Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations (Pioneers in Arts, Humanities, Science, Engineering, Practice, 29) 3030789063, 9783030789060

Citation preview

Pioneers in Arts, Humanities, Science, Engineering, Practice 29

Harvey Starr

Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations

Pioneers in Arts, Humanities, Science, Engineering, Practice Volume 29

Series Editor Hans Günter Brauch, Peace Research and European Security Studies (AFES-PRESS), Mosbach, Germany

More information about this series at http://www.springer.com/series/15230 http://www.afes-press-books.de/html/PAHSEP.htm http://www.afes-press-books.de/html/PAHSEP_Starr.htm

Harvey Starr

Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations

123

Harvey Starr Department of Political Science University of South Carolina Columbia, SC, USA

Acknowledgement Book cover photo by Political Science Department, University of South Carolina, with permission. More on Prof. Harvey Starr, his books, other publications and links to selected media are found at: http://www.afes-press-books.de/html/PAHSEP_Starr.htm.

ISSN 2509-5579 ISSN 2509-5587 (electronic) Pioneers in Arts, Humanities, Science, Engineering, Practice ISBN 978-3-030-78906-0 ISBN 978-3-030-78907-7 (eBook) https://doi.org/10.1007/978-3-030-78907-7 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyediting: PD Dr Hans Günter Brauch, AFES-PRESS e.V., Mosbach, Germany This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Harvey Starr and wife Dianne Luce celebrating their 23rd wedding anniversary in July 2017 at the Grove Park Inn, in Asheville, North Carolina. Source From the author’s personal photo collection

To Dianne with whom all trajectories are positive

Preface and Acknowledgements

This essay was motivated, in part, by the author’s interest in how the concept of “intellectual trajectories” could be incorporated into writing one’s intellectual autobiography, thereby exploring how the personal lives of individual academics intersected with their professional interests. It is hoped that this essay—following my personal story and the development of my research agenda and corpus of work —will provide insights valuable to scholars in terms of what it means to be an academic, and how to approach their research, research design, and the logic of inquiry. Hopefully, this essay might also prod senior scholars to think about their own “political memories” and how they have affected their scholarship and intellectual growth. In addition, it is also to be hoped that graduate students and junior faculty should find useful hints about how to navigate their way through the complexities of becoming a professional academic as well as a successful, productive scholar. And, while this essay reflects a personal trajectory, but it may also move the reader to think about the trajectory of International Relations as a whole. This Preface will briefly introduce the motivation and impetus for the long essay in Chap. 1. That essay is much more personal, with a more in-depth autobiography than is usual in the Springer series. As noted, I do try to make explicit connections between personal and educational experiences and development to my later work as a scholar—both in terms of my process and subjects of study. Early reactions to drafts of that essay indicate I may very well be successful in stimulating other scholars to look at their own “trajectories” and “political memories.” In this Preface I will only provide a brief statement of Acknowledgments, rather than a separate statement as found in the other volumes. A number of sections in Chap. 1 acknowledge the support and assistance of other individuals along my own personal journey, as well as substantial space devoted to my mentor Bruce Russett, and explicit thanks to many others toward the conclusion of Chap. 1. The motivation for writing Chap. 1’s essay arrived serendipitously. In the academic year prior to retiring at the University of South Carolina, I began the process of finding out how to deal with the bureaucratic maze that USC presented to any potential faculty retiree. I had heard how USC figuratively cut all the apron strings the day after retirement became official. As I was still chairing several Ph.D. ix

x

Preface and Acknowledgements

dissertation committees, and had a number of research and publication projects to bring to fruition I wanted to ensure that I would have such things as email, parking, and library privileges. My first encounters with the paperwork involved in retirement, led me to have a conversation with the Provost. He and I agreed that retired faculty represented a major source of untapped talent and resources for a university and I volunteered to undertake a study initially about making it easier for faculty to both retire and retain various links to the university. To do so I wanted to see what sorts of policies, arrangements, programs, and facilities that other institutions had developed to take advantage of the resources provided by retiring faculty, and to help them maintain positive linkages to the institution—positive for both individual retirees and the university. The Provost was most enthusiastic about my proposal. However, shortly after my June 30, 2014 retirement, he announced that he would be moving on to the presidency of a major Midwest institution. His agenda was now focused elsewhere. Without his support, along with my new interests in looking at the synergies between political science and linguistics, the retirement project was put on hold and then abandoned. Before the retirement project was dropped, however, I had begun to contact a number of institutions concerning the role retired faculty played on their campuses. My central mentor across my whole career starting in graduate school—Bruce Russett—had recently retired from Yale after 49 years. He put me on to Yale’s Henry Koerner Center for Emeritus Faculty. I was fascinated by one of the Center’s major activities, to have a series of talks by emeritus faculty on “intellectual trajectories” which would “explore how the personal lives of individual professors intersected with their professional interests…”1 Written versions of these talks, across a broad range of disciplines have, at present, been published in three volumes by The Koerner Center. I found these essays engrossing, stimulating, and revealing. And, while I’m pretty sure the essay to follow pales in comparison to most of those in the three Koerner volumes (and is much, much longer!), I thought, “why not take a shot at linking the personal with an intellectual autobiography which tries to provide a coherent account of ideas and scholarship and my research process across a career of almost 45 years?” 2}I also thought that using the term “trajectory” was perfect— a path, on one hand unbroken, but also characterized by “jumps and breaks,” spikes and drops representing the serendipity that appears in most of the essays in the Koerner volumes. 3 Perhaps, the best way to introduce the essay to follow is to note David E. Apter “Introduction,” in Intellectual Trajectories, Vol. I (The Koerner Center, Yale University, 2009), p. 9. 2 Another factor influencing my motivation to write such an essay was that I wrote the intellectual biography for Bruce Russett, which appeared in the Springer volume on Russett which I also edited (see Starr 2015a). 3 See Apter discussing the essay of English and American Studies scholar Alan Trachtenberg, “Introduction,” p. 15. See also Trachtenberg’s essay, “A Backward Glance,” in Koerner Center, Vol. 1. 1

Preface and Acknowledgements

xi

that it includes the personal, reflects on trajectories, and is simply “An Addendum to a Curriculum Vitae,” the title of Art Historian Walter Cahn’s essay in Koerner, Vol. 1. I do want to acknowledge specifically the assistance of three people who read all or parts of earlier versions of Chap. 1. Zaryab Iqbal, both a departmental colleague and co-author, Marc V. Simon, a former doctoral student of mine as well as co-author, and my brother, Mark Starr were of invaluable help in their reading, their reactions, and their encouragement. And, as I have said once before, the assistance and guidance of Hans Günter Brauch, the editor of the Pioneers in Arts, Humanities, Science, Engineering, Practice series has been equally invaluable. I am always amazed by his energy, his attention to detail, and willingness to help in every aspect of preparing this book for publication. A special thanks to you all. Columbia, SC, USA March 2021

Harvey Starr

Contents

Part I

On Harvey Starr

1 Getting from Then to Now: A Personal Intellectual Autobiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Some Early Autobiographical Context . . . . . . . . . . . . . . . . . . 1.2 The Beginnings of an Intellectual Autobiography . . . . . . . . . . 1.3 The Foundations for The Intellectual Journey: SUNY Buffalo . 1.4 Where the Rubber Meets the Road: Yale . . . . . . . . . . . . . . . . 1.4.1 Yale: Multidisciplinarity, Interdisciplinarity, and Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.2 Bruce Russett: Initial Interactions and Mentoring . . . . . 1.4.3 The Dissertation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Scotland and After . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6 Scholarship: Career Themes . . . . . . . . . . . . . . . . . . . . . . . . . 1.6.1 Conflict: Consequences and Processes . . . . . . . . . . . . . 1.6.2 Opportunity and Willingness as Central to a Scholarly Agenda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6.3 O/W and My Major Scholarly Projects . . . . . . . . . . . . 1.6.4 An Overview of My Research Process . . . . . . . . . . . . 1.7 Concluding Thoughts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7.1 “I Get By With a Little Help From My Friends” . . . . . 1.7.2 Back to Intellectual Autobiography and Trajectories . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 A Selected Bibliography of the Publications of 2.1 Books . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Journal Articles . . . . . . . . . . . . . . . . . . . 2.3 Book Chapters . . . . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

5 5 11 13 19

. . . . . .

. . . . . .

21 23 29 30 35 35

. . . . . . .

. . . . . . .

40 43 58 59 59 62 63

Harvey Starr . . . . . .

69 69 72 78

................. ................. .................

xiii

xiv

Part II

Contents

Texts by Harvey Starr

3 Cumulation from Proper Specification: Theory, Logic, Research Design, and ‘Nice’ Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Applying the Research Triad: Theory, Nice Laws, and Substitutability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Illustrations with Ray’s Guidelines . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

...

83

... ... ...

86 91 94

4 Opportunity, Willingness and the Diffusion of War, 1816–1965 4.1 Opportunity and Willingness in the Diffusion of War . . . . . 4.2 Research Design and Data Generation . . . . . . . . . . . . . . . . 4.3 Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Do Opportunity and Willingness Influence War Diffusion? . 4.5 How Much Do the Variables Contribute to War Diffusion? . 4.6 A Problem? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

97 99 102 107 107 110 117 118 119

5 Democratic Dominoes: Diffusion Approaches to the Spread of Democracy in the International System . . . . . . . . . . . . . . 5.1 Democracy, Dominoes, and Diffusion . . . . . . . . . . . . . . 5.2 Governmental Transitions . . . . . . . . . . . . . . . . . . . . . . . 5.3 Diffusion Approaches to Governmental Changes . . . . . . 5.4 Neighbor Effects and the Interaction Opportunity Model . 5.5 Data and Descriptives . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6 Analysis and Design: Global Effects . . . . . . . . . . . . . . . 5.6.1 Poisson Analyses . . . . . . . . . . . . . . . . . . . . . . . . 5.7 Global Trends Toward Democracy . . . . . . . . . . . . . . . . 5.8 Regional Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9 Bordering Governmental Transitions: Neighbor Effects . . 5.10 Neighbor Effects: A First Cut . . . . . . . . . . . . . . . . . . . . 5.11 Neighbor Effects: A Second Cut . . . . . . . . . . . . . . . . . . 5.12 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

123 123 125 126 128 129 131 131 132 135 137 139 140 142 144

6 On Geopolitics: Spaces and Places . . . . . . . . 6.1 Context: Time, Space, and Determinism . 6.2 Context: Space and Spatiality . . . . . . . . 6.3 Relative Spaces . . . . . . . . . . . . . . . . . . 6.4 To Conclude . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

147 148 153 155 159 160

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

Contents

7 Opportunity, Willingness and Geographic Information Systems: Reconceptualizing Borders in International Relations . . . . . . . . . 7.1 Geopolitics, Borders and IR: A Brief Introduction . . . . . . . . 7.2 Conceptualizing (and Re-conceptualizing) Borders . . . . . . . . 7.3 Methodology: ARC/INFO and the Conceptualization— Operationalization of Borders . . . . . . . . . . . . . . . . . . . . . . . 7.3.1 Opportunity for Interaction (Ease of Interaction) . . . . 7.3.2 Salience (Willingness) . . . . . . . . . . . . . . . . . . . . . . . 7.3.3 A New Dataset on the ‘Nature’ of Borders . . . . . . . . 7.4 The Scientific Utility of a GIS-Generated Border Data Base . 7.4.1 Revisiting IR Hypotheses with More Fully Specified Borders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.2 Revisiting Borders as Part of the War Puzzle . . . . . . 7.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xv

. . . 163 . . . 163 . . . 164 . . . . .

. . . . .

. . . . .

166 168 171 174 175

. . . .

. . . .

. . . .

176 177 179 180

8 Democracy and Integration: Why Democracies Don’t Fight Each Other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1 What Is to Be Explained? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 The Confusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3 Integration: Security Communities and the Democratic Peace . . . . 8.4 What Does Democracy Mean? . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.1 Democracy and Transparency . . . . . . . . . . . . . . . . . . . . . 8.4.2 Lesser Forms of Conflict . . . . . . . . . . . . . . . . . . . . . . . . 8.5 Conclusion: Democracy and Integration . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

183 183 186 187 189 189 192 193 196

9 The Kissinger Years: Studying Individuals and Foreign Policy . . 9.1 Kissinger, His Memoirs, and the Kissinger Industry . . . . . . . . 9.2 The Problem of Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3 Psychological Approaches to the Study of Kissinger . . . . . . . . 9.4 Memoirs, Access, and Confidence: Some Specific Areas of Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.5 Kissinger and Richard M. Nixon: Nixon the Individual/Nixon the Statesman . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.6 Kissinger, Nixon, and the ‘Obstacle’ of the Bureaucracy . . . . . 9.7 Dealing with the Soviet Union . . . . . . . . . . . . . . . . . . . . . . . 9.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

. . . .

199 200 201 203

. . 206 . . . . .

. . . . .

208 213 216 221 222

About the University of South Carolina . . . . . . . . . . . . . . . . . . . . . . . . . . 225 About the Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

Part I

On Harvey Starr

2

Part I: On Harvey Starr

Harvey and Zaryab Iqbal at the 2016 meeting of the International Studies Association, holding their newly published book on state failure from the Stanford University Press. Source From the author’s personal photo collection

Harvey at 38, in his first year as Chair of Political Science at Indiana University. Source From the author’s personal photo collection

Chapter 1

Getting from Then to Now: A Personal Intellectual Autobiography

1.1

Some Early Autobiographical Context

Those readers familiar with my scholarship will not be surprised that I begin with some personal context. From my work based within, and emerging from, the opportunity and willingness theoretical framework, through the foundations of the Most and Starr logic-of-inquiry research program, and my writings on geography and spatiality, context has been a consistent theme. As any grad student who took classes from me can tell you, the Most and Starr goal of uncovering what theories/ hypotheses/frameworks will work under what conditions was central to the study of the relevance and applicability of any theoretical formulation. So, picking up from the Preface, let’s start from the beginning. I was born on November 11, 1946 at Columbia Presbyterian Hospital in Manhattan and lived in Washington Heights, a block off Broadway, just a few blocks south of the George Washington Bridge. My father’s parents had arrived in New York from the Odessa area in Russia around the time of the Russo-Japanese war; as a Jew having miraculously survived the Tsar’s army earlier, my grandfather left before they could conscript him again. My mother’s mother came from Poland as a child. My mother’s father, an Austrian Jew, had actually been born in the United States. I was four and a half when we – father Nat, mother Betty, and older brother Mark – moved to Glen Oaks, Queens, only a mile or so from the Nassau County border. Although those who did not grow up in the New York metropolitan area would consider me a “city boy,” Glen Oaks was a very large garden apartment complex in a distant suburb composed of such apartments and single family houses (with the only high rise building in the entire area being Creedmoor hospital, now called the Creedmoor Psychiatric Center). To us, “the city” meant Manhattan, and it took us well over an hour to get there by bus and then subway. However, I am certainly a child of the urban rhythms that reached even across the width of Queens (well over 20 miles). Although I essentially left Queens at 16, when I went off to SUNY Buffalo (to return only in the summers during college), I still feel the strong pull of “the city” anytime I hear the © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7_1

5

6

1 Getting from Then to Now: A Personal Intellectual Autobiography

saxophones at the opening of Saturday Night Live while the black and white montage of Manhattan flashes on the screen. Any rendition of “New York, New York,” or Billy Joel performing “New York State of Mind” has the potential to bring a tear to my eye. Like most of those reading this, as soon as I learned to read I did so voraciously, always reading at advanced grade levels.1 I was particularly interested in mythology, ancient history, figures such as Alexander the Great, Hannibal, Genghis Khan, and Tamerlane. And then there were the encyclopedias! I discovered in graduate school and throughout my academic career that my very early habit of pulling down a volume of the encyclopedia and going through it page by page, reading those entries that caught my interest, was a rather common habit of other academics. When I first learned to read, our family owned a set of The Book of Knowledge, later replaced by the much more extensive and advanced World Book. As I grew up I must have gone through each volume of both encyclopedias at least a dozen times, each time finding entries that were newly of interest. I have never checked, but I am pretty sure having some form of encyclopedia in the house characterized many lower-middle class families in the area (or perhaps especially Jewish ones!).

Harvey at 14 (lower left) after sophomore year in high school; with mother Betty, father Nathan, and brother Mark. Source From the author’s personal photo collection 1

While I read as early as possible, my parents told me that I did not begin to speak until I was almost three. They also said that I hadn’t stopped talking since—as most of you can attest.

1.1 Some Early Autobiographical Context

7

I attended P.S. 186 from first to sixth grade, living only a few blocks away, and thus able to walk back and forth, including for lunch at home. Many of my friends and classmates from P.S.186 went off to Junior High School 172 with me, both newly constructed buildings. Most of us then migrated to the relatively new (opening in 1955) Martin Van Buren High School, which required taking a New York city bus. Some even followed high school graduation in 1963 by attending SUNY Buffalo as I did (which became part of the New York State system in 1962). A number of those P.S. 186 kids remain friends today (thanks to Facebook). Box 1.1 Harvey Starr: Curriculum Vitae Dag Hammarskjöld Professor in International Affairs Emeritus, University of South Carolina Born: November 11, 1946, New York, New York Married, no children Degrees: B.A., Political Science (with Highest Distinction), Summa Cum Laude, State University of New York at Buffalo, 1967 M.Phil., and Ph.D. in Political Science, Yale University, 1970, 1971 Positions: Dag Hammarskjöld Professor in International Affairs Emeritus, University of South Carolina Emeritus Scholar, Jewish Studies Program, University of South Carolina Dag Hammarskjöld Professor in International Affairs, University of South Carolina, 1989–2014 Chair, Department of Government and International Studies/Political Science, University of South Carolina, 2000–06 Interim Chair, Department of Government and International Studies, University of South Carolina, 1998–2000 Professor and Chair, Department of Political Science, Indiana University, 1984–89 Professor of Political Science, Indiana University, 1983–89 Associate Professor of Political Science, Indiana University, 1977–83 Senior Fellow, Center for International Policy Studies, Indiana University, 1975–80 Assistant Professor of Political Science, Indiana University, 1972–77 Acting Instructor, Yale University, Department of Political Science, 1970– 71 Research Assistant, S.U.N.Y. at Buffalo, Department of Political Science, Summer 1968

8

1 Getting from Then to Now: A Personal Intellectual Autobiography

Harvey in July 1995 at the Imperial Palace in Beijing, as part of the Program for International Studies in Asia (PISA)/People’s University 1995 Summer Workshop on Special Topics in International Studies. Source From the author’s personal photo collection

Visiting Positions: Visiting Fellow, Department of International Relations, Research School of Pacific and Asian Studies, The Australian National University, Spring and Summer 1996 Visiting Researcher, Centre for Defence Studies, University of Aberdeen, Scotland, July–August, 1990. Visiting Lecturer, Centre for Defence Studies, University of Aberdeen, Scotland, July 1985 Leverhulme Visiting Fellow in Politics, University of Aberdeen, Scotland, 1978–79 Visiting Fellow in Politics, Department of Politics, University of Aberdeen, Scotland, 1971–72.

1.1 Some Early Autobiographical Context

9

Harvey at his 2014 retirement party from the Department of Political Science, University of South Carolina. With him is Don Fowler, long time teacher and friend of the Department; a Democratic Party activist who led the South Carolina State Party from 1971 to 1980, also served as National Chairman of the Democratic National Committee from 1995 to 1997. Source From the author’s personal photo collection.

