Letters from an American Botanist the Correspondences of Gotthilf Heinrich Ernst Mühlenberg (1753–1815) Letters Fromletters an American Botanist

The Lutheran Pastor Gotthilf Heinrich and within the botanical discourse of his Ernst Mühlenberg (1753–1815) is remem- time. The result is a multi-faceted depic- bered today as one of the pioneering tion of contemporary standards, codes figures in early American botany, which and pitfalls of scientific communication earned him the posthumous epithet “The in the so-called “Republic of Letters”. As American Linnaeus”. This study traces Mühlenberg’s example shows, the very Mühlenberg’s contributions to American fabric of this Republic – open exchange of botany by reconstructing his vast transat- information – had a strong impact on the lantic correspondence network over a pe- course and outcome of scientific research riod of more than 30 years. Working on the itself. This “Network Factor” becomes tenets of modern network studies and with clearly visible in Mühlenberg’s networking information gathered from close to 700 strategies, which he developed to protect original letters, diaries and publications, his original work against the negative ef- the present study places Mühlenberg both fects of the very medium he was working within his own web of correspondences with. Matthias Schönhofer Letters from an American Botanist The Correspondences of Gotthilf Heinrich Ernst Mühlenberg (1753–1815) Letters fromLetters an American Botanist

www.steiner-verlag.de BEÜ Geschichte Beiträge zur Europäischen Überseegeschichte – 101

Franz Steiner Verlag Franz Steiner Verlag

ISBN 978-3-515-10796-9 Matthias Schönhofer Matthias Schönhofer Letters from an American Botanist beiträge zur europäischen überseegeschichte vormals: Beiträge zur Kolonial- und Überseegeschichte

Im Auftrag der Forschungsstiftung für vergleichende europäische Überseegeschichte herausgegeben von Markus A. Denzel, Hermann Joseph Hiery und Eberhard Schmitt Band 101 Matthias Schönhofer

Letters from an American Botanist The Correspondences of Gotthilf Heinrich Ernst Mühlenberg (1753–1815)

Franz Steiner Verlag Gedruckt mit freundlicher Unterstützung der Gesellschaft für Überseegeschichte (e.V.) und des DFG-Projektes Atlantische Korrespondenzen: Genese und Transformation deutsch-amerikanischer Netzwerke 1740–1870

Umschlagabbildung: Rev. Gotthilf Henry Ernest Muhlenberg by Jacob Eichholtz, 1811, Object number 1984, The Phillips Museum of Art, Franklin & Marshall College, Lancaster, PA, USA

Bibliografische Information der Deutschen Nationalbibliothek: Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über abrufbar.

Dieses Werk einschließlich aller seiner Teile ist urheberrechtlich geschützt. Jede Verwertung außerhalb der engen Grenzen des Urheberrechtsgesetzes ist unzulässig und strafbar. © Franz Steiner Verlag, Stuttgart 2014 Druck: Offsetdruck Bokor, Bad Tölz Gedruckt auf säurefreiem, alterungsbeständigem Papier. Printed in Germany. ISBN 978-3-515-10796-9 (Print) ISBN 978-3-515-10802-7 (E-Book) Für Eva

TABLE OF CONTENTS

I Introduction 11

II List of Abbreviations 15

III Networks and History 16 1. Network Theory Basics 17 2. Applications – Rosenthal et al., Gould and Bearman 25 3. A Critique 31 4. Aims and Methodology – The Plurality of Mühlenberg’s Network 34

IV A Prelude – Mühlenberg’s Correspondences from 1771 to 1784 45 1 The Lutheran Context 45 1.2 Studies at the Francke Foundations (1763–1770) 48 1.3 The Pennsylvania Field 53 1.4 Pious Trade in Halle Medicines I (before 1770) 57 1.5 Pious Trade in Halle Medicines II (after 1770) 61 1.6 Halle’s Private Intermediaries 63 2 The Family Context 70 2.1 Family Life prior to the War of Independence 71 2.2 Brothers–in–Law 77 2.3 A Band of Brothers 81 2.4 The only Son 86 2.5 Conflict with Kunze 89 2.6 In the Wake of the War 93 3. The Scientific Context – the Respublica Litteraria 98 3.1 The Handmaiden of Medicine 102 3.2 American Botanical Fellows 104 3.3 Carl Linnaeus and America 108 3.4 Mühlenberg and Linnaeus 111 3.5 American Botany in the Early National Period (1775–1815) 115 3.6 The Challenges of Independence 116

V Letters from an American Botanist 122 1 Confederate Botany (Phase 1, 1784–1790) 122 1.1 A Franconian Physician at Lancaster – Schöpf 126 1.2 The Prince of Erlangen Science – Schreber 128 1.3 Commercium Litterarium with Schreber and Schöpf 131 1.4 The Hinterlands of Botanical Science 138 1.5 Franklin College and Botany 140 1.6 Old Trade along New Channels 148 8 Table of Contents

1.7 Family and Friends 152 1.8 Network Analysis: Phase 1 154

2 Transatlantic Botany (Phase 2, 1790–1797) 158 2.1 The Philosopher of Kingsessing – William Bartram 165 2.2 More Gardeners and Seedsmen – the Marshalls and William Hamilton 171 2.3 Mühlenberg’s Antagonist – Benjamin Smith Barton 174 2.4 Networks and Network Strategies 178 2.5 A Swabian in Lancaster – Autenrieth 183 2.6 Spinning the Web at the American Periphery 186 2.7 Moravians Botanists – Kramsch and Kampman 188 2.8 Cutler and the Gap in the North 192 2.9 Mitchill, New York and Agriculture 196 2.10 Erlangen falls behind 198 2.11 Hoffmann goes to Göttingen 201 2.12 Palm in Erlangen 204 2.13 Johann Hedwig’s Cryptogamia 208 2.14 James Edward Smith – Linneaus’ English Heir 211 2.15 Changes at the Orphanage 215 2.16 Yellow Fever and American Medicine 221 2.17 Network Analysis: Phase 2 228

3 A Network in Transition (Phase 3, 1797–1802) 233 3.1 The Failures of Smith, Schreber and Hoffmann 236 3.2 The Cryptogamic Circle I 243 3.3 Postbox Halle 254 3.4 Moravians on the Move 256 3.5 Dormant America 262 3.6 Network Analysis: Phase 3 264

4 Network Strategies (Phase 4, 1802–1805) 266 4.1 The South 273 4.2 Europeans in the American Wilderness I – Lyon and Pursh 280 4.3 Europeans in the American Wilderness II – Kin, Enslin and van der Schott 287 4.4 Europeans in the American Wilderness III – Rafinesque 292 4.5 The Cryptogamic Circle II 297 4.6 The Herbarium of André Michaux 301 4.7 Old Europe – Smith, Turner, Hoffmann, Schrader and Schreber 306 4.8 Network Strategies and Publications 312 4.9 The Halle Network 323 4.10 Changes in the Family 327 4.11 Network Analysis: Phase 4 329 Ta ble of Contents 9

5 An American Network (Phase 5, 1805–1811) 332 5.1 The Troubles of Erlangen and Halle 335 5.2 Nulla Salus Bello – Pacem te poscimus 340 5.3 France, England and Sweden 345 5.4 North and South – Peck, Elliott, Dunbar, Moore and Logan 355 5.5 Moravians in the South 372 5.6 The West – Müller and the Planthunters 377 5.7 The Aftermath of the great Expeditions 383 5.8 The Failure of American Botany 390 5.9 Family Matters 399 5.10 Network Analysis: Phase 5 401

6. Towards Botanical Independence (Phase 6, 1811–1815) 403 6.1 Philadelphia and Western Pennsylvania 411 6.2 The patient and successful cultivator – Collins 418 6.3 The Southern Constellation – Oemler, Dörry, Logan, Elliott and Baldwin 425 6.4 The Last of the Moravian Contacts – Schweinitz 441 6.5 Friends and Fraud in New York City 446 6.6 The North and the West – Cutler, Peck, Bigelow, Rich and Moore 453 6.7 The End of an Era 460 6.8 Ars longa, vita brevis 467 6.9 Network Analysis: Phase 6 469

VI Conclusion 472

VII Appendices 483 1. Appendix A – Flow Charts 486 2. Appendix B – Tables 494 3. Appendix C – Lists of Correspondences 535 4. Appendix D – Exchange Charts 550 5. Appendix E – Networks 554 6. Appendix F – Network Documentation 565

VIII Bibliography 569

IX Register of Persons 585

X Register of Places 598

I INTRODUCTION

This study attempts to resconstruct the correspondence network of the Lutheran pastor Gotthilf Heinrich Ernst Mühlenberg (1753–1815), the so–called “American Linnaeus.” Today, Mühlenberg is particularly remembered for his contributions to the establishment of a national scientific infrastructure in the wake of American Independence. From 1771 to 1815, he exchanged letters and specimens with Euro- pean and American botanists, plant collectors and seed traders, but he also wrote to merchants, family members, fellow Lutheran pastors and ordinary citizens. For the present study, 109 direct contacts1 of Mühlenberg could be identified from this period, exchanging a total of 998 letters with him. This number is composed of 693 actual and dated letters, and 297 reconstructed letters, which must be presumed lost or destroyed. Eight additional letters from or to Mühlenberg were undated. Recon- structed letters were identified through references found in the actual source corpus of 693 letters and Mühlenberg’s botanical diaries in the archives of the American Philosophical Society in Philadelphia. In total, a data loss rate of approximately 29,76 % must be assumed,2 which could partially be amended through Mühlen- berg’s diaries. These contained a plethora of crucial information on the development of individual correspondences, the dimensions of his botanical exchanges and personal remarks on most of his contacts and were for the first time systematically read and analyzed in the context of this study. In general, however, historical research has so far eluded Mühlenberg’s network, as a large portion of the letters were scattered across a number of American and European archives. By far the largest collections of Mühlenberg letters are found today at the Lu- theran Theological Seminary Mt. Airy, in the archives of the Historical Society of Pennsylvania, the American Philosophical Society and the Academy of Natural Sci- ences, which are all located in Philadelphia, PA.3 Minor holdings, individual letters and other manuscript material could be located at the Library Company of Philadel- phia, the Mahn Center for Archives and Special Collections, Ohio University, Ath-

1 This number is split up between 107 individual persons and two institutions, counted as individual actors of his network. There was one anonymous letter. For 21 of these 109 confirmed correspondents, no letters have survived, which reduces the core corpus to 693 letters from 88 correspondents. 2 As the reconstructed 297 letters presumably contained even more references to lost letters, this rate must even be assumed slightly higher. The earliest actual letter dates from December 4th 1771, and was written by Mühlenberg to his father Melchior Mühlenberg. The last letter was written by the widow of Mühlenberg’s cousin Carl Daniel Heinrich Bensen (1761–1805), So- phie Bensen, half a year after Mühlenberg’s death in May 1815. Sophie Bensen to Mühlenberg, 09/02/1815, APS Film 1097. 3 See bibliography of manuscript materials in the appendix for further details. In fact, the letters accomodated at Mt. Airy archives are available on microfilm in the A.P.S. reading (Mss.Film 1097), under which label they will also be cited. 12 Introduction

ens, OH, the Archives of the Gray Herbarium, Cambridge, MA and the William L. Clements Library, University of Michigan, Ann Arbor, MI. Further material could be located at Harvard University Archives, Cambridge, MA, the Rare Books and Manuscripts division of the Boston Public Library, Boston, MA, Franklin&Mar- shall College Archives in Lancaster, PA, the Hunt Botanical Institute, Carnegie Mellon University, Pittsburgh, PA, the Library of Congress, Washington D. C, the Trexler Library of Muhlenberg College, Allentown, PA and the William L. Clem- ents Library, Ann Arbor, MI. One Mühlenberg letter is in the private possession of Daniel Weinstock M. D. of Geneva, NY, which has been generously made available to the author by the owner.4 In Europe, the archives of the Francke Foundations at Halle, Germany, the University Archives of the Friedrich-Alexander Universität, Erlangen-Nürnberg, Germany, and the Linnean Society of London hold the greatest number of documents in the collections of Mühlenberg’s correspondents Sebastian Andreas Fabricius (1716–1790), Gottlieb Friedrich Stoppelberg (†1797), Joseph Friedrich Nebe (1737–1812), Johann Christian Daniel Edler von Schreber (1739– 1810) and James Edward Smith (1759–1828). Further material is accommodated at the Museum für Naturkunde, Alexander von Humboldt Universität Berlin, in the archives of the Martin-Luther Universität Halle-Wittenberg and in the historical image and manuscript collections of the Museum der Naturkunde, Alexander von Humboldt Universität zu Berlin. Apart from these manuscript sources, Mühlenberg’s letters have not been edited to date with the exception of William Darlington’s Reliquiae Baldwiniae (1843), containing the complete correspondence between Mühlenberg and William Baldwin (1779–1819), and the edition of Heinrich Melchior Mühlenberg’s letters by Kurt Aland (vols. 1–4) and Hermann Wellenreuther (vol. 5).5 This also reflects the general state of historical literature on Gotthilf Heinrich Ernst Mühlenberg. Compared to the historical literature on the Mühlenberg family, and especially in comparison to past and current research on Henry Melchior Mühlenberg (1711– 1787), Frederick Augustus Conrad (1750–1801) and John Peter Gabriel (1764– 1807), who found a biographer with Henry Augustus Mühlenberg (1823–1854) as early as 1849, Henry, the botanist in the family, has largely been skipped. Apart from a small number of eulogies and entries in biographical dictionaries published between Mühlenberg’s death in 1815 and the American Civil War, it was only with J. M. Maisch’s speech “Gotthilf Heinrich Ernst Mühlenberg als Botaniker (1886)” and William J. Youman’s biographical sketch in “Pioneers of Science in America” (1896) that the historization and documentation of Mühlenberg’s scientific activities began for real. In the 1920s and 1930s, Herbert H. Beck and A. S. Hitchcock unearthed more biographical information on Mühlenberg and his herbarium, al-

4 To Turner, 02/21/1803, Weinstock. 5 The original letters and manuscripts of the Darlington edition are accomodated at Mertz Li- brary, New York Botanical Garden, NY. In the course of the present study on Mühlenberg’s correspondences, which was part of a larger DFG-financed science project entitled Atlantische Korrespondenzen: Genese und Transformation deutsch-amerikanischer Netzwerke 1740–1870, the author also composed an online edition of 100 Mühlenberg’s letters, which will be available from January 2012. Introduction 13 though both men only examined a small portion of the letters used in the present study and made no use of the A.P.S. diaries at all. Mühlenberg’s herbarium, which is today housed at the Academy of Natural Sciences in Philadelphia, was subjected to a thorough examination by Shiu–Ying Hu and E. D. Merril in 1949. Paul A. W. Wallace’s work on the Muhlenberg family (1950) also contained a brief chapter on his botanical activities, which formed the basis of C. Earle Smith’s biographical sketch in 1962. Finally, Wolf-Dieter Müller-Jahncke’s article from 1977 has a special focus on Mühlenberg’s relations with German-speaking botanists, while James Mears traced “Some Sources of the Herbarium of Henry Muhlenberg” in the following year. In none of these articles, however, the extensive correspondence network was ever addressed as a whole, and the author hopes to have filled the gap and therefore to provide a sound basis for further historical and scientific research.6 The following study has been conceived as an ego-network approach to the correspondences of Henry Mühlenberg. For this reason, chapter III contains an introduction to the basics of network theory, their general applicability in historical contexts and their actual application used in this study. Chapter IV covers Mühlen- berg’s biography from his birth to the visit of Johann David Schöpf in late 1783, after which he became an independent transatlantic correspondent and networker. This chapter aims to place him in three individual contexts: within his family, within Pietism in general and the Halle Pietists in detail, and finally within the scientific context of the American Republic of Letters. Chapter V contains the main body of the present study. Here, Mühlenberg’s network will be discussed in six individual subchapters that correspond to consecutive phases in his web of correspondence. Chapter VI contains the conclusion to the study, which has been formatted according to Chicago format. In order to keep reference information about the 697 surviving letters short, Mühlenberg’s name was generally omitted. A regular letter from Mühlenberg (to Zaccheus Collins, for instance) will therefore only be cited as: “To Collins, 07/14/1812, ANSP Coll. 129.” In turn, a letter to Mühlenberg (from Wil- liam Baldwin here) will simply be cited as: “From Baldwin, 05/26/1812, Darling- ton, Baldwiniae, 62.” All other letters are referenced in full. I would like to give my kindest regards and thanks to the many archivists and fellow historians who have listened patiently to my ideas during presentations and visits to archives in the USA and Europe, and have generally helped greatly with comments, remarks and criticism. My special thanks go to my Doktorvater Prof. Dr Mark Häberlein at the University of Bamberg, who offered corrections, advice and directions where it was necessary, but also let me trustfully follow and develop my own ideas. Without this kind mixture of general trust and detailed criticism, it would have been impossible to sustain a high level of motivation for almost four years and

6 In 1978, James Mears stated that “[b]efore any source of specimens of the Muhlenberg Herbar- ium can be complete, Muhlenberg’s botanical manuscripts must be transcribed and associated with the collections.” Mears, “Some Sources,” 155. A list in the archives of the Academy of Natural Sciences in Philadelphia, there is actually a preliminary list entitled “Botanical Corre- spondences of G. Henry Muhlenberg located by 22 September 1981,” which suggests that Mears or one of his colleagues actually started to compile material on Mühlenberg. The list comprises letters from 1781 to 1815. 14 Introduction

to finish the dissertation in the present form. Prof. Häberlein was also a member of the DFG research project Atlantische Korrespondenzen: Genese und Transforma- tion deutsch-amerikanischer Netzwerke 1740–1870, whose other members, Prof. Dr Hermann Wellenreuther, Prof. Dr Claudia Schnurmann, Christina Urbanek M.A., Anna Groeben and Sarah Lentz, I would like to thank for the many inspira- tions and thoughts on historical network studies, delicious food and great company during our workshops. Prof. Dr. Gabriele Lingelbach (University of Kiel) was the second corrector of my dissertation and I kindly thank her for the maximum of professional advice and support she was able to give me in a minimum of time. The Gesellschaft für Überseegeschichte e.V. (GÜSG) has awarded the honor of the Mar- tin Behaim Award to my dissertation, which included the publication of the present text in the Franz Steiner Publishing House (Stuttgart). There, Harald Schmitt was a constant source of tips and support in handling the final edition of my dissertation. Just like Mühlenberg’s own botanical studies, this study would hardly have been possible without a proper network of indefatigable proofreaders, name com- pilers and friends: Dr Toban Szuts (Harvard University, MA), Allen Flint (Ohio University, OH), Wendy Withers-Bassingthwaite (Governors State University, IL), Georg Schafferer M.A., Ms Nina Tschöpp, Marco Eckerlein M.A. and Daniel Gla- ser (all University of Bamberg, Germany), whose help was essential for the final edition and register of the present text. Ms Ulrica Hansson B.A. (University of Sussex, UK) and Dr Mats J. Hansson (Ersta Sköndal University College, Sweden) have gracefully helped me understand and translate some Swedish-language source passages by Olof Swartz. The indurad GmbH (Aachen, Germany) was so kind as to provide the workforce of research assistant Dominik Giesen (University of Kon- stanz, Germany) as further support for compiling the registers of persons and places. Finally, my wife Eva has read the entire manuscript, found more spelling errors than I ever thought possible and offered moral support and understanding whenever 18th century script became all too hard to understand. This book is for her.

