migrants 257 there was little incentive for native workers to leave more secure sectors of the economy,38 the increased demand for workers could only be met by attracting more foreign workers, who, as we saw earlier, came in large numbers. Earlier we have referred to the fact that the increase in the number of immigrants, and in particular that of non-sedentary migrants, for the first time really substantiates the presumed shift in the dual layered immigrant labor market of the Dutch Republic. The main cause for this labor market transformation is thought to have been the emergence of new economic core regions in the traditional migration field of the Dutch Republic. This was first suggested by Jan Lucassen, who showed that in the eighteenth century, an expanding core like Copenhagen started to attract migrants from traditional recruitment areas of the Dutch Republic, such as the Duchies of Schleswig and Holstein.39 The emergence of these new cores led to a process wherein migrants who would otherwise have moved to the Netherlands (the dominant core region), now could choose between different employers, including ones closer to home. From the perspective of the Dutch labor market, the growth of (new) core regions such as Hamburg, Copenhagen and Stockholm acted as intervening opportunities in its traditional hinterland, and it is thought that as a result of this it suffered serious supply constraints.40 Although the causes and the main characteristics of the labor market transformation (as summarized above) are relatively well known, and indeed the migration data presented in the previous section confirms earlier hypotheses about its development, the effects of these changes (i.e. the extent of the supply constraints) are less well understood. Roughly, we can think of two key consequences for the Dutch economy. The first, which is outside the scope of this essay, is the hypothesis of the reduced immigrant consumption.41 This hypothesis argues that the increase of non-sedentary migrants had a dampening effect on wages in the sectors where these migrants worked, while at the same time the demand for

15 (1985), p. 678; De Vries’s analysis of the demographic changes relies on the work of A.M. van der Woude, “Demografische ontwikkeling van de Noordelijke Nederlanden 1500–1800,” in Algemene Geschiedenis der Nederlanden, vol. 8 (Bussum, 1980), pp. 139–56. See also Jan Luiten van Zanden, The Rise and Decline of Holland’s Economy: Merchant Capitalism and the Labour Market (Manchester, 1993), pp. 25–9. 38 De Vries, “Pre-industrial Labour Markets.” 39 Jan Lucassen, “The North Sea: A Crossroad for Migrants?,” in Juliette Roding and Lex Heerma van Voss (eds.), The North Sea and Culture (1550–1800 (Hilversum, 1996), p. 179. 40 Van Lottum, “Labour Migration,” passim. 41 Ibidem, p. 563. 258 jelle van lottum goods and services fell—after all, non sedentary migrants consumed less than their sedentary colleagues—thus affecting wages in these sectors indirectly. More relevant, however, is the second hypothesis, which pro- poses that because of the increased competition for workers, the Dutch Republic failed to attract skilled workers.42 The argument is that because wages in the “new” growing cores started to converge with those in the Dutch Republic, local labor markets were more capable of keeping their skilled workers. In other words, local or national labor markets “creamed off” especially skilled migrants, who were increasingly enticed by better opportunities close by, thus leaving workers with fewer skills to the progressively more “volatile” international labor market, which, as we saw earlier, the Dutch Republic increasingly had to rely on. The foregoing hypothesis thus presupposes that the labor supply to the Dutch labor market underwent significant changes during the eighteenth century. As just mentioned, the data presented here can for the first time test this hypothesis and shed more light on the mechanisms behind it. In the next and final section we will look closer at the aspect of declining skill levels, here we will briefly focus on the general facet of change in the migrant population by investigating alterations in migration patterns, or rather changes in the place of origin of migrants as well as in the age structure. If substantial changes therein exist, this in turn would suggest that the alterations in the composition of the migrants population, as shown in the previous section, were not only a reflection of changes in the migratory behavior, but indeed, as the second hypothesis predicts, was part of a more substantial transformation of the labor supply to the Dutch Republic. Like most economic migrants in the Dutch Republic, migrant workers in the Dutch maritime labor market were part of what Jan Lucassen coined the North Sea system.43 This system consisted of migrants from especially the coastal regions of the northeastern part of the North Sea area (largely excluding England),44 moving to the Dutch Republic in search of work,

42 Ibidem, pp. 560–3. 43 Jan Lucassen, Migrant Labour in Europe, 1600–1900. The Drift to the North Sea (London, 1987). See also: Van Lottum, Across the North Sea, Chapter 2, and idem, “Some Aspects of the North Sea Labour Market, c. 1550–1800,” in Hanno Brand, Poul Holm and Leos Muller (eds), The Dynamics of Economic Culture in the North Sea and Baltic Region (ca. 1250–1700) (Hilversum, 2007). 44 Lucassen, “Crossroads,” p. 180, Map 3; Jan Lucassen, “Mobilization of Labour in Early Modern Europe,” in Maarten Prak (ed.), Early Modern Capitalism. Economic and Social Change in Europe, 1400–1800 (London, 2001), pp. 161–74. Van Lottum, Across the North Sea, Chapter 3 and idem, “Labour Migration.” the economic contribution of labor migrants 259

who were lured by relatively high wages and ample job opportunities.45 When we look at the migrants’ country or region of origin in the dataset in Table 1 below, it becomes clear that indeed the Scandinavian countries and Germany were the main suppliers of maritime workers to the Dutch labor market in both periods. Still however, Table 1 elucidates that important shifts did take place. When we start by looking at the “all migrants” category it becomes evident that not only was Northern Europe indeed the main supplier of workers to the Dutch maritime labor market, but also that its role became more important over time. While in Period 1, still a quarter of the migrant workers came from other European regions, in particular from the South- ern Netherlands and France, by the end of the eighteenth century this was only less than 10 percent. The increase in the share of German and Scandinavian workers as a whole was fairly evenly balanced, but when we look at the shares of sedentary and non-sedentary workers we see greater differences. When comparing the changes in the groups of sedentary and non-sedentary migrants, we find that it are especially the Scandinavians who become more volatile, rising from an already high 42 percent of all non-sedentary migrants to no less than 58 percent.46 If we connect this

Table 1. Origins of workers in the Dutch maritime labor market

Period 1 Period 2

All Sedentary Non- All Sedentary Non- migrants sedentary migrants sedentary Scandinavia 43% 44% 42% 50% 38% 58%

Germany 29% 25% 33% 39% 49% 32%

O ther* 29% 31% 25% 11% 13% 9% Total 100% 100% 100% 100% 100% 100% * mainly from France and Southern Netherlands. Source: Database Prize Papers.

45 See also Lucassen and Lucassen, Winnaars en verliezers, Chapter 5. 46 For specific literature on Scandinavian migration to the Dutch Republic during the early modern period, see amongst others: Sølvi Sogner and Jelle van Lottum, “An Immi- grant Community? Norwegian Sailor Families in Amsterdam in the 17th century,” History of the Family, 12, 3 (2007), pp. 153–68; Sølvi Sogner, “Young in Europe. Norwegian Sailors and Servant-girls Seeking Employment in Amsterdam,” in Jean-Pierre Bardet, François Lebrun and Jacques Dupâquier (eds.), Mesurer et comprendre. Mélanges offerts à Jacques Dupâquier (Paris, 1993), pp. 515–32. 260 jelle van lottum to the figure shown in the previous section, it logically follows that the increase in non-sedentary migration was mainly the effect of more Scan- dinavians joining the Dutch merchant marine.47 German seamen, on the other hand, become relatively more sedentary; in Period 2 about half of all sedentary migrants in the Dutch Republic was of German origin.48 A second aspect that could give us an idea of the extent to which the characteristics of the group of foreign workers in the Dutch maritime labor market changed over time is to look at its age structure. Dutch maritime historian Bruijn has argued that during the early modern period the average age of a European sailor was below the age of thirty.49 Our dataset suggests that those working in the Dutch maritime labor market had a slightly higher average age of around 33, but as we will see below the different categories show important changes over time. Such changes in age structure could of course be an indicator of a variety of possible transformations, varying from changes in the role migration played in the life-cycle of a migrant population, to an indicator of alteration in marital behavior and/or family strategies. However, age can also be seen as an indicator of experience. Of course, there is not a direct correlation between age and experience; all depends on the starting age and the intensity of the training. However, as Bruijn has concluded, because we know that during this period most maritime careers tended to start at a very young age—boys were often in their teens—shifts in the average age may tell us something about the level of experience of sailors at the moment of time they entered the international labor market.50 In Table 2 below, the average age of the three categories of workers is shown for the two periods.

47 This confirms findings based on different sources, Van Lottum, Across the North Sea, pp. 147–50. 48 Here the findings differ from those in Van Lottum, Across the North Sea, p. 150, which states that (Northern) Germany underwent a similar development as the Scandinavian countries (see previous footnote). 49 Jaap R. Bruijn, “Career Patterns,” in Van Royen et al., “Those Emblems”, pp. 27–8. 50 For the use of changing age structure as an indicator for changes in the migrant population see Jelle van Lottum, “Migration to the Netherlands in the First Half of the Nineteenth Century: An Assessment Using the Utrecht Censuses of 1829 and 1839,” Annales de démographie historique, 118 (2009), pp. 198–202. the economic contribution of labor migrants 261

Table 2. Average age of workers in the Dutch maritime labor market Period 1 Period 2 Difference (in years) Native workers 33.1 34.8 1.7 Non-sedentary migrants 34.6 31.1 –3.5 Sedentary migrants 33.4 36.4 +3 Source: Database Prize Papers.

When using the average age as a litmus test for transformations in the migrant population, and indeed a possible indicator for the level of experience, it becomes evident once more that during the eighteenth century significant changes took place. Once again, there are pronounced differences between sedentary and non-sedentary migrant workers. Whereas as the age of Dutch seamen stayed relatively stable during the eighteenth century,51 the average age of two migrant categories changed substantially. The data shows that in Period 1 all three groups of maritime workers have a more or less similar age structure, hovering around or slightly above the age of 33. In Period 2, however, a divergence between the two migrant categories took place. Whereas the group of non-sedentary workers transforms from being (albeit only slightly) the oldest group in period 1 to by far the youngest cohort in period 1, by the end of the eighteenth century sedentary migrants have become the group with the highest average age. Indeed, in Period 2 the age difference between the two is more than 5 years. If we follow Lucassen and Lucassen who argued that an average career of a maritime worker lasted around 12.5 years,52 and if indeed, as Bruijn suggested, most maritime careers started before the age of 20, this is a substantial difference. Moreover, the shifts in the average age as shown in Table 2, do indeed suggest substantial changes in the experience level. The extent to which these changes are genuinely matched by changing skill levels is something we will look further into in the next section. In sum, if we connect the observations of this section with the findings of the previous section, it can be (tentatively) concluded that the change from a mixed labor market where sedentary migrants were slightly in the majority, to one wherein non-sedentary migrants were dominant, was accompanied by substantial change in the migrant populations

51 Confirming the relative stability of the native labor supply, see Van Lottum, “Labour Migration,” p. 559. 52 Lucassen and Lucassen, “Revisited,” p. 63. 262 jelle van lottum themselves. In other words, the labor market transformation not only seems to reflect changing migratory behavior (i.e. that migrants simply became more mobile), but as the second hypothesis suggests, it was accompanied by changes in the migrant population. The non-sedentary migrant workers of Period 1 can therefore not be equated with those in Period 2, and it appears that the same holds true for the sedentary migrants. The proof of the pudding lies, however, in the extent to which these transformations did in fact lead to changes in the skill level of migrants, which is after all the core question of this essay.

The quality of migrants

It is generally understood that there are advantages to the external hiring of skilled workers. By attracting skilled workers, employers can relatively easily increase overall skill levels, and, as we have established in the first section, as a result can boost performance levels. Still, however, and this relates to the observations made in the previous sections, in the long run there are risks attached to overly relying on externally recruited skilled migrant workers. Firstly, non-sedentary migrants in particular have (by definition) a considerably less attachment to the labor market, resulting in a significantly higher turnover rate, which makes the process of hiring more complicated and (potentially) costly. Secondly, and more importantly, a too strong reliance on an external supply of skilled workers can make an economic sector much more susceptible to developments in the sending areas or labor markets elsewhere (the previously mentioned intervening opportunities). Theoretically, the latter could include a range of developments leading to a reduction of the labor supply and/or changes in its character. The causes for this could vary from the distorting effects of wars and other (national or international) conflicts, but can also comprise of changing economic opportunities in other places, thereby shifting the balance in a migrant’s decision-making process away from the traditional attracting core. When this occurs, and when this change is accompanied by a reduction in the supply of skilled workers, the receiving economic sector has to either shift its focus to local workers with lower skill levels, or has to recruit migrant workers with fewer skills.53 The result of both scenarios is, however, that overall skill levels in a sector will fall, which in turn is likely to hamper performance.

53 As we have seen earlier, the hypothesis is that the latter happened to the Dutch Republic during the eighteenth century. the economic contribution of labor migrants 263

This final section aims to determine if the above-mentioned observations can be applied to the early modern international maritime labor market. Did the receiving labor markets profit from attracting migrant workers, and if so, did they remain to do so over longer periods of time? Did the Dutch Republic indeed suffer from the increased competition in the market for sailors by having to attract workers with fewer or lower skills? To provide answers to these questions, of course we need to know the skill levels of the three categories of maritime workers: native workers, sedentary migrants and non-sedentary migrants for the two periods. In the introduction, it was mentioned that in particular numeracy skills had a significant effect on labor productivity; so to determine the relative quality of migrants in the sector, the obvious indicator is indeed the numeracy levels of crews, which in turn can be approximated by the level of age heaping in the group they belong to. In Table 3 below, numeracy levels of crews in Northwestern Europe (Scandinavia, Germany and the Dutch Republic) are shown for the different migratory categories. The numeracy skills shown in the table are expressed in percentages, reflecting the share of individuals within a group that correctly reports their age.54 Unfortunately, because a relatively large sample is needed to calculate numeracy levels, it is not possible to show these levels for individual countries. However, the data in Table 3 still provides a good picture of the development of skills for the different categories of workers in the maritime sector.

Table 3. Numeracy levels among crews in Scandinavia, Germany and the Dutch Republic 1 2 All workers 91% 94% Native workers 90% 97% All migrants 91% 88% Non-sedentary migrants 99% 88% Sedentary migrants 81% 89% Source: Database Prize Papers

54 This is the so-called alternative Whipple Index (or Ŵ), developed by A’Hearn et al., “Age Heaping.” 264 jelle van lottum

Let us start by looking at the overall development of numeracy levels. As became clear in the first section, overall levels of numeracy of crews in Europe increased during the eighteenth century—a development mirror- ing advances made in shipping technology. This is, not surprisingly, also what we see in the “all workers” category in the top row of Table 3. The increase is, however, only marginal with a mere 3 percent increase between Periods 1 and 2. However, more diverging patterns in the development of skill levels become visible when we take a closer look at the differences in skill levels of migrant and native workers in the two periods. Firstly, Table 3 shows that in Period 1, migrants and native workers have more or less similar numeracy levels—90 and 91 percent respectively. However, when we shift to the two migrant categories, it becomes evident that there were marked differences between the two groups. While at the turn of the seventeenth century sedentary migrants appear to have much lower skill levels than their more mobile colleagues; non-sedentary migrants have by far the highest skill levels of workers in the maritime labor market, clearly out-performing all other groups. Based on the data for Period 1, it seems that the answer to the more general question raised in the introduction as to whether migrants in general contributed to overall skill levels of the hosting labor market, and thus to economic performance of the sector, cannot be unequivocal. Although on a whole, foreign-born workers added marginally to the skill level of the native labor pool, it becomes clear that when we distinguish between the two categories of migrants (and this is in itself another proof of the necessity of this distinction), the most flexible group of non-sedentary migrants in fact significantly out- performed all other groups, including local workers. As such, the “brain- gain” they produced compensated for the lesser performance of their sedentary colleagues. By attracting non-sedentary workers in Period 1, maritime employers could therefore potentially gain a great deal which the Northern European fleet in Period 1 particularly took advantage of.55 As the previous paragraphs clearly showed, the eighteenth century brought about significant change in the composition of the migrant workers, and this is no different when we look at the skill level of maritime workers. Firstly, Table 3 shows that skill levels of native workers increased more than the average level, suggesting that employers or local

55 This illustrates the importance of making the distinction between the two migrant categories. the economic contribution of labor migrants 265 communities not only invested more in training, but also that they were able to keep their own workers. Secondly, when comparing the contribution of migrants and native workers to overall skill levels, it is evident that by the end of the eighteenth century the overall contribution of migrants had decreased significantly. In fact, Table 3 shows that as a result of in- migration, overall skill levels in the sector fell strongly. Interestingly, whereas overall levels of sedentary migrants rose at a similar pace as the native workers, which of course implies that the former still had lower levels than the latter, the skill levels of non-sedentary workers witnessed the largest change: they dropped by more than 10 percent. This means that non-sedentary workers changed from being the most skilled group in Period 1, to the least skilled in the late eighteenth and early nineteenth centuries. Finally, if we return to the differences in average ages as shown in Table 2, it becomes clear that the average age of a group indeed cannot be equated to a specific experience level—after all in period 1 the three categories had roughly a similar average age but diverging skill levels. Nev- ertheless, if we look at the group of non-sedentary migrant workers, the fact that both age as well as skill levels fall after period 1 may indeed be an indication that this group did transform from a highly skilled and specialised group of experienced sailors into a much younger, less experienced group of migrants with corresponding lower skill levels.

Concluding remarks

The strand of migration history interested in the link between pre-industrial economic development and population movement is still only in its infancy. By providing a solid quantitative framework Jan and Leo Lucas- sen presented an essential first step in making possible the marriage between (quantitative) economic history and (early modern) migration history. This essay aimed to take the next step by focusing on the quality of migrants (skills) instead of the quantity of them (i.e. migration rates). It is obvious that more in-depth research is necessary to fully understand the effects of population movements on pre-modern economic performance. Given the nature of the data used for this essay, it is inescapable that here key concepts such as skill and economic performance were approximated in a rather crude and imprecise manner. The extending and improving of datasets, which undoubtedly will happen in the future, will make the application of more enhanced tools possible, and will thus considerably 266 jelle van lottum refine the outcome. Nevertheless, some interesting and relevant insights could be distilled from the dataset, which open new avenues for future research. Let us, however, first return to the core question raised in the introduction: if we indeed accept the crude proxies and categorizations which the data forces us to use, did the skills of migrants contribute to overall skill levels in the receiving maritime labor markets (and thus to better economic performance)? As was already pointed out in the previous section, there is no clear cut answer to this question. What did become clear is that the period in which migrants contributed positively to overall skill levels was limited to the late seventeenth, early eighteenth century (Period 1), and was restricted to one migrant category in particular: the non-sedentary migrants. By the end of the eighteenth century (in Period 2) skill levels of all migrant categories (in particular the non-sedentary migrants!) were much lower than that of native workers, and those countries that had to rely on migrant workers saw overall skill levels decline. The analysis made furthermore evident, supporting indirect evidence from other studies, that the Dutch maritime sector suffered most from this development. Here, a too strong reliance on foreigners was clearly felt. Throughout the eighteenth century, because of the expansion of the sector and because natural population growth had halted, a stronger demand for foreign workers grew in the Dutch maritime sector. Neverthe- less, in stark contrast indeed to the earlier period, at the end of the eighteenth century immigrants lowered overall skill levels than anything else. The fact that, because of exogenous developments, the Dutch maritime sector had to particularly attract non-sedentary migrants, made matters only worse. This type of migrant had become considerably more important in the international maritime labor market, but counter to the beginning of the eighteenth century, when this group had by far the best skills, by the end of the century it no longer consisted of skilled (and one may suspect experienced) workers, but of young migrants, with relatively low skill levels. To conclude, as mentioned above, this essay is only a modest next step in fully understanding the effect of skills of migrants for economic development. Nevertheless, it does provide us with some useful insights which may guide future research in this important strand of migration history. Firstly, this essay showed that defining and categorizing migration is crucial. Although this in itself is not a spectacularly new the economic contribution of labor migrants 267 insight,56 what the analysis of the present study points out in particular is that the categorization needs to be done on a sub-sector level. Indeed, it became clear that the (still relatively rough) distinction between sedentary and non-sedentary migrants within the maritime sector was an essential one. These two groups had quite different characteristics, and as a further complication, these characteristics changed strongly over time. It is therefore clear that focusing on specific sectors (thus lumping together different categories of migrants) or picking out one particular sub-group, may give a skewed picture of the role of migration in the promotion of economic performance. Secondly, the changes in skill levels of migrants, or rather their falling levels, illustrate that the impact of migrants can differ strongly over time. The case of the Dutch maritime labor market showed that at the beginning of the eighteenth century, migrants (in particular non-sedentary ones) were an important driver of labor productivity, while in the latter period their role became more much more modest as a (still crucial) factor of production.

56 Jan Lucassen and Leo Lucassen, “Migration, Migration History, History. Old Para- digms and New Perspectives,” in Jan Lucassen and Leo Lucassen (eds.), Migration, Migra- tion History, History: Old Paradigms and New Perspectives (Bern, 1997).

Income Differentials, Institutions, and Religion: Working in the Rhineland or Pennsylvania in the Eighteenth Century1

Richard W. Unger

Rotterdam in the eighteenth century was a place that attracted workers. Jan Lucassen mentioned the labor “pull” the city had as early as his doctoral dissertation.2 It was one of the Holland towns that drew migrant workers from the eastern and southern Netherlands and from the Rhineland. In his first lengthy foray into the history of migration he identified sources of people and patterns of movement, though it would be some time before he wrote about the role of towns like Rotterdam as ports of departure for migrants. The people who passed through Holland harbor towns on their way to the Far East or the New World he called transmigrants. Even if there was a broad range of people among the transmigrants most were laborers in search of higher wages, better working conditions and more opportunities. As many as 500,000 individuals made their way through Dutch ports to the East Indies and stayed. That was a much larger number than the some 15,000 who went to South America and the Caribbean and the 10,000 who went to North America. The Dutch colony of New Neth- erlands in 1664 had a population still short of 10,000. The transatlantic numbers compared to the total of migration were small. More than half of those who left the Netherlands for North America before 1800 were not born in the United Provinces. They were transmigrants taking advantage of opportunities created by the presence of Dutch shipping, trading connections and port facilities.3

1 Barbara Lewis is responsible for compiling data, identifying ships and estimating passenger numbers and I am grateful for her completing the work under difficult circumstances. Eric Leinberger of the UBC Geography Department helped with the creation of graphic material. The University of British Columbia gave financial support to the early exploration of the topic and the Peter Wall Institute for Advanced Studies, UBC, provided facilities for completion of the work. 2 Jan Lucassen, Naar de kusten van de Noordzee: trekarbeid in Europees perspektief, 1600–1900 (Gouda, 1984), pp. 106, 108. 3 Jan Lucassen, “Emigration to the Dutch Colonies and the USA,” in Robin Cohen (ed.), The Cambridge Survey of World Migration (Cambridge, 1995), pp. 21, 24–25. 270 richard w. unger

Rotterdam in the seventeenth and eighteenth centuries was a principal Dutch port of departure for transmigrants to America. Typically they came from the Rhineland and their goal often was Pennsylvania. Ease of access to Rotterdam along streams and waterways certainly helped in attracting the potential voyagers from up the Rhine. They did not come to settle in the Netherlands or to have any part in Dutch society. They were workers and often with families on their way to a new world.4 The Dutch may have been precocious in the years from 1550 to 1750 in their high rates of mobility, especially in moving from the countryside to the growing cities of the Low Countries, but their pattern over time became the norm for others including people from the upper Rhine Valley as far as Switzer- land.5 Mobility in the seventeenth and eighteenth centuries was endemic in southwestern Germany and parts of Switzerland, about 500,000 people leaving the region in the eighteenth century alone. Of those some three- fifths went to Prussia, to Habsburg lands in the Danube Valley and later in the 1700s to Russia. The remaining 200,000 went west across the Atlantic. Many of the villagers from the Palatinate, the source of the great majority of German immigrants, took ship for “the island”, the name they used for Pennsylvania.6 It is extremely difficult to identify the specific reasons for Rhinelanders going to America. The first step in the process of explaining why those people went is to establish how many went, when and how. A range of work based on Philadelphia-generated data already exists so it is possible, expanding on earlier work, to establish the pattern and character of German migration. The obvious explanation for migration, at least to any positivist economic historian, is the possibility of improved welfare. A comparison of income differentials between the Low Countries, the Rhineland and the New World indicates the direct benefit from moving across the Atlantic. Proxy figures for comparative standards of living show that there would have been a strong pull on people to leave Germany for Pennsylvania. Migration data, however, show that far from everyone went and that the flow was erratic. The inconsistency shows economic pull was

4 Lucassen. “Emigration to the Dutch Colonies,” p. 24. 5 Jan Lucassen and Leo Lucassen. “The Mobility Transition Revisited, 1500–1900: What the Case of Europe Can Offer to Global History,” Journal of Global History, 4 (2009), p. 371. 6 Bernard Bailyn, The Peopling of British North America: An Introduction (New York, 1986), pp. 16, 32–3; Marianne S. Wokeck, Trade in Strangers: The Beginnings of Mass Migra- tion to North America (University Park, PA, 1999), p. 8. For a map of the catchment area of migrants, see p. 3. income differentials, institutions, and religion 271 only a partial explanation. An examination of other factors, both push and pull, helps to explain what is apparently inconsistent behavior. Work- ing conditions, potential for economic and legal independence, economic advantages and disadvantages, religious freedom, and potential risks to life and limb must have weighed heavily in any migration decision but institutional factors played a critical role as well. The pattern of movement suggests that individuals were unable to make consistent rational choices. Under those conditions small economic, political, religious or even fashion changes could generate swings in the flow of migrants. The English Quaker William Penn inherited from his father the right to £16,000, a debt owed the elder Penn by Charles II. In exchange the king in 1681 gave the heir a large tract of land in North America, and named it Pennsylvania. Penn’s colony grew quickly in part because from its earliest days it had a reputation for being welcoming to immigrants. Results included a rapid and unprecedented increase in population.7 By 1700 Philadelphia, the provincial commercial and political center of the province, was second in size only to Boston in the American colonies. Population growth was most rapid in the first decade of the colony’s existence, marked again between 1730–60 thanks to substantial immigration, and then slow in last third of the century. By the first two decades of the eighteenth century the colony’s trade to the West Indies

Table 1. Estimated Population of Pennsylvania

Year Population

1690 8,800 1726 40,000 1755 150,000 1790 308,000

Sources: James T. Lemon, The best poor man’s country; a geographical study of early southeastern Pennsylvania (Baltimore, MD, 1972), pp. 23–24 and W. Alan Tully, William Penn’s legacy: politics and social structure in provincial Pennsylvania, 1726–1755 (Baltimore, MD, 1977), p. 66. Lemon’s figures for 1690 and 1790 are for the southeast, the most densely populated part of the colony while Tully’s figures for 1726 and 1755 are for the entire colony.

7 In 1682 23 ships came up the Delaware depositing some 2,000 setters in the new colony. In the following year 20 ships brought an equal number of new residents. Marilyn C. Baseler, “Asylum for Mankind”: America, 1607–1800 (Ithaca, NY, 1998), p. 212; Johan A. van Reijn, “Benjamin Furly. Engels koopman (en meer) te Rotterdam, 1636–1714,” Rotterdams Jaarboekje, 9 (1985), pp. 228–9. 272 richard w. unger and Britain was already flourishing.8 The growth in trade included carrying more immigrants. The influx of new people whose language was not English and with no legal ties or tradition of allegiance to the British crown created anxiety among the original settlers and their offspring. An unexpectedly large number of 363 German immigrants arriving on a single ship in 1717 generated a reaction, the governor of the day calling for a system of registration for aliens. After an aborted start ten years later an entry process became law. The new rule required every male above the age of 16 to sign an oath of allegiance. Two years later the provincial legislators added a declara- tion of fidelity to the proprietor and the signing of an oath of abjuration directed against the Pope and the House of Stuart which had been sup- planted in Britain in 1689 by the king-stadtholder William III and his wife Mary II. From 1727 any captain importing immigrants had to submit a report on arrival with the names of all those he brought into port. Place of origin and occupation for each passenger was only infrequently recorded despite a legislated obligation.9 One unexpected consequence of the legislation sparked by fear of ‘impotent persons’ or ‘sickly vessels’ landing in the province was to create a body of records ideal for genealogists and for historians analysing immigration patterns. Not only were ships’ and captains’ names included but also information about the individuals who chartered the ships and others

8 The tonnage of shipping clearing Philadelphia doubled between 1725 and 1739 and between 1726 and 1755 exports to England almost quadrupled while imports rose in value more than fivefold. W. Alan Tully, William Penn’s Legacy: Politics and Social Structure in Provincial Pennsylvania, 1726–1755 (Baltimore, MD, 1977), p. 66; James T. Lemon, The Best Poor Man’s Country. A Geographical Study of Early Southeastern Pennsylvania (Baltimore, MD, 1972), pp. 23–4; Gary B. Nash, Quakers and Politics: Pennsylvania, 1681–1726 (Princeton, NJ, 1968), pp. 49–50, 56–7, 320. 9 Frank Ried Diffenderffer, The German Immigration into Pennsylvania through the Port of Philadelphia, 1700 to 1775 (Lancaster, PA, 1900), pp. 36–8; Ralph Beaver Strassburger and William John Hinke, Pennsylvania German Pioneers: A Publication of the Original Lists of Arrivals in the Port of Philadelphia from 1727 to 1808 (Baltimore, MD, 1966), vol. I, pp. xvii– xxiii, xxvii–xxviii; Marianne S. Wokeck, “The Flow and the Composition of German Immi- gration to Philadelphia, 1727–1775,” The Pennsylvania Magazine of History and Biography, CV (1981), pp. 251–4. Israel Daniel Rupp, A Collection of Upwards of Thirty Thousand Names of German, Swiss, Dutch, French and Other Immigrants in Pennsylvania from 1727 to 1776, 2nd rev. and enl. edition with an added index (Baltimore, 1980 [1876]), p. 39, gives the text of the 1727 resolution. The desire to keep track of who was coming in was part of a more general fear about the immigrant threat to the integrity of English colonial culture, a fear which reappeared repeatedly among prominent Pennsylvania political figures in the late 1740’s and early 1750’s. Baseler, “Asylum for mankind,” pp. 72–3; Tully, William Penn’s Legacy, p. 49. income differentials, institutions, and religion 273 who acted for the shippers of ‘freights’ as the fare-paying immigrants were called. The lists are not complete, some years being lost, and the captains’ lists are often inconsistent with the lists of those who swore the relevant oaths. Still the ship lists form the only records kept over a long period of time for any ethnic group immigrating to America.10 I.D. Rupp was first to publish the lists in 1856.11 R.B. Strassburger sponsored a new publication which appeared in 1934.12 Rupp, according to Strassburger, had misunder- stood the lists and so had missed or misinterpreted some of the information. The 1934 publication included arrivals down to 1808 while Rupp has stopped at the beginning of the American Revolution, 1775. The declara- tions of abjuration prove to be the most complete and accurate source of information on how many people went to Philadelphia, from where and when on which ships.13 M.S. Wokeck produced an extremely careful and extensive study of the stream of migrants in her doctoral dissertation and then in a later book.14 She compiled data for all German immigrants to America up to 1775 and for entry into all ports, not just Philadelphia. To compensate for instances where the lists only had the names of males aged 16 and above she generated estimates for numbers of women and children.15 She also attempted to tie ship arrivals in Philadelphia with records of departures from the Netherlands and elsewhere in Europe but with limited success. One major result was a massive table of ships carrying German immigrants with departure and arrival dates along with numbers of passengers and a wide range of other information about the transportation of migrants.16 Strassburger’s 40 lists for the years 1785 to 1808 are much less complete but offer a supplement to Wokeck’s table and extend the migration data, confirming many of the trends which she identified at work before the

10 Diffenderffer, German Immigration into Pennsylvania, p. 39; Aaron Fogleman, “Migra- tions to the Thirteen British North American Colonies, 1700–1775: New Estimates,” Jour- nal of Interdisciplinary History, 22 (1992), p. 692; Marianne S. Wokeck, “A Tide of Alien Tongues. The Flow and Ebb of German Immigration to Pennsylvania 1683–1776,” Doctoral Dissertation, Temple University (1982), p. 204; Wokeck, Trade in Strangers, pp. 38–9. 11 Rupp, Collection of Upwards of Thirty Thousand Names. 12 Much of the work of preparing the volume apparently fell to the editor, W.J. Hinke. Strassburger and Hinke, Pennsylvania German Pioneers. 13 Rupp, Collection of Upwards of Thirty Thousand Names, pp. 40–1. Strassburger and Hinke, Pennsylvania German Pioneers, pp. xxiv, xxixi–xxxiii, xl–xli; Wokeck, “Flow and the Composition of German Immigration,” pp. 256–7. 14 Wokeck, “A Tide of Alien Tongues”; idem, Trade in Strangers. 15 Wokeck, “Flow and the Composition of German Immigration,” pp. 264–6. 16 Wokeck, Trade in Strangers, pp. 240–76. 274 richard w. unger

American Revolution.17 The original data are far from perfect. Figures for the number of immigrants and for the tonnage of ships, for example, vary from one list to another. Some ships are not fully documented. The sources record healthy entrants at Philadelphia so those who died en route and ships lost at sea are missing. Occupations are rarely recorded so it is not possible to know directly what kind of work the immigrants did or what skills they brought with them. Despite any shortcomings the source is still extremely valuable because it makes possible following a stream of migrants along a specific path.18 In the early days of the colony settlement was often by religious groups travelling together to Philadelphia. In 1683, Francis Daniel Pastorius laid out a new settlement just to the North and West of the city called the Germantownship. The name soon was shortened to Germantown. The first settlers were families from Cleves, though with roots in the Nether- lands. Starting with 13 families the settlement grew with new immigration which by 1702 totalled perhaps 200 families.19 Despite Penn’s wishes ethnic groups tended to settle close to each other, part of the reason for the growth of Germantown.20 The number of German immigrants was small until about 1707. After that it rose but the flow was spasmodic and irregular.21 Before registration began in earnest in 1727 about 5,000 Ger- man immigrants came through Philadelphia. At first they settled in South- eastern Pennsylvania but by 1720 they were moving to the banks of the Susquehanna River and within a decade new settlers were going to the

17 Strassburger and Hinke, Pennsylvania German Pioneers, p. xliii; Wokeck, “Flow and the Composition of German Immigration,” p. 278. 18 The British government from December, 1773, to March, 1776, did try to record all those leaving the country for the western hemisphere. Customs officers were to report a broad range of information about where people came from, where they were going and why. Record keeping was less than diligent and under reporting was probably in the range of 15–30 per cent. Still that small cross-section of almost 10,000 people departing over a short period does offer points of comparison with the more extensive data on German migration through the Netherlands and to Pennsylvania. Bernard Bailyn and Barbara DeWolfe, Voyagers to the West: A Passage in the Peopling of America on the Eve of the Revo- lution (New York, 1986), pp. 91–5. 19 William Isaac Hull, William Penn and the Dutch Quaker Migration to Pennsylva- nia (Baltimore, MD, 1970 [1935]), pp. 178–81, 189–90, 232–4, 395, 416; Rupp, Collection of Upwards of Thirty Thousand Names, pp. 1–2; Strassburger and Hinke, Pennsylvania German Pioneers, pp. xv–xvi; Stephanie Grauman Wolf, Urban Village: Population, Community, and Family Structure in Germantown, Pennsylvania, 1683–1800 (Princeton, NJ, 1976), pp. 12, 59. 20 Lemon, Best Poor Man’s Country, pp. 43–4; Wolf, Urban Village, pp. 128–9. 21 Diffenderffer, German Immigration into Pennsylvania, pp. 33–4, 42; Wokeck, “Flow and the Composition of German Immigration,” p. 259 n21. income differentials, institutions, and religion 275

Shenendoah Valley in Virginia. The most popular destination by far for Germany immigrants to North America was the port of Philadelphia.22 After 1727 there was a dramatic rise in German immigration. In the peak years between 1749 and 1754 about 37,000 “Palatines” disembarked at Philadelphia. Numbers of arrivals collapsed during the Seven Years’ War but picked up again after 1763 though at a lower level. Only about 12,000 people were enumerated in the 12 yrs before the American Revolution.23 In the late 1720s, an average of 439 German immigrants arrived per year, and that figure rose steadily. At the height of migration from 1749 to 1754 the annual average was about 5,800. From 1763 to 1775, the figure was a much reduced 899. The single biggest years for immigrants and ships were 1738, 1764 and 1773.

8000

7000

6000

5000

4000

3000

2000

1000

0 1683 1693 1703 1713 1723 1733 1743 1753 1763 1773 1783 1793 1803 Number of Persons Figure 1. Number of ‘German’ immigrants entering at Philadelphia

22 Ports in Virginia played only a minor role in welcoming German immigrants. Most came through Pennsylvania and those who settled in that colony were even called Penn- sylvania German into the twentieth century. Wokeck, “Flow and the Composition of Ger- man Immigration,” pp. 249–50; Klaus German Wust, The Virginia German (Charlottesville, VA, 1969), pp. 28–9, 103–4, 195. 23 Wokeck, “Flow and the Composition of German Immigration,” pp. 260–1; Marianne S. Wokeck, “Promoters and Passengers: The German Immigrant Trade, 1683–1775,” in Richard S. Dunn and Mary Maples Dunn (eds.), The World of William Penn (Philadelphia, 1986), p. 261. It is no surprise that between 1745 and 1758 at the peak of immigration Germantown went from 100 to 350 houses and the population of the township grew at an annual rate of 9.6 per cent. Wolf, Urban village, pp. 14, 43. 276 richard w. unger

400

350

300

250

200

150

100

0 1683 1693 1703 1713 1723 1733 1743 1753 1763 1773 1783 1793 1803 Figure 2. Average number of persons per ship

Total German immigration from 1727 to 1775 was around 70,500.24 In 1790 there were about 35,000 Germans in Virginia, some 5 per cent of the population, about 24,500 in Maryland and in Pennsylvania 141,000. As early as the 1750s Germans made up 40 per cent to 45 per cent of the population there. By the end of the century the share was down but still in 1790 the proportion of people with a German or Swiss background in the state as a whole was 33 per cent and about 40 per cent in southeastern Pennsyl- vania.25 In the first three-quarters of the eighteenth century 14 per cent of migrants to the British American colonies came from Germany. Their share was only exceeded by that of Africans who made up 48 per cent. If all immigrants from the British Isles are lumped together then their total was 33 per cent but Germans were more numerous than English, Scottish, northern Irish and southern Irish migrants taken separately. Out of a total of about 278,000 immigrants to British North America from 1700 to 1775

24 They were the largest single group since total immigration, for example, in the shorter period of 1726–1755 was only about 70,000. Tully, William Penn’s Legacy, p. 69. Wokeck, “Flow and the Composition of German Immigration,” p. 266; idem, “Promoters and Passengers,” p. 270; idem, Trade in Strangers, pp. 39–47. 25 Lemon, Best Poor Man’s Country, p. 14; Tully, William Penn’s Legacy, p. 54; Wust, Vir- ginia Germans, pp. 186–7. income differentials, institutions, and religion 277 there were about 84,500 Germans and 83 per cent of them disembarked at Philadelphia.26 The age and gender composition of German migrants changed over time. In the early days typically families came, as in the case of the first residents of Germantown, so there were no extremes in the demographic makeup.27 That was not true of other predominately German settlements, especially in the interior, even though there was always a considerable family component to migration. After the 1750s, the proportion of women and children went down. Through the 1730s and 1740s, men who were related tended to change from being of different generations to being of the same generation. From fathers and sons, they went to being brothers and cousins. Then, from the 1750s, there were fewer and fewer relatives among the male immigrants, and after 1763 German immigrants were more likely to be young single men. The numbers of older men, that is over 50, declined and by the 1750s were all but zero. Young men dominated the numbers on ships with large cargoes of immigrants and with high ratios of passengers per ton, that is on ships that appear to have specialized in carrying immigrants.28 For German migrants the principal port of embarkation was Rotterdam through the middle of the century just as Philadelphia was the principal destination. After 1763 there were more potential receiving ports, and Germans took advantage of possibilities in the Carolinas or New York.29 In Europe as the sources for immigrants expanded to more of Germany the ports of embarkation became more diverse. By the beginning of the nineteenth century Rotterdam and Dutch ports in general no longer dominated the German migrant trade.

26 Fogleman, “Migrations to the Thirteen British North American Colonies,” pp. 698–9; Wokeck, “Flow and the Composition of German Immigration,” p. 266. Wokeck, Trade in Strangers, p. 37, puts the total of German immigration to the British colonies before 1776 at 110,000. Her figure covers a slightly longer period and includes 2,400 immigrants who went to Nova Scotia, pp. 54–5. 27 Wolf, Urban village, p. 45. 28 Wokeck, “Flow and the Composition of German Immigration,” pp. 270–3; idem, Trade in Strangers, pp. 49, 51, 162. British figures for 1773 to 1776 show a similar pattern with the share of single males being 84 per cent for England but only 60 per cent for Scotland, the latter sending more families to America. About 40 per cent of the men were in their 20s but the share was smaller for Scotland where more fathers of families were among the emigrants. Almost 70 per cent of British emigrants travelled separately, not in the company of anyone or any group. Bailyn and DeWolfe, Voyagers to the West, pp. 130–1, 136. 29 Wokeck, “Flow and the Composition of German Immigration,” p. 275. 278 richard w. unger

Figure 3. Ports of embarkation (Number of ships)

The economic advantages of migration for Rhineland Germans were obvious and very sizeable. Extensive long run series of wages and prices are available for Philadelphia in the eighteenth century. Principally but not exclusively based on the work of Anne Bezanson, the data is now available in machine-readable form.30 The figures from Pennsylvania and similar if less complete figures from Boston and Maryland make possible comparisons of standards of living in the New World with parts of Europe. Rob- ert Allen and colleagues have recently calculated earning levels on both sides of the Atlantic and they concluded that income levels in colonies were higher than income levels in mother countries but were correlated with them because of the integrated nature of labor markets.31 Such calculations are filled with dangers and require a number of assumptions made with widely varying levels of confidence. Even with that caveat, the results suggest there was always a good reason to migrate. For data from

30 Peter Lindert and Salvador Puente, Pennsylvania spliced series, 1720–1896, Global Price and Income History Group, University of California Davis, http://gpih.ucdavis.edu/ Datafilelist.htm#NorthAmerica. 31 Robert C. Allen, Tommy E. Murphy, and Eric B. Schneider, “The Colonial Origins of Divergence in the Americas. A Labour Market Approach,” in press, Table 4. income differentials, institutions, and religion 279 the Rhineland, Robert Allen has kindly done a comparable calculation of income levels in eighteenth century Augsburg. Though on the edge of the catchment area for German immigrants to Pennsylvania, it has the advantage of long series of data similar to those for Philadelphia which makes comparison of income levels possible. Philadelphia incomes were high but not considerably higher than in London. The differential grew in the second half of the eighteenth century and so, not incidentally, did emigration from Britain to North America. For Amsterdamers and Londoners, North America promised improvement, but for an Augsburger in the first half of the eighteenth century, the potential gain in income was considerably more—about 100 per cent. The income differential between southern Germany and Pennsylvania became even more dramatic in the second half of the century as incomes fell in Augsburg but rose in Philadelphia. By then the ratio of incomes was better than 3:1. From around 1670 to around 1814 real incomes slowly declined in Germany while they rose in Philadelphia so there any potential for improvement in welfare from emigrating rose over time.32 With such wide income differentials it is surprizing that anyone would have chosen to stay in Europe but they did and in large numbers. What is more as the income differential between Augsburg and Philadelphia grew migration

Table 2. Subsistence Ratios: Laborers

50 Year Period 1700–1749 1750–1799

Augsburg 2.41 1.69 Philadelphia 5.06 5.34 London 4.16 3.51 Amsterdam 4.20 3.77 Antwerp 2.75 2.48 Southern English towns 2.79 2.52 M aryland 3.82 4.32

Sources: Robert C. Allen, Tommy E. Murphy, and Eric B. Schneider, “The Colonial Origins of Divergence in the Americas A Labour Market Approach”, in press, Table 4; Personal communication from Robert Allen.