This seems to be a good place for an important aside.2 As noted I edited the Springer “Briefs on Pioneers in Science and Practice” volume on Bruce Russett (Starr 2015a). This included my writing a “Preface” along with the “Scholarly Biography” which preceded the six articles selected to highlight and represent the major strands of Russett’s work. These two pieces included a number of biographical details, including Russett’s recollection that his “earliest political memories were about war and peace” (p. ix, emphasis added). These prominent early memories included the attack on Pearl Harbor, air raid drills, and the dropping of the atomic bomb on Hiroshima. When I wrote these two sections for the Russett volume I was not sure I had such prominent early memories. But in my present guise as a memoirist, I find I do. Being born in 1946 my first political memories involve the Korean War, particularly the events of December 1950–January 1951 when Chinese forces crossed into North Korea and drove south to recapture Seoul. I recall being greatly dismayed at the headlines about American 2

Again, people who know me understand I speak in an annoying combination of parentheses, footnotes, and perhaps even Ptolemaic epicycles…

10

1 Getting from Then to Now: A Personal Intellectual Autobiography

defeats and American forces being driven back down the peninsula. For the first time, the why, what, and how questions about war (later to encompass all social conflict) became part of my thinking. A recent email exchange with my college roommate and long-time friend – since 1964 – made me bring back another memory, one mirroring those of Russett: the Cold War of the 1950’s and the threat of nuclear war. As with all my contemporaries, I remember the “duck and cover” drills going under our classroom desks; or moving silently and quickly down into the basement of P.S. 186. But I also wrote this in an email to my former roommate: “Do you remember a TV anthology series called ‘Medic’ which starred Richard Boone? There was one episode in 1955, ‘A Flash of Darkness’ which simulated/showed the possible consequences of a nuclear war with the USSR through the lens of a hospital’s ER room. This predated missiles, so the central focus was on the fleet of intercontinental bombers on each side. The part of Queens we lived in was beneath the flight paths to both LaGuardia and Idlewild (later JFK) airports, and for weeks after the show aired, after going to bed, my nine year-old self would listen to the planes overhead and wonder which one might be dropping an atomic bomb. The fact that I still clearly remember such details demonstrates the impact it had on me.” In retrospect, I think this impact found its way into much of my work on conflict, conflict processes, and my initial work on the consequences of war. As suggested by an early reader of this essay, I wonder how many political scientists have been motivated by such childhood political memories. Returning to my chronology, I did very well in elementary school, but indeed, so did most of my friends and classmates. This is relevant to one of the crucial decisions my parents made as sixth grade was coming to a close. New York City at that time had the option that allowed students completing elementary school to go into the Special Progress (SP) program, which permitted students to complete the three years of junior high school in two years. As a November baby I was already somewhat younger than many of my classmates, as well as being rather small (both in height and weight).3 I had qualified to go into an SP class. My parents had qualms about my basically skipping a grade. But as many of my classmates and friends would also be in SP classes, my parents agreed, and I spent the two years of junior high in one of the five SP classes (with students from several other elementary feeder schools). Again, I did well, especially in social studies courses and English. The home room teacher for SP5 – my class – for our two years was also the SP English teacher, Mr. McCollum. Most of our teachers were quite good. A large group of the students across the five SP classes were good friends attending numerous birthday parties as well as bar-mitzvahs and bat-mitzvas, and generally having a good time.

3

Indeed, when I entered Martin Van Buren high school in the Fall of 1960, I am pretty sure I was the second shortest male student out of the 5000 students who attended that school using three different time schedules.

1.1 Some Early Autobiographical Context

11

I was a good student, not one of the top students, but still read a great deal, and began to spend more time drawing. Drawing was a significant activity for me up through college. Mr. McCollum liked to have posters for the classroom illustrating scenes from novels we were reading as well as his extensive list of rules for good grammar which he drilled into us (along with the endless diagramming of sentences!). I did many (most?) of those posters. I was also on the art staff of the small soft cover junior high ‘yearbook’ the Hillside Echo, when we graduated in 1960. In high school I was on the art staff of the student newspaper The Beeline and took over the “Marty Van Buren” comic that appeared in each issue. Marty was quite stylized, and every person who did the comic drew him the same way, but I’m rather struck by the range of topics and situations into which I put him. I also did a great deal of the artwork as a member of the art staff for the high school yearbook, the 1963 Futura. I have to note that the person who also drew many of the posters for Mr. McCollum’s classroom, who sat directly behind me for two years in junior high home room, and who was an important part of the Futura yearbook art staff as well, was by far the better artist, and turned out to be an even better photographer— Robert Mapplethorpe. My last drawing gig was at SUNY Buffalo. In my sophomore year, I drew the sports cartoons for the student newspaper The Spectrum. Indeed, in high school I had given much thought to becoming a commercial artist. I took courses in commercial art from my senior year homeroom teacher, head of the Art Department, Mr. Cascio. I did well enough, but after seeing the quality of the work of the other students, I realized that I did not quite have the talent to make this a career. Similarly, I also had a strong interest in architecture and thought about that as a career. However, perhaps my weakest courses in high school were math classes, and I realized that would be a significant drawback to succeed in that field. My interest in architecture might not surprise those who have been in my office or home, or who have seen the photographs I have taken over the years on trips or during my extended time in Aberdeen, Scotland (especially my sabbatical year in 1978–79). My walls at home are covered in lithographs and photos of buildings of all sorts. My own photos of Scotland and England focus on buildings and ruins of buildings. In Britain I would stop at book stores or print shops and find old architectural drawings. One of my favorite framed posters at home is from an exhibition of the works of Andrea Palladio.

1.2

The Beginnings of an Intellectual Autobiography

Commercial art and architecture were roads not taken, and I knew that by the time I graduated high school. But high school also laid the groundwork for the unexpected success I had at Buffalo, which in turn enabled me to attend Yale for my Ph.D. in Political Science.

12

1 Getting from Then to Now: A Personal Intellectual Autobiography

High school was easy for me. I have no memories of scrambling to get assignments done, or of not turning an assignment in on time, or staying up late to do homework that had been postponed. My English classes were uninspiring, but I spent those years reading across the basic American cannon on my own—from all of Poe, to all of Hemingway, some Fitzgerald, some Jack London, etc. I also began my reading of science fiction and detective novels in earnest. But the bulk of my reading was non-fiction, and I think back to much of it laying the groundwork for my interest in the social sciences and their close observation of human behavior. History was my favorite subject—I found it fascinating, and consistently received my best grades while I found myself cruising through history classes. Jumping ahead a bit, it turns out that I actually earned more credits in History than Political Science as an undergraduate at Buffalo. My grades were just as good as well. I think my exposure to a range of history courses in high school was the basis for my drive to understand the historical context of political phenomena. But I think the most significant foundation for my undergraduate career came in the AP History courses I took—a two semester sequence of Advanced European History (which provided college credits). These were taught by Dr. Bernard Hirschhorn, a Columbia University Ph.D. who had taught in the City University of New York (CUNY) system, before moving full time into teaching at the high school level. Our textbooks were college-level, as were Hirschhorn’s lectures. This was History being taught at a much more challenging level, and I loved it. I had to develop a whole new way of taking class notes and of doing the reading (in terms of underlining, making marginal comments, and preparing separate set of notes from the readings). None of this had been needed in any of my previous high school courses. I also had to learn new ways to study in order to deal with the heavier reading load, more intensive lectures, and harder examinations. I earned high grades both semesters of AP history, but more importantly I had a whole new perspective on learning, and what I needed to do to navigate college courses successfully. Although my brother was already at Hofstra College (yes, before it became Hofstra University), I did not ask him about how people studied at the undergraduate level. I was developing learning strategies that worked for me; and because Advanced European History was so much more challenging, I was also developing the discipline I would need to do well in college. While I was enthralled with History, I decided to major in Political Science as an undergraduate at SUNY Buffalo (UB). In part this was because I saw that I could continue my interest in the historical context of politics. In addition, I came to the conclusion that Political Science gave me a greater range of post-graduate/ professional options. And, I must admit, I decided on my major in part because my older brother was majoring in political science at Hofstra.

1.3 The Foundations for The Intellectual Journey: SUNY Buffalo

1.3

13

The Foundations for The Intellectual Journey: SUNY Buffalo

The day I started my freshman year at Buffalo in the Fall of 1963, I was 16 years old, stood almost 5′6″ and weighed less than 120 pounds. I had never lived away from home—not at a summer sleep-away camp, at relatives, or for a job (as my brother did working at a hotel in the Catskills). Despite a streak of independence that perhaps only my parents had seen, the first thing I realized was that I was truly on my own in the sense that I had, for the first time in my life, full discretionary power over what I would do, when I would do it, and how I would do it. Luckily, the second thing I realized was that I also now had full responsibility for all of this. I would be spending my parent’s money. I would be spending my time and energy. It was on me to make the most—and best—of the opportunities of being away at college. Part of that streak of independence noted above, was my insistence to go away to college, and not live at home (as many of my friends did by going to Queens College in the CUNY system). To meet this broad, new responsibility for my life, I also realized that a great deal of discipline would be needed. I would have to understand time management as never before and stick to it.4 Luckily, the AP history courses I took in high school had already provided me working models for developing responsibility, discipline, and time management. One key element from those AP courses I have come to realize sits directly at the core of much of my approach to academic study, research, and scholarship: the method of note taking I developed for the AP courses. At Buffalo I would buy notebooks with very wide left-side margins. I took class notes on the right side, but when I studied, I also used separate sets of notes from the assigned readings, as well as the marginal notes, underlined, and highlighted passages in my books. As I studied I would, in different colored inks, cross reference material, points, arguments, from the readings with those of the class notes (including arrows, feedback loops, and other diagrammatic structures).5 In retrospect I now realize that I was developing synthesizing skills—skills that helped me bring things together, to make greater sense and a greater whole from the various parts of

4

For example, despite my dedication to reading, I was a full-fledged child of TV, with some of my first memories of Washington Heights involving sitting in front of the black and white Philco TV in our living room. While my childhood was filled with hundreds of hours of TV, I watched almost no TV of any kind my freshman and sophomore years. Except for coverage of the Kennedy assassination, the one TV memory of my freshman year was jamming into the communal TV room one Sunday night to see the Beatles debut on the Ed Sullivan show. 5 Both undergraduate and graduate students will recognize this description—it was exactly how I used the blackboard in my classes—with arrows and loops and diagrammatic structuring. I later learned that studies had indicated that such visual cues were positively related to learning the material.

14

1 Getting from Then to Now: A Personal Intellectual Autobiography

a course. I was also developing a larger feedback loop, moving from studying to synthesis to more listening and reading back to studying, and so forth. Where all this came from, I have not a clue; (possibly the visual aspects of both my drawing and strong interest in architecture?). To jump ahead once more, the idea of synthesis has been central to both my teaching and research. Below I will say more about my take on “theory,” but for now simply note that I have argued that one key question about theory is “where does theory come from?” My answer has always been: just about everywhere. Almost anything can spark a nascent question, hypothesis, proposed relationship, model (and importantly, so can “political memories”). The key is looking across prior literatures, drawing from unexpected sources, and putting things together in unexpected ways—synthesis! Across my career I have been adept at seeing how things go together, how work across disciplines and the sub-disciplines can dovetail, overlap, reinforce, or challenge each other; how widely diverse works could add up to larger pictures and broader explanations. ‘Scientific’ cumulation is not possible without it; the pursuit of synthesis has also been an important component of my interest in cumulation (e.g., Most/Starr 1989). Generations of graduate students know that at the beginning of a seminar I tell them that other faculty will hone their critical skills by asking them to search for flaws in theoretical logic, research design, and methodology; (and I certainly did much of this as well, especially in the research design/logic of inquiry courses I have taught). But I also asked graduate students to keep other questions in mind when doing any reading: How is this material useful to this course? To one’s research? To the issues/questions at hand? Where does it fit in? How does it fit in? What does it add that is new? If the answers to these questions is “not much”, then students have learned an important strategy to read and think critically. A former doctoral student has written about this aspect of my work in a letter of recommendation for an academic award for which I had been nominated6: I believe that Harvey’s greatest contributions have resulted from his skill in collaboration. He has a way of bringing together both people and ideas and creating interactions that improve both the people and the ideas… I was once told that Harvey is one of the best discussants to have on your panel. He makes connections between papers that others don’t see, and helps panelists find ideas in literatures that they had not yet considered.

These study habits, along with responsibility and discipline, made me great at exams. These led to great grades. Great grades, great faculty, and great preparation were all necessary to have the opportunity to attend Yale University for graduate work. But let me start with History. As noted earlier, I took more credits in History than Political Science at UB. This deepened my understanding of the context within which political phenomena occurred, and the courses on Asia and Southeast Asia as

6

I got it!

1.3 The Foundations for The Intellectual Journey: SUNY Buffalo

15

well as American diplomatic history were quite important in understanding the unraveling saga in Indochina during the 1963–67 period I was at Buffalo. One key history teacher, in a two-course sequence on South and Southeast Asia which introduced multiple non-western religions as key elements of the historical context, was Theodore Friend, whose book (1965) on the Philippines won the 1966 Bancroft Prize. Only a sophomore, I felt I came into my own in terms of expressing ideas (and connections!) in that class. He was quite encouraging. At the beginning of my senior year Dr. Friend (as he shall always be to me), sent me a note saying he had written a recommendation for me, and adding, “be sure not to underestimate yourself” – something I continue to fight against to this day. Note that from 1973 to 1982 Theodore Friend served as President of Swarthmore College. The Political Science Department had some young, energetic, and well-trained faculty, and some slightly older ones of national prominence.7 It helped prepare me for graduate research and for a career in international relations (although I did not quite know it then that I would wind up in international relations—or IR). The only two courses in IR I took at Buffalo were from Glenn Snyder, who seemed older than his 40 years. I now know (but not as an undergraduate), that Snyder’s Deterrence and Defense (1961) was one of the seminal works produced by the first wave of security studies scholars, that included other giants such as Bernard Brodie, Thomas Schelling, and Herman Kahn. As Robert Powell has noted, these were the scholars that produced the “foundational ideas” for the study of deterrence. As an undergraduate in 1965–66 I was learning about deterrence, strategy, and game theory that included such things as the Prisoner’s Dilemma and Chicken. A few years later as a Yale grad student I discovered that some of my classmates in IR courses were not quite as familiar with these ideas! In graduate school I came to appreciate Snyder’s place in IR and would later continue to assign his articles in both undergraduate and graduate classes. I know that much of my early interest in alliances was affected by his scholarship. While I almost always point to Bruce Russett’s graduate IR course as my gateway to the work of Harold and Margaret Sprout (which informed the opportunity and willingness framework and my interests in geography), it was actually in Glenn Snyder’s 200-level International Politics course in Spring 1965 that first introduced me to the Sprouts. I am happy that we had a continued (if sporadic) correspondence during my career, and that I participated on a panel in his honor a year before he died in 2013. The Buffalo department also provided a splendid introduction to the research process, and what it meant to do social science research. This occurred through the departmental Honors Program. In my junior year, the Honors course was taught by one of those younger and quite energetic faculty, Richard M. Johnson, who later had a long career at the University of Illinois, Chicago. It was a course in research design along with the philosophy of science that informed and evaluated the

7

I still see and correspond with one of those young professors, the Africanist Claude Welch, who came to UB a newly minted Oxford Ph.D. in 1964 at age 25. He retired in 2017 but is still going strong.

16

1 Getting from Then to Now: A Personal Intellectual Autobiography

Harvey with Meredith Reid Sarkees, Glenn Snyder, and Claudio-Cioffi Revilla on the occasion of “A Roundtable in Honor of Glenn H. Snyder;” October 2012 at the annual conference of the International Security Studies Section of the ISA, the International Security and Arms Control Section of the APSA, and Triangle Institute for Security Studies. Source From the author’s personal photo collection

choices made in design. I was introduced to the likes of Thomas Kuhn, Karl Popper, Ernst Nagel, and Carl Hempel, as well as such exotic topics as content analysis, logic, statistics, and data gathering. He was my undergraduate adviser and did a good job of guiding the honors group (only about eight in my junior year, but just four in my senior year).8 One quick story: I was on a long string of A’s in my courses, starting with three in my second semester as a freshman, but all A’s across my sophomore year and the first semester of my junior year. But in the Spring semester of my junior year I had three final exams scheduled on the same day, with Johnson’s exam scheduled the day before. I asked him if I could take his exam a day early to give me more prep time for the day of the triple. He agreed, and on that day let me take it in his office. He just asked me not to tell the others what was on

8 My Research Design paper for Johnson’s Honor’s Course was, “Policy Dilemma of the American Public: Attitudes Concerning the Distinction Between the National Interest/Foreign Policy of the United States and the Military Policy of the United States.”

1.3 The Foundations for The Intellectual Journey: SUNY Buffalo

17

the exam, to leave it on his desk when I was finished, and then left. I continue to be amazed that such a simple matter of trust still resonates so profoundly with me.9 The two-semester Honors course in my senior year 1966–67 was run by Lester Milbrath, who had been hired the year before, moving from Northwestern. A year younger than Glenn Snyder, Milbrath had just published perhaps his most well-known work Political Participation: How and Why Do People Get Involved in Politics? (1965). The senior year Honors experience was completely different. The year-long course was about “legitimacy,” how people understand the concept and ways to teach the concept. After going into conceptualization, operationalization, measurement, validity, etc., and writing out different ways to present the elements of legitimacy, Milbrath arranged for us to engage in an experiment at Niagara Falls High School. We created an experimental design of pre-tests, different treatments (that is presentations on legitimacy), post-tests, and control groups, using groups of students taking civics and American government courses. All of the class presentations were made by the Honors students. The pre- and post-tests were collected and brought back to UB. Before going to the high school, each of us had developed a research topic with Milbrath, including a hypothesis (or set of hypotheses) about what might affect perceptions of or meanings of legitimacy. Thus, each Honors student covered some major aspect of the entire project. Upon returning from Niagara High School, the tests were graded, and the scores entered on computer punch cards along with the other student variables.10 Yes, the Honors students used such ‘hi-tech’ equipment as the card puncher, used another machine to catch punching errors, and did some preliminary runs on the counter-sorter that could be used for some of the hypotheses. Graduate students then did the necessary statistical work to generate results. The Honors students then wrote up the full research report/paper.11 This structure of a semester reading about philosophy of science and research design culminating in a research proposal, plus a year of hands-on field research employing experimental design, was an almost perfect preparation for social science graduate study.12 In the Fall semester of my first year of graduate study at Yale, all the first-year students had to take the research design and philosophy of science course, taught that year by (the great) Robert Lane. I watched as many of

The next day I aced the first two final exams and received A’s in both courses. The third exam of the day was Glenn Snyder’s 400-level strategy course—got an A on the first essay, a B+ on the second essay, a B+ on the exam and a B+ in the course. All the grading was done by Snyder’s teaching assistant, whose name I still remember to this day, but will not reveal. 10 Of course, no student identities were involved, only code numbers that could connect scores with the appropriate experimental group and the other variables that had been included. 11 My Honors paper for Milbrath was: “Niagara Falls Project: Legitimacy and Learning Techniques.” 12 The fact that the Niagara Falls Project involved a research team, along with learning the basics of philosophy of science and research design, exposed me to the fundamental idea that science is a “social enterprise.” This important but underappreciated truism is developed at greater length below. 9

18

1 Getting from Then to Now: A Personal Intellectual Autobiography

Starr photo taken in Fall semester of senior year at SUNY Buffalo (1966), turning 20 years old. Source From the author’s personal photo collection (SUNY Buffalo yearbook)

the other students from top Ivy League schools struggled to see how this material fit with the undergraduate curriculum they had taken which stressed political philosophy or ‘policy’ studies. But the Buffalo faculty had prepared me well. I was already familiar with the material, did well in the course and, if I had taken the

1.3 The Foundations for The Intellectual Journey: SUNY Buffalo

19

‘normal’ comprehensive exams two years later, I would have had Research Methods as one of my two fields. Ironically, Buffalo also prepared me in a way that I only came to understand years later when I taught at Indiana. I noted that in high school math was probably my worst subject. At Buffalo I was required to take two semesters of math, and rather than moving on to calculus (or pre-calculus), I took the two-course sequence in “Significance of Mathematics.” The student name for this rather large lecture course was “Mickey Mouse Math.” I still have the textbook. Part one was “Mathematics as a Search for Patterns” and had chapters on geometry, arithmetic (including the Fibonacci sequence), and “New Patterns.” The lattermost chapter included matrices, sets, Venn diagrams, etc. Other parts of the book included logic, probability and statistics, typology, with a final chapter on the “elementary concepts of the calculus.” Two sections of the book were about mathematics as a way of thinking and mathematics as a tool. In retrospect these topics helped me to synthesize and organize my thinking about empirical phenomena through the search for patterns, commonalities, and differences. Other topics were similarly of importance to research design and logic of inquiry. Again, in retrospect I’ve seen these courses as being much more relevant to my research as a social scientist than learning the math central to the development of formal models (which required a depth of training or a turn of mind that I would never achieve). Just as the high school AP courses set the stage for college, so did the UB Honors program set the stage for graduate study. At age 20 I graduated from SUNY Buffalo in May 1967 Summa Cum Laude, with Highest Honors in Political Science, Phi Beta Kappa. I was accepted into every graduate program to which I applied with full four-year funding. Among others these included Syracuse, Cornell, Northwestern, Michigan, Harvard, and Yale (and Chicago, which did not offer any funding). Having won a Herbert H. Lehman Fellowship from the State of New York, my stipends would have been greatly enhanced had I attended a school in New York.