II LIST OF ABBREVIATIONS

AFst Archiv der Franckeschen Stiftungen, Halle, Germany ANSP Academy of Natural Sciences, Philadelphia, PA APS American Philosophical Society, Philadelphia, PA BPL Boston Public Library – Rare Books and Manuscripts, Boston, MA Clements Lib University of Michigan, William L. Clements Library, Ann Arbor, MI FMC Archives and Special Collections Franklin & Marshall College, Lancaster, PA HSP Historical Society of Pennsylvania, Philadelphia, PA HUB Museum der Naturkunde der Humboldt Universität zu Berlin, Historische Bild und Schriftgutsammlungen, Germany HUH Harvard University Herbaria, Archives Gray Herbarium, Cambridge, MA Hunt Hunt Institute of Botanical Documentation, Pittsburg, PA LibComp Library Company of Philadelphia, Philadelphia, PA LoC Library of Congress, Manuscript Division, Washington D. C. LSoL Linnean Society of London, London, UK MCollege Muhlenberg College, Trexler Library, Allentown, PA NCHGS Northampton County Historical and Genealogical Society, Easton, PA OUAr Ohio University, Mahn Center for Archives and Special Coll., Athens, OH UAE Archiv der Friedrich–Alexander Universität Erlangen, Germany UAH–W Archiv der Martin–Luther Universität Halle–Wittenberg, Germany UoV University of Virginia, Albert and Shirley Small Special Collections Library, Charlottesville, VA Weinstock Private Property of Daniel Weinstock M. D., Geneva, NY YUL Yale University Libraries – Manuscripts and Archives, New Haven, CT

III NETWORKS AND HISTORY

The development of social network studies into a scientific paradigm largely took place after World War II, though important groundwork had already been done by the late nineteenth and early twentieth century. The names of Émile Durkheim (1858–1917), Georg Simmel (1858–1918) and the sociometric school of Jacob Moreno (1889–1974) are most often invoked in this context.1 Network studies’ first boom came in the 1940s and 1950s with the rise of British anthropology and its new focus on urban environments and the rules governing human sociability. At that time, some of the most basic concepts of network studies still in use today, such as “centrality”, “multiplexity”, “transitivity” and “structural equivalence,”2 were first conceived and applied. Scholars like Albert Radcliffe-Brown (1881–1955) and the “Manchester School of Anthropology” of Max Gluckman (1911–1975), Elizabeth Bott and John A. Barnes (b. in 1918) pioneered this new field of study and set the frame for all subsequent developments. The 1960s and 1970s saw further elabora- tions in theory and methodology and the firm establishment of social network studies within the scientific landscape with the founding of the International Network for Social Network Analysis (INSNA) and specialized journals.3 During the 1960s, the network paradigm was first accepted into American sociology as an independent research perspective, the basis for which had been laid earlier by the translation of Simmel’s work into English after World War II.4 Ste- phen Berkowitz and Barry Wellman have described this paradigmatic shift as “a broadly based scholarly movement away from the Aristotelian-Linnean tradition of analysing things in terms of the intrinsic characteristics of their individual parts.”5

1 “Theoretical ‘precursors’ of network analysis have often been invoked in passing – especially Durkheim and Simmel – but network analysis, itself a constellation of diverse methodological strategies, has been systematically grounded in the conceptual frameworks they elaborated. In addition, there has been a notable absence in this literature of any sustained consideration of the potential usefulness of network analysis for historical investigation.” Emirbayer, “Problems of Agency,” 1412. For Simmel, Moreno et al. see also Faust and Wasserman, Analysis, 11f; Di- az-Bone, Ego-zentrierte, 6–21; Marsden and Lin, Social Structure, 9; and Degenne and Forse, Introducing, 28. 2 Faust and Wasserman, Analysis, 11–13; Wellman, “Structural analysis,” 21; and Lipp, “Räum- liche Muster,” 52. 3 “Since the seminal works of Barnes (1954) and Bott (1971) sociological studies utilizing network analysis have appeared with increasing frequency; a veritable explosion of such work has taken place over the last 15 years, particularly with the founding of two specialized journals Social Networks and Connections, in the late 1970s.” Emirbayer, “Problems of Agency,” 1411; Degenne and Forse, Introducing, 12. Actually, Connections was first published as the INSNA’s internal newsletter. Marsden and Lin, Social Structure, 9. 4 Berkowitz and Wellman, “Introduction,” 1; Wellman, “Structural Analysis,” 22. 5 According to Berkowitz and Wellman, this encompasses nearly all fields of scientific study, including physics, quantum physics, biology and linguistics. Berkowitz and Wellman, “Intro- Network Theory Basics 17

In American sociology and historiography, network concepts first came into broad use in the field of “Community studies.”6 Here, they helped to replace the “Commu- nity-lost-theory,” which contended that social changes set off by nineteenth century industrialization had atomized communities and led to a complete loss of social cohesion. Network scholars successfully challenged this view and were able to prove that socially cohesive forces had merely changed, rather than disappeared altogether. The “new” communities had to be looked for in more extensive geogra- phic contexts – they were “liberated” as new forms and patterns of connectivity appeared.7 With the publication of Stanley Wasserman and Katherine Faust’s seminal “Social Network Analysis: Methods and Applications“ in 1994, which provides an extensive overview of the discipline’s potential and scope, network studies reached a new level of sophistication, while the book itself has since become the “Bible of network analysis“.

1. NETWORK THEORY BASICS

At its most fundamental level, network studies differentiate between actors within a system and the kinds of ties linking them, and it is the latter that are emphasized in the analysis.8 Network studies “analyse the ordered arrangements of relations that are contingent upon exchange among members of social systems. They map these structures, describe their patterns (often using a set of tools derived from ma- thematical graph theory), and seek to uncover the effects of these patterns on the behavior of the individual members of these structures – whether people, groups or organizations.”9 Individual attributes10 assigned to actors, or classifications such as

duction,” 4. Emirbayer and Goodwin agree to this view: “From this historical vantage point, contemporary network analysis can be viewed as part of a second crucial watershed period in American sociology, one in which empirical research is now directing its attention back again to the systemic level, this time assisted by the development of quantitative techniques and methods of a highly sophisticated nature.” Emirbayer and Goodwin, “Problems of Agency,” 1417. 6 “The network approach has been of special importance in the study of the urban community, in large part because it is well suited to test the implications of theories of social change which suggested that industrialization and bureaucratization were to create a mass society marked by the atomization of individuals and the withering away of informal relations of friendship, neighboring, and helping to support.” Marsden and Lin, Social Structure, 14. See also Bender, Social Change, especially chapter 4: “Social Networks and the Experience of Community;” Wellman and Wetherell, “Historical Communities;” Wellmann, “Studying Personal Communi- ties;” 62f. and Russo, Families and Communities, passim. 7 Diaz-Bone, Ego-zentrierte, 21. 8 “It is a comprehensive paradigmatic way of taking social structure seriously by studying directly how patterns of ties allocate resources in a social system. Thus, its strength lies in its integrated application of theoretical concepts, ways of collecting and analyzing data, and a growing, cumulating body of substantive findings.” Wellman, “Structural Analysis,” 20, 30. See also Orser, Race and Practice, 119; Wasserman and Faust, Analysis, 4. 9 Berkowitz and Wellman, “Introduction,” 3. 10 Individual attributes of actors are not fully discarded, though. In other ways, they can be fruit- 18 Networks and History

age, sex, income, class-affiliation etc. are deemphasized in favor of looking at the actual connections between actors, which sometimes contradict any a-priori cate- gorizations11 and promise to yield more realistic and reliable models of social structure. This is what Mustafa Emirbayer and Jeff Goodwin have termed the anticategorical imperative, which “rejects all attempts to explain human behavior or social processes solely in terms of the categorical attributes of actors, whether individual or collective.”12 In the course of its development, the network paradigm has been applied to all kinds of networks between people, groups of people, companies, organizations, NGOs and nations. Therefore, the term actor is commonly employed as a neutral agent to refer to all of these, whereas the term relation is used in reference to the sum of all ties that connect actors in a network.13 A tie between two actors is referred to as a dyad,14 and is the most commonly used unit of analysis in network approaches in which measurements are taken. Fur- thermore, there are triads (between three actors) and larger subgroups, which can be singled out as units of analysis according to the respective aims, interests and intentions of a particular research project at hand.15 Clusters and Cliques, two closely resembling but different forms of subgroups, will be of most interest in the study of Mühlenberg’s botanical network. While a clique is defined strictly as “a set

fully combined with network data in order to enrich the analysis. In network studies, individual non-structural attributes are referred to as “compositional attributes.” Faust and Wasserman write that “(...) attributes of the actors may also be included. Measurements on actors will be referred to as network composition. Complex network data sets may contain information about the characteristis of the actors (such as the gender of people in a group, or the GNP of nations in the world), as well as structural variables.” Faust and Wasserman, Analysis, 21, 29. 11 “One problem is that the categorical attributes of individuals are not powerful predictors of subsequent action – at least in this period. Both historians and sociologists, wedded to the traditional categories of class, status, and party as the units of analysis, have failed to identify the tangible bases of action which are, in historical context, often independent of the strata or class memberships of individuals.” Bearman, Relations, 8. For a network criticism of Pierre Bourdieu’s social categories see Lipp, “Räumliche Muster,” 51f. 12 Emirbayer and Goodwin,”Problems of Agency,” 1412. 13 “The collection of ties of a specific kind among members of a group is called a relation.” Faust and Wasserman, Analysis, 19. 14 “Dyadic analyses focus on the properties of pairwise relationships, such as whether ties are reciprocated or not, or whether specific types of multiple relationships tend to occur together.” Faust and Wasserman, Analysis, 18. In Alexander Pyrges’ analysis of the Ebenezer network from 1732–1828, a dyadic relationship is merely one of four types of “mutually engaging and multi-directional” relationships under scrutiny. The three other types consist of “enclosings,” “forwarders” or “switchers,” and letters transported by migrants. Pyrges, “Ebenezer Network,” 59. 15 Groups, however, are not to be confused with cliques among actors. Whereas groups can be identified on the basis of distinguishing features, a clique is defined as a minimum set of three actors with a maximum number of links in between them. Faust and Wasserman explain that “[o]ne must be able to argue by theoretical, empirical, or conceptual criteria that the actors in the group belong together in a more or less bounded set. Indeed, once one decides to gather data on a group, a more concrete meaning of the term is necessary. A group, then consists of a finite set of actors who for conceptual, theoretical, or empirical reasons are treated as a finite set of individuals on which network measurements are made.” Faust and Wasserman, Analysis, 18f. Network Theory Basics 19 of completely interconnected nodes”16 with the minimum of three nodes, clusters can be identified more loosely on the basis of common traits, qualities or properties. This use of the term has to be distinguished from the method of “hierarchical clus- tering,” a data analysis approach applied in blockmodelling techniques, which “groups entities into subsets, so that entities within a subset are relatively similar to each other.”17 In this study, however, the term cluster will be used in a more general sense,18 to refer to subgroups that show particularly tight interconnectedness within specific geographical boundaries – mostly within cities.19 The discussion of subgroups has become a major aspect of social network analysis, as it helps to identify centers of heightened cohesion, collaboration and action and therefore to uncover the internal dynamics of a social system.20 This dissection of networks into smaller units corresponds to the different levels of analysis to which network studies are typically applied.21 The ultimate goal will be to reintegrate all these individual levels into a single organic panorama that allows for deductions, generalizations and projections of how actors in a specific network are going to act and how the network as a whole is going to evolve. The anticategorical imperative brings the ties connecting actors to the center of attention. Most of the methodological refinements of the past 40 years of network research have exclusively dealt with ties, or, more precisely, with ways to refine our understanding and application of them. Consequently, ties have been described and differentiated in increasingly subtle and detailed ways, in order to interpret any complex arrangement of relations that might appear in a network. For instance, we speak of directional or non-directional, valued or weighted ties when a tie between two actors is not simply dichotomous (existent/non-existent) but takes on specific “quality.” A directional tie stands for an asymmetric or non-reciprocal relationship, such as trade among nations or choices of friendship among children.22 In a graphic representation of a network, non-reciprocal ties are represented through arcs, poin-

16 Degenne and Forse, Introducing, 80; “A clique in a graph is a maximal complete subgraph of three or more nodes. It consists of a subset of nodes, all of which are adjacent to each other, and there are no other nodes that are also adjacent to all of the members of the clique.” Faust and Wasserman, Analysis, 254; See also Diaz-Bone, Ego-zentrierte, 43. 17 Faust and Wasserman, Analysis, 381. For an example of how clusters can be applied, see the discussion of Rosenthal’s study below on pages 18f. 18 Faust and Wasserman, Analysis, 19. 19 This study’s focus will be on the “Republic of letters” and how Mühlenberg participated in it. The social organization of scholars in early modern Europe and the young American republic was a decidedly urban phenomenon, which implies that much of its developments originated from these “clusters of science.” See below on pages 98f. 20 A study on the Chicago-based Western Electric’s Hawthorne plant conducted by Roethlisberger and Dickinson in 1939 found that the existence of cliques and subgroups among workers had immense impact on the factory’s productivity. Ever since, it has become “essential to study networks and identify subgrous from the standpoint of cohesion for a proper understanding of a social system.” Degenne and Forse, Introducing, 79. 21 “Social networks can be studied at several levels: the actor, pair or dyad, triple or triad, sub- group, and the group as a whole.” Faust and Wasserman, Analysis, 97. 22 “The import/export of goods between nations is an example of a directional relation. Clearly goods go from one nation to another; one nation is the source and the other is the destination of 20 Networks and History

ting from one actor to another. This construct will be helpful to depict flows of information within Mühlenberg’s botanical network. “Valued” or “weighted,” on the other hand, are synonymous adjectives ascribed to ties that “take on a range of values, indicating the strength, intensity, or frequency of the ties between each pair of actors.”23 In a correspondence network, however, we observe that some contacts exchange a smaller number of letters with each other while others confide much more personal information to one trusted correspondent than they would to any other. All of these network-relevant data would be lost in a dichotomous representation or discussion of the network. Therefore, the idea of valued ties is particularly helpful in order to come up with a viable graphic reflection of the various personal relationships Mühlenberg had with his correspondents. A variation of the same theme is Mark Granovetter’s often quoted concept of the “strength of weak ties,” which can serve as a working model for the development of Mühlenberg’s network and the dynamics of information exchange. Grano- vetter defines strong ties as a “combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and the reciprocal services which cha- racterize the tie,”24 and argues that if actor A in a triad has strong ties to actor B and actor C, it is highly unlikely that B and C have not established contact as well.25 Due to this we understand that networks have a natural tendency towards fragmentation and strongly interconnected subgroups, while information flow between subgroups is primarily channelled through weak ties. In Granovetter’s words, “this means that whatever is to be diffused can reach a larger number of people, and traverse greater social distance (...), when passed through weak ties rather than strong.”26 In order to substantiate his observations, Granovetter used examples from “diffusion studies,” which concentrate on such issues as the spread of diseases, technological innovations and the importance of weak ties on the job market.27 Therefore, it will be an important step in our analysis to distinguish strong and weak ties in Mühlenberg’s network and to assess their impact on the development of his correspondence and the information flow within it.28

the goods.” Faust and Wasserman, Analysis, 121. See also Diaz-Bone, Ego-zentrierte, 40; and Breiger, “Analysis,” 507. 23 Faust and Wasserman, Analysis, 44. For other examples of weighted ties, see Degenne and Forse, Introducing Social Networks, 79–82, or the discussion of Rosenthal et al., “Social Move- ments,” below on pages 18f. Signed graphs, however, indicate the specific quality of a relationship, such as “A loves/hates B” or “C is in war/allied with D,” depending on the scenario. Faust and Wasserman, Analysis, 136. 24 By implication, this definition of a strong tie also makes it impossible for any actor to maintain an indefinitely large number of strong ties, thus preventing the possible emergence of “a network of strong ties.” Granovetter, “Strength,“ 1361. See also Degenne and Forse, Introducing, 109f. 25 This is Granovetter’s Forbidden Triad. Granovetter, “Strength,” 1363. 26 Granovetter, “Strength,” 1364–66. 27 Granovetter, “Strength,” 1367–68. For further discussion see “Weak ties in egocentric networks” in Granovetter, “Strength,” 1369f.; and Diaz-Bone, Ego-zentrierte, 44f. 28 “In Europe, Lefebvre (1989), in his classic work, described the spread of a wave of rural panic in 1789 through weak ties (travelers, messengers, family, friends) between villages. (...) Like- wise, Gould (1991, 1993, 1995) described how weak ties among Parisian neighborhoods Network Theory Basics 21

Network study’s signature focus on ties and relations, however, has not resulted in a complete abandonment of social categories. In fact, much of the explanatory power of network approaches lies in their ability to incorporate social categorizations and describe how they play out in network projections of social formations and action. As Christopher Ansell and John Padgett have pointed out in their 1993 study on the Medici network: “We do not argue (...) that social attributes and groups are irrelevant to party formation; merely that their role needs to be understood within a deeper relational context. There is no simple mapping of groups or spatial dimensions onto parties; social attributes and group interests are ‘merely’ cognitive categories, which party mobilization, networks, and action crosscut.”29 Therefore, social markers such as class, group interests etc. must not be seen as in opposition to, but rather as an expression of network structure. Barry Wellman has observed that “people belong to networks as well as to categories. Structural analysts believe that categorical memberships reflect underlying structural relationships, that is, pat- terned differences in the kinds of resources with which they are linked. They do not treat social class, for example, as a set of statuses occupied by members of a population, but as a summary label for economic relations of power and dependency.”30 Analogous to this correlation of social categories and network structure, individual behavior31 must also be seen as a function of the actor’s relational embeddedness. Ultimately, this means that the social composition of Mühlenberg’s correspondence network must be taken into account.32 Ties and relations are distinct features of network terminology and theory. While this focus remains fundamental, network studies differ greatly in their methodological design and conception of the network under scrutiny. The most fundamental distinction is the one between whole-network analysis and ego-centered network analysis.33 In the latter case, a network is presented from the perspective of

helped in the mobilization for the Paris Commune of 1871.” Barkey and van Rossem, “Net- works of Contention,” 1349. 29 Padgett and Ansell, “Robust Action,” 1274. 30 Wellman, “Structural analysis,” 32. 31 “Zu den theoretischen Annahmen der Netzwerkanalyse gehört, daß Netzwerke und die Bezie- hungsmuster, die sich ergeben, wie auch der Typ der Beziehungen, maßgeblich die Hand- lungsspielräume von Menschen bestimmen, und daß diese interaktiven Prozesse für das Ver- halten eine größere Rolle spielen als soziale Attribute wie Schicht oder Einkommen.” Lipp, “Räumliche Muster,” 56. See also Degenne and Forse, Introducing, 4. 32 With respect to social diversity, Rainer Diaz-Bone suggests to identify “status groups” within the ranks of the alteri, based on their socioeconomic potential. He warns that the degree of social diversity directly correlates with the total dimensions of the network, which can lead to data distortion. “Die Verwendung der Netzwerkgröße als Maß für die Spannweite der Ausprä- gungen im Netzwerk ist nur dann sinnvoll, wenn begründet werden kann, daß sich die alteri unähnlich sind. Dann ist die Netzwerkgröße ein Maß für Range im Sinne der Diversität (…) der alteri (…) und es kann angenommen werden, daß eine Größenzunahme [des Netzwerks] die Variation der Merkmalsausprägungen erhöht.” Diaz-Bone, Ego-zentrierte, 57, 60. 33 Network analysts also differentiate between one-mode and two-mode networks. “One-mode networks, the predominate type of network, study just a single set of actors, while two-mode networks focus on two sets of actors, or one set of actors and one set of events.” Faust and Wasserman, Analysis, 35–42. For instance, a two-mode network set in the early modern scien- 22 Networks and History

one single focal actor, ego, whose personal ties and social embeddedness define the boundaries of the network. Ego’s ties to other actors – called alteri – and the ties these alteri maintain among each other constitute its social tissue.34 Typically, personal network studies are survey-based and conducted to study social support and family networks.35 Barry Wellman has observed that “[r]ather than showing the universe as it is perceived by an outside observer, [personal networks] provide Pto- lemaic views of networks as they may be perceived by the individuals at their centers.”36 The main advantage of ego-centered network approaches lies, therefore, in the reduction of the often forbidding complexity and size that make whole-network approaches so unwieldy. Heuristically, they offer a micro-perspective on people’s connectedness and allow us to identify relations that typically occur within specific kinds of networks. Elizabeth Bott is usually credited with the invention of the concept of ego-networks in her 1957 study Family and Social Network. Roles, Norms, and External Relationships in Ordinary Urban Families. In what has become a classic of network literature, Bott compares the personal networks of 20 London-based couples and shows that there is a close correlation between family structure and household- role segregation.37 In this study, a lesser degree of overlapping in the personal networks of couples was found to correlate directly with a stricter segregation of household roles. Additionally, Bott contends that higher levels of interconnectedness among personal network alteri result in a heightened importance and observance of social norms, which, in the case of her 20 London couples, was typically expressed through local or “quarter”-patriotism. To describe this network property, Bott employed the terms “close-knit” and “loose-knit” communities, which have

tific community could be conceived as consisting of individual scientists and contemporary scientific communities. For alternative methodological categorizations of networks see also Diaz-Bone, Ego-zentrierte, 46–48; and Breiger, “Analysis,” 507–509. 34 Faust and Wasserman, Analysis, 42f. 53f.; Lipp, “Räumliche Muster”, 53; Depending on the theoretical framework, the alteri’s contacts to other alteri – actors ego is not linked to in any way – may also figure in an ego-network. See Granovetter, “Strength,” 1370. Granovetter fol- lows here A. L. Epstein’s distinction between “effective” and “extended” networks. Granovet- ter, “Strength,” 1369, 1372. 35 “Untersucht wird dabei zum Beispiel, wie groß die Wahlfreiheit der Personen eigentlich ist und wie stark ihre Freundschaftsnetze auf sozialer und kultureller Homologie aufbauen.” Lipp, “Räumliche Muster,” 53. See also Faust and Wasserman, Analysis, 42f. 53f. 36 “These egocentric network studies have documented the pervasiveness and importance of connectivity, thereby rebutting mass society contentions that recent large-scale social transforma- tions have produced isolation and alienation. Numerous scholars have described how networks link individuals through strong and weak ties, situate them in larger social systems, and affect the flows of resources to and from them.” Wellman, “Structural analysis,” 27. 37 “Die klassische und oft gegengeprüfte Untersuchung von Elizabeth Bott demonstriert am Beispiel von Londoner Familien, daß zwischen der Struktur des Familiennetzwerks und der Geschlechterrollendifferenzierung in einer Familie ein enger Zusammenhang besteht. (…) Wie auch eine ganze Reihe von Panelstudien zu Botts Thesen zeigten, fiel im Hinblick auf den Grad der Rollensegregation die Struktur des Familiennetzes auffällig stärker ins Gewicht als harte Faktoren wie Beruf, Ausbildung, Mobilität oder Wohngebiet.” Lipp, “Räumliche Muster,” 55. For Bott, see also Degenne and Forse, Introducing, 46. Network Theory Basics 23 subsequently been replaced by the term network density.38 Ever since, density has come to be strongly associated with group pressure, social norms and control.39 In conclusion, whole-network approaches seek to detect network structures within larger social groups, in order to detect flows of information, coalitions and potential lines of conflict. Inevitably, the network’s boundaries have to be defined in a way that neither includes redundant parts nor excludes essential ones, both of which would compromise the results in a negative way. Ego-networks, on the other hand, construct networks from the point of view of one focal actor and focus on its internal dynamics to make abstract claims about people in similar kinds of networks or compare it to similar kinds of personal networks.40 Since the 1940s, a variety of different measurements, descriptive and interpretative tools have been developed41 to discuss the overall constellation, quality and quantity of relations governing a network. It is the careful application of these that gives texture to networks and significance to their systematic study and description. Although there is a basic set of interpretative tools that are commonly applied to virtually all network approaches, their choice is in part predetermined by the initial choice of network perspective.42 Apart from the size of the network, which refers to the sheer number43 of all alteri named by ego (n-1) at a given time, density, centrality, betweenness and multiplexity are all important measurements to describe the specific features of an early modern correspondence network. Density (Delta Δ) indicates the proportion of actually existing ties of one particular type of relation to all potential ones. For example, a network featuring ten