32 Ulrich Pfister, Consumer Prices and Wages in Germany, 1500–1850, Center for Quan- titative Economics, Westfälische Wilhelms-Universität Münster, http://www1.wiwi.uni- muenster.de/cqe/forschung/publikationen/cqe-working-papers/CQE_WP_15_2010.pdf, pp. 17, 24–5. 280 richard w. unger decreased so income differential was not the defining reason for migration from Germany to Pennsylvania in the eighteenth century. Other factors must have been at work in the labor market influencing the decisions of people from the Palatinate. Economic barriers to migration to the New World kept people back or made them decide to migrate north or east in Europe rather than across the Atlantic. The cost of transportation to Philadelphia from Rotterdam varied from about five to eight English pounds. That might or might not include the cost of the voyage from the Upper Rhine to Rotterdam and the cost of accommodation in Holland while waiting for the departing ship to be ready to take on passengers. It would have taken the unskilled worker in Philadelphia from about 45 to about 75 days to earn enough to cover the direct charges for transportation, that is based on daily laborers’ wages in the 1780s. The sum was far from trivial. It would have appeared to be more of a hurdle to an Augsburg laborer who would have faced giving up all of his income for up to a year to pay for himself. For a family the burden would have been even greater but at least children travelled for half fare and infants for nothing.33 From the perspective of the upper Rhine Valley the cost of transportation was a serious barrier and one which rose over time as fares went up and wages did not.34 The fare itself was only a portion of what the passenger had to pay. The lump sum might or might not include the cost of the river trip, some or all of the costs of the layover in Rotterdam or charges for any cargo brought along, whether household items or goods the emigrant planned on selling on arrival to pay for some or all of the fare. It almost certainly did not include the cost of food on board. Unscrupulous captains could charge interest at exorbitant rates on the deferred payment of any extras and since bills were often in one currency but paid in another those same men could use exchange rates to their advantage. Immigrants were often ignorant of what was being done or in no position to challenge the actions of their carriers.35 Saying with

33 Wokeck, Trade in Strangers, pp. 86, 100, 155–6. For Philadelphia wages and conversion to grams of silver, see Peter Lindert and Salvador Puente, Pennsylvania spliced series. For exchange rates see Allen–Unger Global Commodity Prices Database, http://www.gcpdb .info/index.html,. 34 Other reports of fares from Pennsylvania in 1767 and from Britain in the 1770s suggest that figures of five to eight pounds were high and that prices may have been considerably less. Bailyn and DeWolfe, Voyagers to the West, pp. 166–7; Strassburger and Hinke, Penn- sylvania German pioneers, p. xxxvii. 35 Wokeck, Trade in Strangers, p. 102. income differentials, institutions, and religion 281 certainty how much of a barrier transport costs were to emigration is not possible. However, without doubt migrants had to weigh carefully their ability to pay for getting to Pennsylvania. There were three sources of funding for such a voyage. The migrant could have money of his or her own. This could take the form of cash, acquired through the sale of assets at home or of goods acquired in Europe and then sold in Philadelphia. Better off Europeans, those who typically did not plan on being unskilled laborers, paid in full and a sizeable portion if not the majority of migrants were in that category. The second option was to have relatives or friends in the New World pay the costs. That implies that there was string migration where contacts through kin groups, religious affiliation or geographical proximity led people to leave their homes for a new land and join people they knew or knew of. The third option was for the migrant to sell his or her labor services over a period of time, usually three to four years, in exchange for the payment of the accumulated debt. The term of such indentures suggests that the cost of getting to Pennsylvania was a year’s wages or somewhat more. The servant had to earn enough for room and board, to repay the debt and cover interest on money lent during the period of the contract. When a ship arrived at Philadelphia the captain or his agent would advertize those available for indenture. Potential investors in the future labor of the migrants would come on board the ship and examine who was being offered. Over the course of the century ships spent less time in port so captains or their agents had to sell their debts more quickly. The period of indenture was negotiable. It depended on the age, sex, skills and health of the immigrant, among other things. Once agreed the government certified the transactions producing a document attesting that the indentee was property of the lender for the specified period. At least the new immigrant did not have to find a job or face a period with no income while searching for work or getting established. Being tied legally to an employer of unknown character, though, could pose a significant risk.36 The numbers of indentured servants rose in the early years of the colony but varied over the course of the century, possibly rising again but from a higher base after 1750.37 Skilled craftsmen in their 20s or 30s obviously had a much

36 Strassburger and Hinke, Pennsylvania German Pioneers, p. xxxvii; Wokeck, Trade in Strangers, pp. 140–1, 150–3, 155–6. 37 Lemon, Best Poor Man’s Country, p. 10. Nash, Quakers and Politics, p. 322. 282 richard w. unger better chance of a favorable indenture contract and an increase in trained migrants might offer one explanation for a rise in such arrangements.38 Obviously there were advantages for all parties from the system of indentures but there were also downside risks. Neither party knew about the character or background of the other. It is not surprizing that relations in some cases deteriorated and the servant ran away. Leaving was often easy since there was a large area of lightly settled country to the west where the reach of the law was in doubt. Even language was not a problem for runaway ‘Palatines’ since there were by the mid eighteenth century many German speakers and German communities in central Pennsylvania, Maryland and Virginia. Getting away from a difficult employer, avoiding working off the debt incurred for getting to America and the possibility of entering a free labor market with potentially higher income must have enticed people to break their contracts. The fact that runaways comprised perhaps only some 3 per cent of all indentees is, then, a surprise. Those who fled are well documented since the owners advertized in newspapers, sometimes in German, to get their property back, offering rewards as was done for runaway slaves.39 The availability and quality of transportation could have deterred migration. The numbers of ships involved in the trade and the efficiency of shipping and connections between Rotterdam and Philadelphia, with some fluctuation, rose in the course of the eighteenth century. The number of ships carrying immigrants into the port of Philadelphia peaked in the boom around 1750. Even though the number of immigrants travelling each year fell after 1765 the number of ships carrying them did not fall as rapidly, a fact reflected in the falling average of immigrants per ship.40 It is not obvious that travellers enjoyed better conditions on vessels with fewer passengers.

38 From the end of 1772 to early in 1776 nearly half of British emigrants to North Amer- ica were indentured. The rate among English migrants was a full 68 per cent but for Scots it was considerably lower at just 18 per cent, in part because the latter were more likely to emigrate as part of a family though about one-third of the Scottish indentees went with other family members. By the 1760s in Britain the process of indenture had become for- malized with standard printed forms to be completed and signed by the parties involved. The same would have applied to German immigrants entering Philadelphia. Bailyn and DeWolfe, Voyagers to the West, pp. 175–7, 182, 313–5. 39 Bailyn and DeWolfe, Voyagers to the West, p. 350. Between pages 352 and 353 they offer a sketchbook of runaway servants with the text of the advertisements, two of the 20 being in German. Wokeck, Trade in Strangers, pp. 100–1. Wust. Virginia Germans, p. 106. 40 Compare figures 2 and 4. income differentials, institutions, and religion 283

0 1683 1693 1703 1713 1723 1733 1743 1753 1763 1773 1783 1793 1803 Figure 4. Number of ships entering Philadelphia with ‘German’ immigrants

There was some specialization in the boom years of the German migrant trade in mid century but in most cases vessels carried people in addition rather than in place of trade goods. At least 170 different ships made the 324 voyages bringing Germans to Philadelphia from 1727 to 1775 so some 154 trips were by ships which had carried passengers previously. A few ships specialized in the migrant trade especially during the mid century boom, making direct passages back and forth from the Netherlands without diversions to the West Indies or other destinations on the way back from Pennsylvania.41 Presumably there were advantages to travelling on a ship fitted out to handle passengers commanded by an experienced captain. Some captains specialized in the immigrant trade. Dealing with what could be a difficult cargo and the accounting problems coming from multiple charges required more than the ordinary navigational and leadership qualities of a captain. A total of 75 voyages or more than 20 per cent of all trips were under the command of just ten masters.42

41 From 1727 to 1775 four ships made five voyages, two made six, one made seven and five made eight voyages while in addition a number of ships made two or three trips carrying migrants. Twelve ships were responsible for almost 25 per cent of all trips. 42 One captain made 13 crossings with immigrants between 1747 and 1771, another 11 in the much shorter period from 1763 to 1773. One made eight crossings, three made seven and four made six with immigrants. Strassburger and Hinke, Pennsylvania German Pioneers. Of the 402 ships carrying people from the British Isles to America between the end of 1773 and early 1776 only 28 ships made two voyages and only one made three. The 284 richard w. unger

300

250

200

150

100

0 1727 1737 1747 1757 1767 Figure 5. Average tonnage of ships carrying immigrants

The vessels, even those specializing in the trade, tended to be small by standards of the day. In the eighteenth century the typical cargo vessel was a sailing packet of around 400–500 tons. Tonnage is recorded for only a few of the migrant ships entering Philadelphia but for those the overall average reported was only 170. The figure rose slightly over the years but only slightly. Records of 16 vessels show the smallest to have been just 21 metres long, the largest 33 metres. Still there were large ships in the trade with 22 vessels each delivering more than 400 German passengers.43 Only 37 vessels were described by type and of those 13 were called galleys though they were not and could not be like the contemporary rowed sailing barges of the Mediterranean. Ships of that design could not have survived the trip

incidence of multiple voyages would have been higher over a longer period of course. One half of them carried fewer than five passengers. The average number of passengers was 21.7, considerably lower than that for contemporary voyages carrying Germans to Phila- delphia. On the other hand, the few specialized ships carried large numbers. Just 22 of the 432 voyages were responsible for moving almost 40 per cent of the total of emigrants. In more or less the same period, that is from 1773 to 1776, half of the eighteen ships landing Germans in Philadelphia carried 82 per cent of migrants. The higher share for Rotterdam and similar ports of embarkation was presumably the result of the vessels leaving assembly points for migrants where large numbers of “freights” could be found. In Britain specialist ships tended to sail from London, northern English and Scottish ports where the largest share of migrants took ship. Bailyn and DeWolfe, Voyagers to the West, pp. 94–7, 102. 43 Diffenderffer, German Immigration into Pennsylvania, pp. 46–8; Wokeck, “Promoters and Passengers,” p. 269. income differentials, institutions, and religion 285 across the North Atlantic. Two were called ships but that rather general term may have been to distinguish them as three-masted vessels and so different from the remainder: snows, pinks, brigs and bilanders, which were all two-masted. The emigrant trade presumably benefited from other long term efficiency improvements in eighteenth century shipping.44 The carriage of Germans to Philadelphia was never a serious burden to the carrying capacity of the British merchant marine. Even at the height of the migration in 1751 probably only about 3200 tons of shipping made the voyage to Philadelphia, less than 1 per cent of the tonnage of the British merchant marine at the time.45 A second barrier for the potential migrant were the costs of getting established in the British colonies. For skilled tradesmen apparently such costs were close to zero because of the labor shortage which was typical of Pennsylvania through mid century. As Philadelphia grew to be a city requiring a broader range of services and as industries became established, such as weaving in Germantown, finding work became easier. Lit- eracy rates rose over the course of the eighteenth century among German immigrants which suggests both better schooling at home and more marketable abilities.46 Most immigrants from Germany, though, were farmers. They were interested in taking up land, a commodity in short supply in the Palatinate but easily available in Pennsylvania. What is more the soils of the southeastern part of the colony were very rich and the landscape even had similarities in appearance and climate to what the migrants had known at home. The cost of land to the settler might be as low as zero. The proprietor of the colony, Penn and his heirs, owned the land and people bought from the proprietor or designated agents or those who had already bought tracts of land. In the interior it often took some time for officials to catch up with individuals and charge them for land they had settled. Peo- ple in Pennsylvania tended to move and German immigrants were part of the pattern, often choosing to go out to the frontier of settlement. Over the eighteenth century land values rose and holdings on average became

44 Richard W. Unger, “Ship Design and Energy Use, 1350–1875,” in Richard W. Unger (ed.), Shipping and Economic Growth, 1350–1850 (Leiden, 2011), pp. 254–9. Ships carrying emigrants from Britain in the 1770s were also small, most of them being between 100 and 300 tons. Bailyn and DeWolfe, Voyagers to the West, p. 316. 45 Richard W. Unger, “The Tonnage of Europe’s Merchant Fleets 1300–1800,” The Ameri- can Neptune, 52 (1992), pp. 257, 260–1. 46 Wolf. Urban village, p. 105; Wokeck, Trade in Strangers, p. 50. After 1765 in the Vir- ginia back country for example more Germans were identified with crafts. Wust, Virginia Germans, p. 70. 286 richard w. unger smaller. Moving inland was a stop gap solution to the growing pressure on land since as population rose so did land values.47 More expensive land would have been a deterrent to migrants. There were additional startup costs for farmers including clearing the land and buying livestock. The foresightful immigrant would have weighed those financial burdens as well when thinking about moving to the New World. Working conditions in Pennsylvania were both an inducement and a barrier to migrants. Many worked in primary agriculture. Land was relatively plentiful but developing it required both capital and considerable labor including skills not always known to farmers from the densely settled Rhineland. The better off might be able to set up their own farms. By the mid eighteenth century skilled artisans could find work in established handicraft enterprises so for them working conditions were not much different from those at home. For unskilled laborers the tasks might be similar and the legal structures might be more to their advantage in Pennsylvania but there was potential for disaster in going to work for an unknown employer in unfamiliar surroundings. In indentures the laborer had no control over the work or working conditions. The uncertainty of labor status must have stopped some men and women from leaving the Rhineland. Though religion influenced decisions about migration the simple view that people left Europe seeking and finding religious freedom in the New World obscures the role of belief in the character and timing of decisions of whether or not to move. William Penn and his fellow Quakers had limited success in recruiting members of other religious groups to join in his experiment. Penn made extensive efforts to recruit settlers in England and on the Continent. He went to the Rhineland on an extended trip in 1677, one which started and ended in Rotterdam. He travelled in the company of a number of very prominent Dutch and English Quakers, visiting religious groups and individuals with practices similar to those of his own sect. His last visit to the Continent in 1686 brought even more migrants to Pennsylvania. He also wrote tracts about the colony which friends and agents in Rotterdam quickly translated into Dutch and German and circulated widely along with positive letters from early settlers. The appeals were directed at a wider audience but co-religionists or potential co-religionists were the most vulnerable targets.48 Religious groups tended

47 Lemon, Best Poor Man’s Country, pp. 42, 71; Tully, William Penn’s Legacy, pp. 55–7; Wokeck, “Flow and the Composition of German Immigration,” p. 277; Wolf, Urban Village, pp. 63–9. 48 Hull, William Penn and the Dutch Quaker Migration, pp. xi, 2–58, 64–106, 106–22, 123–77, 236, 301, 308–10, 318–22; van Reijn, “Benjamin Furly,” p. 226; Julius Friedrich Sachse, income differentials, institutions, and religion 287 to come as families, providing a demographic profile similar to that in European settlements which was an advantage in the new colony. The settlement of Germantown was not unique in that way. Migration for religious reasons, though, declined as a share of the total almost as soon as the colony became established.49 By the 1730s religious migrants were a small minority and almost irrelevant among those arriving in Philadelphia by the 1750s. Pennsylvania had a variety of religions present by mid century and an individual’s prosperity did not depend on religion or origin. The economic status of various national groups in the colony was more or less the same.50 Religion may have been important in the early days of the colony and important over the longer term in bringing more people through string migration but its role faded as economic considerations which included working conditions became paramount. Violence, or more generally the potential risk of surviving the voyage and settling in a new land compared to staying at home, was a factor in the pattern and flow of migrants. There was the chance of dying on the trip. Conditions of travel were harsh. Boatmen ferrying people down the Rhine were not known for their consideration for passengers. On board ships crossing the Atlantic space was limited and disease could and did spread rapidly in cramped quarters. Ships sank disappearing completely and presumed lost with no one knowing when or how or why people on board died.51 The profitability of a voyage depended on the number of ‘freights’, captains being paid a fixed sum for each one so they had an obvious incentive to pack as many people on board as was feasible. They also had incentives to limit food consumption since carrying provisions took up space that could be devoted to fare-paying passengers. On the other hand the men who had extended credit to passengers wanted them to arrive in good health. Sickly immigrants were harder to sell as indentured

Benjamin Furly and Symon Jansz Vettekeücken, “Benjamin Furly,” The Pennsylvania Maga- zine of History and Biography, 19 (October 1895), p. 287; Wokeck, “Promoters and Passen- gers,” p. 259. 49 The electors Palatine from 1690 to 1742 were Roman Catholics and though they could not change the predominately Protestant nature of their lands their beliefs implied a threat to longstanding Protestant domination. Diffenderffer, The German Immigration into Pennsylvania, pp. 10–1, 16–7; Strassburger and Hinke, Pennsylvania German Pioneers, pp. xv–xvi; Wolf, Urban village, p. 74. 50 Lemon, Best Poor Man’s Country, pp. 15–7. 51 Gemeente Archief Rotterdam, Notarieel Archief, 3271/332, 3281/267, 3283/238, for example. The Friends Generous of 120 tons, chartered on 26 June, 1773, left Rotterdam for Cowes and Philadelphia but never arrived in the New World. As late as 27 April 1779, the Rotterdam agents were still producing notarial acts acknowledging the loss and presumed deaths of certain passengers. 288 richard w. unger servants. Voyages across the ocean lasted as little as one month, good for all those involved, but the average was more like two months and some took as much as four.52 Extended travel time threatened the health of passengers and the profits of captains and merchants. Mortality rates among German immigrants may well have been considerably lower than the norm for all immigrants but they were still at least two to three times those at home in Germany. Governments in the colonies passed legislation around mid century to end the worst abuses of crowding and cheating immigrants of goods they had brought with them on board. Pennsylvania also from 1765 insisted that vessels with 50 passengers or more carry a surgeon. Efforts to reduce mortality through legislation had little effect on loss of life from disease once on land. Faced with new microbes and in a weakened state, as in the East Indies, a number of new arrivals died soon after embarkation.53 The violence of war or the potential for violence acted as a deterrent to migration. During periods of fighting among states the numbers of Ger- mans going to Philadelphia usually fell to zero as shipping was all but suspended. The years of the Seven Years’ War and the War of the American Revolution saw virtually no new immigrants. The Peace of Amiens of 1802 which proved a year-long truce in the wars between revolutionary France and Britain coincided with a jump in migrants. There was an even larger increase in 1804 and 1805, years when Prussia and France were at peace. War broke out between the two in 1806 and migration fell dramatically. It was not just war in Europe that might have influenced decisions. In the Seven Years’ War German settlers on the frontier in the Blue Moun- tains and along the Susquehanna River were killed by Indian allies of the French. Rupp writing a only century after what he called massacres of an inflated estimate of 3,000 Germans attributed the deaths to ‘the relent- lessly cruel savage’.54 How well known those events were in the Rhine Valley is not clear but the general danger from attacks on isolated settlements must have been common knowledge. The devastating wars of Louis XIV down to 1713 were often fought in the Rhineland and coincided with rising taxation by German states to pay for those wars as well as the ones of the Holy Roman Empire against the

52 Bailyn and DeWolfe, Voyagers to the West, p. 330; Diffenderffer, The German Immi- gration into Pennsylvania, pp. 29–30, 55–6; Wokeck, Trade in Strangers, pp. 78–9, 90. 53 Baseler. “Asylum for Mankind,” pp. 93–8, 109–10. 54 Lemon, Best Poor Man’s Country, p. 143; Rupp, Collection of Upwards of Thirty Thou- sand Names, p. 17. income differentials, institutions, and religion 289

Ottoman Turks. The burden fell on lands which had not yet recovered from the disasters of the Thirty Years War.55 The peace in 1748 ending the War of the Austrian Succession may help to explain the sudden rise in migrant numbers in 1749. Perhaps as or more influential were recent changes in the laws on conscription in the Palatinate. Young men engaged or newlywed were no longer exempted from military service. Only the well off could buy their way out of the army so many apparently took the new threat of being forced into a violent profession as a reason to leave and go to America. Certainly the changes help to explain the spike of young men landing in Philadelphia from 1749 to 1754.56 There were institutional barriers to migration. Jan Lucassen himself has emphasized the importance of organizational factors in shipping in general in the period and it is not surprizing that they would have had an influence in the specific case of German migration to America.57 Informa- tion or more precisely the lack of it was a barrier to emigration. People in the Rhineland would not have known much if anything about Penn- sylvania, despite the pamphlets of Penn and the letters from emigrants published in German. Efforts to disseminate information had by definition to go through literate people and that usually meant religious leaders. Often they had no incentive to promote migration and saw people leaving their congregations to be a bad thing. Other sources of information and encouragement were necessary. Making the transfer from one continent to another was a complex process. It was virtually impossible for an individual to navigate all the different bureaucratic and financial barriers on his or her own. The agents who emerged to act for immigrants and pave the way served over time as promoters of migration. Those men became part of an institutional structure which lowered or eliminated barriers in what became a transatlantic labor market. Their charges, however, created another financial barrier. Merchants in Rotterdam arranged the four to six week trip down the Rhine. They sent out their own agents to recruit potential migrants in the Rhine Valley and arranged for the layover in Rotterdam which was

55 Christopher R. Friedrichs, Urban Society in an Age of War: Nördlingen, 1580–1720 (Princeton, NJ, 1979), pp. 293–4; Pfister, Consumer Prices and Wages in Germany, pp. 17, 23; Wokeck, Trade in Strangers, pp. 6–7. 56 Wokeck, “Flow and the Composition of German Immigration,” pp. 267, 270–1. 57 Jan Lucassen, “Work on the Docks: Sailors’ Labour Productivity and the Organization of Loading and Unloading,” in Richard W. Unger (ed.), Shipping and Economic Growth, 1350–1850 (Leiden, 2011), pp. 269–78. 290 richard w. unger often as long as the Rhine journey. The city had a population of only some 45,000 in the first half of the eighteenth century so 1,000 or 2,000 or more German transmigrants passing through the city in the spring and summer, normal for the 1740s and early 1750s, presented a significant burden on resources and drove up the cost of any time spent there. Merchants in Rotterdam saw to the chartering, outfitting and provisioning of ships for the transatlantic voyage. British Navigation Laws allowed only British- owned and British-operated ships to travel between the mother country and the colonies and also required all ships going to the colonies to come from British ports. Rotterdam merchants then had to charter English ships for the trade from London merchants and also arrange for a stop in Eng- land which could last for one or two weeks to clear customs and to get a favorable wind. The correspondents in England looked after formalities as well as ensured that the vessel picked up any necessary supplies and, if possible, more cargo.58 Rotterdam businessmen also contracted with the captains and set payments to them. Merchants increasingly advanced money to migrants to pay their passage money. As agents, promoters, organizers and financial backers of the trade the merchants were critical to its functioning smoothly or even functioning at all. Until there was a group of men able to take on those tasks Rhinelanders found it difficult even to consider migrating to America. There was a colony of English traders in Rotterdam in the seventeenth century. Between 1698 and 1729 about one in eight of new burghers registered in the city was English. There were Scots as well with their own church as early as 1643 when they numbered between 400 and 500. Both groups proved important to the economy of the city.59 Among the Eng- lish was a number of Quakers and most notably one Benjamin Furly. He was a friend of William Penn among other prominent English Quakers, a man of considerable learning and linguistic skill and a strong supporter of the colonial effort in the New World. The common religious ties with

58 P.W. Klein, “’Little London’: British Merchants in Rotterdam during the Seventeenth and Eighteenth Centuries,” in D.C. Coleman, Peter Mathias, and Charles Wilson (eds), Enterprise and history: Essays in Honour of Charles Wilson (Cambridge, 1984), pp. 117–8; Strassburger and Hinke, Pennsylvania German Pioneers, pp. xxxiii–xxxiv; Wokeck, “Pro- moters and Passengers,” pp. 261–2; Wokeck, Trade in Strangers, p. 82. 59 Douglas Catterall, Community without Borders: Scot Migrants and the Changing Face of Power in the Dutch Republic, c. 1600–1700. Studies in Medieval and Reformation Thought, v. 86 (Boston, 2002), pp. 346–50; idem, “ ‘Secondo il resoconto di sua madre che ancora abita a Oostenhuysen’: migranti e politiche della migrazione nella societá urbana nordeuropea,” Quaderni storici, 36 (2001), p. 36; Klein, “ ‘Little London’,” pp. 121, 126, 132. income differentials, institutions, and religion 291 merchants in England and among fellow Quakers in the Netherlands made him a central figure in establishing institutions necessary to promote migration. Furly became Penn’s agent in the Netherlands when Pennsylvania was set up, joined Penn on the visit to Germany in 1677 to recruit settlers, invested in Pennsylvania property and set to work in the very early days of the colony, organizing passage for the first settlers of Germantown.60 Religious ties were important to trading connections in the years around 1700 and there was a number of prominent and successful Quaker trading families emerging in London open to contact from and with Furly. English Quakers, because they were excluded from a number of paths, may have found themselves more likely to engage in dangerous or risky trades like those to America which suited Furly and fellow Rotterdam merchants in finding correspondents. The British in Rotterdam dominated trade with the British Isles.61 They had extensive connections across the North Sea and knew the ins and outs of British trade regulations. They also had the advantage of knowing about Dutch regulations and even influenced the rules which applied to migrants in the United Provinces. The Dutch government, concerned about the large numbers of transients, promulgated laws in 1735, refined in 1739 and reissued largely unchanged in 1764, which required emigrants to present valid passports to enter the country and show proof they could pay for their transatlantic passage before they could board a ship at Rot- terdam. The merchants were made responsible for the emigrants leaving the country and for their not causing any problems while in the Repub- lic, paying for a military escort for the migrants and covering any associated fees. To guarantee that all the costs imposed by the legislation were paid merchants had to post bonds with the government. Their knowledge and connections allowed some Rotterdam merchants to specialize in the transmigration trade. From 1730 to 1775, just 12 handled arrangements for 145 ships, which carried close to 90 per cent of migrating Germans for

60 Furly made his home on the Scheepmakershaven into a center for discussion of political, religious and philosophical questions, attracting resident as well as exiled Eng- lishmen, among them his long time friend John Locke and Pierre Bayle. When Furly died in 1714 his library of 4,000 titles in 25 different languages was sold at auction. The fullest treatment of Furly and his career is William Isaac Hull and Thomas Kite Brown, Benja- min Furly and Quakerism in Rotterdam (Swarthmore, PA, 1941). On Furly’s life and career, pp. 3–170. Hull, William Penn and the Dutch Quaker Migration, pp. 253, 328–9; van Reijn, “Benjamin Furly,” pp. 219–21, 224, 229–33, 238; Sachse, Furly and Vettekeücken, “Benjamin Furly,” pp. 277–87, 295. 61 Klein. “ ‘Little London’,” pp. 128–9. 292 richard w. unger whom agents are known. Some merchants went so far as to purchase blanket passports in advance for large numbers of migrants, in one case as many as 3,000.62 Three firms stood out as specialists in the trade: James and Patrick Crawfurd, Archibald, Isaac and Zacharias Hope and John Sted- man. Together they may have seen off as many as half the migrants. Sted- man had been a captain and commanded six voyages between 1731 and 1738. He settled in Rotterdam and not long after two brothers established themselves as his agents in Philadelphia. While Isaac and Zachary Hope and John Stedman dominated the business before 1739 the Crawfurds got into the trade post 1763.63 The Hopes’ Quaker connections were probably responsible for early involvement and then domination, along with John Stedman, of the migrant trade up to the start of the Seven Years’ War.64 To be effective, the Rotterdam merchants had to have correspondents in London and in Philadelphia. John Stedman had his brothers in Pennsylva- nia, and the Hopes had the firm of Ben and Sam Shoemaker there. Between them they were responsible for handling about half the immigrant vessels entering Philadelphia with Germans. The function of the Philadelphia merchants changed over time. Before the 1730s migrants often paid their fares in advance or paid half in Rotterdam before embarkation, the latter being normal to protect merchants from losses and to give them capital to charter and outfit ships for the voyages. The receiving merchant then only had to settle accounts on arrival. As the flow of migrants increased, as government regulation grew and as Rotterdam merchants more actively recruited and financed the travel of those people, Philadelphia agents had to make sure migrants complied with the laws of registration, arrange for indentures and make sure funds were remitted to Rotterdam, all in addition to their earlier duties. Philadelphia merchants came to invest in the trade as well, chartering local ships and financing voyages with correspondents in Rotterdam supplying them with freights. Once established after 1763 and even more so after the American Revolution the Philadelphians

62 Wokeck, Trade in Strangers, pp. 63–5. 63 The Crawfurds had come from Scotland to Rotterdam in the 1730s. By the early nineteenth century the richest member of the British community was among its number, indicating that the migrant trade could prove profitable. For the Crawfurds continuing activity see for example, Gemeente Archief Rotterdam, Notarieel Archief, 3271/332 (1773). Klein, “ ‘Little London’,” pp. 131–2; Wokeck, “Promoters and Passengers,” p. 261; idem, Trade in strangers, pp. 70–5, 107. 64 Marten G. Buist, At spes non fracta: Hope & Co. 1770–1815: Merchant Bankers and Dip- lomats at Work (The Hague, 1974), pp. 4–11, 543n2. Hull and Brown, Benjamin Furly and Quakerism in Rotterdam, pp. 245–53, on the history of the Hopes in Rotterdam. income differentials, institutions, and religion 293 could cast their nets wider, finding immigrant cargoes in other places. The ports of embarkation of German immigrants became more diverse through 1808, a sign of the maturity of the Philadelphia merchants.65 The connection and cooperation between Dutch and English merchants declined in the second half of the eighteenth century as Amsterdam became more a financial than a trade center and Britain developed direct trades. The American Revolution further eroded what had been a productive exchange of funds, goods and information.66 The informal arrangements of the first third of the eighteenth century with their roots in religious ties changed in the face of growing migration into an institutional structure spanning the Atlantic with nodes in Rotterdam, London and Philadelphia. The business supplied seamless services to migrants easing the process and opening the door for many more people to move to the New World. The institutions of the international labor market once in place evolved and adjusted to new and reduced circumstances after 1763 but especially after 1783. The continuing trade and its diversification to include ports of embarkation as different as Lisbon in Portugal and Fred- erikstad in Norway in the first years of the nineteenth century indicate the importance of institutions in paving the way for migration in the case of Germans going to Pennsylvania and to migration in general. The income differentials between the Rhine Valley and Pennsylvania were massive. Given the economics of the situation in a frictionless labor market migration from western Germany in the eighteenth century should have been greater, more consistent and risen over time. Like German migrants who went to find work in the western Netherlands, described and analysed by Jan Lucassen,67 those same people and their relatives searching for a better standard of living logically should have kept going through Rotterdam and on to settle in North America and especially in prosperous Pennsylvania. Conditions for migration improved as the eighteenth century went on. The regulations of governments both in Pennsylvania and

65 Wokeck, “Promoters and Passengers,” pp. 263–6, 273; Wokeck, “A Tide of Alien Tongues,” pp. 151–4; idem, Trade in Strangers, pp. 93–7, 110–2. 66 The decline in the size and prosperity of the British community in Rotterdam was a reflection of the change. K. Jennifer Newman, “Anglo-Dutch Commercial Co-Operation and the Russia Trade in the Eighteenth Century,” in Wiert Jan Wieringa and Association internationale d’histoire des mers nordiques d l’Europe (eds), The Interactions of Amster- dam and Antwerp with the Baltic Region, 1400–1800 = De Nederlanden en het Oostzeegebied, 1400–1800: Papers Presented at the Third International Conference of the “Association Inter- nationale D’histoire Des Mers Nordiques De L’europe,” Utrecht, August 30th–September 3rd 1982 (Leiden, 1983), pp. 102–3; Klein, “’Little London’,” pp. 130–1. 67 Lucassen, Naar de kusten van de Noordzee. 294 richard w. unger the Netherlands worked to promote crossing the Atlantic by regulating conditions on board emigrant ships. Those vessels became more efficient over time and regular trading connections lowered overall travel times. The image of the poor but heroic migrant, faced with unspeakable treatment on board and dishonest brokers and merchants trying to rob them at every turn, undoubtedly had some basis in fact. Travellers’ accounts attest to that though the most extreme cases were the ones likely to be reported.68 Privation and corruption did not apply in all cases. There were positive reports from migrants as well and protection against exploitation increased over time. The reward of higher income in America was and remained very large by any measure and offered a continuing pull. The world of the eighteenth century was complex, and more complex than those living in later centuries often like to admit. In prevailing conditions of limited information for potentail migrants making a rational choice about where to sell their labor was all but impossible. Rather than being overloaded with masses of data ‘Palatines’ knew little about the possibilities open to them and might even have faced conflicting or incomplete reports about what to expect in the New World. Even small considerations then would have shifted the balance. The presence or threat or war, the data shows, could yield a dramatic change in migrant flows. Shifts in government policy, for example with conscription rules, could tip the balance for many. The possibility of alternative sites for migration, the competition among different potential destinations, could affect the flow of migrants to any one destination.69 Wage differentials or income differentials counted no doubt in the decisions of those Germans who made the voyage to Philadelphia. Higher incomes were by no means the only or even the dominate force in creating and directing the flow of migration. The costs of moving, the risks involved at home, on the voyage and in the new land, religious beliefs and the institutional structures in place to handle the process all affected the choices of Germans considering the trip to find a better standard of living. Small differences in circum-

68 Diffenderffer, The German Immigration into Pennsylvania, p. 57; Wokeck, Trade in Strangers, pp. 113–63. 69 Alternatives existed in the seventeenth and eighteenth centuries for people throughout northern Europe. In the seventeenth about 250,000 people from Britain moved to Ireland and in the same years Germans moved repopulating devastated areas within Ger- many. Lucassen and Lucassen, “The mobility transition revisited, 1500–1900,” p. 357. In the late eighteenth century Rhinelanders had options other than Pennsylvania in the Danube Valley, Prussia and later Russia, in the last case invited by Czarina Catherine the Great as noted above. income differentials, institutions, and religion 295 stances could translate into large differences in flows of people. Incomes were on average dramatically higher in Philadelphia but in specific years the enticement may not have been so great. Income differentials counted but even, great as they were, they were not the sole or even the most important consideration for people choosing to leave the known and voyage to the imperfectly known. Migrants wanted to know not just what the remuneration was for their labor but also the character, the conditions of and the nature of work. The mystery of what the New World would be like was a force in shaping what was an emerging international labor market.

CONTEXTUALIZATIONS

Labor Laws in Western Europe, 13th–16th Centuries: Patterns of Political and Socio-Economic Rationality

Catharina Lis and Hugo Soly

Since the 1990s, anthropologists, historians, and sociologists studying labor relations have increasingly emphasized that freedom and unfreedom are relative notions. They agree that the many middle grounds between free labor and forced labor need to be situated in specific historical and institutional contexts, but they differ regarding both the theoretical aspects of the issue and in their interpretation of the empirical data concerning free versus unfree.1 With respect to wage labor, Jan Lucassen has argued that the crucial question in all cases is whether one is free “to choose one’s own employer and therefore one’s labour conditions,” and that “in most historical situations free wage labourers didn’t have, according to their capacities, a complete free choice of employer.”2 Pre-industrial Europe was no exception. On the contrary, in various areas west of the Elbe and the Danube, the ruling classes tried from the thirteenth century onward to curtail the freedom of wage dependents as much as possible, based both on economic considerations and on the desire to maintain public order. This restrictive policy, which figured in labor laws, poor laws, and penal laws, merits special consideration, because proletarianization was a more massive phenomenon in Western Europe than anywhere else in pre-industrial society. As early as 1350, one third of English rural households is assumed to have derived a substantial share of its income from wage labor. In most other countries, the proportion was probably smaller, and in many cases wage labor constituted a part-time activity or a stage in the life-cycle. There were adult men and women who performed wage labor full-time,

1 See esp. Tom Brass and Marcel van der Linden (eds), Free and Unfree Labour: The Debate Continues (Bern, 1997); Stanley Engerman (ed.), Terms of Labor: Slavery, Freedom, and Free Labor (Stanford, 1999); Robert J. Steinfeld, Coercion, Contract, and Free Labor in the Nineteenth Century (Cambridge, 2001); Simon Deakin and Frank Wilkinson, The Law of the Labour Market: Industrialisation, Employment, and Legal Evolution (Oxford, 2005); Marcel van der Linden, Workers of the World: Essays toward a Global Labor History (Leiden, 2008), chapters 2–4; Alessandro Stanziani (ed.), Le travail contraint en Asie et en Europe, XVIIe–XXe siècle (Paris, 2010). 2 Jan Lucassen, “Free and Unfree Labour Before the Twentieth Century: A Brief Over- view,” in Brass and Van der Linden, Free and Unfree Labour, p. 47. 300 catharina lis and hugo soly whether as servants living in their employer’s household or as casual laborers, but far more smallholders, cottagers, and sub-tenants derived part of their subsistence from selling their labor power. While the size of the different categories is impossible to determine, the total number of wage dependents was undoubtedly growing everywhere.3 Employers were concerned with controlling these workers, who in many cases were personally free. This was the hardest to do in sparsely populated areas, where serfdom had never come about (Norway), or where it had virtually disappeared in the course of the thirteenth century (Sicily and Portugal). In most countries west of the Elbe, feudal relations were still significant around 1300, but a large proportion of the wage dependents consisted of personally free workers, which meant that individual landlords were not in a position to subject them to extra-economic coercion. Late-medieval labor laws need to be considered in this context. The successive epidemics, commonly designated as the Black Death, and which felled one third to one half of the European population during the fourteenth century,4 revealed how important wage labor had become in many areas west of the Elbe: several governments imposed measures prohibiting pay raises (as well as price increases in some cases) and/or restricting the geographic mobility of wage dependents. In a comparative study of labor laws promulgated in different monarchies and city-states after the Black Death, Samuel Cohn warns against viewing this legislation as a universal and uniform reaction to the same problems. He emphasizes the great diversity of responses and the absence of a comprehensive logic: “These legislative acts splintered in a multitude of directions that to date defy any obvious patterns of economic or political rationality”; in his view, they are “better understood in the contexts of fear and anxiety that sprung forth from the Black Death’s new horrors of mass mortality and destruction.”5 The author rightly argues that no single demographic or economic model explains the legislative variety, and that the timing and nature of the measures cannot be attributed to a specific political

3 The rise of wage labor in the countryside in northwest Europe between the eleventh and fourteenth centuries has not been investigated in great depth. Isaac Joshua, La face cachée du Moyen Âge (Paris, 1988), pp. 179–241, is the only author who discusses rural social change from this perspective. 4 Ole J. Benedictow, The Black Death, 1346–1353: The Complete History (Woodbridge, 2004), pp. 380–4, even advances the hypothesis that 60 percent of Europe’s population died in the epidemic. 5 Samuel Cohn, “After the Black Death: Labour Legislation and Attitudes Towards Labour in Late-Medieval Western Europe,” Economic History Review, 60 (2007), pp. 457–85. labor laws in western europe, 13th–16th centuries 301 regime or composition of the ruling elites. He is also right in his assertion that governments operated in a state of disruption and confusion that was hardly conducive to coherent actions. Still, we believe that patterns of socio-economic and political rationality are discernible. Understanding late-medieval labor laws requires keeping in mind that the “labor problem” was not identical everywhere and did not manifest itself in the same manner and with the same intensity everywhere, as well as that political-institutional options for intervening varied considerably. Not all aspects are within the scope of a brief essay. We will not discuss the various ways in which producers exploited their workers or the changing economic fortunes of wage dependents, their relations with different categories of employers, or their responses to the labor laws. Instead, we examine how public authorities defined the “labor problem,” to what extent they were able to legislate in this field, and which conditions were necessary to ensure continuity. We consider whose interests the measures served, and to what extent interventions were promoted by converging interests or thwarted by divergent ones. Nor should the rural elites be the sole focus. Sustained enforcement of countrywide labor laws required a broader social base, which meant that some aspects of such laws would need to benefit wealthy farmers as well. In several areas urban elites were also important and sometimes even decisive operators. In short, we devote special attention to political balances of power, to social differentiation within the peasantry, and to the relative autonomy and economic importance of cities.

Labor laws before the Black Death

Nearly all authors conducting comparative research on the late medieval crisis have presented the countrywide labor laws promulgated at the time in many parts of Western Europe as a new phenomenon. We agree that these laws were a milestone, in that they defined wage dependents as a problem category everywhere, and that forms of labor regulation in many areas were henceforth complemented by poor relief provisions and penal measures. High mortality rates alone, however, do not explain why so many public authorities attempted to mobilize and discipline the labor force after the Black Death. In various countries measures to this effect were taken far earlier, when there was no absolute labor shortage. Several English agrarian bylaws from the late thirteenth and early fourteenth centuries that were compulsory for all villagers comprised 302 catharina lis and hugo soly measures ordinarily regarded as characteristic of post-plague legislation. In some places able-bodied wage dependents were prohibited from moving elsewhere during the harvest season and were obliged to work; elsewhere customary wages were set for reapers; and in still other places labor coercion was applied to prohibit the able-bodied from gleaning or requesting alms, thereby depriving them of alternative sources of income. The explicit stipulations, in some instances, that tenants might compete with lords to recruit wage workers, indicates a demand among wealthy farmers as well.6 This is understandable, because according to the Hun- dred Rolls of 1279–80, the top 20 percent of all holdings each comprised at least thirty acres (12 ha).7 Such tenants were able to farm at a scale sufficient to produce surpluses for the market, take on more land, and hire servants and casual laborers.8 The problem was that demand for and supply of casual labor were not synchronous at critical moments. Densely populated counties such as Essex, for example, were areas of extensive geographic mobility.9 The ambition of manorial lords and wealthy farmers to mobilize and control labor manifested in measures to regulate itinerant men and women in search of jobs, suggesting that efforts were made to restrict the geographic mobility of wage dependents. Between 1272 and 1305 no fewer than 31 royal commissions were formed to curtail “vagrancy,” and a statute from 1331 authorized village constables to imprison suspected “idlers.”10 England was not the only country where labor regulations were introduced prior to the Black Death. Elsewhere, an absolute labor shortage

6 Warren Ortman Ault, “Some Early Village By-Laws,” English Historical Review, 45 (1930), pp. 208–31, at 211–5, and the book by the same author Open-Field Husbandry and the Village Community: A Study of Agrarian By-Laws in Medieval England (Philadelphia, 1965), pp. 13–16, 43–6; Elaine Clark, “Medieval Labor Law and English Local Courts,” Ameri- can Journal of Legal History, 27 (1983), pp. 330–53; Anthony Musson, “New Labour Laws, New Remedies? Legal Reaction to the Black Death “Crisis,” in Nigel Saul (ed.), Fourteenth Century England, vol. 1 (Woodbridge, 2000), pp. 73–88, at 75–9. 7 Junichi Kanzaka, “Villein Rents in Thirteenth-Century England: An Analysis of the Hundred Rolls of 1279–80,” Economic History Review, 55 (2002), pp. 593–618, at 599. 8 Bruce M. Campbell, “The Agrarian Problem in the Early Fourteenth Century,” Past and Present, 188 (2005), pp. 3–70, at 46–7, 67. 9 L.R. Poos, A Rural Society after the Black Death: Essex, 1350–1525 (Cambridge, 1991), p. 160. 10 Chris Given-Wilson, “Service, Serfdom and English Labour Legislation, 1350–1500,” in Anne Curry and Elizabeth Matthew (eds), Concepts and Patterns of Service in the Later Middle Ages (Woodbridge, 2000), pp. 21–37, at 29; Elaine Clark, “Institutional and Legal Responses to Begging in Medieval England,” Social Science History, 26 (2002), pp. 447–73, at 459. labor laws in western europe, 13th–16th centuries 303 was not a necessary prerequisite for imposing regulations and control either. Moreover, several thirteenth-century rulers attempted to regulate the labor supply and/or terms of employment throughout their kingdoms. While the socio-economic trends that motivated their legislative initiatives are not always clear, the measures demonstrate in any case that only a strong monarchy could promulgate labor laws that applied throughout the territory and were implemented effectively. Another factor determining whether such laws could be sustained for extended periods was the stability or instability of the political configuration. The Constitutions of Melfi—also known as the Liber Augustalis— promulgated in 1231 by Emperor Frederick II for the Kingdom of Sicily restricted the power of the nobility and placed the cities under the direct authority of the ruler and his ministers. All freemen and citizens were henceforth equal before the law. This meant that the vast majority of the population basically had legal security, as serfdom had all but disappeared. As far as labor was concerned, the royal bailiffs were ordered to set maximum wages for various occupations—mostly artisans, but also grape-gatherers and reapers—and to issue heavy penalties to those who demanded or paid higher rates. Both these initiatives were new, as no earlier examples of such government measures have been identified in European history.11 What motivated Frederick II is unclear. Did the urban elites support his measures to curtail the power of the aristocrats, and did the legislation serve to protect their interests? This seems plausible, as the erosion of central authority led city councils to control wages and regulate the labor market independently from the late thirteenth century onward, especially when the steep population decline after 1311–13 caused a labor shortage that became far more serious in the 1320s and 1330s. Wages were capped, and workers were prohibited from abandoning their employer and migrating.12 During the second half of the fourteenth century, a period of political fragmentation and economic segmentation, leading feudal families attempted in vain to reintroduce serfdom. The cities, in turn, continued to take measures to regulate the local labor market.13

11 Liber Augustalis, trans. James M. Powell (New York, 1971), pp. 133–5; Hermann Dilcher, Die sizilische Gesetzgebung Kaiser Friedrichs II. Quellen der Constitutionen von Melfi und ihrer Novellen (Cologne and Vienna, 1975), pp. 693, 695–6. 12 Clifford R. Backman, The Decline and Fall of Medieval Sicily: Politics, Religion and Economy in the Reign of Frederick III, 1296–1337 (Cambridge, 1995), pp. 38, 99–100. 13 Steven R. Epstein, An Island for Itself: Economic Development and Social Change in Late Medieval Sicily (Cambridge, 1992), pp. 88–91, 330, 373. 304 catharina lis and hugo soly

The case of Castile reveals how decisive power relationships were for implementing countrywide labor laws. In 1268, Alfonso X not only proclaimed price freezes but also decreed wage ceilings for both urban artisans and rural workers, suggesting that employers already faced a “labor problem” at the time. This is no surprise, as thousands of Muslim artisans had left the cities of Andalusia over the preceding decades, and after the revolt of 1264 the ruler had expelled the Muslims still remaining in the countryside. While Alfonso’s wage ceilings accommodated complaints from urban entrepreneurs, the debasement of the coinage and especially the levying of special taxes greatly irritated urban elites and middle groups. He moreover incurred the wrath of many noblemen, which instigated political tensions and conflicts.14 These circumstances ruled out implementing national wages. In Norway, the strong central authority and the convergence of interests between the primary social groups made implementation of labor laws possible. Well before the Black Death, elaborate countrywide labor legislation was introduced, based essentially on coercion, and remained in effect for centuries: the law code (Landslov) that King Magnus VI Lagabøte (“Law-Mender”) promulgated in 1274 throughout the country, including the Faroe Islands and Shetland. One of the articles prohibited those with little or no property from leaving their place of residence between Easter and Michaelmas. His aim was to mobilize the labor of those who were landless or nearly landless, as since the middle of the century many had complained that “everybody wants to go trading and no one wants to work for the farmers.” In 1291, it was stipulated by law that every able- bodied man and woman who owned no land was required to perform wage labor.15 The fact that Norway was a unified state, where in the late Middle Ages the monarchy remained very powerful and the aristocracy comparatively weak, explains why labor legislation could be implemented. But the question remains: why was it enforced for so long? Four factors came into

14 Maria del Carmen Carlé, “El precio de la vida en Castilla del Rey Sabio al Emplazado,” Cuadernos de Historia de España, 15 (1951), pp. 132–56; James Todesca, “Wages and Cur- rency in Thirteenth-Century Castile,” Paper presented at the symposium on Wages and Currency: Global and Historical Comparisons, held on 24–25 May 2002 in Amsterdam and Leiden. 15 See esp. Sølvi Sogner, “The Legal Status of Servants in Norway from the Seventeenth tot the Twentieth Century,” in Antoinette Fauve-Chamoux (ed.), Domestic Service and the Formation of European Identity: Understanding the Globalization of Domestic Work, 16th– 21st Centuries (Bern, 2004), pp. 175–87, at 178. labor laws in western europe, 13th–16th centuries 305 play. First, land ownership was allodial, i.e. non-feudal: the land belonged to free and sovereign owners. Second, the—relatively small—population in the thirteenth century was personally free, both legally and in practice; serfdom had never been significant, and slavery no longer existed by then.16 Third, no manorial system of large-scale farming had come about. The land held by the elite consisted largely of scattered holdings and even parts of holdings run by tenant farmers who paid rent; work on the central farm, often the seat of the owner, was done by servants and casual laborers. Last but not least: social differentiation within the peasantry was very advanced. There was both a top stratum of wealthy farmers and a broad stratum of cottagers and nearly landless.17 Demand for labor was therefore substantial, and wage dependents were the only option, as no other forms of labor mobilization existed. It was therefore essential that free wage dependents be prevented from obtaining other sources of income or migrating to cities, and this was precisely the objective of the legal measures taken by the Crown. The epidemic of 1362–63 exacerbated the “labor problem,” not only because of the intrinsic mortality, but also because many cottagers were now able to lease larger, more viable holdings. The amendments of 1364 and 1383–84 to the existing laws stipulated that royal officials were responsible for the labor supply in the countryside; in the spring and autumn, they were expected to chase from the cities all able-bodied men and women lacking adequate means of support and not yet working the fields. In the sixteenth century the labor laws were tightened: all the unemployed—except for tailors, cobblers, and carpenters—had to enter the service of farmers for at least half a year; “servants” were not allowed to trade independently in the countryside and were required to give three months notice, and “vagrants” were subject to severe punishment.18 In other Scandinavian countries the population was also fairly small and personally free in the thirteenth century, and the available evidence suggests a rise in the number of wage dependents there as well. Still, no central measures were taken to regulate rural labor in Denmark and

16 Ruth Mazo Karras, Slavery and Society in Medieval Scandinavia (New Haven and Lon- don, 1988). 17 Knut Helle, “Down to 1536,” in Rolf Danielsen et al. (eds.), Norway: A History from the Vikings to Our Own Times (Stockholm, 1995), pp. 53–6; Eljas Orrman, “Rural Conditions,” in Knut Helle (ed.), The Cambridge History of Scandinavia, Vol. I: Prehistory to 1520 (Cam- bridge, 2003), pp. 299–303, 308–11. 18 Sogner, “The Legal Status,” pp. 178–80. 306 catharina lis and hugo soly

Sweden. This was attributable to a combination of power relations and ownership structures. Between 1250 and 1350, Denmark was the scene of political turmoil and even signs of disintegration, leading the peasantry to seek protection from noblemen and religious institutions. Unlike in Norway, the Danish aristocracy controlled much of the land, a manorial system of large-scale farming came about, and the tenanted farms of the nobility and religious institutions were exempt from royal taxes. The strong aristocracy, able to exercise power in society independently from the Crown, devised separate strategies for mobilizing and controlling labor. In the century after the Black Death, large Danish landowners increased their authority over their servants and tenant farmers (both legally and in practice), regulated the mobility of the rural population, and enserfed the farmers on Sjaelland (Zealand) and the other islands.19 Clearly, some governments promulgated countrywide labor legislation long before the Black Death, and mobilizing wage labor at conditions that favored employers was the main objective. Whenever the interests of noblemen converged with those of wealthy farmers, the work requirement was focused on agriculture (Norway), whereas when the interests of rulers converged with those of cities, setting wage ceilings was a primary interest (Sicily and Castile). The implementation of legislation depended on the presence of a strong monarchy on the one hand and on converging interests on the other hand. Where rulers faced political oppositions and conflicts, wage ceilings became impossible to enforce, as the example of Castile demonstrates. Wherever the aristocracy was unwilling to cede power and tried to devise a separate labor mobilization strategy, no countrywide labor laws were promulgated (Denmark and Sweden), or they were short-lived (Sicily). In Norway, both noblemen and wealthy farmers championed labor legislation targeting the agricultural sector. Economic and demographic changes had increased the numbers of cottagers and landless, but the absence of serfdom meant that individual landowners were unable to exert extra-economic coercion; they had to rely on the monarchy to do so. Even the aristocracy, who owned comparatively little land, needed the central government to compel wage dependents through legislation to offer their labor power in the countryside. This countrywide labor legislation therefore survived and was enforced effectively.