1.4

Where the Rubber Meets the Road: Yale

Why grad school? Why Yale? While at Buffalo I took the law school admissions exams (LSAT’s) as well as the GRE’s for graduate school, and I did apply to about half a dozen law schools. While not aiming at the very top, I did apply to, and was accepted by, very good law schools such as New York University, Chicago, and Georgetown. The decision-making process concerning whether to go to graduate school or law school was purely internal, with no outside input. The main reason I applied to both was, given my academic success as an undergraduate, to have a wider range of post-graduate options. The internal list of positives and negatives quickly led me to take the graduate school route. I realized that I did not like having

20

1 Getting from Then to Now: A Personal Intellectual Autobiography

others dictate/constrain my time and activity. I did not like such organized activities growing up—Little League lasted two years (although I was thoroughly mediocre), and the Cub Scouts lasted two meetings. While law school clearly set out a path that could lead to a much larger income, the thought of having others controlling my workload—the things I could work on and those I could not—for a number of years and with hours that had to be put in, was repellent. I concluded that “punching a clock” was not my thing. I have been reminded by one of the people who was hired when I was a department chair that I said, “this is not a job, it’s a lifestyle.” Somehow, I had some understanding of this even before grad school. I think part of having the temperament for being successful in academia requires understanding this distinction. I knew I had the discipline for the time management that a Ph.D. and a scholarly career required. Another primary element of having the “right” temperament is being able to manage one’s time, to juggle the time requirements needed for research and publication, along with the time needed for one’s teaching and service (departmental, university level, and professional). I understood that there were academic deadlines—tenure and promotion being the most prominent. But having control over one’s time, research program, daily and weekly activities, etc., mostly fell on the individual scholar. For me, this sounded about right. While understanding what was the ‘right’ temperament for an academic career developed over time, beginning in grad school—I think my own streak of independence, my propensity to observe people and their behavior, along with my sense of responsibility, formed the foundation for my having the proper temperament. If choosing grad school over law school was an ‘internal’ decision, choosing which graduate school was another matter. I recall conversations with the Buffalo faculty, especially Richard Johnson and Lester Milbrath. Some of the younger faculty that I had, and liked, were of the newer empirical, ‘scientific’ trend. I asked for guidance as to which was “the best” graduate program in political science. Not surprisingly, conversations usually led quickly to Harvard or Yale (Theodore Friend received his undergraduate degree from Yale). I also recall that Yale was presented in terms that I interpreted as being the best “social scientific” of the top departments, with faculty that stressed empirical analysis in their work. These included a number of giants in behavioral studies such as Robert Dahl, Charles Lindblom, Robert Lane, or Karl Deutsch. These descriptions of the Yale faculty appealed to me. As an undergraduate I had to take the micro/macro sequence in Economics; and I also took introductory courses in Anthropology and Psychology. These courses opened my mind to areas never thought about before, and, again, I liked how they approached the study of human behavior. Indeed, I took Psychology in my senior year at UB to get the requisite credits to graduate—but, had I taken Psychology in my freshman year, there is a better than even chance it would have been my major(!).

1.4 Where the Rubber Meets the Road: Yale

21

I have startled more than a few people when I tell them I did not visit a single campus before choosing Buffalo; nor did I visit either Yale or Harvard before deciding on Yale. Regarding graduate school I was concerned with the quality and nature of the faculty and the programs. There was, however, one other factor that affected my decision. The parts of my brain concerned with responsibility and discipline told me that New Haven would be far less distracting than Boston—and that the energy and concentration that would be needed in graduate school would be far greater than anything I had previously experienced. Those parts of my brain were absolutely correct.13

1.4.1

Yale: Multidisciplinarity, Interdisciplinarity, and Process

It is unsurprising that graduate school played a foundational part in the initial stages and then development of a lifelong research program with all of its various branches (including those that didn’t quite pan out and had to be pruned…). Thus, this section on Yale will note only some of the broadest influences on the career to follow. Following sections dealing with the thematic substance of my scholarly work will also reference back to initial work and ideas originating in graduate school, and the faculty that helped along the way. The same background that the Honors Program at UB gave me for doing well in Political Science, as noted by my first term experience in Robert Lane’s class, also grounded me into the social science enterprise as a whole. The UB faculty who advised me were correct that the Yale program would be strongly focused on the social sciences. Their observations had picked up on an important aspect that cut across teaching at Yale. David Apter, the renowned comparative politics scholar, and co-editor of the first volume of emeriti essays for the Koerner Center, also wrote the Introduction to that volume. He provided brief overviews of the individual contributors, noting again and again how each scholar “built bridges” across multiple disciplines, and how the 22 emeriti—reflecting the culture of Yale—“refused to hesitate at disciplinary boundaries.” I have characterized one of the central aspects of the graduate courses at Yale as being multidisciplinary. I was taking courses that would lead me to a doctorate in Political Science, yet the heavy reading in each course included substantial amounts of material from the other social sciences. Comparative politics courses were heavy My first two years at Yale I lived in the Hall of Graduate Studies. Political Science also had offices and seminar rooms in HGS. I could sleep, eat, and go to class in the same building. Straight out the front entrance of HGS it was a short half-block to a side entrance of Sterling Memorial Library, where the great bulk of my time during those two years was spent in the Graduate Reserve Book Room (which I always thought would be a good name for a folk-rock band). If I went out the main entrance to Sterling, it was another half block to cross the street to the Yale post office where I would collect my mail. My universe was thus (usefully) constrained.

13

22

1 Getting from Then to Now: A Personal Intellectual Autobiography

on the policy making process and the role of economics, especially the course from Charles Lindblom. Other courses on policy, urban politics, American government, and IR included a range of economic models. I was introduced to the work of Mancur Olson, Buchanan and Tullock and others; work based on formal models of rational choice and game theory. Courses taught by Doug Rae, Bruce Russett, and Robert Dahl, among others, would lead me back to new (and to me, mind blowing) material about collective goods and the nature of collective action. In this regard Doug Rae’s course on empirical theory was particularly important. Extremely important to my later work on conflict, was the heavy dose of anthropology, sociology, psychology, and social psychology that appeared in almost every course, particularly James David Barber’s course on decision making, and Russett’s IR Theory course. These materials were never presented as something separate from Political Science that, oh by the way, could be useful to us, but as part of the tapestry of literature relevant to whatever issues the specific Yale course was dealing with at that point in the semester. As noted in the sub-head for this section, all of this is why I see a key contribution of the Yale program as being ‘multidisciplinarity’—thinking broadly as an empirically oriented social scientist first, then as one specializing in Political Science. A quick Google provides a useful definition of ‘multidisciplinarity,’ as an activity that draws on knowledge from different disciplines but stays within their boundaries. I think this is well illustrated in the Koerner Center essays. On the other hand, interdisciplinarity “analyzes, synthesizes and harmonizes links between disciplines into a coordinated and coherent whole.” The multidisciplinarity of Yale harmonized wonderfully with my innate sense of synthesis. Its multidisciplinarity naturally drew upon and drew out my undeveloped but clear tendency toward the synthesis of interdisciplinarity. The best example of this is the “opportunity and willingness” framework (O/W), its utility and applicability, e.g.: “…O/W has moved well beyond an organizing principle to a model of causal complexity with great potential” (Starr 2018, vol. 3: 1). Ironically, the courses at Yale included no systematic literature from Geography, an area with which I am now identified. It was in Russett’s IR theory course that I was assigned the work of Harold and Margaret Sprout—both an inspiration for the later development of opportunity and willingness, and for extensive forays into political geography, geopolitics, and spatial approaches to IR phenomena. Geography has been the area where I have the longest list of crossover activities— publishing in Geography journals, reviewing for such journals, presenting papers at Geography conferences, giving talks at Geography departments, and in grants. But note that I have engaged in similar activities with: Economics, History, Psychology, International Business, and most recently in Linguistics.14 14

I have presented papers or served as a discussant at the annual meetings of the: American Historical Association, Association of American Geographers, International Society of Political Psychology, and the Linguistic Society of America. I have also reviewed NSF proposals not only for the Political Science, but six other programs including Geography, Economics, Decision Science, Law, and Methodology/Measurement/ Statistics. The activities with Linguistics can be

1.4 Where the Rubber Meets the Road: Yale

23

Thus, the multidisciplinarity I have followed across my career has been broadened into interdisciplinary contributions that “cross boundaries” (see Starr 2006a, b) through synthesis. This includes the work on the logic of inquiry (best represented by Most/Starr 1989, 2015; Goertz/Starr 2003; Cioffi-Revilla/Starr 1995; as well as Starr 2005a reproduced in Part II below) along with the elaborations and applications of “opportunity and willingness.” I also think of my time at Yale as the merging of multidisciplinarity and process. Rather than absorbing material (as one generally does as an undergraduate), this is when I fully began to think in terms of ‘how’ and ‘why’ questions rather than “what” in order to produce new material of my own. While I have studied “’war,’ my main interests have always been in social conflict more generally and especially conflict processes. The empirical theory and research courses, early reading in IR such as Ken Boulding’s Conflict and Defense (1962) among many others, led me to think about dynamics, moving parts, feedback loops, and the like. Graduate school is when I fully encountered context, and the feedback loops from context to agent, from agent to behavior, from that behavior to a newly formed context, and so on. However, my classes made me see that this context was, indeed, multidisciplinary, and it would be counterproductive to limit context to “the political.” Multidisciplinarity, interdisciplinarity, and process, I think usefully and concisely summarize important elements of the Yale experience.15

1.4.2

Bruce Russett: Initial Interactions and Mentoring

A large number of the academics I have known, if not a majority, have acknowledged the singular importance of an individual in their development; one with a foundational impact as they emerged, grew, and established their own place in Political Science. Usually this person is called a mentor.16 For me, Bruce Russett is such a foundational mentor. This is not a secret. In the (disguised) festschrift I edited for Bruce (Starr 2006a, b), the dedication read, “To Bruce Russett: Teacher, mentor, colleague, friend, and most of all—inspiration—to us and to a discipline.” His work, his teaching, his co-authorship, and his assistance and model as a mentor, have indeed been foundational in my thinking and the shape of my career. When I arrived at Yale I had not yet decided on a subfield or specialization within political science. I know that things have changed radically since I applied to Yale in 1966–67 and today. I was not asked, nor did I feel compelled to declare a major field in my applications for graduate school. This is not the case today. found on the webpage for ConflictAnalytiX (https://www.conflictanalytix.com/), the consultancy founded by Stan Dubinsky (a linguist) and myself. 15 Note that one purpose of Iqbal/Starr (2016) was to bring systematic analysis to bear on the full range of the state failure process. 16 This is not to deny that we all might have a series of mentors at different stages of a career, or at different institutions, etc.—only that one person usually stands out as this foundational mentor.

24

1 Getting from Then to Now: A Personal Intellectual Autobiography

Photo of Prof. Bruce M. Russett with several ofhis former students at his retirement dinner in New Haven, 2011: from left to right, Roy Licklider, Bethany Lacinia, Bruce Russett, Fred Chernoff, David Kinsella, and Harvey Starr. Source This photo is from the personal photo collection of Bruce M. Russett who granted permission

I understand most of the reasons why, but I feel that the pressure to select a subfield and even a specific area of that subfield before entering a graduate program is much too premature. As happened to me at Buffalo, each graduate course was a revelation, introducing me to new realms of knowledge, new phenomena, new ideas, new theories. And, as I have explained, the fun was in finding where and how they overlapped, helped or challenged each other, and provided broad new perspectives or frameworks for my understanding of the political, social psychological, and economic world.17

I have often spoken of things that I read in graduate school that “made my brain explode.” These were many and varied. But I have identified about a dozen books (trying to find articles would have been quite overwhelming) that could be included as such readings. Most clearly at the top of such a list would be Anatol Rapoport’s Fights, Games, and Debates and Karl Deutsch’s The Nerves of Government. Four more books that would complete the top half of those dozen or so works: Thomas Schellings’ The Strategy of Conflict, Lewis Coser’s The Functions of Social Conflict, Kenneth Boulding’s Conflict and Defense, and Charles Lindblom’s The Intelligence of Democracy. Note how this list reflects the multidisciplinarity discussed above: Schelling, Boulding and Lindblom were economists; Coser was a sociologist; and Deutsch is a political scientist (although he also held a law degree). Rapoport defies categorization. A Ukrainian Jew, he was initially trained in Vienna as concert pianist who moved into both conducting and composition. With the rise of Nazism, he moved to the United State and received a Ph.D. in mathematics from the University of Chicago. After serving in WWII he returned to Chicago and joined the Committee on Mathematical Biology. He was Professor of Mathematical Biology at the University

17

1.4 Where the Rubber Meets the Road: Yale

25

In my second year of classes I had not yet chosen a major field. Given the constraints of required courses, I had one open slot. Given my previous courses I was basically asked to choose either IR or American Politics—either Bruce Russett’s International Relations Theory course or James David Barber’s course on the American Presidency. I heard from other grad students that Barber would assign each student a president and, guided by a codebook, amass data and material that would be used to develop a psychological/political character study of that president. That did not interest me at all.18 Various references from other courses and student comments about the topics and readings in Russett’s class, led me to choose it. Once I saw the course syllabus, and started the readings I was hooked, especially as I now had taken a year of other graduate courses that provided me with all sorts of connections, methods, and literature to plug in. Perhaps the best of these earlier courses I took in my first semester. One reason I went to Yale was because of Karl Deutsch, and his quantitative approach to IR. Alas, in the Fall of 1967 when I got to Yale, I discovered that Deutsch had moved to Harvard, so I never had the opportunity to take a course from him. However, a recently minted Yale graduate student of Deutsch, with a position at Columbia University in Manhattan, had agreed to train up to New Haven once a week to teach the course on the quantitative study of international relations—Don Puchala. Twenty- two years later Puchala and I became colleagues when I joined the faculty at the University of South Carolina. Puchala had left Columbia for South Carolina in the early 1980s to become Director of the Institute for International Studies.19 His course at Yale was important for my formative view of research design and methods as it developed into my broader interests in the logic of inquiry. In turn, these were central to my later work with Ben Most, especially our 1989 book, Inquiry, Logic and International Politics. This book was submitted to and accepted by the University of South Carolina Press in its series, “Studies in International Relations,” edited by Don Puchala and Charles Kegley.20 Over the course of my career some of the best, most interesting and fun conversations/arguments about the study of IR and IR theory have been in Don’s office in the Institute, steps away from my own office. Bruce Russett was born in 1935, eleven years before me, and received his Ph.D. from Yale in 1961, ten years before I did. Karl Deutsch was his dissertation advisor as well. When I took Russett’s IR Theory course in my second

of Michigan, and then Professor of Mathematics and Psychology at the University of Toronto. Kopelman (2019) pays tribute to Rapoport: “Rapoport’s contributions spanned scientific disciplines and included the application of mathematical models to biology and the social sciences, alongside metatheoretical work bridging semantics, ethics, and philosophy.” 18 As you might have guessed, all of this culminated in Barber’s most famous work, the 1972 book, The Presidential Character. 19 Later to become the Walker Institute of International and Area Studies. 20 It is important to know that we signed the contract with the University of South Carolina Press in 1984, years before there was ever a thought of moving from Indiana. The book had a delivery date for a completed manuscript of 1987, which was pushed back due to the sudden and untimely death of Ben Most in November 1986.

26

1 Getting from Then to Now: A Personal Intellectual Autobiography

year he had already published five (of his 27) books, and about 25 articles in refereed journals (with a few book chapters thrown in), including the top journals in Political Science and IR. That is, 25 of his over 250 such publications.21 His IR Theory syllabus was long, quite extensive in its bibliography, and served as the template for my own IR theory syllabus, especially in its early incarnations. Every section of that Russett syllabus brought something new. Given his background in economics, and his place as one of the pioneers to get IR scholars to learn and use economics in their work, I was introduced to a number of economic concepts, models, and findings. As noted above he introduced me to the work of Harold and Margaret Sprout and their “ecological triad” of environment, entity, and environment-entity relationship—which became a central influence and feature of the opportunity and willingness framework. He assigned work on individual decision makers and psychological approaches; work on strategic studies allowing me to draw on my courses with Glenn Snyder with its readings on game theory and formal models. Finally, I must note that Russett’s IR theory course introduced me to the pioneering work in international integration theory by Karl Deutsch and Ernst Haas. Having models that helped me understand processes of cooperation and coordination were crucial complements to understanding conflict processes. Deutsch’s concept of a “pluralistic security community,” where sovereign states could cooperate in the formally anarchic environment of the international system while retaining their independence, was basic to understanding ‘order’ in such systems. It would also later provide me with a fresh perspective on the democratic peace. Suffice it to say Russett’s course itself was a template on how to apply work from across the social sciences to the study of IR in all its manifestations. Indeed, Russett had also assigned J. David Singer’s classic 1961 article from Behavioral Science, “The Relevance of the Behavioral Sciences to the Study of International Relations.” In Russett’s course I had also found a model, although not consciously understood at the time, for being a leader in trying to bring other disciplines to the attention of IR scholars, and their incorporation into the work on IR. My presidential addresses to both the Peace Science Society (International) and the International Studies Association stressed the interconnections between IR and Geography (see Starr 2003, 2013a, b). My most recent, and ongoing, work has tried to do the same with Linguistics. While Russett’s seminar had been a great experience, especially in the papers I wrote, Russett was but one of the Yale faculty whose intellect, research, and courses had influenced me. The role of mentor developed slowly. As I approached time for comprehensive examinations, I took advantage of a newly effected change. Students would have a choice of traditional exams (which at the time I seem to remember were closed book), or of writing three article length and article worthy

21

To get a full feel for Russett’s range of interests and expertise, and publications, see Starr (2015a).