38 Bott, Family and Social, 10, 63f., 73f. 39 Lipp, “Räumliche Muster,” 57. 40 For a slightly different use of the term “actor-centered” in connection to network studies see Lothar Krempel and Carola Lipp’s discussion of the 1848/49 petition movement in the Baden-Württembergian town of Esslingen, “Petitions and the Social Context of Political Mo- bilization in the Revolution of 1848/49: A Microhistorical Actor-centered Network Analysis.” Their approach focuses on overlapping memberships in local political factions and efforts to account for the spread of the petition movement during the revolutionary events of 1848/49. Krempel and Lipp largely base their argumentation on visualizations of the Esslingen network and a correlation model of the citizens’ social standing to their degree of exposure to other activists. Krempel and Lipp, “Petitions,” passim. 41 “Many of the key structural measures and notions of social network analysis grew out of keen insights of researchers seeking to describe empirical phenomena and are motivated by central concepts in social theory. In addition, methods have been developed to test specific hypotheses about network structural properties arising in the course of substantive research and model testing.” Faust and Wasserman, Analysis, 4. 42 Methods and concepts come not only associated with specific perspectives, but also pertain to different levels of analysis. Properties and measures such as Prominence, Centrality and Pres- tige are all structural variables of individual actors, while measures and methods like Connect- edness, Density and Diameter are qualifications about the entire network. Wasserman and Faust, Analysis, 25. For another discussion of network measurements in personal networks see chapter 2.6 “Maße zur Beschreibung der Struktur von ego-zentrierten Netzwerken” in Di- az-Bone, Ego-zentrierte, 56. 43 Rainer Diaz-Bone introduces Burt’s Range concept as an alternative way to capture the correlation of network size and its degree of internal socioeconomic variety. Diaz-Bone, Ego-zentrierte, 57. 24 Networks and History

potential ties and another six ties actually linking a number of actors has a density of 0.6. Network density is routinely correlated with interpretations of social control, observance of norms, mobilization and information flow.44 Centrality as a measurement used to identify actors “that are extensively involved in relationships with other actors”45 was first described and employed in the 1940s46 and has since become one of the most widely used and modified network measurements. Katherine Faust and Stanley Wasserman distinguish actor centrality, degree centrality, close- ness centrality and betweenness centrality, all of which differ slightly in their definition and interpretation of centrality.47 Generally speaking, “sociological and economic concepts such as access and control over resources, and brokerage of information, are well suited to measurement. These concepts naturally yield a definition of centrality since the difference between the source and the receiver is less important than just participating in many interactions. Assuming that one is studying a relevant relation (such as communication), those actors with the best access, most control, or who are the most active brokers, will be the most central in the network.”48 Following this assumption, central actors will be viewed as information multipliers and bridges between otherwise unconnected nodes or parts of the network. This role of central actors will also be referred to as betweenness and can be interpreted in terms of communication control and power.49 In this respect, actors with a high degree of betweenness are considered to occupy a strategic position in Mühlenberg’s network, allowing them to control flows of information and resources.50 It is neces-

44 “Densely knit networks, for example, could mobilize resources more rapidly than networks in which resources would flow to some members through longer chains.” Wellmann, “Studying Personal Communities,” 63. See also Faust and Wasserman, Analysis, 101; Degenne and Forse, Introducing, 21; Diaz-Bone, Ego-zentrierte, 58. 45 Faust and Wasserman, Analysis, 173. 46 “The definition of centrality was first developed by Bavelas (1948, 1950). The idea was applied in the late 1940’s and early 1950’s in laboratory experiments on communication networks (rather than from observed, naturally occurring networks, directed by Bavelas (...).” Faust and Wasserman, Analysis, 173. 47 For an elaborate discussion see Faust and Wasserman, Analysis, 173–193; Lipp, “Räumliche Muster,” 59; and Emirbayer and Goodwin, “Problems of Agency,” 1419–21. 48 Faust and Wasserman, Analysis, 174. 49 “Interactions between two nonadjacent actors might depend on the other actors in the set of actors, especially the actors who lie on the paths between the two. These ‘other actors’ potentially have some control over the interactions between the two nonadjacent actors.” Faust and Wasserman, Analysis, 188. Pyrges calls these positions “switchers.” “Switchers are nodes whose power stems from the connecting, mediating, channeling or supervising of communca- tion rather than from the control of scarce resources.” Pyrges, “Ebenezer Network,” 59. 50 Carola Lipp and Lothar Krempel’ study of the 1848 petitions movement, betweennes turned out to be a far more decisive measure to gauge the network’s dynamics. Whereas the most central actors showed very poor connections to the periphery, actors with a high degree of inbetween- ness provided cohesion and peripheral integration, which turned out to be crucial for mobilization. “Personen, die Positionen in der Mitte des Organisationsnetzes einnahmen, gehörten zwar zentralen Organisationen an, besaßen aber in vielen Fällen wenige Verbindungen in die Außen- bereiche des Netzes, da sie Kontakt nur mit ihnen nahestehenden Personen pflegten. Sie übten damit geringe Effekte auf die Gesamtbewegung aus und trugen eher zur Schließung des Netzes bei, statt es zu öffnen. Umgekehrt erwiesen sich Personen, die gleichzeitig Verbindungen zum Applications 25 sary to add, however, that both centrality and betweenness must be treated with caution in a personal network, as the focal person will naturally appear “central” due to the underlying methodological setup. Experience shows that people in networks tend to be linked through various channels, such as kinship, friendship, professional association, common interests or memberships. This fact has been conceptualized as multiplexity, which indicates the simultaneous existence of different kinds of relations within the same set of actors, and represents a measurement for network cohesion and integration.51 In Mühlenberg’s case, correspondents may appear as relatives, as partners in trade with Halle medicines and books, as professional colleagues or friends, and sometimes in more than one of these contexts. Hunter Dupree has pointed out the republic of letters’ tendency to combine various kinds of relations into one network: “The overlapping membership in the eighteenth century made possible the combining of cultivators, practitioners, and researchers into one society. At the same time each of these members were also connected with other groups, a profession, a church, a political club, a social class, or a family. This multiple mapping of an individual into many templates has often misled observers, because the concept of a single relationship as the one that defines identity to the exclusion of all others is a fallacy.”52 Consequently, a careful distinction of Mühlenberg’s and his correspondents’ various and mutual affiliations will be of high importance in this study.

2. APPLICATIONS – ROSENTHAL ET AL., GOULD AND BEARMAN

In 1985, Naomi Rosenthal, Meryl Fintgrud, Michele Ethier, Roberta Karant and David McDonald collaborated on a network analysis of “organizational affiliations of 19th century women reform leaders in New York State as a case study of relations among social movements.”53 Their study demonstrates how central measurements of network analysis, e.g. centrality, clusters and strong/weak ties can be applied to arrive at a thorough and original structural analysis of historical social movements. Its specific strength lies in the comparison of the authors’ findings with standard historiographical narratives, which exemplifies how network approaches can enrich historical investigation. In the following, their study and two other historical network studies by Roger V. Gould and Peter S. Bearman will be summarized and discussed, in order to flesh out the theoretical context provided in the preceding

Zentrum und zu Gruppierungen hatte, die sich in den Randzonen des Netzes bewegten, für die politische Mobilisierung als sehr bedeutsam. Als Dazwischenstehende sorgten sie für die Ver- breitung von Ideen in die Außenbezirke des Netzes und integrierten die sozial marginal gela- gerten Gruppen.” Krempel and Lipp, “Petitions,” 60. 51 Diaz-Bone, Ego-zentrierte, 59; Breiger, “Analysis,” 508; Degenne and Forse, Introducing, 21; Granovetter, “Strength,” 1361. 52 Dupree, “National Patterns,” 23. 53 “We argue here that social movements usually appear within the context of, and depend on the existence of, other social movements.” Rosenthal, “Social Movements,” 1022f. 26 Networks and History

subchapter. This will also help to make clear the enormous theoretic flexibility and applicability of the Network paradigm. In a first step, Rosenthal et al. compiled a list of 202 women residing and working in New York State between 1820 to 1914 for at least ten years, all of whom were active in social reform movements.54 Based on this first list, three further ones were generated that featured (1) a total of 1,015 organizations the women participated in, (2) respective memberships in these organizations, and (3), most importantly, all pairs of any two of these 1,015 organizations having at least one woman activist in common, constituting a dyad. As it turned out, only 10,393 of all 514,605 potential dyads in this network actually existed.55 Somewhat arbitrarily, as they admitted themselves, Rosenthal et al. “labeled ties of one and two as weak, three as moderate, and four or more as strong”56, thereby adding a weighting of ties on which the main part of their interpretation relies.57 The vast majority of these dyads had only one member in common (94.3 %), 4.2 % were linked by two, 1.0 % by three and only 0.7 % had four or more mutual memberships.58 The network they come up with clearly demonstrates the interconnectedness of the various social movements in New York state, membership patterns and the three most important organizations in New York State, the WRA [Woman’s Rights Asso- ciation], the NWSA [National Women Suffrage Association] and the NAWSA [Nati- onal American Women Suffrage Association].59 Furthermore, Rosenthal et al. contended that the “pure number of ties that one organization has with others is not enough to indicate its influence or the degree to which that organization might or

54 Rosenthal, “Social Movements,” 1027. Rosenthal et al. relied on biographical dictionaries of and on the period while imposing several restrictions on the material that finally brought the number down to 202 women. 55 Rosenthal, “Social Movements,” 1030: “The matrix, then, consists of a tally of links between dyads. Looking, for example, at the dyad created by the NAACP [National Association for the Advancement of Colored People] and the Socialist party, we see three links (a tie of three) because three women belonged to both organizations.” 56 Rosenthal, “Social Movements,” 1031. Similarly, Barkey and van Rossem define strong ties as the product of friendship and frequency of contact, which could also lead to socioeconomic advantages, whereas mere acquaintance was defined as a weak tie. Barkey and van Rossem, “Networks of Contention,” 1362. 57 Therefore, this is a “weighted analysis,” as ties in this network do not simply constitute an affiliation between any two organizations but are also qualified in terms their strength or weak- ness. 58 Rosenthal, “Social Movements,” 1032. 59 “Centrality analysis identifies the most central nodes in the network. (...) This technique allows for a focused analysis on the most important individual organizations in the network, the most important links among organizations, the most important groups of organizations, and the overall density or looseness in the network. (...) The centrality scores for the top three organizations are relatively high. (...) These are the Woman’s Rights Conventions (with 253 links to 154 organizations), the NWSA (209 links to 114 organizations), and the NAWSA (329 links to 232 organizations) (...) The centrality of these three groups in the network confirms historical contentions that, at least nationally, suffrage and women’s rights organizations dominated the 100 years of reform activity in which women were involved.” Rosenthal, “Social Movements,” 1030–36. Applications 27 might not be central in the network or be in a position to exert leverage in it.”60 Drawing on the distinction between weak and strong ties, they seek to empirically measure individual organizations’ potential community outreach based on network data. This approach, however, is the direct opposite of an analysis relying on the organizations’ essential attributes, such as membership numbers or financial resources. For instance, whereas the WCTU [Woman’s Christian Temperance Union] and the NWSA were roughly equal as far as number of active members (24 out of 202 in both cases), number of weak ties (WCTU: 142, NWSA 95) and moderate ties (three in both cases) were concerned, the NWSA was found to be connected through 16 strong ties to third parties, which gave it a much higher centrality than the WCTU, despite its larger number of weak ties.61 Next, Rosenthal et al. went on to identify clusters and areas of heightened collaboration among organizations, thereby disclosing the “skeleton, so to speak, of the network.”62 For this purpose, they excluded the top three organizations on their list, as these, in their view, primarily functioned as “umbrella groups” that “served as links between diverse organizations and provided a means for unifying groups.”63 By doing so and by limiting the tie strength to four, they finally came up with an elite group of 36 groups linked to each other through at least four common members and among which five distinct clusters emerged. Furthermore, each cluster’s most central organizations were identified by “peak analysis”, in order to “locate those organizations that functioned as bridges between clusters (because they belonged to more than one cluster).”64 Finally, a periodization of the network into three chrono- logically successive parts was established, each distinguished by individual network shapes and contents.65 Although the study of Rosenthal et al. has been criticized for its lack of explanatory value with regard to the very distinctions it establishes,66 it provides an excellent framework for the study of American women

60 Rosenthal, “Social Movements,” 1032. 61 Rosenthal, “Social Movements,” 1032. 62 Rosenthal, “Social Movements,” 1037. 63 Rosenthal, “Social Movements,” 1036. “Yet the very density of this network masks any distinct groupings that might exist. We suspected this as because of the “bridging function” played by the three most central organizations and the large number of weak ties among groups. We wanted to see if it were possible to demonstrate underlying groupings and hubs of activity if the influence of these highly central organizations was removed and only strong interlocks were considered. Rosenthal, “Social Movements,” 1037. 64 Rosenthal, “Social Movements,” 1037. 65 Rosenthal, “Social Movements,” 1042f. 66 “Rosenthal et al.’s delineation of these three periods of women’s reform activity by means of relations analysis surely ranks as a significant and worthy contribution. But its limitations are also considerable: the study provides little systematic explanations as to precisely why these changes occured from one historical period to the next, settling instead for a succession of static ‘map configurations’ or relational “snapshots” of network patterns. The individual and social actions that led from one structural configuration of reform activity to the next are left unana- lyzed, as are the developments in social structure and cultural and political discourse that underlay and motivated them. At best, Rosenthal et al. treat these various developments in their analysis as exogenous variables.” Emirbayer and Goodwin, “Problems of Agency,” 1426. Al- though Emirbayer and Goodwin’s objection makes sense within their own line of arguments, it 28 Networks and History

reform activism in the 19th century and a reliable model for examining large-scale social developments. In their own words, they “have shown that network analysis can be successfully applied to historical data, that it can both amplify and correct historical propositions about structural links, and that empirical work of this kind leads to a better theoretical understanding of social movements.”67 In his 1996 essay “Patron-Client Ties, State Centralization, and the Whiskey Rebellion,” Roger V. Gould has approached the 1794 Whiskey Rebellion in Wes- tern Pennsylvania from a network perspective. Specifically, his argument focuses on a splinter group of elite landholders whose persistent participation in the infa- mous uprising constituted a puzzling anomaly that could never be resolved within standard historical narratives.68 Initially, a large portion of the local elites had sided with small agrarian protesters against new taxes on liquors and helped to give organizational shape to the movement. Later, most of this landed elite came to be co- opted into the ranks of the new federal tax bureaucracy. For reasons hitherto unex- plained, a tiny fraction of this local elite actually remained with the resistance movement and galvanized public discontent into a mob of some 5,000 armed farmers and liquor-producers, which prompted President Washington to dispatch 15,000 federal troops to quell the rebellion in early August of 1794. Gould acknowledges that patronage ties and the lure of official careers were useful instruments in the hands of state centralizers to undermine local networks and loyalties. He criticizes, however, that historiography has so far treated these ties merely as an “invention” of the state builders, completely ignoring the fact that these structures largely conflated with those of rebellious landholders. In Gould’s scenario, individual landholders’ decisions to continue or discontinue their support for the insurgents can best be explained by looking at the changing configuration of social ties during the months preceding the uprising. Gould contends that rebellious elites “occupied structurally disadvantaged positions (...) in the network of patronage relations that linked many of the region’s prominent officeholders (including

has to be admitted that Rosenthal et al. never claim to explain the changes and structures un- veiled in their study and clearly see their work as a starting point for further research rather than an end in itself: “Network analysis provides a fresh look at activity previously described by historians. Where the two agree, our faith in the validity of the conclusions increases. In those areas where our findings diverge from those of historians, our analysis points out directions for further research, new hypotheses, and new historical syntheses that incorporate the structural configurations created by overlapping affiliations.” Rosenthal, “Social Movements,” 1045. 67 Rosenthal, “Social Movements,” 1050. For another synopsis of this study see Lipp, “Räumli- che Muster,” 61f. 68 Gould first dismisses a number of traditional explanations that include “Scots-Irish traditions of animosity towards tax collectors,” economic consequences of the new taxes on distilled spirits after 1791 or political ambitions of the elites. He argues that hatred of taxes was too general a phenomenon to account for the uprisings in western Pennsylvania in 1794, citing the War of Independence as another instance of conflict over taxation. Additionally, the whiskey tax of 1791 had had little impact on producers as it was passed on to consumers, with whiskey prices rising by 25% in a period when prices for other grain products fell by 20%. Finally, political ambitions as a motive must be confined to the earliest phase following the passage of the law until 1793, according to Gould. Gould, “Whiskey Rebellion,” 408–12. Applications 29 the newly appointed tax inspector and all the other federal officials in the area) to one another.” These “structurally disadvantaged positions” could take on two forms. In the most extreme form, revolutionary elites found themselves completely cut off from the new patron-client system, while others found their position weakened through the emergence of an alternative network for their clients, which deprived them of their local monopoly on public support. Bringing together data from tax records, official records of postcolonial Pennsylvania and surety bonds, Gould seeks to reconstruct this newly emerging client-patron network, for which surety bonds proved to be especially helpful. After 1789, it had become compulsory for potential office-holders in Pennsylvania to present to the Supreme Executive Coun- cil one or several persons willing to sign a surety bond on the applicant’s behalf. As these bonds represented an enormous financial risk, it was common practice within the established elite to post sureties on behalf of each other, whereas political outsiders had to rely on other outsiders for support. As Gould further elaborates, it was predominantly this practice that created an “unconnected elite,” which faced the highest risk of being completely cut off from political influence once federal efforts to collect the whiskey tax were underway.69 With regard to previous class-based interpretations of the Whiskey rebellion, Gould argues that “standard dimensions of social categorization (class, gender, race, nationality) are best seen as contingent outcomes rather than autonomous or primordial bases of difference.”70 Similarly, he contended in his 1993 essay “Trade cohesion, class unity, and urban insurrection: artisanal activism in the Paris Com- mune” that the mobilization of insurgents during the Parisian commune of 1871 was not rooted in artisanal organizations, but rather in Parisian neighborhoods. Ba- sing his argument on trial dossiers and official records on the insurgents, his approach on the subject provides a fresh explanation of cross-trade participation during the insurgency. Also, it replaces previous interpretations, which had taken this social phenomenon for a sign of an early class-consciousness cutting across older systems of allegiance and identity formation.71 In a related manner, Peter S. Bearman starts his 1993 study on the transforma- tions of gentry social relations in Norfolk, England between 1540 and 1640, with the acknowledgment that “[t]he solution to understanding elite social action cannot be further subclassification from received categories. Categorical models alone rarely partition people in a way that conforms with observed action, because individual activity in the world is organized through and motivated not by categorical affiliations but by the structure of tangible social relations in which persons are