19 Helle, “Down to 1536,” pp. 53, 64, 101; Eljas Orrman, “The Conditions of the Rural Population (c. 1350–1520),” in Helle, The Cambridge History of Scandinavia, Vol. I, pp. 581–3, 600–2. labor laws in western europe, 13th–16th centuries 307

Agrarian interests and countrywide labor laws after the Black Death

While Norway after the Black Death continued and reinforced the existing wage labor policy, several central governments promulgated the first national labor laws around that time. In Provence, ruled by the Counts of Anjou, who were also the Kings of Naples, the ordinance that the Grand Seneschal Raymond II d’Agoult proclaimed on 5–6 September 1348 was the very first attempt in Europe to cap prices and wages following the outbreak of the plague. The action was short-lived, however, for the “war of the seneschals” prevented the representatives of the House of Anjou from enforcing the legislation or taking new initiatives.20 The labor laws promulgated in England, Portugal, and some parts of Central Europe, especially in Upper Bavaria and Tyrol, on the other hand, remained in force, essentially serving agrarian interests. The difference was that the English and Portuguese kings primarily considered the interests of the aristocracy, whereas the rulers of Tyrol were more interested in improving the economic position of the wealthier farmers. All areas had a powerful central authority, and the objective was the same everywhere: to control rural labor, as manifested in comparable legal regulations. English legislation merits special consideration, however, because it gave rise to government measures that became increasingly consistent and lasted for centuries, coinciding with the rise of agrarian capitalism. As the modalities and enforcement of the Ordinance of Labourers (1349), the Statute of Labourers (1351), the Statute of Cambridge (1388), and the subsequent statutes (enacted in the 1440s and 1490s), have been examined by many authors since the pioneering work by Bertha Haven Putnam,21 we will cover only some of the highlights here.22 Various objectives of

20 Robert Braid, “Et Non Ultra: Politiques royales du travail en Europe occidentale au XIVe siècle,” Bibliothèque de l’École des Chartes, 161 (2003), pp. 437–91, here 438, 442–5, 454, 474–9. 21 Bertha Haven Putnam, The Enforcement of the Statutes of Labourers during the First decade after the Black Death, 1349–1359 (New York, 1908). 22 In addition to the references in notes 6–10, see esp. Simon A.C. Penn and Christopher Dyer, “Wages and Earnings in Late Medieval England: Evidence from the Enforcement of the Labour Laws,” Economic History Review, 43 (1990), pp. 356–76; Robert J. Steinfeld, The Invention of Free Labor: The Employment Relation in English and American Law and Culture, 1350–1870 (Chapel Hill, 1993), pp. 28–9, 35–6, 53–4; Robert C. Palmer, English Law in the Age of the Black Death, 1348–1381: A Transformation of Governance and Law (Chapel Hill, 1993); Chris Given-Wilson, “The Problem of Labour in the Context of English Govern- ment,” in James Bothwell, P.J.P. Goldberg and W.M. Ormrod (eds), The Problem of Labour in Fourteenth-Century England (York, 2000), pp. 85–100; A.L. Beier, “A New Serfdom: Labor Laws, Vagrancy Statutes, and Labor Discipline in England, 1350–1800,” in A.L. Beier and 308 catharina lis and hugo soly the English labor laws, such as controlling and inspecting wage-rates and curtailing labor mobility, derived from local antecedents but henceforth became applicable countrywide and transgressors were subject to fines and even corporeal punishment. The main objective, however, was new in its conception: all able-bodied men and women, whether free or unfree, and younger than sixty, without visible means of subsistence, and lacking a specific occupation were required to serve anybody who claimed their labor. What “visible means of subsistence” were depended on how the authorities defined work; activities by cottagers that figured in a multi- form subsistence base were regarded as useless or even as manifestations of idleness, whenever they limited the local labor supply. Persons with little or no land were forced to accept cheap annual contracts or otherwise risk being accused of “wandering about idly” or, worse still: being labelled as vagrants. From 1349, wage dependents were moreover required to work for rates that applied throughout the country or risk stiff fines. Never before had free persons been required by law to perform wage labor.23 The law stipulated that recalcitrant workers might be summarily punished; testimony from two witnesses was sufficient to convict them. The compulsory labor clause elicited enormous resistance,24 as did the contract clause, since the law declared anybody departing prior to “completing the work” to be in breach of contract and therefore punishable. Employer-employee relationships were formulated exclusively in master- servant terms, with employment contracts by definition favouring the master. A.L. Beier has rightly noted that the labor legislation drafted in England from the mid-fourteenth century onward, entailing a wide range of coercive measures, “might be considered England’s version of the “second serfdom’ that the landed classes of eastern Europe were imposing around the same time, only in the English experience it was wage labor that was mandated.”25 English labor laws were indisputably intended primarily to accommodate the demand on the part of the aristocracy and the gentry for cheap

Paul Ocobock (eds.), Cast Out: Vagrancy and Homelessness in Global and Historical Per- spective (Athens, OH, 2008), pp. 35–63; Judith M. Bennett, “Compulsory Service in Late Medieval England,” Past and Present, 209 (2010), pp. 7–52. 23 For a different view, see Musson, “New Labour Laws,” pp. 76–7 (though he admits that he is dealing with manorial regulations). 24 See inter alia Nora Kenyon, “Labour Conditions in Essex in the Reign of Richard II,” in E.M. Carus-Wilson (ed.), Essays in Economic History, II (1962), pp. 91–111; Penn and Dyer, “Wages,” pp. 366–70. 25 Beier, “A New Serfdom,” p. 56. labor laws in western europe, 13th–16th centuries 309 and docile labor. After 1349, labor availability became a serious problem, less because of an absolute labor shortage than because wage dependents could obtain better material conditions by migrating to areas where wages were higher. The elite therefore had good reason to restrict the leeway of cottagers and landless as much as possible. But labor laws also served the interests of wealthy farmers, even though their objectives did not always coincide with those of their social superiors. Administrative-legal reforms enforced the rights of all employers with respect to their employees, so that lords could no longer claim priority. Indications abound that wealthy farmers did indeed use this legislation, both to discipline wage dependents and to sue neighbors for paying excessive wages or stealing their servants.26 Since many labor law violations were defined as refusing to work, and involuntary unemployment among able-bodied men and women lacking a means of subsistence was regarded as such as well, labor legislation was inextricably linked with both the prohibition of begging and prosecution of vagrancy and with highly restrictive poor relief. The different strands of legislation continued to coexist and complement one another in the early modern period. The Statute of Artificers, enacted in 1563, following a mortality crisis in the late 1550s and a rise in nominal wages,27 reaffirmed all fundamental principles set forth in late-medieval laws, elaborated them in detail, and introduced some innovations: compulsory service, the contract clause, restricted labor mobility, official wage rates, attempts to tie workers to particular occupations and employers, and apprenticeship clauses (an innovation).28 Fragmentary data for Norfolk suggest that the enforcement system was comprehensive and relatively efficient.29 In the sixteenth century, curtailment of vagrancy and selection of the “deserv- ing” poor—who qualified for support—were provided for in separate laws

26 L.R. Poos, “The Social Context of Statute of Labourers Enforcement,” Law and His- tory Review, 1 (1983), pp. 27–52; Clark, “Medieval Labor Law”; Penn and Dyer, “Wages,” p. 359; Given-Wilson, “Service,” pp. 22–3, 27, 30; Jane Whittle, The Development of Agrar- ian Capitalism: Land and Labour in Norfolk, 1440–1580 (Oxford, 2000), pp. 296–7; Bennett, “Compulsory Service,” p. 47. 27 Donald Woodward, “The Background to the Statute of Artificers: The Genesis of Labour Policy, 1558–63,” Economic History Review, 33 (1980), pp. 32–44; John S. Moore, “Demographic Dimensions of the Mid-Tudor-Crisis,” Sixteenth Century Journal, 41 (2010), pp. 1039–63. 28 The Statute made it illegal to practise any craft without serving seven years as an apprentice first. On these clauses (repealed in 1814), see Margaret Gay Davies, The Enforce- ment of English Apprenticeship, 1563–1642 (Cambridge, MA, 1956). 29 Whittle, The Development of Agrarian Capitalism, pp. 275–304. 310 catharina lis and hugo soly but continued to correlate with the labor laws. The chief objectives of the Elisabethan poor laws of 1601 were to regulate employment and to discipline the labor force.30 A strong state, growing significance of wage labor, and convergence of agrarian interests: Portugal resembled England in these respects. This is also apparent from the labor laws that the Portuguese monarchy enacted in 1349 and the years that followed. Both noblemen and wealthy farmers had an interest in regulating rural labor, as the small population of the country meant that a demographic crisis had serious consequences. Besides, the majority of rural dwellers was free; serfdom had virtually disappeared during the thirteenth century. Most members of the elite did not manage their lands directly but divided them into small units that they leased to tenants, resulting in many areas in a complex patchwork of minifundia worked by jornaleiros.31 Precisely because mobilizing wage labor was so important for different social groups, the central government took measures: wage ceilings were set, jornaleiros were prohibited from leaving their work on farms and migrating to the cities, and “unauthorized beggars” were subject to prosecution. A passport system was introduced to control the comings and goings of farm workers and to distribute labor as equitably as possible among the landlords.32 The Lei das Sesmarias, promulgated by King Fernando I in 1375, went much further: not only were itinerant workers labelled and prosecuted as vagrants, but countrydwell- ers with property below a specified value were required to remain on the land, together with their sons and grandsons, and they were forced to accept low wages. Begging was restricted to those who were weak, elderly, or infirm, i.e. those who were unable to work.33 Whether these measures produced the effects envisaged in the long run is doubtful. At any rate, the strong growth of the cities reveals that migration was impossible to prevent. In the late fourteenth century, protest movements repeatedly materialized as well, both in the countryside and in the cities. Eventually,

30 As Richard H. Tawney, The Agrarian Problem in the Sixteenth Century, ed. Lawrence Stone (New York, 1967), p. 47, put it: “Villeinage ceases, but the poor laws begin”. See Beier, “A New Serfdom”. 31 Anthony R. Disney, A History of Portugal and the Portuguese Empire from Beginnings to 1807, vol. 1: Portugal (Cambridge, 2009), p. 98. 32 Antonio Henrique de Oliveira Marques, History of Portugal, Vol. I: From Lusitania to Empire (New York, 1972), p. 110; Humberto Baquero Moreno, Marginalidade e Conflitos Sociais em Portugal nos Séculos XIV e XV (Lisbon, 1985), p. 27. 33 Virginia Rau, Sesmarias Medievais Portuguesas (Lisbon, 1982; originally published in 1945); Laurinda Abreu, “Beggars, Vagrants and Romanies: Repression and Persecution in Portuguese Society (14th–18th Centuries),” Hygiea Internationalis, 6 (2007), pp. 41–66, at 44. labor laws in western europe, 13th–16th centuries 311 general wage rates ceased to apply. The penal laws against begging and vagrancy, however, remained in effect and were tightened in the sixteenth century.34 Labor laws that primarily served agrarian interests were promulgated in two other areas after the Black Death: Upper Bavaria and Tyrol, which had the same ruler in the mid-fourteenth century. In 1352 Louis V, Duke of Bavaria, stated “that we have seen the weakness and the harm which have been brought about in our land of Upper Bavaria by farmhands and workers, that every man seeks the highest wage he can secure and will undertake no cultivation.” Accordingly, he set maximum wages and provided severe punishment for all violators. He did the same for Tyrol and stipulated that whoever paid or received higher wages would be fined or imprisoned. In addition, he prohibited domestic servants and casual laborers from migrating to another community, on pain of having all their possessions seized; they had to continue working for their current employer at the same rate of pay. Whether the measures benefited wealthy farmers in Upper Bavaria as well is unclear, but they certainly did so in Tyrol. For political-military and tax-related reasons, the peasantry was protected by the sovereigns. Nor did they lack political influence: representatives of the peasantry sat in the Landtage as a separate estate alongside the aristocracy, the clergy, and the cities.35

Royal interventionism and urban wealth

After the Black Death, rulers in several countries promulgated labor laws that in theory applied throughout their territory but were intended primarily to enable urban employers to mobilize the labor of wage dependents as cheaply as possible. The rulers involved had good reasons to pursue such a policy. Promoting the economic interests of urban elites and middle groups benefited the royal treasury and moreover helped them increase their control over local politics. In some cases rulers relied on a traditional

34 Disney, A History of Portugal, Vol. 1: Portugal, p. 110; Abreu, “Beggars,” pp. 44–54. 35 Karl Moeser, “Die drei Tiroler Wirtschaftsordnungen aus der Pestzeit des 14. Jah- rhunderts,” in Ernest Troger and Georg Zwanowetz (eds), Beiträge zur geschichtlichen Landeskunde Tirols: Festschrift für Franz Huter (Innsbruck, 1959), pp. 253–63; Friedrich Lütge, “The Fourteenth and Fifteenth Centuries in Social and Economic History,” in Gerald Strauss (ed.), Pre-Reformation Germany (New York, 1972), pp. 316–79, 349–51; Michael Mit- terauer, “Die Wirtschaftspolitik der österreichischen Landesfürsten im Spätmittalter und ihre Auswirkungen auf den Arbeitsmarkt,” in Hermann Kellenbenz (ed.), Wirtschaftspolitik und Arbeitsmarkt (Vienna, 1974), pp. 15–46, at 20–2. 312 catharina lis and hugo soly oligarchy and in others on those without privileges, but their measures were in any case economically beneficial to urban employers. In Castile the labor laws applied in all cities of the kingdom, whereas elsewhere they catered to a single, major city that was politically and economically very influential: Barcelona, Paris, and Vienna. The labor legislation in Castile and Aragon might initially seem to concern the countryside, as both kingdoms were sparsely populated and had been experiencing migration waves that seriously impacted the agrarian output in different regions for decades. The outbreak of the plague further complicated mobilizing labor. Still, the labor laws that came about primarily concerned the cities. Why? In Castile cities exercized a type of “collective lordship,” implying extensive jurisdictional powers over rural society, and they experienced rapid economic growth, enabling them to become increasingly important in politics. For tax-related, economic, and political reasons, the monarchy therefore had every interest in controlling the urban centers. The ordenamientos promulgated by Pedro I in 1351 capped wages. The king did not take more extensive measures, as following the premature death of his predecessor, he operated in a complex field of forces, where he soon clashed with the exceptionally powerful aristocracy. Moreover, he had to let the city councils adapt wages to reflect the changed local circumstances, making a centralized economic policy impossible. After the end of the civil war in 1369, the effort was resumed by his successor Enrique of Trastamura, who ordered that master artisans and urban workers be controlled, and that wage increases be prevented: all able-bodied people over twelve without any means of subsistence were required to work for the legal wage, and unwilling workers were to be flogged.36 Over the decades that followed, the monarchy promulgated poor laws, stipulating that anybody was allowed to capture itinerant “beggars” and have them

36 The basic study of the late medieval Spanish labor laws remains Charles Verlin- den, “La grande peste de 1348 en Espagne: contribution à l’étude de ses conséquences économiques et sociales,” Revue Belge de Philologie et d’Histoire, 17 (1938), pp. 103–46. See also Adeline Rucquoi, “Pouvoir royal et oligarchies urbaines d’Alfonso X à Fernando IV de Castille,” in Genesis medieval del estado moderno: Castilla y Navarra, 1250–1370 (Vall- adolid, 1987), pp. 173–92; Clara Estow, Pedro the Cruel of Castile, 1350–1369 (Leiden, 1995), pp. 40–57; Isabel del Val Valdivieso, “Urban growth and Royal Interventionism in Late Medieval Castile,” Urban History, 24 (1997), pp. 129–40; Teofilo F. Ruiz, Spain’s Centuries of Crisis, 1300–1474 (Oxford, 2007), pp. 30–7, Pablo Sánchez León, “Changing Patterns of Urban Conflict in Late Medieval Castile,” in Christopher Dyer, Peter Coss and Chris Wick- ham (eds), Rodney Hilton’s Middle Ages: An Exploration of Historical Themes (Oxford, 2007; Past and Present Supplement 2), pp. 220–1. labor laws in western europe, 13th–16th centuries 313 work without pay for a month. Enforcement, however, was left up to the local councils, and this policy was continued in the fifteenth and sixteenth centuries.37 In Aragon the weak monarchy had to reconcile different and often conflicting political and economic interests of rural and urban elites in the three realms of the confederation, thereby excluding broad social support for labor laws. The ordinance that King Pedro IV proclaimed for the County of Catalonia in 1349 applied in particular to the port of Barce- lona, the economic center of his kingdom, where according to the ruler, workers were demanding wages that were four or five times what they had received before the plague. He assigned a commission to investigate whether their demands were justified. If not, they would have to continue working for the old rates. He prohibited master artisans from forming associations to obtain higher wages and threatened violators with corporeal punishment. The next year the king tried to implement this wage policy in rural areas as well, but in 1352 he had to abandon this effort.38 The noblemen preferred a strategy that placed them in charge: tightening seigneurial control. As Paul Freedman has noted: “They combined tightening their hold on tenants with a measure of concessions with regard to specific obligations.” This inconsistent approach, though effective in the short and medium term, was reversed by the peasant revolts in the late fifteenth century.39 After the Black Death, the French rulers were in no position to regulate economic affairs adequately throughout the kingdom. In 1328 the change of dynasty upset the balance of power, and the catastrophic start of the Hundred Years’ War further heightened political tensions and deepened regional divisions. The Great Ordinance that John II, surnamed “the Good,” proclaimed in January 1351 applied only to the part of France where the king truly exercised power: Paris, the capital and largest city of France (and of Europe north of the Alps), and the surrounding area (the Île-de-France), where the domaine royal was situated. In addition to

37 Linda Martz, Poverty and Welfare in Habsburg Spain: The Example of Toledo (Cam- bridge, 1983), pp. 7–22. 38 Verlinden, “La grande peste”. For the political, social, and economic context in Ara- gon, see esp. Melanie V. Shirk, “The Black Death in Aragon, 1348–1351,” Journal of Medieval History, 7 (1981), pp. 357–67. 39 Paul Freedman, The Origin of Peasant Servitude in Medieval Catalonia (Cambridge, 1991). See also the interesting comments of Robert Brenner, “The Rises and Declines of Serfdom in Medieval and Early Modern Europe,” in M.L. Bush (ed.), Serfdom and Slavery. Studies in Legal Bondage 5 (London and New York, 1999), pp. 247–76, at 270–1. 314 catharina lis and hugo soly setting ceilings on prices and wages, the ordinance included an extensive series of measures to curtail begging. Unemployed, able-bodied men and women were required to accept any work offered to earn their keep. Both the mendicants and the inhabitants of Paris were prohibited from giving alms to those capable of working, and this category was very comprehensive indeed, excluding only the blind, the disabled, and other “unfortunate persons.” In a new ordinance from November 1354, it was argued that the merchants deplored the “high cost of workers,” who not only demanded “unreasonable wages” but moreover worked only when it suited them. According to the authorities, setting maximum rates of pay had been pointless, as, contrary to all good customs and practices, many now refused to work for day wages, unless their employer provided food as well, whereas others would work only for piecework rates to circumvent the legislation, and still others moved to regions where maximum rates were not enforced. What was more: “Increasing numbers of greedy, lazy workers spend their time at the tavern, where they indulge in games and leisure pursuits.” The ordinance listed several occupations for which those practising them were henceforth required to appear at a set venue before sunrise, requiring them to accept any job they were offered at the official daily wage. All those who failed to do so or were caught during working hours—from sunrise to sunset—casting dice or hanging about a tavern would be labelled as unwilling to work and punished. The king also took measures to restrict migrations by rural dwellers, on the one hand because individual employers in the Paris region could not mobilize legally free wage dependents through coercion,40 and on the other hand for fear of disturbances of public order. Migration was strictly prohibited, except for seasonal workers, and even they were allowed to move on to another place only if they found no work where they lived. Able-bodied beggars/ vagrants would be imprisoned, exhibited at the stake, and/or branded.41 The ordinance of 1354 essentially applied throughout the kingdom, but

40 On the legal condition of the peasantry in the Paris region, see Guy Fourquin, Les campagnes de la région parisienne à la fin du Moyen Âge (Paris, 1964), pp. 166–8, 189–90. 41 Bronislaw Geremek, Le salariat dans l’artisanat parisien aux XIIIe–XVe siècles. Étude sur le marché de la main d’oeuvre au Moyen Âge (Paris, 1968), pp. 122, 126–42, and the articles of the same author: “La lutte contre le vagabondage à Paris aux XIVe et XVe siècles,” in Luigi De Rosa (ed.), Ricerche storiche ed economiche in memoria di Corrado Barbagallo (Naples, 1970), Vol. 2, pp. 211–36, at 230–6, and “Le refus du travail dans la société urbaine du Bas Moyen Âge,” in Jacqueline Hamesse and Colette Muraille-Samaran (eds), Le travail au Moyen Âge. Une approche interdisciplinaire. Actes du Colloque international de Louvain-la- Neuve, 21–23 mai 1987 (Louvain-la-Neuve, 1990), pp. 379–94, at 388–9. See also Epstein, Wage Labor, pp. 233–6, and Françoise Auffray, Charles V, le Sage (Paris, 1994), pp. 33, 57, 87. labor laws in western europe, 13th–16th centuries 315 whether it was enforced outside the Île-de-France is doubtful. At any rate, the French monarchy allowed local councils to decide whether or not to set wage ceilings to regulate the labor market. This practice continued in the fifteenth and sixteenth centuries. In Austria the wage policy introduced by Duke Albert II in 1352 affected only the citizens of Vienna, as the influence of the central authority was severely curtailed by aristocratic and ecclesiastical overlords elsewhere in the territory. It was part of a systematic policy on taxes and economic affairs, since the year before the duke had granted the merchants in the capital new privileges to trade with other countries and control the main road to Venice. Viticulture and wine trade were the main sources of income for many residents of the capital, which is why the ordinance primarily concerned this sector. It remained in effect for a long time, because the highly labor-intensive viticulture continued to expand. In the 1350s and 60s new maximum wages were repeatedly proclaimed for vineyard workers, and over the following decades the focus on their activities intensified, as excise duties on wine consumption accounted for a progressively larger share of tax revenues. In the fifteenth century maximum wages were frequently set for this category of workers. These measures indisputably benefited the merchants and wealthy master artisans of Vienna, but they also helped line the coffers of the central government, enabling it to strengthen its authority over the capital.42

Urban interests and labor laws

Rulers were not the only ones who took measures after the Black Death to mobilize the labor of wage dependents and minimize labor costs. In some parts of Western Europe city councils were especially active in this respect. Their interventions obviously served the interests of the urban elites and middle groups, but they varied considerably. Differences in any case existed between (1) city-states in North and Central Italy, where the labor laws applied primarily to the rural part of their territory, and (2) politically influential economic centers, which were concentrated mainly in heavily urbanized regions, such as the Low Countries, the Rhineland, and Southern Germany.

42 Mitterauer, “Die Wirtschaftspolitik,” pp. 21, 24–5; Peter Csendes and Ferdinand Opll (eds), Wien: Geschichte einer Stadt von den Anfängen bis zur Ersten Türkenbelagerung (Vienna, 2001), pp. 123–4, 128, 222–5 (with references to the literature). 316 catharina lis and hugo soly

Protecting the monopolies and privileges of their wealthy citizens was of course the main objective of the city-states in North and Central Italy. While they reacted in various ways to the labor shortage, most drafted legislation that applied mainly to the rural part of their territory, as small, competing political entities had ample reason to secure their supply and tax base. In 1349 Florence subjected farm workers to a series of measures, of which the general designation left nothing to the imagination: contra laboratores (against the workers). Their purpose was not merely to obstruct the mobility of servants and casual labor but also to tie mezzadri (share- croppers) to the land. Anybody leaving the farm where he worked without permission would receive a fine beyond the means of even the wealthiest peasant, basically forcing him into exile. In 1355 the punishments became still more draconian: workers damaging the farm of a Florentine citizen by neglecting the work on the fields or in the vineyards were labelled rebels and as such liable to be executed. Following the second outbreak of the plague (in 1362–63), this policy proved untenable and even counterpro- ductive for both economic and political reasons. The stick made way for the carrot: between 1364 and 1434 no fewer than 27 laws were adopted “for the benefit of the peasants,” greatly improving conditions for mezzadri and farm workers. Florence was unique, however, as following the Black Death laws in the other Italian city states curtailing the mobility of farm workers and restricting the rights of mezzadri did not threaten offenders with very severe punishments. Orvieto and Siena even quickly reversed their measures, granting tax exemptions to workers arriving in the city or surrounding villages to promote repopulation of their territories. As for the policy regarding master artisans and workers in the cities, Florence once again differed from the other Italian city-states, but in no case were attempts made to reduce wages to their pre-plague levels, on the contrary: they were revised upward.43 In areas where no city-states arose, and where rulers did not promulgate labor laws, or where their ordinances went unheeded, the councils of major economic centers had every reason to take charge. This obviously required a certain measure of political autonomy and convergence of local interests. When it became clear that the central authority was not in a position to enforce wage-rates in Provence, the council in Marseille, the harbor city where the plague had first manifested in France, and where maritime trade continued to decline, engaged in very intensive legislative

43 For an excellent analysis, see Cohn Jr., “After the Black Death,” pp. 465–78. labor laws in western europe, 13th–16th centuries 317 efforts: between 1348 and 1368 over thirty labor market regulations were proclaimed, requiring all able-bodied men and women to work, setting maximum wages, prohibiting casting dice during working hours, and so on and so forth.44 In urbanized areas, such as the Low Countries, the Rhineland, and Southern Germany, overlords did not promulgate labor laws, because they had to consider the divergent economic interests and political power positions of both aristocratic landowners and urban elites and guild-based organizations. Noblemen and wealthy farmers ordinarily favoured uniform low wages, whereas urban merchants and master artisans by definition attributed great value to the level of skill and consequently regarded wage differentiation as a socio-economic necessity. Further complicating matters: cities sometimes had common political interests, but they tended to compete economically, which meant that they rarely synchronized their labor regulations. Some city councils actively recruited certain categories of—generally—skilled workers and did not hesitate to offer them additional benefits to entice them to leave their present place of work. The attitudes of city councils toward wage labor consequently varied far more than those of central governments, because of all the factors involved: the extent of political autonomy, the nature of the urban economy, the composition of the elite, the local balances of power, and relations with other cities. Nearly everywhere, however, urban policy consisted of (1) a very limited selection of economic sectors where wage ceilings applied, and (2) consensus about the work requirement for able-bodied wage dependents, in which penal laws against begging and poor relief provisions were crucial. The earliest labor laws in the Low Countries concerned the wages and working conditions of urban textile workers. From the late thirteenth century (and perhaps even earlier) in the large wool production centers in the County of Flanders efforts were made to keep master artisans and wage workers “in their place,” which included manipulation of the fullers’ wages by the drapers.45 In the first half of the fourteenth century measures

44 Braid, “Et Non Ultra”; Jean-Paul Boyer, “1245–1380: l’éphémère paix du prince,” in Mar- tin Aurell, Jean-Paul Boyer and Noël Coulet, La Provence au Moyen Âge (Aix-en-Provence, 2005), pp. 254–5, 263–6, 275, 278. 45 Marc Boone, “Social Conflicts in the Cloth Industry of Ypres (late 13th–early 14th Centuries): The Cockerulle Reconsidered,” in Marc Dewilde, Anton Ervynck and Alexis Wielemans (eds), Ypres and the Medieval Cloth Industry in Flanders: Archaeological and Historical Contributions (Zellik, 1998), pp. 147–55. 318 catharina lis and hugo soly to regulate the labor market followed in other cities in the Low Coun- tries. Some magistrates threatened “servants” who left their employer before completing the work with stiff fines. After 1350, such interventions occurred in far more cities and with increasing frequency, and the penalties for violations rose, especially those concerning breach of contract.46 This was also the case in many large and medium-sized economic centers in the Rhineland and Southern Germany. The authorities capped wages for journeymen masons and carpenters, threatened those who refused to accept this rate or abandoned their employer before completing the job with a ten-year expulsion, and allowed immigrants to come work in the city without requiring them to join the masons’ guild. In many cities both workers demanding wages higher than the official rates and employers who paid them risked fines and/or imprisonment. Punishments were also meted out to those accepting a new and possibly better paid job before finishing the one they were presently doing, as well as to those regarded as vagrants declining work offered by the city council at the fixed rate.47 In the sixteenth century city councils in the urbanized regions of North- west Europe continued to take measures to regulate the local labor market,48 while the higher authorities restricted their efforts to prosecuting and punishing “idlers.” In 1531 Emperor Charles V ordered the city councils in the Low Countries to reorganize poor relief provisions immediately based on uniform principles: all able-bodied men and women were required to work, begging was strictly prohibited, and all existing relief funds were to be centralized to enable efficient control. In practice, however, he left implementation of such a programme to the local authorities—which fifteen or sixteen cities did effectively.49 In 1561, the Estates of Brabant proposed setting maximum wages for all categories of wage earners in the Habsburg Netherlands (the “Seventeen Provinces”). Both nominal

46 J.W. Bosch, “Rechtshistorische aanteekeningen betreffende de overeenkomst in het huren van dienstpersoneel,” Themis, 92 (1931), pp. 355–418. 47 See esp. Ernst Kelter, “Das deutsche Wirtschaftsleben des 14. und 15. Jahrhundert im Schatten der Pestepidemien,” Jahrbuch für Nationalökonomie und Statistik, 165 (1953), pp. 161–208; Rainer Schröder, Zur Arbeitsverfassung des Spätmittelalters. Eine Darstellung mit- telalterlichen Arbeitsrechts aus der Zeit nach der grossen Pest (Berlin, 1984), pp. 157, 184–5; Eberhard Isenmann, Die deutsche Stadt im Spätmittelalter (Stuttgart, 1988), p. 390. 48 See, for example, Etienne Scholliers, “Vrije en onvrije arbeiders, voornamelijk te Antwerpen in de 16e eeuw,” Bijdragen voor de Geschiedenis der Nederlanden, 11 (1956), pp. 285–322. Maximum wages were set both in cities where craft guilds had political input and in those where they did not. Corporative organizations often set wages on their own. 49 Catharina Lis and Hugo Soly, Poverty and Capitalism in Pre-Industrial Europe (Brigh- ton, 1982 edn), pp. 87–8. labor laws in western europe, 13th–16th centuries 319 and real wages had indeed risen sharply because of the mortality of 1558, the strong economic growth since 1559, and the plummeting prices as a consequence of abundant harvests. The proposal was debated at length by the representatives of other provinces and the central government, but opinions were deeply divided. A bill providing for maximum wages and designating those refusing to work for the official rates as idlers, which meant imprisonment or being sentenced to the galleys, was not passed in the end.50 When massive emigration to the North after 1585 instigated a labor shortage and concurrent wage increases, while the devastation in rural areas drove prices up sharply, the royal officials were ordered in 1588 to set wage ceilings and maximum prices in each province. One year later the central government issued a new ordinance demanding rigid compliance with official rates, achieving little or no effect, and subsequently abandoned the effort. Labor regulation was once again left to the local councils.51

Conclusion

In the Central Middle Ages wage labor became increasingly significant in many parts of Western Europe. While the prevalence and nature of wage labor varied considerably from region to region, in the countryside and cities alike, labor by wage dependents became a progressively larger share of the total labor output. Increasing intra-rural and rural-urban mobility promoted the rise of a free labor market, but these dynamics by no means met the needs of the employers, whether they were feudal lords or wealthy farmers. In their view, the different forms of wage labor had in common not only that the employees concerned had to submit to the authority of an employer but also—and above all—that they could not dispose freely of their labor power: those who were not economically independent had no “property in labor” and consequently did not have

50 Charles Verlinden and Jan Craeybeckx, Prijzen- en lonenpolitiek in de Nederlanden in 1561 en 1588–1589. Onuitgegeven adviezen, ontwerpen en ordonnanties (Brussels, 1962), pp. 7–23. See the pertinent remarks of Bas J.P. van Bavel, “Rural Wage Labour in the Sixteenth- Century Low Countries: An Assessment of the Importance and Nature of Wage Labour in the Countryside of Holland, Guelders and Flanders,” Continuity and Change, 21 (2006), pp. 37–72, at 66. 51 Verlinden and Craeybeckx, Prijzen- en lonenpolitiek, pp. 23–32; Etienne Scholliers, Loonarbeid en honger. De levensstandaard in de XVe en XVIe eeuw te Antwerpen (Antwerp, 1960), pp. 144–8, and the article by the same author: “Le pouvoir d’achat dans les Pays-Bas au XVIe siècle,” in Album Charles Verlinden (Ghent, 1975), pp. 305–30. 320 catharina lis and hugo soly the right to deprive others of the use and enjoyment of their labor. The thirteenth-century countrywide labor laws promulgated in Norway demonstrate that wage dependents were not regarded as “free workers,” even though they were all personally free. The authorities deprived them of the right to refuse work, if they had no sources of income, or to migrate and/ or negotiate their wages. From the mid-fourteenth century, elites everywhere took note of wage labor and increasingly labelled the phenomenon as “problematic.” After all, all sources of cheap labor were on the verge of drying up simultaneously: the massive death rate enabled smallholders to acquire—additional—tracts of land, reducing their dependence on wage labor, and encouraged the landless to migrate to the places with the most favourable working conditions. In many countries, the public authorities therefore restricted the leeway of wage dependents as much as possible. Able- bodied men and women without visible means of subsistence had a duty to work; if not, they risked being labelled as idlers, beggars, or—still worse—vagrants. During the Great Revolt of 1381 the rebels demanded not only that King Richard II free all serfs in England, but also that “no one be required to serve unless it is of their own free will and according to free contract,”52 which basically entailed abolishing labor legislation. The notorious response from the king to a delegation of the rebels that they were and would remain serfs reflected little insight into the future of the seigneurial regime, but his subsequent prediction (“you will remain in bondage, not as before, but incomparably harsher”) was right on target, as far as wage dependents were concerned. Even though they were not serfs or tenants burdened with labor services, they were not free to enter contract relationships as free and equal partners. Labor legislation of a comparable scope did not come about in all areas west of the Elbe. Explaining all the spatial variations would require far more research, but the data available help us discern various patterns. First, the need for labor laws was not equally great everywhere. In countries where the feudal forces were strong, and the monarchy was relatively weak, the old strategies of extra-economic coercion persisted for a long time. This was the case not only in Denmark and Sweden in the thirteenth century (and afterwards) but also in Catalonia and Scotland

52 Quoted in Kellie Robertson, “Branding and the Technologies of Labor Regulation,” in Kellie Robertson and Michael Uebel (eds), The Middle Ages at Work: Practicing Labor in Late Medieval England (Basingstoke, 2004), pp. 133–53, here p. 140. labor laws in western europe, 13th–16th centuries 321 in the late Middle Ages. While the Catalonian aristocracy eventually lost out, the Scottish magnates managed to obtain power in vast territories and to control rural labor for centuries without any interference from the monarchy.53 Second, a strong central authority and convergences of political and economic interests were essential conditions for implementing a countrywide labor legislation that benefited both landlords and wealthy farmers, as the cases of England, Portugal, and Tyrol demonstrate. In these countries labor legislation served essentially the same purposes: averting pay increases, restricting the mobility of wage workers, and enforcing contracts that favoured employers, especially in the countryside. The fact that only the English government measures were sustained for centuries and were adapted during the early modern period to accommodate the economic changes is undoubtedly attributable to the rise of agrarian capitalism. Third, countrywide labor legislation came about in several areas only because the local ruler hoped to derive political, economic, and/or tax benefits from it, which basically meant labor regulation in and around urban centers—all cities in Castile, but only the most affluent city in Aragon (Barcelona), France (Paris), and Austria (Vienna), as the central authority had little or no influence elsewhere. Finally, in the most economically progressive parts of Western Europe the urban elites were the ones who regulated labor. Italian city-states with fairly vast territories obviously acted as central governments and focused on rural wage dependents. In the Low Countries, the Rhineland, and Southern Germany, rulers and overlords were unable to impose uniform measures to control labor, but the councils of important economic centers had sufficient political autonomy to control industrial labor themselves. The absence of a centrally conceived policy and inter-city rivalry resulted in a wide variety of—largely discontinuous—measures. Nearly all local authorities, however, were willing to accommodate the demand from merchant entrepreneurs and wealthy master artisans to regulate the labor force, as is clear from the way they used the poor relief system as a direct or indirect means of labor coercion.

53 T.C. Smout, A History of the Scottish People, 1560–1830 (London, 1969), pp. 31–9.

The First “Male Breadwinner Economy”? Dutch Married Women’s and Children’s Paid and Unpaid Work in Western Europe Perspective, c. 1600–1900

Elise van Nederveen Meerkerk

Introduction

Exactly 65 years ago, in 1947, Marga Klompé was selected as a delegate to visit the United Nations. Soon after, Klompé became Member of Parlia- ment for the Roman Catholic Party, and in 1956 she was appointed the first female Minister in the Netherlands. Klompé, the daughter of a small entrepreneur, studied Chemistry and Physics in Utrecht, and worked as a teacher. During the Second World War, she was active in the resistance against the German occupiers. Her life was quite unusual, and not only because she became a prominent person in politics.1 Unlike most women in this period, Klompé pursued a paid career and did not marry—rather striking in a period when women’s labor force participation was at an all time low. Indeed, statistics show that about 65 years ago, the share of women active in the labor market was historically low in most West- European countries, and proportionally lowest in the Netherlands.2 Instead, the “male breadwinner model” was in full bloom at the time: the husband was the only member of the family active in market work, while the wife’s responsibility was to do unpaid reproductive tasks, such as cooking, cleaning and the bearing and raising of children. However, this “male breadwinner” family-model was in truth an exception in history. Presumably, it had already existed as an ideal for much longer, but the historical evidence shows that it was realized among large segments of Western European societies only from in the late 19th century, and dominated until the mid-1960s approximately, after which more and more women across Europe (re-)entered the labor market.3 For many

1 (18 July 2011). 2 Hettie A. Pott-Buter, Facts and Fairy Tales about Female Labor, Family and Fertility. A Seven-country Comparison (Amsterdam, 1993), p. 21. 3 Jan de Vries, The Industrious Revolution. Consumer Behavior and the Household Econ- omy, 1650 to the Present (Cambridge, 2008), especially pp. 186–237. 324 elise van nederveen meerkerk non-Western countries, it is questionable to what extent the model of the “male breadwinner” has actually ever been achieved for large segments of the population. But even in European countries, such as Belgium or Spain, many families remained dependent upon the earnings of women and children until well into the 20th century.4 Moreover, the concept of the “male breadwinner” is focused on work for wages, while other forms of remuneration were important as well, perhaps even more so for women and children. Undeniably, the economic contributions of women and children were very important, both in pre-industrial and in industrializing Europe throughout the 17th, 18th and 19th centuries, whether directly in terms of wage labor, or more indirectly, by bringing home income in kind, or by being active in their husbands’ and fathers’ enterprise. Nevertheless, many historians have claimed that there was one society where the ideal of bourgeois domesticity—and therewith the model of the breadwinning husband and father—already pervaded most social echelons during the 17th century. In the prosperous Dutch Republic, the “first modern economy,” many women did not “have to” work, because of the relatively high standard of living compared to other Western Euro- pean countries. This supposedly “constrained their work lives already in the 17th century”, and would explain the historically low labor force participation of Dutch women compared to many other countries.5 In what follows I will show that the participation of Dutch women and children in gainful employment was not lower, and perhaps even higher than in surrounding countries such as England and Germany, at least until the second half of the 19th century. Insofar as we can make reliable measurements of women’s and children’s work activities, we may just have to conclude that they differed greatly according to geographic region and economic structure. Also, it is important to explore the relationship between the share of women and children working for wages and other forms of gainful employment. For the Northern Netherlands, I will try to estimate the involvement of women and children throughout the period under investigation, and make sketchy comparisons with England and

4 Angélique Janssens, “The Rise and Decline of the Male Breadwinner Family?,” Inter- national Review of Social History, Supplement 5 (1997), pp. 2–23, at 6–9. 5 Jan de Vries and Ad van der Woude, The First Modern Economy. Success, Failure, and Perseverance of the Dutch Economy, 1500–1815 (Cambridge, 1997), pp. 604–5 (quote on p. 605). See also: Simon Schama, The Embarrassment of Riches. An Interpretation of Dutch Culture in the Golden Age (London, 1987), p. 407; Pott-Buter, Facts and Fairy Tales, pp. 48, 282; Els Kloek, Vrouw des huizes. Een cultuurgeschiedenis van de Hollandse huisvrouw (Amsterdam, 2009), p. 77. the first “male breadwinner economy”? 325

Germany, in order to evaluate the Dutch findings. For the end of the chosen period, 1900, I will provide estimates of Dutch women’s and children’s share in various forms of work activities. Inevitably, the question of definition arises—what is work? In Chris and Charles Tilly’s description, work is “any human effort adding use value to goods and services”.6 Accepting this definition means looking beyond the traditional focus on “wage labor,” and including many other kinds of economic activities, both market-oriented and non-market. Figure 1 tabu- lates the forms of work which were important in the period and regions under study here. Although throughout Western European history many people have been involved in non-wage work, either for their own use or for the market, unpaid and non-monetary paid work have probably been most important for women and children.8 More often than men, they were responsible for cleaning and mending the house, or “assisting” in their husbands’ and fathers’ workshop or farm. Of course it is a question for debate—which I will not enter into here—to what extent the various contributions of

non-market work market work housework subsistence work in own or family business work in someone else’s agriculture business or household

self-employed assisting employed family members unpaid unpaid wages / payment in profits unpaid wages kind

Figure 1. A basic taxonomy of work and labor relations in Western Europe, 1600–19007

6 Chris Tilly and Charles Tilly, Work under Capitalism (Boulder, 1998), p. 22. 7 This is a modified version of the scheme used by Pott-Buter, Facts and Fairy Tales, p. 7. Drawing from the more complex IISH-taxonomy of labor relations (), I have chosen to add subsistence labor, as this was very important in the regions and period under investigation. For parallel reasons, forms of forced and tributary labor have been omitted in this scheme. 8 See, for example: Jane Humphries an K.D.M. Snell, “Introduction,” in Penelope Lane, Neil Raven and K.D.M. Snell (eds), Women, Work and Wages in England, 1600–1850 (Wood- bridge, 2004), pp. 1–14, at 5–6. 326 elise van nederveen meerkerk family members were pooled, and how they were redistributed,9 but it is evident that there were large gender differences in the allocation, remuneration and valuation of work. Nevertheless, the unpaid or indirectly paid labor efforts of all family members were vital for the survival of families, or even for some extra room for consumption above subsistence level.10 Taking together women’s and children’s work by no means implies that I think they are always similar, or that there are no differences between boys’ and girls’ labor activities.11 On the contrary, I will try to show the similarities as well as the variance in work patterns between (adult) women and children, and between boys and girls. However, for the sake of a useful analysis of household labor, I do think it is valid to distinguish between the presumed “male breadwinner” on the one hand, and his “family members” on the other.