1.4 Where the Rubber Meets the Road: Yale

27

papers. One paper had to be a survey of the literature, one had to be a piece of empirical research, and the third could be any type of paper. Only two could deal with the same field. I chose the paper option of comps, being the first Political Science student to do so. My logic was that I had always done very well at exams, so why do more. On the other hand, I needed much more experience in writing papers.22 For the literature review, my paper was on “The Foreign Policy Making Process—the Decision Maker and his Environment;” (which also foreshadowed opportunity and willingness). For the third paper, my topic was, “The Urban Highway and the Urban Setting: Government Action and Group Interest.” This paper reflected my interest in policy and the policy making process, something covered extensively in the courses I took from Charles Lindblom, Doug Rae, and what were essentially two other courses on urban politics(!).23 The research paper came out of one of the articles assigned by Russett, the classic article by Mancur Olson and Richard Zeckhauser, “An Economic Theory of Alliances” (1966). Both Russett’s and Rae’s courses included readings dealing with collective goods and collective choice, a topic that fascinated me and later showed up in quite a few of my publications, particularly the earlier ones. My comprehensive exam research paper was, “Is There ‘An Economic Theory of Alliances’: An Investigation into the Collective Good Approach to Alliance Behavior.” In the paper I applied the Olson and Zeckhauser model regarding the burden sharing of goods with collective good properties (and research design with some modifications) not only to NATO as in Olson and Zeckhauser, but to the Warsaw Pact, SEATO, CENTO, the Rio Pact, and the Arab League. Russett was my advisor on this paper, and guided its set of revisions. This paper became my very first publication. At the time I was planning and writing these papers Russett was working on his book, What Price Vigilance: The Burdens of National Defense, published by Yale University Press in 1970. Chapter 4 of the published volume was called “Alliances and the Price of Primacy.” Russett had taken my paper, re-ran the analyses with an updated and better data set, and edited it with further revisions. This was not an edited volume, but Russett was clear in giving me co-authorship of Chap. 4. After the usual caveat that he was solely responsible for any errors of fact or interpretation, he wrote, “…my student Harvey Starr shares with me the authorship and responsibility of Chap. 4 on

22

Although I should note that the paper I wrote for Puchala was also the paper for my course on Southeast Asia (Puchala and Robert Tilman had agreed to allow me to submit the same paper for their seminars). Puchala liked the paper, but Tilman liked it so much he sent it to another scholar at an Asian studies journal asking whether he thought it was submission ready. While this was as far as it went, I was greatly encouraged that a Yale faculty member thought my first-year paper was good enough for an external academic to look at. 23 Writing this makes me realize how and why I had the background to so easily (and positively) interact at Indiana with Elinor (Lin) Ostrom across a range of topics, especially policy related. While Russett was a foundational mentor, Lin was extremely important in helping me understand what being a professional academic meant, especially in departmental service.

28

1 Getting from Then to Now: A Personal Intellectual Autobiography

alliances” (p. xii). At the bottom of the first page of Chap. 4 was the note: “This chapter was written with the collaboration of Harvey Starr.” Here was an element of mentorship I have tried hard to emulate across my career— a “nuturant” and “generative” approach to the professionalization of graduate students and assistance to their careers. Those two adjectives were used to describe Karl Deutsch in a Biographical Memoir written by Merritt et al. (2001). Mentorship of this sort is only one of the several ways in which I felt myself a “grandson” of Deutsch, reflecting both scholarly and professional influences he passed on to Russett, who passed them on to me. Russett’s generosity in giving me co-authorship only became apparent to me as my career unfolded, and I observed how many scholars I knew who would have simply thanked me as one of their graduate students in a footnote, or with a note that said, “the chapter had benefited from the assistance of Harvey Starr.”24 As the study of alliances was one major research component of my early career, this publication was quite important; it again gave me confidence that I could do the work needed to be successful.25 Note also that “What Price Vigilance?” won the American Political Science Association’s 1971 Gladys Kammerer Award for the best publication on policy in the United States. With no humility whatsoever, let me note that in the book review of Vigilance in the Western Political Quarterly26 the reviewer noted: “The chapter on alliances and the price of primacy is worth the price of the book. It is a very worthwhile and thoughtful examination of alliances from a public versus private goods perspective.” The papers option for comprehensive exams also taught me how to deal with setbacks and reinforced the need for discipline. While I was the first student to take the papers option, I was also the first student to fail the papers option. Note here that a key element of academic temperament is the ability to deal with failure, especially in the form of rejection. Rejection of submitted publications, grants proposals, etc., will be an ongoing occurrence in the research life of any academic. Learning how to take rejection and use it to improve the research and/or its written presentation will be central to academic success. In order to pass the papers option, all three papers were required to be considered a pass. The lit review and alliance research paper easily passed. But the urban highways paper failed. My committee told me that I had written a polemic not a research paper, and that it needed to be re-written. Using my research materials, which included documents and news reports that I acquired in a trip to Boston, and similar materials collected in New Haven, I produced a social science research paper, whose conclusions still permitted me to summarize the negative effects of how the type of highway that was built, and where it was located, affected poor and non-white neighborhoods in America’s urban renewal project of the 1950s–60s,

The Merritt et al. piece noted that Deutsch also helped graduate students “with the challenges of getting a job, writing assignments, conference invitations, and other opportunities,” (pp. 7–8). 25 It also led to my 1974 article in International Organization, “A Collective Goods Analysis of the Warsaw Pact After Czechoslovakia.” 26 This journal later changed its name to the more recognizable Political Research Quarterly. 24

1.4 Where the Rubber Meets the Road: Yale

29

with Boston and New Haven as case studies. The re-write passed, and I received a stark reminder of what it means to be a social scientist.27

1.4.3

The Dissertation

Most people assume that Bruce Russett was my dissertation chair, but he was not. It was John D. Sullivan. I cannot fully remember the process which brought us together, since I did not take a course from him. However, we did interact in a variety of departmental settings, and he was a young and enthusiastic assistant professor of IR with publications and interests in alliances. I think a key factor was that Russett was on leave in 1969–70, and that Jack (as I called him; but never “Bruce” with Russett) wanted to chair a dissertation committee as he went into the tenure process. Russett was a member of my committee, as was the noted historian Gaddis Smith, known for his work on American foreign policy and diplomacy. In large part, the research design and proposal were developed with Jack Sullivan. The dissertation was titled, “To the Victor: An Analysis of the Distribution of Payoffs, Spoils and Losses Among War Coalition Participants.” We liked the topic because it was somewhat different in looking at the end of war rather than its causes or beginnings, which had been the focus of most of the extant literature. In follow-on publications I began to cut into the feedback loops from the endings and consequences of war to new wars, and so forth across time. The dissertation also began a career long interest in the consequences of war and conflict.28 The investigation of the distribution of war payoffs took a quantitative, model testing approach, creating a novel dataset with the operationalization of variables capturing coalition member participation, pre-war goals, types of payoffs (or losses), amounts of payoffs (or losses) and the most challenging variable, “Degree of Fulfillment.” For those readers who are old enough to appreciate this—the data for each war coalition studied took five computer punch cards which filled three boxes (and were color coded in an attempt to make sure they were in the correct order when looking at the cards). These data were analyzed using a primitive forerunner of SPSS. The above reflected the Harvey Starr who became the social scientist. But this project also captured the attention of Harvey Starr who studied history. For each war I spent hours in Sterling Library reading through (and coding) the texts of preand post-war treaties. Most of these, thankfully, were included in the British Foreign and State Papers series. However, there were also treaties that I had to read in French (which I basically could back then with the help of a dictionary) or Spanish. I also used scores of books on war, or wars of specific states or regions. Still, my work on this paper led to another “publication”—a letter to the New York Times Book Review pointing out critical omissions in a review of a book titled Superhighway, Superhoax (July 19, 1970, pp. 34–35). 28 Indeed, my last major research project, on state failure, with Zaryab Iqbal, was concerned with the consequences of internal social conflict and external conflict on the occurrence of state failure. 27

30

1 Getting from Then to Now: A Personal Intellectual Autobiography

The dissertation appendix, “List of Sources Used for the Collection of Data for Wars,” was 10 pages, only one page shorter than the appendix containing the references in my bibliography. This combination of quantitative social science and history would prove to be a key element in my receiving an offer from Indiana University in January 1972. Regarding the dissertation, Russett emerged as my key mentor at the dissertation defense (called a “colloquium” at Yale) which consisted of me, the three-person committee, and I think one other faculty member sitting in. Russett asked for extensive revisions—which were not difficult. Most involved adding appendices presenting as much data and data analysis as possible, a defense of methodological choices, and presentation of checks on the reliability of the coding and methods. After the defense I spent much more time huddling with Russett than Sullivan. My dissertation was defended, and then fully completed for Sullivan and Russett in June 1971. I was 24 years old. Because I missed the May degree date, I was not officially the recipient of a Ph.D. until the December 1971 degree period (and by then I was 25). One of the regrets of my life was that I was unable to attend the Yale degree ceremony in mid-December. I was in Aberdeen, Scotland.

1.5

Scotland and After

As noted, one reason why Jack Sullivan became my dissertation advisor was that Russett spent 1969–70 as a Visiting Professor at the Institut d’Etudes Europeennes, Universite Libre de Bruxelles. In addition to doing research, he gave talks across Europe. One visit was to the Department of Politics at the University of Aberdeen in Scotland. Apparently, he was very well received, and the Professor of the Department (that is, department chair) Frank Bealey, asked if Russett would want to come to Aberdeen in 1971–72 to teach a senior honors course in Quantitative International Relations. Russett demurred, as he had just been on leave. He did tell Bealey, however, that he had a graduate student who would be an excellent replacement.29 My initial year looking for a position had not been very successful, in part because I could not arrive with a degree in hand. My one viable offer was a one-year job at Swarthmore—to replace Robert Keohane (!) who would be on sabbatical. With both Puchala’s course as well as Russett’s seminar, I was able to quickly put together a syllabus for the quantitative IR course and sent it off to Bealey. I received a letter offering me a position as Visiting Fellow in Politics for 1971–72.30 29

In 1970–71 the Yale Department made me an Acting Instructor. This was a position offered to only one or two individuals in any given academic year and provided graduate students the only opportunity to teach their own courses. I taught one each semester—one on the policy-making process, and one on comparative democracy. I assume Russett knew of this and the Department’s belief in my ability to teach on my own. 30 When I asked my first wife (who is confirmed Anglophile) if she would rather spend a year in Philadelphia or Scotland—the answer was a no brainer.

1.5 Scotland and After

31

This year in Scotland (my first of two, returning in 1978–79 during the sabbatical that was awarded after receiving tenure at Indiana) was one of the most salient events of my life. While I was away in Scotland Russett found me a publisher for the dissertation. The relatively new Lexington Books (then a subsidiary of D.C. Heath) accepted the book manuscript with revisions that were not onerous. The dissertation was published as War Coalitions: The Distribution of Payoffs and Losses in 1972. I was now able to go on the job market again with a book accepted for publication. Another regret is that I never took parts of the dissertation to send out as submissions to journals.31 However, in Aberdeen I was already doing the analyses and writing a follow up publication, the Sage International Studies Series Monograph, Coalitions and Future War: A Dyadic Study of Cooperation and Conflict, which appeared in 1975.

Harvey in Summer 1979 after first sabbatical in Scotland, back on Long Island with brother Mark (holding nephew Gary), father Nathan, and nephew Monte (who has never lived this photo down!). Source From the author’s personal photo collection

31 Bruce Bueno de Mesquita has generously stated on several occasions that War Coalitions was underappreciated, being “decades ahead of its time.” He noted that by looking at what war participants can get, or expect to get out of war, I “anticipated the game theoretic revolution in international relations, being the first to closely investigate expected outcomes as a way to understand initial decisions about getting into war in the first place.” (Comments from his presentation at the panel, “On the Career Contributions of Harvey Starr,” Annual Meeting of the International Studies Association, 2016, Atlanta.)

32

1 Getting from Then to Now: A Personal Intellectual Autobiography

Today we would call my year at Aberdeen a post-doc. My teaching load was light. The quantitative IR course was taken by fourth-year, or honors, students (most students would take only three years to complete an undergraduate degree). There were four students in the class. One was Trevor Salmon, who did his Ph.D. at St. Andrews, took a position there, and later became Jean Monnet Professor of European Integration Studies. Subsequently, he moved back to Aberdeen and served as department chair. The year in Aberdeen was life changing. I made life-long friends. Perhaps the closest of my friends was Clive Archer, who was a year younger than I was, finishing his doctorate while a faculty member in the Department of Politics. I would joke he was the brother I never had (a joke my brother Mark never appreciated). Clive was an expert in Scandinavian politics and small states, international organizations and the EU, and defense studies. He traveled and researched extensively in the Scandinavian countries and across Europe. In recognition of his work on Scandinavia, he was awarded the Knight First Class, Royal Norwegian Order of Merit in 1996, and the Commander of Order of the Finnish Lion in 2003. In recognition of his work on the EU, he also was awarded a personal Jean Monnet chair through the EU (which he received after moving to Manchester Metropolitan University). We saw each other whenever he came to the U.S. or I went to the UK. In Fall 1978 he was in Poland, and I stayed at his flat in Aberdeen adjacent to the university. I helped him arrange a semester at the University of Iowa, in the Fall of 1986. He became good friends with Ben Most and was the one to phone me to tell me that Ben had died of a sudden heart attack. Since retirement I have made several trips to the UK, so I was glad to be able to see him relatively soon before he died of cancer in September 2016.

Harvey with Clive Archer in July 2009 at Chatsworth House in Derbyshire, England. Source From the author’s personal photo collection

1.5 Scotland and After

33

The year in Aberdeen provided the opportunity for me to become a confirmed Anglophile, coming back to Aberdeen for a second academic year in 1978–79, making multiple trips backs (minimally one week, usually much longer). In early 1972 my first wife and I took the Spring vacation to drive around Britain, going from Scotland through England and back up to Scotland—from being at Land’s End, the very southwest tip in Cornwall, to John O’Groats at the very northeast tip of Scotland. We put 3500 miles on the car—which was a new 1971 Saab with American specifications (meaning, of course, left hand steering). Saab shipped the car back to the U.S. for us – and I drove it until I left Indiana for South Carolina in 1989! The social scientist in me drew on sociology and anthropology, as I tried to observe—as an outsider—another culture as closely and deeply as I could. I did this through TV, being awestruck at the discovery of something completely different— Monty Python (the best of a set of comedy shows, with a perspective on humor, especially social commentary, that was quite different from American TV). This British sense of comedy played into my lifelong addiction to words and wordplay (as well as my lifelong addiction to humor and comedy of all kinds).32 Again, such comedy was rare on U.S. TV at that time (I don’t think I saw George Carlin until the mid-1970s). A nascent interest in linguistics appeared as well during this time, with my fascination with different British and American usage, spelling, idioms, slang, and the like, as well as the range of different regional and class-based accents in the UK. TV also initiated my ongoing love of soccer (it is the beautiful game). I spent the year reading British novels that might shed light on the key socio-political elements of British culture especially class; comparisons across English, Welsh, and Scottish cultures; as well as the differences between the English and American educational systems. Two prime examples were the groups of novels by C.P Snow in his “Strangers and Brothers” series, and Anthony Powell’s 12-volume “A Dance to the Music of Time,” but I also got into Evelyn Waugh among others. Sprinkled through were autobiographies and memoirs of British politicians. This year in Scotland not only resulted in lifelong friends and an exposure to the different cultures of the UK, but the opportunity to work—to turn my dissertation into a book and finalize that manuscript, do the data analyses and complete a first draft of the Sage monograph, and publish my first journal article.33 All of these scholarly activities were factors leading to two job offers that I received in early 1972. This subsection is titled, “Scotland and After.” The ‘After’ refers to my career that followed at Indiana University and then the University of South Carolina. Here, I will only briefly discuss the job search process in late 1971 and early 1972. As 32

Thanks Dad! When I discovered that Graham Allison’s work on organizational process and bureaucratic politics models of foreign policy making were unknown to the Aberdeen faculty, I was moved to write an essay that became “‘Organizational Process’ as an Influence on National Security Policy,” which was published in the British journal International Relations in the November 1972 issue.

33

34

1 Getting from Then to Now: A Personal Intellectual Autobiography

noted, the CV that accompanied my placement file would now be able to include the International Relations article as accepted for publication and the forthcoming book from Lexington Books, as well as my teaching experience as a Visiting Fellow in Politics at Aberdeen; the latter being something a bit exotic to separate me from the host of other job seekers. Whatever it was, my file produced positive results. By November into early December I received two invitations to interview. One was from the University of Rochester. Since my wife and I had planned to fly back for Christmas with her parents in Buffalo, we arranged to return a few days early and to have my interview at Rochester before Christmas, a drive of less than two hours from Buffalo.34 The other interview invitation came from Dalhousie University in Nova Scotia, which would be for January, and which I wanted to put off as long as possible. To be honest, the Rochester interview is a blur in my memory. A one-day affair, I remember people being kind, I remember Bill Riker as an imposing presence, and I remember a typical cold and snowy Western New York day. But that’s about it. I must have been better than OK, because at some point after returning to Scotland, I received a job offer! I think that was sometime in late January (or perhaps early February). Again, I’m not sure if I knew that my file had been sent to Indiana—my file would have been sent out to a number of schools by the Yale Department’s placement director—but after returning from the Christmas trip to the U.S. one evening the phone rang (actually the rather inexpensive phone in the duplex flat we were renting more closely resembled a chirp) and Dina Zinnes from Indiana was on the phone. We conducted a trans-Atlantic phone interview (I only recall speaking with Dina and do not remember anyone else on the line). Not long after that call I received a job offer from Indiana. Later I was told that the cost of flying me over was not in the budget so that a phone interview was approved (Rochester did not pay for my flight either as I told them we were already planning to go to Buffalo to spend Christmas with my wife’s family). Again, I was later told that I was a first choice as a compromise between the Indiana faculty that wanted a quantitatively oriented scholar, and those that wanted an applicant with a more traditional approach to IR (a rather common battle across political science departments in the 1970s—and after). My background in history, and the heavy use of primary historical materials in the dissertation stood me in good stead with the latter group, and the use of statistics for my analysis across several competing hypotheses and models won the support of the former group. The offers from Rochester and Indiana were comparable, with the Indiana offer somewhat stronger in terms of salary. I have always joked that the difference was that Bloomington was a nine-hour drive from Buffalo, but that Rochester was less than two hours away from my in-laws (who would become ex-in-laws in 1979). The true reason was that Indiana was a much larger department, with four other faculty

34

But, as noted, still too late for the earlier December Yale graduation ceremony.

1.5 Scotland and After

35

members in IR, including Zinnes who was a major figure in the field. At Rochester I would be the second person in IR, but also be responsible for research methods. At that time Indiana had around half a dozen faculty with significant expertise in research methodology. I simply felt more comfortable joining the larger department and I took the Indiana offer.35 Earlier, I had written to Dalhousie returning the airline ticket they had sent. I told them I had two offers, would accept one of them, and thanked them for their interest. The wrote back and thanked me, noting that many job seekers would have taken the interview for practice or perhaps to negotiate a larger salary offer! I have decided to place this next paragraph in the text rather than a footnote. I have joked, both privately and in public presentations, that I have been the most influential figure in the contemporary study of IR. No, I am not that much of an egomaniac. My assertion is based on the fact that Rochester advertised their IR opening again the following year and hired Bruce Bueno de Mesquita. If I had taken the Rochester job, they would never have hired Bruce, and he would never have had that unique relationship with the brilliant Bill Riker which he has always said was so important to his scholarly development. I rest my case.

1.6 1.6.1

Scholarship: Career Themes Conflict: Consequences and Processes

As highlighted above, from my earliest work onwards I have been concerned with conflict processes and the consequences of conflict. My dissertation and the book it generated were about what wars cause not the causes of war. They were concerned with what happened when wars ended, and the effect that wars had on the rewards and punishments for both the losing side and the members of the winning coalitions. The findings from the dissertation project generated a question that was central to much of my work that followed: what were the effects of war or other forms of international conflict on subsequent war/conflict?