69 Gould, “Whiskey Rebellion,” 412–17. For another discussion see Orser, Race and Practice, 122f. 70 Gould, “Whiskey Rebellion,” 403. 71 Gould, “Paris Commune,” passim. “Gould concludes thim this that “cross-neighborhood solidarity” was a significant feature of the Parisian insurrection. (...) [He] thus shows that structural analysis needs to take into account not only individual-level variables such as those that are employed in the McAdam studies, but also the complex influences of multiplex or overlapping networks of social ties.” Emirbayer and Goodwin, “Problems of Agency,” 1421. For other discussions see Lipp, “Räumliche Muster,” 59; and Barkey and van Rossem, “Networks of Contention,” 1348. 30 Networks and History

embedded. To make sense of action, we have to make sense of the complex clusters of social relations surrounding individuals, which often drive them to act at cross- purposes.”72 Based on the observation that participation in the English civil war often cut across established class boundaries,73 the study relies on blockmodelling74 to identify and describe historical actor groups according to their structural position and embeddedness. Blockmodelling techniques operate on the idea that actors with a similar configuration of ties are structurally equivalent and therefore entertain similar sets of ambitions and interests.75 Bearman establishes his “equivalency classes” through an examination of shifting landholding patterns and the changing importance and distribution of kinship ties76 and concludes that “[r]evisionist historiography has failed to offer a consistent theory of action, because the fact of connectivity alone does not provide an operational mechanism for aggregating individuals into equivalency classes. The provision of such an operational mechanism is the strength of the broader accounts of the civil war developed by sociologists. These accounts which recognize that a class of individuals acting in pursuit of similar interests must, at the same time, share something in common. What they share is a position, however, rather than an attribute.”77 This commonality, Bearman argues, soon found expression in abstract rhetorics, which both propelled social change and allowed for the formation of new interest–based alliances that transcended the local scene and eventually set the stage for the civil war.78 With the notable exception of religious beliefs, Bearman denies that either centralizing state powers or any other social category can exclusively be claimed as the driving force of elite action and mobilization. Instead, the combined effects of shifting social patterns in local contexts explain best both factionalization and individual motives in the civil war.79 Blockmodelling techniques have also been successfully employed in other historical network studies,80 of which John Padgett and Christopher K. Ansell’s account of “Robust action and the rise of the Medici, 1400–1434” stands out as pro-

72 Bearman, Relations, 10. 73 Bearman, Relations, 9. 74 According to Degenne and Forse, blockmodelling techniques were first developed by ethnolo- gists to describe kinship patterns. “Equivalence is a cognitive operation that the sociologist performs to describe relations and social roles; the issue is how to actually construct them from the data. Most roles take on meaning in the context of daily living, e.g. father/son, employer/ employee and shopkeeper/client. This is because of the institutionalized definition these roles enjoy. But other roles are fuzzier. ‘Boss’ and ‘expert’ mean very different things depending on the context.” Degenne and Forse, Introducing, 83, 93. See also Faust and Wasserman, Analysis, 394–97. 75 Degenne and Forse, Introducing, 88. 76 “The structure of local landholding led to the crystallization, into parties and factions, of networks of elite alliance and opposition which would in time come to be defined in terms of abstract rhetoric.” Bearman, Relations, 46. 77 Bearman, Relations, 12. 78 Bearman, Relations, 175–77. 79 For criticism of Bearman’s approach see below on pages 24f., or Orser, Race and Practice, 120f. and Emirbayer and Goodwin, “Problems of Agency,” 1439. 80 For more historical network studies and theoretical discussion not explained in full detail here, please refer to Burton and White, “Regional Comparisons;” McAdam, “Freedom Summer;” A Critique 31 bably the most often quoted historical network study so far. Padgett and Ansell show that by employing a “focused marriage strategy” and by taking advantage of “structural holes,” the Medicis showed a great deal of network awareness and were thus able to turn the political developments in their favor.81 Similarly, Karen Barkey and Ronan von Rossem have examined rural “Networks of Contention” in the se- venteenth century Ottoman empire. In their article, they “argue that actors that are in similar positions, facing similar environments, will display similar behavior. Vil- lages that are positioned similarly in the overall intervillage network should show similar patterns of behavior.”82

3. A CRITIQUE

Network studies have frequently been criticized for their predominant focus on structural patterns to explain human agency. In fact, Berkowitz and Wellman admit that “the commonly heard objections to structuralism apply to network theory as well.”83 In their joint 1994 essay “Network Analysis, Culture, and the Problem of Agency,” Mustafa Emirbayer and Jeff Goodwin have subjected historical network studies to a “theoretically informed assessment and critique”84 and it is primarily their work that I am condensing here to briefly summarize the use of network concepts in the present study on Mühlenberg’s transatlantic correspondences. Within the corpus of network literature, Emirbayer and Goodwin have identified three types of approaches that have conceptualized “the relationships among culture, agency, and social structure (...) in varying degrees of theoretical sophistication.” These, in turn, have “led to varying degrees of difficulty in elaborating satisfactory explanations of historical processes.”85 Generally speaking, the three approaches – labeled structuralist determinism, structuralist instrumentalism and structuralist constructionism – can be distinguished through their respective ways of incorpora- ting and integrating context data into their network perspective. Thus, Emirbayer and Goodwin’s critical essay is helpful to avoid some of the pitfalls associated with research that seeks to integrate structural and cultural contexts into a stringent and source-informed line of historical arguments.

Häberlein, Brüder;” Krempel and Lipp, “Petitions;” and Wellman and Wetherell, “Historical communities.” 81 Emirbayer and Goodwin, “Problems of Agency,” 1433. For additional, more extensive discussions of “Robust Action” see; Orser, Race and Practice, 120–21; and Lipp, “Räumliche Mus- ter,” 57f. 82 Barkey and van Rossem, “Networks of Contention,” 1350. See also Orser, Race and Practice, 120–21. 83 Berkowitz and Wellman, “Introduction,” 4. Rainer Diaz-Bone briefly discusses the structuralist basis of network studies with regard to the development from Ferdinand de Saussure (1857– 1913) to Claude Lévi-Strauss (1908–2009) and Barry Wellman’s (b. 194) criticism of formal- ism. Diaz-Bone, Ego-zentrierte, 22. 84 Emirbayer and Goodwin, “Problems of Agency,” 1412. 85 Emirbayer and Goodwin, “Problems of Agency,” 1428. 32 Networks and History

Emirbayer and Goodwin categorize those studies under structuralist determinism, which “[neglect] altogether the potential causal role of actors’ beliefs, values, and normative commitments – or, more generally, of the significance of cultural and political discourses in history.”86 As an example of this, they cite Rosenthal’s study on nineteenth-century women reform, criticizing its almost complete lack of “systematic explanations as to precisely why these changes occurred from one historical period to the next, settling instead for a succession of static ‘map configurations’ or relational ‘snapshots’ of network patterns. The individual and social actions that led from one structural configuration of reform activity to the next are left unanalysed, as are the developments in social structure and cultural and political discourse that underlay and motivated them. At best, Rosenthal et al. treat these various developments in their analysis as exogenous variables.”87 With regard to Mühlenberg’s network, this calls attention both to contemporary standards of scientific communication and exchange in the so-called Republic of Letters and to internal developments and discourses in the science of botany at the turn of the 18th and 19th century. Studies filed under Structuralist Instrumentalism, however, take personal motivations of historical actors more seriously than structural determinism, although they ultimately “conceptualize their activity in narrowly utility-maximizing and instrumental forms.”88 In a discussion of Robert Gould’s study of the Paris Com- mune of 1871, for instance, Emirbayer and Goodwin claim that “[Gould] never provides a plausible causal account as to why Parisians would have risked their lives for the Commune in the first place. (...) Gould’s assumptions about the purely instrumental foundations of political mobilization – whether class- or status-based – in the present stage of his research prevent him from analysing such cultural and normative influences in a fully satisfactory manner.”89 “In a more complete historical explanation,” they argue, “Gould would have to direct far more attention than he does to the specifically cultural bases of (cross-) neighborhood solidarity and their influence upon individuals’ projects of action.”90 This directs our focus to the ques- tion of potential motivations on the part of Mühlenberg and his correspondents in continuing their mutual botanical exchanges over a long period of time, which often

86 Emirbayer and Goodwin, “Problems of Agency,” 1425. 87 Emirbayer and Goodwin, “Problems of Agency,” 1426. “[Structural Determinism] (...) ‘ruth- lessly abstracts’ the formal or ‘objective’ dimensions of social relations from their cultural and intersubjective contexts so as to be able to represent and analyze such relations with sophisticated technical tools; in the process, however, it drains such relations of their active, subjective dimension and their cultural contents and meanings.” Emirbayer and Goodwin, “Problems of Agency,” 1427. 88 Emirbayer and Goodwin, “Problems of Agency,” 1425. 89 Emirbayer and Goodwin, “Problems of Agency,” 1429. 90 Emirbayer and Goodwin, “Problems of Agency,” 1429. In the same vein, they criticize Peter Bearman’s study for “its tacit instrumentalism – that is, its tendency to devote almost all of its analytical attention to uncovering the ‘structural preconditions’ for this elective affinity, rather than to also exploring the independent causal significance of these discursive frameworks themselves. (...) Indeed, Bearman seems to attribute little more than purely material interests (in money, status, and power) to the historical actors at the center of his account.” Emirbayer and Goodwin, “Problems of Agency,” 1439. A Critique 33 entailed great financial risks. I have tried to conceptualize these inner motivations in Exchange Charts, which depict the chronological flow of botanicals and scientific information within individual exchanges.91 Finally, in the approach which they term Structuralist Constructionism, the two authors see “the most successful of all in conceptualizing human agency and the potentially transformative impact of cultural idioms and normative commitments on social action.”92 As an example, they cite Doug McAdam’s 1986 essay “Rec- ruitment to High-Risk Activism: The Case of Freedom Summer,” wherein the author sought to explain patterns of participation in the Civil Rights movement. Based on the observation that most of those, who finally participated in the Freedom Sum- mer Project,93 had already taken part in similar events or organizations before, McAdam states that “structural availability” eclipses “attitudinal affinity” as an explanation for participation.94 Though attitudinal affinity is a necessary prerequisite, he makes the much more important second point that participants slowly began to “grow more comfortable with the role of activists” during their earlier involvement in previous organizations. This implies a theory of identity conversion, “that takes seriously the formation of motivations and identities without sacrificing at all the moment of ‘structural location.’”95 Emirbayer and Goodwin’s essay raises important points for a fruitful application of network theory to Mühlenberg’s correspondences. Most importantly, their reassessment of Rosenthal’s, Gould’s and Bearman’s different approaches shows that historical network studies need to integrate contemporary social and political circumstances, types of social relations and discourses into their interpretation of social structure.

91 See Appendix D, page 535f. 92 Nevertheless, the two remain skeptical even about this model: “However, even this perspective falls short in understanding the full complexities of the theoretical interconnections among culture, agency, and social structure. It too pays insufficient attention to the structuring influences of cultural and political discourses upon historical actors.” Emirbayer and Goodwin, “Problems of Agency,” 1424. 93 The Freedom Summer Project was launched in June 1964 by the activits of the Council of Federated Organizations (COFO) and the Student Nonviolent Coordinating Committee (SNCC) in order to raise the number of registered voters among African Americans in the state of Mississippi. The ten-week event was met with heavy resistance by local groups of the Ku- Klux-Clan and culminated in the deaths of three SNCC and COFO activists in an ambush by clansmen on June 21st 1964. See Sally Belfrage, Freedom Summer, passim. 94 Emirbayer and Goodwin, “Problems of Agency,” 1431. McAdam admits, however, that this holds mainly true for high-risk/-cost activism. McAdam, “Recruitment,” 73. Emirbayer and Goodwin add that both Gould’s Paris Commune and Bear’s English Civil War also belong to this group of high-risk/- cost activism. Emirbayer and Goodwin, “Problems of Agency,” 1431. 95 “Unlike many other network analysts, McAdam recognizes that actors can undergo far-reach- ing processes of identity formation in the course of their involvements in extraordinary affairs. Such an insight is especially important to bear in mind when analyzing their participation in “high-risk/cost” activities such as Freedom Summer – or, for that matter, in any major social or political movement.” Emirbayer and Goodwin, “Problems of Agency,” 1432. 34 Networks and History

4. AIMS AND METHODOLOGY – THE PLURALITY OF MÜHLENBERG’S NETWORK

The so-called Republic of Letters, the transconfessional and intercultural network of scientific collaboration in the early modern period, constituted the primary sphere of communication, in which Mühlenberg’s correspondences between 1771 and 1815 must be situated and analyzed. As Mühlenberg became active as an independent botanical correspondent after the war for American independence in 1784, the present network analysis will be confined to the years between 1784 to 1815. The years from 1771, from when the first known Mühlenberg letter dates, and late 1784, however, will be treated in the “Prelude”-chapters IV.1 through IV.3.96 Following Emirbayer and Goodwin’s suggestions and critique, chapters V.1 through V.6 aim to make the Republic’s communication standards, implicit norms and structure an integral part of its network-analytic approach from 1784 to 1815.97 In order to achieve a sound evaluation of Mühlenberg’s role as networker, scientific organizer and founding figure of botanical sciences in America during these 31 years, the focus will be put on the ebbs and flows of individual correspondences, the identification of exchange patterns, the flow of resources and the rise of collaborative clusters of scientists between Mühlenberg as the focal actor and his alteri, respectively among the alteri themselves. Thus, it is sought to unearth the “subnets of the system” (Daniel Roche), which practically equal “centers of heightened cohesion and collaboration.”98 On the one hand, Roche’s observation suggests that Mühlenberg’s network must be seen as one of these subnets of contemporary scientific exchange itself. On the other hand, it also suggests that even the network of one individual consisted of several subnets – both in space and time. As will be shown, the distinction between the European and the American wing of Mühlenberg’s net-

96 Only 50 out of a total of 990 known Mühlenberg letters were written from 1771 to 1784, 36 of which alone were exchanged with Mühlenberg’s father Heinrich Melchior Mühlenberg. 97 “Examples of cultural or discursive structures that need to be analyzed internally (as well as in their interplay with network structures) include the civic rights discourse of Freedom Summer, and the socialism and republican patriotism of the Paris Commune. (...) We propose that these cultural formations are significant because they both constrain and enable historical actors, in much the same way as do network structures themselves. (...) Cultural structures constrain actors, to begin with, by blocking our certain possibilities for action, as, for example, by render- ing it inconceivable for the oligarchs of 15th century Florence to have pursued marriage ties with nonpatrician new men, even when it might have been materially advantageous for them to do so. Emirbayer and Goodwin, “Problems of Agency,” 1438. See also Diaz-Bone, Ego-zentrierte, 35; Berkowitz and Wellman, “Introduction,” 6; Lipp, Struktur, 56; Barkey, Contentions, 1376f. For the history, norms and communication structures of the Republic of Letters, see below on page 98f. 98 Kempe, “Anglo-Swiss,” 75. For subnets in the case of Christoph Jacob Trew (1695–1769), see also Schnalke, “Vernetzen,” 172, and Wellman, Structural Analysis, 45: “Given asymmetric ties and bounded network clusters, resources do not flow evenly or randomly in a structure. The density of clusters, the tightness of boundaries between them, and the patterns of ties within and between clusters all structure resource flows. Because of their structural locations, members of a social system differ greatly in their access to these resources. Indeed, unequal access to scarce resources may lead to greater asymmetry in ties.” Aims and Methodology 35 work is crucial to any deeper understanding of its dynamics. Apart from this transatlantic separation, the European and the American webs were also composed of local centers of collaboration or botanical interest – predominantly cities like Erlangen, London, Göttingen, Paris, Philadelphia, Boston and New York, with their scientific infrastructure. In a geographical sense, these distinctions will inform the present study. More important, however, is the temporal aspect. Speaking of “Mühlenberg’s network” ignores the fact that there were actually many “Mühlenberg networks” from 1784 to 1815, each of which was defined by a unique social composition, Mühlenberg’s evolving botanical interests, a growing and shrinking range of possible botanical, professional, kinship or economic contacts, personal developments, and other, secondary circumstances. Therefore, the present study engages in separate discussions of six consecutive network configurations, which were primarily identified by the author through their individual social composition. Chapters 3.1 through 3.6, which represent the main body of the text, concur with six individual network configurations between 1784 and 1815. Flow Chart A99 gives a graphical impresson of the fact that Mühlenberg corresponded with different sets of synchro- nously active contacts 1784 to 1815.100 These six individual phases are:

Phase 1 1784 – about 1790 Phase 2 1790 – about 1797 Phase 3 1797 – about 1802 Phase 4 1802 – about 1805 Phase 5 1805 – about 1811 Phase 6 1811 – about 1815

These phases lasted for an average of five to six years and were defined by a relative stability of their social composition, while they were separated by six brief periods of change, characterized by the entry of entirely new correspondents, the deaths of old correspondents, or temporary communication-breakdowns with singular contacts. Some of these contacts were discontinued for good, while others were only resumed at a later point. These periods of gradual change will be routinely described and analyzed in the introductory notes V.1 to V.6 in the form of brief narrative accounts of internal developments and external historical circumstances. It is important to point out, however, that this periodization has been developed independently of external historical periodizations, although contemporary developments and events – such as war – induced communication problems during the French Revolutionary Wars – often influenced the course of Mühlenberg’s network decisively.

99 Appendix A, Flow Chart A, 483. 100 See Appendix A, 483–485. From 1771 to 1784, Mühlenberg merely appears as a sporadic correspondent of relatives, professional colleagues and friends. As this study’s focus lies on his principal role as scientific organizer and botanical correspondent, Mühlenberg’s youth, education, family relations and first 13 years of tentative correspondences will be treated in Chapter IV“A Prelude – Mühlenberg’s Network from 1771 to 1784.” 36 Networks and History

Some more methodological remarks are necessary to render my analytical approach clearer. These subdivisions are merely a heuristical help to understand the course and social dynamics of Mühlenberg’s network better. Some of the events, developments and facts I will cite in order to support my approach of “consecutive phases” were certainly felt by Mühlenberg himself, but it is very unlikely that he would have spoken of individual “phases” in his network himself. Therefore, their establishment is a historical projection from a network-driven perspective that allows to organize a massive amount of data, accumulated over 31 years letter-writing, into meaningful subsets of letters, correspondents, and illustrative charts and graphs. It must further be admitted, however, that truly clear-cut separations between any two of these network phases are hardly ever feasible, primarily because individual correspondences often tend to continue over these artificial thresholds.101 These “blurry edges” entail the rather sticky problem of deciding where to draw the line and which letters will fall into what chapter. This step, however, is of essential importance in order to provide a sound basis for a number of statistical breakdowns in individual chapters. Specifically Appendix B contains tables, for which definite lines of separation are essential to avoid data confusion and logical inconsistencies.102 In order to overcome this problem, I have decided to take the first letter of a completely new correspondent in a new phase as the threshold in any of these six transitional phases, which generally separate periods of stability and are roughly dated 1784, 1790, 1797, 1802, 1805 and 1811. These were the years when the Müh- lenberg network was significantly re-shaped, and it is in fact one preliminary result of the present study that early modern communication networks seem to have un- dergone phases of renewal after certain periods of time. In the first footnote of chapters V.1 through V.6, these first letters of one new correspondent will be cited as “cornerstone-letters” – their dates will practically separate the present from the preceding phase.103 Consequently, chapters V.1 through V.6 all work on the basis of mini-corpora of letters, which were isolated from the main corpus of 998 letters