Women’s and children’s wage work in the pre-industrial and industrializing economy

Although wage work was probably—in a quantitative sense—the least important form of women’s and children’s labor around 1600, it became increasingly important in most of Western Europe in the period under study. This was part of a larger process of “proletarianization” of the workforce, which started long before industrialization. Not only did urbanization of a large part of England and the Northern Netherlands lead to increased specialization and a larger dependency on the market, but large parts of the countryside became ever more involved in market production as well. During this development of “proto-industrialization,” peasants and farmers began to perform non-agricultural activities such as spinning and weaving for the market in order to supplement their income. In many cases, they also became less and less independent producers, because larger entrepreneurs started employing a growing number of textile-

9 See, for example: Michiel Baud, “Families and Migration: Towards an Historical Analysis of Family Networks,” Economic and Social History of the Netherlands, 6 (1994), pp. 83–107, at 94–7. 10 De Vries, Industrious revolution. 11 In fact, in my earlier work I have tried to show that this was not the case. See, for example: Elise van Nederveen Meerkerk, “Couples Cooperating? Dutch Textile Workers, Family Labor and the “Industrious Revolution,” c. 1600–1800,” Continuity and Change, 23 (2008), pp. 237–66; idem and Ariadne Schmidt, “Between Wage Labor and Vocation: Child Labor in Dutch Urban Industry, 1600–1800,” Journal of Social History, 41, 3 (Spring 2008), pp. 717–36. the first “male breadwinner economy”? 327 producing peasant households for wages.12 Because both in cities and in rural areas the textile industry was of great importance, and because women and children were traditionally very much involved in this sector, the “commodification” of labor affected them at least as much as it did men. Moreover, they worked primarily in other low-skilled, poorly paid, and often casual and irregular occupations much more often than men, since they were for instance excluded from most guild-organized trades. The relatively greater wage-dependency of women and—to a lesser extent—children made them more vulnerable to the swings of the early modern economy.13 Moreover, in most of history, women’s and children’s work has been structurally undervalued. Varying over time and region, women’s wages constituted on average one-third to two-thirds of adult male wages.14 Chil- dren’s wages were often even lower.15 Various explanations have been given for the relatively low wages for women and children. First and foremost, physical differences attributed to men and women and to adults and children are put forward, most notably by neo-classical economic historians. Since—according to their theory—wage formation is based on demand and supply, wage differentials have to be explained from productivity differences, generally caused by women’s and children’s lesser body strength, and women’s responsibilities for the household and childcare.16 Other historians favour explanations of wage discrimination, either directly, or indirectly, so that women were excluded from several crafts and trades, and thus pushed them into “overcrowded” work activities such as

12 Sheilagh Ogilvie and Marcus Cerman, “The Theories of Proto-industrialization,” in Sheilagh Ogilvie and Marcus Cerman (eds), European Proto-industrialization (Cambridge, 1996), pp. 1–11, at p. 4. 13 Katrina Honeyman and Jordan Goodman, “Women’s Work, Gender Conflict, and Labour Markets in Europe, 1500–1900,” Economic History Review, 44, 4 (1991), pp. 608–28, at 610. 14 This was true both in agriculture and industry. See e.g. Joyce Burnette, “An Investiga- tion of the Female-Male Wage Gap during the Industrial Revolution in Britain,” Economic History Review, 50, 2 (1997), pp. 257–81; Deborah Simonton, A History of European Women’s Work, 1700 to the Present (London and New York, 1998), p. 45. 15 In the Leiden textile industry, they constituted 33 to 44 percent of adult wages. Van Nederveen Meerkerk, De draad in eigen handen. Vrouwen en loonarbeid in de Nederlandse textielnijverheid, 1581–1810 (Amsterdam, 2007), p. 291. 16 John Hatcher, “Debate. Women’s work Reconsidered: Gender and Wage Differen- tiation in Late Medieval England,” Past and Present, 173 (2001), pp. 191–8; Joyce Burnette, Gender, Work and Wages in Industrial Revolution Britain (Cambridge, 2008), pp. 12–3, 327. 328 elise van nederveen meerkerk hand-spinning.17 That aside, we should not overlook the possibility that women and children—or families—deliberately chose to allocate their labor in a certain fashion, for instance because they placed family responsibilities before profit maximization.18

Women’s and children’s wage work in the “first modern economy” One important characteristic of the premodern Dutch economy was that agriculture was highly specialized and commercialized, and that the share of people working in the primary sector was already low around the turn of the 16th century. De Vries and Van der Woude estimate that in 1500 around 555,000 people were working in agriculture (54% of the total population), 740,000 people in 1675 (38%), and around 1,833,000 people (40%), including women and children.19 Through the 19th century, this proportion declined to circa 30% in 1900. Moreover, wage labor was very important early on. Bas van Bavel has calculated for the provinces of Hol- land and Guelders that, in the 16th century, agricultural wage labor in the countryside accounted for respectively 48% and 57% of the rural population. He believes that the majority of wives and young children were not overwhelmingly engaged in wage labor in agriculture, but rather worked unpaid on their family farm. However, they probably did perform wage labor in rural proto-industries, such as spinning and weaving which, in the province of Holland, were already highly commodified around 1600.20 Although the Dutch Republic was renowned for its commerce and trade, export industries also became very important in the 17th century. Tex- tiles from Leiden and Haarlem, earthenware from Delft, clay pipes from Gouda, and many other products were sold all over Europe. Especially the growing textile industry offered ample employment to married women and children—the preparatory tasks for weaving were labor-intensive, and easily combinable with other tasks. As opposed to what many historians have believed, however, textile production by Dutch married women and children often did not constitute “ancillary tasks” for their husbands

17 See for a recent overview of the debate: Elise van Nederveen Meerkerk, “Market Wage or Discrimination? The Remuneration of Male and Female Wool Spinners in the Seventeenth-century Dutch Republic,” Economic History Review, 63, 1 (2010), pp. 165–86. 18 Pamela Sharpe, Adapting to Capitalism. Working Women in the English Economy, 1700–1850 (Basingstoke, 2000), p. 9. 19 De Vries and Van der Woude, First Modern Economy, pp. 229–33. 20 Bas van Bavel, “The Transition in the Low Countries: Wage Labour as an Indicator of the Rise of Capitalism in the Countryside, 1300–1700,” Past and Present, Supplement 2 (2007), pp. 286–303, at 300–2. the first “male breadwinner economy”? 329 and fathers, but independent wage work, both in the urban and the rural, more proto-industrial context.21 For several localities in the Northern Netherlands, I have estimated the input of women’s and children’s wage labor in the textile industry. Many thousands pairs of hands were required for providing the local textile industries (both linen and woollen) with enough spun yarn. Even if we account for imported yarn, this bottleneck in the production of cloth was fulfilled by the labor of married women and children, ranging from a small percentage in non-textile areas, to about 25% of all wives and children in some urban and proto-industrial textile regions in the 17th century.22 Among poor laboring families, married women’s earnings on average comprized 20% of family income, and once children grew older, their contributions totalled even more than that.23 In the 18th century, both the decline of the urban textile industries and proletarianization probably negatively influenced women’s and children’s wage work. As they generally worked in the worst-paid segments of the economy, the segmentation of the labor market became more pronounced. This could even be seen in the hand-spinning industry of the city of Leiden, where male spinners entered the highest and best paid segments.24 In the same century, however, wage labor became more important in the countryside. This is illustrated by the fact that in the proto-industrialized town of Tilburg, 38% of all married women around 1800 and almost 32% of all children under 18 are identified as wage workers in the wool industry.25 Given the importance of hand-spinning for women and children, we can imagine that the effects of mechanization of exactly this labor-intensive part of textile production in the Nether- lands from circa 1830 onwards must have been dramatic. In 1849, just over 1% of all women over 16 worked in the textile industry, and for children, numbers were even more negligible.26

21 Van Nederveen Meerkerk, Draad in eigen handen, p. 235. 22 Ibidem, pp. 93–4, 139–41. 23 Van Nederveen Meerkerk, “Couples cooperating?,” pp. 257–8. 24 Elise van Nederveen Meerkerk, “Segmentation in the Pre-industrial Labour Market: Women’s Work in the Dutch Textile Industry, 1581–1810,” International Review of Social History, 51, 2 (2006), pp. 189–216. 25 Regionaal Historisch Centrum Tilburg, Registers van volkstellingen van de gemeente Tilburg, 1810–1840, inv. nrs. 1275–1277, Registers van de Volkstelling 1810. 26 CBS, Uitkomsten volkstelling 1849. Unfortunately, this census does not distinguish between married and unmarried women. 330 elise van nederveen meerkerk

In 1899, the national census provided occupational data according to marital status. Of all women mentioning an occupation, 77% indicated they were unmarried, 13% were widowed or divorced, and only 10% were married. Figure 2 shows the division of married women’s labor in several sectors, according to the official statistics. These figures suggest that only 5% of all married women mentioned had an occupation; 3% indicated being self-employed, and just 2% as employed for wages. But as the introduction to the census states, due to the casual nature of women’s work, or even shame to report the fact that they were working, their share in wage labor was understated, and in reality undoubtedly larger.27 More- over, a lot of work by married women was done in the family business, but went unrecorded (this type of work is discussed in more detail below). As for children’s work, it appears that 118,983 Dutch children in 1899 were registered with an occupation; most of them in wage labor (5.9% of all children under 16). Elsewhere, I have shown that these were mostly children between 12 and 16 years of age, but as the census was taken after the first child labor law (which prohibited children under 12 to work for the market), we may assume that their share too, will have been higher than reported.28 What is clear from all this, is that the share of wage labor among married women and children in the Northern Netherlands was very high, especially in the urban and rural handicraft industries. With the rapid decline of hand-spinning and other small-scale manufactures in the 19th century, especially married women’s but also children’s opportunities in wage labor declined, although especially children in the age groups 12–16 found some alternatives in factory work.29

Wage work in England

In 1600, around 70 per cent of the English population depended on agricultural activities for their income, but this proportion dropped to around 25 percent in 1900. Such a shift in the occupational structure of course had large consequences for men’s, women’s and children’s work patterns. Simultaneously, the proportion of small, almost entirely self-supporting farms diminished during the course of the 17th and 18th century. Land

27 CBS, Inleiding beroepentelling 1899, pp. 185–9. 28 Elise van Nederveen Meerkerk, “Child Labor in Proto-industrial and Industrializing Netherlands,” in Hugh Hindman (ed.), The World of Child Labor: An Historical and Regional Survey (New York, 2009), pp. 625–8. 29 Ibidem, p. 627. the first “male breadwinner economy”? 331

100% 0 1256 90%

80% 19239 3484 (2%) 70% 6840

60%

50% 6152 605 1507 14099 Wage Work 40% Self-employed 30% 24127 4309 3% 20% 5114

10%

0% 0 0

ions Other 8,247) Agriculture Transport Manufacture

Domestic Service Free Profess (total = 83 Trade and All Married Women Figure 2. Married women’s work per sector in the official census statistics, Netherlands 1899 ownership became concentrated in the hand of fewer farmers, who increasingly practised large-scale farming for the market. This development not only led to impoverishment and a greater dependence of the rural population on wage labor, but also to shifting work allocation between the different family members.30 Most women in English agriculture worked as mothers or daughters in their own family farms. Female day laborers throughout the period were present, but always greatly outnumbered by men. Women were often hired on a seasonal basis—usually in the spring and early summer months—for tasks such as hay-making, weeding and stone picking. But day-laboring in agriculture declined faster among women than among men in the period 1750–1850.31

30 Louise Tilly and Joan Scott, Women, Work and Family (Second edition, New York and London, 1987), p. 14. 31 Sharpe, Adapting to Capitalism, pp. 73–8; Nicola Verdon, “A Diminishing Force? Reas- sessing the Employment of Female Day Labourers in English Agriculture, c. 1790–1850,” in Lane, Raven and Snell, Women, Work and Wages, pp. 190–211. 332 elise van nederveen meerkerk

Moreover, those women engaged in agricultural wage labor were usually unmarried—as a stage in their life cycle they worked for wages and/ or lodging on another farm. Many of them probably were young children, who in 1851 still consisted of about 9% of all wage workers in agriculture in England and Wales.32 Indeed, wage contributions to the family income in agricultural areas in the period 1790–1865 were much higher by children (ranging from average 7.4–21.9%) than by wives (0.5–11.6%).33 Despite this general picture, there were regional differences, varying from pasture to arable land, and high- and low-wage agricultural regions. In the far North for example, where there was a shortage of labor, more women were employed in agricultural wage labor. Here, even a system of “bondage” existed, in which male day laborers were required to bring a woman—usually his wife or daughter—who was available for work at a set wage.34 In the urban pre-industrial context, women probably more often worked for wages. For late 17th century London, Earle has shown that 60% of the married women testifying in court cases stated that they were wholly or partly involved in gainful work, about half of them (30% of all married women) being wage workers. Most of them were working in domestic service and laundry, textile production or making and mending clothes.35 Children too often worked for wages, either in a putting-out system, as “apprentices” in guilds, or via the widely spread parish apprenticeship system. Boys and girls worked in a variety of trades, often for low wages or paid in kind, because they were supposed to be learning skills. Habitually, they were employed in low-skilled productive work such as the textile industry, girls probably more so than boys, serving both as sources of cheap labor and as a way to relieve local poor administrations of some of their burden.36 In smaller towns too, for instance in Southern England, the textile industry employed about 40% of the population, a large share of which were married women and children. It is clear that industrialization in these areas greatly affected their work opportunities:

32 Peter Kirby, Child Labour in Britain, 1750–1870 (Basingstoke, 2003), p. 57. 33 Sara Horrell and Jane Humphries, “Women’s Labour Force Participation and the Transition to the Male Breadwinner Family, 1790–1865,” Economic History Review, 48, 1 (1995), pp. 89–117, at 102. 34 Burnette, Gender, Work and Wages, pp. 56–7. 35 Peter Earle, “The Female Labour Market in London in the Late Seventeenth and Early Eighteenth Centuries,” Economic History Review, 42, 3 (1989), pp. 328–53, at 337, 348–52. 36 Katrina Honeyman, Child Workers in England, 1780–1820. Parish Apprentices and the Making of the Early Industrial Labour Force (Aldershot, 2007). the first “male breadwinner economy”? 333 at the beginning of the 19th century, most of the small-scale handicraft industries had disappeared.37 Despite the loss of spinning, Horrell and Humphries have proposed that married women’s labor force participation rates experienced a short revival in the second quarter of the 19th century, only to decline again after 1850. They relate this to changes in demand rather than changes in supply—having work obviously remained necessary for the larges part of the population. In fact, women’s wage levels fell sharper than men’s and children’s after 1840, which in their opinion explains the withdrawal of (married) women from the wage labor market.38 Thus, industrialization had a differential effect on wives’ and children’s work. Both categories started to withdraw from the wage labor market, but it has to be noted that for married women in England, wage labor had already been less important in the early modern period. The number of children under 14 working in factories instead increased in the third quarter of the 19th century, and only declined again towards the end of the century. Neverthe- less, for children too, more informal forms of work, often paid in kind, remained more important than wage labor.39

Germany More than England and the Dutch Republic, Germany—or, perhaps more precisely, the German territories until 1871—formed a predominantly agrarian society until well into the 19th century, although there were of course variations between regions. In the countryside, mainly single women were employed for wages, in seasonal agricultural labor, grape- vines, or needlework, laundry, and spinning.40 The rise of textile manufacturing in rural areas (“proto-industry”) from the 17th century onwards unmistakably contributed to a higher involvement of wives and children in wage labor. However, because households were often contracted as a

37 Neil Raven, “A “Humbler, Industrious Class of Female”. Women’s Employment and Industry in the Small Towns of Southern England, c. 1790–1840,” in Lane, Raven and Snell, Women, Work and Wages, pp. 170–89, at 181–2. 38 Horrell and Humphries, “Women’s Labour Force Participation,” pp. 108, 113. 39 Kirby, Child Labour, pp. 75, 132. 40 Christina Vanja, “Zwischen Verdrängung und Expansion, Kontrolle und Befreiung— Frauenarbeit im 18. Jahrhundert im deutschsprachigen Raum,” Vierteljahrschrift für Sozial- und Wirtschaftsgeschichte, 79 (1992), pp. 457–82, at 467. 334 elise van nederveen meerkerk whole, many historians have stated that it is hard to establish concrete numbers of independent individuals involved in proto-industrial labor.41 In Germany, most of married women’s wage work activities in crafts and proto-industries were in milling, spinning and tailoring. Just as in the Dutch Republic, there are indications that wage work was often not done in the “family economy,” but that the wife, poor as she might be, independently sold her yarn in the market. However, there are also many examples of guilds and community courts trying to interfere in (married) women’s wage work, for instance by imposing piece-rate ceilings for spinning, or by interfering in employment relationships.42 This probably restricted Ger- man women’s independent (proto-industrial) wage work opportunities in rural areas. In towns, women also performed mostly low-paid tasks such as spinning and needlework, and their work identity and capacity to organize were generally low.43 If they were masters’ wives, women sometimes performed wage work, but more generally they worked alongside their husbands in their trade, or conducted an independent business.44 Arguably, married women’s possibilities to be active in the labor market were overall greater than those of unmarried women, because, “after carrying out their core domestic responsibilities, they did not have to crowd into “unregulated” [often wage labor—EvNM] sectors as unmarried women did, but rather could be active across the entire spectrum of the economy.”45 Indeed, many married women were involved in market- work, both in rural and urban areas, but most likely there were not a great many in wage-work activities. Let us now take a closer look at non-wage work activities.

41 Peter Kriedte, Hans Medick and Jürgen Schlumbohm, Industrialisation before Indus- trialization: Rural Industry in the Genesis of Capitalism (Cambridge, 1981), pp. 61–2; Rita Bake, Vorindustrielle Frauenerwerbsarbeit. Arbeits- und Lebensweise von Manufakturarbeit- erinnen im Deutschland des 18. Jahrhunderts unter besonderer Berücksichtiging Hamburgs (Cologne, 1984), p. 17. 42 Sheilagh Ogilvie, A Bitter Living. Women, Markets and Social Capital in Early Modern Germany (Cambridge, 2003), pp. 165–6. 43 Merry Wiesner, “Spinsters and Seamstresses: Women in Cloth and Clothing Produc- tion,” in M.W. Ferguson, M. Quilligan and N.J. Vickers (eds), Rewriting the Renaissance. The Discourses of Sexual Difference in Early Modern Europe (Chicago, 1986), pp. 191–205, at p. 203. 44 Christine Werkstetter, ‘. . . da ich meinem Vater Tochter, Gesell, Junge und handt- langer gewesen’. Arbeitsfelder, Ausbildung und ‘work identity “von Frauen im Augsburger Zunfthandwerk des 18. Jahrhunderts,” in Mark Häberlein and Christof Jeggle (eds), Vor- industrielles Gewerbe. Handwerkliche Produktion und Arbeitsbeziehungen in Mittelalter und früher Neuzeit (Konstanz, 2004), pp. 163–79, at 165. 45 Ogilvie, Bitter Living, pp. 145–6. the first “male breadwinner economy”? 335

Self-employment and working in the family business

The ideal of domesticity spread among the “middling sorts” earlier than among lower-class households, but only gradually and, in most regions, not until the 19th century. Wives and children of farmers, artisans and shopkeepers, generally worked in the family business. As Davidoff and Hall put it, for early modern women of the middle classes “marriage was indeed a “trade” and as economic actors they appear as shadows behind the scenes of the family enterprise”.46 Nevertheless, married women also quite often had a commercial enterprise of their own, apart from her husband’s trade. Both forms of work by middle class women and children will be discussed in this section.

A business of her own: female entrepreneurs

The visibility of women in public space, in commerce and in the market of the Dutch Republic was proverbial. But to what extent did it concern married women, and how does the image of the active Dutch businesswoman correspond with their actual involvement in trade, compared to other societies? Admittedly, the proportions of women among recorded “pre-industrial” businesspeople were rather high, ranging from 18.5% in Amsterdam, 30.5% in ’s-Hertogenbosch and 38% in Leiden around 1750.47 However, these figures concerned all widows and unmarried women who were listed as household heads in local censuses. Many more women, also married ones, conducted a business, but they are often obscured in the historical records. Sometimes, we can get a glimpse of the true situation when censuses mention two occupations for the male household head: combinations such as farmer-shopkeeper or carpenter-innkeeper suggest that apart from the husband’s occupation as a self-employed farmer or craftsman, somebody else—i.e. his wife—was responsible for another line of business.48 There are indications that, in the Netherlands, the opportunity for married women to have an independent business was larger than in (for instance) England or Germany, because the special legal status of femme sole trader in the Netherlands, which enlarged their legally

46 Leonore Davidoff and Catherine Hall, Family Fortunes. Men and Women of the Eng- lish Middle Class 1780–1850 (London, 1987), p. 273. 47 Based on: Danielle van den Heuvel, Women and Entrepreneurship. Female Traders in the Northern Netherlands, c. 1580–1815 (Amsterdam, 2007), pp. 283–6. 48 De Vries and Van der Woude, First Modern Economy, p. 601. 336 elise van nederveen meerkerk sanctioned roles, applied to all married businesswomen, and not only to specific branches.49 Another indication for the activity of married female traders can be found by “zooming in” on a specific retailing business: the tea and coffee trade. In the course of the 18th century, this business became increasingly popular, because tea and coffee were increasingly consumed by people from most social classes. Interestingly, this trade not only became feminized during the 18th century, it also became increasingly dominated by married women. Of all registered tea and coffee dealers in the 18th century, more than 37% were married women. Their husbands appear to have been active mostly in the lower echelons of guilded crafts (46%) and wage workers in the textile industry (30%). Most of these women selling tea and coffee had a number of small children in their care during their economically active years.50 Presumably, especially for married women from the lower middling classes, being an independent businesswoman was of crucial importance. In England too, retailing and victualling were important for married women from the lower and middle classes. Peter Earle found that almost 19% of the 417 London wives in his sample worked in retailing around 1700, and for the 18th century, these percentages may have been higher.51 But not only women from the lower classes were active in business. There are also many examples of women from the higher social strata active as businesswomen and (financial) brokers.52 For Germany, there are instead indications that married women retreated from large-scale business in the 18th century, but retailing and hawking remained very important to them until the beginning of the 20th century.53 In the Netherlands too, self-employment of married women in the retailing sector remained the most important in the 19th century as well. Whereas their share as independent producers in agriculture and manufacturing was small (respectively 0.6% and 0.5% of all married women in

49 Van den Heuvel, Women and Entrepreneurship, p. 59. 50 Danielle van den Heuvel and Elise van Nederveen Meerkerk, “Huishoudens, werk en consumptieveranderingen in vroegmodern Holland. Het voorbeeld van de koffie- en theeverkopers in 18de-eeuws Leiden,” Holland. Historisch Tijdschrift, 42 (2010), pp. 102–24. 51 Earle, “The Female Labour Market,” p. 339; Amy Erickson, “Married Women’s Occu- pations in Eighteenth-century London,” Continuity and Change, 23 (2008), pp. 267–307, at 296–8: 27% of all married women in retailing, but Erickson only looks at married women stating an occupation. 52 See e.g. Nicola Phillips, Women in Business, 1700–1850 (Woodbridge, 2006); Hunt, The Middling Sort, p. 129. 53 Vanja, “Zwischen Verdrängung und Expansion,” pp. 464–5. the first “male breadwinner economy”? 337

Table 1. Self-employed women according to marital status and sector, Netherlands 1899 Widows % Married % Unmarried % Total % of all women over 16 Agriculture and 14,124 67.3 5,114 24.4 1,756 8.4 20,994 1.3 fisheries Manufacturing 6,954 19.6 4,309 12.2 24,131 68.2 35,394 2.1 Trade and 15,816 43.6 14,099 38.8 6,393 17.6 36,308 2.2 services Total 36,894 39.8 23,522 25.4 32,280 34.8 92,696 5.6

1899), almost 1.7% of all married women had a small business, a figure approaching the official share of married women working for wages in the same year.54

Working in the family business in England, the Netherlands and Germany

Despite regional differences, women’s and children’s work in general tended to be more seasonal than men’s, and they were employed for labor-intensive tasks such as haymaking, planting, spreading manure, weeding and threshing. Both in England and the Netherlands, women’s and children’s work on the farm were much more common in the arable areas than in pastoral farming, except for dairy-making.55 Dairying was often associated with women and girls. Both at small-scale farms and at larger commercial dairy farms, women and girls remained important until the arrival of mechanization of dairying in the late 19th century.56 It is highly likely that for married women and children in England and the Northern Netherlands, agricultural activities of these kinds were more

54 Jacques van Gerwen, “The Netherlands 1900 and 2000,” 7; CBS, Inleiding, pp. 190–4; CBS, Uitkomsten voor het geheele Rijk. 55 Harriet Bradley, Men’s Work, Women’s Work. A Sociological History of the Sexual Divi- sion of Labour in Employment (Cambridge, 1989), p. 82; Willemien Schenkeveld, “Het werk van kinderen in de Nederlandse landbouw, 1800–1913,” Tijdschrift voor Sociale en Economis- che Geschiedenis, 5, 2 (2008), pp. 28–54. 56 Simonton, European Women’s Work, pp. 30–1; Jan Luiten van Zanden, De economische ontwikkeling van de Nederlandse landbouw in de negentiende eeuw, 1800–1914 (Wagen- ingen and Utrecht, 1985), p. 74. 338 elise van nederveen meerkerk often performed for the market than was the case in Germany, where a much smaller percentage of women’s and children’s work in agriculture was commercially oriented.57 At the end of the 18th century, when agriculture in Germany intensified and both potato-growing and animal husbandry spread, German men started working for wages more often, and the involvement of German women and children in the family farm increased.58 Despite great regional variations in the character of agriculture in the Northern Netherlands, the tendency from the 17th century to circa 1850 was scaling-up. De Vries and Van der Woude estimate that, in 1500, around 555,000 people were working in agriculture (54% of the total population), compared to 740,000 people in 1675 (38%), and 833,000 people around 1800 (40%), including women and children. Although for their purposes “it is not necessary to estimate the precise labor force participation rates of women and children,”59 I will make the effort here. Bas van Bavel has calculated for the provinces of Holland and Guelders that in the 16th century, non-wage labor in agriculture accounted for respectively 68% and 42% of the population. In the second half of the 17th century, the percentage in Holland would have dropped to 40%, remaining fairly constant thereafter.60 Presumably, these percentages are lower than elsewhere in the Northern Netherlands, but if we want to calculate minimum numbers of assisting family members, we are probably on safe ground if we stick to 65% of non-wage labor for the first period, 45% for the two following periods, and 35% in 1900.61 If we assume that 50% of those working in non-wage labor in agriculture were wives and children (subtracting male and female household heads and live-in servants) and that subsistence labor became less important (20% in 1500, 10% in 1675, and 1800, 5% in 1900), we can roughly reconstruct the importance of assisting family labor for the market (see Table 2). Between 1800 and 1910, the number of agricultural enterprises in the Netherlands rose from approximately 73,500 to circa 118,900 (an increase of almost 62%). From the second half of the 19th century, specialization and mechanization led to a declining share of agriculture in the total labor force, as well as decreasing work opportunities in agricultural wage

57 Ogilvie, A Bitter Living, pp. 116–7, 141. 58 Vanja, “Zwischen Verdrängung und Expansion,” pp. 460–2. 59 De Vries and Van der Woude, First Modern Economy, pp. 229–33, citation on p. 229. 60 Van Bavel, “Transition in the Low Countries,” pp. 300–2. 61 Van Zanden, De Nederlandse landbouw, p. 332. the first “male breadwinner economy”? 339

Table 2. Rough estimates of assisting labor on family farms, the Netherlands 1500–1900 Agrarian In non- of whom Minus subsistence Assisting family population wage wives and by women and members in labor children children agriculture c. 1510 555,000 360,750 180,375 36,075 144,300 (15% of total population) c. 1675 740,000 330,000 165,000 16,500 148,500 (7,8% of total population) c. 1800 833,000 374,850 187,425 18,743 168,682 (8,0% of total population) c. 1900 790,27862 402,127 231,579 11,579 220,000 (4,3% of total population) labor. According to Van Zanden, these developments—in combination with increasing living standards and compulsory schooling—led to a retreat of married women and children under 16 from the agricultural wage labor market, and resulted in a higher work effort of especially married women in their own family farms (52% of all women in agriculture in 1810, compared to 67.5% in 1910). Moreover, while in an absolute sense the number of (married) women and children in agriculture rose, their relative share dropped even faster than the percentage of adult males in agriculture.63 In the urban environment, women and children also worked in their husbands’ and fathers’ crafts and trades. Using court records from 17th century to 19th-century London, both Peter Earle and Amy Erickson estimated proportions of 10–17% of wives working in a trade similar to that of their husbands, although Erickson suggests that, in fact, women of the “middling sorts” were probably working together with their husbands

62 This number is higher than in the official CBS-statistics, because I have worked the other way around for 1900. I have reconstructed the agrarian labor force beginning with Van Zanden’s estimates on the input of family labor (c. 110,000 Dutch farms on average requiring the labour of 1 wife and 1 child), added 5% subsistence labor, then added to this the self-employed in agriculture, and eventually adding all wage laborers. This amounts to slightly over 790,000 people working in agriculture. 63 Van Zanden, De Nederlandse landbouw, pp. 74–8. 340 elise van nederveen meerkerk more often.64 She assumes that at least 15 percent of men in the London livery companies worked together with their wives in a “skilled capacity,” and that even many more wives worked in a related or unrelated entrepreneurship apart from their husbands, as we have seen above.65 In Dutch towns as well, wives appear to have been working in their husbands’ crafts or businesses, although it is difficult to estimate their numbers, because they are often obscured in the historical records. One indication for women’s participation in guilds is the fact that, although they could not become guild members when they were married, they did so from time to time as widows. In the city of Gouda, for instance, half of the total of 28 guilds contained female members, ranging from 1% (Shoemakers’ & tanners’ guild) to 84% (Tailors’ guild). On average—when excluding the tailors’ guild—16% of all members of guilds with female membership were widows. These widows worked in a wide array of occupations, ranging from weaving and rope spinning, tasks traditionally associated with women, to wood-working, brick-laying, and pipe-making.66 This implies that, during their marriage, many women worked alongside and learned skills from their husbands. In the city of Leiden too, we find 18th century evidence that weavers taught “their sons, wives or daughters” the trade, and they were even explicitly exempted from paying an apprenticeship fee.67 In German towns, the share of women working in guild crafts seems to have been lower, which Ogilvie believes to be caused by the more restrictive role guilds played regarding women’s work in Germany. Nevertheless, she states that a married woman was “the only female permitted to work at all aspects of craft [. . .] if her husband held the requisite guild licence”.68 This implies that in reality, although much of married women’s work in guilds is not made explicit in historical records, it was important in Ger- many as well (perhaps even more so than in the Dutch Republic, since women’s formal role was more limited in Germany). Nevertheless, there are indications that in all of the three regions studied, the formal exclusion of women existed, and started during childhood. Although there were

64 Earle, “Female Labour Market,” p. 338; Erickson, “Married Women’s Occupations,” p. 276. 65 Erickson, “Married Women’s Occupations,” p. 294. 66 Schmidt, “Women and Guilds,” p. 179. 67 Regionaal Archief Leiden, Archief Hallen, inv. no. 221, 5–9–1767. See also: Van Ned- erveen Meerkerk, “Couples cooperating?,” p. 246. 68 Ogilvie, Bitter Living, pp. 142, 330. the first “male breadwinner economy”? 341 exceptions, girls were usually not permitted in formal (guild) apprentice- ships, which suggests that, more than boys, they were likely rather to work unpaid in their parents’ workshop and learn “on the shop floor.”69 Married women however not only assisted their husbands in craft industries; they were very active in commercial businesses as well. Many married women worked in the family business, alongside their husbands, often conducting transactions in their husbands’ name.70 This was not only the case in the Dutch Republic, but also in England and Germany, where married women often worked in the trade of their husbands or had an independent business.71 It is difficult to establish how many wives helped their husbands in their business, but we do know that in many trades, the fact that women had reproductive and housework duties did not prevent them from being involved in a business, even though married women might be formally restricted in doing so. In Dutch food markets, for instance, butchers’ wives were not allowed to permanently assist their husbands in the meat hall, but they often sold the residues of the meat in the offal hall.72 Amy Erickson estimates that 35–59% of couples from the “middling sort” in 18th-century London shared a business, and that their relative “prosperity was probably built on the very fact of their running a joint establishment or complementary enterprises”.73 In the 19th century, it is likely that the share of family members helping their husbands and fathers in manufacture diminished, because of the decline of traditional crafts and industries. Nevertheless, this was a gradual process, and differed tremendously per branch. Even in industrializing England, many children continued to work in their fathers’ workshops in urban settings. Sometimes, they were even hired together with their parents in the factories.74 Also, commercial family enterprises did not yet change remarkably during the 19th century. If we assume that in the Netherlands some 35% of married men in manufacturing were generally assisted by their wives and by at least one of their children, and

69 Van Nederveen Meerkerk, Draad in eigen handen, p. 275; Honeyman, Child Workers in England, pp. 151–2; Ogilvie, Bitter Living, pp. 98–9. 70 Van den Heuvel, Women and Entrepreneurship, p. 67. 71 Hunt, Middling Sort, p. 126; Ogilvie, Bitter Living, p. 167. 72 For instance, because they could not obtain membership of the retail guild in their own name. Danielle van den Heuvel, “Partners in Marriage and Business? Guilds and the Family Economy in Urban Food Markets in the Dutch Republic,” Continuity and Change, 23 (2008), pp. 217–36, at 221–3. 73 Erickson, “Married Women’s Occupations,” pp. 284–5. 74 Kirby, Child Labour, pp. 72–3. 342 elise van nederveen meerkerk that this was the case in 50% of all commercial enterprises, we estimate the number of assisting family members in small-scale manufacturing and business around 1900. Of all 109,299 self-employed married men in manufacturing, there would be 38,255 wives and as many children assisting him. Of all 68,973 married men in retailing and commerce, there would be 34,487 wives and as many children assisting him.75

Housework and subsistence labor

Throughout history, the work performed—usually by women—in the home, such as washing, cooking, mending, cleaning, child minding, etc. tends not to be seen as “economic activity,” but rather as “reproductive labor,” in order to sustain the (male) productive wage laborer.76 As Figure 1 above shows, however, I explicitly choose to consider both housework and the unpaid labor performed by women and children for subsistence, because they are, as physical activities adding value to goods and services, the same as any other forms of (paid or unpaid) work. Despite the fact that a large percentage of women in Europe never married, the majority of them did, and many of them devoted a considerable part of their time to housewifery. Both in urban and (especially) in rural regions, “housewifery” was generally much more broadly defined than in our present- day notion of “housework.” Among other things, apart from cleaning, food preparation and minding children, it comprized spinning, dairying, poul- try farming and many other activities.77 The wide array of female tasks in fact implied that cleaning, mending, cooking and childcare took quite a modest proportion of women’s time.78 Moreover, not only wives, but children—and especially girls—also performed a wide range of domestic tasks, either to assist their mothers or to replace them, if they worked outside the home, or had died.79 Because the lines between housework and subsistence labor of women and children were often blurred, and because they represent the two

75 CBS, Beroepentelling 1899. 76 Edward Higgs, “Women, Occupations and Work in the Nineteenth Century Cen- suses,” History Workshop, 23 (Spring 1987), pp. 59–80, at p. 60. 77 Jane Whittle, “Housewives and Servants in Rural England, 1440–1650: Evidence of Women’s Work from Probate Documents,” Transactions of the Royal Historical Society, 15 (2005), pp. 51–74, at p. 52. 78 Simonton, European Women’s Work, p. 20. 79 Kirby, Child Labour, p. 10. the first “male breadwinner economy”? 343 economic activities about which we are probably least informed as regards the epoch from the 17th through to the 19th century, I will treat them together in this section. For the sake of analysis, developments in England, the Northern Netherlands and Germany will be described separately. It is important to note that within these countries too, various different patterns existed, depending on the structure of the local economy, the availability of alternative work for women and children, and the standard of living.

An early ideal of bourgeois domesticity? The case of the Northern Netherlands

Many historians have suggested that the ideal of domesticity pervaded social strata in the Dutch Republic already from the 17th century. How- ever, as was shown above, Dutch women from all social echelons, and from all marital statuses, also were remarkably economically active. Also, child labor was prevalent in most social strata in the early capitalist Dutch economy. How can these findings be squared with the cult of domesticity that both historians and contemporaries have commented upon? The Dutch Republic, and certainly the coastal provinces, was highly urbanized and commercialized. Many foreigners visited the towns of the Republic and were amazed by its prosperity, its business and cleanliness, and—especially—by the influence of Dutch women on these characteristics. Dutch women were, according to many travellers’ accounts, not only very independent and visible in business and public life, but they were also presumed to be dominating the household—hence their obsession with keeping their houses and sidewalks remarkably clean. Of course, these are stereotypes, and it is very difficult to substantiate claims of these accounts with historical evidence. Also, it is very difficult to find out what Dutch housewives actually did, and, particularly, how much time they devoted to “housework.” Prescriptive literature from the 17th century tells us that their main tasks were taking care of the meals and cleaning of the house.80 The main duties of urban women can also be derived from a glance at the more “professional” mothers: the binnenmoeders (orphan mothers) of urban orphanages. “Mothers” and other women working in the orphanage

80 Kloek, Vrouw des huizes, pp. 85, 105. 344 elise van nederveen meerkerk were held responsible—and increasingly so—for household management such as cooking and cleaning, and the care for small children. Education and work outside the orphanage increasingly became the responsibilities of the male regents and binnenvaders (orphan fathers) in the course of the 17th and 18th centuries. Thus, the changing norms and ideologies about the role of fathers and mothers, at least became more pronounced in these institutionalized “substitute” families—the orphanages—during the early modern period.81 To what extent this sharper division of tasks was also implemented in the private family sphere, remains unclear. Further down the social ladder, women and children undertook a whole range of activities in order to survive, but it is clear that even for the poorest families, it was accepted that married women in the first place took care of small children and/or sick family members. Although the poor relief authorities generally expected women and children to bring in some income, mothers and wives were never pressured into earning money if their time was consumed with caring for household members. The small dwellings occupied by poor households—sometimes no more than a basement in another family’s house—and the relatively simple items found in their inventories, suggest that cleaning and cooking did not take up most of women’s time. Because children from a certain age were generally able to earn more than their mothers, it is unlikely that they (even the girls among them) very often substituted their mothers in their housework duties. More likely, they were engaged in wage labor, as we have seen.82 Not much is known about the household and subsistence work of women and children in agriculture in the early modern period, due to a paucity of historical records. Although there were large regional differences, between 1600 and 1900, one thing stands out very clearly: the number of self-sufficient farmers was very small, and the scaling up of farming and the pervasion of wage labor increased markedly. Because agriculture was so commercialized, and much of the grain for consumption was imported from the Baltic, women’s and children’s work in subsistence agriculture was probably much less prevalent. Much of their work efforts would have been market-oriented.83

81 Ariadne Schmidt, “Managing a Large Household. The Gender Division of Work in Orphanages in Dutch Towns in the Early Modern Period, 1580–1800,” History of the Family, 13 (2008), pp. 42–57. 82 Hilde van Wijngaarden, Zorg voor de kost. Armenzorg, arbeid en onderlinge hulp in Zwolle, 1650–1700 (Amsterdam, 2000), pp. 190–1, 263–4. 83 De Vries and Van der Woude, The First Modern Economy, pp. 195–6, 543–61. the first “male breadwinner economy”? 345

In the coastal provinces of Zealand and Groningen during the 19th century, married women fully retreated from work in the family farm only in the richer farmers’ households. In other provinces, employment of married women probably increased during the period, due to intensification of agriculture, and a relatively fast growth of the number of farms.84 But again, most of this employment should not be labeled as “subsistence work,” but rather as work for the market, considering the high degree of commercialization of Dutch farming in this period. Remarkably, the first Dutch law against child labor (1874) did not even cover agricultural work, and most parents did not consider the unpaid work which children did on their own farm to be impermissible, but instead regarded it as “natural.”85

“Good housewifery” in the industrializing English economy

Though husbandry semeth, to bring in the gains, yet huswifery labors, seeme equall in paines (Thomas Tusser, 1573).86 At the beginning of the 17th century, English peasants’ and farmers’ wives were supposed to be able to cook, bake, wash, clean the house, and over- see the servants in the case of a larger household. Depending on the type of farm and region, other regular tasks were milking, brewing, making hay, harvesting grain, picking fruit and spinning. In other words, the farmer’s housewife’s duties were a combination of a daily routine and seasonal activities. Although these tasks were to a large extent self-supporting, the lines between work for subsistence and the market were blurred: not infrequently women sold the surpluses of their dairy products, eggs, and fruit and vegetables grown in their garden or orchard in the marketplace. Sometimes, this practice was even bestowed on rural wives at marriage, referred to as “pin-money,” which they could use to buy some extras for the household.87 Peasant children too became accustomed to working on their parents’ plot of land from a very young age: picking stones, weeding and planting crops. From early on, there was a clear division of work according to sex: boys less than 10 years of age started as bird scarers or

84 Van Zanden, De Nederlandse landbouw, pp. 74–5. 85 Schenkeveld, “Het werk van kinderen,” pp. 33, 38. 86 As cited in Whittle, “Housewives and Servants,” p. 51. 87 Whittle, “Housewives and Servants,” pp. 62–4, 69–70; Simonton, European Women’s Work, p. 20. 346 elise van nederveen meerkerk shepherds, whereas girls generally took on more domestic tasks similar to their mothers’.88 For the wives and children of small peasants and agricultural laborers—categories becoming more predominant in the 17th and 18th century—perspectives were somewhat different. First of all, in the absence of a considerable area of their own arable land, most of them performed part time wage work in agriculture on other people’s farms, which obviously left them less time for housework.89 Apart from the time constraints, their houses were usually quite small and simple, thus requiring less attention. In order to survive, women and children performed all kinds of activities, varying from spinning or gleaning (the gathering of leftover wheat and fruits on arable fields, in some counties a communal right for the landless and poor) to begging.90 The above-mentioned developments of commercialization and mechanization in the countryside, both in agricultural work and in hand- spinning, greatly affected women’s and children’s work. Where English women had previously been involved in a variety of agricultural tasks, in the course of the 18th century an increasing number of them “undertook the back-breaking task of gleaning for several weeks from morning till night because their work yielded a vital piece in the jigsaw of makeshifts that kept their household economies afloat”.91 This greater importance of gleaning—on average consisting of some 6 to 10% of family incomes— was probably highly correlated with the disappearance of hand-spinning in the countryside, as spinning was the first occupation to become mechanized after 1750. Like rural women, women in cities were considered responsible for the direct needs of the households—cleaning, mending, cooking and childcare. Although urban women were definitely less involved in subsistence labor and more dependent on the market for products of daily consumption, the processing of items bought did take some of their time. Apart from a greater involvement in market consumption, her work activities besides housework were probably also more in the sphere of market work, either for wages, or in her husband’s workshop or business.92 In the more

88 Kirby, Child Labour, p. 56. 89 Simonton, European Women’s Work, p. 29. 90 Peter King, “Customary Rights and Women’s Earnings: The Importance of Gleaning to the Rural Labouring Poor, 1750–1850,” Economic History Review, 54, 3 (1991), pp. 461–76. 91 King, “Customary rights,” p. 474. 92 Simonton, European Women’s Work, p. 21. the first “male breadwinner economy”? 347 plebeian households, both in the early modern period and the factory workers’ households in the 19th century, probably less time was spent on domestic chores. Dwellings were small and diets were simple, so housework and food preparation took relatively little effort, at least until the 1850s.93 Child minding was of course an important task for poorer households as well, and this was often left to the older girls in the house, apart from the other household tasks they performed.94 Much prescriptive literature from the 17th century points to the importance of men bringing in the money, and wives performing duties in the home. Still, a few contemporary authors also acknowledged that a wife should bring in income of her own.95 From the late 18th century onward, however, the ideal of bourgeois domesticity was more actively propagated by social movements such as the Clapham Sect. In the decades around 1800, the bourgeois ideal of the family, with married women staying at home, became the dominant culture. Even if many working-class families still dearly needed the income of wives and children, from circa the 1830s onwards, the ideal was promoted for working-class households too, and if married women entered into employment, they ought not to be seen.96 Most probably, the pervasion of this ideology went hand in hand with declining possibilities for women to work outside the home, and it is not always clear which was cause and which was effect.97

Germany: from Hausmutter to Hausfrau

In agrarian Germany too, women’s perceived role in society changed in the late 18th, early 19th century. Interestingly, in the 17th century the general custom was still that the Hausmutter, the mother of the house, together with the Hausvater (father of the house) shared the responsibility for the household (das ganze Haus), and thus the farm. Her duties concerned (the supervision of ) work in the kitchen and cellar, tending small animals, cleaning and production of clothes, spinning, weaving and other work with wool and flax. Around the turn of the 19th century, however, the Hausmutter increasingly came to be described as Hausfrau

93 De Vries, Industrious Revolution, pp. 192–3. 94 Kirby, Child Labour, p. 80. 95 Whittle, “Housewives and Servants,” p. 64. 96 Catherine Hall, “The Early Formation of Victorian Domestic Ideology,” in Sandra Bur- man (ed.), Fit work for women (Canberra, 1979), pp. 15–32, at 30–1. 97 Simonton, European Women’s Work, pp. 18–9. 348 elise van nederveen meerkerk

(housewife). Like in England and the Netherlands, the ideal of domesticity and a male breadwinner model clearly became more pervasive in this period.98 For the county of Württemburg, in Southern Germany, Sheilagh Ogilvie has calculated which percentages of rural and urban married women were involved in all kinds of work, including housework and subsistence labor (see Table 3). When we separately identify these two types of work, it becomes clear that, indeed, housework and care for relatives was “only” reported as their primary task by around 25% of all women, both in towns and in villages. Of course, subsistence agriculture was performed by a larger percentage of the rural women, but nevertheless 12% of the urban women indicated that they were involved in some kind of non-market agricultural activities, mostly tending small animals and growing fruits and vegetables. Although these percentages are based on relatively few cases, they seem to be quite representative, compared to what we know from regions outside Würt- temburg.100 Ogilvie’s figures thus lead us to conclude that housework pur sang was relatively unimportant to married women in Germany, too. Sub- sistence agriculture was overall of more importance, and even played a role in the urban setting.