35

It turns out there was another major plus for taking the Indiana job. There were no women on the Buffalo faculty until my senior year. There were no women on the Yale faculty, nor on the Politics faculty at Aberdeen. At Indiana, however, there were several women faculty in the department when I arrived (with more to come in my first few years). More importantly, two of those women became important role models: Dina Zinnes, essentially the leader of the IR faculty with whom I worked in many different arenas, both research and teaching; and, as noted, Lin Ostrom with whom I connected initially given our mutual interests with collective goods and collective choice. Later, Lin chaired the department for the four years prior to my becoming chair in 1984. As chair (both at IU and South Carolina) my first question when confronted with any issue, problem, or situation was WWLD? (What Would Lin Do?)

36

1 Getting from Then to Now: A Personal Intellectual Autobiography

Harvey at 26, during tour of southern Indiana, taken in Spring 1973, first year on the Indiana University faculty. Source From the author’s personal photo collection

The Sage monograph that was started during my first year at Aberdeen looked at whether and how the distributions of post-war payoffs affected the subsequent conflict behavior of coalition dyads (from both winning and losing coalitions)— allying again with coalition partners, or warring against them. Key factors for those dyads that became ‘Enemies’ in the ‘Nextwar’ were: a dyad composed of two major

1.6 Scholarship: Career Themes

37

powers, with a common border, and differences in the distribution of territorial spoils in the original coalition. This study, importantly, introduced borders as a key factor in the conflict behavior of states. Thus, as early as my dissertation and its first follow-on project I was looking at how war impacts future war; how war participation affected the probability of subsequent war participation in the same state—that is reinforcement or repeated war behavior of the same country. With the use of a border variable, these two studies set the stage for studying how war/conflict behavior in one state would affect the war behavior of other states—that is the spatial diffusion of war/conflict. This evolved into the diffusion project with Ben Most: how involvement in large-scale military conflict (internal or external) affected the conflict behavior of other states. Because we needed to separate the effects of each, the work with Most was concerned both with looking at both positive and negative reinforcement (a war experience making a state more or less likely to engage in war again), as well as positive or negative spatial diffusion (making war more or less likely in other states). The work on diffusion pushed us to consider the geographic context as well as a set of conflict processes. Each state existed within a context of contiguous states or a geographic regional context. This necessitated the use of a dataset of borders between states, as well as the borders of remaining colonies and territories. Most and I created the first interstate border data set (see Starr/Most 1976), covering the 1945–65 period, looking not only at contiguous borders between states, but borders between states and territories, and territory-territory borders. A conflict process approach was required in hypothesizing why distance/proximity between states would be important for the spatial diffusion of war. One argument was that states that were close—that touched through contiguous borders—were more important or relevant to each other. We were thus identifying “relevant dyads” although we did not specifically use that term; and we may have been the first to use this concept in a major journal article. The geographic context was thus seen as an important factor in the ‘opportunity’ dimension of states. Box 1.2 Harvey Starr: Selected Honors, Awards, and Grants Phi Beta Kappa, 1967 New York State Herbert H. Lehman Fellowship, 1967 (declined) Yale University Fellowship, 1967–7 Indiana University, Summer Faculty Fellowship, Summer 1973, 1977 Center for International Policy Studies Faculty Research Seed and Supplementary Grants, January 1976, 1977 (Indiana University) Supplementary and Renewal Grants from the National Security Education Program New York University, 1981 National Science Foundation Research Grants, 1984, 1998; conference grant 2010 University of South Carolina, Research and Productive Scholarship Grant, 1996

38

1 Getting from Then to Now: A Personal Intellectual Autobiography

University of South Carolina, 1998 Russell Award for Research in Humanities and Social Sciences in recognition of outstanding research and scholarship Choice named Anarchy, Order And Integration one of its “Outstanding Academic Books,” 1998 Appointed to the National Advisory Board for the NSF supported Integrative Graduate Education and Research Training (IGERT) grant to the State University of New York at Buffalo for a multidisciplinary doctoral program in Geographic Information Science, 1999–2003 Distinguished Alumnus Award, Department of Political Science, University at Buffalo (first recipient; 2009) Panel on: “Inquiry, Logic And International Politics 25 Years On,” Annual Meeting of the American Political Science Association, August 2014, Washington DC 2015 Lifetime Achievement Award from the Conflict Processes Section of the American Political Science Association, “in recognition of scholarly contributions that have fundamentally improved the study of conflict processes.” Presidential Theme Panel: “On the Career Contributions of Harvey Starr,” International Studies Association Annual Meeting, March 2016, Atlanta. International Studies Association Section on Political Demography and Geography’s 2016–17 Distinguished Scholar (Myron Weiner Award), 2017

The border dataset we created pre-dated that of the Correlates of War project (COW). Once the COW data became available, we (including me, Most, and later Randy Siverson) abandoned our data for the far more extensive and detailed COW data. The work on diffusion with Randy Siverson, subsequent to the untimely death of Ben Most, returned to alliances, adding the strategic/political/alliance context to the geographic context in exploring factors that led to the growth and spread of war; see Siverson and Starr, “Opportunity, Willingness and the Diffusion of War, 1816– 1965” included in Part II (see also Starr 1991a). The opportunity and willingness framework/model was explicitly presented and used in the 1991 book with Siverson where the key factors were borders and alliances. Before turning to Opportunity and Willingness, it is important to note that the various papers of mine, and those co-authored with Most and then with Siverson spent a good deal of time in the conceptualization of ‘diffusion,’ and the ways of getting at its various meanings, forms, and processes before moving on to the actual research analyses. Zaryab Iqbal and I were similarly confronted by the same problem of conceptualizing “state failure” before we could proceed with our project; in particular avoiding definitions and conceptualizations which became tautologies when the factors used to define the concept were claimed to be the very factors in causing state failure.

1.6 Scholarship: Career Themes

39

Harvey at the 2015 Annual Meeting of the American Political Science Association, speaking after being presented the Conflict Processes Section Lifetime Achievement Award. Source From the author’s personal photo collection

A different way of looking at diffusion as demonstration effects or models, (along with some spatial contagion effects), was taken in my analysis of diffusion of democracy in the international system, as represented by “Democratic Dominoes: Diffusion Approaches to the Spread of Democracy in the International System” included in Part II of this volume; (also Starr/Lindborg 2003). Indeed, the initial impetus of the later Starr and Iqbal project on failed states was to test the proposition that state failure was ‘contagious.’ Spoiler alert: a chapter in Iqbal/Starr (2016) was devoted to the consequences of state failure. It investigated the question of diffusion, and while several of the elements related to state failure showed some levels of diffusion, state failure itself was not found to diffuse to contiguous neighbors or the local region. As I have noted throughout this essay, the study of borders, spatial diffusion, and the geographic element of context (opportunity!), have all been critical to my position that political scientists, and particularly IR scholars, need to pay more attention to spatiality and geographic factors in their analyses. Thus, in my own research agenda the possibility of a variety of diffusion effects was applied to war, interstate and intrastate conflict, transitions to democracy, and the occurrence of state failure.

40

1.6.2

1 Getting from Then to Now: A Personal Intellectual Autobiography

Opportunity and Willingness as Central to a Scholarly Agenda

While I had become familiar with the Sprouts’ work on different models linking geographic factors with behavior and decision making, the true origins of opportunity and willingness came during my first two years at Indiana, while I was developing the first syllabus for a new graduate seminar on “International Conflict.” Note that I did not use ‘war’ in the course title, as I wanted to begin the course with the basics of all social conflict and assigned a good deal of work by sociologists and psychologists (and some anthropologists). Note also that I was again reflecting the multidisciplinary grounding from graduate study. I initially thought about organizing the course by levels of analysis but found that such an approach artificially separated individuals and their identity groups, individual decision makers, small groups of decision makers, governments, and states, from the state system environment and the environments each of them provided. When thinking about organizing the course around disciplines, I again ran into artificial barriers without integrating mechanisms. I needed something that could help synthesize across levels of analysis and across disciplines. It was then I thought back to agent-structure models that placed some units of action within their environment, the interrelationships between agents/entities and their environment(s), and the impacts on outcomes.36 The opportunity and willingness framework or approach (abbreviated hereafter as O/W), drawing on the Sprouts, was an agent-structure approach. It was also all about ‘process’ in the interaction of the entity and environment. It promoted the use of multiple levels of analysis, each of which could provide context for human agents, as well as the dynamics of decision making and strategy (which was captured by willingness). O/W was first presented in my 1978 article which appeared in International Interactions (after it was rejected for different and rather strange reasons by several journals). Most scholars, however, were introduced to O/W when a somewhat revised version appeared as a chapter in Most and Starr, Inquiry, Logic and International Politics (1989). Most and Starr will be discussed more fully below, but let me just note here that O/W and the logic of inquiry arguments in Most and Starr are intimately connected. An introduction to the two key concepts is in order before a brief overview of O/ W (all of which draw from my 2017 discussion of the evolution and use of O/W).37 Opportunity is based on the Sprouts’ “environmental possibilism” which was developed as one alternative to the “environmental determinism” of early—generally Realist—geopolitical thinking. It refers primarily to the “possibility of The “agent-structure problem” is a complicated one, relating to ontological and epistemological debates. See Gil Friedman and Harvey Starr, Agency, Structure, and International Politics (1997) for a fuller discussion of this literature, and how we argue that the opportunity and willingness approach handles many of the issues and problems raised in that literature. 37 See also the ‘Preface’ to the 2015 re-issue of the Most and Starr book. 36

1.6 Scholarship: Career Themes

41

interaction.” Opportunity means that “conditions exist that permit sufficient interaction between states, that adequate capabilities and resources exist to allow certain kind of action to take place, and that decision makers perceive both the range of possible interactions and the extent of capabilities” (Russett/Starr 1981: 23). Important sources of possibility include geopolitical factors such as proximity and borders, as well as the various means by which humans manipulate this environment—especially technology (see also Most/Starr 1989: 30–31; Starr, 2013a, b). Opportunity also refers to the existence of capabilities that permit the creation of opportunities, capabilities which might permit and/or promote interaction. There are two related but distinct ways to look at levels of capabilities and possibilities. First, as Most and Starr (1989: 31) note, “some capability—technology, ideology or religion, form of government, manner of organizing people to some task, etc.— must be created/invented so as to be part of the range of possibilities available to at least some members of the international system.” The second level involves how the various capabilities or factors in the environment are distributed among states or other international actors. For example, nuclear weapons do exist in the post-World War II system, but not all states have them. Willingness, closely related to the Sprouts’ “cognitive behaviorism,” refers to “the choice (and process of choice) that is related to the selection of some behavioral option from a range of alternatives. Willingness thus refers to the willingness to choose (even if the choice is no action), and to employ available capabilities to further some policy option over others” (Most/Starr 1989: 23). That is to say that willingness deals with the motivations of people (Friedman and Starr’s focus on foreign policy decision-making elites) to take advantage of opportunities. Willingness deals with perceptions, emotions (e.g., fear, insecurity, revenge), and all of the factors related to the choice within the set of perceived available alternatives. While originally developed to organize thinking about international politics, O/ W has evolved into a guide for generating IR theory, developing research designs to study IR, and ways to evaluate those theories.38 It does this by showing how to synthesize and bring together apparently disparate hypotheses and evidence to bear —crossing a variety of analytic boundaries—and by pulling together what we know across levels of analysis, academic disciplines, and the sub-disciplines of political science. O/W compels scholars to cross, link, and synthesize levels of analysis— complementing theories built around levels of analysis, while at the same time moving them forward in order to deal with the complex causation they have to confront. This complex causation derives from the logical features of O/W, which regards opportunity and willingness as jointly necessary conditions for the occurrence of any event. A discussion of the characteristics of necessity and sufficiency as causal processes leads to the conclusion that not only does this joint necessity distinguish O/W from theoretical approaches that are deterministic, monocausal, or are

38

This section draws directly from Starr (2018).

42

1 Getting from Then to Now: A Personal Intellectual Autobiography

concerned only with either opportunity or willingness, but is the beginning of a logical story that demonstrates how this framework can deal with causal complexity. The joint necessity of opportunity and willingness, along with its two corollaries of ‘substitutability‘ and ‘nice laws,’ forces a researcher to more fully specify the logical and substantive structure of the theoretical statements under investigation. It also ensures that the research design is relevant to the theory and set of research hypotheses—such that there is a coherent relationship among the components of logic, theory, and method, as later stressed in Most/Starr (1989). At the end of the logical story of opportunity and willingness, it can be seen that O/W has moved well beyond an organizing principle and is a model of causal complexity of great potential. The two corollaries of substitutability and nice laws even more closely link O/W to the logic of inquiry developed in Most/Starr (1989) and demonstrate why its use has become widespread in the development of research design (in many cases using the terms and meanings of Most and Starr without attribution, indicating common usage in the field). The first corollary (Most/Starr 1989, Chap. 5) is substitutability. Substitutability refers to the existence of a set of alternative modes of response by which decision-makers could deal with some situation. Substitutability indicates that any single cause may have multiple effects, and that any single effect may have multiple causes. Different problems may lead to similar responses; a given problem may be dealt with in multiple ways. This may also be expressed as “equifinality” (Goertz/Mahoney 2005) or multiple paths to the same outcome.39 Substitutability lies at the heart of the use of agent-structure frameworks like opportunity and willingness, where the choices of agents are analyzed in regard to the contextual structures within which they exist. In Most and Starr we argue that opportunity and willingness can operationally occur or be made available in any number of alternative, non-unique ways. However, it is not simply that substitutable modes (or redundancy) are related to greater ‘complexity’ in the world—these modes actually make political behavior in a complex world more possible (for a full demonstration of this, see Cioffi-Revilla/Starr 1995). And, I have argued, this is a view of complexity which has not been adequately addressed by Realism, Neo-realism, nor most variants of transnational liberalism. The second corollary introduced by Most/Starr (1989) involved nice laws. The contextuality of O/W, along with its introduction of complex causal relationships, additionally suggests that it may be useful to expect that social laws may not always hold, operate, or apply empirically. Most and Starr argue that there are a variety of social laws, each of which is ‘true,’ but which should be expected to hold only under certain—perhaps very special—conditions. Thus, any one or more ‘contending’ theories may be true under different conditions or contingencies (i.e., nice laws). The Nice Laws concept is one way of arguing that the contexts of

Note that substitutability was the subject of a special issue of the Journal of Conflict Resolution: “Substitutability in Foreign Policy: Applications and Advances” (Palmer 2000).

39

1.6 Scholarship: Career Themes

43

opportunity must be as clearly and as fully specified as possible along with any other variables/factors/concepts used in research. Nice Laws also appealed to my desire to see how things fit together rather than pit explanations against each other in a zero-sum battle of constant critiques (which also often characterize Realist writings).

1.6.3

O/W and My Major Scholarly Projects

1.6.3.1

World Politics: The Menu for Choice

O/W would become one of the analytic principles found throughout the Russett and Starr undergraduate textbook, World Politics: The Menu for Choice. In the Spring of 1978 when my promotion and tenure file was moving through Indiana University (with every signal that it would be successful) I received a phone call from Bruce Russett. I was surprised but quite pleased that Bruce was asking me to come on board as a co-author of an undergraduate textbook for W.H. Freeman (publishers of Scientific American among other things). I was surprised because I knew Russett was working on a textbook when I was his TA for the undergrad IR course at Yale. But Russett’s many research projects and other responsibilities had led to work on textbook being put off, and he had come to the conclusion that he needed a co-author. Me!? The fact that he had such confidence in me, and felt that we would make a good fit, was quite a compliment. As I was planning for the 1978–79 sabbatical I would have after receiving tenure (returning to the University of Aberdeen as a Leverhulme Visiting Fellow in Politics), this seemed like perfect timing. This invitation was yet another element of Russett’s mentoring. Rather than my simply being given tasks, we sat down in New Haven and blocked out what the book would look like, and which of us would write the first draft of which chapters. The first chapter indicated the reader would find a volume organized around “levels of analysis and constraint.” It would introduce students to the Sprouts’ “entity-environment relationship” indicating how policymakers crafted decisions under constraint. This was done through a discussion of Russett’s idea of “the menu” and a section on opportunity and willingness; both reflecting the Sprouts. As noted in the 10th edition of the textbook (2013: 15–16): The person (entity or actor) who enters a restaurant is confronted by a gastronomical environment—the menu. The menu provides a number of behavioral opportunities, not determining the diner’s choice but constraining what is possible (pizza, lasagna, and linguini are possible in an Italian restaurant, but chicken chowmein and matzo ball soup generally are not). The menu also affects the probability of the diner’s choice through price, portion size, specials, and the restaurant’s reputation for certain dishes. In an Italian restaurant whose menu proclaims that it has served pizza since 1910 and offers over fifty varieties at low prices, a diner is most probably going to order a pizza. The restaurant,

44

1 Getting from Then to Now: A Personal Intellectual Autobiography however, offers other selections as well, and the probabilities they will be ordered are affected by how a diner sees those choices. Knowing a patron’s palate and resources, as well as the patron’s perception of the menu, permits us to analyze and predict his or her choice of entrée [emphasis added].

The menu analogy is helpful for understanding that the opportunities presented to international actors are constrained in various ways and that these constraints affect the willingness of decisionmakers to act. Constraints can be external or, as is so often the case, they can be self-imposed. For nearly three decades, the academic writings of former U.S. Secretary of State Henry Kissinger have stressed the domestic and international constraints on the foreign policy decisionmaker. The skilled diplomat, Kissinger has emphasized, understands these constraints and learns to work within them to achieve his or her desired aims. The skilled diplomat also realizes that the menu changes, and is therefore on the lookout for better choices as new selections become available. The first edition of Russett and Starr was published in 1981, the 10th edition appeared over 30 years later in 2013.40 With the 5th edition David Kinsella, who received his Ph.D. at Yale in 1993 under the direction of Russett, came on as the third author; with the 10th edition the order of authorship changed to Kinsella, Russett and Starr. While never one of the best-selling IR textbooks, it was an influential one. It was not unusual for a colleague to tell me or Russett that they used Russett and Starr to write their lectures while assigning a much simpler text to their undergrads. Indeed, this textbook was cited in many research articles— Google Scholar reports that across the years The Menu for Choice was cited almost 1000 times. The first edition not only introduced O/W and ‘Menu’ to undergrads, it was one of the first to link theory, methods, and research design in an undergrad text, interweaving levels of analysis with opportunity (contexts) and willingness (individual idiosyncrasies, psychology, and organizational role). It also contained one of the first undergraduate presentations of interdependence, collective goods, ecological/environmental issues regarding “limits to growth’’, the “tragedy of the commons” and related game theoretic analyses. Although the Russett and Starr textbook was written for undergrads, we also made sure to cite the most recently published works based on empirical evidence to support our examples or assertions. At this point, the text’s multidisciplinary nature should not surprise the reader. My interests in the international system—reflecting context, opportunity, and the interdependence of systems which is best reflected by collective goods—were also pulled together in my book Anarchy, Order, and Integration (1997a). This book is another example, along with the textbook, of my career-long project (which began

40

Note also the international editions: Indian second edition by Vakils Feffer & Simons, Ltd., Bombay, 1986; Italian third edition by Il Mulino; Taiwan fourth edition by Wu Nan Book Co.; Russian fifth edition by Svetlychor; Japanese sixth edition by Japan Uni Agency; seventh edition in English with Chinese introduction, Peking University Press, 2004; Chinese ninth edition. Impressive, but Russett and I are pretty sure we never saw any substantial royalties from these foreign editions.