101 In a similar vein, Alain Degenne and Michel Forse have observed that “[a]s a rule, the best of personal network studies only capture a very rough structural picture of volume, frequency, multiplexity or density, and the statistical gain is a structural loss. For better results, structural analysis needs to operate on total networks. (...) The two basic judgements concern the best population breakdown and which relations to study. As with personal networks, partitioning will always be somewhat arbitrary. No network has ‘natural’ frontiers; researches impose them. (...). Because partitions are always somewhat arbitrary, we must ever bear in mind that there is always something tentative about network analysis data.” Degenne and Forse, Introducing, 22. 102 See tables e, g, i, k, m, n, Appendix B, 488f. 103 The following seven letters have been picked as “cornerstone-letters” and represent the first instance of correspondence in a new phase of Mühlenberg’s network. A note by Henry to his father, dated December 04, 1771, AFSt M.4 C18, stands both as the the earliest surviving letter from his hand and the first letter treated in chapter 2, “A Prelude.” The other six are: From Fabricius, 01/20/1784, APS Film 1097; To Humphrey Marshall, 01/18/1790, HSP Soc. Coll.; From Nebe, 06/06/1797, [reconstructed]; From Brickell, 01/21/1802, [reconstructed]; To Nebe, 10/25/1805, AFSt M.4 D6; To Baldwin, 01/07/1811, Darlington, Baldwiniae, 15. For reconstructed letters mentioned here, see Appendix C, 500, 520, . Aims and Methodology 37 located as of August 2011.104 In contrast to this method, Hächler, Stuber and Lien- hart have used Albrecht von Haller’s (1708-1777) biography, itinerary and changing abodes as the primary structuring element in their edition of Haller’s correspondences.105 Mühlenberg’s permanent immobility as Lancaster’s Lutheran pastor from 1780 to 1815 rules out this possibility for the present study. Alexander Pyrges has introduced a similar method of network periodization in his analysis of the “Ebenezer network” from 1732 to 1828. The main difference, however, is Pyrges’ primarily quantitative approach.106 Whereas Appendix A and Appendix B primarily feature graphic and statistical data on the macro-development of Mühlenberg’s network, Appendix C and Appen- dix D contain micro-level information on individual correspondences. Appendix C features lists of the entire corpus of Mühlenberg’s surviving and reconstructed letters in alphabetical order of contacts. The development, temporary discontinuations and final endings of individual correspondences can here be seen at a glance.107 The charts in Appendix D illustrate the chronological developments of specific botanical exchanges, through which the economic nature of individual contacts will be analyzed.108 Whereas the left column of the charts provides archival information on the respective letters from which information was retrieved, the right column features the actual exchange chart of botanicals, pieces of information or other items mentioned in these letters.109 This helps to carry the idea of “transfer of knowledge” a step further, which is a traditionally defining feature of the republic of letters in

104 This number is composed of 693 surviving letters and an additional 297 letters, which were reconstructed from information in the core corpus of 693 letters. Eight surviving letters were undated. For a graphic representation of surviving and reconstructed letters, see table a, Appen- dix B, 486. See also lists of correspondences, Appendix C, 494–534. 105 Hächler, Stuber and Lienhart, Hallers Netz, 65f. 106 Pyrges primarily focuses on the intensity of contacts in individual phases, represented by active nodes in the network. Also, Pyrges’ essay does not have a focal actor, including data from 300 individuals at 60 different places. See Pyrges, “Ebenezer Network,” 58; 66. For more on network visualization, see Faust and Wasserman, Analysis, 11; 70; 76; Breiger “Analysis,” 508; Hächler et al., “Exploration,” 348–50; Petersen, New World Botany, 210. See also Lipp, Struk- tur, 61f.: “Wie dieses Beispiel zeigt, liegt die Stärke der Netzwerkanalyse in der Visualisierung, die intuitiv anspricht und Verflechtungen sichtbar macht (…). Oder wie Barry Wellman es formuliert: ‘A basic strength of the whole network approach is, that it permits simultaneous views of the social system as a whole and of the parts that make up the system.’ Komprimierung ohne Informationsverlust ist die Stärke des Konzepts.” See also Faust and Wasserman: “The visual representation of data that a graph or sociogram offers often allows researches to uncover patterns that migh otherwise go undetected.” Faust and Wasserman, Analysis, 94. 107 In several cases, apparent “pauses” are due to a scarcity or lack of surviving letters, archival collections or other historical evidence rather than an actual temporary or final discontinuation. For a precise delineation of individual correspondences, confer to the respective passages in the main text. See also Appendix C, 494f. 108 See Appendix D, 535f. 109 To note, no distinction has been made between exchanges that actually took place and merely projected transfers, or even only desired items. The goal is to develop a “supply and demand”-perspective on the mutual expectations in an individual barter and thus to approach an answer as to what exactly made two exchange partners in a specific situation interesting to each other. 38 Networks and History

historiography. Specifically in the given botanical context, this makes sense, as herbaria – collections of identified and described specimens – were and still continue to be seen as the places where original botanical knowledge was stored. In the context of this study, “knowledge” and its transfer will therefore predominantly be treated as a set of material and immaterial commodities pertaining to Mühlenberg’s specific field of interest, botany, rather than in a philosophical, psychological or psychosocial sense.110 In the charts in Appendix D, there are eleven different categories of transferred goods. Most commonly, Mühlenberg provided [plants], [seeds], [minerals] or [specimens] to his European correspondents, who generally returned their plant identifications [plant ident.], [books] or their [own works] to Lancaster. In nearly all of Mühlenberg’s American correspondence, we find this pattern reversed.111 The item [own works] is especially interesting, as these tended to be botanical publications containing plant information on specimens Mühlenberg had originally submitted to the respective author. Other than that, [info] refers to general information on other botanists, publication plans, political news and, in rare cases, even gossip on the republic of letters. Whenever [post] is listed in an exchange chart, a correspondent forwarded or received letters for his counterpart, which, in times of unsafe and mostly improvised postal ways, was a valuable service and a commodity of exchange in itself. The tag [opportunity] refers to occasional opportunities to sell books, medicals or seeds, usually granted through access to one’s own correspondence network. Finally, [equipment] represents extremely rare occasions when valuable scientific instruments were part of an exchange. The following passage from a letter by Mühlenberg to the English botanist James Edward Smith (1759–1828) illustrates the procedure. According to my Promise, Mühlenberg began in Decem- ber 1794, I beg leave to send you a third Fascicle of dried Plants found near Lan- caster this Year, the second went last Spring with the Pigon consistin of Lichens, (...) [The] third Paquet is numbered from no. 351–474. I begin with Grasses, my favorite and add such Plants as are amongst my Adversaria. Further down in the letter, Mühlenberg thanked Smith for his plant identifications, which had been sent in response to an earlier package of plants.112 In the respective exchange chart, this passage would add a [specimen] tag to the “Mühlenberg to Smith” column and a [plant ident.] tag to the “Smith to Mühlenberg” column. This practical approach to the practice of scientific exchange is rooted in a dy- namic understanding of “science” itself as a cultural practice of exchange and collaboration rather than as a solitary activity of isolated individuals.113 This also cor-

110 For an introduction to different concepts of knowledge and the “Plurality of Knowledges,” see Burke, Knowledge, 13f; 81f. 111 Compare, for instance, Mühlenberg’s exchange patterns with Baldwin and with Schreber. See Appendix D, 535f., 545f. 112 Your instructive Observations on the Plants no. 1–15, 18–28, 46–76 (exc. 61. 63) gave me so much Satisfaction and Pleasure that I wait with real Impatience for a Continuation of your Remarks and Nomenclature of the rest. To Smith, 12/10/1794, APS Film 6 Linn. Soc. 113 See, for instance, Rigby, “Seaborne,” 99: “In David Philip Miller’s essay in Visions of Empire, Miller questions the familiar ‘great man’ interpretation of history by reading Banks as only the visible tip of botanical science’s iceberg, the other nine tenths of which is formed by the nurse- Aims and Methodology 39 responds to a recent change of perspective in the historiography of sciences, which tends to approach its object in terms of a multi-layered, collaborative process, and integrates contemporary social, political and scientific discourses to explain the dynamics, shapes and floating outcomes of knowledge production.114 Peter Burke has been the champion of this new take on the socio-historical conditions of knowledge, which ties in well with some of the main insights of network theory.115 As will be shown, late 18th century botany mainly focused on Carl Linnaeus’ (1707– 1778) 24th “cryptogamic” class in the sexual system, which brought more and more cryptogamists into Mühlenberg’s network around the turn of the century. In turn, the increasingly difficult conditions for trade and mail transport on account of continued warfare in Europe and on the Atlantic after 1805 were one of the main factors in the gradual Americanization of Mühlenberg’s network from 1800 to 1815. Finally, Appendix E holds six network drawings generated with UciNet© network visualization software that reflect the individual social compositions in phases one through six. These will be discussed, using proper network terminology, in the concluding subchapters of chapters V.1 through V.6.116 At this point, however, it must be stated clearly that the approach outlined above, which mainly crystallizes in the separation of six network phases, uses network theory as a heuristical guide- line rather than as a method of “social computation,” as it is mainly understood in the modern social sciences. The reasons for this self-limitation are based in the very limitations of the source materials used in this study themselves. The German his- torian Verena Kücking has recently pointed out that the typical limits of written sources and the data gaps one encounters in almost any type of historical tradition necessarily render any network analysis in a strictly sociological sense practically inapplicable in historical contexts.117 In the present study, four major problems

rymen, surgeons, gardeners, plant collectors, missionaries, ships’ captains, and traders who play equally important parts in plant transportation.” Ultee writes: “An isolated scholar was all the more dependent on letters to remain in contact with the Republic, for journals and books gave him only partial information. Through letters he might touch the Republic in its very es- sence.” Ultee, “Republic,” 104. See also Hächler, Stuber and Lienhart, Hallers Netz, 3; Kempe, “Anglo-Swiss,” 73–75; Goldgar, Impolite Learning, 4; Hagner, “Ansichten,” 13; Herren, “Er- weiterung,” 197; Schnalke, “Sammeln,” 171; Kempe, “Anglo-Swiss,” 73. 114 See Burke, Knowledge, 11f.; Hagner, “Ansichten,” 1423f.; Hächler, Stuber and Lienhart, Hallers Netz, 13; Emirbayer and Goodwin, “Problems of Agency,” 1438. See specifically Peter Bearman’s approach in Relations into Rhetorics which focuses on discursive liguistic developments that have ushered in a reification of social discourses in popular mobilization and insti- tutionalization. Bearman, Relations, 12. 115 See first chapter in Burke, Knowledge; and Hächler, Stuber and Lienhart, Hallers Netz, 13. See also Emirbauer and Goodwin, “Problems of Agency,”1443: “The fundamental point that we wish to make (...) is that network analysis errs seriously in ignoring the conceptual insights shared by all of these various theories, in particular the notion that agency and structure inter- penetrate with one another in all individual units (as well as complexes) of empirical action, and that all historical processes are structured at least in part by cultural and political discourses, as well as by networks of social interaction.” 116 See below on pages 122f., 158f., 233f., 266f., 332f. and 403f. 117 “In den letzten Jahren hat die Analyse und Darstellung von Netzwerken ein zunehmendes In- teresse auch unter Historikern und Historikerinnen erfahren. Während zeitgenössische empiri- 40 Networks and History

were met in the reconstruction of Mühlenberg’s network which must be carefully taken into account in the construction and discussion of the network drawings. First, through a close reading of the surviving 693 dated, eight undated letters, Mühlenberg’s diaries and his notebooks, 297 letters could be reconstructed which must be presumed lost today.118 This equals a data loss rate of 29,76 %, which must actually be assumed even higher, considering the potential information in these 297 letters themselves.119 Also, the preservation rate is highly unequal and ranges from 100 % (see Mühlenberg’s correspondence with William Baldwin) down to 0 % preservation (see Frederick Kampmann).120 Another problem showed in the fact that an unquantifiable part of contemporary scientific communication was conducted orally in Roche’s “local clusters of collaboration,” which corresponds to the well-known fact that the republic of letters was an essentially urban phenomenon. In the exemplary case of Johann Christian Daniel Edler von Schreber (1739–1810) and Johann David Schöpf (1752–1800), who both lived and worked in the duchy of Ansbach-Bayreuth, but also in many other cases, a significant amount of oral communication must be assumed which went completely unrecorded.121 Thirdly, the language of the letters is frequently too vague and opaque to derive truly reliable network information from them. A simple and very common phrase like Mr. Le Conte says this is not Pinus scrotina Michaux122 in a letter from Zac- cheus Collins (1764–1831) to Mühlenberg in 1812 can be interpreted in multiple ways. Although it suggests that John Eaton LeConte (1784–1860) and Collins met, the verb “say” was also often used in Mühlenberg’s correspondences to make a quote from someone else’s scientific publications. In any event, in order to avoid confusion, data for the network drawings was only gathered from passages with clear references to visits or other contacts. Your friend Mr. Bartram, whom I visited last week, Collins wrote to Mühlenberg in the same letter, reciprocated in warm

sche Wissenschaften, wie die Soziologie und Ethnologie, Netzwerke nahezu ausschließlich anhand von eigens erhobenen Daten analysieren, steht die historische Netzwerkforschung hier vor einem fachspezifischen Problem: Die zur Verfügung stehenden Daten sind in der Regel sehr lückenhaft und bei den wenigsten Themen ist es möglich, Zeitzeugen zu befragen oder auf Befragungsdaten zurückzugreifen. Darüber hinaus beschäftigen sich Historiker und Historiker- innen in der Regel mit Zeitspannen, so dass auch die Komponente chronologischer Dynamik von Netzwerken ein zentrales Anliegen ist.” Kücking, “Historische Netzwerkforschung,” (see online references). 118 The total number of Mühlenberg’s actual and reconstructed letters is 998. Of these, 693 have been preserved, 297 were reconstructed and eight were undated. See also Appendix C, 494f. 119 297 letters out of 998 = 29,76 %. For reconstructed letters, see especially the lists of correspondences in Appendix C, 494-534. 120 See Appendix E, 495, 513. 121 See also Barkey, Contentions, 1357: “Systematic data about social relations in historical societies is scarce. In particular, little or no systematic records exist that allows us to reconstruct the intravillage as well as the intervillage social structures. Birth and marriage records, even when available, only shed light on one small aspect of the social structure and the networks within and between villages. Court records, on the other hand, can provide information on a broader range of activities and relations, conflictual as well as cooperative ones.” 122 From Collins, 09/24/1812, HSP Coll. 443. Aims and Methodology 41 terms your expressions of esteem and remembrance. 123 This is sound evidence for a personal contact between Collins and Bartram. The problem of vague language, however, also extends to entries in older biographical dictionaries and articles that often feature lists of contacts of Mühlenberg, or of one of his correspondents, or brief allusions to botanical exchanges. In the cases where these could not be proven wrong,124 they need to be considered network information, too. In a similar vein, I do not consider the forms of address and the rhetoric of the letters a particularily reliable source of information on the state and development of individual correspondences. This is mostly due to the changes in literary style and taste in contemporary letter-writing, which became gradually detached from former, highly stylized forms of epistolary exchange and began to include less formal ways of addressing a potential correspondent.125 Reinhard Nickisch’s study on style

123 From Collins, 09/24/1812, HSP Coll. 443. See also Hächler, Stuber and Lienhart, Hallers Netz, 93. 124 A certain “F. Moore” was proposed by C. Early Smith Jr. as a Mühlenberg correspondent in 1954. This claim has already been refuted by Mears. See Smith, “Pioneer,” 443; Mears, “Her- barium,” 169. According to Youman, John Linnaeus Edward W. Shecut (1770–1836), was also a Mühlenberg correspondent. In fact, Shecut is only mentioned in a couple of letters in 1812. In January of that year, Mühlenberg wrote to Stephen Elliott that [b]y a Letter of D[octor] Bald- win I see you have some valuable Botanists in Charles Town Noyzette, Shecut Author of a Flora Caroliniana! not finished and Maywood. Is there no Possibility of getting a Copy of this Flora (…)? To Elliott, 01/05/1812, HUH Elliott Papers. A month later, Elliott answered from Savan- nah: Of our Botanists in Charleston I am afraid I must give an unfavorable account. D: Shecut is no Botanist and undertook to compile a work which he termed Flora Caroliniensis the first number contained an Introduction to Botany, in which he borrowed so much from D: Barton that the D[octor]? not very civilly threatens him with a prosecution. From Elliott, 02/14/1812, HSP Coll. 443. This terminated Mühlenberg’s interest in Shecut, no direct contact was ever established. See Youman, “Muhlenberg, 60. For more on Shecut, see also: Maisch, Mühlenberg als Botaniker, 13; Greene, American Science, 112; Petersen, New World Botany, 341; Ewan and Ewan, Barton, 628. According to Renate Wilson, Albrecht von Haller († 1777) was a Mühlenberg correspondent, too, which is impossible, as von Haller died years before Mühlen- berg became an active botanical networker in 1784. See Wilson, “Second Generation,” 236. The same is true for Greene’s claim that André Michaux (1746–1802) exchanged letters with him, when it was actually his son Francois André Michaux (1770–1855) that corresponded with him. Greene, American Science, 255. In another case, Maisch, Youman and Ewan and Ewan contend that Mühlenberg corresponded with William Aiton (1731–1793) of Kew Gardens. In fact, Mühlenberg tried to contact Aiton in 1792, but the respective letter did not reach Aiton before his death in spring 1793. See Flora Lancastriensis APS 580 M89f, entry for 12/07/1792; Ewan and Ewan, Barton, 571; Maisch, Mühlenberg als Botaniker, 34; Youman, “Muhlenberg,” 69. Furthermore, no evidence of contact with August Johann Georg Karl Batsch (1761–1802) could be found in Mühlenberg’s letters and diaries, contrary to the claims of Schiedt, “First President,” 512; Maisch, Mühlenberg als Botaniker, 34; Youman, “Muhlenberg,” 69; Hu and Merril, “Publications,” 3. Maisch and Youman have also named the botanist Johann J. J. Dille- nius (1687–1747), who died 11 years before Mühlenberg’s birth. See Maisch, Mühlenberg als Botaniker, 34; Youman, “Muhlenberg,” 69. Finally, no evidence of sustained contact could be found in the cases of James Dickson (1738–1822), Lewis Weston Dillwyn (1778–1855) and Albrecht Willhelm Roth (1757–1834). See Mears, “Herbarium,” 170. 125 Nickisch, Stilprinzipien, 161f.; 205; Kempe, “Anglo-Swiss,” 74. This trend also concurred with contemporary developments in literature, which saw the rise of epistolary novels such as Sam- uel Richardson’s (1689–1761) Pamela (1740). Based on a turn towards a less stylized and more 42 Networks and History

principles in German letters of the 17th and 18th century names Benjamin Neukirch (1665–1729), Johann Christoph Stockhausen (1725–1784), Johann Wilhelm Schau- berts (1720–1751) and Christian Fürchtegott Gellert (1715–1769) as the main champions of these developments. Specifically Gellerts Briefe, nebst einer Prakti- schen Abhandlung von dem guten Geschmack in Briefen (1751) summarized and boosted this change of taste, which favored “natural emotions” and dismissed the “Kanzleystil” (“chancellory” or “formal” style) as too artificial and stiff to reflect a writer’s inner emotions and intentions. This change becomes especially explicit in a direct comparison between Heinrich Melchior Mühlenberg and his son Henry. Whereas Mühlenberg senior continued to emphasize the importance of the knowledge of formal letters styles,126 his son’s letters show a much more individual and less stylized approach to his choice of language and form.127 Typically, letters of introduction in Mühlenberg’s correspondences, through which either he or his op- ponent sought to establish contact, were filled with polite introductions, namedrop- ping of common friends or – in the case of a previous personal acquaintance – with memories of the time spent together.128 A critical reading of Mühlenberg’s letters with a focus on these literary topoi would certainly confirm that he was aware of styles and writing-techniques, but would ultimately reveal little or nothing about