Table 3. Married women’s and children’s housework and subsistence activities according to court records in Würrtemburg, c. 1650–180099 Married Married All married Daughters Sons women rural women urban women (n = 136) (n = 212) (n = 210) (n = 187) (n = 397) Non-commercial 22% 12% 17% 22% 40% agriculture Housework (indoor) 15% 15% 15% 4% 0% Unpaid care 7% 12% 9% 10% 1% Gathering 6% 3% 4% 13% 10% Stealing 1% 1% 1% 0% 2% Total 51% 35% 46% 49% 53%

98 Ibidem, pp. 19–20. 99 Based on: Ogilvie, Bitter Living, pp. 116–7, Table 3.10 and pp. 142–3, Table 4.1. 100 Ogilvie, Bitter Living, p. 143. the first “male breadwinner economy”? 349

Even more than their mothers, the work of offspring was characterized by either household or subsistence work: 49% of the girls’ and 53% the boys’ work was described as such. Especially for boys this meant working on their parents’ farm. But for girls too, though they more often helped in the house or with (child) care, subsistence agriculture was the most important task. For both boys and girls, the gathering of wheat, fruits and other products came second. Although I have excluded forced labor from the range of labor relations in Figure 1, in some parts of Germany agriculture still relied on feudal services, and it was not uncommon that daughters were sent out as servants to pay for their parents’ duties, or that peasant wives were required to work for 20–25 days a year as labor rent.101 Another distinct feature of German society was migratory labor. After the end of a period of turmoil and wars, the population started to grow quickly in the second half of the 17th century. Because of impartial inheritance rights, the number of small peasants and landless agricultural laborers looking for work increased. Many of them found—seasonal or permanent—work in the Dutch Republic, where wages were relatively higher than in the bordering German lands.102 This resulted in a tradition of migration which continued well into the 19th century. Often, seasonal migrants left their families at home, and their wives and children were responsible for working their own plot of land for part of the year.103

Conclusions

The work of wives and children is often extremely difficult to trace in the historical records. Even where we do have national statistics available for the past, they tend to under-record the participation of women and children. Partly, this has to do with the reluctance of the actors themselves to report their work activities, but more often, their work was simply not regarded “gainful employment” for statistical purposes, because it was seasonal, performed within the family business, or not paid in wages. Estimating the participation of married women and children is a very uncertain task, and it is difficult to make comparisons across time and

101 Simonton, European Women’s Work, p. 29. 102 Jan Lucassen, Naar de kusten van de Noordzee. Trekarbeid in Europees perspektief, 1600–1900 (Gouda, 1984), pp. 168–71. 103 Ibidem, pp. 123–4. 350 elise van nederveen meerkerk region, especially since much of the literature on women’s and children’s work (most notably before 1800) is very descriptive and sketchy, and does not enable much quantification. However, some general trends can be discerned in work patterns and labor relations of married women and children from the 17th century through to the 19th century in the Nether- lands, England and Germany. A first conclusion is, that there are probably more similarities than differences in the three regions under study here. Throughout the period, both married women and children were very much actively involved in all types of work and labor relations. Wage labor among them appears to have been the least important form of work during the period, but the developments in the share of female and child wage laborers were far from linear. Most probably, fluctuations in demand influenced women’s and children’s work, which was often performed in trend-sensitive, low-paid, and labor- intensive branches, more dramatically than men’s. In the Netherlands, the favorable economic trend in the 17th century, which boosted a number of urban export industries, increased the share of wage laborers—in some periods and regions up to 40% of all people not heading a household! In England, industrialization in some parts had similar effects, although the fast decline of hand manufactured textiles also caused great under- employment in many regions dependent on rural industry. In Germany, proto-industry particularly led to the rise of female and child workers for the market, although the sources make insufficiently clear whether this was done as wage labor or family labor. In Germany, more than in England and the Netherlands, women and children were working in agriculture throughout the period studied—but across the board not so much in wage labor as on the family farm, either for subsistence or for the market. As regards subsistence agriculture, we may conclude that this was already of little importance in the Netherlands around 1600, became less important in England in the 17th century, and started to decline in Germany in the 18th century. The division of tasks varied per type of agriculture, but generally, women and children were most involved in weeding, picking stones and haymaking, dairying and taking care of the garden. From age 12, the gender division of labor became more pronounced between boys and girls. In the 18th century, agricultural changes, mechanization (of spinning) and proletarianization sometimes led to more work for women and children in subsistence agriculture (e.g. gleaning in England), whereas in Germany their share in assisting labor for the market may have increased. the first “male breadwinner economy”? 351

“Assisting labor,” in the family farm, in the craftsman’s workshop, and in the store or market stall, was probably the most typical form of wives and children’s work in the entire period. Although on a micro-level we can sometimes make statements about this type of work, the actual range and importance of assisting labor very much depend on educated guesses. Especially in small-scale agriculture and retailing, these activities were important, and it is not surprising that the self-employed today still rely on the input of two marriage partners and (even young) children. In the end, it is very difficult to distinguish between assisting labor, subsisting labor and housework, because many of the tasks were overlapping and/or performed by the same persons year round. We can however safely conclude that the percentage of married women only performing housework was very low from the 17th century up to the middle of the 19th century. We have seen that in all three countries under investigation, the actual generalization of the male breadwinner ideology among all social strata was accomplished after 1850. In the 17th century, the share of married women and children in wage labor in the Dutch Republic was not lower, and perhaps even higher, than in other countries, due to the flourishing export industries in the cities. Consequently, in the 18th century, opportunities in retailing and trade expanded greatly. Why, then, this stubborn image of the “Dutch domestic housewife”? It is probably a relic of more recent times. Indeed, in the second half of the 19th century, the official share of working women was relatively low, for instance compared to England and Germany. Even if we try to adjust the official statistics of the 1899 census with estimates on assisting labor and unrecorded wage work, by using the minimally estimated figures given in the various section of this paper, over 60% of the married women do not seem to have been economically active around 1900, although this would be the absolute maximum (see Table 4). All in all, it is evident that the bourgeois ideal of domesticity did not gain ground among the lower social classes in the Dutch Republic much earlier than, for instance, in England or Germany. In all these countries, the cult of domesticity only pervaded the lower classes in the second half of the 19th century. Perhaps the level of subsistence labor in agriculture by Dutch women was lower than in England, due to the already highly commercialized and specialized agriculture in the Netherlands around 1500. But women and children were highly involved in the family farm. And in cities as well, their economic activities were largely directed towards the market, albeit often performed as “assisting labor.” Although 352 elise van nederveen meerkerk

Table 4. Estimates of married women and children in various types of labor relations, the Netherlands 1899 Labor relation Married women Children under 16 Number % of total Number % of total (n = 838,247) (n = 2,009,700) Affluent** 8,382 1.0% 20,097 1.0% Too old/too young* 21,192 2.5% 337,630 16.8% Subsistence agriculture** 41,912 5.0% 100,485 5.0% Assisting work in family 187,790 22.4% 187,790 9.4% business** Agriculture 110,000 13.1% 110,000 5.5% Manufacture 43,303 5.2% 43,303 2.2% Trade & services 34,487 4.1% 34,487 1.7% Self employed 23,522 2.8% 0.0% Agriculture 5,114 0.6% 0.0% Manufacture 4,309 0.5% 0.0% Trade & services 14,099 1.7% 0.0% Employed 39,844 4.7% 158,983 7.9% Free professions 605 0.0% 0.0% Wage work (officially 19,239 2.3% 118,983 5.9% recorded) Wage work (unrecorded)** 20,000 2.4% 40,000 2.0% Only housework (?) 515,605 61.5% * Married women over 71 years, and children under 5 years of age ** Estimates contemporary travelers who visited the Northern Netherlands were amazed about the clean houses and streets, and many historians have continued this image of the early bourgeois ideal of domesticity, recent research has shown that women of all social echelons were active agents in the labor market, either assisting their husbands, or as self-employed or wage working women. The period of the male breadwinner model, with the housewife at home, and the children at school, was thus a relatively short episode in Dutch history, although Marga Klompé in her days could probably not have foreseen its rapid decline when she started her political career, some 65 years ago. Wage Labor and the Household Economy: a Russian Perspective, 1600–2000

Gijs Kessler

Drawing on examples from Russian labor history of the last four centuries, I want to engage with a subject central to the work of Jan Lucassen: the complementarity of wage labor and other economic activities within the household economy of the early modern and pre-industrial periods. Lucas- sen’s work focused primarily on North-Western and Southern Europe, but his insights into the place of wage labor in pre-industrial household economies are very pertinent for the Russian experience as well. Indeed, Russian history is a real treasure trove of examples illustrating the very mechanisms and patterns Lucassen identified in his work on labor migration in the North Sea Basin and Western and Southern Europe across the period 1600–1900.1 Seasonal labor migration and involvement in off-farm activities by peasants with an economic base in subsistence-type farming have even historically occurred on such a scale, and for such a long period in Russia, that the Russian language offers specific terms for some of these phenomena—otkhod or otkhodnichestvo (seasonal work involving out-migration by peasants), and promysly (peasant crafts and trades). Unfortunately, though, much of the historiography on these phenomena has been caught up in the highly politicized debates on Russian labor history spawned by the rise and fall of Soviet socialism. In what follows, I will trace the ways in which Russian, Soviet and West- ern historiography have conceptualized and dealt with the issue amidst changing political and academic ideologies, and combine the evidence from the Russian case with the insights developed by Lucassen and others, to provide a long-term and comparative perspective on the rise of wage labor and processes of economic specialization within the household economy. I argue that systematic reliance of households on wage labor as one of many sources of income is as good, or perhaps an even better indication of economic modernization than outright proletarianization and household specialization in wage labor as such. Diversification of

1 Jan Lucassen, Migrant labour in Europe, 1600–1900: The Drift to the North Sea (London, 1987). 354 gijs kessler sources of income emerges as a strategy of modernization and adaptation to a changing environment, whereas specialization can be expected to occur in periods of social and economic stability.

Agrarian overpopulation: a contested notion

Scholarly engagement with the issue of peasants’ participation in waged labor and other off-farm economic activities has been dominated by notions on the existence of surplus labor and “agrarian overpopulation” in the Russian countryside. Having started out as an essentially political debate between liberals, populists and Marxists over the prospects for economic development in the late 19th century Russian Empire, the issue became the subject of serious academic discussion from the 1920s on.2 Principally, there were three sides and two positions to the debate. One position was that of the liberals, who interpreted peasant involvement with hired labor and other seasonal off-farm activities as indicative of the existence of a labor surplus locked within a low-productivity agricultural sector. This surplus was seen as an impediment to economic growth, but once released, also a potential source for industrial and economic expansion. Marxist analysts basically followed this line of argument, but preferred to speak of “agrarian overpopulation” rather than labor surplus, underlining the plight of the millions of peasants involved.3 The second position was that of the populists, who contested liberal and Marxist images of a Russian peasantry locked in a straitjacket of primitive low-productivity agriculture, and criticized the very notion of a labor surplus in peasant agriculture, stressing the importance of its quintessential institution, the obshchina (redistributive land-commune) in matching land to labor resources. Whereas initial populist thought of the nineteenth century had hinged on the idealization of Russian peasant farming, the works of the economist A.V. Chayanov, published in the 1920s, lifted the populist position to a wholly new level of analysis. Chayanov’s theory stressed the fundamentally different principles on which the peasant household economy operated.4 Crucial in this respect

2 For an outline of the original debate and its repercussions for historiography, see Peter Gatrell, The Tsarist Economy 1850–1917 (London, 1986), pp. 1–28. 3 Lev E. Mints, Agrarnoe perenaselenie i rynok truda v SSSR (Moscow, 1929), pp. 1–6. 4 A.V. Chayanov, Ocherki po ekonomike trudovogo sel’skogo khoziaistva (Moscow, 1924) and idem, Organizatsiia krest’ianskogo khoziaistva (Moscow, 1925), published in an English translation in Daniel Thorner et al. (eds), A.V. Chayanov on the Theory of Peasant Economy (Madison, 1986). wage labor & household economy: a russian perspective 355 was the valuation of labor, which was based on the contribution it made to meeting the needs of the household, rather than on its marginal productivity. The intensity of labor efforts on the farm and the involvement in non-agricultural activities therefore varied in accordance with the family cycle and the consumer-producer ratio within the household. When there were many non-working mouths to feed, peasants would intensify efforts, but when the consumer-producer ratio improved, the contribution of each producing member to the overall labor effort would be toned down. Accordingly, what might seem to be “idle hands” at first sight, was by no means indicative of a labor surplus, but merely a reflection of a particular stage in the family cycle of a household. By the same token, involvement in wage labor or other off-farm activities reflected a phase of intensification in the overall labor effort of the household at a particular point in its development. The three sides in the debate all rated peasant involvement in wage labor differently. Liberals saw it as the prime indication of an unused or insufficiently utilized pool of labor in the agricultural sector. Marxists regarded it as a symptom of poverty and dislocation, brought about by the onset of capitalist relations and exploitation in agriculture. Populists claimed it was not indicative of any sort of problem or change, because it simply was part and parcel of the peasant economy as it had existed for centuries. As Chayanov and the populist school of thought became a victim to repression in the Soviet Union from the late 1920s, and their insights were largely forgotten, the liberal and Marxist view dominated the scene for most of the 20th century in, respectively, Western and Soviet historiography. In the West, the works of Alexander Gerschenkron received attention above all.5 In the Soviet Union, the existence of agrarian overpopulation and the characterization of peasant labor migration (otkhod) as a flight from rural poverty became the standard interpretation.6 Only after Chayanov’s work had been rediscovered in the West and published in an English translation in the late 1960s did the populist interpretation gain some new followers. Inspired by authors like Teodor Shanin, a “neo-Chayanovist” approach to Russian peasant studies emerged.7 Peasant

5 Alexander Gerschenkron, “Agrarian Policies and Industrialization. Russia 1861–1917,” in Peter Mathias and M.M. Postan (eds), Cambridge Economic History of Europe (Cam- bridge, 1965), Vol. 6-II, pp. 706–800. 6 Mints, Agrarnoe perenaselenie, pp. 1–6; Gatrell, Tsarist Economy, pp. 89–91. 7 Teodor Shanin, The Awkward Class: Political Sociology of Peasantry in a Developing Society. Russia, 1910–1925 (Oxford, 1972); idem, Russia as a Developing Society (London, 1985). 356 gijs kessler involvement in wage labor and labor migration gained appreciation as an economic strategy in itself, rather than being merely a flight from poverty, or a process of urbanization thwarted by bad institutions like the commune. An influential work of the time was that of Robert Eugene Johnson, who referred to otkhodniki as “peasant-proletarians,” who were routinely embedded in both urban and rural sector of the economy, and used the money earned in industrial employment to pay taxes, purchase additional grain for consumption, or rent or purchase additional land.8 These new perspectives developed in close interaction with the spread of peasant studies into the developing world, where much similar patterns of outside-earnings and cross-sectoral economic strategies were identified. An influential theory among Marxist anthropologists in the 1970s and 1980s spoke of the “articulation of modes of production,” to explain how subsistence farmers were partially integrated into labor and commodity markets. A crucial factor in this process was the monetization of traditional transactions, which forced peasants to “raid” the cash-economy by devoting part of their land to the cultivation of cash-crops, often exotic ones like coffee, or periodically engage in hired labor.9 Some of this work also spoke of a “peasant-proletarian” existence to describe such situations; and, as Johnson explicitly recognized in his 1979 study, patterns of migration similar to the 19th century Russian one, in which peasants incorporated outside earnings into village structures, could be found in many places across Latin America, Southern Africa and Asia.10 Notwithstanding these new insights, the perception of wage-labor as an economic activity essentially at odds with peasant subsistence farming has proven surprisingly tenacious in Russian historiography. If not the letter, the tone of analysis has primarily remained one of tension and conflict between the different modes of production and lifestyles, best exemplified by recurring references to the suspension of production at factories

8 Robert E. Johnson, Peasant and Proletarian. The Working Class of Moscow in the late 19th Century (New Brunswick, 1979), pp. 40, 50. 9 Pierre-Philippe Rey, Les alliances de classes: sur l’articulation des modes de production (Paris, 1973); Samir Amin, Le développement inégal essai sur les formations sociales du capitalisme périphérique (Paris, 1973); Claude Meillassoux, Femmes, greniers et capitaux (Paris, 1975). For an example of the forcible integration of Cameroonian peasants in cash- crop cultivation through a monetization of the bride price, see Peter Geschiere, “Imposing Capitalist Dominance through the State: The Multifarious Role of the Colonial State in Africa,” in Wim van Binsbergen and Peter Geschiere (eds), Old Modes of Production and Capitalist Encroachment (London, 1985), pp. 94–144. 10 Johnson, Peasant and Proletarian, p. 156. wage labor & household economy: a russian perspective 357 over the summer, to release the workforce for the harvest season.11 Such perspectives have fed perhaps as much on more widely held views of a quintessential peasant aversion to markets, reinforced by peasant studies in the developing world, as on deep-seated affinities with proletarianization and class formation processes within the field of Russian labor history itself. As a result, analysis of the complementarity of wage labor and subsistence agriculture has tended to take a backseat in existing studies, and what is lacking is a proper understanding of what Lucassen has called the “work cycle” of such households.

Perspectives from the sideline

In what amounts to a refinement of Chayanov’s insights, Lucassen intro- duces the concept of the “work cycle,” essentially a graphic representation of the complementarity of the economic activities of a household as a circle divided in twelve segments, depicting the distribution of various kinds of work and the respective earnings involved over a typical year.12 Through the use of outer and inner layers, it also renders the combination of economic pursuits which coincide in time. Additional labels indicate the division of labor between the sexes and generations, which allows a household to accommodate simultaneous demands on its labor reserves. The work cycle, therefore, is more than just a “presentation tool” because it gives due attention to the factor of time in the analysis of cross-sectoral or multiple income household strategies. As Lucassen shows, many of the types of work performed by migrant laborers have a seasonal character which is no less pronounced than that of agriculture itself. Because temporary employment was often sought over substantial distances, peasant migrant laborers operated on very tight schedules, even profiting from slight climatological differences between their regions of origin (the “push-area”) and the regions of destination (the “pull-area”) to take on one and the same kind of work in two locations in succession. Mowers from the Western areas of Germany, for example, would combine work in the hay harvest in the specialized dairy farming provinces of Friesland in the East, and in Holland in the West of the Netherlands, with only a two-weeks’ interval in-between. Non-agricultural jobs in construction, brick-making, bleacheries and the transport sector

11 Ibidem, p. 35. 12 Lucassen, Migrant labour, pp. 95–9. 358 gijs kessler also were more than often tied to a specific season.13 Finally, there was of course the work on their own farms, which required a timely return to the region of origin, in as far as it was not performed by the women or adolescent members of the household remaining behind. Analysing these various activities as one work-cycle of interlocking elements emphasizes complementarity over conflict, and goes a long way towards explaining the tenacity and longevity of multiple income household strategies. Most instructive from the Russian point of view—in which seasonal labor migration and off-farm earnings by subsistence farmers are generally associated with backwardness, and the thwarted and distorted economic development of the country—is Lucassen’s analysis of the rise and fall of the North Sea System of labor migration as intricately related to the development of commodity production and labor markets, in what was at the time of its emergence by far the most economically developed part of Europe—the provinces of Holland and adjacent areas on the North Sea coast.14 Seasonal labor markets and seasonal labor migration were part and parcel of this process of economic development; the areas of origin in the East of the Netherlands and in Westphalia were just as much involved in this, be it in a different way than the regions of destination in the capitalist West. From such a perspective, the integration of wage labor and outside earnings into a household work cycle based on subsistence farming emerges in a somewhat different light: it was a strategy of improvement through a diversification of income-earning activities, rather than a stop-gap to stave off poverty, or a forced engagement with the cash economy. Recent comparative work by Bas van Bavel on the incidence of rural wage labor in the Low Countries and Northern Italy from the 12th to the 17th century even inspires us to take this argument a step further—it can be read to suggest that it is not so much just the incidence of wage labor, but rather the co-existence of wage labor with other economic activities which is indicative of processes of economic development and economic growth.15 Comparing the incidence of wage labor and the share of wage

13 Ibid., pp. 52–92. 14 Ibid., pp. 133–41. 15 Bas van Bavel, “Rural Wage Labour in the Sixteenth-century Low Countries: an Assessment of the Importance and Nature of Wage Labour in the Countryside of Holland, Guelders and Flanders,” Continuity and Change, 21, 1 (2006), pp. 37–72; idem, “The Transi- tion in the Low Countries: Wage Labour as an Indicator of the Rise of Capitalism in the Countryside, 1300 1700,” Past and Present, 195, supplement 2 (2007), pp. 286–303; idem, “Markets for Land, Labor, and Capital in Northern Italy and the Low Countries, Twelfth to Seventeenth Centuries,” Journal of Interdisciplinary History, 41, 4 (2011), pp. 503–31. wage labor & household economy: a russian perspective 359 labor in total labor inputs in agriculture, proto-industry, crafts and para- agrarian work (mainly diking) in three parts of the Low Countries (Hol- land, Flanders and the Guelders river area) in the mid-16th century with data on wage labor for Central and Northern Italy, Van Bavel finds that the highest share of wage labor, and the highest degree of proletarianization are found in the Guelders river area, the least economically developed and urbanized region of the four.16 This was related to a process of accumulation of land leased by large-scale tenant farmers, which deprived others from access to the land, and essentially forced the latter to rely on hired labor to provide for the means of existence. At the same time, Flan- ders and Northern-Central Italy, both highly economically developed and urbanized areas, showed a low incidence of rural wage labor, although for slightly different reasons. Flanders was dominated by proto-industry based on family labor, whereas in Italy urban landlords relied on sharecropping arrangements to bind rural labor to the land. Most interesting from the point of view of my argument, however, is the situation in Holland, by far the most capitalist and dynamic part of the world at that point. Although the incidence of rural wage labor was also high here, what was most characteristic for the region was that rural households regularly combined wage labor with other economic activities.17

A closer look at the Russian evidence

Armed with these insights, let us now return to the Russian case. No comprehensive studies on the historical dynamics of peasant labor migration for off-farm earnings exist, but based on a case-study of brick-making, a seasonal industry, we can trace the emergence of peasant migrant laborers, or otkhodniki to the late 17th century, when they started to replace urban craftsmen in certain stages of the production process.18 The immediate background to this development was an increase in construction activity, fuelled by the general economic expansion of the mid- and late

16 In particular van Bavel, “Rural wage labour,” p. 62 and van Bavel, “Markets,” p. 527. 17 Van Bavel, “Transition,” pp. 300–1. 18 Gijs Kessler and Jan Lucassen, “Labour Relations, Efficiency and the Great Diver- gence. Comparing Pre-industrial Brick-making across Eurasia, 1500–2000,” in Maarten Prak and Jan Luiten van Zanden (eds), Technology and Human Capital Formation in East and West (Leiden, 2012, forthcoming). On the late 17th century appearance of peasant seasonal laborers, see N.V. Voronov, “O rynke rabochei sily v Rossii v XVIII veke (po materialam kirpichnoi promyshlennosti),” Voprosy Istorii, 55, 3 (1955), p. 93. 360 gijs kessler

17th century. Until then, brick-making had been carried out exclusively by urban craftsmen, mobilized through the so-called Masonry Chancellery (Prikaz kamennykh del), instituted in 1584 at the end of the reign of Ivan IV to assign the necessary labor force and craftsmen to important construction projects commissioned by the state or the church. This essentially unfree and cumbersome system of labor allocation, however, could not cope with the rising demand for bricks during the economic upswing of the late 17th century, and was gradually replaced by the subcontracting of peasant work gangs before being finally abolished in 1700.19 To what extent the brick-making industry was unique in this respect we do not know, for lack of specialized studies on other seasonal industries over a longer time-frame. But recently published data for 1678 from the Global Collaboratory on the History of Labor Relations strongly suggest that, in other sectors of the economy as well, peasant wage labor only just made its appearance in the late 17th century. The data set relies on the 1678/79 taxpayers’ census books to classify the population of the Russian state at that time according to the taxonomy of labor relations developed within this Collaboratory.20 Starting from the broadest possible definition of work as “any human effort adding use value to goods and services”,21 from hunting-gathering and independent production to household work and slavery, the Collaboratory uses the term labor relation as a common denominator for the wide variety of hierarchical, non-hierarchical, familial, forced, voluntary, monetary, tributary and other social relations under which this work is performed. The taxonomy orders this variety into four larger overarching categories—non-working (too young, too old, or unable to work), subsistence labor (household producers, including kin producers), tributary labor (slavery, feudal labor services and convict labor) and commodified labor (self-employment, entrepreneurial, and wage labor). Acknowledging the historical occurrence of multiple income- earning activities, the set-up of the database is such as to allow for the assignment of up to three labor relations for each individual or group of individuals. The 1678 data for Russia are plotted in Figure 1. Together, the four categories comprise the total population. Those too young (≤6) or too old (≥60) to work are included in the first column to

19 A.N. Speranskii, Ocherki po istorii Prikaza kamennykh del Moskovskogo gosudarstva (Moscow, 1930), in particular pp. 7–13, 186–91, 194–6. 20 On the Global Labour Collaboratory and its taxonomy, see [October 16, 2011]. 21 Chris Tilly and Charles Tilly, Work under Capitalism (Boulder, Colo, 1998), p. 23. wage labor & household economy: a russian perspective 361

8 subsistence + 7 commodi ed

Millions 6 subsistence + 5 tributary 4

3 2 1 subsistence

0 non-work subsistence tributary commodi ed Source: Compiled by author on the basis of The Global Collaboratory on the History of Labour Relations 1500–2000, https://collab.iisg.nl/group/labourrelations/, Dmitry Khitrov, Russia 1650 ( June 2011). Figure 1. Labor relations Russia 1678—Total population the left, the self-employed, entrepreneurs and full-fledged wage or salary earners in the first column to the right, those exclusively performing feudal labor services or other types of unfree labor in the second column from the right, and those engaged in subsistence labor or combining subsistence labor with other economic pursuits in the second column from the left. Not surprizingly for a peasant society—as Russia was at this time—this last category comprises the overwhelming majority of the population. Within this category, the large share of people who combined subsistence agriculture with tributary labor (4,655,210 or 62.8%) reflects the recent imposition of what has been called the second serfdom in the Russian lands, considered to have been finalized by the Ulozhenie of 1649.22 Another 2,655,094 people (35.8%) were only engaged in subsistence labor;

22 The term “second serfdom” and the associated interpretation of the imposition of serfdom as a system has recently been challenged, see Alessandro Stanziani, “The Legal Status of Labour from the Seventeenth to the Nineteenth Century: Russia in a Comparative European Perspective,” International Review of Social History, 54, 3 (2009), pp. 359–89, and idem, “Revisiting Russian Serfdom: Bonded Peasants and Market Dynamics, 1600s–1800s,” International Labor and Working-Class History, 78 (2010), pp. 12–27. However, the imposition of feudal labor obligations on a significant part of the peasant population it effectively amounted to, is undisputed. 362 gijs kessler and only 104,260 peasants (1.4%) combined farming with either wage labor or non-agricultural crafts and trades. Generally, the literature regards the rise of peasant seasonal labor migration (otkhod) and wage labor as a phenomenon of the late 18th century.23 It is considered to have been related to the transfer of church and monas- tery landholdings to the state, which commuted feudal labor obligations for the peasants on these lands into the quitrent (obrok) customary for state peasants, increasing their freedom of movement and resulting in a growing supply of peasant labor to industry.24 Indeed, with the exception of kiln-men and overseers, late 18th century Moscow brick-making predominantly relied on a workforce of peasant migrant laborers from the surrounding provinces.25 It should be noted that, at this time, such laborers still were mostly serfs, drawn either from seigneurial estates, or from state or crown land. As said, the latter two categories paid a quitrent, and were therefore free to engage in wage labor, but many serfs from seigneurial estates had to perform labor services to their lords, and needed permission to go and work elsewhere. Popular belief notwithstanding, serfdom and peasant labor migration were by no means incompatible. It has been well-documented that serfs not only migrated for work elsewhere of their own accord, but also were in many cases facilitated and even encouraged in this by estate owners, particularly in less fertile regions where income from agriculture alone did not suffice to support a household.26 Data on labor relations for 1795, the next historical cross-section defined by the Global Collaboratory of Labor Relations, confirm this increase in peasant involvement in wage labor compared to the late 17th century (Figure 2). In the subsistence category, by now some 3,203,098 peasants (11.4%) were involved in wage labor next to farming. Meanwhile, manu- mission and the transfer of peasants from physical labor service to the payment of a quitrent had brought down the share of peasants performing tributary labor for their lords to 44%, almost equalling the share of peasants who were engaged in farming only (45%).

23 David Moon, The Russian Peasantry 1600–1930: The World the Peasants Made (New York, 1999), p. 150; V.M. Kabuzan, Narodonaselenie Rossii v XVIII—pervoi polovine XIX v. (po materialam revizii) (Moscow, 1963), p. 122. 24 Voronov, “O rynke,” pp. 97–8. 25 N.V. Voronov, “Stachka moskovskikh kirpichedel’chev letom 1779 g.,” Istoricheskie zapiski, 37 (Moscow, 1951), p. 294. 26 Tracy K. Dennison, The Institutional Framework of Russian Serfdom (Cambridge (UK) and New York, 2011), pp. 149–80. wage labor & household economy: a russian perspective 363

30 subsistence + 25 commodi ed Millions

20 subsistence + tributary 15

5 subsistence

Peasant involvement in wage labor on a much greater scale made its appearance in the decades after the 1861 abolition of serfdom. This was not so much a consequence of the reform as such, but a corollary to the industrialization and economic expansion of this period, which fuelled the demand for labor in various regions of the country. The number of wage laborers in late 19th century and early 20th century Russia has been a subject of lengthy debate and controversy in historiography—not in the least because of its role in class formation, which had direct implications for the legitimacy of Soviet rule. Little consensus has been reached, and so far no comprehensive and authoritative data-set on the occurrence of wage labor exists. The principal problem is that no solid figures exist on the number of peasant migrant labourers. Statistics on the number of passports for internal migration issued to peasants can provide some indication. The average annual number of passports issued rose from 3.6 million in the 1870s to 4.7 million in the 1880s, 6.2 million in the 1890s, and 8.8 million for the years 1906–1910.27 These data include a certain amount of double-counting, because passports had to be regularly renewed, and of course they leave aside non-migratory employment, but they no

27 Jeffrey Burds, Peasant Dreams and Market Politics. Labor Migration and the Russian Village, 1861–1905 (Pittsburgh, 1998), p. 22. 364 gijs kessler doubt accurately reflect the order of magnitude of peasant involvement in wage labor at the time. The most important regions of destination for peasant migrant laborers were the towns and factory settlements of St. Petersburg, Moscow and their hinterland, as well as the smaller industrial centers of the wider Cen- tral Industrial Region. They attracted peasant migrant laborers from the rural areas of the region itself, as well as from the adjacent provinces of the Non-Black-Earth Region and the North-West. Not only did peasants provide the bulk of new entrants into the industrial labor force, there also were several important seasonal branches of the economy which explicitly relied on a workforce combining work in agriculture with seasonal wage employment. Most importantly, this was the construction sector, which was confined to the frost-free season, and relied almost entirely on a peasant migrant workforce of otkhodniki. Brick-making is another example (about which I will say more below). Although migrant laborers were mostly men, women also came to the towns for off-farm employment, mostly in domestic service, but also in other jobs.28 Peasant migration for outside earnings had a profound impact on the regions of origin, particularly in areas like the Central Industrial Region (where up to one third of adult males migrated for off-farm earnings during the late 19th century). Labor migration in these areas, less fertile to start with and traditionally reliant on supplementary non-agricultural earnings, had received a huge boost from the arrangements surrounding the abolition of serfdom. Throughout the country, peasants had received land at emancipation, for which the state had compensated the nobles. Peasants had to pay back these sums to the state, through so-called redemption payments. These payments, as well as land taxes, were levied in cash. In these areas with poor soils, the proceeds of agriculture alone were insufficient to meet these payments. The system of taxation therefore operated on the implicit assumption that peasants would tap the cash economy through “off-farm” earnings. Gradually however, and particularly after the

28 Dmitrii N. Zhbankov, “O gorodskikh otkhozhikh zarabotkakh v Soligalicheskom uezde, Kostromskoi gubernii,” Iuridicheskii vestnik, 9 (1890), pp. 130–48; D.N. Zhbankov, Otkhozhye promysly v Smolenskoi gubernii (Smolensk, 1896); R. Munting, “Outside Earn- ings in the Russian Peasant Farm. The Case of Tula Province 1900–1917,” Journal of Peasant Studies, 3 (1976), pp. 429–46; Johnson, Peasant and Proletarian; Joseph Bradley, Muzhik and Muscovite. Urbanization in Late Imperial Russia (Berkeley, 1985); Barbara Alpern Engel, Between the Fields and the City: Women, Work and Family in Russia 1861–1914 (Cambridge and New York, 1994); Evel G. Economakis, “Patterns of Migration and Settlement in Pre- revolutionary St. Petersburg: Peasants from Iaroslavl and Tver Provinces,” Russian Review, 56, 1 (January 1997), pp. 8–24. wage labor & household economy: a russian perspective 365 still outstanding redemption payments were annulled in 1905, migration for outside earnings became a channel for a major injection of wealth into the countryside, feeding a culture of acquisition and consumption which radically altered all aspects of rural life, including the very design of peasant houses and interiors.29 Employment opportunities and the growth of wage labor were not limited to the non-agricultural sector. Commercial sugar beet cultivation in the Western parts of Ukraine, for example, relied heavily on wage labor, and from the Central Black Earth Region up to one million seasonal agricultural laborers annually went to work in the steppe grain belt on the Northern and Eastern shores of the Black Sea. In these thinly populated regions, large grain-growing estates were producing for export markets and many extra hands were needed at harvest time. The seasonal migration to the steppe-belt offers a spectacular example of the complementarity of economic activities within the “work cycle” of a household described by Lucassen for the North Sea System. Because the harvest in the steppe areas usually took place somewhat earlier than in the Russian agricultural heartland, migrant laborers from the Southernmost areas of the region of origin managed to be back home in time to bring in the harvest on their own farms.30 Brick-makers also migrated over considerable distances. Moscow and St. Petersburg, as well as Warsaw, then still part of the Russian Empire, were the main production areas.31 In Moscow, workers came predominantly from the rural areas of the province of Kaluga to the South-West, and of Riazan to the South-East. To what extent these workers still combined wage labor with farming is a moot question; workers from Kaluga apparently supplemented their earnings from work in the brick-fields

29 Burds, Peasant Dreams, pp. 22–3, 141–85. 30 Robert Grube, Das Wandergewerbe in Russland (s.l., 1904), pp. 3–51; Simon Sagorsky, Die Arbeiterfrage in der Südrussischen Landwirtschaft (Munich, 1907); Timothy Mixter, “The Hiring-Market as Workers’ Turf: Migrant Agricultural Laborers and the Mobilization of Collective Action in the Steppe Grainbelt of European Russia, 1853–1913,” in Esther Kings- ton-Mann and Timothy Mixter (eds), Peasant Economy, Culture and Politics of European Russia, 1800–1921 (Princeton, 1991), pp. 294–340. For an overview of the rise and fall of this Steppe migration system, see Gijs Kessler, “Russische und ukrainische Saisonarbeitskräfte in den Getreideanbaugebieten Neurußlands und des nördlichen Kaukasus im späten 19. und frühen 20. Jahrhundert,” in Klaus J Bade et al. (eds), Enzyklopädie Migration in Europa: Vom 17. Jahrhundert bis zur Gegenwart (Paderborn, 2007), pp. 902–4. On wage labor in sugarbeet cultivation, see Robert Edelman, Proletarian Peasants. The Revolution of 1905 in Russia’s Southwest (Ithaca, 1987), pp. 53–6. 31 Based on employment statistics for 1900 and 1902, see A.V. Pogozhev, Uchet chislen- nosti i sostava rabochikh v Rossii. Materialy po statistike truda. (Sankt Peterburg, 1906), pp. 48–224. 366 gijs kessler primarily with the cutting and transport of firewood and the production of hempseed oil.32 Despite this, they were unmistakably from a peasant background, which leaves us to suppose that agricultural activities of some sort were indeed part of the work cycle of the household, but were taken care of by the other members of the family. The geographical origins of peasant migrant workers in the St. Peters- burg brick industry were more diverse. About half of all workers were from Vitebsk province in modern-day Belarus, around 10% each from Novgorod province, the St. Petersburg province itself, and from the Kaluga province, but from two different districts than those sending workers to Moscow. The remaining 20% of workers came from a great variety of other places. Depending on their origin, between 71% and 94%, and on average 86% of these workers also held a land allotment.33 The comparative approach taken here has caused us to reject traditional perspectives on this mass phenomenon as something transitional and innerly contradictory. But what were the benefits which peasants derived from such multiple income household strategies? As suggested by Joseph Bradley in his study of peasant migrants in late 19th and early 20th century Moscow, crucial was perhaps the instability and insecurity of non- agricultural employment in these early years of economic and industrial take-off. Frequent fluctuations in the business-cycle offered opportunities, but could also easily be reversed and result in mass lay-offs and unemployment. Also, most employment available in the urban economy was of the informal and casual type, intermittent even under normal circumstances, let alone in times of crisis. Peasants would therefore move in and out of urban and wage employment according to the economic tide—tapping the cash economy when opportunities presented themselves, and switch- ing back to farming during the leaner years.34 What was true in case of unemployment was of course just as relevant in case of illness, disability and old age: the family farm in the village offered the only social safety net available, and was therefore not just one, but in many ways an essential element in peasant multiple income strategies.

32 A.V. Pogozhev, Kirpichnogoncharnoe proizvodstvo Moskovskogo uezda. Opyt “san- itarno-promyshlennogo issledovaniya”, Sbornik statisticheskikh svedenii po Moskovskoi gubernii. Otdel sanitarnoi statistiki (Moscow, 1881), T. 3, Vypusk 2, p. 189. 33 T.F. Sanotskii, Kirpichnoe proizvodstvo na r. Neve i ee pritokakh (Sankt Peterburg, 1904), pp. 66–7, 74. 34 Bradley, Muzhik and Muscovite, pp. 90, 183–92. wage labor & household economy: a russian perspective 367

What is important to emphasize, is that within this “dual strategy” a switch-back to farming was by no means just an option of last resort. Dur- ing the 1880s, for example, when an industrial recession combined with a lowering of redemption payments, many peasant workers returned home and reinvigorated peasant agriculture in the Central Industrial Region.35 Indeed, within the Russian system of land tenure and inheritance, the value attached to keeping a foothold in farming favored migration strategies in which the men would work elsewhere, while the women and children stayed behind. The Russian countryside was characterized by the East European Marriage Pattern, as distinguished by John Hajnal, combining universal and early marriage with co-residence of married couples from two or more generations in multiple or joint family households.36 Brides moved in with the household of the groom, adding to the family pool of labor. Due to early marriage and high fertility, households could grow to contain substantial numbers of people, and would at a certain point be partitioned into several new core-households, usually when the head of household, or bolshak, passed away. The inheritance system was one of partitional inheritance, in which all married sons who were part of the household would receive equal shares.37 Accordingly, for a peasant migrant who wanted to insure himself of the right to the full share at the time of partition, it would have been essential to be considered a member of the household. Participating in the agricultural work was one way to achieve this, but adding a young bride to the household would certainly also have been regarded as a contribution to the family labor pool, and perhaps this was a factor explaining why marriage was, allegedly, generally considered to be a prerequisite for otkhod.38

A longer-term perspective

Summarizing, in the light of the evidence for the Russian case the complementarity of wage labor with other economic activities within the framework of the household economy emerges as a strategy of risk-aversion

35 Burds, Peasant Dreams, pp. 125, 139. 36 John Hajnal, “European Marriage Patterns in Perspective,” in David Edward Charles Eversley et al. (eds), Population in history. Essays in Historical Demography (London, 1965), pp. 101–43. 37 Cathy Frierson, “Razdel: The Peasant Family Divided,” Russian Review, 46, 1 (January 1987), pp. 35–52. 38 Burds, Peasant Dreams, pp. 72–5. 368 gijs kessler and flexibility well-adapted to the volatility of economic opportunities in emerging commodity markets. But how should we understand this phenomenon from a long-term perspective? Having explored the causal connection between the rise of wage labor as an element within composite household economic strategies on the one hand and economic modernization on the other hand, the question emerges how these mechanisms evolve in the longer run. From a late 20th century European perspective (but perhaps much less so from an Indian or African perspective) systematic complementarity of wage labor with other economic activities clearly stands out as a thing of the past. How should we interpret this tendency towards specialization? Several factors would appear to be involved. In the first place, multiple income strategies offer comparative advantage above all in situations of economic modernization or transformation, because of general volatility and the unpredictability of the eventual vector of change. But once economic systems consolidate and employment opportunities stabilize, specialization on particular niches in the labor market comes to offer greater benefits than “double-dipping” and risk-aversion. Stable employment opportunities make it possible for households to have their economic base in wage labor. And makes it easier for young married couples to secede from the parental household, accelerating processes of urbanization, and fundamentally changing the equation. What is striking in this respect is that both Russia (with its ubiquitous peasant-proletarians at the start of the 20th century) and Western Europe have eventually followed largely the same developmental path, albeit with interesting differences attributable to the legacy of the Soviet period. First among these differences is a much more rapid and much more total “proletarianization” of the population—due to the repressive policies of the Soviet state towards self-employment and entrepreneurial economic activities, waged and salaried employment virtually became the only economic activity open to the population by the 1930s.39 And a second important difference with Western Europe was the timing and magnitude of the increase in female labor participation rates, which became all but universal in the Soviet Union at the late 1960s.40

39 Gijs Kessler, “Work and the Household in the Inter-war Soviet Union,” Continuity and Change, 20, 3 (2005), pp. 412–3. 40 Andrei Markevich, “Soviet Urban Households and the Road to Universal Employ- ment, from the End of the 1930s to the End of the 1960s,” Continuity and Change, 20, 3 (2005), p. 455. wage labor & household economy: a russian perspective 369

What is more, the similarity between the two also extends to the move away from strict specialization towards a renewed complementarity of economic activities in the last decade of the 20th century and the first decade of the 21st century. Both in the West and in the East these were times of fundamental changes in employment patterns, initially of a different nature, but later blending into one. In Russia, the implosion of the Soviet economic system and the collapse of real wages gave rise to a resurgence of income diversification and side-earnings, mostly within the wage labor segment, but often also involving self-employment, entrepreneurial activity and even subsistence agriculture.41 Meanwhile, in the West, economic policies aimed at the liberalization and deregulation of labor markets created more flexible employment structures, and stimulated entrepreneurship and self-employment. The dominance of wage labor and a single nine-to-five life-long job now largely belong to the past, and a new period of income diversification and complementarity appears to be around the corner.

41 Sergey Afontsev, “Post-Soviet Urban Households: How Many Income Sources Are Enough?,” Continuity and Change, 21, 1 (2006), pp. 131–57.

Cane Sugar and Unlimited Supplies of Labor in the 1930s: New Thinking and the Origin of Development Economics

Ulbe Bosma

A full history of the collapse of the international sugar market in the 1930s with its global and local repercussions has yet to be written.1 There are, to be sure, fine national and even regional studies already. Nigel Bolland for example has enriched the historiography of labor movements with a painstaking comparison of the labor rebellions in the British West Indies of the 1930s. Yet his narrative stays within the contours of the failed quest for post-colonial unity by the various parts of the British Caribbean. In truth, the topic involves much more than that. The mixture of national- isms and labor resistance were also vociferous in the Dutch and French Caribbean, while states like Cuba and Haiti, under the tutelage of the United States, sought both political and economic nationhood.2 Through- out the Caribbean region, hunger rapidly eroded the already contested legitimacy of colonial rule in the 1930s. Resistance threatened to culmi- nate in outright uprisings of the entire society, should colonial governments remain impassive about the suffering of the workers (which was the case in Surinam, for example). But even the more enlightened colonial governments might also find themselves in situations where only bullets could quell the social unrest resulting from the economic slump in their territories. The 1930s crisis in the Caribbean manifested the vulnerability of plantation labor, and not for the first time. It inspired Nobel Prize Laureate W. Arthur Lewis to formulate his theory of “unlimited supplies of labor”— the argument being that plantation wages could be kept at an incred- ibly low level thanks to the abundance of labor supplies in comparison to

1 Bill Albert and Adrian Graves (eds), The World Sugar Economy in War and Depression, 1914–40 (New York, 1988) provide some overview, but not in a systematic comparative way. 2 Haiti was brought under U.S. control in 1915—a situation that was deemed to expire in 1936, but was in fact ended in 1934 because the Roosevelt administration expedited the ending of military occupation (though the American co-management of the Haitian debts continued until 1947). The American occupation had not done anything to reduce the dismal poverty and appalling illiteracy. Other American interventions had been made in Nicaragua and the Dominican Republic, whereas Cuba was a semi-dependency under the Platt-agreement. 372 ulbe bosma land and capital. Low labor productivity in subsistence agriculture was a symptom of hidden unemployment,” Lewis argued, and this kept wages for unskilled labor in the highly developed parts of the colonial economies at their low level. Critics have argued against the notion of “hidden unemployment,” pointing out that low labor productivity might also be a consequence of the labor intensive ways of food production in traditional agriculture. In this case, all labor is needed for subsistence.3 This argument does not apply to the cane-sugar belts of the world, where immense amounts of labor were either directly available or imported by plantation holders. It is important to understand that, according to Lewis, “unlimited supplies of labor” are not a natural condition of the Global South, but specifically a feature of densely populated (and often heavily colonized) countries. Plantation owners exacerbated these conditions, either by shifting production to locations where labor was abundant, or by creating abundant supplies of labor through attracting migrant labor. My aim in this essay is to revisit the labor unrest of sugar producing areas in the Caribbean of the different colonial empires in the 1930s; locate them in the wider framework of global history; and relate their divergent trajectories to the supplies of labor. By doing so, one can, as it were, “zoom in” on the formative stages of development economics, which gave a central place to the question why workers of the world were so unequally paid, regardless of their productivity. At the same time, I intend to demonstrate how the global cane sugar economy, dominated by Cuba and Java in the early 20th century, offered an excellent “laboratory” to develop theoretical notions which later became the centerpiece of development economics. In this sense, the making of global labor history and the making of development economics are intimately connected.