1.6 Scholarship: Career Themes

45

in Russett’s graduate seminar on IR Theory) of demonstrating the weaknesses of “Realism” and how other approaches, and combinations of approaches, better explain the phenomena of international relations.41 I always enjoyed teaching 100- or 200-level IR courses with Russett and Starr, and I know that a small but loyal group of political scientists had success as well using the text across a range of institutions. I take great pride in producing a well-received and important undergrad text which represented many of the central themes of my research program. I am even more proud that I was also the co-author of a book that had a major impact on the teaching of IR/methods courses at the graduate level—Inquiry, Logic and International Politics. Box 1.3 Harvey Starr: Selected Professional Activities and Appointments President, 2013–14, International Studies Association, 2012–13 President, Peace Science Society (International), 2000–01 Vice President, American Political Science Association, 1995–96 Vice-president, International Studies Association, 2011–2012 President, Conflict Processes Section, American Political Science Association, 1992–1995. President, Midwest Section of the Peace Science Society (International), January 1978–April 1980 Editor, International Interactions, 1991–2000 Associate Editor, Journal of Politics, 2000–03 Associate Editor, Teaching Political Science, 1978–1981 Editorial Boards/Advisory Boards: • • • • • • • • •

American Political Science Review; International Studies Quarterly; International Interactions; Journal of Politics; Comparative Political Studies; Journal of Territorial And Maritime Studies; Mershon International Studies Review; Asian Journal of Peacebuilding; “Dilemmas in World Politics” Series, Westview Press NSF Activity:

• NSF Advisory Panel for Political Science, 1992–93 • SES Presidential Young Investigator Advisory Panel, 1990

41 See, for example the discussion of “substitutability” in Most/Starr 1989, especially the new Preface to the 2015 re-issue. Let me also note that Choice named Anarchy, Order, and Integration one of its “Outstanding Academic Books” in 1998.

46

1 Getting from Then to Now: A Personal Intellectual Autobiography

• NSF Panel for the Social and Behavioral Dimensions of National Security, Conflict, and Cooperation (NSCC) competition, “Conflict” sub-panel, December 2008 • Proposals reviewed for programs in: Political Science; Decision Science; Geography and Regional Science; Geography and Spatial Sciences; Economics; Law and Social Sciences; Methodology, Measurement and Statistics

Harvey at the 2014 Annual Meeting of the International Studies Association, where, as ISA President, he is awarding Bear Braumoeller the 2014 ISA Best Book Award for The Great Powers and the International System: Systemic Theory in Empirical Perspective. Source From the author’s personal photo collection

Other Grant Review Activity: • • • •

National Endowment for the Humanities; Israel Science Foundation; Canada Foundation for Innovation; Swiss National Science Foundation. Selected Conference/Convention Activity:

• Program Co-Chair, Midwest Political Science Association 1982 Annual Meeting

1.6 Scholarship: Career Themes

47

• Program Co-Chairman for the Conference, “Research Agendas for the Study of International Conflict,” the Annual Meeting of the Midwest Section of the Peace Science Society (International), Ann Arbor, 1978 • Program Co-Chairperson for the Conference, “New Directions in International Relations Teaching and Research,” the Annual Meeting of the International Studies Association Midwest Section, Bloomington, 1976 • Section Chair, “International Relations: National Security and Conflict Analysis,” American Political Science Association 1984 Annual Meeting • Program Chairman, Peace Studies Section of the International Studies Association for the Annual Meeting of the International Studies Association, St. Louis, 1977 • Conference Co-Organizer for the Conference, “American (Jewish) Humor in an Era of Ethnic Sensitivity and Cultural Competence: Intersections of Religion, Race, Ethnicity, and Gender in American Humor,” April 2018, University of South Carolina.

1.6.3.2

Inquiry, Logic and International Politics [ILIP]

The genesis of ILIP can be found in the borders/diffusion project that Ben Most and I began while he was still a graduate student at Indiana. Ben had taken both my IR Theory and International Conflict graduate seminars—unusual since his first field was comparative politics. I was not even on Ben’s dissertation committee, which was co-chaired by John Gillespie and David Collier (who moved to Berkeley and became a leading scholar in comparative methods as well as Latin American studies). Ben’s dissertation was on public policy in Argentina over the 1930–1970 period.42 As a scholar of policy and policy processes Ben was interested in the work on the diffusion of policy—across the American states, or in international systems (a true comparatist!). I had started to develop an interest in how large-scale armed conflict (war) in one state (or dyad) would affect subsequent war behavior in other states or dyads. This common interest in diffusion brought us together, and the collaboration began. We initially worked on the borders data set, which led to a series of articles looking at the spatial diffusion of war. But, even more importantly we ran into a variety of problems regarding research design—particularly at a time when geographers were only beginning to develop spatial statistics (unknown to us and most political scientists). Ben came at these 42

At the time of his death Ben had a draft of a manuscript which expanded and updated the dissertation to turn it into a book, especially expanding his analysis of policymaking processes. I directed the enterprise of turning this draft into a book (Most 1991).

48

1 Getting from Then to Now: A Personal Intellectual Autobiography

problems from a policy analysis perspective, I tried to set them in the context of a philosophy of science perspective—which I later always referred to as the “logic of inquiry.” We both realized there were logical inconsistencies in much of the IR literature—with flawed research design leading to non-cumulative research results, and research that simply made no sense. In working out both these problems and their solutions we presented conference papers which eventuated in three articles. We realized that with my article on O/W we had the basis for a book that we hoped could be used in graduate level courses on research design and logic of inquiry. A revised and updated version of my O/W article became Chap. 2 in ILIP (with an introductory chapter to be written much later). The three articles would become Chaps. 3, 4, and 5—on basic logic and research design, conceptualizing war, and foreign policy substitutability and nice laws, respectively. Another conference paper would form the basis of Chap. 6, followed by a conclusion (not yet drafted). Both O/W and a focus on war became running themes throughout the book. As noted, for many international relations scholars, ILIP was their introduction to opportunity and willingness. As one form of an agent-structure model, opportunity and willingness was well suited for helping ILIP discuss the running use of the study of war as a substantive example of the logic of inquiry arguments presented there. Such an agent-structure framework was also useful in providing a range of research questions that went beyond those depending on standard linear correlation and regression-type analyses. While a proper overview of ILIP would take a whole other article, let me try to summarize what Most and I were trying to do (see Starr 2015b for a fuller discussion of the content and impact of ILIP over its first 25 years in print). In ILIP Most and I introduced a set of elements that were claimed to be useful—indeed critical in several senses—for good model building, research design, and the assessment of the validity of systematic empirical inquiry. While explicit in our thinking about how to do “good” research that would more efficiently and profitably lead to “better” cumulation in the research of international relations scholars, we were less explicit in framing the book as a truly “critical” commentary on the research practices of many IR scholars. In retrospect I have argued that ILIP is a significant piece of critical analysis. Drawing on our work on diffusion and a critical reading of the IR literature, we addressed a set of impediments to cumulation (as conceptualized and discussed at length by Zinnes; e.g. 1980) which included a self-critical appraisal of the failure of quantitative scholars to match theory and logic to research design and method (see also Starr 2005a, b). We identified three elements—theory, logic, and method– as the components of the “research triad,” which served as the core of our presentation. We argued that “scholars need to recognize the existence of a research triad... and that each leg of this triad is critical for advancing our knowledge of international phenomena...” (ILIP:2). Scholars were likened to ‘jugglers’ as “each element of the triad needs to be held in the air at the same time in a complex set of interrelationships, indicating that for the juggler to be successful, all of the balls (elements of the triad) must be kept going simultaneously” (ILIP:10).

1.6 Scholarship: Career Themes

49

The obvious position students of research design must support is that the methods selected by any researcher must be appropriate to the questions that are being investigated. Those questions, in turn, must be theory-driven. Thus, any researcher must begin with well-grounded and well-specified theory. Thus, ILIP stresses (in Chap. 1) the dynamic feedback loops that exist among theory, research design, and findings, as each informs and modifies the other. Theory thus not only affects the design and the research product but is itself continually modified and updated by the research process and the results of that process, in a dynamic combination of induction and deduction. As the research process unfolds, and theory is modified, the researcher may be led to investigate additional phenomena originally omitted from the study. This view means that theory is central to the enterprise of empirical research. ILIP also presents a set of elements that comprise what we called our “critical logic”: the form of the relationship under study; matching the logic of theory/ models to the logic of research design; development of process models; broader versions of the agent-structure problem; identifying the appropriate units of analysis; case selection; and such vital linking and synthesizing concepts as ‘substitutability‘ and ‘nice laws.’ All of these issues have become part of broader dialogues, debates, and progressive Lakatos-like developments in the study of international politics. The argument in ILIP that greater attention needed to be paid to the form of the relationship which had to reflect the theoretical model being used and had to be logically connected to the research design. This was most clearly reflected in Chap. 3’s discussion of necessity and necessary causal relationships. Note that this chapter was reprinted in the Gary Goertz and Harvey Starr edited volume, Necessary Conditions: Theory, Methodology and Applications (2003). This edited volume was the main product of work that I co-authored with Goertz on the nature of causality, necessary causation, sufficient causation, and the problems of research design that would arise if the differences between necessity and sufficiency were not recognized and taken into account. Much of our work began with the analyses found in ILIP. The Goertz and Starr project moved beyond IR, as Most and I hoped would be the case with ILIP. Both ILIP and the Goertz and Starr projects fit comfortably as complements to Charles Ragin’s various quantitative comparative analysis (QCA) models.43 When I was an assistant professor at Indiana University beginning the diffusion project with Most (and the methodology papers as well), Ragin was an assistant professor in Sociology. I was unaware of Ragin’s 1987 book The Comparative Method while I was writing the introduction and conclusion, converting our 43

It should be noted that the Goertz and Starr volume has been used extensively in Comparative Politics graduate courses and in the scholarly research associated with the American Political Science Association’s Section on Qualitative and Multi-Method Research. Discussions of necessary conditions as well as references to Goertz and Starr have frequently appeared in the Qualitative and Multi-Methods Newsletter and at the annual summer Institute for Qualitative and Multi-Method Research.

50

1 Getting from Then to Now: A Personal Intellectual Autobiography

conference paper into chapter six, and re-writing and linking the other chapters of ILIP after Most’s death. Similarly, Ragin who left Indiana for Northwestern in 1981, was not aware of the Most and Starr articles that would become ILIP. Neither Ragin or I can identify where the commonalities in our work came from (although we were all associated with the Gillespie and Zinnes Ford Foundation supported Center for International Policy Studies at Indiana), but we have discussed both the common and complementary foundational elements in his work and ILIP.

1.6.3.3

Two-Level Security Management Project

The Two-Level Security Management Project, developed with Marc Simon, neatly pulls together major themes already noted: conflict processes, context, O/W, substitutability, and the flexibility of using multiple methodologies to get at the research questions being investigated. The general concern of this project was the viability or security of governments as they make decisions within a two-level game of internal and external conflict. It deals with willingness in terms of the choices made by decisionmakers regarding internal security vis-à-vis external security. The project is set within the context of my previous theoretical work on opportunity, willingness, and substitutability and substantive work on the diffusion of violent conflict. It was focused on the investigation of the study of conflict processes that cut across levels of analysis, and on the broader relationships between internal and external conflict, including the more specific relationship between revolution and war.44 As a student of conflict processes, I wrote a series of conference papers (culminating in my 1994 article) looking at the commonalities between revolution and war (which also sent me back to both old and new material written by sociologists).45 I argued that social conflict at any level could be seen as a mechanism by which order is established, challenged and re-established. This view could be found in the two literatures presenting theories of revolution as well as that developed around the concept of “general,” “hegemonic,” or “system change” war. Both general war and revolution could be conceptualized as agents of change through the organized use of violence. What I have called the “common logic” of internal and external conflict involved a common concern of governmental decisionmakers with the “viability” of the government, from either internal or external threat, and how the government responds to that threat, given the government’s “capacity” or resources. The modeling of this linkage is based on four components of a common logic presented in Chapter 5 of ILIP (and as refined in collaboration with Mike

This work also includes the study of “protracted social conflict,” as exemplified in Starr (1999). It was during this period I initiated a correspondence with Charles Tilly concerning the various areas of overlap of our work, the differences as well as complementarities. I very much value this experience, and only wish it had been more extensive.

44 45

1.6 Scholarship: Career Themes

51

McGinnis at Indiana University).46 Each of these components affects the perceptions that decisionmakers have of a state’s viability (security), and its capacity (resources) to defend itself against internal opposition or external threat: – – – –

Ci, a state’s (state i) defense capacity Ri, the external risks the state faces in the international system Si, the strength of the government in the face of domestic opposition Ti, the threat the government faces from domestic sources.

Factors such as “defense capacity” and “strength of government” reflect the finite resources available to decisionmakers as well as the substitutable choices available to them as they seek to achieve and maintain state viability. While the papers in this project also employed empirical analyses, the primary method for evaluating the hypotheses generated by the “common logic” were investigated through the use of computer simulations—which looked at the dynamics of the two-level interaction of managing domestic and external security. These simulations (developed by Marc Simon whose expertise, based on his background in mathematics, was lightyears beyond my own) were used to investigate such diverse topics as the rise and decline of systemic hegemons and challengers as well as the survival of new and/or unstable democracies. For example, ‘endangered’ democracies, with larger democratic allies in a system (or region), could rely on those other democracies to deal with the “external risks” faced by the endangered democracy. This, in turn, permitted the government of an endangered democracy to recover security through increased domestic spending and increased legitimacy, as well as buying off domestic threats rather than spending resources to deter them.47 Quick note: My work on two-level games, the relationship between internal and external conflict, and especially “protracted social conflict,” have been important in creating a foundation for my most recent work regarding IR and linguistics. This work investigates the importance of language and linguistic factors in ongoing ethnic conflicts – and indeed even the definition and conceptualization of ‘ethnicity’. I have also been addressing the synergies between political science and linguistics in studying the role and impact of language in conflict processes across all levels of analysis.

1.6.3.4

GIS Project on Reconceptualizing the Nature of Borders

One additional research project based on geography, space, context, and O/W needs to be mentioned: a project using GIS (Geographic Information Systems) to investigate more deeply the ‘nature’ of contiguous borders between states. My interests and work on geographic context and diffusion, and thus the study of borders has been described above. “On Geopolitics: Spaces and Places,” included in 46

See Starr/McGinnis (1992); Simon et al. (1994). See: Simon/Starr (1996, 1997, 2000); Starr/Simon (2017).

47

52

1 Getting from Then to Now: A Personal Intellectual Autobiography

Part II is one attempt to pull this work together. The GIS project merits a separate discussion because it highlights the multidisciplinarity of my work, and how it can draw upon both O/W and the lessons from the Most and Starr work on the logic of inquiry. It exemplifies the latter’s call to more deeply and fully specify one’s concepts and their measurement, and to see how so-called ‘competing’ theories or hypotheses can both be true under different conditions. This project sprung from a set of conversations I had with a colleague at South Carolina who was, at that time, chair of chair of the Geography department. He also ran the Geography department’s center for ESRI (Environmental Systems Research Institute) which produced GIS software such as ARC/INFO. After receiving a University seed grant to do a preliminary study (Starr/Bain 1995), I submitted an NSF proposal which was co-funded by the Political Science and Geography Programs—I was told it was the first proposal to receive such dual funding. The NSF funding allowed me to hire two Geography doctoral students to translate my research design into the proper GIS analyses of the Digital Chart of the World (produced by ESRI for the Defense Mapping Agency). These two grants led to a number of conference papers and presentations and six journal articles—two in Geography journals and four in IR journals, including major empirical analyses co-authored with Dale Thomas.48 “Opportunity, Willingness and Geographic Information Systems” has been included in Part II to present the project in detail. As noted earlier, the “relevance” of states to one another can be operationalized by their interactions and the opportunities for such interactions. Generally (with the exception of major powers) interactions are increased with greater proximity, with contiguous land borders providing the closest indicator of proximity (reflecting opportunity). Borders also are important for a variety of social, political, psychological, and economic reasons—and have a special salience (reflecting willingness). Most studies of borders simply used the number of land borders a state might have, using an ‘on/off’ measure of whether or not a dyad had a contiguous land border. The GIS project looked more specifically at measures of “ease of interaction” and the ‘salience’ of any sector of a border or summary measures of that border. The character of this work may be illustrated by concluding comments in the 2005 International Studies Quarterly article by Starr and Thomas (136–37): Borders matter. Proximity remains an important conceptual variable in the study of major power general war, the diffusion of international conflict, and crisis behavior. Whether or not states share borders has been a principal means of operationalizing proximity. However, a shared contiguous land border may not adequately reflect the expected underlying behavior. Two different views on the relationship between contiguity and conflict from the literature were presented– each view, however, represented a linear (positive or negative) relationship between them. We proposed a curvilinear relationship, with the low occurrence of conflict at both the lowest and highest levels of ease of interaction (opportunity) and salience

48

See Starr/Bain (1995), Starr/Thomas (2002, 2005) and Starr (2005a) for Political Science articles. The two Geography articles were Starr (2001) and (2002).

1.6 Scholarship: Career Themes

53

(willingness). In finding support for our hypothesis we have made two significant contributions to our understanding of the political geography of conflict and cooperation. We have examined and demonstrated the utility of moving beyond a simple ‘on-off’ dichotomous view of contiguous land borders to examine the terrain and human activity along shared border areas. Secondly, we have demonstrated that the relationship between borders/contiguity and conflict behavior is non-linear. Neither contribution could have been made using previously collected data on contiguity which simply noted whether or not two states shared a land border. These results could only be possible with the type of data provided by the GIS project on the nature of borders. Such results provide evidence that scholars should be moving beyond the simple dichotomous ‘on-off’ view of contiguous land borders to look at the nature of shared land borders. Failure to look beyond the presence/absence of shared borders will lead to model misspecification and an overall loss of model power. A border is not a border is not a border. Territory, per se is neither a necessary nor sufficient reason for conflict (a la one traditional view of territory as the causes/stakes of conflict, exemplified by Vasquez but also such scholars as Goertz and Diehl, e.g. 1992), nor does it automatically create greater opportunity for conflict (a la Most and Starr, e.g. 1980). Some types of borders are dramatically less likely to lead to military conflict than others, while some types are more likely to lead to cooperative activity, and scholars need to account for these differences. Additionally, such analyses need to take place within a broader research context in which space and spatiality play a larger role – one co-equal with that of time. Ease of interaction and salience represent not only opportunity and willingness but also variation across both time-space and cost-space. Our findings highlight the importance of incorporating spatial components into the theory and methodology of our research design.

1.6.3.5

Democracy and the Democratic Peace

While many scholars will know me by work related to the democratic peace, my work on democracy was not conceived or held together by the framework of a “research project,” but followed from my interest in conflict processes. As with my teaching interests in international order (and thus international law), my concern with democracy and the democratic peace came from a logical need to see the flip side of the study of conflict. That is, I was concerned with what conditions, factors, processes, and mechanisms led to the escalation of social conflict and perhaps to the threat and/or the actual use of force—and which ones led to de-escalation and the ‘peaceful’ management of conflict. How, and in what ways, was order established and maintained in the ‘anarchic’ international system? This perspective stemmed both from a conflict processes approach, integration theory, and to my excellent introduction to democratic theory in the Yale seminar taught by Robert Dahl (how could it be anything but excellent?).