individual language in letters, the genre’s success must be seen in its promise to give the audi- ence a much more intimate look into the minds of the novels’ protagonists. Müller, Brief, 155. 126 See the following passage: Indeßen legte ich der Committee die Documenta, Power of Attorney [Vollmacht], Rechnungen, Bande [Schuldscheine], Quitungen und alles übrige zu dem Hoch- gr[ä]fl[ichen] S[olms-] R[ödelheimischen] Legat gehörige vor, und sie saßen lange genug und hatten hinlängliche Zeit und Gelegenheit alles durch zu sehen. Ob die Herren /:H[err] Lew[is] Weiss aus genommen:/ den stilum curiae [Kanzleistil] in den deutschen Documenten verstun- den, oder nicht verstehen wolten, das weiß ich nicht. Mühlenberg senior to J. C. Kunze, 02/01/1779, Aland, Mühlenberg Korrespondenz V, (letter 725). 127 This becomes obvious in an early letter to Schreber: Mit meiner botanischen Schwäche und meiner überall sichtbaren Eilfertigkeit die ich bei der Menge meiner übrigen Geschäfte nicht wohl vermeiden kann, und meine Amerikanische Schreibart ohne die geringste Formalität, bitte ich besonders zu übersehen. To Schreber, 06/16/1788, UAE Briefnachlass Schreber. 128 The so-called “talk among absentees” (“Gespräch unter Abwesenden”) was one of the most common stylistic devices of contemporary epistolary theory and also appears frequently in Mühlenberg’s letters. See the following passage from a letter by Schöpf, whom Mühlenberg had met in late 1784: “Ihre schriftliche Unterhaltung wird uns daher auch in Zukunft so angenehm als interessant sein u[nd] bleiben.” From Schöpf, 04/03/1786, HSP Soc. Coll. A similar passage is contained in his first letter to his first correspondent at Halle, Gottlieb Friedrich Stoppelberg († 1797): “Ich erinnere mich nie ohne besondere Vergnügen der Zeit da ich unter Ihrer Leitung mich in Halle den Wißenschaften widmete, und eine Rechnung von Ihrer Hand dat[iert] Jun. 20. 1790 erneuerte Ihr Andenken so lebhaft in meinem Gemüth, daß ich mir so gleich vornahm, bei erster Gelegenheit mich Ihrer vorigen Freundschaft zu empfehlen, (…).” To Stoppelberg, 03/14/1791, AFSt M.4 D3. Generally, the individual letter was perceived both as a “mirror of the soul” of the author and a physical representation of his body. See Mauelsha- gen, Netzwerke, 133: “Frühneuzeitliche Gelehrtenbriefe sind voll von Beschreibungen des memorialen Aktes der imaginären Vergegenwärtigung des anderen, der aus solchen Gemein- plätzen gelebte Realität machte” See also Müller, Briefe, 138–140; 150; Barner, “Freund- schaft,” 34; Steinke, Brief, 42f. Aims and Methodology 43 the development of individual contacts, which can be followed more precisely in the “Exchange Charts,” Appendix D.129 Finally, there is the problem of temporarily resting contact activity, when a contact was interrupted by long breaks of no letters, although passages from other letters suggest that Mühlenberg was actually still expecting a response from these “sleeping” contacts. This is also a source-based problem, as in most of these cases, the information available today simply does not suffice to decide whether Mühlen- berg considered the tie “dead” or whether his attention was only directed towards other, more promising contacts at the time. A case in point is the contact with Jo- hann David Schöpf (1752–1800), which began in 1784 with Schöpf’s visit to Lan- caster and was effectively discontinued in 1791, when his professional career pre- vented him from maintaining his correspondence with Mühlenberg until 1800, when he died unexpectedly at 48 years of age. Until then, Mühlenberg was actually expecting the contact to continue at some point. As the network drawings are cons- tructed from his ego-perspective, the Schöpf-contact must therefore remain present until 1800, even though not a single instance of direct contact between the two men in Lancaster and Erlangen could be found from 1790 to 1800. Consequently, only the death of a correspondent ended a contact for real, both in Mühlenberg’s percep- tion and in the network drawings in Appendix E.130 In rare cases, however, he was aware that a contact had come to an end even before, as was the case with Constan- tine Samuel Rafinesque (1783–1840). In this case, Mühlenberg showed extreme reluctance to conduct the botanical exchange from the start and was thoroughly relieved when Rafinesque left the United States in 1804 for Europe. Therefore, their communication was only active briefly from April 1803 to August 1805, when he submitted a final letter from Palermo to which Mühlenberg never answered.131 In order to integrate this highly segregated data base into cohesive and interpre- table UciNet©-network visualizations, four types of ties have been used, whose graphical appearance reflect the reliability of the underlying data. The skeleton of the six network drawings is composed of the core corpus of 693 letters, the 297 reconstructed letters and those of Mühlenberg’s contacts among each other, as far as they could be located. Ties based on this material are the most reliable and the only quantifiable source of network information. Therefore, they will be valued according to the number of letters in a specific contact during any given network phase, which will help to keep track of the development of individual contact inten- sities over time. Ties of this sort in Appendix E will be represented by solid and valued lines. Secondly, Mühlenberg’s letters and diaries contained a plethora of information on individual contacts, which are certainly accurate, but most often not very precise. A passage like Collins’ note on his visit to Bartram’s garden, quoted above, cannot be ignored, but it cannot be quantified either. In cases like this, a basic connection between two individuals was assumed that is represented by a tie

129 See Appendix D, 535f. 130 Consequently, the Schöpf-contact is also represented until 1800 in Flow Chart A, Appendix A, 483. The Flow Charts and the Network Visualizations are generally based on the same source material to avoid inconsistencies. 131 See Flow Chart A, Appendix A, 483 and respective list of correspondences, Appendix C, 528. 44 Networks and History

of a dashed line of value one. Thirdly, a dashed-dotted line of value one was used to represent ties that are based on vague information in secondary literature references,132 while dotted lines of value one represent “sleeping” contacts in the phases where no contact information both for a continuation or, in fact, a discontinuation, could be found. Furthermore, correspondents located in the United States are represented by triangular nodes, those in European countries by circular nodes.133 Correspondents that share the same place of residence also share the same node color, in order to illustrate the local clusters of collaboration. The sources used in the composition of these six networks are noted in Appendix F.134 With the exception of Network Phase 6, tie strengths in all network drawings generally correspond to the actual number of letters exchanged.135

132 Specifically older research literature has a lot of references on individual relationships, where the author omitted proper source references or where the wording is simply too vague to be entered into the network matrix. In those cases, where supportive information for ties like these could be found in other sources or other articles, these ties were integrated as solid grey lines of value one. For a list of persons wrongly cited as Mühlenberg’s correspondents, see above on page 41, note 124. 133 Please note that the primary distinction has not been drawn according to the nationality of individual correspondents, but with regard to their location, in order to reflect the respective growth of Mühlenberg’s European and American networks of correspondence and their changing relationship to each other. 134 See Appendix F, Network Documentation, 554f. 135 In Network Phase 6, the high number of letters exchanged with William Baldwin (93 letters), Zaccheus Collins (83) and Stephen Elliott (36) render a graphical representation of a 1:1 ratio impossible. For this reason, the highest tie strength has been preset to 31, which equals one third of Baldwin’s letters. See Appendix E, Network F: Phase 6, 552.

IV A PRELUDE – MÜHLENBERG’S CORRESPONDENCES FROM 1771 TO 17841

Long before Mühlenberg began to weave his own web of correspondences in the mid-1780s, he had grown into an intricate network of family members and fellow Lutheran pastors, working in the Pennsylvania field and at the Halle orphanage. During his various sojourns at Philadelphia in the 1770s, he must also have begun to take notice of the local scientific community, whose undisputed center Philadel- phia was at the time. Since mid–18th century, the American branch of the “Republic of letters” had begun to organize itself under the tutelage of Benjamin Franklin, who was also the first American to be accepted by European scientists as one of their peers.2 It is in these three contexts - Mühlenberg’s Lutheran-Pietist background, his family connections in Pennsylvania, and the nascent American scientific community – that I want to present him, in order to provide the background for a network-theory driven analysis of his correspondences after 1784. During the fall of 1783, Mühlenberg housed a traveling German naturalist, Jo- hann David Schöpf, who crossed through Lancaster in late November. To Mühlen- berg, the visit opened the doors to the European theater of science. By far the bigger portion of his letters3 were written in the years between Schöpf’s visit and 1815, and bear witness to the fact that his scientific interests should soon claim a considerable amount of his time and energies.4 It is tempting to conceive him only as a botanist, and quite easy to forget that he was at the same time a father to eight children and a professional Lutheran minister by vocation. Therefore, this chapter’s aim is to situate Mühlenberg in a context of actual and potential channels of communication in between 1770 to 1783, focusing on his professional and familial ties, and his place within the Republic of Letters, especially within the development of its American offshoot in the course of the 18th century.

1 THE LUTHERAN CONTEXT

It was during Mühlenberg’s stay at the Halle Orphanage from 1763 to 1770 that the young student of the Lateinschule sent his first letters across the Atlantic to let his

1 All data in this chapter is based on letters sent or received between Mühlenberg’s letters to his father H. M. Mühlenberg, 12/04/1771, Aland, Mühlenberg Korrespondenz IV, (letter 575), and 01/02/1784, Aland, Mühlenberg Korrespondenz V, (letter 924). 2 For his originial research on electricity, Franklin was awarded with the Royal Society’s Copley Medal in 1753. Greene, American Science, 37. Jefferson’s Notes on the State of Virgina, however, was seen much more critical by European scientists. Hindle, Pursuit of Science, 255, 322. 3 See table a, Appendix B, on page 486. 4 See table b, Appendix B, on page 486. 46 A Prelude

parents know about his progress in learning.5 Unfortunately, not a single line of his hand from this period has survived. The fact, however, that it was possible to submit letters to his native Pennsylvania with relative ease must have given him his first taste of the global network through which the Orphanage staff coordinated their missionary efforts, raised money to support the Francke Foundations, and sent Bi- bles, books and medicines to all corners of the world. The later success and expan- sion of the initially humble Francke Foundations at Glaucha near Halle would have been unthinkable without August Herman Francke’s (1663–1727) talent to make connections to dignitaries and people in key positions. Carl Hildebrand von Can- stein (1667–1719) and Gneomar Dubislav von Natzmer (1654–1739) were his most effective political agents, while Francke himself managed to secure the crown’s benevolence time and again by clever political maneuvers.6 Another link to minor and major courts and regional commercial networks were well-connected women from the higher and lesser nobility, as Renate Wilson has argued. These fundatrixes not only offered social and financial support7 for Francke’s enterprises, but also came to be a vital part of the network via marriage policies and resulting family connections.8 Thus, only a few years after the Foundations’ establishment in 1698, a network had evolved which “spanned continents and oceans and survived its founder by many decades.”9 In Europe, Francke’s most important single contact during this early phase, Heinrich Wilhelm Ludolf (1655–1712), managed to secure the Halle Pietists stable links with the courts of Russia, England and Denmark,10 thus functioning as door- opener to most prestigious fields of influence and evangelicalisation. By 1706, again mainly through Ludolf’s support, Francke’s emissaries were firmly established in Russia,11 London and India. Bartholomäus Ziegenbalg (1682–1719) and

5 Meine 2 Söhne in den gesegneten Anstalten haben geschrieben und auch durch Hn: Br: Helmuth mündlich bitten laßen, daß sie nun gern die studia academica anfangen mögten, wenn Sr: Hochw: Herr Dir: und Consistorial= Rath Dero hoch=väterl: Genehmhaltung dazu zu geben geruhen würden. H. M. Mühlenberg to G. A. Francke and F. M. Ziegenhagen, 04/15/1769, Aland, Mühlenberg Korrespondenz IV, (letter 463). 6 In the kingdom of Prussia of King Frederick I (1657–1713), Francke responded to the mon- arch’s need for field clergy, thus securing longterm state support for his Foundations. Brecht, “Pietismus,” 500f. 7 Financial aid mostly came in the form of direct, personalized support for missionary candi- dates, whose passage, supply of clothing and medications were paid for. Other forms of support consisted in collections organized by these Fundatrixes, or bequests to be transferred to the missions. Renate Wilson quotes the case of Willhelmine Sophie von Münchhausen née Wan- genheim (no data available) to illustrate this practice. Wilson, Pious Traders, 36. 8 Most prominently, Martin Brecht highlights the case of Sophie Luise von Mecklenburg (1685– 1735), who clung to her Lutheran faith after her marriage to Frederick I., which opened new doors to Francke. Brecht, “Pietismus,” 499. See also Wilson, Pious Traders, 33. 9 Müller-Bahlke, “Communication,” 139. 10 Ludolf managed to make most of these connections because of his position as secretary to King George of Denmark, the latter husband of Queen Anne. Brecht, “Pietismus,” 511, 514–517. Renate Wilson has argued that “establishing and maintaining this link was possible by a constellation of political and religious trends.” Wilson, Pious Traders, 2. 11 In the case of Russia, contacts could soon be extended to include the Russian court and the The Lutheran Context 47

Heinrich Plütschau (1676–1752) began proselytizing in the Danish colonies in the name of king Frederick IV of Denmark (1671–1730),12 while Anton Wilhelm Böhme13 (1673–1722) figured as Francke’s first link to London’s Court Chapel, the center of English pietism, before Friedrich Michael Ziegenhagen (1694–1776) came to occupy the position from 1722 to 1776.14 Thomas Müller-Bahlke has described the stunning case of the fraudulent “Pastor Schultz,” which illustrates the dimensions and efficiency the Halle network had already attained by the 1730s. When Schultz was traveling in southern Germany under the false guise of a Halle representative to collect donations for the mission in North America,15 Francke and Ziegenhagen quickly activated their contacts in the region to get hold of him. Through their joint efforts and with the help of their local correspondents, they finally managed to make the case publicly known, have Schultz arrested and his false credentials and collections confiscated.16 Francke’s original intention in setting up his network had been to achieve independence from state and church interference, a goal which must be seen in the wi- der context of European Reformation. In the beginning of the Protestant movement, various unconnected groups and individuals saw themselves confronted with a system of deep-rooted familial ties, trade privileges conferred along confessional lines and the challenge of Catholic counter-reformation. In order to resist and survive, mutual support across territories and nations needed to be established.17 Even the so-called “Republic of Letters,” the early modern communication system of scientists and natural philosophers, was essentially an outgrowth of religious persecution and joint efforts to re-establish contacts within a new, confessional Europe. Mutual moral support and offers to take shelter from persecution, however, would scarcely have sufficed to sustain the movement for long, if not a system of financial support had come into place as well. The Francke Foundations yield an exceptional example for this. After Francke’s ideas of Christian charity, epitomized by the Orphanage, had acquired credit and respectability in Pietist-Lutheran circles, he began to make use of his contacts to raise the necessary funds for the cause. Donations were a first step, but trade in books and medicines promised both to augment financial resources and to be an

immediate surroundings of the Czar himself. Brecht, “Pietismus,” 519. 12 Brecht, “Pietismus,” 527. 13 Boehme’s predecessor J. W. Mecken had denied further service to the crown in consequence of the king’s attendance to Anglican masses. It was again Ludolf who acted as the pivotal agent to replace Mecken with Boehme. Brunner, Halle Pietists, 51; Splitter, Pastors, 14. 14 “In Ziegenhagen, Halle not only had a close friend and correspondent, but someone who filled the office for fifty-four years.” Brunner, Halle Pietists, 57; Brecht, “Pietismus,” 521. 15 Glatfelter, Pastors II, 32. 16 Müller-Bahlke, “Communication,” 143. 17 Wilson, Pious Traders, 25f. Apart from this rather utilitarian interpretation of Pietist networks, D. F. Durnbaugh has pointed out that exchange and the continuation of contact with like- minded Christians on a global scale was characteristic of pietist thought. Durnbaugh, “Com- muncation,” 34. For brief sketches of similar pietist networks, see Durnbaugh, “Communca- tion,” 35f. 48 A Prelude

effective means to spread their ideas of charity and reform.18 By mid-century, inner- European19 and transatlantic commerce with products from the Medikamentenexpe- dition and Halle’s Bible printing press had developed into a major business.20 Fran- cke and his successors relied on a widely ramified network of sales agents, intermediaries and accredited pharmacies, apothecaries and physicians to market their products. In North America, the Mühlenberg-family and John Christopher Kunze (1744–1807) acted on behalf of Halle in this business.21 Apart from this developing trade, Halle’s network also served to handle all financial transactions associated with the trade itself and external financial occurrences such as bequests and legacies. Additionally, it gave the Halle Pietists access to potential investors and served them as security for low-interest credit, given to a cause which promised to be both a safe investment and a good Christian deed.22 Halle’s network not only served material purposes, but was also a working communication system that allowed every single pastor and associate of Francke’s to participate in the worldwide progress of God’s word and the Christian mission. Halle publications such as the Halle’sche Korrespondenz (1707), the Merkwürdige Nachricht aus Ost-Indien (1708), the later Halle’schen Berichte and the Merkwür- dige Nachrichten von einigen evangelischen Gemeinden, sonderlich in Pennsylvanien,23 which combined mission reports from India and North America, played a crucial role in this transatlantic news exchange. According to common Christian belief, the Kingdom of God would come again once the gospel was heard in every far corner of the earth,24 and the desire to stay in touch with recent developments created the need for centralized, communal platforms that everyone could easily access and subscribe to. Thomas Müller-Bahlke has called this a “minor masterpiece” of “contemporary public relations.” 25

1.1 Studies at the Francke Foundations (1763–1770)

The ten-year-old Henry Mühlenberg and his two older brothers Johann Peter Gab- riel and Frederick Augustus Conrad embarked for Europe on April 27, 1763 aboard

18 Müller-Bahlke, “Communication,” 144; Wilson, “Second Generation,” 255; Wilson, Pious Traders, 29; Brunner, Halle Pietists, 129. 19 Apart from German territories, Russia soon came to be one of the most profitable markets for Halle medicines, bibles and diverse pedagogical materials to support evangelisation and charity modelled after the Halle Orphanage. Brecht, “Pietismus,” 518f. Brunner points out that “Francke was his own bookseller and publisher and had the financial backing and official influence of Canstein, who insisted that Bibles and New Testaments be available at the cheapest possible prices. Halle was far more of a model to the great nineteenth-century Bible societies in Britain and North America than the SPCK.” Brunner, Halle Pietists, 135. 20 In 1765, net profits reached 34,000 Rth. Wilson, Pious Traders, 93. 21 Wilson, Pious Traders, 31. 22 Wilson, Pious Traders, 30. 23 Brecht, “Pietismus,” 527. 24 Wellenreuther, “Atlantische Welt,” 15. 25 Müller-Bahlke, “Communication,” 140. The Lutheran Context 49 the ship of “Captain Budden” to begin a seven-year course of studies at the Francke Foundations, where their father had also lived and studied theology.26 On June 15, 1763, the three brothers were introduced to court preacher Friedrich Michael Zie- genhagen (1694–1776) in London by a certain “Chirurgus Meyer,” whom Heinrich Melchior Mühlenberg had selected as travel guide for his sons. In a letter to Gotthilf August Francke (1696–1769), Ziegenhagen acknowledged that he had only come to see Peter and Frederick, as the youngest son Henry had (…) wrestled with the boy of a skipper and came out of it with a black eye. This is why I have not seen the same.27 Continuing via Amsterdam and the ancestral Mühlenberg family hometown of Einbeck, where the company briefly lost track of Henry, who only managed to proceed with the help of a stranger carrying him on his back all the way to Einbeck,28 the party arrived at the Orphanage in early fall of 1763. While Peter was immedia- tely admitted to the Lateinschule on September 2, 1763,29 Frederick and Henry had to wait until December 3 of the following year for their admission.30 As far as the contents and subjects of his education at Halle are concerned, Mühlenberg rarely ever alluded or commented on them in his later writings, which leaves us with what is generally known about the contemporary Halle curriculum. In addition to his major subject theology, he underwent intense schooling in He- brew, Latin, Ancient Greek and French,31 which was supplemented by courses in

26 Until then, the Mühlenberg sons had received their basic schooling at their Trappe homestead and later at Philadelphia. Muhlenberg-Richards, “Gotthilf Heinrich Ernst Muhlenberg,” 147. According to J. Bennet Nolan, this Captain Budden was one of Philadelphia’s best known and most trusted captains to undertake the transatlantic crossing. Nolan, Smith Family, 55. Beck, however, claims that the name of the ship was Captain Budden. Beck, “Muhlenberg, Botanist“, 99. 27 Ziegenhagen wrote that Mühlenberg had (…) mit einem Schiffer Jungen geschlagen und ein blaues Auge bekommen, und folglich habe ich selbigen gar nicht gesehen. Later in the same letter, he also wrote: Am 15 aber fand sich ein Chirurgus Nahmens Meyer (...) der sich meist im Jahr in Pensylvanien aufgehalten, sonst aber zu Einbeck wohnhaft ist, bey mir ein mit dem äl- testen Sohn des Herrn Mühlenbergs. (...) Ehe und bevor ich selbiges erörterte, zeigte H Meyer an, daß Er nicht allein den ältesten Sohn, sondern auch die 2 folgend, also 3 Söhne des H[errn] M[ühlenberg] mit sich von Philadelphia herausgebracht. Johan Peter (16 Jahr) Frederick Au- gust 12 Jahr und Heinrich... 10 Jahre alt. Frage: Wo sollen diese Kinder hin? R[esponsio] alle 3 nach Halle, und werden der Disposition, Direction und Verordnung zu Hochw. des Herrn D. Franckens gäntzlich übergeben. Ziegenhagen to Francke, 06/29/1763, AFSt/M 1 D 8 : 14. 28 Youman, “Muhlenberg,” 59; Muhlenberg-Richards, “Gotthilf Heinrich Ernst Muhlenberg,” 148. 29 Matrikel der Lateinischen Schule, Mühlenberg, Jo. Peter Gabr. 02.09.1763 AFSt/S L 4 255 – 9141. 30 Matrikel der Lateinischen Schule, Muhlenberg, Frid. Aug. Conrad 03.12.1764 AFSt/S L 4 293 – 9329; Muhlenberg, Henry Heinr. Ernst 03.12.1764 AFSt/S L 4 293 – 9330. See also Wilson, Pious Traders, 23. 31 Wallace, Muhlenbergs, 59. In his later letters, Mühlenberg frequently complained about his lack of proficiency in French, which kept him from reading French botanical publications: Should You be so fortunate as to receive the Continuation of Michaux, you will add very much to the Obligations I owe you already by letting me have a Sight of them. Pray is the promised Translation published? The French Language I only understand imperfectly and I purpose to wait for a Translation. To Collins, 03/19/1812, ANSP Coll. 129. 50 A Prelude

history, geography, mathematics and music. Periodically, botany also appeared in the curriculum of the Halle Lateinschule during the 18th century,32 but for lack of any direct statement by Mühlenberg, we must assume that he never attended any courses in the favorite science of his older days. The only two statements about his education at Halle from his own hand do not mention botanical studies, but rather imply that his interest in plants only developed at a more mature age.33 Renate Wil- son has contended that the elder Mühlenberg’s second thought about sending his sons to Halle was to provide one of them the oppportunity for medical training, which Peter was supposed to receive in the original scheme.34 When he refused to follow his father’s wish, no successor for this arrangement seems to have been picked.35 In any case, neither Frederick nor Henry seem to have excelled in their subsequent studies at Halle.36