Early excursions into development economics

In the 1930s, hunger pervaded the cane sugar exporting colonies of the world, and hunger marchers took to the streets throughout the French, British, Spanish and Dutch Caribbean. Cuba even experienced a revolution in 1933. But in contrast to widespread labor unrest in the Caribbean sugar plantations, the Asian sugar complex of Java—which suffered even

3 See Robert L. Tignor, W. Arthur Lewis and the Birth of Development Economics (Prince- ton, 2006), pp. 98–9; and for a well-known critique see the book of his co-laureate Schultz, Transforming Traditional Agriculture (New Haven: Yale University Press, 1964). new thinking and the origin of development economics 373 more from the slump of the world sugar market—did not experience any large scale unrest. In Java, collective action in the sugar industry had been squashed in the early 1920s, and failed to be organized in the 1930s, despite resentment and anger among the workers. I would argue that, in the case of Java, it was the very existence of “unlimited supplies of labor” that seriously undermined the basis for any collective action. One may even wonder what workers in tropical dependencies like Java had to gain from their participation in the global economy under these conditions. Wages stayed at an extremely low level, and the majority of the Javanese peasants living around sugar factories continued to be poor, in spite of the fact that tropical plantation agriculture had reached levels of efficiency that compared well with those of the industrial world. It was in this context that J.H. Boeke launched his concept of “dualism” in his inaugural lecture as Professor in Tropical-Colonial Economy at the University of Leiden in 1930—a crystallization of ideas he had been developing from the early 1920s onwards.4 The colonial state and colonial business had inserted the village economies in a monetized world, but in such a way that “the obligations of the village outweigh its assets.” In other words, economic dualism meant a net loss for Javanese rural society.5 Lewis’s notions of economic dualism and unlimited supplies of labor strikingly resemble Boeke’s, although Lewis probably never read Boeke before publishing his article on the unlimited supplies of labor, and did not share Boeke’s preoccupation with the cultural dimensions of colonial economics. Having been involved in building a system of cooperative credit in colonial Indonesia, Boeke discarded the explanation of marginal productivity as “a niece piece of economic theory” which was unable to come to grips with the cultural divide between the oriental pre-capitalist village economy and Western capitalism. It is true that Lewis linked his concept of “unlimited supplies of labor” to notions of “traditional” and “modern” sectors of the economy, in order to give his theory worldwide coverage—he has even been criticized for that—but with his Caribbean background he would never resort to the “oriental communal village” as

4 J.H. Boeke, “Auto-activiteit naast autonomie. Voordracht van Dr. J.H. Boeke in de ver- gadering van het Indisch Genootschap te ‘s-Gravenhage op 13 october 1922,” Blaadje van het Volkscredietwezen 11, 3 (1923), pp. 53–77; Idem, Dualistische economie. Rede gehouden bij de aanvaarding van het ambt van gewoon hoogleraar in de tropisch-koloniale staathuishoud- kunde aan de Rijksuniversiteit te Leiden op 15 januari 1930 (Leiden, 1930). 5 J.H. Boeke, Economics and Eonomic Policies of Dual Societies as Exemplified by Indone- sia (New York, 1953), p. 69. 374 ulbe bosma an explanatory variable in the way that Boeke did. At any rate, Fabian concerns about poverty in the West Indies and Dutch leftist thinking about the Indonesian economy seem to have developed independently from each other. The only possible exception is J.S. Furnivall’s concept of the “plural society,” which resulted from his study of colonial Indonesia, and which is indebted to Boeke’s thinking.6 Furnivall was a member of the Fabian society, and in one of the society’s publications we can find him in the same list of contributors as Lewis. Nevertheless, it seems that his ideas were not influential among his fellow Fabians.7 Against the backdrop of the lack of intellectual exchange between the colonial experts of the Fabian society and those of the Dutch Labor Party, the resemblances in their economic thinking are striking. The work of Lewis also shares some important concerns with the work of the Dutch economist and statistician Jacob van Gelderen, who came to work at the Netherlands Indies statistical office in 1919. More than Boeke, van Gelderen was inclined to involve demographics and the cyclical nature of the global economy in his analysis of the tropical economies. In 1913, thirteen years before Kondratiev’s discoveries, van Gelderen had already identified the two-generation economic cycle. But because van Gelderen published only in Dutch at the time, he never won international recognition for it. Van Gelderen was not cited by Lewis in his 1970s publication on the Kondratiev cycle, for example.8 Van Gelderen played his role in propagating economic planning as a prominent member of the Dutch Labor Party. Moreover, in 1937 he acted as the Dutch signatory for the London Sugar Convention, which regulated global sugar production in a way which, according to both van Gelderen and Boeke, could “hardly be neglected in the further development of international economic policy.”9 What these excursions in early development economics by Boeke, Fur- nivall, Van Gelderen and Lewis demonstrate is that a new chapter was being opened in the thinking about labor in tropical dependencies subjected to the inequalities and unfairness in the world market—new thinking which left the classical Ricardian perspective behind, but which also

6 Julie Pham, “J.S. Furnivall and Fabianism: Reinterpreting the ‘Plural Society’ in Burma,” Modern Asian Studies 39, 2 (2005), pp. 321–48, at 321. 7 See A. Creech Jones and Rita Hinden, Colonies and international conscience. Report by [. . .] to the Fabian Colonial Bureau (London: Fabian Publications, 1945). 8 See W. Arthur Lewis, Growth and fluctuations, 1870–1913 (London, 1978). 9 Boeke, Economics and Economic Societies, p. 142; J. van Gelderen, The Recent Develop- ment of Economic Foreign Policy in the Netherlands East Indies (London, 1939), p. 62. new thinking and the origin of development economics 375 went beyond debates on “colonial drainage,” because it focused on the global economy rather than on the economic relationship between colony and mother country. It identified mass unemployment in plantation societies in the Global South, and the ensuing weakness of the position of labor. It also advocated the need to organize demand at a national level alongside international economic planning to alleviate the consequences of the cyclical depressions. And last but not least, this new thinking arose from a consciousness that the lack of coordination of the sugar market between the World Wars had led to the traumatic experiences of overproduction and the ensuing sugar slump, which caused deprivation and hunger both in the Caribbean and in Java. Since early modern times, the global sugar market had been a playfield of national interventions; it was a market that, it was now believed, could only reasonably function with a firm degree of coordination. The story about the global sugar market, and about sugar and labor, cannot be told without extending the comparison to Cuba. This is in many ways a comparison of contrasts. Being the prime sugar producer in the world, Cuba did not at all rely on unlimited supplies of labor, but rather on abundant supplies of land. Cuba was in fact outcompeting the other Caribbean sugar producers, thanks to both its fertility and its privileged access to the market of the United States (under the Platt agreement). In other words, low wages were not a result of fierce product competition on the world market but the consequence of an abundant supply of labor and land, and resource monopolies. I will first assess the impact of the global sugar crisis of the 1930s on the living conditions of the cane workers in the Caribbean and Java, and the forms of resistance to that. The big difference between Cuba’s revolutionary trajectory since 1933, and the almost complete absence of collective action in Java’s sugar industry in these years, hinges on the existence of abundant supplies of labor in the latter. By pointing out these differences I seek to explain why the concept of “economic dualism” fitted the Brit- ish West Indies and the Dutch East Indies rather than Cuba, but at the same time demonstrates why Boeke’s perspective erred in being “cultural- ist.” Second, I will identify connections between the ideas of progressive economists and other specialists on labor issues in a colonial economy. Thirdly, I will suggest how these ideas could give birth to a new way of thinking about labor in commodity producing countries. The similarities between British Fabians and Dutch Social Democrats on colonial issues, I will seek to demonstrate, had their provenance in the social setting of sugar production for the world market. 376 ulbe bosma

The crisis of the Ricardian theory of world trade

The birth of development economics and the birth of labor movements in the Global South more or less coincided, and both seemed to hold the promise of an evolutionary end to the colonial relationship. In 1939, W. Arthur Lewis published his Labour in the West Indies: the Birth of a Workers Movement for the Fabian Society. Lewis wrote his pamphlet one year before the Royal Moyne Commission would investigate the labor riots in the West-Indies. The relevance of Lewis’ publication stretches far beyond the British West Indies. The 1930s were a period of economic misery for commodity producers and for cane sugar producing countries in particular. All over the world cane sugar producing countries suffered heavily from the collapse of the global sugar market. Not just in Cuba and Java, the largest cane sugar exporting countries at the time, but also in the British West Indies, Haiti and Suriname, which were clearly waning sugar colonies but had been relying on the remittances of their migrant workers at the Cuban plantations. The sugar crisis that evolved from the late 1920s onwards was the second of its kind in a generation. The aftermath of the sugar crisis of 1883, which had lasted for about 20 years, had been a dismal period for most cane sugar-producing areas in the world in spite of the fact that global consumption was soaring and that Cuba, the largest cane sugar exporter, was crippled because of its War of Independence. In fact, an extensive investigation into the condition of sugar production in the British West Indies concluded in 1897 that the industry was in danger of being close to extinction.10 There had been some hope of improvement after the Brus- sels Convention of 1902 had forced beet sugar producers in the world to scale down the exports of their bounty-fed sugar. This had drastically reduced the share of beet sugar in total world sugar production. In these first decades of the 20th century, Cuba and Java rapidly expanded their production, but most smaller sugar producing colonies particularly in the Caribbean became increasingly marginalized. The situation was succinctly summarized by the West Indian Sugar Commission in 1929: Cuba is producing cheap cane by simply allowing it to grow year after year on immense expanses of fertile land, much of it recently in forests, with little cultivation and low yields per acre. Java and Hawaii produce cheap cane by

10 The Morris report cited in Report of the West India Sugar Commission 1929–1930 (Lon- don, 1930), p. 37. new thinking and the origin of development economics 377

intensive cultivation of small areas, giving high yields per acre. The cane- fields of the British West Indian Colonies have been under cultivation for so long and the area is so limited that Cuban methods would be disastrous.11 Elsewhere in the Caribbean, the situation was not much better. The sugar industry in Martinique was waning; it survived thanks only to its rum production, for which there was excess demand during World War I. An emerging sugar industry in Haiti was simply nipped in the bud, because of lack of access to the United States market.12 Meanwhile, the British Caribbean sugar production was either stagnating or declining. In the British West Indies, total sugar production stood at 338,392 tons in 1928, which was just 1/8 of Java’s total sugar output and 1/12 of Cuba’s. Jamaica’s exports showed a slight increase, but were not very important, since sugar represented only 19 percent of of its output value, and only 3 percent of its labor force was employed in sugar—many Jamaicans found work in sugar production in Cuba.13 In other words, only the largest sugar producers that were able to hold on a solid chunk of the world market seemed to have a future. The Java sugar factories had benefited most from the Brussels Convention, which allowed them to take large parts of the markets for industrial sugar in India and China. Under the Platt agreement, Cuba meanwhile enjoyed prefer- ential access to United States’ market. In the first years of the 20th century, major American companies like the United Fruit Company bought very large areas of land in Cuba for a pittance.14 The period between 1914 and 1919 then witnessed an enormous influx of US capital in the Cuban sugar industry; the Cuban sugar industry went through a period of fren- zied growth.15 The poor Oriente became a sugar hub, and by 1917 large parts of the forests in this Eastern part of Cuba were felled.

11 Report of the West Indian Sugar Commission, p. 45. 12 Jacques Adélaide-Merlande, Les origines du movement ouvrier en Martinique, 1870–1900 (Paris, 2000), p. 109; Edouard de Lepine, La crise de fevrier 1935 à la Martinique. La marche de la faim sur Fort-de-France (Paris, 1980), p. 54; Guy Pierre, “The Frustrated Development of the Haitian Sugar Industry beween 1915/18 and 1938/39: International Financial and Commercial Rivalries,” in Albert and Graves, World Sugar Economy in War and Depres- sion, pp. 121–9. 13 Report of the West Indies Sugar Committee, pp. 39–40. 14 Robert B. Hoernel, “Sugar and Social Change in Oriente, Cuba, 1898–1946,” Journal of Latin American Studies, 8, 2 (November 1976), pp. 215–49, at 229. 15 César J. Ayala, “Social and Economic Aspects of Sugar Production in Cuba, 1880–1930,” Latin American Research Review, 30, 1 (1995), pp. 95–124, at 110. 378 ulbe bosma

In 1925, the Java sugar factories had about 180,000 hectares planted with cane.16 This production was realized by about 1 million cultivators, 250,000 cane-cutters and a workforce of 150,000 in the factories, leading to a total number of workers of 1.4 million. That these figures mean a massive substitution of land for labor can be illustrated by a comparison with Cuba, which produced 5 million tons of sugar in 1925 with probably about 315,000 cane-cutters, to clear 747,780 hectares planted with cane.17 The planting in Cuba was done by about 50,000–75,000 planters and their workers, a fraction of the number of cultivators that were in the cane fields in Java.18 In terms of labor input per hectare, the Cuban figure of 0.5 compared with 7.7 in Java. In Cuba, Haitian and Jamaican workers did the heavy and lowest-paid work, and they suffered from discrimination. But in the 1910–1920 period, even their wages reached levels which were unknown in the Netherlands Indies, amounting to as much as five dollars per day for cane cutters.19 In Java, the hundreds of thousands of male and female workers in the sugar fields hardly benefited from high sugar prices of the early 1920s. Yet, considerable segments of the better positioned workers in Javanese rural society got their share, in spite of the high dividends that were siphoned off to the Netherlands. An increase in welfare seems to have taken place, and a class that had some purchasing power must have grown too. Elson mentions the rising consumption, and an increase in the pilgrimages to Mecca.20 From 1919 to 1928, the world production of beet sugar increased from 4.5 million to 9 million tons, whereas the world output of cane sugar grew from 11 to 16 million tons. In these eleven years, the Java sugar industry more than doubled its output to almost 3 million tons, whereas the Cuban

16 H.C. Prinsen Geerligs, Handboek ten dienste van de suikerriet-cultuur en de rietsuiker- fabricage op Java Vol. 4: De rietsuikerindustrie in de verschillende landen van productie: historisch, technisch, economisch en statistisch overzicht over de productie en den uitvoer van de rietsuiker (Amsterdam, 1924–1931; 2nd revised print), p. 147. 17 Ibidem, 217. 18 Data for the 1930s reveal that the cane was planted and tended by 25,000 to 37,000 colonos, of whom 18,000 were just labradores, or cultivators. Only the top layer consisted of few hundred planters who had been owners of the disappeared factories of the 19th century. These facts taken together suggest that only about 50,000–75,000 laborers were involved in the agricultural part of sugar production in the 1920. See César J. Ayala, Ameri- can Sugar Kingdom. The Plantation Economy of the Spanish Caribbean, 1898–1934 (Chapel Hill, 1999), pp. 123, 125; Ayala, “Social and Economic Aspects,” p. 106. 19 Hoernel, “Sugar and Social Change,” p. 238. 20 R.E. Elson, The End of the Peasantry in Southeast Asia. A Social and Economic History of Peasant Livelihood, 1800–1990s (Houndmills, 1997), pp. 236–8. new thinking and the origin of development economics 379 factories reduced their output, after a peak in 1925 of 5 million tons, to their previous 1919 level of about 4 million tons.21 While the Java sugar producers trusted that demand of their sugar would further expand in India and China, Cuba—with the moderate nationalist Gerardo Machado at the helm since 1925—reduced its production by 20 percent. This Cuban policy fitted Machado’s nationalist agenda to stem the influx of labor and capital that accompanied the burgeoning sugar industry. Western Cuba began to diversify its commodity production, to sever itself from production within colonial terms of trade. By contrast, the Eastern part of Cuba was controlled by American capital; there, factories had increased their economies of scale and reduced the costs of labor and land to sharply competitive levels. The Java sugar industry also drastically raised the efficiency of its factories, and increased the yields per hectare thanks to abundantly available labor.

The collapse of the global sugar market and its social consequences

Without any international regulatory device in place—the Brussels Sugar Convention had expired in 1923—the major sugar producers were left to hammer out a new mechanism to keep global sugar output in check. The government of Cuba, which had already shown its willingness to curb its production in 1927, tried to get around the table with the other two giant sugar exporters in the world (Czechoslovakia and Java) to discuss production restrictions. But the plan failed, because the Java sugar factories were confident that they could continue expanding their markets in Asia by keeping the coolie wages at the lowest possible level, while developing both the agricultural as well as the industrial side of sugar production to the most advanced technical levels.22 The intransigence of the Java sugar producers set in motion a sugar war, with dumping practices that brought about nothing less than a disaster of global proportions. In turn, dumping only hastened protectionist measures. India, an immense producer of sugar but not industrially, closed its markets for foreign sugar to protect its agriculture in 1931. For Java, this meant a loss of about 30 percent of its export market. Meanwhile, the United States decided to protect its beet sugar industry. Cuba’s factories discovered that they could only partly rely

21 H.C. Prinsen Geerligs and R.J. Prinsen Geerligs, Cane Sugar Production 1912–1937. A Sup- plement to The World’s Cane Sugar Industry. Past and Present (London, 1938), pp. 78, 98. 22 Jacob van Gelderen, Wereldcrisis (Batavia, 1931), pp. 58–9. 380 ulbe bosma on the American market, because of competition from beet manufacturing and sugar from The Philippines, Puerto Rico and Hawaii which could be imported in America duty free.23 Thus, Cuban exports to the United States fell from 3.8 million tons in 1929 to just 1.4 million in 1933, and the total number of sugar mills reduced from 163 to 125 over the same period. From 1929 to 1934, world sugar prices sank by almost 50 percent, to stay more or less at that level until 1937.24 By 1935, Cuban production had plummeted to half a million tons—one sixth of its production in 1930—and only recovered to 1.6 million tons in 1940.25 The Java sugar industry lost its Indian market for good. The consequences for the workers of the Java sugar industry were severe, as has been described in a contemporary report: . . . those no longer receiving an income in cash from the sugar or other source were forced to fall back on the native community and in many cases went to strengthen the group of so-called “kampong coolies” (casual laborers working around the villages), persons in the villages living from hand to mouth doing odd jobs as they may appear [. . .]. Again, it is probably principally this group whose state of nutrition has fallen off in the past few years [. . .]. Real symptoms of a deterioration in the diet of some groups of the people are furnished by reports that locally now and then the emergency foodstuffs are being eaten, such as cassava refuse and the pith of banana stems, which were formerly eaten in the more remote districts. Other unfa- vorable phenomena are the higher death rate in 1934, and the fact that the workers are becoming exhausted more rapidly, which was observed among those set to preparing the soil again for the planting of sugar cane in 1935.26 In Cuba, wages on the cane fields had sunk by 30 percent between July 1931 and the end of 1932, down to 20–25 cents, a decline comparable to the one in Java.27 In 1934, the Vice-President of the United Fruit Company reported that:

23 Brian H. Pollit, “The Cuban Sugar Economy and the Great Depression,” Bulletin of Latin American Research, 3, 2 (1984), pp. 3–28, at 15. 24 Anno von Gebhardt, Die Zukunftsentwicklung der Java-Zucker-Industrie unter dem Einfluss der Selbstabschliessungstendenzen auf dem Weltmarkt (Berlin, 1937), pp. iii, 56. 25 The Convention assigned 1,050,000 tons for the Netherlands and its colonies and 940,000 for Cuba; together 5/9 of the total quota of 3,622,500 tons. Gebhardt, Zukunfsent- wicklung der Java-Zucker-Industrie, p. ix. 26 A.M.P.A. Scheltema, The Food Consumption of the Native Inhabitants of Java and Madura. National Council for the Netherlands and the Netherlands Indies of the Institute of Pacific Relations (Batavia, s.a. [1936]), p. 62. 27 “General Survey of Wages in Cuba, 1931 and 1932,” Monthly Labor Review, 35 (Decem- ber 1932), p. 1404. new thinking and the origin of development economics 381

. . . the threat of communism is very real. U.S. tariff policies have destroyed its [Cuba’s] very life—the sugar industry of our best neighbor. [. . .] Cuba is economically ruined. The great mass of her people are underfed and under- clothed. They are so beset with the consequent unrest that they are prey to communists who are flooding to Cuba. There are great and numerous strikes and seizures of property throughout Cuba by workers.28 The crisis led to an exodus from the Cuban countryside. Coffee and tobacco production were able to absorb part of the labor force, but most went to the cities in “caravans of hunger,” as they were called by contemporaries.29 Many migrant laborers returned from Cuba to Haiti and the West Indies, and some to Suriname, triggering off organized resistance in their home countries. For example, the impoverished Surinamese migrant laborers returning home gave an important impulse to labor agitation in their colony. In October 1931, a hunger riot occurred in Paramaribo, and a leftist labor movement emerged with strong Communist influences, which however was immediately repressed.30 Although Martinique and Guadeloupe had a guaranteed market in France in the 1930s, a drastic reduction of the wages was implemented in 1935, which led to the hunger march in February of that year.31 The West India Royal Commission of 1938–1939 concluded, as regards Jamaica, that: [. . .] the nutritional state of a distressingly large proportion of the laboring classes and of children is considered by some observers to be defini- tively bad, the chief causes begin adverse economic conditions, poverty, low wages, unemployment, illegitimacy and overlarge families.32 The commission corroborated the findings of the young economist W. Arthur Lewis, who depicted in his Labour in the West Indies. The Birth of a Worker’s Movement the following scene:

28 Quoted in Robert Whitney, State and Revolution in Cuba. Mass Mobilization and Political Change, 1920–1940 (Chapel Hill and London, 2001), p. 119. 29 Barry Carr, “Mill Occupations and Soviets: The Mobilisation of Sugar Workers in Cuba, 1917–1933,” Journal of Latin American Studies, 28, 1 (February 1996), pp. 129–58, at 132–3; Hoernel, “Sugar and Social Change,” pp. 243, 249. 30 Ulbe Bosma, Terug uit de koloniën. 60 jaar postkoloniale migranten en hun organisa- ties (Amsterdam, 2009), p. 78. 31 Lepine, La crise de février 1935, p. 71. 32 West India Royal Commission 1938–1939, cited in ILO, Labor Policies in the West Indies Studies and Reports New Series, No. 29 (Geneva, 1952), p. 22. 382 ulbe bosma

Malnutrition plays havoc with productive efficiency and resistance to disease. There are thousands of people too poor to eat as much as is necessary, and any teacher can give cases of children coming to school on a breakfast of sugar and water with no prospect of lunch. But far greater numbers eat enough and are yet malnourished because their diet is unbalanced. There is an abundance of starchy foods, but milk, meat and other fats are so expensive as to be beyond the reach of the working classes, except as Sunday luxuries.33

The birth of labor movements among cane sugar workers

The social unrest in Cuba that followed the drastic reduction of its sugar production led to the overthrow of president Machado’s government in 1933. This sugar baron, who had been working hard to put ceilings on the Cuban sugar production, became a victim of the global sugar battles he had been desperately attempting to pacify.34 Social unrest in cane sugar-producing areas was nothing new. It had occurred in Java and the Caribbean from the late 19th century onwards. Around 1900, extensive cane burnings took place in Java, strikes erupted in Martinique and labor unrest surged in Guyana.35 The period 1917 to 1920 again saw widespread labor resistance in the cane sugar industry, both in the Caribbean and in Java. Worldwide reverberations of the Bolshevik Revolution coincided with high sugar prices and extremely high profits for the factories, while their workers suffered from rising living costs.36 What also contributed to the revolutionary fervor were the West Indian soldiers returning from the front. In the 1920s, extensive labor migration in the Caribbean carried Spanish anarchism, Jamaican Garveyism and—last but not least—Com- munism to almost every sugar island of the Caribbean. In these revolutionary times, given a volatile political situation back home in Europe, colonial administrations were inclined to take an accom- modating stance. In 1919, labor unions were legalized in Jamaica, and in

33 W. Arthur Lewis, Labour in the West Indies. The Birth of a Worker’s Movement. With a Preface by A. Greech Jones, M P. (London, 1977; first print 1938), p. 17. 34 Barry Carr, “Identity, Class and Nation: Black Immigrant Workers, Cuban Commu- nism, and the Sugar Insurgency, 1925–1934,” The Hispanic American Historical Review, 78, 1 (February 1998), pp. 83–116, at 84. 35 For Guyana, see Nigel O. Bolland, On the March. Labor Rebellions in the British Carib- bean, 1934–39 (Kingston and London, 1995), p. 14. 36 Ibidem, p. 22; Elspeth Locher-Scholten, Ethiek in fragmenten. Vijf studies over koloniaal denken en doen in de Indonesische Archipel, 1877–1942 (Utrecht, 1981), p. 101. new thinking and the origin of development economics 383

1920 they were legalized in Trinidad. In Cuba, strikes broke out in 1917 with demands for an eight hour working day, plus the recognition of labor unions. US marines intervened on the plantations. Since these Cuban labor movements operated in areas with a high concentration of sugar factories, and were led by more urbanized and more skilled workers, they were however able to survive repression. In 1925, labor actions led to the formation of the National Confederation of the Cuban Workers (CNOC). In the same year, the Cuban Communist Party (PCC) was formed. Inter- estingly enough, the CNOC was able to forge unity between field and factory workers. The combination of revolutionary reverberations and high sugar prices led even in Java’s sugar industry to the emergence of Indonesian labor unionism. In the Netherlands Indies, labor unions had been legalized already in the first decade of the 20th century, but these unions were dominated by European residents of the colony. In 1919, the Indonesian labor union of sugar workers (Personeel Fabriek Bond) which had a membership of 30,000 was recognized by the colonial government as a legitimate representative of labor.37 It was definitely a breakthrough. The sugar industry, however, vehemently opposed any collective action, and after a year of strikes in 1920, the sugar factories were able to pressure the colonial government into repressive policies which continued until the Japa- nese occupation and the Indonesian revolution of 1945 ended colonial rule. An important difference was that the higher echelons of the sugar industry consisted of Europeans, and that the colonial divide coincided with a racial divide. This seriously impeded collective action—exactly the reverse situation that applied in Cuba. In the labor resistance of the Cuban sugar industry, the communists played central role. The Surinamese Otto Huiswoud, since 1929 a candidate-member of the Executive Committee of the Communist Interna- tional, was strongly in favor of cross-racial solidarity, and had travelled extensively in the Caribbean in 1931.38 Huiswoud’s influence was visible via the Caribbean bureau of the Comintern, and the Communist Party of

37 Locher-Scholten, Ethiek in fragmenten, p. 101. 38 Ken Post, Arise Ye Starvelings. The Jamaican Labor Rebellion of 1938 and its Aftermath (The Hague, 1978), p. 3. See also Maria Gertrudis Cijntje-van Enckevort, “The life and work of Otto Huiswoud: professional revolutionary and internationalist (1893–1962)” (Disser- tation: UWI, Mona Campus, 2000), pp. 62–3. In 1930 Huiswoud had published his most important intellectual contribution as a communist activist in The Communist under the title “World Aspects of the Negro Question.” The point he made was that the Caribbean was part of the West and that class struggle was the right course and not Garveyite “Back 384 ulbe bosma the United States which both directed their attention to the sugar industry around 1930.39 Together with his wife Hermine Dumont, Huiswoud also visited Cuba in 1931, and that experience may have contributed to the staunchly cross-racial character of the Cuban communist movement and its affiliated labor union. In the early 1930s, the CNOC denounced the deportation of foreign workers by the Machado government.40 From August to October 1933, field and mill workers seized factories and estates, establishing Soviets. It was a revolt stemming partly from hunger and partly out of fear that there would be no sugar campaign, because the sugar could not be sold. There was an active participation of the Antil- lianos, and of both the Jamaicans and the Haitians.41 The Communist Party continued to hold an influential position in Cuba in the 1930s. After the removal of Grau, the new strongman of Cuba Ful- gencio Batista needed an ally against the leftist nationalist movement, of which an unlikely alliance between him and the Cuba’s communist party was the result. By February 1939, the Cuban Communist Party counted 23,300 members.42 Nonetheless, it was not able to change the sentiment against massive imports of black labor for plantations. This was associated with slavery and as ploy of plantation capitalists to bring wages down. In contrast to other cane sugar producing societies, the population of Cuba was dominated by white Creoles. From 1919 to 1953, the white population of Cuba accounted for more than 70 percent of the total, peaking at 75 percent in 1943.43 In the early 1930s, there were still between 150,000 and 200,000 Carib- bean migrant workers in Cuba.44 Under Machado, the expulsion of these workers had begun. During the short first presidency of Ramón Grau San Martín (September 1933–January 1934), who was outspoken in his objective to end the dependency on the United States, the deportation of Caribbean workers continued. As part of his “Cuban national”-oriented policies, Grau had determined that at least half of the labor force in every to Africa.” Otto Huiswoud, “World Aspects of the Negro Question,” The Communist. A Mag- azine of the Theory and Practice of Marxism-Leninism, IX, 2 (February 1930), p. 132. 39 Carr, “Mill Occupations and Soviets,” p. 136. 40 Alejandro de la Fuente, “Two Dangers, One Solution: Immigration, Race and Labor in Cuba 1900–1930,” International Labor and Working-Class History, 51 (Spring, 1997), pp. 30–49, at 43–5. 41 Carr, “Identity,” p. 114. 42 Samuel Farber, Revolution and Reaction in Cuba, 1933–1960 (Middletown, 1976), p. 85. 43 Alejandro de la Fuente, “Race and Inequality in Cuba, 1899–1981,” Journal of Contem- porary History, 30, 1 (January 1995), pp. 131–68, at 135–6. 44 Carr, “Identity,” pp. 83, 86–8. new thinking and the origin of development economics 385 enterprise must be Cuban.45 In the course of the 1930s, when communists had obtained an influential political position in Cuba, tens of thousands of migrant workers were rounded up and deported. These were mostly Haitians who, in contrast to Jamaicans, were not literate, and unable to hide out in an urban environment. In 1937, almost 25,000 Haitians were expelled from Cuba, and another 5,000 were expelled in 1938 and early 1939.46 The conditions for collective action in the British West Indies were less favorable than in Cuba, but a lot better than in Java. The sugar factories in the small islands knew exactly who the troublemakers were among their workers, as Lewis recounts.47 But the role of the colonial governments was distinctly different in the West Indies. Given the 1929 Colonial Develop- ment Act submitted to Parliament when Lord Passfield (Sidney Webb) was Secretary of the Colonies, these administrations had to pursue policies which accepted the existence of labor unions, even if these were kept under control.48 In these Fabian policies in the British West Indies, the political line was to develop “responsible” trade unionism, encouraged by the Colonial Office in collaboration with the TUC (Trade Union Congress). The political margins for the colonial governments continued to be small; either they had to take on the plantation owners, or they had to accept that workers who were paid a pittance would continue to cause unrest. While a revolution took place in Cuba, and hunger marchers took to the streets all over the West Indies, nothing of the kind happened in Java. The degree of urbanization may have played an important role in these divergent trajectories, as Lewis suggests for the West Indies. Urbanization in Cuba and the West Indies, involving about 50 percent of the population, was much more advanced than in Java, where—as Boeke pointed out—only 6 percent of the population was urban.49 In an urban environment, the mobilization of workers and an organizational infrastructure could be created and maintained more easily. The three different political trajectories of Cuba, the French/British Caribbean and the Netherlands Indies strongly suggest that the presence of collective action was directly correlated to the abundance of labor, and to the racial composition of the

45 Marc C. McLeod, “Undesirable Aliens: Race, Ethnicity, and Nationalism in the Com- parison of Haitian and British West Indian Immigrant Workers in Cuba, 1912–1933,” Journal of Social History, 31, 3 (Spring 1998), pp. 599–623, at 604. 46 Ibidem, p. 606. 47 Lewis, Labour in the West Indies, p. 39. 48 George C. Abbott, “A Re-Examination of the 1929 Colonial Development Act,” Eco- nomic History Review, 24, 1 (February 1971), pp. 68–81, at 71. 49 Boeke, Economics and Economic Policies, p. 17. 386 ulbe bosma rank and file of sugar workers. Class solidarity in Cuba was effective, but only because labor was scarce, and kept relatively scarce because migrant labor was expelled during the Depression years. For other sugar producing areas in the world economy, mass unemployment was the rule. The Cuban situation, involving more a “colonial settler state” than an “exploited colony,” was exceptional. Moreover, it was able to stay competitive on the world sugar market by economizing on labor. The British West Indies lost out in the global competition, which led to mass unemployment; the French Antilles only survived thanks to metropolitan protection. Java became the quintessential example of competition based upon unlimited supplies of labor. But when India established a prohibitive tariff wall for foreign sugar in 1931, it dealt a deadly blow to the Java sugar industry.

Unlimited supplies of labor, colonial and post-colonial answers

The 1930s were not the first time that overproduction and the ensuing slump in the sugar market inflicted hardship on cane sugar exporting colonies. Neither was it the first time that the structurally awkward conditions of the sugar workers had been brought to notice of the colonial authorities. Any attempts to have the village economy profit from sugar industry failed. In fact, Boeke concluded, the Javanese farmer was hardly better off in the early 1920s than in the 1880s. Prices on the world market stagnated from 1883 onward, until the Brussels Convention reduced the dumping practices of beet sugar producers in 1903.50 The late 19th and early 20th century were a time of rationalization and reorganization of the sugar industries, creating a period of boom and bust that lasted until the 1920s. According to Bolland, wages in the British West Indies were in 1905 the same as they were thirty years earlier, and the same applied in the Netherlands Indies, which prompted the colonial government to start an extensive “Declining Welfare Investigation” in Java in 1905.51

50 For labor unrest in the Caribbean in the late 1890s and early 1900s in British Carib- bean see Bolland, On the March, p. 7. For Cuba, where rising labor conscience became mixed with the insurrection against Spain see Dumoulin, “El primer desarrollo.” For Mar- tinique see, Adélaïde-Merlande, Les origines; For Java, see Elson, R.E., Javanese Peasants and the Colonial Sugar Industry. Impact and Change in an East Java Residency, 1830–1940 (Singapore, 1984). 51 Bolland, On the March, p. 15. new thinking and the origin of development economics 387

The idea that colonial governments were responsible for the welfare of their subjects spread with leaps and bounds in the early 20th century. The British Labor Party, for example, felt that the concept of the League of Nation’s Mandate should not just apply to the former German and Turk- ish dependencies, but might well apply to its classical colonies also.52 Ini- tially, the progressive Governor-General of the Netherlands Indies Count John Paul van Limburg Stirum (1916–1921) tried in the same vein to establish a minimum wage and an enquiry into the conditions of the workers in the sugar factories and fields.53 Not only had wages been practically stagnant for more than twenty years, they had actually declined in view of rising costs of living. Wages were adjusted in the early 1920s, but for Boeke, who developed the new scholarly tradition of Indonesian economics, it was clear that the sugar factories thoroughly changed the social fabric of rural Java, undermining its economic basis. And this conclusion was drawn in the halcyon days of the Java sugar industry. The depression years were still to come. In spite of Boeke’s misgivings, the sugar industry absorbed large amounts of labor and even confronted labor shortages, which were addressed by a modicum of mechanization—haulage and to a lesser degree ploughing for example—and an optimal division of labor by gender and age. The problem was however that the sugar factories, as I have indicated, had to operate in a very competitive world market facing regular slumps, and that the general trend across several decades was one of declining prices. The economist van Gelderen—based at the statistical bureau in the Neth- erlands Indies in the 1920s, and since 1928 professor of economics in Bata- via (Jakarta)—postulated that from the early 19th century, every seventh or eighth year in the economic cycle exhibited a downturn, with all the consequences of economic misery and unemployment. This he attributed, like most contemporary economists, to the production cycle of industries. In 1930, these short-term downturns coincided with the lowest point in the Kondratiev cycle.54 But van Gelderen also perceived this slump as different from an ordinary cyclical situation—the economy, he considered, might not recover without government interference. It is no surprise therefore that Van Gelderen played an influential role in the group that wrote the Labor Plan (Plan van de Arbeid) published in 1935 by the Dutch

52 Creech Jones and Hinden, Colonies and International Conscience, pp. 4–5. 53 Verslag van de arbeidscommissie betreffende de wettelijke vaststelling van minimum- loonen voor werknemers op Java en Madoera (Batavia, s.a.). 54 Van Gelderen, Wereldcrisis, p. 7. 388 ulbe bosma

Labor Party, which was strongly Keynesian in its approach. The protectionist tide of the 1930s, and the Dutch government policy of clinging to the gold standard, was extremely detrimental to the Indonesian economy. In the chapter on Indonesia which Van Gelderen wrote for the Labor Plan, he diagnosed that deflation and underinvestment kept unemployment in the colonies at an appallingly high level, and had resulted in a serious contraction of the available currency in the Javanese village economies.55 The same was also signalled by Boeke as a severe and combined consequence of deflationary policies and a dualistic economy.56 The Dutch Labor Party and Arthur Lewis supplied by and large the same set of recommendations for respectively Java and the West Indies. One recommendation was to open up the markets of the industrialized countries to tropical commodities—while Lewis was hoping for the United States market, Van Gelderen expected a liberal attitude from Japan—and another recommendation was indigenous industrialization, in order to generate more and different sources of income.57 Both Lewis and Van Gelderen devoted considerable attention to the issue of taxation, to spread the profits of commodity production. In the 1930s, Lewis noticed that direct taxation was a good means to redistribute income, though he also admitted that it could only be applied with restraint, since it would deter foreign investment otherwise.58 In practice, vested interests in Barbados, St. Kitts and Antigua heavily reliant on sugar production had been able to keep wages down. The taxes which the island elites imposed were indirect taxes “to keep the burden of taxation mainly on the masses.”59 This of course was a direct consequence of the weak economic and political position of these masses. The pressure of metropolitan investors to keep taxes down had also been felt by Jacob van Gelderen, who in 1922 defended the existing progressive company profit tax in the Netherlands Indies against the interests of colonial

55 SDAP, Het Plan van de Arbeid (Amsterdam, 1935), p. 300. About Van Gelderen see Wicher Schreuders, “Jacob van Gelderen,” Biografisch Woordenboek van het Socialisme en de Arbeidersbeweging in Nederland 4 (1990), pp. 60–2, consulted at . Schreuders, “Jacob van Gelderen.” 56 Boeke, Economics and Economic Policy, p. 237. 57 Any hopes that Japan, that had a stake in the Java sugar industry would helping Java by opening up its markets in exchange for its own export of industrial goods to Java disappeared in 1937. See G. Roger Knight, “Exogenous Colonialism: Java Sugar between Nippon and Taikoo before and during the Interwar Depression, c. 1920–1940,” Modern Asian Stud- ies, 44, 3 (2010), pp. 477–515. 58 Lewis, Labour in the West Indies, p. 48. 59 Ibidem, p. 51. new thinking and the origin of development economics 389 enterprise. He also pointed out that the income on capital was excessively high compared to Europe, and that the land was leased by the plantations at rates which were ridiculously low.60 Being a government statistician, Van Gelderen also explored a demographic angle, reckoning that the average population density of Java in the late 1920s was 679 per square mile. That made Java one of the most densely populated places on earth, more than even England and Bengal, or Saxony and Rhineland, though less than Jiangsu (the coastal province of China where Nanking is located).61 The point was that all highly densely populated areas in the world were much more urbanized than Java. The stagnation of the non-agricultural economic sectors, combined with rapid demographic growth, had created abundant supplies of labor, which had brought Java to the top of world sugar production. Now, however, these factors were the primary cause of a prolonged crisis, precisely because any other sources of investment for diversification, let alone industrialization, were lacking. The only way out was to borrow and invest loans in the construction of public works, thus generating a multiplier effect. But such Keynesian advice found no sympathetic hearing among the top colonial officials in the Netherlands Indies.62 Any attempt to challenge colonial dualism was bound to fail, because, in the minds of the colonial bureaucrats and colonial enterpreneurs, the colonial order based upon economic laissez-faire was an unalterable condition. In reality, the economy of Indo- nesia slumped so badly that the population was virtually reduced to bare subsistence, in part utilizing abandoned plantation lands. Ironically, it was precisely the Java sugar industry that had produced exemplars for future economic policies, holding out the promise of a more balanced development. The Javanese sugar industry had been placed under the most stringent forms of national coordination, and since 1937 was monitored by an International Sugar Board. Moreover, the collapse of the Java sugar industry forced the colonial government of Indonesia to engage itself more and more with economic life. In the course of the 1930s, it even began to consider encouraging small industries, a policy that had been swept aside under the pressure of the powerful sugar industry

60 See J. van Gelderen, De theoretische grondslag der progressieve winstbelasting. Publi- catie No. 1 van de Vereeniging “De Taak” (Batavia-Weltevreden, 1922). 61 J. Van Gelderen, Western Enterprise and the Density of the Population in the Nether- lands Indies (Weltevreden, 1928), p. 89. 62 D.M.G. Koch, Verantwoording. Een halve eeuw in Indonesië (The Hague/Bandung, 1956), p. 175. 390 ulbe bosma in the 1920s, that was only interested in reducing wages. There was no doubt, Van Gelderen declared in 1939, that the “ideal of free exchange of products should be maintained,” but at the same time he hoped that the economic nationalism of the 1930s would provide the instruments for a more balanced progress in the future.63 One year after Van Gelderen wrote his English booklet about economic policies for the Netherlands Indies, Lewis had written his essay for Fabian Society on the labor movement in the West Indies. Both publications were guided by the traumatic experience of mass unemployment and hunger in the plantations societies, as a consequence of the sugar slump and collapse of the global sugar market. But an analysis of why the global sugar industry combined exceptionally high productivity gains with poverty for its workers, would be succinctly summarized only fifteen years later, in the 1950s, when Lewis published in his famous article on the unlimited supplies of labor: We have here the key to the question why tropical produce is so cheap. Take for example the case of sugar. This is an industry in which productivity is extremely high by any biological standard. It is also an industry in which output per acre has about trebled over the course of the last 75 years, a rate of growth of productivity which is unparalled by any other major industry in the world-certainly not by the wheat industry. Nevertheless workers in the sugar industry continue to walk barefooted and to live in shacks, while workers in wheat enjoy among the highest living standards in the world. The reason is that wages in the sugar industry are related to the fact that the subsistence sectors of tropical economies are able to release however many workers the sugar industry may want, at wages which are low, because tropical food production per head is low.64 Under conditions where colonial enterprise was not able to provide work, unlimited supplies of labor did not exist in many parts of Africa. But they did exist in India, Egypt and Jamaica. In his article on the unlimited supplies of labor, Lewis hinted at the reason: In the first place, an unlimited supply of labor may be said to exist in those countries where population is so large relatively to capital and natural resources, that there are large sectors of the economy where the marginal productivity of labor is negligible, zero, or even negative. Several writers have drawn attention to the existence of such “disguised” unemployment in the agricultural sector, demonstrating in each case that the family holding is so small that if some members of the family obtained other employment

63 Van Gelderen, Recent Development, p. 90. 64 Lewis, “Unlimited Supplies of Labor,” p. 140. new thinking and the origin of development economics 391

the remaining members could cultivate the holding just as well (of course they would have to work harder: the argument includes the proposition that they would be willing to work harder in these circumstances).65 In 1979, W. Arthur Lewis won the Nobel Prize mainly because of his article on the unlimited supplies of labor. To a certain extent, this award vindicated the work of his Dutch colleagues, who as early as the 1920s and 1930s, had discarded Ricardo’s theory of comparative advantage. The workers the Global South, in the words of Patrick O’Brien, had to work in the margins of a “restricted range of natural advantages allowed by their ecologies provided opportunities for more intensive participation in international trade.”66 Their opportunities were even more restricted by massive dumping practices and protectionist measures on behalf of the beet sugar industry in the North. This problem now became part of the debates on colonial development from the 1930s onwards. In that sense, the movement of the workers in the West Indies spearheaded a revolution, not only politically but also intellectually. The labor movement and the labor parties called for the application of Keynesian remedies of investments in public works and industrialization, to get out of the vicious circle of underutilization of labor. The Fabians and the Dutch Social Democrats clearly took a reformist stance, perceiving the future independence of the colonial dependencies as an outcome of economic and political reform, and not the other way round. By taking such a position, they laid the intellectual basis for the development policies in the first decades after 1945. After all, these policies did not threaten the economic hegemony of the United States and Europe. Cuba’s revolutionary trajectory, challenging the economic hegemony of the United States, was exceptional in that respect, a “singularity” that hinged on its capacity to regulate labor supplies at the expense of workers from the waning sugar colonies. In conclusion, the workers’ unrest in the Caribbean in the 1930s, as well as the issue of the relationship between the Java sugar industry and the native rural economy, became important building blocks for the development policies of the 1950s and 1960s, which advocated government planning and international coordination. The origins of this new era of “development” were intimately connected to the sugar crisis of the 1930s, which legitimated government intervention to create a social market economy.