54

1 Getting from Then to Now: A Personal Intellectual Autobiography

And, as noted, my first major pieces on this topic were generated by questions of the diffusion of democracy. My firm grounding in Deutschian integration literature (as well as the work of Ernst Haas), and how it informed my use of interaction opportunity models, was another component of looking at democracy in the international system. Indeed, in 1991 I attended Bruce Bueno de Mesquita’s 1991 summer Institute at Stanford on rational choice/formal modeling. During that institute I realized that I could make a logical argument demonstrating that the democratic peace was one form of Karl Deutch’s broader theory of integration (especially his concept of pluralistic security communities). This argument appeared in Starr (1997b) (“Democracy and Integration,” also included in Part II). As described in the abstract to Starr (1992): In the spirit of Lakatosian cumulativeness looking for explanations with excess empirical content—this commentary supports one explanation of the ‘why democracies do not fight democracies’ question. The model supported is an expected utility formulation by Bueno de Mesquita & Lalman based on the logical relationships between states which are ‘doves’ and ‘non-doves’. The same explanation for democracy-to-democracy peace provided by the Bueno de MesquitaLalman analytics, based on the ability to 'separate' states into doves and non-doves, can be used to explain the linkages between integration and the Deutschian concept of ‘security community.’ This argument was an important component of my discussion of the place of the democratic peace in an interdependent but anarchic world system (e.g. Starr 1997a, 2006a, 2013a). It also appears in the Russett and Starr chapter, “From Democratic Peace to Kantian Peace: Democracy and Conflict in the International System” (2000). This became an oft cited piece, as it summarized and evaluated the substantial literature on the democratic peace that had appeared by the end of the 1990s.

1.6.3.6

Decision-Making, Cognitive Behaviorism, and Willingness: The Kissinger Project

While most of my work within the O/W agent-structure approach has dealt with ‘structure’ and the opportunities provided by various overlapping contexts, I have always argued that the study of ‘agents’ is just as important for a full understanding of the phenomena of international politics, as both opportunity and willingness are necessary components of any real-world behavior or activity. My position on willingness being as important (or perhaps even more important) as opportunity is perhaps most fully articulated in Friedman/Starr (1997), and most compellingly presented in Chap. 6 of ILIP where Most and I used simulations to demonstrate that war occurrence was more sensitive to small changes in willingness than changes in system structure-based opportunities. The importance of willingness is also found in every edition of the Russett and Starr textbook. Especially in the first editions, the textbook provides discussions of models of decision making and decision makers, not only as they sit within the context of the international system, but their own

1.6 Scholarship: Career Themes

55

societies, and governmental structures, particularly their specific roles within the government. But the decision-making processes and choices of individuals are also affected by the idiosyncratic factors of any individual; factors that shape how that individual sees the world. As noted in the first edition of Russett and Starr, “the psychological environment limits the menu just as the other environments we studied limit what is possible and probable” (1981: 295). We noted two basic assumptions—that foreign policy is made and implemented by people, and that individuals can indeed make a difference in the foreign policy process. One alternative to geographic determinism provided by the Sprouts, as noted, was “cognitive behaviorism,” meaning that the possibilities provided by the environment had to be seen and understood, thus affecting the probability that any alternative choice would be selected. Some of my earliest writing on alliances and the foreign policy process noted that this view was not new, that the Sprouts and such scholars as Snyder et al. (1962) were stressing that decision makers could only work with the world as it existed in their heads, which provided them with their “definition of the situation.” These ideas on how decision makers defined their views of the world, their operational codes, their styles of behavior, and personalities were all important components of ‘willingness’ (and, again, also constituted critiques of Realism). My desire to use these ideas to investigate willingness more closely led me to my project on Henry Kissinger which I started in the mid-1970s, resulting in a set of articles and book chapters, and my 1984 book. The model for this study was Ole Holsti’s (1962) classic article on the operational code of John Foster Dulles, and his image of the enemy. Holsti demonstrated that Dulles had a “closed image,” within an “inherent bad faith” model of the Soviet Union, and that this image shaped his policies and behavior. Indeed, I saw the Kissinger study explicitly as a replication of Holsti’s work, looking at someone who held the same government position in dealing with the Soviet Union during the Cold War, but whose background, life experiences, and policies looked so different. I wanted to see if these differences were related to differences in the ‘openness’ of the image Kissinger held of the USSR (as well as the PRC): that is, whether the image was open to change given new information, or ‘closed’ as was Dulle’s image of the USSR. I used Holsti’s design to first identify Kissinger’s operational code. Holsti generously provided a copy of his detailed codebook for looking at writings made prior to taking office to generate an operational code; as well as the specific method of content analysis to be made on speeches, interviews, and writings made during Dulles’ tenure as Secretary of State.49 I quickly learned that I could not jump into the content analysis of Kissinger’s words as National Security Advisor/Secretary of State, or even the operational code prior to these positions, without understanding his life and how he used words. Thus, I first had to look at all the extant biographical and psychobiographical writing about Kissinger to get a firm grounding of those things that helped make an

49

I also made extensive use of Stephen Walker’s (1977) article on Kissinger’s operational code; it too used Holsti’s work as a template.

56

1 Getting from Then to Now: A Personal Intellectual Autobiography

operational code analysis possible. The operational code consisted of five broad ‘philosophical’ and five ‘instrumental’ questions.50 I had one major advantage over the Dulles study, in that I could then turn to the extensive literature of Kissinger’s academic writings on international politics, starting with his senior thesis at Harvard, and his doctoral dissertation which became A World Restored (1957), and moving on to the books and articles written while a faculty member at Harvard. Of these, I found many of the articles to be the most important, as Kissinger the academic was directing his writing to policy makers as policy prescriptions or arguments as to how policy should be made. It was fascinating to see how the method and substance of Nixonian foreign policy corresponded to the arguments in these academic works. These works were vital in developing answers to the broad questions posed by the operational code framework, and the specific coding procedure needed to perform the Evaluative Assertion Analysis type of content analysis (developed by the psychologist Charles Osgood) that would later be performed on Kissinger’s words while in office. As one might imagine this project led to a substantial number of articles, conference papers, and invited talks. I’ll just comment on one article, because it reflects themes developed above. At a conference panel (in early 1980 I believe), another participant was Doug Stuart of the University of Southern California (or, as we say at South Carolina, “the other USC”). He was also using Holsti’s work as a model for his study of John Kennedy. We got to talking, and the result was an article published in Political Psychology in 1981, “The ‘Inherent Bad Faith Model’ Reconsidered: Dulles, Kennedy, and Kissinger.” This article is a useful example of the importance of attending conferences, meeting people with similar interests, having people learn about one’s ongoing projects and you learning about theirs, and —as a “social enterprise”—building on each other’s work (here, both Holsti’s and Stuart’s). The content analyses on the openness of images of the Soviet Union showed that Dulles’ images were the most closed and inflexible, that Kissinger’s were the most open and flexible, with Kennedy almost exactly in between. The two methodologies —operational code (more ‘qualitative’) and content analysis (more quantitative’) were both necessary and complemented each other very well. But the article’s most interesting discussion was about how the words of these decisionmakers changed with different audiences, in different historical contexts, and different occasions. I had assumed at the time that I would continue this research program, because it fit with earlier work and findings on willingness, but I kept putting it off for projects dealing with the democratic peace and then, the Two-Level Security Management Project, and then the GIS project on the nature of borders. Obviously, I never did get back to this area of research, but it gave me confidence that there were a range of methodologies—both ‘qualitative’ and As examples, the first philosophical question: “What is the ‘essential’ nature of political life? Is the political universe essentially one of harmony or conflict? What is the fundamental character of one’s political opponents?” The third instrumental question: “How are the risks of political action calculated, controlled, and accepted?”

50

1.6 Scholarship: Career Themes

57

‘quantitative’—for getting at a variety of the psychological components of willingness in a systematic and replicable manner. I was happy to get at least some exposure to one form of systematic content analysis that produced quantitative data; (I used an early form of content analysis, one that is like a horse and buggy compared to the supersonic jets of the computational linguistics and discourse analysis used by both contemporary linguists and political scientists). But I also made good use of biography, psychobiography, and the systematic reading of Kissinger’s academic works to tease out his operational code. One of my most cited articles from this project (Starr 1980), “The Kissinger Years: Studying Individuals and Foreign Policy,” is included in Part II. It focused on the utility of memoirs to gain ‘access’ to the thinking and foreign policy decision making processes of political leaders. I have additionally used this project in writing about the logic of inquiry and in my graduate courses as an example of the difficulties of defining what exactly is a “case study,” as well as the complexity of trying to distinguish between ‘quantitative’ and ‘qualitative’ studies (a distinction that, in the end, I find to be of little value). Was the Kissinger Project a case study? Well, there was only one subject— Henry Kissinger. Was that one case? There were two sets of analysis, Kissinger’s images of the Soviet Union and his images of China. Did that mean the project was composed of two comparative cases? But the documents that were content analyzed were created during the period Kissinger was National Security Advisor/Secretary of State, from 1969 to 1977. So, this was time series data—and did the words of Kissinger captured in these documents represent different cases? Was each document a case? Was each of the almost 4000 “evaluative assertions” found in these documents ‘cases,’ aggregated for quantitative analysis? There were several analyses of different groups of documents (and thus assertions); was each of these a ‘case’—such as looking at different types of documents meant for different audiences (e.g., Congressional testimony or media interviews)? When I would teach this topic I had to go back to the whole issue of case selection and the primacy of theory: exactly what questions were being asked/ investigated to get at the empirical linkage between one’s theory or model or hypothesis and traces of what has actually occurred in the world. In turn such questions would force a researcher to return to the issues of validity, reliability, and ultimately replicability. Doing so only revealed the blurred distinctions between what is qualitative and quantitative (which would require too much space to go into here). Let me also note a related lesson to be learned in developing case studies. In Iqbal/Starr (2016) I developed case studies of state failure for each country in a series of matched-case dyads. As each research source that I used to create the case study was longer and more detailed, the characterizations that could be made, along with the conclusions that could be drawn changed! This only reinforced my observations that data—whether considered ‘qualitative’ or ‘quantitative’—were always required to meet the full range of criteria for good research design (e.g., beginning with the criteria noted above such as validity and reliability). Similarly, the issue of how fine-grained the data (or the cases selected) might be, reflects the same research problems raised by how fine-grained a grid for the use of GIS might

58

1 Getting from Then to Now: A Personal Intellectual Autobiography

be, or even how large (or small) time lags might be in looking for causal effects. And, always, in what ways does one’s theory justify the choices a researcher might make.

1.6.4

An Overview of My Research Process

This last section in the discussion of the Kissinger project provides a useful segue to one final commentary on how I have approached my research. As I have alluded to above, the undergraduate experience is usually about the student being a consumer of knowledge, but for me the foundation for how I would approach research, the research process, and the nature of knowledge originated in my undergraduate training. The Honors courses I took as a junior and senior were all about understanding the research process and the principles of good research design. Those principles, in turn, emerged out of understanding the philosophy of science that was concerned with the nature of theory and the sorts of generalizations we can make (a spectrum that runs from, for example, Kuhn to Popper); concepts and conceptualization; the nature of causality; and the nature of evidence (especially important to understanding the scientific method). I feel these courses began my thinking that eventuated not only in incorporating theory, methods, and findings from other disciplines, but also in my use of a wide array of research approaches and methodologies. One of the key lessons that I took away from these courses, but of course much more fully developed in grad school, was that the principles of good research design apply no matter what approach or methodology is used, and no matter what developments or changes have occurred in research technologies, in data collection, or analytical tools (e.g. the development of spatial regression techniques; see Darmofal 2015). All of these issues were later stressed in ILIP, and especially the primacy of theory to guide one’s research design and choice of methodological tools. I summarized much of this in the new Preface to ILIP in the re-issued 2015 edition. There I noted (p. xx) that Most and I: wanted to exhort researchers to look carefully at their theory and the research questions generated, and apply only those methods of analysis appropriate to the broader issues/ questions that concerned the researcher. To do so meant, and still means, that we should be expansive and inclusive in terms of the items we include in our methodological toolboxes. In turn this means that the methods we choose will reflect the theory used, the aims of the research (from description and association, to causation, explanation, and/or prediction), and the nature of the causality involved, rather than tailor our theory and the aims of research to the methods we want to use. Our methods should not be the proverbial “hammer in search of a nail,” nor should we act as the child with the hammer who discovers that everything needs to be pounded.51

51

The way in which Most and I thought about these logic of inquiry issues was very similar. I do not want to undervalue his impact on our work or my thinking.

1.7 Concluding Thoughts

1.7 1.7.1

59

Concluding Thoughts “I Get By With a Little Help From My Friends”52

In the discussion of my key questions about theory, one was “where does theory come from?” I answered, ‘everywhere,’ which included the range of extant literatures not only in political science but a variety of other disciplines. This point was also central to the concern in ILIP with cumulation in the enterprise of studying empirical phenomena. These perspectives are, in turn, vital in understanding that science is a social enterprise. Both imply that all research (please let me know about possible exceptions) flows out of previous work—previous speculations, concepts, more formal hypotheses, frameworks, theories, typologies, data sets, and research projects. But science as a social enterprise is also central to the second major question I would tell students to ask about theory: how do we go about evaluating and comparing the explanatory/causal claims of theory? My answer would always return to the research processes reflected in “the scientific method.” The norms of scientists require them to be explicit and transparent about all the elements in, and stages of, the research process so that the research is capable of being evaluated (and in some way replicated) by other scientists. Based on this process, and the evidence presented, which theory or hypothesis appears to have, in John Dewey’s terms, greater ‘warrant’—a question that would be considered by other scholars and investigators. While we all, as social scientists, should understand this, we have generally not done a very good job in helping laymen understand—that those who practice science are actively asking to be proven wrong by the critiques and research of other scientists. Science is, indeed, a social enterprise. Understanding where theory comes from and how that theory can be evaluated flows from the community of scientists. As such, no matter how much my knack for synthesis is just something about how my brain is wired, so much of my scholarship has depended on those who mentored me, taught me, the literatures to which I was exposed or discovered on my own, those students that I, in turn, taught, as well as departmental colleagues and scholarly colleagues both nationally and internationally. And, more specifically on the group of exceptional scholars with whom I have had the privilege of working with on research projects and co-authoring. Some of these individuals have been noted above as colleagues in major research projects with multiple co-authored works, others have only one or possibly two items on which we worked together. Some have started out as grad students who I have taught and/or supervised. Examples, in temporal order: Ben Most (IU), Chuck Ostrom (IU), Maria Papadakis (IU), Phil Schrodt (IU), Marc Simon (IU), Will Bain (USC), Jeffrey Morton (USC), Gil Friedman (USC), Dale Thomas (USC), Christina 52

This is the title, taken from the classic Beatles song, of one of the most influential early articles on collective goods in international relations by Frohlich/Oppenheimer (1970). I have found it appropriate in so many areas of IR; see especially Iqbal/Starr (2016) and Starr (1997a).

60

1 Getting from Then to Now: A Personal Intellectual Autobiography

Lindborg (USC), Roger Chi-Feng Liu (USC).53 Most of the co-authors who were departmental colleagues or colleagues elsewhere have been noted above. An alphabetical list would of such faculty colleagues would include: Claudio Cioffi-Revilla, Stan Dubinsky, John Freeman, Gary Goertz, Jeff Hart, Frank Hoole, Zaryab Iqbal, Dave Kinsella, Michael McGinnis, Randy Siverson, Douglas Stuart, Jack Sullivan, and Dina Zinnes. I think it is important to note these lists include some of the most iconoclastic and critical observers of standard/cookbook approaches to methods and the logic of inquiry; e.g. Ben Most, Phil Schrodt, Gil Friedman, Gary Goertz, and Dina Zinnes. To all of you—my great respect and thanks. Before concluding this section I’d like to say a few words more about two people who have had a career-long impact on my work—one who has been central to the academic voyage described above, and one who has only been mentioned a few times. I call them “the two Bruces.” I have discussed the impact of Bruce Russett, as the mentor at many points of my training and development as a scholar (although I have probably omitted as many episodes of his wisdom and assistance as I have presented). Simply, without his presence at key points in my career, I am not sure if that career would have been the same, and certainly not as successful. Much less well known has been Bruce Bueno de Mesquita’s (BdM) influence. We first met around 1974–75 at one or more conferences (when we both were regularly attending APSA, the Midwest, ISA, and Peace Science). One month younger than I was, we both received our Ph.Ds in 1971. Our initial common interest was in coalitions—my dissertation work on the dynamics of how wartime coalitions divided spoils (and losses), and his on a theory of how coalitions work, looking at the party-coalition process in the Indian Parliament (Bueno de Mesquita 1975). A secondary interest was Indian politics and culture, an area in which I took both undergraduate and graduate level courses. Interestingly, both Ben Most and BdM started off as Comparative Politics scholars (I know I converted the former, but not the latter!). BdM and I met frequently at conferences in various capacities and followed and commented on each other’s work. I always benefited from our conversations, his comments on my work, and advice in general – which would give me a different perspective and help me think through my research strategies (as well as commiserations in regard to departmental administrative duties).

53 There is also an inevitable, and tragic downside—the loss of friends and colleagues much too early and wondering about what important future work was lost due to their sudden deaths. Ben Most died at age 38 of heart failure (while in training for his first marathon which he planned to run with his brother in Philadelphia). Gil Friedman died at 42, unexpectedly after a short bout with cancer. His career at Tel Aviv University was just moving into its most productive phase when he passed away. At the time Ben died he was just getting a promising grad student through his dissertation proposal defense at Iowa—Steve Poe. I then served on Steve’s dissertation committee after Ben’s death. Steve went on to establish his place as a major scholar in the study of international human rights while at the University of North Texas, editing ISQ, and winning innumerable teaching awards. He too succumbed to a heart attack in his forties. While I have seen myself as a grandson of Deutsch, I always thought of Steve in the same light.

1.7 Concluding Thoughts

61

Group photo taken after the Presidential Theme Panel: “On the Career Contributions of Harvey Starr,” International Studies Association Annual Meeting, March 2016, Atlanta. The group includes the panel participants and some of the audience—all good friends and colleagues. From left to right: Zaryab Iqbal, Shannon Blanton, Marc Simon, Harvey Starr, Bruce Bueno de Mesquita, Scott Gates, Sami Baroudi, Bruce Russett, and Roger Chifeng Liu. Source From the author’s personal photo collection

I was always astounded by the energy, productivity, scope of activities (and let’s face it—the sheer brilliance) of the two Bruces. I learned to be less hard on myself when I finally understood that we were wired differently, and that I could watch one (or two) NFL games on a Sunday and not lecture myself that Russett would have written an article in that time. BdM has an extraordinary range of expertise— knowledge of history, command of languages, methodological expertise in game theory and formal modeling, and ultimately one of the most successful predictive models (and consultancies) in political science. Another lesson learned (and this will come across with an arrogance not intended) was that the two Bruces occupied the same place in an IR Hall of Fame as did, say, Hank Aaron or Willie Mays or Ted Williams in the Baseball Hall of Fame. While I too have come to consider myself a member of the IR HOF, it was more on the relative level of a Yogi Berra or Duke Snider or Harmon Killebrew (or some might say more accurately Ralph Kiner or Ryne Sandberg—look ‘em up, people). I would just like to take this opportunity publicly to acknowledge my special thanks to the two Bruces.