32 “Als eigene Lehranstalt wurde die Lateinschule im September 1697 eingerichtet. Hier sollten befähigte Knaben auf das Universitätsstudium vorbereitet werden. (…) Der Lehrplan der Lateinischen Schule legte den Schwerpunkt auf die Vermittlung der alten Fremdsprachen Latein, Griechisch und Hebräisch. Außerdem wurde Geschichte, Geographie, Mathematik, Musik und Botanik unterrichtet, wie Francke 1701 berichtete. Das ganze 18. Jahrhundert hin- durch blieben dies die wesentlichen Unterrichtsfächer trotz kleinerer Veränderungen. So wurde schon bald Physik in den Lehrplan aufgenommen, Anatomie und Malen traten zeitweise hinzu, während der Botanikunterricht phasenweise wieder ausgesetzt wurde,” Müller-Bahlke, Latein- schule, 40f. In 1733, the Halle physician Friedrich Hoffmann (1660–1742) praised the local conditions for science: “Denn es fehlet ihnen weder an genugsamen Vorrathe der kostbarsten Instrumenten, so zur Erforschung der Natur und Kräften der Elementen (…) noch an auserlese- nen Naturalien, welche von allen, die sie sehen, admiriret werden. Hiernechst findet man auch daselbst schöne Gelegenheit in der Anatomie durch Secirung der Thiere, in der Materia Me- dica, und in der Botanic zu profitiren,” Müller-Bahlke, “Naturwissenschaft,” 367. See also Aland, Korrespondenz IV, 10; Wallace, Muhlenbergs, 74f; Roeber, “Helmuth,” 82. 33 When Mühlenberg was accepted into the Berlin Gesellschaft Naturforschender Freunde in 1799, he was asked to give a short CV for the society, which briefly mentions Halle: Weil Sie es ausdrücklich verlangen, daß ein jedes Glied der Gesellschaft etwas gewißes von seinem Leben melden soll, so merke ich folgendes an. Ihr Freund, Henry Heinrich Ernst Mühlenberg ist in Pensilvanien d. 17. Nov. 1753 gebohren. Sein Vater war der durch die Pensilvanische Nach- richt bekannte D. Heinrich Melchior Mühlenberg. Von 1763–1770 war er in Halle und stud- ierte Theologie. Im Sept. 1770 wurde er von dem Lutherischen Ministerio zum Prediger ordiniert. Nachdem er von 1771 bis 1777 mehrentheils in Philadelphia als Prediger gestanden mußte er mit der Hälfte seiner Gemeinsglieder Philadelphia verlaßen, das von den Engländern besetzt worden. Von seinen Büchern entfernt und in der Stille des Landlebens fing er an sich in der Natur umzusehen, und ohne Buch und Lehrer die Natur zu studieren. Er wurde bekanter mit ihr und gewann sie alle Tage lieber. To GNF, 10/02/1799, HUB GNF S, Mühlenberg H. E. See also Wilson, “Second Generation,” 238. 34 Wilson quotes from a letter of Mühlenberg to Inspektor Schulenburg dating from March 1759, in which the two men discuss some details on the planned studies at the Foundations. Wilson, Pious Traders, 121. In a letter to Francke, dated June 29, 1763, Ziegenhagen also mentions an apparent disposition of the oldest Mühlenberg son to become a surgeon. Ziegenhagen to Francke, 06/29/1763, AFSt/M 1 D 8 : 14. 35 Instead, Peter soon began an apprenticeship with a merchant Niemeyer at Lübeck. Aland argues that sending Peter off to Lübeck was their father’s original intention plan rather than an alternative to Halle. Aland, Korrespondenz IV, 8. 36 Aland, Korrespondenz IV, 10; Wallace, Muhlenbergs, 69. Muhlenberg-Richards, however, claims that “[a]t Halle he showed such marked proficiency in his studies, and such dilligence in The Lutheran Context 51

Not much is known about Henry’s social relations and friendships at the Orpha- nage, either.37 Apparently, he made friends with an “Accise-Einnehmer Garliep,” to whom he submitted one of his first letters after his departure from Halle in 1770,38 detailing the circumstances of their voyage back to North America and his first experiences as a pastor in the Pennsylvania field. In a later letter to Sebastian Andreas Fabricius (1716–1790) from 1775, Georg Christian Knapp (1753–1825), the son of the former Orphanage-principal Dr Johann Georg Knapp39 (1705–1771), is named as one of his closest friends.40 Another vague trace of Mühlenberg’s personal relations from 1763 to 1779 comes from a letter to Gottlieb Friedrich Stoppelberg († 1797), Mühlenberg’s second permanent correspondent after Fabricius. From him, he received 1795 greetings by the old gardener Schönberg, with whom both Henry and Frederick obviously shared a friendship.41 Above all, however, Inspektor Fabricius appears as the most cherished and re- spected contact during these seven years. For the three young Mühlenbergs, Fabri- cius came to be their substitute father, which echoes throughout their subsequent letters until Fabricius’ death in 1790.42 Fabricius had already been a close friend of their father’s, and their friendship was probably the basis for the relationship with his sons.43 Equally tight bonds must be assumed to Justus Henry Christian Helmuth (1745–1825) and John Frederick Schmidt (1746–1812), Henry’s future colleagues in Pennsylvania: The 23rd of July 1768 – This day I was obliged to leave Halle, and take leave of all my benefactors, and friends, (...), Schmidt noted in his diary much later, (...). I was accompanied out of Halle, for a few miles, by Mr. Inspector Fabri- cius, Peter Muhlenberg, and Frederick Augustus Muhlenberg. The evening was fine

their preparation, that he attained the head of his class.” Muhlenberg-Richards, “Gotthilf Hein- rich Ernst Muhlenberg,” 148. 37 Wallace reports an episode of Henry harrassing a younger schoolboy in the presence of school- mates and teachers. The incidence is recorded in a letter by Knapp. Wallace, Muhlenbergs, 76f. 38 To Accise Einnehmer Garliep, Halle, 04/05/1773, AFSt/M 4 C 17. 39 Knapp took over when Gotthilf August Francke died in 1769, but died himself only two years later. 40 2. An des sel. Hr. D. Knapps Herren Sohn, meinen ehemaligen guten Freund, die ihm schuldi- gen 4 r. ohne zu forschen wie. Der Herr segne ihn überschwänglich! To Fabricius, 01/03/1775, AFSt/M 4 C 17 : 26. 41 Empfehlen Sie mich bey Gelegenheit besonders Ihrem 2n Herrn Bruder bestens, den auch der alte Gärtner Schoenberg grüßen läßt. From Stoppelberg, 05/15/1795, APS Film 1097. 42 Meine Brüder sind noch beide in öffentlichen Ämtern und sind für die teutsche Nation in diesem Lande eine gute Stütze. Wir denken bei jeder Zusammenkunft noch mit vieler Rührung an die väterliche Liebe die Sie zu uns getragen, und mir insondernheit ist der Name Fabricius un- vergeslich und theuer, so wie ich Gott stets überhaupt danke daß er mich in den gesegneten Anstalten des Waisenhauses hat auferzogen werden laßen. To Fabricius, 11/01/1785, AFSt/M 4 D 20. 43 See H. M. Mühlenberg to Fabricius, 03/31/1773, Aland, Mühlenberg Korrespondenz IV, (letter 604): Mein Hochgeehrt und hochgeschätzter Herr, Bruder, Gönner und vieljäriger Woltäter, ja Vater wolte ich sagen, wenn wir nicht fast von einerlei Alter wären! Weil ich in einen Jüngern Jaren eine starcke EinbildungsKrafft, die man sonst Imagination nante, besaß, so kan mich Dero Werteste Person noch lebhafft vorstellen, vermöge der von Gott verliehenen Gelegenheit Dieselben im Winter 1742 (…) von Person kennen zu lernen. 52 A Prelude

and I rode on praising the goodness of God, to Barbury.44 Half a year before they finally left Halle, Henry and Frederick finished their studies. For the remaining time, Henry enrolled at the University of Halle on October 10, 1769 as a student of theology.45 Heinrich Melchior Mühlenberg had agreed to this short-time enrollment on the single condition that the two brothers would not be allowed to move into town during their studies.46 On May 4, 1770, they started their journey from Halle across Germany via Aschersleben, Wolfenbüttel, Lüneburg to Altona,47 where the party probably met with Jacob Gysbert van der Smissen (1746–1829) of the van der Smissen & Söhne Company, a long-standing Halle contact and agent in their transatlantic communi- cations. Unfortunately, the letter from van der Smissen to Fabricius, in which the incident is recorded, does not clearly reveal whether the meeting took place or not.48 In Altona, they embarked for London to meet with Ziegenhagen’s assistant Friedrich Willhelm Pasche49 (1728–1792) in Kensington, who had already arranged for their passage back to Pennsylvania. On June 26, 1770, Pasche informed Fabri- cius in a short letter that the three had arrived safely at Kensington a few days earlier and that Ziegenhagen had plans to employ them during their short stay before they would continue their travels.50 After a month with Pasche, the three finally left

44 Nolan, Smith Family, 51. 45 UAH-W, Matrikel 1767–1781. See also Müller-Jahncke, “Linnaeus Americanus,” 1323. 46 Meine 2 Söhne in den gesegneten Anstalten haben geschrieben und auch durch Hn: Br: Helmuth mündlich bitten laßen, daß sie nun gern die studia academica anfangen mögten, wenn Sr: Hochw: Herr Dir: und Consistorial= Rath Dero hoch=väterl: Genehmhaltung dazu zu geben geruhen würden. Ich überlaße es gäntzlich der höhern und beßern Einsicht Hochwürdig- ster Väter, nur mit demüthigstem Vorbehalt, daß sie nicht in die Stadt ziehen, sondern in den Anstalten bleiben mögten. Ich und mein Weib wollen hertzlich gern alles bezahlen, welches wir desto beßer können, wenn wir aus der Barrenhiller Sache nach und nach entwickelt werden. H. M. Mühlenberg to G. A. Francke und F. M. Ziegenhagen, 04/15/1769, Aland, Mühlenberg Korrespondenz IV, (letter 463). 47 The details of this journey are outlined in a statement from Kunze’s hand, in which he reported back to Halle all costs accrued until their arrival at Kensington on June 21, 1770. The full itinerary took included Könnern, Aschersleben, Halberstadt, Wernigerode, a brief detour to Ein- beck, Wolfenbüttel, Braunschweig, Gamsen, Uelzen, Lüneburg, Hamburg, Harburg, Altona, Altona and per Haxney-coach to Kensington. Verrechnung der Reisekosten von Johann Chris- toph Kunzens und der zweyen Söhne des Herrn Pastor Mühlenbergs von Philadelphia Reise von Halle bis London angetreten d. 4. Mai 1770 und durch Gottes Gnade vollendet d. 21. Jun ebend. Jahres. AFSt/M 4 A 8 : 8a. 48 Jacob Gysbert van der Smissen to Fabricius, 05/16/1770, AFSt/M 4 A 8 : 17. 49 Pasche was Ziegenhagen’s associate and a lecturer at the local chapel. It was him who informed the elder Mühlenberg about Ziegenhagen’s death in 1776. Brunner, Halle Pietists, 86, 196, 197. 50 As it turned out, neither of them felt ready to answer Ziegenhagen’s wish on account of fatigue in the wake of the passage to London. Daß der liebe H. Kunze mit den beyden jungen H. Mühlenberg am 21sten hujus abends um 5, endlich Gottlob! wohlbehalten hier angekommen sind, hat der ältere H. Mühlenberg mit lezter Post an E[uer] Hochwürden vorläufig berichtet. Der theure Herr Hofprediger (...) wünschen, daß Sie ihnen, während ihres hiesigen Aufen- thaltes, einige nüzliche Dienste thun könnten, Sie finden Sich aber wegen anhaltender großer Entkräftung nicht im Stande, solchen Wunsch zu erfüllen. Ihre Bagage wird morgen oder über- morgen vom Schiff und Customhouse losgemacht und herausgebracht werden können. Aber die Pensilvanische und unsre Continuations-Liste sind noch nicht hier. (...) Pasche to Fabricius, The Lutheran Context 53 for Philadelphia aboard the “Duchess of Gordon” with Captain Winn on July 24, and after a brief delay at Deptford, they finally left European waters on the morning of July 28, 1770.51 It was Mühlenberg’s only and last visit to the Old World.

1.2 The Pennsylvania Field

During the period between the close of the Seven Years’ War and the beginning of the American Revolution, the ranks of both Reformed coetus and Lutheran ministerium were crucially reinforced by a fresh wave of “young, able and committed pastors,”52 amidst whom were Henry, Frederick and Kunze, who disembarked on American soil on September 27, 1770.53 Almost instantly upon arrival, the two young Mühlenbergs were invited to preach a Thanksgiving sermon at Philadelphia’s newly built Zion’s church under the critical eyes of their father.54 Quite contrary to the two brothers’ apparently low motivation during their Halle studies, Henry seems to have left a lasting impression on his father and colleagues.55 After this first test, they were admitted to oral exams, which they passed, and received their ordination in early October of 1770.56 At first, Henry was sent out to serve congregations at

06/26/1770, AFSt/M 4 A 8 : 8a. In a letter to Vetter Bense at Einbeck, Mühlenberg indeed indicates that the passage to London was a rather unpleasant experience: Wir kamen ohne einig Hinderniß von Einbeck nach Clausthal u. nach Wernigerode, von wo wir auch bis auf Hamburg mit vieler Liebe befördert worden. O wie viele Freunde u. Gönner haben wir da gefunden! (…) Von Hamburg nach London war freylich nicht sehr angenehm, aber auch darinnen erkennen wir die Hand Gottes (…). To Vetter Bense, 04/03/1773, AFSt/M 4 C 17. 51 Pasche to Fabricius, 07/24/1770, AFSt/M 4 A 8 and Pasche to Fabricius, 07/27/1770, AFSt/M 1 D 11 : 13. 52 Glatfelter, Pastors II, 341. In a letter to an unidentified “Accise Einnehmer Garliep“ at Halle, Mühlenberg makes the same observation: Gleich nach unserer Ordination wurden wir zu einigen verlassenen Gemeinen ausgesandt, einige 60–80 Meilen von der Stadt Philadelphia, endlich auf diesen Tag haben wir gleiche Beschäftigungen gehabt, weil wir meist die jüngsten sind u. vielleicht noch mehr ertragen können als andere die schon viel von ihren Kräften im Werk des Herrn verzehret haben. To Accise Einnehmer Garliep, Halle, 04/05/1773, AFSt/M 4 C 17. 53 To Vetter Bense, 04/03/1773, AFSt/M 4 C 17. 54 To Accise Einnehmer Garliep, Halle, 04/05/1773, AFSt/M 4 C 17. See also Wallace, Muhlen- bergs, 77. 55 See the following passage from Schmidt’s diary: The learning of this young Mr. Muhlenberg considering his age and the length of his stay at Halle, not quite seven years, is remarkable. Nolan, Smith Family, 62. 56 Mühlenberg gives contradictory information on behalf the exact date of his ordination. Im Sept. 1770 wurde er von dem Lutherischen Ministerio zum Prediger ordiniert. To GNF, 10/02/1799, HUB GNF S, Mühlenberg H. E. Henry and his older brother Frederick obviously passed their exam during the 1770 synod, as indicated in a letter to Vetter Bense at Einbeck: Gleich darauf gingen wir mit auf ein General Synode od[er] Versammlung der hiesig H[erren] Prediger wo wir auch beyde nach gehörigen Examine zum Predigtamt ordiniert u, in das Ministerium aufgenommen worden. Möchten wir doch allezeit unseren Beruf würdig wandeln. To Vetter Bense, 04/03/1773, AFSt/M 4 C 17. In a letter of the same date to Vetter Eicke at Clausthal, however, Mühlenberg dates the event in October: Der andere Bruder wurde mit mir 1770 im Oct[ober] öffentl. von den Ministerien examiniert u. darauf zum Predigtamt ordiniert. To Vetter Eicke, 04/03/1773, AFSt/M 4 C 17. In his letter to Accise Einnehmer Garliep, Halle, 04/05/1773, 54 A Prelude

Barrenhill, Cohansey, Frankfurt and Pikestown for the following four years,57 before he attained his first fixed position as Adjunctus (helper) to his father, and later as third pastor, at Philadelphia’s ever growing Lutheran congregation.58 In the en- suing decade, he would primarily move and act within the Pennsylvanian branch of Halle’s network. North America had been brought to August Herman Francke’s attention by Phi- lip Jacob Spener (1635–1705), who had himself taken notice of German immigrants to America from the Palatinate when Anton Wilhelm Böhme began to include descriptions of their fate in his reports from London.59 Before, Böhme had also translated Francke’s Fußstapffen into English under the title Pietas Hallensis, which came to the attention of the Congregationalist preacher and later president of Yale Cotton Mather (1663–1728) at Boston. With him, Francke subsequently enga- ged in a correspondence that lasted from 1711 to 1726.60 It was Halle’s first permanent link to the British colonies in the New World, although there were no attempts to take care of the Lutheran fold during the elder Francke’s lifetime, despite its significantly growing numbers after 1709.61 Only after the fate of the Salzburg refugees came to public notice in 1733, his son Gotthilf August (1696-1769) seriously began to consider a new transatlantic field of mission. After failed attempts to establish David Weisiger62 at the Salzburg community at Ebenezer, it finally fell to the

AFSt/M 4 C 17, Mühlenberg confirms October as the correct date. According to Glatfelter and G. H. G., they were both ordained October 25, 1770. Glatfelter, Pastors I, 93, 94; Dictionary of American Biography, s.v. “Mühlenberg, Frederick Augustus Conrad;” Wallace, Muhlenbergs, 78. 57 Dieser Mangel der Arbeiter ist die Ursache daß ich und mein Bruder eher von privat Studiis haben ablassen und an den öffentlichen Arbeiten teilnehmen müssen als sonst unser Wunsch war. Er stehet neu in Neu-York nicht ohne Beifall und Segen. Ich bediente eine Zeitlang die Filiale in Barrenhill, Cohenzy, Frankfurt und Pikestown und half meinem Vater in seinen wöchentlich Geschäfte, dann wurde ich Adiunctus meines Vater in Philadelphia und Raritan bis ich d. 6ten Jun. 1774 den ruf eines dritten Predigers hier völlig annahm. To Fabricius, 01/03/1775, AFSt/M 4 C 17 : 26. According to a letter to the Gesellschaft Naturforschender Freunde, Mühlenberg had accepted the position as third pastor at Philadelphia and was frequently commissioned to tend to temporarily vacant congregations until well into 1777: Nach- dem er von 1771 bis 1777 mehrentheils in Philadelphia als Prediger gestanden mußte er mit der Hälfte seiner Gemeinsglieder Philadelphia verlaßen, das von den Engländern besetzt worden. To GNF, 10/02/1799, HUB GNF S, Mühlenberg H. E. 58 In his letter to his friend Garliep at Halle in 1773, Mühlenberg writes that he was covering an average of 100 to 170 miles per month on horseback in Phladelphia’s hinterland, serving in 385 baptisms, 221 burials and 1,777 communicants for Philadelphia in 1772 alone. To Accise Ein- nehmer Garliep, Halle, 04/05/1773, AFSt/M 4 C 17. 59 Boehme himself took care of some 53 Palatinates emigrating via London under the guidance of Joshua Kochertal in 1708. In consequence of Boehme’s support and a small donation by Queen Anne to help the refugees, the following year saw 800 emigrants. At the end of this first German mass-migration to America, roughly 10–14,000 Palatinates took the route via London. Fourty years after the events, Germans in North America still testified to Heinrich Melchior Mühlen- berg how much they had valued Boehme’s support. Brunner, Halle Pietists, 58, 63–66. 60 Brecht, “Pietismus,” 524–526; Roeber, “Pietismus,” 683. 61 Glatfelter, Pastors II, 3. 62 Müller-Bahlke, “Communication,” 145. The Lutheran Context 55 former Orphanage teacher Johann Martin Boltzius (1703–1765) to give order and stability to the refugees’ church organization in the south of England’s American colonies.63 Alexander Pyrges has highlighted and analyzed the functional differen- tiation of communication in the Ebenezer network from 1732–1828. While Halle’s London contacts coordinated the actual transatlantic transfer based on their first hand experiences with the Navigation Acts and corollary regulations, Halle and Augsburg provided the marketing necessary to raise moneys in donations.64 On the American side, Charleston emerged as the Ebenezer network’s “postal center” and Halle’s primary point of contact in their North American trade and philanthropic mission until well into the 1760s.65 In Pennsylvania, the main area of German immigration during the 18th century, patterns of organized Lutheran church life first became visible after 1717, when Anthony Jacob Henkel (1668–1728), a native of Nassau, arrived in Pennsylvania and settled at Falckner’s Swamp.66 Henkel acted independently of European church authorities when he organized Pennsylvania’s first Lutheran congregation at New Hanover, and soon extended his activities to Providence, Goshenhoppen, German- town and Philadelphia, as Glatfelter assumes.67 After his death in 1728, German immigrants again depended largely on occasional services by Swedish pastors and the two John Casper Stoevers, father (1685–1739) and son (1707–1779), for their spiritual needs. In 1733, the united congregations of Philadelphia, New Hanover and Providence finally sent Christian Schultz and two laymen on a mission to Eu- rope to support and present the urgency of their cause.68 By then, and mostly as a consequence of the lack of support from German state authorities for emigrants, the Royal Chapel at London under Boehme and Ziegen- hagen gradually became the main communication agent for Germans abroad to their native countries.69 Especially after Boltzius’s success, Ziegenhagen began to establish ties with Lutherans in New York. It was through this channel that another desperate call for a minister by their Pennsylvania brethren was transmitted to him. Ziegenhagen and others forwarded this call to Gotthilf August Francke at Halle, with whom he stood in correspondence ever since his student days at Jena.70 Never- theless, it took Halle several more years before the imminent threat of Nikolaus Ludwig von Zinzendorf’s (1700–1760) plan to establish Moravian mission settle-