65 Ibidem, p. 141. 66 Patrick Karl O’Brien, Colonies in a Globalizing Economy 1815–1948 (London School of Economics, Working Paper No. 08/04,2004), p. 32. sources

Unwritten Autobiography: Labor History Libraries before World War I

Jaap Kloosterman

The writer Walter Mehring inherited his father’s books in the First World War, and lost them in the Second. When he mentally reconstructed the library afterwards, he described the result as the “autobiography of a culture”. “Es lag mir nicht so sehr an den Büchern im einzelnen,” he explained, “sondern an jener historisch, ästhetisch, philosophisch einma- ligen Konfiguration wie sie sich in der Bibliothek meines Vaters, in seinem speziellen Horoskop des xix. Jahrhunderts eingestellt hatte”.1 This view of a library as telling a story is no longer just the fruit of a writer’s imagination. Since almost any collection of books—indeed, almost any collection—tends to appear as the product of deliberate choice, historians have become interested. They have begun to realize that “the history of book ownership, and the formation of collections, [. . .] provides a window into earlier tastes and fashions,” and they regret that in modern repositories, “collections, received over the years, have been divided up across the rest of the stock”.2 Why somebody systematically brought together certain books is in fact a question that might produce rather illuminating answers. Like the famous collections of early modern Europe, those of the 19th century laid the groundwork for new disciplines, among them labor history. The beginnings of the modern historiography of labor, labor conditions and labor movements are normally linked to its first institutionalization in the 1890s.3 Yet earlier in the century, “individual scholars had already eagerly scratched the surface” in what in hindsight is a “fuzziness [that]

1 Translated, “I was not so much attached to the details of the books themselves, but to that historically, aesthetically and philosophically unique configuration which the library of my father represented, as his special horoscope of the 19th century.”—Walter Mehring, Die verlorene Bibiothek: Autobiographie einer Kultur (Düsseldorf, 1978), p. 23. 2 David Pearson, “Introduction: From Texts to Collections,” in Giles Mandelbrote and Barry Taylor (eds), Libraries within the Library: The Origins of the British Library’s Printed Collections (London, 2009), pp. 1–7, at p. 3. 3 Lex Heerma van Voss and Marcel van der Linden (eds), Class and Other Identities: Gender, Religion and Ethnicity in the Writing of European Labour History (New York, 2002), introduction. 396 jaap kloosterman makes this ‘prehistory’ of labour history even more interesting”.4 The same can be said of the collectors who made research possible in the first place. These people had various motives, but seldom cared much about the past. Of course, some of them were historians; and organized labor sometimes wanted to know and preserve its history. Yet it was chiefly the contemporary problems associated with industrialization—the unequal distribution of the blessings of progress, the inadequacy of the schools, or the dangers of the new factories—that provided the incentive to collecting. Those problems elicited a response not just from socialists, but also from Christians and liberals, from employers driven by moral considerations or calculations of risk, from social scientists with new methods and instruments, and sometimes from a legislating state. In many cases, the response involved setting up a library for the convenience of researchers or the education of the public, often as part of one of the new museums that graced the century.5 Of the many stories contained in those collections, few have been written, and those sometimes unevenly. This essay seeks to identify the most promising candidates for a role in a future autobiography of labor history.6

Useful knowledge

The most conspicuous common denominator of labor history libraries was their aim to convey useful knowledge. Since the 18th century, “useful knowledge” covered a broad category of concepts ranging from the use of the empirical method in science to the need for vocational schools—and many other things that furthered economic rationality and avoided scho- lastic disputes. From the American Philosophical Society (1743) through Britain’s Society for the Diffusion of Useful Knowledge (1826) to the Ency-

4 Jan Lucassen, “Writing Global Labour History c. 1800–1940: A Historiography of Con- cepts, Periods and Geographical Scope,” in his (ed.), Global Labour History: A State of the Art (Bern, 2006), pp. 39–89, at p. 49. 5 Walter Hochreiter, Vom Musentempel zum Lernort: zur Sozialgeschichte deutscher Museen 1800–1914 (Darmstadt, 1994); Tony Bennett, The Birth of the Museum: History, The- ory, Politics (London, 1995); Dominique Poulot, Une histoire des musées de France: xviiie– xxe siècle (Paris, 2005). 6 It is based on my “Do Rjazanova: Razmyšlenija o pervych bibliotekach posvjaščennych rabočej istorii,” in I.Ju. Novičenko and I.B. Cvetkova (eds), Izvestnyj i neizvestnyj David Borisovič Rjazanov (1870–1938) (Moscow, 2011), which in turn borrowed from Jaap Kloost- erman and Jan Lucassen, Rebels with a Cause: Five Centuries of Social History Collected by the iish (Amsterdam, 2010), introduction. labor history libraries before world war I 397 clopédie des connaissances utiles of the 1830s, these ideas had fostered what amounted to an informal international movement, which set out to explain scientific principles and technological innovations to the public at large, including the working classes.7 Although serious moral concerns underlay this educational activism, it also sought to appeal to human curiosity. As Henry Brougham, one of the founders of the sduk as well as of the University of London, wrote in his preliminary treatise to the Library of Useful Knowledge, “The practical uses of any science or branch of knowledge are undoubtedly of the highest importance; and there is hardly any man who may not gain some positive advantage in his worldly wealth and comforts, by increasing his stock of information. But there is also a pleasure in seeing the uses to which knowledge may be applied, wholly independent of the share we ourselves may have in those practical benefits”.8 This was in fact one of the assumptions behind the many industrial exhibitions and World’s Fairs, which play an important supporting role in our story. The tradition of those exhibitions went back to the mid-18th century, and was particularly well established in France from 1798 on.9 It was relatively weak in Britain until the Royal Society of Arts started to organize an annual series in 1846. Interestingly, these exhibitions contained historical elements presented for emulation. Most notably, the event of 1850 showed medieval decorative art with a view to inform modern-day artisans in hopes of an improvement of British industrial design.10 This anticipated the medieval court which would be devised a year later by

7 Alan Rauch, Useful Knowledge: the Victorians, Morality and the March of Intellect (Durham, nc, 2001); and, from a different perspective, Joel Mokyr, The Gifts of Athena: Historical Origins of the Knowledge Economy (Princeton, nj, 2002). For a Russian example, see Joseph Bradley, Voluntary Associations in Tsarist Russia: Science, Patriotism, and Civil Society (Cambridge, ma, 2009), chap. 4.—The aps still has as its motto, “Promoting Useful Knowledge.” 8 [Henry Brougham], A Discourse of the Objects, Advantages, and Pleasures of Science, new ed. (London, 1828), pp. 4–5. Brougham typically demonstrated the usefulness of science by arguments like, “The art of good and cheap cookery is intimately connected with the principles of chemical philosophy, and has received much, and will yet receive more, improvement from their application” (ibid., p. 158). 9 For a bibliography, see Kenneth E. Carpenter, “European Industrial Exhibitions before 1851 and Their Publications,” Technology and Culture, 13 (1972), pp. 465–86. For the literature on the international fairs, see Alexander C.T. Geppert, Jean Coffey, and Tammy Lau, International Exhibitions, Expositions Universelles and World’s Fairs, 1851–2005: A Bibliogra- phy (2006), online at the Henry Madden Library of the California State University at Fresco (). 10 Elizabeth Bonython and Anthony Burton, The Great Exhibitor: The Life and Work of Henry Cole (London, 2003), pp. 99–103. 398 jaap kloosterman

Augustus Pugin for the Great Exhibition of the Works of Industry of All Nations, held at the Crystal Palace constructed for the purpose in Hyde Park, London. In this way, the first World’s Fair revealed an ambiguity that would prove divisive: was the knowledge behind the artisan’s craft “useful” in the same way as that underpinning the steam engine? The two very different answers given to that question both provided an incentive to collecting. Those favoring the skills of the craftsman over the machine could find support in the new interest for folklore. Although this term was introduced only in 1846, enthusiasm for the creativity and wisdom of the common people was of course much older. The 19th-century folklorists established important museum collections and pioneered new ways to display them—for example, the “open-air museums” first created in Scan- dinavia and linked to the name of Artur Hazelius, or the “period rooms” that appeared in many European exhibitions and museums at the end of the century.11 The definition of folklore was extraordinarily broad, and encompassed such immaterial topics as dances, songs, fairy tales, dialects, or recipes. There was inevitably much of interest to labor history, if only because of efforts to study traditional crafts and customs, and preserve pre-industrial tools. Those who were more impressed by the progress of science and industry were interested in another kind of museum, whose prototype was the Conservatoire des Arts et Métiers founded by l’abbé Henri Grégoire in Paris in 1794, on the principle, “Il faut éclairer l’ignorance qui ne connaît pas, et la pauvreté qui n’a pas le moyen de connaître”.12 The institution received the mission “de recueillir les machines, outils, desseins, descriptions, procédés relatifs au perfectionnement de l’industrie, et d’en répan- dre la connaissance dans toute l’étendue de la République.”13 It could start its work on the basis of a number of existing collections, including the

11 Sten Rentzhog, Open Air Museums: The History and Future of a Visionary Idea (Öster- sund, 2007); Adriaan A.M. de Jong and Mette Skongaard, “The Hindeloopen and the Amager Room: Two Examples of an Historical Museum Phenomenon,” Journal of the His- tory of Collections, 5 (1993), pp. 165–78. 12 Translated, “We must enlighten ignorance which does not know, and poverty which has no means of knowing.” 13 Translated, “To collect the machines, tools, designs, descriptions, and processes related to the development of industry, and to spread this knowledge throughout the whole of the Republic.”—Alain Mercier, Un conservatoire pour les arts et métiers (Paris, 1994), p. 20; Josiane Boulad-Ayoub, L’abbé Grégoire apologète de la République (Paris, 2005), p. 104.—In 1900, the Conservatoire created the first French chair for l’histoire du travail, occupied by Georges Renard. labor history libraries before world war I 399 various machines and prototypes left to Louis xvi by Jacques de Vaucan- son. A product of the French Revolution, the Conservatoire was from the beginning conceived as a government-funded school. The modernist currents that arose in the mid-19th century were of a different character. A typical representative was Thomas Twining iii, a scion from a wealthy family of tea merchants and high-ranking British-Indian officials. In his youth, part of which he spent on the Continent, he had become partly blind and disabled, but he was greatly interested in technical education and more generally in applying modern science for the benefit of workers, their health, and their quality of life. At the time of the Great Exhibition, he floated the idea of a “working man’s museum”, and he succeeded in having a section on “domestic economy” added to the Paris World’s Fair of 1855, which included “such articles as their cheapness, combined with good quality, may render particularly useful to a simple home”.14 In the same year, he pleaded for a more permanent exhibition in a pamphlet entitled Special Museums for the Working Classes; and in 1860 he opened his own Museum of Domestic and Sanitary Economy at his estate in Twickenham, near London. Twining labelled his ideas “practical bionomy” or the “science of common life”. This not only included knowledge about “the things to be adopted or eschewed” in the working man’s household, but also insight into those principles of physics, chemistry, physiology and natural history that were needed to understand “the reason why”.15 One problem he evidently encountered was the lack or insufficiency of the foundation on which to build, and his bionomy was also intended as a form of primary education. The Twickenham Economic Museum, as it was known, sought to realize this program not only in exhibitions, but also through free popular lectures and a library. Unfortunately, it burned down in 1871. The books were saved, but dispersed after Twining’s death in 1895. The museum had to decide how to bring order to its holdings. Similar questions had already been raised at the Great Exhibition, when countless exhibits had to be sorted and the scientist Lyon Playfair “brought Prince

14 B.L. Pearce, Thomas Twining of Twickenham: His Work, his Museum, and the Perryn House Estate (Twickenham, 1988), p. 9.—The collection exhibited in Paris ended up in the South Kensington Museum, now the Victoria & Albert, built from the proceeds of the Great Exhibition. 15 Thomas Twining, Science for the People: A Memorandum on Various Means for Propa- gating Scientific and Practical Knowledge among the Working Classes, and for Thus Promot- ing Their Physical, Technical, and Social Improvement (London, 1870), pp. 10–1. 400 jaap kloosterman

Albert, who had firm notions about the philosophical basis of classification, to agree on a system similar to that used by botanists to classify plants”.16 In the case of the Twickenham library, the solution was found in a catalogue produced as a guide for readers that was at the same time systematic and moral.17 It nicely illustrates the interest of such early catalogues, of which unfortunately few remain: mostly self-designed, they both expressed and shaped the contours of the new field.18

Safety first

In addition to specifically economic and scientific topics, the Twickenham catalogue contains headings for hygiene, accidents, “industrial pathology”, public health, temperance, and the like. This complex, with its focus on safety, was another important factor in the establishment of collections— again, often in the guise of a museum equipped with a specialized library and naturally animated by a strong educational urge. In many cases, the initiative came from businessmen, or from the factory and labor inspectors who appeared on the scene with the introduction of appropriate legislation in some European countries. Although there were predecessors, especially in Germany,19 perhaps the most prominent example in this field was the Gewerbe-hygienisches Museum of Vienna. It was largely the work of Franz Migerka, who already in 1857 had written about the usefulness of industrial exhibitions and, a

16 Bonython, Burton, Great Exhibitor, p. 130. See also Steve Edwards, “The Accumulation of Knowledge or William Whewell’s Eye,” in Louise Purbrick (ed.), The Great Exhibition of 1851: New Interdisciplinary Essays (Manchester, 2001), pp. 26–52. 17 Handbook of Economic Literature, Being a Descriptive Catalogue of the Library of the Twickenham Economic Museum, or Repertory of Useful Knowledge for Every-day Life: Part i: Domestic and Sanitary Economy (London, 1862). There were no further parts. 18 Library standardization was just beginning. Antonio Panizzi’s “ninety-one rules” for the alphabetical catalogue were issued in 1841; Charles Cutter’s rules for the dictionary catalogue, in 1876; the Instructions élémentaires et techniques of Léopold Delisle, in 1890; the Preussische Instruktionen, in 1899. In subject cataloguing, Melvil Dewey’s decimal classification dates from 1876, the Library of Congress Classification, from 1897. In 1895, Henri de la Fontaine and Paul Otlet created the Office international de Bibliographie, which designed the Universal Decimal Classification.—Systematic studies of collecting, such as David Mur- ray’s Museums: Their History and Their Use (Glasgow, 1904) or Julius von Schlosser’s Die Kunst- und Wunderkammern der Spätrenaissance: ein Beitrag zur Geschichte des Sammel- wesens (Leipzig, 1908), begin to appear at the turn of the century. 19 See Olaf Hartung, Museen des Industrialismus: Formen bürgerlicher Geschichtskultur am Beispiel des Bayerischen Verkehrsmuseum und des Deutschen Bergbaumuseums (Köln, 2007), chap. 2. Many of the most important special libraries in 19th-century Germany were technical, see Wolfgang Schmitz, Deutsche Bibliotheksgeschichte (Bern, 1984), pp. 138–9. labor history libraries before world war I 401 decade later, launched the idea of a museum dealing with health and safety issues in the workplace. Together with his wife, Katharina, he founded the first domestic science schools for Austrian housekeepers, and created a special exhibition on women’s work as part of the Vienna World’s Fair of 1873.20 Migerka, who was Austria’s representative to the Philadelphia fair of 1876, became central factory inspector in 1883, and continued to enlarge the collection he had built. He initially presented part of it in the Tech- nologisches Gewerbe-Museum founded in 1879, on the model of the Paris Conservatoire, by Wilhelm Exner. An engineer and politician interested in the history of technology, Exner had already advocated the creation of an Austrian “trades museum” after the Paris fair of 1867, and been involved in the Viennese exhibition of 1873. Yet, in 1890 Migerka finally succeeded in opening an independent museum for industrial safety and the preven- tion of accidents. It set up its own library, and in 1898 published its first catalogue.21 The museum attracted much interest. In 1893, Dutch industrialists opened a similar institution in Amsterdam, which came to be known as the Veiligheidsmuseum, after having sent the engineer H.W.E. Struve to study the Viennese collections.22 A German by birth, Struve was the co- author of an important statistical report on industry in the Netherlands, and one of the first labor inspectors in the country. He became a long- time vice-president of the museum, which also established a specialized library.23 In 1901, the Dutch example in turn inspired a group of Canadians led by Louis Guyon, an early factory inspector in Quebec, to organize an exhibition in Montreal, which claimed to be the first in North America.24 In 1903, Berlin-Charlottenburg saw the opening of the Ständige Ausstellung

20 Jutta Pemsel, Die Wiener Weltausstellung von 1873: das gründerzeitliche Wien am Wendepunkt (Wien, 1989), p. 74. The “Pavillon der Frauenarbeit” was a “first” in the world according to Maria Grever, “Reconstructing the Fatherland: Comparative Perspectives on Women and 19th Century Exhibitions,” in Maria Grever and Fia Dieteren (eds), Een Vaderland voor Vrouwen: A Fatherland for Women: The 1898 “Nationale Tentoonstelling voor Vrouwenarbeid “in Retrospect (Amsterdam, 2000), pp. 13–29, at p. 17. 21 Stefan Poser, Museum der Gefahren: die gesellschaftliche Bedeutung der Sicherheits- technik: das Beispiel der Hygiene-Ausstellungen und Museen für Arbeitsschutz in Wien, Berlin und Dresden um die Jahrhundertwende (Münster, 1998). 22 Struve produced a Lijst der voorwerpen van het Gewerbe-hygienische Museum te Wee- nen opgenomen 18–22 April 1893 of interest to the Dutch. Migerka became a “corresponding member” of the Veiligheidsmuseum. 23 Its books and posters were later transferred to the International Institute of Social History (iish). 24 Jean-Claude Dionne, “Documents pour l’étude des expositions et musées pour la pré- vention des accidents et des maladies du travail au Québec au début du siècle,” Labour/le Travail, 40 (1997), pp. 199–211, at pp. 202–4. 402 jaap kloosterman für Arbeiterwohlfahrt, while the Paris Conservatoire from 1905 on housed the Musée de la Prévention des Accidents et de l’Hygiène industrielle. In the United States, the Social Gospel movement added workplace accidents, which were growing alarmingly, to the long list of social evils that it was fighting. A major role was played by the League for Social Ser- vice founded in New York in 1898 by Josiah Strong, a pioneer and leader of the movement, and William H. Tolman, a former minister. They were no socialists—in Strong’s view, socialism “attempts to solve the problem of suffering without eliminating the factor of sin”—but were convinced that it would be in everybody’s interest to improve working conditions.25 The League arranged an American exhibit at the “social economy” section of the Paris World’s Fair of 1900, featuring the welfare programs of H.J. Heinz and the National Cash Register Co., among others.26 In 1906–1907, it established the American Museum of Safety, with money from the steel industry and Tolman as its director. Among the services it offered was, again, a library.27

Social economy

When the League for Social Service was still groping for the right institutional form of its activism, Tolman was reported as trying to establish a “social museum” or “museum of social economy” in New York, on the model of the Musée Social in Paris, of which he was the American cor- respondent.28 This shows how the themes of useful knowledge and safety in the workplace were continuous with the equally loose notion of “social economy,” which since the days of Jean Simonde de Sismondi constituted a critical counterpart to classical economics, and found its practical expression in such workers’ organizations as co-operatives. It also intro- duces, with the Musée Social, one of the most influential early initiatives in labor history collecting.

25 Charles Howard Hopkins, The Rise of the Social Gospel in American Protestantism 1865–1915 (New Haven, ct, 1940), p. 75; Dorothea R. Muller, “The Social Philosophy of Josiah Strong: Social Christianity and American Progressivism,” Church History, 28 (1959), pp. 183–201. 26 On this occasion, Tolman published Industrial Betterment, with a French translation, Progrès industriel. 27 Hopkins, Social Gospel, pp. 259–262; Daniel Nelson, Managers and Workers: Origins of the New Factory System in the United States 1880–1920 (Madison, wi, 1975), pp. 32–3, 109–11.—In 1906, United States Steel had instigated the Safety First movement. 28 New York Times, 13 April 1902. labor history libraries before world war I 403

In France, one major exponent of social economy was Frédéric Le Play, a widely-travelled mining engineer who in 1855 published Les Ouvri- ers européens, and is today seen as an early sociologist. He was also the organizer of the 1855 Paris World’s Fair. Le Play left the structuring of the exhibition’s section on “domestic economy” proposed by Twining to the care of Augustin Cochin, who co-operated with other liberal Catholics such as Armand de Melun, the organizer of an international charity meeting that took place during the fair.29 On the basis of these contacts, Le Play founded the Société d’Economie sociale, which aimed at the scientific study of manual workers and their living conditions, and counted among its members men like the engineer and economist Michel Chevalier, the banker Emile Pereire, and Louis-René Villermé, a pioneer of occupational medicine.30 They represented the various currents behind the Musée Social—former Saint-Simonians, the emerging “science of society” with its interest in statistics, Protestantism. The efficient cause of the museum was an effort to keep together the documentation assembled in the social economy section of the Paris World’s Fair of 1889. This collection was transferred to—and still is at—5, rue Las-Cases, in the seventh arrondissement of Paris, the home of Count Aldebert de Chambrun. Chambrun was married to Marie-Jeanne Godard- Desmarest, heiress to the Baccarat crystal fortune; together they had decided to spend their wealth “pour Dieu et pour les pauvres”. After the death of his wife in 1891, Chambrun endowed chairs of social economy at the old Sorbonne (the sociologist Alfred Espinas) and the young Ecole libre des Sciences politiques (Emile Cheysson, an engineer and statistician who had worked with Le Play on the World’s Fair of 1867), and pondered the foundation of an institute of sociology. In 1894, however, he funded the creation of the Musée Social instead. Among his co-founders were Cheys- son, the economist and politician Léon Say (a grandson of Jean-Baptiste), and Jules Siegfried, a rich social entrepreneur and politician. The museum

29 Among the participants was another early sociologist; Anne Lhuissier, “Alexis de Tocqueville et l’économie sociale chrétienne: sociétés alimentaires et classes ouvrières,” Genèses, 37 (1999), pp. 135–55. 30 Maguelone Nouvel, Frédéric Le Play: une réforme sociale sous le Second Empire (Paris, 2009). See also the various studies of Antoine Savoye on Le Play and the beginnings of sociology.—The full name of the organization was Société internationale des Etudes pratiques d’Economie sociale. Twining was a member. Le Play would also organize the Paris World’s Fair of 1867. 404 jaap kloosterman was patronized by the elite of the Third Republic, and is now the oldest existing institution collecting in the field of social history.31 According to its rules, the Musée Social aimed “de mettre gratuitement à la disposition du public, avec informations et consultations, les documents, modèles, plans, statuts, etc., des institutions et organisations sociales qui ont pour objet et pour résultat d’améliorer la situation matérielle et morale des travailleurs”, while at the same time “s’interdit toutes discussions politiques et religieuses”.32 It gave special attention—“soin tout particulier”—to setting up a library, which in 1896 featured around 6,000 volumes and several dozens of serial subscriptions from various countries.33 It developed auspiciously, and became a model that was widely adopted abroad. In the Netherlands, Arnold Kerdijk, a progressive liberal, in 1899 succeeded in establishing the Centraal Bureau voor Sociale Adviezen, which like the Musée Social offered legal and practical advice for the creation of workers’ associations in several spheres, while upholding strict political neutrality.34 The first board included Jacques van Marken, a social entrepreneur in the food and chemical industry who shared an interest in “social engineering” with Cheysson, as well as Alfons Ariëns, an important organizer of the Roman Catholic workers, Sybrand Talma, a Protestant labor leader, and Willem Vliegen, one of the founders of the Dutch Social- ist Party. The liberal economist Willem Treub, later a cabinet minister, became director of the cbsa office, to be succeeded in 1905 by the social democrat Dirk Hudig. The organization sought to write the history of the labor movement, for which it set up a “document commission” consisting

31 Janet R. Horne, A Social Laboratory for Modern France: The Musée Social and the Rise of the Welfare State (Durham, nc, 2002); Colette Chambelland (ed.), Le Musée social en son temps (Paris, 1998); Christian Topalov (ed.), Laboratoires du nouveau siècle: la nébuleuse réformatrice et ses réseaux en France, 1880–1914 (Paris, 1999). 32 Translated, “To make freely available to the public, with information and consultation, the documents, models, plans, laws, etc., of the institutions and social organizations which have as their purpose and result the improvement of the material and moral situation of the workers,” while it at the same time “prohibited all political and religious discussions.” 33 Le Musée social [. . .]: statuts–organisation–services (Paris, 1896), pp. 13, 40–47; Horne, Social Laboratory, pp. 154–60. 34 A circular letter (5 July 1899) situated the cbsa “in the footsteps of the ‘Musée Social’ “in France and of the ‘Centralstelle für Wohlfahrtseinrichtungen’ in Duitsland” (iish, cbsa records, 2).—The Centralstelle, born in 1891, co-founded the Ständige Ausstellung in Char- lottenburg; see Poser, Museum der Gefahren, part c. labor history libraries before world war I 405 of representatives of the Dutch ideological “pillars”. The cbsa also built a library, which issued several catalogues from 1905 on.35 In Germany, Peter Schmidt, the librarian of the Statistical Bureau of Saxony, wrote a well-documented article about the Musée Social in the journal of the social-reformist Centralverein für das Wohl der arbeitenden Klassen.36 The Austrian-Hungarian journalist and peace activist Leopold Katscher wrote a pamphlet on the international impact of the Paris example, which was reprinted within the year.37 In Switzerland, the Zürich minister Paul Pflüger, a social democrat who in 1900 travelled to Paris to see both the new World’s Fair and the Musée Social, set up the Zentralstelle für soziale Literatur der Schweiz, which would develop into the country’s foremost labor history center.38 In Italy, Luigi Einaudi published an article on the Musée Social on the pages of la Stampa in 1898, in which he emphasized the richness of its library.39 Earlier, in 1893, the fortune left by Prospero Moisè Loria, a Jew- ish entrepreneur and a friend of the socialist Osvaldo Gnocchi Viani, had served to create the Società Umanitaria in Milan. This institution of broad social and educational reform in 1906 founded the Museo Sociale, which typically established a library on social problems.40 So did the Museo Social that operated in Barcelona from 1909 on; Cebrià de Montoliu, a reformer and translator of John Ruskin, became its librarian.41 In the United States, Francis G. Peabody, another leader of the Social Gospel movement, in 1903–1907 founded the Social Museum at Harvard

35 The library and the cbsa records are now at the iish.—Hudig in fact wrote a history of Dutch trade-unionism. 36 P. Schmidt, “Sociale Museen,” Der Arbeiterfreund, 37 (1899), pp. 268–90. 37 Leopold Katscher, Die sogenannten “Sozial-Museen”: Museen für Arbeiterwohlfahrt und Sozialpolitik und das Pariser “Musée social” als Vorbild (Leipzig, 1904). 38 Jacqueline Häusler, 100 Jahre soziales Wissen: Schweizerisches Sozialarchiv 1906–2006 (Zürich, 2006). 39 Cristina Accornero, “Metodo positivo, musei e laboratori: un esempio di interdisci- plinarietà nella Torino positivista di inizio secolo: il Laboratorio di Economia politica e il Regio Museo Industriale” (2004), p. 13, “Sulla circolazione internazionale delle idee: il Laboratorio di Economia Politica e il Musée Social di Parigi” (2005), both papers online at the Università di Torino ( and , respectively).—The Museo Industriale had been founded in Turin in 1862, in part to accommodate materials from that year’s World’s Fair in London. 40 Riccardo Bauer, La Società Umanitaria (Milano, 1958), pp. 26–7. An early overview of the collection is Il primo fondo librario del Museo Sociale (Torino, 1906). 41 José A. Benevent Oltra, “Del ‘Museo Social de Barcelona’ al ‘Institut Psicotècnic de la Generalitat de Catalunya’: origen, evolución y desaparición de una institución pionera y modélica de orientación psicopedagógica (1909–1939),” Revista Española de Orientación y Psicopedagogía, 19 (2008), pp. 79–100, 212–34, at p. 82. 406 jaap kloosterman

University, where he taught theology and lectured on social ethics. In The Social Museum as an Instrument of University Teaching, he explicitly praised the “admirable library of reference” of the Musée Social.42 In Bue- nos Aires, Tomás Amadeo, an agricultural engineer who had visited the Musée Social, established in 1911 the Museo Social, which would organize the first meetings of mutual benefit societies, co-operatives and housing associations in Argentina.43 In 1895, the Musée Social had sent a team of researchers across the Chan- nel to study the British trade union movement; they inevitably talked to Sidney Webb. This was precisely around the time when Webb, together with other members of the Fabian Society, was busy founding the Lon- don School of Economics, with money left by another Fabian, Henry Hunt Hutchinson. Webb, who was of the opinion that “anyone who studied the facts of economics was bound to become a socialist”, had in mind something “along the lines of the Ecole Libre des Sciences Politiques in Paris”.44 The Sciences Po, as it is known, had been established by the sociologist Emile Boutmy in 1872 with the support of a group of (often Protestant) businessmen and intellectuals, in response to the double catastrophe of the Franco-Prussian War and the Paris Commune.45 The school, which had a firmly international outlook, was designed as a center of history and the social sciences, while securing the collaboration of politicians and top civil servants as teachers. It became one of the great French Ecoles; it also built one of the world’s largest libraries in political science, broadly conceived. Immediately after the successful foundation of the lse, Webb set out to build just such a library in London. In April 1896, he explained why it was needed, arguing that even “the magnificent library of the British Museum” could not be expected to collect “the publications of such important

42 Francis G. Peabody, The Social Museum as an Instrument of University Teaching [ followed by] A Classified List of the Collections in the Social Museum of Harvard University to January, 1908 (Cambridge, ma, 1908), p. 4, online at Harvard University; the first part is translated from Internationale Wochenschrift, 23 November 1907.—Peabody knew Europe well; see Jurgen Herbst, “Francis Greenwood Peabody: Harvard’s Theologian of the Social Gospel,” Harvard Theological Review, 54 (1961), pp. 45–69. 43 Hebe Carmen Pelosi, El Museo Social Argentino y la Universidad del Museo Social Argentino: historia y proyección (1911–1978) (Buenos Aires, 2000). 44 Sydney Caine, The History of the Foundation of the London School of Economics and Political Science (London, 1963), p. 29; Ralf Dahrendorf, lse: A History of the London School of Economics and Political Science 1895–1995 (Oxford, 1995), p. 4. 45 Pierre Favre, Naissances de la science politique en France (1870–1914) (Paris, 1989), chap. 1.—Boutmy was on the board of the Musée Social. labor history libraries before world war I 407 voluntary development as Trade-Unionism, Friendly Societies, and Co-operation” and many documents from abroad. He noted that other countries were better equipped, citing the example of France and Cham- brun’s support “for the establishment of a library and center of research in Paris (Le Musée Social), not unlike what is now proposed for London”. In November 1896, the British Library of Political Science was officially opened, to become another magnificent library.46

Historians

The Library Committee of the blps counted among its first members Herbert Somerton Foxwell, a British economist and statistician. It was an excellent choice. Foxwell was, in the words of John Maynard Keynes, “seriously addicted to bibliophily, indeed to bibliomania, as such”.47 Although he was the successor of his friend Stanley Jevons in the Economics chair of the University of London, he is remembered above all for the more than 70,000 volumes he is estimated to have collected during his life.48 As a matter of fact, he tended to buy more books than he could afford, and had to sell his library twice—each time using part of the proceeds of the sale to start all over again. It first happened in 1901, when he sold over 30,000 titles on economic history to the Goldsmiths’ Company, which deposited them at the University of London. For thirteen years after, the old craft guild financially enabled Foxwell to make further acquisitions, until he fell out with the university and never set foot in the library again. The collection he amassed from then on he later sold to Harvard, to be transferred there after his death. It is now at the Kress Library of Harvard Business School, so named after Claude W. Kress, the philanthropist who

46 Caine, Foundation, chap. 6. 47 John Maynard Keynes, Essays in Biography (London, 1981), pp. 267–96, at p. 289. For more on Foxwell’s “magpie habits,” see R.D. Freeman, “The R.D. Freeman Collection of Foxwell’s Papers: Its Rescue,” Journal of the History of Economic Thought, 28 (2006), pp. 489–95, at p. 493. 48 This exceeds the 50,000 books left in 1872 by Thomas Phillipps, whose collection is considered the largest ever brought together by a single person. Admittedly, Sir Thomas left an additional 60,000 manuscripts as well as innumerable paintings, drawings, maps, etc.; see Werner Muensterberger, Collecting: An Unruly Passion: Psychological Perspectives (New York, 1994), chap. 6. The social democrat P.A. Pijnappel, who left 70,000 volumes to the library of the University of Amsterdam in 1936, likely holds the record for the largest single bequest. 408 jaap kloosterman funded the purchase from Foxwell.49 Together, the holdings of the Gold- smiths’ and Kress libraries constitute the best economic-historical collection in the world. Meanwhile, in some places, historians were successfully trying to persuade industrialists that historical collections might be useful. This manifested itself first of all in Germany, where Armin Tille had called for a better understanding of the country’s economy and a deeper analysis of its possible future development, exclaiming, “Der Vorschlag lautet einfach: Gründet Wirtschaftsarchive!”50 In reply, the Chambers of Commerce of the Rheinland and Westphalia set up the Rheinisch-Westfälisches Wirtschafts- archiv in Cologne in 1906. This was an independent repository of business archives and other documents related to the economic history of the region; it was entirely funded by the business sector itself. The birth of the rwwa took place in the context of a whole series of similar foundations such as Cologne’s Handelshochschule (1901), the Kolonialinstitut in Hamburg (1908), and the Institut für Seeverkehr und Weltwirtschaft at the University of Kiel (1914). Later in 1906, another economic archive was established in Saarbrücken; and the German example was soon followed by the Schweizerisches Wirtschaftsarchiv in Basle (1910) and the Nederlands Economisch-Historisch Archief in The Hague (1914). The institute in Kiel would eventually grow into one of the largest economic libraries in the world; the neha’s library became one of the best in the area of economic history.51 The printed collections of the German archival institutions were mostly much smaller. This didn’t make them less valuable, however, since they typically held materials that were not collected elsewhere, such as annual reports, company histories and trade periodicals.52

49 Ruth R. Rogers, “The Kress Library of Business and Economics,” Business History Review, 60 (1986), pp. 281–8. 50 Armin Tille, Wirtschaftsarchive (Berlin, 1905), p. xiii; see Klara van Eyll, Vorausset- zungen und Entwicklungslinien von Wirtschaftsarchiven bis zum Zweiten Weltkrieg (Köln, 1969). 51 Frieda Otto, “Die Bibliothek des Instituts für Weltwirtschaft: von der wirtschaftswis- senschaftlichen Seminarbibliothek zur sozialwissenschaftlichen Forschungsbibliothek,” in Anton Zottmann (ed.), Institut für Weltwirtschaft an der Universität Kiel 1914–1964 (Kiel, 1964), pp. 67–115; J.J. Seegers, “‘De bibliotheek mag zich voortdurend meer in de belangstel- ling verheugen, zoowel wat bezoek als wat de schenkingen betreft’: 75 jaar Economisch- Historische Bibliotheek,” in E.J. Fischer, J.L.J.M. van Gerwen, and J.J. Seegers (eds), De Vereeniging het Nederlandsch Economisch-Historisch Archief 1914–1989 (Amsterdam, 1989), pp. 67–93. 52 Ulrich S. Soénius, Zukunft im Sinn—Vergangenheit in den Akten: 100 Jahre Rheinisch- Westfälisches Wirtschaftsarchiv zu Köln (Köln, 2006), pp. 60–4. labor history libraries before world war I 409

As a rule, 19th-century libraries refused to accept “grey”, let alone “sub- versive” documents, and institutions that collected printed matter on a large scale and without too much bias, such as the British Museum since the days of Panizzi, were rare. From this perspective, the development of private collections was only natural. In a typical example, when H.P.G. Quack set out to write his six-volume history of socialism in 1875, he soon discovered that he would have to gather the sources himself. He left the University of Utrecht and took a better-paid job at the Dutch central bank in order to pay for his travel through Europe, to buy the materials he could not obtain in any library.53 In Holland at least, the remedy for the ill came when Quack donated his books to the University Library of Amsterdam in 1912, but it was quite common for a private library to be sold or dispersed at an auction after the death of its owner. And anyway, private libraries were not normally accessible to third parties; indeed, precisely because they were valuable, they often constituted the jealously-guarded working capital of the scholars who had brought them together, and whose reputation might be closely related to the reputation of their treasures. At any rate, before the First World War, it was unusual for such collections to become available in public institutions. There were exceptions other than Quack. When the writer Rossend Arús, a republican and freemason, died in 1891, he left his house and his books to the people of Barcelona, who could choose from 24,000 volumes when the Bibliotéca Pública Arús opened there in 1895.54 Likewise, Jules Perrier, who fled to Geneva after the Paris Commune, made the bequest of his library to the Bibliothèque publique et universitaire, although after he died in 1904 the bpu in a Pavlovian reflex threw away much that it found of no interest.55 In 1906, the collection of George Jacob Holyoake, the pioneer of the British co-operative movement, was divided by his widow between the library of the Co-operative Union in Manchester, which was in fact established on the basis of this gift, and the Bishopsgate Institute in London, founded in 1895 by William Rogers, the rector of St Botolph’s

53 H.P.G. Quack, Herinneringen uit de eerste veertig levensjaren (Amsterdam, 1907), pp. 278–81. 54 Jordi Galofré, Rossend Arús i Arderiu (1845–1891) (Barcelona, 1989), chapters 8 and 13. 55 Marc Vuilleumier, “Un donateur oublié: le citoyen Jules Perrier,” “La Collection Jules Perrier: manie de collectionneur ou passion révolutionnaire?,” Musées de Genève, 1, no. 5 (May 1960), pp. 7–9, no. 9 (October 1960), pp. 15–17; Maria Hunink, “Das Schicksal einer Bibliothek: Max Nettlau und Amsterdam,” International Review of Social History, 27 (1982), pp. 4–42, at pp. 13–4. 410 jaap kloosterman church.56 Other libraries brought together in these decades were purchased after the First World War by a number of new institutions with money to spend.57 The books of the law historian Otto von Gierke and the economist Carl Menger went to Japan; those of Lucien Descaves on the Paris Commune, Max Nettlau on anarchism, and Henri Rollin on the French Revolution, to Amsterdam; and those of the sociologist Carl Grünberg and the lawyers Theodor Mauthner and Wilhelm Pappenheim, to Moscow. A precondition for these private collections was the existence of a mature market for old books. From the mid-19th century on, Europe’s great capitals all witnessed the rise of competent antiquarian booksellers issuing specialized catalogues, from Stargardt in Berlin through Clavreuil in Paris to Quaritch in London—to name a few that still exist. Around the turn of the 20th century, Paris had the “damals noch so reichen Quais”.58 The British Antiquarian Booksellers Association, the oldest of its kind, was founded in 1906. To study its members’ catalogues, Foxwell told Keynes, “was a liberal education—‘they know so much more than I do’. [. . .] In the last year of his life, he said to me with a smile that he had got to the point where he much preferred the catalogues to the books”.59

The labor movement

In the labor movement, migrant organizations were among the first to set up their own libraries. As a famous example, in 1846, the German workers educational society in London had “etwa 500 Bänden, ferner haben wir

56 Roy Garratt, “The Co-operative Union Library,” Manchester Region History Review, 5 (1991), pp. 35–9; Raphael Samuel, “The Bishopsgate Institute,” History Workshop, 5 (1978), pp. 163–73. 57 Notably the Bibliothèque-Musée de la Guerre (1917), the Hoover Institution on War, Revolution and Peace (1919), Tokyo University of Commerce (1920), the Marx-Engels Institute (1921), and the iish (1935). 58 Rudolf de Jong, “Biographische und bibliographische Daten von Max Nettlau, März 1940,” International Review of Social History, 14 (1969), pp. 444–82, at p. 453. Nettlau’s unpublished memoirs, held at the iish, afford exceptional insight into the ways and means of collectors in the fin-de-siècle. A well-known antiquarian bookseller of Berlin similarly called those years the “goldene Zeitalter der Bibliophilie” (‘golden era of bibliophily’), but also noted, “In dieser Zeit war die Zahl der Liebhaber und Kenner schöner und seltener Ausgaben buchstäblich an den Fingern abzuzählen” (‘Those days the number of lovers and connoisseurs of rare and beautiful editions could literally be counted on one’s fingers.’) Hugo Streisand, Ein halbes Jahrhundert mit Büchern 1901–1951 (Berlin, 1951), p. 57. 59 Keynes, Essays, p. 291. Foxwell died in 1936. labor history libraries before world war I 411

Erd- und Himmelsgloben, Landkarten, deutsche, französische, englische und skandinavische Zeitungen”.60 Such collections were usually a mix of political and historical works, popular science, and classic and progressive belles-lettres. Russian migrants routinely created them. Paris came to host two of the largest—the Russian Library, later named after the writer Ivan Turgenev, who had funded it in 1875; and the library of the Party of Socialists-Revolutionaries, known as the Lavrov-Goc collection, after two of its major benefactors.61 Similar institutions existed in Germany and Swit- zerland, especially in university towns.62 The most prestigious movement library was undoubtedly that of the most prestigious movement, the Socialist Democratic Party of Germany. It had sprung from an appeal by August Bebel in early 1878, and because of Bismarck’s anti-socialist laws was originally built in exile from a stream of donations from party members solicited by Hermann Schlüter, who ran a bookshop in Zürich-Hottingen. First taken to London in 1888, the Parteiarchiv—the terms “archive” and “library” were used interchange- ably—could finally be transferred to Berlin at the beginning of 1891.63 It was located at the Öffentliche Bibliothek und Lesehalle founded by Hugo Heimann, a well-to-do social democrat who was inspired by the example of the British public libraries. Its opening in 1899 was an instant success: at the dawn of the new century, the number of users already exceeded 100,000 a year. Yet the Parteiarchiv proper, which incorporated the private libraries of Karl Marx and Friedrich Engels, could only be consulted with the permission of the party’s executive. Its more than 8,000 volumes, separately kept, were made accessible through a printed alphabetical

60 Armin Matthäus Kuhnigk, Karl Schapper: ein Vater der europäischen Arbeiterbe- wegung (Bad Cambach, 1980), p. 86. Lists of books from 1845 and 1864 survive; Jacques Grandjonc, Karl-Ludwig König, and Marie-Ange Roy-Jacquemart (eds), Statuten des “Com- munistischen Arbeiter-Bildungs-Vereins” London 1840–1914 (Trier, 1979), pp. 37–9, 53–64. Much of the library ended up in the iish. 61 T.L. Gladkova, T.A. Osorgina (eds), Russkaja obščestvennaja biblioteka imeni I.S. Tur- geneva: sotrudniki, druz’ja, počitateli (Paris, 1987); Maria Hunink, De papieren van de revo- lutie: het Internationaal Instituut voor Sociale Geschiedenis 1935–1947 (Amsterdam, 1986), pp. 90–5. The Lavrov-Goc collection, now in large part at the iish, had author and numerus currens catalogues, which survive. 62 For two out of several documented examples, see Willy Birkenmaier, Das russische Heidelberg: zur Geschichte der deutsch-russischen Beziehungen im 19. Jahrhundert (Heidel- berg, 1995); Alfred Erich Senn, Nicholas Rubakin: A Life for Books (Newtonville, ma, 1977). 63 By then, Schlüter had moved to the United States; his own library is now in the University of Wisconsin at Madison (Schlueter-Walling Collection). 412 jaap kloosterman

Figure 1. The cyrillic alphabetical catalogue of the Lavrov-Goc library used by the Russian Party of Socialists-Revolutionaries. Courtesy International Institute of Social History, Amsterdam [IISG Lavrov-Goc Library Collection, 2] catalogue as well as a subject catalogue designed by Max Schippel.64 As a specialized historical institution it remained distinct from the many public and educational libraries that were created by socialists and other Bildung-oriented Germans, including factory owners, from the late 1890s on.65

64 Rüdiger Zimmermann, Das gedruckte Gedächtnis der Arbeiterwegung bewahren: die Geschichte der Bibliotheken der deutschen Sozialdemokratie, 3rd ed. (Bonn, 2008), online at the Friedrich Ebert Stiftung (05888.pdf ); Hans-Peter Harstick, Richard Sperl, and Hanno Strauss (eds), Die Bibliotheken von Karl Marx und Friedrich Engels: annotiertes Verzeich- nis des ermittelten Bestandes (= mega2, iv, 32, Berlin, 1999), pp. 7–84; Paul Mayer, “Die Geschichte des sozialdemokratischen Parteiarchivs und das Schicksal des Marx-Engels- Nachlasses,” Archiv für Sozialgeschichte, vi–vii (1966–1967), pp. 5–198; Günter Benser and Michael Schneider (eds), “Bewahren–Verbreiten–Aufklären”: Archivare, Bibliothekare und Sammler der Quellen der deutschsprachigen Arbeiterbewegung (Bonn, 2009). 65 Dieter Langewiesche and Klaus Schönhoven, “Arbeiterbibliotheken und Arbeiter- lektüre im Wilhelminischen Deutschland,” Archiv für Sozialgeschichte, xvi (1976), pp. 135–204. labor history libraries before world war I 413

In 1892, the Swedish trade unions, which had set up their own libraries, decided to merge a number of them into Stockholms Arbetarbib- liotek. Soon after this was moved to the new People’s House opened in 1901, it acted as host to the Arbetarrörelsens Arkiv, or Labor Movement Archives, which was, again, a combination of what we would today call an archive, a library, and even a museum, created for the express purpose of the study of the labor movement. As in the German case, the attitude of the leadership was important: Hjalmar Branting strongly supported the new institution, which in 1906 became the common property of the social democratic party and the trade union confederation. Contrary to the German situation, though, it was open to the general public.66 The model proved congenial to the labor movement in all of Scandinavia. From 1906 on, the socialist parties and trade unions went to work on the creation of similar institutions in Oslo, Copenhagen and Helsinki; by 1909 they had all succeeded. At around the same time, the United States saw the birth of what is now the Tamiment Library. This was originally the library of the Rand School of Social Science established in New York in 1906 and named after Carrie Rand, a rich widow from Iowa, whose money made it possible. Its instigator, George D. Herron, Rand’s son-in-law, was a minister who first became interested in the Social Gospel movement and then joined the Socialist Party of America. The school was intended for workers, but became a broadly-conceived cultural center. It attracted as teachers many distinguished intellectuals, who often donated books, as did Eugene Debs, the party’s presidential candidate.67

Among the teachers at the Rand School was Bertrand Russell, who was also a colleague of Foxwell at the lse. Both had visited the Paris World’s Fair of 1889—Russell as a seventeen-year old boy, and Foxwell as a par- ticipant in the international monetary conference, of which Léon Say was a vice-president. At another of the dozens of meetings held during the fair, one could have met Bebel, Nettlau, and the eighteen-year old

66 Hans Larsson, Tidstecken: Stockholms arbetarbibliotek och samhällskroppens utformn- ing, 1892–1927 (Stockholm, 1989); Martin Grass, with Hans Larsson, Labour’s Memory: the Labour Movement Archives and Library 1902–2002 (Stockholm, 2002). 67 Daniel Bell, The Tamiment Library (New York, 1969). The library was later taken over by Camp Tamiment, a summer resort for socialists at the eponymous lake in Pennsylvania, which had been set up by the Rand School to help cover its expenses; hence its present name. After the camp and the school folded in 1963, it became part of New York University. 414 jaap kloosterman

David Rjazanov, the future director of the Marx-Engels Institute.68 Similar conjunctions may be found at all the World’s Fairs of the 19th century, which served as a rallying point for many of the protagonists of our story. They did not necessarily come to like each other as a result—indeed, the reverse was often true—but precisely because of that they contributed in diverse and complementary ways to the construction of some of the major source-bases of labor history. Of course, circumstances mattered. To stay with our example, the 1889 fair, which was intended to celebrate the French Revolution, opened its doors when the Boulangisme crisis was still raging. Small wonder, then, that the life of the Musée Social remained intertwined with the vicissitudes of the Third Republic. Likewise, each of the institutions created in its footsteps had its own particular context: they differed as much from the Musée as they resembled it. Yet taken together, their histories are indispensable for an understanding of how the community of early labor historians was born, and why it took certain roads and shrank from others.

68 There were 49 officially approved international conferences; l’Exposition de Paris de 1889, 16 (15 June 1889), p. 126. This did not include such meetings as the third Volapük convention and what was later known as the founding congress of the Second International. labor history libraries before world war I 415

Figure 2. The catalogue of the Twickenham Museum library, designed as a guide for workingman families; from a collection brought together by Leon Kashnor. Courtesy International Institute of Social History, Amsterdam [IISG E 300/75] 416 jaap kloosterman

Figure 3. Pamphlet on “social museums” by Leopold Katscher (1853–1939), in a series opened by Werner Sombart in 1904. Courtesy International Institute of Social History, Amsterdam [IISG BRO D 4060/75] Notes on Contributors

Ulbe Bosma (1962) is Senior Research Fellow at the International Insti- tute of Social History in Amsterdam. His books include Sugarlandia Revis- ited (New York: Berghahn, 2007; edited with Juan Giusti-Cordero and G. Roger Knight), and Being “Dutch” in the Indies: A History of Creolisation and Empire, 1500–1920 (Singapore: NUS Press, 2008; with Remco Raben). His articles have appeared in Comparative Studies in Society and History, Journal of Global History, International Review of Social History, Interna- tional Migration Review, and other journals.

Jaap R. Bruijn (1938) is Professor Emeritus of Maritime History at the University of Leiden. His numerous books include The Dutch Navy of the Seventeenth and Eighteenth Centuries (Columbia, SC: University of South Carolina Press, 1993) and Commanders of Dutch East India Ships in the Eighteenth Century (Woodbridge, UK, and Rochester, NY: Boydell and Brewer, 2011). Together with Jan Lucassen he edited Op de schepen der Oost-Indische Compagnie: vijf artikelen van J. de Hullu (Groningen: Wolters- Noordhoff/Bouma’s Boekhuis, 1980) and together with Jan Lucassen and Paul van Royen “Those Emblems of Hell”? European Sailors and the Mari- time Labour Market, 1570–1870 (St. John’s, Newfoundland: International Maritime Economic History Association, 1997).

Karel Davids (1952) is Professor of Economic and Social History at the VU University Amsterdam. He is the author of The Rise and Decline of Dutch Technological Leadership. Technology, Economy and Culture in the Netherlands, 1350–1800. 2 volumes (Leiden: Brill, 2008). With Jan Lucassen he edited A Miracle Mirrored. The Dutch Republic in European Perspective (Cambridge: Cambridge University Press, 1995; Dutch edition 2005).

Femme S. Gaastra (1945) is Professor Emeritus of Maritime History at the University of Leiden. His most important publications are Bewind en beleid bij de VOC. De financiële en commerciële politiek van de bewindhebbers, 1672–1702 (Zutphen: De Walburg Pers, 1989) and The Dutch East India Company: Expansion and Decline (Zutphen: De Walburg Pers, 2003). 418 notes on contributors

Lex Heerma van Voss (1955) leads the Research Group for Political- Institutional History at the Huygens ING Institute in The Hague and is Professor of the History of Labor at Utrecht University. He is a former staff member of the International Institute of Social History in Amsterdam. He is co-editor of Class and Other Identities. Gender, Religion and Ethnicity in the Writing of European Social History (Oxford and New York: Berghahn, 2002); Between Cross and Class. Comparative Histories of Christian Labour in Europe, 1840–2000 (Bern: Peter Lang, 2005); and The Ashgate Companion to the History of Textile Workers, 1650–2000 (Farnham and Burlington, VT: Ashgate, 2010).

Danielle van den Heuvel (1977) is British Academy Postdoctoral Fellow at the Faculty of Economics and Girton College, and affiliated lecturer at the Faculty of History, University of Cambridge. She is the author of “Bij uijtlandigheijt van haar man”. Echtgenotes van VOC-zeelieden aangemonsterd voor de kamer Enkhuizen (1700–1750) (Amsterdam: Aksant, 2005) and Women and Entrepreneurship. Female Traders in the Northern Netherlands, c. 1580–1815 (Amsterdam: Aksant, 2007; winner of the Thirsk-Feinstein Dissertation Prize 2008 and the IEHA PhD Dissertation Prize 2009) and of articles in i.a. Signs, Continuity and Change and The History of the Family.

Karin Hofmeester (1964) is Senior Research Fellow at the International Institute of Social History in Amsterdam, and Professor of Jewish Culture at the University of Antwerp, Belgium. Her most important book is Jewish Workers and the Labour Movement. A Comparative Study of Amsterdam, London and Paris (1870–1914) (Aldershot: Ashgate, 2004). Together with Jan Lucassen and Christine Moll-Murata she coordinates the “Global Col- laboratory on the History of Labour Relations”. She is currently working on a project called “Luxury and Labour. A global trajectory of diamond consumption and production”.

Chitra Joshi (1952) teaches history at Indraprastha College, University of Delhi. She is the author of Lost Worlds. Indian Labour and its Forgotten Histories (Delhi: Permanent Black, 2003 and London: Anthem Press, 2005) and of numerous book chapters and articles in i.a. Indian Economic and Social History Review, International Review of Social History and Studies in History. She is currently working on the history of roads and labor in nineteenth-century India. notes on contributors 419

Gijs Kessler (1969) is Senior Research Fellow and Senior Collector East- ern Europe and Russia at the International Institute of Social History, Amsterdam, and a founding member of the Interdisciplinary Centre for Studies in History, Economy and Society, Moscow, Russia. He is co-editor of A Dream Deferred: New Studies in Russian and Soviet History (Berne: Peter Lang, 2008) and published articles in a number of journals, including Cahiers du Monde Russe, Continuity and Change and The History of the Family. He lives and works in Moscow.

Jaap Kloosterman (1948) is a former General Director of the International Institute of Social History in Amsterdam. Together with Jan Lucassen he published Rebels with a Cause. Five Centuries of Social History Collected by the IISH (Amsterdam: Aksant, 2010). He is currently writing a comparative history of secret societies.

Marcel van der Linden (1952) succeeded Jan Lucassen as Research Director of the International Institute of Social History in Amsterdam in 2001. He is also Professor of Social Movement History at the University of Amsterdam. Recent publications include Transnational Labour His- tory: Explorations (Aldershot: Ashgate, 2003; Spanish translation 2006), Western Marxism and the Soviet Union: A Survey of Critical Theories and Debates since 1917 (Leiden: Brill, 2007, Chicago: Haymarket, 2009; Chinese and Korean translations forthcoming), and Workers of the World: Essays toward a Global Labor History (Leiden: Brill, 2008; Brazilian edition forthcoming; winner of the René Kuczynski Prize 2009).

Catharina Lis (1945) is Professor Emeritus of Social History at the Vrije Universiteit Brussel. She is the author of Social Change and the Labouring Poor: Antwerp, 1760–1860 (New Haven: Yale University Press, 1986). With Hugo Soly she has published Poverty and Capitalism in Pre-Industrial Europe (Brighton: Harvester Press, 1979), and Disordered Lives: Eighteenth-Century Families and Their Unruly Relatives (Cambridge: Polity Press, 1996). With Jan Lucassen and Hugo Soly she edited Before the Unions: Wage Earners and Collective Action in Europe, 1300–1850 (Cambridge, Cambridge Univer- sity Press, 1994), with Jan Lucassen, Maarten Prak and Hugo Soly Craft Guilds in the Early Modern Low Countries: Work, Power and Representa- tion (Aldershot: Ashgate, 2006), and with Josef Ehmer The Idea of Work in Europe from Antiquity to Modern Times (Aldershot: Ashgate, 2009). 420 notes on contributors

Jelle van Lottum (1976) is a Birmingham Fellow at the University of Bir- mingham. His publications include Across the North Sea. The Impact of the Dutch Republic on Labour Migration, c. 1550–1850 (Amsterdam, 2007), and articles in Continuity and Change, Economic History Review, History of the Family, Rural History and other journals. Having been principle investiga- tor on projects at the Universities of Cambridge and Oxford, he now leads the ESRC funded project “Migration, human capital and labour productivity: the international maritime labour market in Europe, c. 1650–1815”.

Leo Lucassen (1959), is Professor of Social and Economic History and the chair of the History Department at the University of Leiden, and Jan Lucassen’s brother. Together with Jan Lucassen he edited Migration, Migration History, History: old paradigms and new perspectives (Bern 1997) and (with Patrick Manning) Migration History in World History. Multidis- ciplinary approaches (Leiden and Boston 2010). His publications include The Immigrant threat. The integration of old and new migrants in West- ern Europe since 1850 (Urbana and Chicago 2005) and The Encyclopedia of Migration and Minorities in Europe. From the 17th century to the present (New York 2011) edited with Klaus Bade et al. In 2011 the book Winnaars en verliezers. Een nuchtere balans van vijf eeuwen immigratie (Amsterdam) appeared which he co-authored with Jan Lucassen.

Elise van Nederveen Meerkerk (1975) is a postdoc researcher at the International Institute of Social History in Amsterdam and teaches history at the University of Leiden. She is the author of De draad in eigen handen. Vrouwen en loonarbeid in de Nederlandse textielnijverheid 1581– 1810 (Amsterdam: Akant, 2007) and co-author of several other books. She also published articles in a number of journals, including Continuity and Change, Economic History Review, International Review of Social History and Journal of Urban History.

Maarten Prak (1955) is Professor of Social and Economic History at the University of Utrecht. He (co-)authored and (co-)edited numerous books and articles on West European early modern history. His books include Gezeten burgers. De elite in een Hollandse stad: Leiden 1700–1780 (Amster- dam: De Bataafsche Leeuw, 1985), Republikeinse veelheid, democratische enkelvoud. Sociale verandering in het Revolutietijdvak: ’s-Hertogenbosch 1770–1820 (Nijmegen: sun, 1999) and The Dutch Republic in the Seventeenth Century (Cambridge: Cambridge University Press, 2005; originally in Dutch 2002; Hungarian edition 2004). notes on contributors 421

Ratna Saptari (1959) teaches anthropology at the University of Leiden and is a former staff member of the International Institute of Social His- tory, Amsterdam, where she coordinated the Changing Labour Relations in Asia (clara) programme. In addition to publishing several articles and book chapters, she edited a number of books including The Household and Beyond: Cultural Notions and Social Practices in the Study of Gender in Indonesia (London and Copenhagen: Curzon Press-NIAS, 2000) and Labour in Southeast Asia: Local Processes in a Globalized World (London: RoutledgeCurzon, 2004; with Becky Elmhirst).

Willem van Schendel (1949) is a Senior Research Fellow and Senior Collector South Asia of the International Institute of Social History, Amsterdam, and Professor of Modern Asian History at the University of Amsterdam. His recent books include The Bengal Borderland: Beyond State and Nation in South Asia (London: Anthem Press, 2005), Global Blue: Indigo and Espionage in Colonial Bengal (Dhaka: University Press, 2006; with Pierre-Paul Darrac) and A History of Bangladesh (Cambridge: Cam- bridge University Press, 2009).

Hugo Soly (1945) is Professor Emeritus of Early Modern History at the Vrije Universiteit Brussel. His many publications include Urbanisme en kapitalisme te Antwerpen in de 16de eeuw (Brussels: Gemeentekrediet, 1977), and with Catharina Lis Poverty and Capitalism in Pre-Industrial Europe (Brighton: Harvester Press, 1979), and Disordered Lives: Eighteenth-Century Families and Their Unruly Relatives (Cambridge: Polity Press, 1996). He (co-)edited numerous books on the social history of pre-industrial Europe, including Before the Unions: Wage Earners and Collective Action in Europe, 1300–1850 (Cambridge, Cambridge University Press, 1994), Craft Guilds in the Early Modern Low Countries: Work, Power and Representation (Alder- shot: Ashgate, 2006), and Learning on the Shop Floor: Historical Perspec- tives on Apprenticeship (Oxford: Berghahn, 2007).

Richard W. Unger (1942) is Professor Emeritus of Medieval History at the University of British Columbia, Vancouver. His numerous books include The Ship in the Medieval Economy, 600–1600 (Montreal: McGill- Queen’s University Press, 1980), A History of Brewing in Holland 900–1900: Economy, Technology and the State (Leiden: Brill, 2001), Beer in the Middle Ages and the Renaissance (Philadelphia: University of Pennsylvania Press, 2004; paperback 2007), and Ships on Maps: Pictures of Power in Renaissance Europe (London: Palgrave Macmillan, 2010). Together with Jan Lucassen 422 notes on contributors he published “Labour Productivity in Ocean Shipping, 1500–1850,” Interna- tional Journal of Maritime History, 12, 2 (2000) and “Shipping, Productivity and Economic Growth,” in a volume he edited, Shipping Efficiency and Economic Growth 1350–1850 (Leiden: Brill, 2011).

Jan Luiten van Zanden (1955) is Distinguished Professor of Global Eco- nomic History at the University of Utrecht and a former staff member of the International Institute of Social History, Amsterdam. His recent books include The Strictures of Inheritance. The Dutch Economy in the Nineteenth Century (Princeton: Princeton University Press, 2004; with Arthur van Riel), A History of Royal Dutch Shell, vol. I: From Challenger to Joint Indus- try Leader, 1890–1939 (Oxford: Oxford University Press, 2007; with Joost Jonker) and The Long Road to the Industrial Revolution. The European Economy in a Global Perspective, 1000–1800 (Leiden: Brill, 2009).

Erik-Jan Zürcher (1953) is General Director of the International Insti- tute of Social History, Amsterdam, and Professor of Turkish Studies at the University of Leiden. His many publications include The Unionist Factor. The Role of the Committee of Union and Progress in the Turkish National Movement (1905–1926) (Leiden: Brill, 1987; Turkish edition 1987), Turkey: A Modern History (London: I.B. Tauris, 1993; translated into eight other languages) and The Young Turk Legacy and Nation Building (London: I.B. Tauris, 2010). In 2009–2012 he headed an international research group involved in a project on military labor called “Fighting for a Living”. TABULA GRATULATORIA

Mieke Aerts Marcel van der Linden Bert Altena Catharina Lis & Hugo Soly Klaus J. Bade Jelle van Lottum Rana Behal Henk Looijesteijn Sabyasachi Bhattacharya Leo Lucassen Aad Blok Kees Mandemakers Ulbe Bosma Patrick Manning Jaap R. Bruijn Bauke Marinus P.J.E.M. van Dam Sjarrel Massop Karel Davids Christine Moll-Murata Gita Deneckere Bert De Munck Josef Ehmer Elise van Nederveen Meerkerk P.C. Emmer Irina Novichenko Myriam Everard Rinus Penninx Marijke van Faassen Roger de Peuter Femme S. Gaastra Maarten Prak Oscar Gelderblom Richard N. Price Hilde Greefs Wilfried Reininghaus Kees Groenendijk Matthias van Rossum Lex Heerma van Voss Eric de Ruijter Danielle van den Heuvel Huub Sanders Jack Hofman Ratna Saptari Karin Hofmeester Willem van Schendel Chitra Joshi Marlou Schrover Gijs Kessler Pieter Spierenburg Paul M.M. Klep L. Jeroen Touwen Jaap Kloosterman Richard W. Unger Ad Knotter Loes van der Valk Liesbeth Laman Boudien de Vries Serge Langeweg Peer Vries Götz Langkau Anne Winter Ursula Langkau-Alex Jan Luiten van Zanden Marco H.D. van Leeuwen Erik-Jan Zürcher J. Thomas Lindblad

INDEX

Abdullah Fauzia, 33, 155 Bavel, Bas van, 319, 328, 338, 358, 359 Abu-Lughod, Janet, 30 Bayle, Pierre, 291 Accidents, 402 Bebel, August, 411, 413 Africa, 1, 6, 7, 21, 36, 45, 50, 126, 134, 136, Beggars, 310, 311 141, 143, 172, 186, 356, 384, 390 Beier, A.L., 307, 308, 310 Age, 30, 48, 242, 250, 256, 263, 289, 307, 324 Belgium Ahuja, Ravi, 27 Antwerp, 20, 24, 43, 44, 45, 46, 144, 147, Aksan, Virginia, 173, 181 293, 319 Albert II, Duke, 315 Dendermonde, 231 Alfonso X, 312 Flanders, 4, 198, 240, 317, 319, 358, 359 Ali Pasha, Mehmed, 173, 176 Benin, 134 Allen, Robert, 248, 249, 278, 279, 280 Bergsma, Adriaan, 228, 232 Alting, Willem Arnold, 228, 232 Bezanson, Anne, 278 Amadeo, Tomás, 406 Bhattacharya, Sabyasachi, 11, 96 American Museum of Safety, 402 Bibliothèque publique et universitaire, 409 American Revolution, 273, 275, 292, 293 Birth rate, 241 Anglo-Indians, 50 Bishopsgate Institute, 409, 410 Antigua, 388 Black Sea, 365 Antiquarian Booksellers Association, 410 Blacksmiths, 56 Arbetarrörelsens Arkiv (Stockholm), 413 Blackwater, 186 Argentina, 154, 406 Boeke, J.H., 15, 373, 374, 375, 385, 386, 387, Buenos Aires, 12, 154, 157, 160, 162, 163, 388 167, 169, 406 Bolland, Nigel, 371, 382, 386 Ariëns, Alfons, 404 Bombay Vigilance Association, 163 Armenians, 50 Bondage, 313 Armies, 173, 174 Bookkeepers, 60, 197, 198, 201 Arús, Rossend, 409 Boot, H.M., 191, 192, 200, 207, 211, 212 Asia, 1, 7, 9, 11, 20, 30, 50, 61, 68, 78, 99, 110, Boreel, Balthasar, 227, 233 126, 134, 136, 141, 148, 173, 174, 181, 188, Boulangisme, 414 189, 193, 195, 215, 223, 226, 227, 228, 229, Box-making, 113 230, 231, 232, 233, 234, 235, 356, 378, 379 Brabant, Estates, 318 Aslam, Ishrat, 27 Bradley, Joseph, 337, 364, 366, 397 Austria, 315, 321, 401 Branting, Hjalmar, 413 Vienna, 303, 311, 312, 315, 321, 400, 401 Braudel, Fernand, 125, 127, 128, 129, 130, 132, 145, 146 Babur, Sultan, 31 Braverman, Harry, 99, 100 Bailey, Beth, 173, 179, 187 Brazil, 20, 31, 36, 37, 38, 41, 46 Balkan Peninsula, 179 Bahia, 36 Baltic region, 240, 258, 293, 344 Mato Grosso, 36 Bangladesh, 47 Minas Gerais, 36, 37, 38 Dhaka, 48, 50, 53, 58 Rio de Janeiro, 21, 36, 38 Jessore, 49, 50, 54, 60, 62, 81, 85 Brigands, 58 Banians, 24 Britain, 36, 42, 51, 66, 103, 112, 156, 180, 185, Barbados, 388 186, 191, 192, 212, 249, 272, 279, 280, 282, Barendse, R.J., 25, 30 284, 285, 288, 293, 294, 327, 332, 396, 397 Batista, Fulgencio, 384 England, 20, 23, 24, 41, 50, 127, 130, 151, Bavaria, 307, 311 187, 237, 248, 258, 272, 277, 286, 290, 426 index

291, 302, 307, 308, 310, 320, 321, 324, Nanking, 389 325, 326, 327, 330, 332, 333, 335, 336, Shanghai, 154, 157, 159, 160, 162, 163, 165, 337, 341, 342, 343, 348, 350, 351, 389 168 Essex, 302, 308 Chinese, 10, 20, 59, 60, 61, 68, 105, 118, 154, London, 4, 6, 14, 20, 21, 23, 24, 35, 37, 38, 157, 158, 160, 163, 165, 167, 175, 178, 179, 39, 40, 43, 44, 48, 50, 54, 62, 64, 66, 71, 180, 186, 187, 188, 225 75, 77, 91, 92, 99, 100, 101, 105, 112, 125, Chinese Exclusion Act (1882), 158 127, 128, 130, 131, 134, 139, 142, 154, 156, Chinese Revolution, 163 159, 170, 231, 233, 240, 252, 255, 258, Christians, 396 279, 284, 290, 291, 292, 293, 313, 321, Churches, 402 324, 327, 332, 335, 336, 339, 340, 341, Clapham Sect, 347 353, 354, 355, 356, 367, 374, 379, 381, Clark, Alice, 131, 302, 309 382, 391, 395, 396, 397, 398, 399, 400, Clerks, 193, 195, 203, 207, 208 405, 406, 407, 409, 410, 411 Cochin, Augustin, 403 Manchester, 239, 257, 400, 409, 410 Cohn, Samuel, 300, 316 Norfolk, 23, 309 Cold War, 186 Wales, 50, 332 Colgong factory, 56 British Museum, 406 Colombia, 30 Brougham, Henry, 397 Colonial Development Act (1929), 385 Bruijn, Jaap R., 9, 13, 215, 221, 223, 224, 251, Communist Party (Cuba), 383, 384 254, 260, 261 Conservatoire des Arts et Métiers, 398 Brussels Sugar Convention (1902), 377, Contagious Diseases Acts (1864, 1866, 379, 386 1869), 156 Burawoy, Michael, 100, 101 Coolies, 51, 54, 56, 57, 59, 64, 68, 70, 110, Burkina Faso, 134 116, 379, 380 Cope, Joseph, 39 Cambridge Group, 237 Costa Rica, 141 Canada, 21 Crisis Import Ordinance (1933), 103 Montreal, 401 Cuba, 15, 371, 372, 375, 376, 377, 378, 379, Quebec, 401 380, 381, 382, 383, 384, 385, 386, 391 Caribbean, 11, 15, 53, 54, 228, 269, 371, 372, Czechoslovakia, 379 373, 375, 376, 377, 378, 382, 383, 384, 385, 386, 391 Dairying, 337 Carpenters, 56 Dam, Pieter van, 23, 25, 26, 27, 28, 29, 34, Cashiers, 210 193, 194, 196, 200, 201, 207, 211 Catherine the Great, 294 Danube, 270, 294, 299 Cellini, Benvenuto, 31 De Beers Company, 21 Centraal Bureau voor Sociale Adviezen, 404 De Soto, Hernando, 133, 134, 135, 145 Central Industrial Region, 364, 367 De Vries, Jan, 151, 191, 192, 198, 200, 202, Centralverein für das Wohl der 203, 208, 210, 211, 238, 240, 243, 244, 245, arbeitenden Klassen, 405 255, 256, 257, 323, 324, 326, 328, 335, 338, Chambrun, Aldebert de, 403 344, 347 Chapmen, 127 Death rate, 241 Charles V, 312, 318, 319 Debs, Eugene, 413 Chayanov, A.V., 354, 355, 357 Denmark, 128, 219, 220, 222, 231, 253, 305, Chevalier, Michel, 403 306, 320 Cheysson, Emile, 403, 404 Aarhus, 220, 225 Child labor, 4, 14, 25, 26, 40, 43, 45, 57, 100, Copenhagen, 20, 231, 257, 413 103, 106, 111, 112, 113, 116, 121, 135, 138, 139, Sjaelland, 306 324–330, 332, 333, 335, 337, 338, 339, Descaves, Lucien, 410 341–352 Dewey, Melvil, 400 China, 6, 10, 13, 19, 20, 158, 171, 172, 173, 178, Diamonds, 19–46 180, 181, 182, 183, 188, 223, 377, 379, 389 Briljant cut, 34, 35, 39 Canton, 225 Cutting, 31, 33, 35, 36, 39, 44, 45 Jiangsu, 389 Polishing, 20, 31, 33, 34, 35, 42, 43, 44, 45 index 427

Rose cut, 31, 32, 39, 42 Nantes, 231 Table cut, 31, 32, 42 Paris, 35, 42, 129, 130, 138, 145, 154, 156, Dinabandhu Mitra, 48, 72 231, 259, 299, 300, 312, 313, 314, 321, Division of labor, 26, 113, 176, 193, 195, 350, 356, 377, 396, 398, 399, 401, 402, 403, 357, 387 404, 405, 406, 407, 409, 410, 411, 413, Domestic servants, 51, 57, 59, 60, 63, 160, 414 166, 170, 252, 311, 331, 332, 364 Provence, 54, 307, 316, 317 Dominican Republic, 371 Strasbourg, 139, 147 Dumont, Hermine, 384 Frank, Andre Gunder, 25, 30, 173, 181, 188, Dutch Republic, 3, 14, 125, 128, 151, 174, 180, 272, 299 188, 215, 216, 230, 231, 251, 254, 255, 256, Frederick II, Emperor, 303 257, 258, 259, 260, 263, 290, 324, 328, Freedman, Paul, 313 333, 334, 335, 340, 341, 343, 349, 351 French Revolution (1789), 251, 399, 414 See also: Netherlands Fryer, John, 35 Furber, Holden, 30 East India Company (Dutch), 2, 9, 13, 20, Furly, Benjamin, 271, 286, 287, 290, 291, 292 23, 29, 191–212, 215–235 Furnivall, J.S., 374 East India Company (English), 13, 20, 23, 35, 39, 40, 41, 42, 43, 46, 76, 78, 79, 80, Gaastra, Femme, 13, 193, 194, 195, 197, 203, 187, 191, 192, 197, 200, 207, 211, 212 210, 215, 226, 227 Ecole Libre des Sciences Politiques, 406 Gandhi, Mahatma, 73 Edwardes, Stephen M., 162 Garveyism, 382 Edwards, Richard, 112, 118, 121, 400 Geertz, Clifford, 133 Ehmer, Josef, 6 Gelder, Roelof van, 216 Einaudi, Luigi, 405 Gelderen, Jacob van, 374, 379, 387, 388, Elbe, 299, 300, 320 389, 390 Elephant servant, 57 Gender, 99, 128, 134, 135, 154, 327, 332, 344, Elson, R.E., 378, 386 395 Employers, 300 Geological Survey of India, 43 Engels, Friedrich, 271, 410, 411, 412, 414 Germany, 126, 128, 131, 137, 142, 222, 231, Engerman, Stanley, 170, 299 233, 253, 255, 259, 260, 263, 270, 276, 277, England: see Britain 279, 280, 285, 288, 289, 291, 293, 311, 315, Erickson, Amy, 336, 339, 340, 341 317, 318, 321, 324, 325, 333, 334, 335, 336, Espinas, Alfred, 403 337, 338, 340, 341, 343, 347, 348, 349, 350, Exner, Wilhelm, 401 351, 357, 400, 405, 408, 411 Augsburg, 279, 280 Fabian Society, 374, 375, 376, 385, 390, 391, Berlin, 31, 75, 230, 231, 318, 380, 401, 408, 406 410, 411, 412 Factories, 49, 50, 56, 58, 103, 117, 128, 402 Bremen, 220, 231, 234 Fahmy, Khaled, 173, 190 Cleve, 230 Faroe Islands, 304 Cologne, 132, 303, 334, 408 Fernando I, 310, 312 Frankfurt, 231 Fines, 85, 114, 118 Hadersleben, 223 Fogel, Robert W., 170 Hamburg, 171, 220, 231, 257, 408 Fontaine, Henri de la, 126, 127, 128, 129, 130, Hanau, 231 132, 138, 143, 145, 147 Hannover, 231 Food, 27, 125, 136, 341, 380 Holstein, 234, 257 France, 51, 54, 128, 129, 138, 139, 150, 171, 178, Kiel, 408 181, 188, 231, 251, 253, 254, 255, 259, 288, Leer, 233 313, 316, 321, 377, 381, 396, 397, 403, 404, Lippe-Detmoldt, 8 406, 407 Lipstadt, 231 Auvergne, 141, 143 Magdeburg, 231 Île-de-France, 313, 315 Rhineland, 269, 270, 278, 279, 286, 288, Marseille, 316 289, 315, 317, 318, 321, 389 Metz, 145, 231 Saarbrücken, 408 428 index

Sauerland, 140 Holy Roman Empire, 288 Saxony, 389, 405 Holyoake, George Jacob, 409 Schleswig, 219, 257 Home production, 12, 104, 106, 108, 110, 111, Westphalia, 358, 408 112, 114, 115, 116, 117, 119, 120 Gerschenkron, Alexander, 355 Hope & Co, 38, 292 Gewerbe-hygienisches Museum, 400 Horrell, Sara, 332, 333 Ghana, 133 Howard, Henry Earl, 23, 24, 28, 402 Accra, 133 Hudig, Dirk, 404, 405 Gierke, Otto von, 410 Hufton, Olwen, 129, 131, 138, 139, 141, 143, Gildemeester, Daniël, 37, 38 146, 150 Gnocchi Viani, Osvaldo, 405 Huiswoud, Otto, 383, 384 Godavari river, 21 Humphries, Jane, 325, 332, 333 Golconda mines, 21 Hunger, 371, 372, 375, 381, 384, 385, 390 Gomashta, 52, 56 Hunt Hutchinson, Henry, 406 Grau San Martín, Ramón, 384 Great Depression, 12, 380 Immigrants, 10, 251, 272, 280 Great Exhibition of the Works of Industry Indenture, 179, 281, 282, 286, 288–289, 292 of All Nations, 398 India, 2, 5, 6, 8, 9, 10, 11, 12, 13, 19, 20, 21, 22, Great Revolt of 1381, 320 23, 24, 25, 26, 27, 28, 29, 30, 31, 33, 34, 35, Grégoire, Henri, 398 36, 37, 38, 39, 40, 42, 43, 44, 45, 46, 47, Grünberg, Carl, 410 48, 50, 54, 61, 62, 64, 71, 73, 75, 76, 77, Guadeloupe, 381 78, 79, 80, 81, 82, 87, 88, 89, 92, 93, 94, Guatemala, 141 95, 96, 99, 162, 165, 172, 173, 175, 176, 180, Guilds, 7, 99, 125, 340, 341 182, 187, 191, 192, 194, 197, 198, 200, 207, Gustavus Adolphus II, 176 211, 212, 215, 217, 226, 250, 376, 377, 379, Guyana, 382 386, 390 Guyon, Louis, 401 Allahabad, 42, 87, 93 Andhra Pradesh, 21 Habsburg Empire, 20, 188, 318 Assam, 66 Haiti, 371, 376, 377, 381 Benares, 42, 81, 87, 93 Hajnal, John, 367 Bengal, 11, 30, 42, 47, 48, 49, 50, 53, 55, Hart, Keith, 133, 134 57, 58, 59, 60, 61, 62, 64, 65, 66, 69, 70, Harvard, 405, 406, 407 72, 73, 79, 86, 88, 95, 229, 231, 389 Harvard Business School, 407 Bihar, 47, 50, 56, 61, 66, 72, 84 Havelock Ellis, Henry, 161 Bijapur, 23, 29 Haven Putnam, Bertha, 307 Bombay [Mumbai], 12, 30, 42, 43, 45, 79, Hawkers, 125–151 80, 83, 89, 90, 96, 154, 155, 158, 160, Haze de Georgio, Jeronimo de, 210 162, 163, 165, 168 Hazelius, Arthur, 398 Calcutta [Kolkata], 33, 39, 48, 50, 53, 60, Health, 76, 154 62, 73, 79, 81, 82, 88, 90, 93, 95 Heimann, Hugo, 411 Central Provinces, 79 Heinz, H.J., 171, 402 Chhota Nagpur, 59, 64, 65, 66, 67, 68 Herron, George D., 413 Cawnpore, 87, 93 Hershatter, Gail, 154, 155, 159, 160, 162, 164, Coromandel Coast, 20, 23 165, 168 Cuddapah [Kadapa], 21, 39, 40, 41 Heyne, Benjamin, 40 Deccan plateau, 21, 23 Hippler, Thomas, 173, 174, 189 Goa, 20, 21, 24 Hogendorp, Dirk van, 229, 232 Gujarat, 24, 43, 44 Hoh-Cheung Mui, 130 Hyderabad, 42 Holland, 4, 13, 99, 102, 128, 132, 148, 195, Krishnagar, 62 203, 221, 229, 232, 237, 238, 239, 240, 241, Madhya Pradesh, 21 242, 243, 244, 245, 252, 257, 269, 280, 319, Madras [Chennai], 20, 23, 24, 27, 39, 328, 336, 338, 357, 358, 359, 409 42, 50 See also Dutch Republic, and Masulipatnam, 23 Netherlands Midnapur, 83, 90 index 429

Murshidabad, 50 Kahars, 76 Orissa, 21, 47, 88, 94 Katscher, Leopold, 405, 416 Oudh, 76 Kaya, Yelda, 173, 179 Purnea, 50 Kenya, 134, 157, 168, 169 Ramallakota, 23, 25, 28, 33, 34, 42 Nairobi, 12, 154, 155, 157, 159, 161, 162, 166, Rungpore, 82, 84 167, 169 Singhbhum, 90 Kerdijk, arnold, 404 Indian Penal Code, 157 Keynes, 407, 410 Indians, 19, 29, 34, 43, 54, 161 Kipling, Rudyard, 75, 87, 96 Indigo, 47–73 Klompé, Marga, 323, 352 Indigo Commission, 48, 62 Koh-i-noor, 31 Indigo factories, 49, 55 Kolonialinstitut, 408 Indonesia, 103, 104, 105, 373, 374, 388, 389 Kondratiev, Nikolai, 374, 387 Borneo, 36 Kress, Claude W., 407, 408 Cheribon, 228, 230 Kress Library, 407, 408 Jakarta [Batavia], 215, 387 Kretek, 105 Japara-Rembang residency, 108 Kristna river, 21 Java, 12, 15, 68, 101, 102, 103, 104, 105, 106, Kudus Cigarette Workers’ Union, 119 108, 109, 110, 111, 113, 115, 116, 117, 118, 119, 120, 121, 229, 231, 372, 373, 375, Labor costs, 40 376, 377, 378, 379, 380, 382, 383, 385, Labor history, 153 386, 387, 388, 389, 391 Labor laws, 301, 311 Jogja (Yogyakarta], 108, 110 Labor Party (Netherlands), 374, 388 Kediri, 111, 113, 115 Labour Party (Britain), 387 Kudus, 105, 106, 108, 109, 111, 114, 115, 116, Lal Behari Day, 48, 72 117, 119 Landsknechte, 175 Madiun, 113, 119 Lange, Pieter de, 23, 24, 25, 26, 28, 29, 34 Malang, 116 Latin America, 6, 126, 134, 172, 356, 381 Pekalongan, 108, 113 Laundry work, 159, 160, 332, 333 Ponorogo, 109, 113 Lavrov-Goc collection, 411 Semarang, 105, 113, 209 Le Play, Frédéric, 403 Industrial Revolution, 151, 191, 249, 327, League for Social Service, 402 355 League of Nations, 156 Industrious Revolution, 151, 323, 326, 347 Leonhard, Jörn, 173, 185 Institut für Seeverkehr und Weltwirtschaft, Levant, 23 408 Lewis, W. Arthur, 15, 269, 371, 372, 373, 374, Internal contract system, 112 376, 381, 382, 385, 388, 390, 391 Italy, 173, 185, 186, 198, 315, 316, 358, 359, Liber Augustalis, 303 405 Liberals, 355 Florence, 37, 131, 316 Libraries, 395 Milan, 405 Library of Useful Knowledge, 397 Orvieto, 316 Library standardization, 400 Sicily, 300, 303, 306 Limburg Stirum, John Paul van, 387 Siena, 316 Luxury, 19, 35, 44, 49, 157 Venice, 20, 31, 146, 315 Ivan IV, 360 Male breadwinner, 323–326, 348, 351, 352 Ivan Turgenev, 411 Managers, 48–53, 57, 59, 60, 61, 67, 70, 71, 196 Jamaica, 377, 382, 390 Manning, Patrick, 6, 7 Janissaries, 173, 186 Marginality, 150 Japan, 388, 410 Marken, Jacques van, 404 Jevons, Stanley, 407 Martinique, 377, 381, 382 John II ‘The Good’, 313 Marx, Karl, 171, 410, 411, 412, 414 Johnson, Robert, 91, 173, 187, 248, 356, 364 Mary II, 272 Jute, 72, 73 Mauthner, Theodor, 410 430 index

Maximum wages, 318 225, 227, 230, 233, 234, 237, 238, 239, McKeown, Adam, 6 240, 241, 244, 245, 248, 251, 252, 256, Mehring, Walter, 395 259, 293, 304, 323, 324, 327, 335, 344, Melun, Armand de, 403 378, 381, 388, 396, 401, 408, 409, 410, Mendez, Albaro, 24 411, 412, 415, 416 Menger, Carl, 410 Arnhem, 138 Mercenaries, 9, 172, 174, 179 Brabant, 142, 151 Merchants, 24, 25, 79, 131, 226, 289, 290 Delft, 194, 195, 219, 226, 231, 328 Methwold, William, 23, 24, 25, 26, 27, 29 Den Bosch, 139, 335 Mexico City, 12, 99, 154, 158, 162, 165 Dokkum, 225 Middle East, 6, 13, 76, 95, 172, 174, 185, 186, Enkhuizen, 195, 218, 225, 226, 252 187, 188, 189 Friesland, 128, 225, 357 Migerka, Franz, 400, 401 Gouda, 4, 128, 247, 269, 328, 340, 349 Migration, 5, 6, 128, 129, 134, 136, 141, 144, Groningen, 8, 9, 216, 345 215, 226, 247, 249, 251, 252, 255, 257, 258, Guelders, 319, 328, 338, 359 260, 267, 269, 270, 274, 286, 287, 291, 314, Haarlem, 328 326, 363, 364, 365 Hoorn, 195, 219, 226, 231 Migration rates, 6, 247, 248, 265 Leiden, 2, 3, 6, 9, 25, 30, 36, 54, 139, 148, Military labor, 171–190 182, 193, 198, 216, 226, 248, 285, 289, Miners, 24–29, 33, 36, 40, 41, 45, 46, 160, 171 293, 299, 304, 312, 327, 328, 329, 335, Ming dynasty, 173, 178, 179, 180, 181, 186, 188 336, 340, 359, 373 Mining methods, 26 Middelburg, 229 Moch, Leslie, 5, 6, 141, 144, 146, 150, 247 Nijmegen, 3, 231 Montoliu, Cebrià de, 405 Rotterdam, 14, 62, 195, 221, 225, 226, 231, Mortality, 217, 288 269, 270, 271, 277, 280, 282, 284, 286, Mossel, Jacob, 229 287, 289, 290, 291, 292, 293 Mui, Lorna H., 130, 146, 147 The Hague, 38, 50, 138, 141, 192, 195, 197, Mumbai, see Bombay 202, 203, 210, 211, 215, 227, 228, 239 Musée Social, 15, 403, 404, 405, 406, 407 Tilburg, 329 Museo Industriale, 405 Utrecht, 2, 3, 129, 138, 195, 209, 260, 293, Museo Social, 405, 406 323, 337, 382, 409 Museo Sociale, 405 Vlissingen, 230 Museum of Domestic and Sanitary Zealand, 306, 345 Economy, 399 Zwolle, 131, 145 Nettlau, Max, 409, 410, 413 National Cash Register Co., 402 New York University, 413 Navigation Laws (Britain), 290 Nicaragua, 371 Nederlands Economisch-Historisch North Sea system, 4, 252, 258, 358, 365 Archief, 408 Norway, 219, 222, 223, 225, 293, 300, 304, Nederveen Meerkerk, Elise van, 14, 148, 151, 305, 306, 307, 320 323, 326, 328, 329, 330, 336, 341 Egersund, 223 Netherlands, 1, 3, 5, 7, 8, 10, 15, 102, 103, 104, Numeracy, 263 109, 126, 128, 137, 186, 192, 193, 198, 200, 209, 215, 227, 228, 230, 231, 233, 235, 237, Oberpfennig, Hannelore, 128, 129, 141 251, 253, 256, 257, 259, 260, 269, 270, 273, Ogilvie, Sheila, 126, 130, 131, 142, 145, 327, 274, 283, 291, 293, 294, 323, 324, 326, 329, 334, 338, 340, 341, 348 330, 331, 335, 336, 337, 338, 339, 341, 343, Ordinance for Regulation of Enterprises 348, 350, 351, 352, 357, 358, 374, 378, 380, (1935), 103 383, 385, 386, 387, 388, 389, 390, 401, 404 Ordinance of Labourers (1349), 307 Amsterdam, 3, 5, 6, 9, 10, 20, 38, 43, 45, Ordonnantie Bia Tembakau (1932), 109 46, 105, 129, 130, 131, 138, 139, 144, 145, 153, 172, 192, 193, 194, 195, 196, 197, 198, Palanquin bearers, 78 200, 201, 202, 203, 204, 206, 207, 208, Panizzi, Antonio, 400, 409 209, 210, 211, 212, 215, 217, 221, 222, 223, Pappenheim, Wilhelm, 410 index 431

Paris Commune (1871), 406, 409, 410 Prussia, 181, 185, 270, 288, 294 Party of Socialists-Revolutionaries, 411, See also Germany 412 Puerto Rico, 380 Patras, Abraham, 216 Pugin, Augustus, 398 Pattamars, 76 Pawning, 164, 165, 169, 185 Quack, H.P.G., 409 Peabody, Francis G., 405, 406 Quitrent, 362 Peace of Amiens (1802), 288 Pearl river, 225 Rand School of Social Science, 413 Pearls, 37, 38 Reid, W.M., 50, 51, 67 Peasantry, 72, 73, 78, 158, 172, 173, 176, 181, Reijden, B. van der, 104, 105, 106, 107, 108, 183, 185, 187, 190, 301, 305, 306, 311, 313, 109, 110, 111, 112, 113, 114, 115, 117, 118, 120 314, 316, 326, 327, 345, 346, 349, 353–357, Remuneration, 328 359–368, 373 Renaissance, 20, 137, 138, 139, 142, 147, 148, Peddlers, 125–151 334 Pedro I, 312, 313 Renard, Georges, 398 Pedro IV, 313 Rent, 25, 204 Penn, William, 271, 272, 274, 275, 276, 280, Replenishment Scheme (1962), 44 285, 286, 289, 290, 291, 307, 308, 309 Resistance, 12, 64, 70, 71, 72, 73, 90, 94, 95, Penner river, 21, 41 105, 112, 117, 118, 119, 147, 154, 183, 186, 189, Penninx, Rinus, 3, 10, 144, 251 190, 211, 308, 323, 371, 372, 373, 375, 376, Peons, 56 381, 382, 383, 386, 391 Pereire, Emile, 403 Responsible autonomy, 101 Perquisites, 207 Richard II, 320 Perrier, Jules, 409 Rjazanov, David, 396, 414 Peru, 126, 133, 134 Robertson, George, 39, 320 Lima, 133 Rogers, William, 408, 409 Petitions, 79, 84 Rollin, Henri, 410 Pflüger, Paul, 405 Roosevelt, Franklin D., 371 Phillipps, Thomas, 407 Ross, C., 2, 40, 41 Pijnappel, P.A., 407 Rousselet, Louis, 42 Pillage, 179, 180, 181 Rovinello, Marco, 173, 175, 189 Pimps, 165, 169 Roy, Kaushik, 36, 173, 180, 410, 411 Planters, 50, 51, 56, 58, 59, 73 Royal Moyne Commission, 376 Platt-agreement (1901), 371 Royal Society of Arts, 397 Playfair, Lyon, 399 Royal Society of England, 23 Plettenberg, Joachim, 228 Runners, 75–97 Pleyte, C.M., 102 Rupp, I.D., 272, 273, 274, 288 Plantations, 55, 66, 182, 371, 372, 373, 375, Ruskin, John, 405 376, 378, 383, 384, 385, 389, 390 Russell, Bertrand, 413 Police, 58, 84, 130, 155, 156, 159, 162, 166 Russia, 21, 162, 178, 182, 185, 231, 270, 293, Poland 294, 353, 355, 360, 361, 363, 364, 365, 368, Dantzig [Gdansk], 231 369, 397 Elbing [Elblag], 220, 223 Kaluga, 365, 366 Pomerania, 222 Moscow, 230, 231, 354, 356, 362, 364, Stettin, 222 365, 366, 396, 410 Warsaw, 186, 365 Novgorod, 366 Populists, 355 Riazan, 365 Portugal, 20, 21, 24, 25, 36, 37, 38, 45, 46, St. Petersburg, 231, 364, 365, 366 176, 223, 307, 310 Vitebsk province, 366 Prager, Lyon, 39 Russian Library, 411 Prize Paper Archive, 250 Productivity, 248, 249, 256, 289 Sailors, 9, 250, 251, 259, 289 Prostitution, 12, 118, 153–170, 190, 225 Sautijn, Willem, 227 432 index

Sawyers, 56 Syria Say, Léon, 403, 413 Aleppo, 180 Scandinavia, 253, 255, 259, 263, 305, 306, Damascus, 180 413 See also Denmark, Norway, Sweden Tagore, Dwarkanath, 48, 50 Schippel, Max, 412 Tallett, Frank, 173, 181, 188 Schlüter, Hermann, 411 Talma, Sybrand, 404 Schmidt, Peter, 139, 326, 340, 344, 405 Tamiment Library, 413 Schrover, Marlou, 128, 129, 137, 140 Tavernier, Jean-Baptiste, 23, 24, 25, 26, 27, Schweizerisches Wirtschaftsarchiv, 408 33, 34 Seasonal labor, 9, 172, 252, 353, 358 Thedens, Johannes, 216, 233 Second International, 414 Thévenot, Jean de, 33 Second serfdom, 361 Tigers, 91 Shipping, 80, 194, 215, 216, 248, 250, 256, Tille, Armin, 408 285, 289 Tilly, Charles, 184, 325 Shoemaker, Ben and Sam, 292 Tilly, Chris, 184, 325 Shoemakers, 340 Tokugawa Japan, 93 Siegfried, Jules, 403 Tolman, William H., 402 Simonde de Sismondi, Jean, 402 Trade Union Congress (Britain), 385 Sinclair Bart, John, 76 Trastamura, Enrique of, 312 Sinor, K.P., 43 Treub, Willem, 404 Sirdars, 58 Tributary labor, 175–179, 325, 360–363 Social Gospel movement, 402 Trinidad, 383 Social Museum, 405, 406 Società Umanitaria, 405 Uganda, 166 Société d’Economie sociale, 403 Ukraine, 365 Soeara Oemoem, 109 Unfree labor, 299 Soldiers, 171–190, 226 United Fruit Company, 377, 380 Somerton Foxwell, Herbert, 407 United Kingdom, see Britain South Africa, 2, 43, 186 United Nations, 323 Soviet Union, 355, 368 United States of America See also Russia California, 134, 158, 160, 161, 278, 397 Spain, 27, 139, 251, 312, 313, 324, 386 Delaware, 271 Statute of Cambridge (1388), 307 Germantown, 274, 275, 277, 285, 287, 291 Statute of Labourers (1351), 307, 309 Hawaii, 376, 380 Stedman, John, 292 Iowa, 413 Strassburger, R.B., 272, 273, 274, 280, 281, Maryland, 276, 278, 282 283, 287, 290 Pennsylvania, 14, 154, 270, 271, 272, 273, Strong, Josiah, 402 274, 275, 276, 278, 279, 280, 281, 282, Struve, H.W.E., 401 283, 284, 285, 286, 287, 288, 290, 291, Subsistence labor, 14, 325, 338, 339, 342, 293, 294, 413 344, 345, 346, 348, 349, 351, 360, 361 Virginia, 173, 181, 275, 276, 282, 285, 310 Sugar, 371, 376, 377, 378, 379, 380, 381, 382, University of Amsterdam, 2, 407 386, 388, 389 University of Kiel, 408 Surinam, 224, 371 University of London, 397, 407 Sweden, 176, 178, 219, 222, 231, 253, 306, Urbanization, 240, 364, 385 320, 413 Uruguay, 163 Göteborg, 231 Stockholm, 224, 257, 305, 413 Vagrancy, 142, 307, 308 Switzerland, 231, 270, 405 Veiligheidsmuseum, 401 Basle, 408 Verelst, Harry, 39 Bern, 5, 10, 153, 231, 250, 267, 299, 304, Versluys, J., 102 396, 400 Villermé, Louis-René, 403 Geneva, 134, 381, 409 Viticulture, 315 Zürich, 172, 405, 411 Vliegen, Willem, 404 index 433

VOC: see East India Company (Dutch) Weber, Max, 174 Vredenburg, E., 43 West Indian Sugar Commission, 376, 377, 381 Wage labor, 350 White, Luise, 154, 155, 159, 161, 162, 166, 168, Wageningen School, 237 170, 203 Wages, 10, 59, 61, 279, 289, 303, 304, 307, 308, Wiesner, Merry, 131, 132, 137, 138, 139, 142, 309, 325, 327, 331, 332, 333, 373, 380, 387 147, 148, 334 Walmart, 45 William III, 272 Wars William V, 229 Fourth Anglo-Dutch War (1780–84), Winter, Anne, 93, 142, 150, 151 223, 251 Wokeck, M.S., 270, 272, 273, 274, 275, 276, Franco-Prussian War (1870–71), 406 277, 280, 281, 282, 284, 285, 286, 287, French Revolutionary and Napoleonic 288, 289, 290, 292, 293, 294 Wars (1793–1815), 251 Women, 12, 14, 56, 57, 69, 73, 99, 121, 126, Seven Years’ War (1756–63), 275, 288, 127, 128, 130, 131, 134, 135, 136, 137, 138, 292 139, 142, 144, 145, 147, 148, 149, 153, 154, Third Anglo-Dutch War (1672–74), 251 155, 156, 158, 159, 160, 162, 163, 164, 167, Thirty Years’ War (1618–48), 147 170, 242, 252, 323–352, 364, 368, 378, 401 Vietnam War (1955–75), 177 World Bank, 133 War of the American Revolution Woude, Ad van der, 191, 192, 200, 208, 238, (1775–83), 251, 288 243, 245, 257, 324, 328, 335, 338, 344 War of the Austrian Succession (1740–48), 289 Zanden, Jan Luiten van, 7, 13, 200, 237, 239, War of the Spanish Succession (1701–14), 249, 256, 257, 337, 338, 339, 345, 359 251 Zelinsky, Walter, 247 Warsaw Pact, 186 Zentralstelle für soziale Literatur der Way, Peter, 171, 173 Schweiz, 405 Webb, Sidney [Lord Passfield], 385, 406