63 In England, the case of the Salzburg emigrants had come to the notice of James Oglethorpe (1696–1785) through Ziegenhagen, who stood in correspondence with Samuel Urlsperger at Augsburg. Brunner, Halle Pietists, 165. See also Roeber, “Pietismus,” 684; Müller-Bahlke, “Communication,” 146. Interestingly, Boltzius and the physician Ernst Thilo were obviously also the first ones to include natural observations and plants in their reports back to Halle. Wil- son, “Second Generation,” 236. 64 Pyrges, “Ebenezer Network,” 61–63. 65 Pyrges, “Ebenezer Network,” 64. 66 New Hanover. Glatfelter, Pastors II, 19. 67 Glatfelter, Pastors II, 21. 68 Glatfelter, Pastors II, 32. 69 Brunner, Halle Pietists, 67. 70 Brunner, Halle Pietists, 68; Glatfelter, Pastors II, 32. 56 A Prelude

ments in North America made Pennsylvania a top priority on their agenda.71 Shortly after his 30th birthday, Heinrich Melchior Mühlenberg accepted Francke’s call to Pennsylvania and went to London, where he should stay nine weeks with Ziegen- hagen to learn English and prepare for his new task.72 Heinrich Melchior Mühlenberg’s success in establishing a working church order in Pennsylvania can be seen as a masterpiece of organizational intelligence and demonstration of high awareness of the importance of networking. Upon the founding of the Lutheran ministerium in 1748, he not only managed to bring together all properly ordained ministers into one administrative central body, but also effectively curbed all other itinerant, independent and self-appointed Lutheran pastors’ claims to a legitimate practice in Pennsylvania by excluding them from this very body.73 On the personal side, his marriage with Anna Maria Weiser (1727–1802), the daughter of Indian interpreter and eminent Pennsylvanian political figure Con- rad Weiser (1696–1760) in 1745, opened him doors to the political establishment, which set an example for the next generations of Lutheran pastors to learn from.74 On the other side of the Atlantic, Mühlenberg’s reliable and exhaustive correspondence soon corroborated Halle’s trust in their new emissary and secured the fledgling new churches the necessary funding to unfold and expand its activities. Thus, after 1742 Pennsylvania slowly grew into the global Halle network through which growing amounts of Halle medicines and books found their way to the emerging Pietist project in Pennsylvania.75 In return, Francke and Ziegenhagen long claimed complete control over Mühlenberg’s colonial network.76

71 Splitter, Pastors, 18. 72 Ziegenhagen undoubtedly occupied a key position in Halle’s missionary network. From 1722 to 1776 literally every new missionary to North America and East India passed through Lon- don, which invested enormous influence in the court chaplain. Brunner writes that “the siginif- icance of the Lutheran ministers in London for the development of the infant German Lutheran Church in America has been underestimated. If, as has been suggested, Pietism was the domi- nant influence in the early years of the American Lutheran Church, then the contributions of ministers like Boehme and Ziegenhagen deserve greater appreciation. They became the key contact point of Lutherans in America and provided important pastoral care through their pro- curement of Bibles and devotional books, their written spiritual counsel and admonition, and their preparation of clergy en route to the new world.” Brunner, Halle Pietists, 68. See also Baglyos, “American Muhlenbergs,” 46. 73 Splitter, Pastors, People, Politics, 19; Baglyos, “Muhlenbergs,” 47f. 74 “Johann Dietrich Heinzelmann, one of Mühlenberg’s vicars, married into the Weiser family three years after his arrival in 1751; he died in 1756. Johann Helferich Schaum, who had come in 1747, married into the well-off Pickele family of Raritan, but his wife soon died. Justus Heinrich Christian Helmuth, a man of driving ambition and much talent who would succeed Mühlenberg as head of the Lutheran ministerium in the 1780s, married one of the Keppele daughters fourteen months after arriving in 1769. This marriage in particular ensured that the second generation of Halle clergy had a continued link to the tight web of Pennsylvania Ger- man and Dutch trading interests.” Wilson, Pious traders, 134. 75 Müller-Bahlke, “Communication,” 149. “Trade in books and medicines and the resulting network of transatlantic correspondence was an indispensable medium through which the Halle Pietists transmitted financial support to their North American diaspora and reinforced its evan- gelical mission.” Wilson, “Second Generation,” 255. 76 Francke and Ziegenhagen “held central positions in the ecclesiastical domain[;] they cared for The Lutheran Context 57

1.3 Pious Trade in Halle Medicines I (before 1770)

During the 1760s, Halle’s main focus in its North American trade gradually shifted from Ebenezer, Georgia, to Pennsylvania, where Mühlenberg and the ministerium worked hard to replicate not only Halle’s institutions and infrastructure, but also set up an equivalent internal communication system through which news, letters, medicines, books and materials circulated freely.77 This system not only enhanced domestic demand, as Halle products became available to an ever-growing number of potential consumers in Lutheran congregations, but also allowed the Orphanage agents to find access to local trade networks and even nourished inter-confessional trade, especially in medicines.78 It is difficult to pin down the exact number of Lu- therans in the Pennsylvania field upon Henry Mühlenberg’s arrival in 1770. In 1773, Mühlenberg wrote to his Vetter Bense at Einbeck that among this raw lot, God has also hid His own here and there, whose solace will be His word. There are 26 regularly ordained Lutheran pastors in this country, and especially Philadelphia has a strong and thriving community, with many that let their light shine!79 Glatfelter’s catalogue of Lutheran pastors lists a total of 63 regularly ordained Lu- theran pastors,80 with whom Mühlenberg could legimitately have communicated,

a network of mission projects around the world [and] they felt themselves responsible for each field of missionary labor.” Baglyos, “Muhlenbergs,” 49. 77 This complies with Alexander Pyrges’ periodization of the Ebenezer network, which basically ceased to exist during the 1770s. Pyrges, “Ebenezer Network,” 66; See also Wilson, “Second Generation,” 233. By 1720, the Halle Orphanage had an equal system at their disposal, which connected both distributors and consumers of medicines and books in Germany, central Europe and Russia and England. The same channels were used to purchase basic materials for the medical recipes, most of which had been composed by Christian Friedrich Richter (1676– 1711), his brother Christian Sigismund Richter (1676–1739) and later by David Samuel Madai (1709–1780). Wilson, Pious Traders, 69f., 78, 87, 92. 78 Wilson, Pious Traders, 4, 155. 79 Unter dieses rohen u[nd] wilden Haufens hat auch Gott hie u[nd] da die Seinigen verborgen, deren Trost sein Wort ist, wie deren auch 26 rechtmässige lutherische Prediger in hiesigen Lande seyn u[nd] insbesondere in Philadelphia eine grosse u[nd] stark blühende Gemeine ist, davon viele ihr Licht leuchten lassen. To Vetter Bense, 04/03/1773, AFSt/M 4 C 17. 80 These are: John George Bager (1725–1791); Peter Bentz (1791–1802); Nehemiah Bonham (1765–1846); Jacob van Buskirk (1739–1800); William Carpenter (1762–1833); John Casper Dill (1758–1824); George Frederick Ellissen (died ca. 1798); John Michael Enderlein (1726– 1800); John Frederick Ernst (1748–1805); Christian Espich (1766–1844); Jacob Frank (gave up licence around 1778); Theophilus Emanuel Frantz (ca. 1753; died after 1782); John Andrew Friderichs (ca. 1712; died after 1786); Frederick William Geissenhainer (1771–1838); John Siegfried Gerock (1724–1788); Jacob Goering (1755–1807); William Anton Graaf (1727– 1809); Philip Jacob Grotz (died ca. 1809); John Christopher Hartwick (1714–1796); Justus Henry Christian Helmuth (1745–1825); Benjamin Henkel (1765–1792); Paul Henkel (1754– 1825); Frederick William Jasinsky (died 1815); John Andrew Krug (1732–1796); John Chris- topher Kunze (1744–1807); John Daniel Kurtz (1764–1856); John Nicholas Kurtz (1720– 1794); John William Kurtz (1732–1799); Frederick William Lange (Long) (died ca. 1812); John Daniel Lehman (1754–1810); John Christian Leps (died after 1787); Anthony Ulrich Luetge (1754–1796); John Samuel Mau (1752; died ca. 1830); Frederick Valentine Melsheimer (1749–1814); August Frederick Meyer (1766–1832); Henry Moeller (1750–1829); Frederick Augustus Conrad Mühlenberg (1750–1801); Henry Melchior Mühlenberg (1711–1787); John 58 A Prelude

based on common professional standing. At the same time, however, the existence of contemporary interconfessional dialogue among Lutherans and Reformed clergy is a well–known fact today. Regular communication among the majority of Luthe- ran pastors largely took place during annual meetings of the ministerium, while personal correspondence usually remained confined to a few colleagues. Trade in medicines linked suppliers, distributors and consumers in a transatlantic business. According to Renate Wilson, about 14 % of the 250 ministers listed in Glatfelter’s work had some medical training and access to medications, and therefore took part in this exchange throughout the 18th century.81 When Henry, Fre- derick and Kunze returned to Pennsylvania in 1770, products from the Medikamen- tenexpedition had already become a household name82 both in the Old and the New World, and served Lutheran pastors as an important additional source of income to augment their often meagre ministerial stipends.83 In the 1770s, Halle’s Pennsylva- nia market niche gained in importance for the mother institution, as conditions and opportunities for safe investment in Europe deteriorated and the economic and spiritual development of the Foundations showed first signs of stagnation.84 The decades after August Herman Francke’s death in 1769 therefore saw a complete over- haul of Halle’s transatlantic trading channels to North America.85 Wherever Halle was active in missionary and commercial activities, trade in medicines, Bibles, books, tracts and other products relied on a system of trusted intermediaries, brokers and traders that provided for the safe and quick transportation of commodities, and the professional handling of all associated business, such as insurances, financial transactions and customs regulations.86 Surely, neither transatlantic trade itself nor this manner of organization was an original Halle

Peter Gabriel Mühlenberg (1746–1807); George Henry Pfeiffer (1747–1827); John Frederick Ries (died 1791); Frederick Augustus Rudolphus Benedictus Ritz (died 1811); Conrad Roeller (died 1795); John F. Ruthrauff (1764–1837); Frederick David Schaeffer (1760–1836); John Helferich Schaum (1721–1778); John Frederick Schmidt (1746–1812); John George Schmucker (1771–1854); John Daniel Schroeter (died after 1817); Christopher Emanuel Schultze (1740– 1809); John Schwarbach (1719–1800); Adolph Spindler (1743–1828); John Stauch (1762– 1845); John Michael Steck (1755–1830); John Casper Stoever, the younger (1707–1779); Christian Streit (1749–1812); John Lewis Voigt (1731–1802); John Frederick Weinland (1744– 1807); George Joseph Wichterman (died after 1817); John Guenther Wiegandt (Weagant) (ca. 1762–1835); Charles Frederick Wildbahn (1733–1804); John Conrad Yeager (Jaeger) (1768– 1832); John David Young (Jung). The list does neither include pastors not approved by the ministerium and Halle, like Adam Henry Meyer, Balthasar Meyer or Peter Mischler, nor pastors deemed unfit for the service on account of sickness, debts etc. like John George Butler. Glatfelter, Pastors I, passim. 81 Wilson, Pious Traders, 122–126. 82 Medications from the Orphanage are one of the earliest known cases of active branding to en- hance brand recognition and consumer loyalty. Concoctions like Essentia dulcis came with an enclosed product sheet and a uniform packaging and design, which was also supposed to prevent brand piracy. Wilson, Pious Traders, 71. 83 Wilson, Pious Traders, 4. 84 Wilson, Pious Traders, 143, 156. 85 Wilson, Pious Traders, 131. 86 “Both domestic and foreign trade operated through associated apothecaries and commercial houses as well as private agents, university professors, Pietist clergy, and Lutheran missionaries The Lutheran Context 59 invention,87 but the elder Francke and his successors had been particularly inge- nious in implementing and sustaining a system that included both their Lutheran brethren as recipients and intermediaries, and private intermediaries, such as book traders, professional trading agents and other merchants.88 The latter usually resi- ded in key locations, especially commercial centers like Frankfurt, Leipzig and London, or in seaports like Amsterdam, Rotterdam, Hamburg or Altona, all of which were located on transatlantic sailing routes.89 When regular trade with Pennsylvania started in the 1750s, Halle’s first constant North American correspondents primarily charged with financial affairs were Peter Brunnholtz90 († 1757) and John Diedrich Matthias Heintzelman91 (1726– 1756), whose responsibilities soon comprised every aspect of the commerce from trade matters in medicines and books to the stewardship of legacies and bequests.92 Their respective deaths in 1756 and 1757 left Halle temporarily devoid of able American contact persons sufficiently familiar with the routines and workings of their business. At about the same time, new conflicts in Europe flared up and soon erupted into a war, which should last seven years in Europe was well as in North America. As a consequence, transatlantic commerce came to a near halt during these years, and it was probably this first experience of a long period without any support or advice from Europe that gave the elder Mühlenberg the idea to send his sons to Halle to expose them to training, knowledge and learning which would support their cause in times of interrupted contact. Impediments caused by war and armed conflict on both sides of the Atlantic remained a central theme in the development of the Halle trade and frequently appear in Henry Mühlenberg’s correspondences with European and American contacts.93

in the Baltic territories, Russia, and India.” Wilson, Pious Traders, 71. See also Müller-Bahlke, “Communication,” 145. 87 See for instance Rosalind Beiler’s essay on the transatlantic trade between Caspar Wistar and Georg Friedrich Hölzer. Beiler holds that “[l]inking the two men were trustworthy brokers, who used their own connections to carry out the partner’s transactions. Thus the particular dangers of trade between continental Europe and the British colonies directly shaped transatlantic commercial channels and required a high degree of reciprocity and trust for success.” Beiler, “Cas- par Wistar,” 172. 88 In the case of Caspar Wistar, Rosalind Beiler speaks of “a constantly changing group of agents with connections to other transatlantic networks (...). The fluid nature of the brokers carrying out his trade added a final set of risks to Wistar’s transatlantic networks.” Beiler, “Caspar Wis- tar,” 172. 89 Wilson, Pious Traders, 78. 90 Brunnholtz was the first properly ordained Lutheran minister sent by Halle after the elder Mühlenberg. Glatfelter, Pastors II, 23. 91 Wilson, Pious Traders, 78, 141. 92 Wilson, Pious Traders, 141. 93 Complaints and notices about war and its costs can be found throughout the roughly 40 years of Mühlenberg’s surviving correspondences, which saw the American War of Independence, the European coalition wars in the wake of the French Revolution, the Quasi War of 1798– 1800, the Barbary War of 1801–1805, Napoleon’s Continental System and the War of 1812: In 1792, Mühlenberg wrote to Schreber: Wenn doch durch den unglücklichen Krieg in Europa unser Briefwechsel nicht unterbrochen wird! Von Holland und nach Holland gehen fast immer Schiffe und Briefe können schnell gehen. Wir haben in den hintern Gegend auch einen kleinen 60 A Prelude

In the wake of the Seven Years’ War, Halle found its transatlantic network se- verely eroded, and without a financial trustee on the other side of the Atlantic, trade picked up again only at a very moderate pace. On August 12, 1766, Gotthilf August Francke even informed Heinrich Melchior Mühlenberg that large revenues from their North American trade were not to be expected and for that reason, no “special person” would be necessary to take care of business correspondence. In case that such a person would be needed in the future, Francke added that he considered Mühlenberg’s own son Peter no adequate choice for such a responsible position.94 Presumably, Francke wrote this in response to an earlier suggestion by Mühlenberg himself. During the following years, Heinrich Melchior Mühlenberg and his son-in- law Christopher Emanuel Schultze (1740–1809), who had married his daughter Eva Elizabeth in 1765,95 took care of financial matters, e.g. the handling of John Frederick Handschuh’s bequest after his death in 1764.96 In the spring of 1769, a letter from the elder Mühlenberg to Pasche in London states that Schultze was by now fully responsible for all cases of transatlantic financial stewardship.97 Appa- rently, Mühlenberg was already considering ways how to relieve his burden to some degree,98 and the fact that Schultze badly needed relief also confirms Renate Wilson’s assertion that Halle’s niche in Pennsylvania’s medical and print market grew in importance around 1770.99

Krieg mit den Indiandern oder einem Theil derselben. To Schreber, 09/20/1792, UAE Brief- nachlass Schreber. 94 [O]bs 2. Was den Arzney= und Bücherhandel betrifft; so ist davon nicht so viel Profit zu er- warten, daß eine besondere Person um desselben willen unter halten werden und für die Anstalt viel übrig bleiben könte. Am wenigsten würde das eine Sache für den Sohn des H. P: Mühlen- bergs seyn, zu dem man auch nach seinem bisherigen Betragen das Vertrauen nicht faßen könnte; wie denn überhaupt zu besorgen ist, daß dabey entweder hiesige Anstalten, oder die dortige Cassa nicht ohne Schaden bleiben dürfte. G. A. Francke to H. M. Mühlenberg, 08/12/1766, Aland, Mühlenberg Korrespondenz III, (letter 380). 95 Glatfelter, Pastors I, 125. 96 For Handschuh’s bequest, see especially H. M. Mühlenberg to F. W. Pasche, 11/07/1767, Aland, Mühlenberg Korrespondenz III, (letter 413). 97 Dahero nahm die noch übrige 20 £ sterl. von der Arzeney, nemlich an hiesiger Curr: 30 £ und den Wechsel von 100 £ sterl. der an Curr[ency] pr[o] Qentum] 165 £ trug, und 14 £ 17 sh[illing] 4 d. [pence] aus der alten Brunnh[olz] und Handsch[uh] Bücher Cassa welche H. Pfr: Schultz in Verwaltung hat, nahm, sage ich, die 209 £ 17 sh: 4 d. Curr: und legte von dem Meinigen so viel zu, als die äuserste Noth erfoderte, und die bey gelegte Rechnung zeiget; H. M. Mühlenberg to G. A. Francke and F. M. Ziegenhagen, 04/15/1769, Aland, Mühlenberg Korrespondenz IV, (letter 463). 98 One of the elder Mühlenberg’s ideas was to install Helmuth at Philadelphia, whom he considered most apt to aid Schultze: H[err] Helmuth wäre meines geringen Erachtens noch der einzige Mann, der sich wegen seines lebhaften Temperaments, Stimme und übrigen Gaben für den zweyten Prediger und Seelsorger in Philadelphia schicken möchte; alsdenn wäre aber der großen Gemeine in Lancaster übel gerathen, und sie würden ihn gewiß nicht mit guten Willen loslaßen. Ich wüßte aber nicht, wie H[err] Pfarr[er] Schultz nur allein die Zeit und Kräfte er- fordernde Correspondence bestreiten könnte! H. M. Mühlenberg to F. W. Pasche, 08/12/1769, Aland, Mühlenberg Korrespondenz IV, (letter 479). 99 Wilson, Pious Traders, 143, 156.