62

1.7.2

1 Getting from Then to Now: A Personal Intellectual Autobiography

Back to Intellectual Autobiography and Trajectories

Part of the biothat I have recently written for the “Who We Are” section of the web page for the Dubinsky and Starr consultancy ConflictAnalytiX (Dynamic Political and Ethnolinguistic Analysis of Global Conflict Systems and Particular Conflict Cases) includes this description of my research interests, directed toward potential clients: His broad research interests include theories and methods in the study of international relations, war and international conflict, geopolitics and diffusion analyses, and domestic influences on foreign policy. His primary research focus has been on the study of conflict processes—including the causes of conflict, the consequences of conflict, the spread of conflict, and the relationships between conflict within states and between states, and between the governments of states and internal opposition groups to state governments– including civil war, and “protracted social conflict” among ethnic groups, and between such groups and governments. Also related is his most recent research on the causes and consequences of failed states. His overall theoretical framework of “opportunity and willingness” has led him to study decision making and psychological approaches, as well as the contexts within which the decision making process takes place, especially geo-spatial and geopolitical studies. His work includes the use of statistical methods, game theory and formal models, simulation, and GIS (geographic information systems), as well as case studies. I think this paragraph basically captures my general areas of interest and methods, but it does not capture what I have tried to present in this essay on how I got there (in the same way that a review of my CV would be quite incomplete). I hope that saying more about the journey—early influences, the choices made, the roads taken and those avoided (e.g., art, architecture, law, or not pursuing willingness after the Kissinger project), and the lines connecting the evolution of research projects moving out of, or branching off of earlier ones—has helped make sense of my scholarly journey. I would like to end almost where I began. In his contribution to the first Intellectual Trajectories volume, the sociologist Kai Erickson (and son of the famed psychologist Erik Erikson) provides a major caveat to the enterprise of autobiography in general. He opens his piece with this paragraph (Erikson 2009: 110): I once attended an annual meeting at which the convener asked us to go around the table and say a word about what we had done in the year since we last met. Howard Zinn, a wonderful historian and a wonderful person, was among the last to make a report. When his turn came he said: “If I had known I was going to be asked to do this, I might have had a different year.” That’s how I feel now. If I had known I was going to have to account for myself in this way, I might have had a different life. Erikson’s main point applies to the brief webpage bio immediately above, as well as to the entire intellectual autobiography enterprise—we write such accounts in the present and try to make sense of what we have become from the perspective

1.7 Concluding Thoughts

63

of the present. Erikson says it more elegantly: “The point I am making, of course, is that what an individual becomes will provide the filter through which one sifts out the relevant detail of that life” (2009: 111). Erikson cites Sartre when he says such endeavors always “start at the end.” I am certain my ‘sifting’ has been no different, but hopefully of some use to anyone interested enough to read this piece. Erikson will also help me conclude in a most apposite way: “Trajectories do not come to an end until they turn into obituaries, and it is too early for that. So I won’t bring these remarks to a close. I’ll just stop talking” (2009: 119).

References Apter, David E., 2009: “Introduction”, in: Apter, David; Dallai, Patricia (Ed.): Intellectual Trajectories, Vol.I (New Haven: The Koerner Center): 6–16. Barber, James David, 1972: The Presidential Character (Englewood Cliffs, NJ: Prentice-Hall). Boulding, Kenneth, 1962: Conflict and Defense (New York: Harper & Row). Bueno de Mesquita, Bruce, 1975: Strategy, Risk and Personality in Coalition Politics: The Case of India (Cambridge, UK: Cambridge University Press). Cioffi-Revilla, Claudio; Starr, Harvey, 1995: “Opportunity, Willingness, and Political Uncertainty: Theoretical Foundations of Politics”, in: Journal of Theoretical Politics 24: 447–476. Darmofal, David, 2015: Spatial Analysis for The Social Sciences (Cambridge, UK: Cambridge University Press). Erickson, Kai, 2009: “So Far”, in: Apter, David; Dallai, Patricia (Ed.): Intellectual Trajectories, Vol.I (New Haven: The Koerner Center): 110–119. Friend, Theodore, 1965: Between Two Empires: The Ordeal of The Philippines, 1929–1946 (New Haven, CT: Yale University Press). Frohlich, Norman; Oppenheimer, Joe, 1970: “I Get By with a Little Help from My Friends”, in: World Politics 23: 104–120. Goertz, Gary; Mahoney, James, 2005: “Two-Level Theories and Fuzzy Set Analysis,” Sociological Methods & Research, 33: 497–538. Goertz, Gary; Starr, Harvey (Eds.), 2003: Necessary Conditions: Theory, Methodology, and Applications (Lanham, MD: Rowman & Littlefield). Holsti, Ole, 1962: “The Belief System and National Images: A Case Study”, in Journal Of Conflict Resolution, 6, 3: 244–252. Iqbal, Zaryab; Starr, Harvey, 2016: State Failure in the Modern World (Stanford: Stanford University Press). Kissinger, Henry. 1957: A World Restored (Boston: Houghton Mifflin). Kopelman, Shirli, 2019: “Tit for Tat and Beyond: The Legendary Work of Anatol Rapoport”, in Negotiation And Conflict Management Research, 13, 1: 16–84. Merritt, Richard; Russett, Bruce; Dahl, Robert, 2001: “Karl Wolfgang Deutsch, 1912–1992: A Biographical Memoir”, in: Biographical Memoirs, 80: 3–22. Most, Benjamin A, 1991: Changing Authoritan Rule and Public Policy in Argentina, 1930–1970 (Boulder, CO: Lynne Reinner Publishers). Most, Benjamin A; Starr, Harvey, 1980: “Diffusion, Reinforcement, Geopolitics, and the Spread of War”, in: American Political Science Review, 74: 932–946. Most, Benjamin A; Starr, Harvey, 1989: Inquiry, Logic, and International Politics (Columbia: University of South Carolina Press). Re-issued with new Preface 2015. Olson, Mancur; Zeckhauser, Richard, 1966: “An Economic Theory of Alliances”, in: Review Of Economics And Statistics, 48: 266–279.

64

1 Getting from Then to Now: A Personal Intellectual Autobiography

Palmer, Glenn (Ed.), 2000: “Special Issue: Substitutability in Foreign Policy: Applications and Advances”, in: Journal Of Conflict Resolution, 44: 1. Russett, Bruce, 1970: What Price Vigilance? The Burdens of National Defense (New Haven: Yale University Press). Russett, Bruce; Starr, Harvey, 1981: World Politics: The Menu for Choice (New York: W.H. Freeman). Russett, Bruce; Starr, Harvey, 2000: “From Democratic Peace to Kantian Peace: Democracy and Conflict in the International System,” in: Midlarsky, Manus I., (Ed.), Handbook of War Studies II (Ann Arbor, MI: University of Michigan Press): 93–128. Simon, Marc; Starr, Harvey, 1996: “Extraction, Allocation, and the Rise and Decline of States”, Journal of Conflict Resolution, 40, 1: 272–297. Simon, Marc; Starr, Harvey, 1997: “A Two-Level Analysis of War and Revolution: A Dynamic Simulation of Response to Threat”, in: Geva, Nehemia; Mintz, Alex (Eds.), Decision-Making on War and Peace: The Cognitive-Rational Debate (Boulder, CO: Lynne Rienner Publishers): 131–159. Simon, Marc; Starr, Harvey, 2000: “Two-Level Security Management and the Prospects for New Democracies: A Simulation Analysis”, in: International Studies Quarterly, 44, 3: 391–422. Simon, Marc; Starr, Harvey; McGinnis, Michael, 1994: “A Two-Level Analysis of War and Revolution: A Dynamic Simulation of Response to Threat”, Paper Delivered at the Conference on “Decision Making on War & Peace: The Cognitive-Rational Debate”, a Symposium Sponsored by the Military Studies Institute, Texas A&M University, March 10–11. Siverson, Randolph; Starr, Harvey, 1990: “Opportunity, Willingness and the Diffusion of War, 1816–1965”, in: American Political Science Review, 84, 1: 47-67. Siverson, Randolph; Starr, Harvey, 1991: The Diffusion of War: A Study of Opportunity and Willingness (Ann Arbor: University of Michigan Press). Snyder, Richard; Bruck, H.W.; Sapin, Burton, 1962: Foreign Policy Decision Making (New York: Free Press). Snyder, Glenn, 1961: Deterrence and Defense (Princeton, NJ: Princeton University Press). Starr, Harvey, 1972a: War Coalitions: The Distribution of Payoffs and Losses (Lexington, Mass.: D.C. Heath). Starr, Harvey, 1972b: “‘Organizational Process’ as an Influence on National Security Policy”, in: International Relations (London), IV: 176–186. Starr, Harvey, 1974: “A Collective Goods Analysis of the Warsaw Pact After Czechoslovakia”, in: International Organization, 28: 521–532. Starr, Harvey, 1975: Coalitions and Future Wars: A Dyadic Study of Cooperation and Conflict (Beverly Hills, California: Sage Publications, 1975). Starr, Harvey, 1978: “‘Opportunity’ and ‘Willingness’ as Ordering Concepts in the Study of War”, in: International Interactions, 4: 363–387. Starr, Harvey, 1980: “The Kissinger Years: Studying Individuals and Foreign Policy” in: International Studies Quarterly, 24, 4: 465–496. Starr, Harvey, 1984: Henry Kissinger: Perceptions of International Politics (Lexington: University Press of Kentucky). Starr, Harvey, 1991a: “Joining Political and Geographic Perspectives: Geopolitics and International Relations”, in: International Interactions, 17: 1–9. Starr, Harvey, 1991b: “Democratic Dominoes: Diffusion Approaches to the Spread of Democracy in the International System”, in: Journal of Conflict Resolution, 35: 356–381. Starr, Harvey, 1992: “Democracy and War: Choice, Learning, and Security Communities”, in: Journal of Peace Research, 29: 207–213. Starr, Harvey, 1994: “Revolution and War: Rethinking the Linkage Between Internal and External Conflict”, in: Political Research Quarterly, 47, 2: 481–507. Starr, Harvey, 1997a, Anarchy, Order, and Integration: How to Manage Interdependence (Ann Arbor: University of Michigan Press). Starr, Harvey, 1997b: “Democracy and Integration: Why Democracies Don’t Fight Each Other”, in: Journal of Peace Research, 34, 2: 153–162.

References

65

Starr, Harvey, (Ed.), 1999: The Understanding and Management of Global Violence: New Approaches to Theory and Research on Protracted Conflict (New York: St. Martin’s Press). Starr, Harvey, 2001: “Using Geographic Information Systems To Revisit Enduring Rivalries: The Case of Israel”, in: Geopolitics, 5, 1: 37–56. Starr, Harvey, 2002: “Opportunity, Willingness and Geographic Information Systems: Reconceptualizing Borders in International Relations”, in: Political Geography, 21: 243–261. Starr, Harvey, 2003: “The Power of Place and the Future of Spatial Analysis in the Study of Conflict”, in: Conflict Management and Peace Science, 20: 1–20. Starr, Harvey, 2005a: “Cumulation from Proper Specification: Theory, Logic, Research Design, and ‘Nice’ Laws”,” Conflict Management And Peace Science, 22, 4, 353–363. Starr, Harvey, 2005b: “Territory, Proximity, and Spatiality: The Geography of International Conflict”, in: International Studies Review, 7, 3: 387–406. Starr, Harvey (Ed.), 2006a: Approaches, Levels and Methods of Analysis in International Politics: Crossing Boundaries (New York: Palgrave Macmillan). Starr, Harvey, 2006b: “Introduction: The Future Study of International Relations, Two-Level Games, and Internal- External Linkages”, in: Starr, Harvey (Ed.), Approaches, Levels, and Methods of Analysis in International Politics: Crossing Boundaries (New York: Palgrave Macmillan): 1–9. Starr, Harvey, 2013: “On Geopolitics: Spaces and Places”, in: International Studies Quarterly, 57, 3: 433–39. Starr, Harvey, (Ed.), 2013a: On Geopolitics: Space, Place, and International Relations (Boulder, CO: Paradigm Publishers). Starr, Harvey, 2015a: Bruce Russett: Pioneer in the Scientific and Normative Study of War, Peace, and Policy (Cham, Switzerland: Springer International Publishing AG). Starr, Harvey, 2015b: “Preface: Inquiry, Logic and International Politics 25 Years On: The Expanded Coverage of a Critical Logic”, in: Most, Benjamin A.; Starr, Harvey: Inquiry, Logic and International Politics (Columbia, SC: University of South Carolina Press): xi–xxviii. Starr, Harvey, 2018: “Opportunity and Willingness: From ‘Ordering Concepts’ to an Analytical Perspective for the Study of Politics”, in: Thompson, William R. (Ed.), Oxford Encyclopedia of Empirical Theory of International Relations (New York: Oxford University Press). Starr, Harvey; Bain, Will, 1995. “The Application of Geographic Information Systems (GIS) to International Studies”, in: International Studies Notes, 20, 2: 1–8. Starr, Harvey, Liu, Roger Chi-feng, Dale Thomas, G., 2016: “The Geography of Conflict: Using GIS to Analyze Israel’s External and Internal Conflict Systems”, in: Starr, Harvey; Dubinsky, Stanley (Eds.), The Israeli Conflict System: Analytic Approaches (London: Routledge): 125–150. Starr, Harvey; Lindborg, Christina, 2003: “Democratic Dominoes Revisited: The Hazards of Governmental Transitions, 1974–96”, Journal of Conflict Resolution, 47, 4: 490–519. Starr, Harvey; McGinnis, Michael, 1992: “War, Revolution, and Two-Level Games: A Simple Choice-Theoretic Model”, Paper delivered at the Twenty-Sixth North American Meeting of the Peace Science Society (International), Pittsburgh. Starr, Harvey; Most, Benjamin, 1976: “The Substance and Study of Borders in International Relations Research”, in: International Studies Quarterly, 20: 581–620. Starr, Harvey; Simon, Marc V., 2017: “The Dynamics and Processes of Conflict: Linkages Between Internal and External Conflict within Differing Contexts”, in: Stohl, Michael; Lichbach, Mark I.; Grabosky, Peter (Eds.), States and Peoples in Conflict: Transformations of Conflict Studies (London and New York: Routledge): 95–112. Starr, Harvey; Dale Thomas, G., 2002: “The ‘Nature’ of Contiguous Borders: Ease of Interaction, Salience, and the Analysis of Crisis”, in: International Interactions, 28: 213–235. Starr, Harvey; Thomas, G. Dale, 2005: “The Nature of Borders and Conflict: Revisiting Hypotheses on Territory and War”, in: International Studies Quarterly, 49, 1: 123–139.

66

1 Getting from Then to Now: A Personal Intellectual Autobiography

Trachtenberg, Alan, 2009: “A Backward Glance”, in: Apter, David; Dallai, Patricia (Eds.), Intellectual Trajectories, Vol. I (New Haven: The Koerner Center): 246–252. Walker, Stephen, 1977: “The Interface Between Beliefs and Behavior: Henry Kissinger’s Operational Code and the Vietnam War”, in: Journal Of Conflict Resolution, 21, 1: 129–168. Zinnes, Dina A., 1980: “Three Puzzles in Search of a Researcher”, in: International Studies Quarterly, 24: 315–342.

Chapter 2

A Selected Bibliography of the Publications of Harvey Starr

2.1

Books

War Coalitions: The Distribution of Payoffs and Losses (Lexington, Mass.: D.C. Heath, 1972). Coalitions and Future Wars: A Dyadic Study of Cooperation and Conflict (Beverly Hills, California: Sage Publications, 1975). A Collective Goods Approach to Understanding Transnational Action (with Charles W. Ostrom, New York: Learning Resources in International Studies; Consortium for International Studies Education, Number 18; 1976). World Politics: The Menu for Choice (with Bruce Russett, New York: W.H. Freeman, 1981). Second edition 1985; third edition 1989; fourth edition 1992; fifth edition 1996; sixth edition 2000, with David Kinsella as third author (New York: Bedford/St. Martin’s); seventh edition 2003 (Belmont, CA: Wadsworth); eighth edition 2006; ninth edition 2009 (copyright 2010); tenth edition as Kinsella, Russett and Starr appeared 2012 (copyright 2013). Indian second edition by Vakils Feffer & Simons, Ltd., Bombay, 1986; Italian third edition by Il Mulino; Taiwan fourth edition by Wu Nan Book Co.; Russian fifth edition by Svetlychor; Chinese fifth edition by Huaxia Press; Japanese sixth edition by Japan Uni Agency; seventh edition in English with Chinese introduction, Peking University Press seventh edition, 2004; Chinese ninth edition.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7_2

69

70

2 A Selected Bibliography of the Publications of Harvey Starr

Henry Kissinger: Perceptions of International Politics (Lexington: University Press of Kentucky, 1984). Inquiry, Logic, and International Politics (with Benjamin A. Most, Columbia: University of South Carolina Press, 1989; re-issued 2015). – re-issued 2015 with A New Preface by Harvey Starr, pp. xi–xxviii. – Chapter 3, reprinted in Gary Goertz and Harvey Starr, eds., Necessary Conditions: Theory, Methodology, and Applications (Lanham, MD: Rowman & Littlefield, 2003), pp. 25–45. – Chapter 4, reprinted in Paul F. Diehl, ed., War, Vol.I, Concepts, Measurement And Patterns in the Study of War (Thousand Oaks, CA: Sage Publications, 2005), pp. 41–59. Choices in World Politics: Sovereignty and Interdependence (editor with Bruce Russett and Richard Stoll, New York: W.H. Freeman, 1989). The Diffusion of War: A Study of Opportunity and Willingness (with Randolph M. Siverson, Ann Arbor: University of Michigan Press, 1991). Anarchy, Order, and Integration: How to Manage Interdependence (Ann Arbor: University of Michigan Press, 1997). Agency, Structure, and International Politics: From Ontology To Empirical Inquiry (with Gil Friedman, London: Routledge, 1997). The Understanding and Management of Global Violence: New Approaches to Theory and Research on Protracted Conflict (editor, New York: St. Martin’s Press, 1999). Necessary Conditions: Theory, Methodology, And Applications, (editor with Gary Goertz, Lanham, MD: Rowman & Littlefield, 2003). Approaches, Levels and Methods of Analysis In International Politics: Crossing Boundaries (editor, New York: Palgrave Macmillan, 2006).

2.1 Books

71

Dealing with Failed States: Crossing Analytic Boundaries (editor, Milton Park, UK and New York: Routledge, 2009). On Geopolitics: Space, Place, and International Relations (Boulder, CO: Paradigm Publishers, 2013).

72

2 A Selected Bibliography of the Publications of Harvey Starr

Bruce Russett: Pioneer in the Scientific and Normative Study of War, Peace, and Policy (editor, Cham, Switzerland: Springer International Publishing AG, 2015). The Israeli Conflict System: Analytic Approaches (editor with Stanley Dubinsky, London: Routledge, 2016). State Failure in the Modern World, (with Zaryab Iqbal, Stanford: Stanford University Press, 2016). Editor, Special Issue on “Failed States,” Conflict Management And Peace Science (vol. 25, no. 4, 2008). Co-Editor with Zaryab Iqbal, Special Issue, “Spaces and Places Geopolitics in an Era of Globalization,” International Studies Review (vol. 17, no. 1, 2015).

2.2

Journal Articles

“‘Organizational Process’ as an Influence on National Security Policy,” International Relations (London), IV, 1972, pp. 176–186. “A Collective Goods Analysis of the Warsaw Pact After Czechoslovakia,” International Organization, 28, 1974, pp. 521–532. –Reprinted in Keith Hartley and Todd Sandler, eds., The Economics of Defense, Vol. I (Northampton, MA: Edward Elgar, 2001), pp. 305–316. “The Quantitative International Relations Scholar as Surfer: Riding the ‘Fourth Wave’”, Journal of Conflict Resolution, 18, 1974, pp. 336–368. “Developing the Relevance Potentialities of National Security and Foreign Policy Research: Some Proposed Criteria,” Policy Sciences, 6, 1975, pp. 161–173 (with U. Arad, W. Bacchus, E. Gonzalez).

2.2 Journal Articles

73

74

2 A Selected Bibliography of the Publications of Harvey Starr

“The Substance and Study of Borders in International Relations Research,” International Studies Quarterly, 20, 1976, pp. 581–620 (with Benjamin A. Most). “Physical Variables and Foreign Policy Decision-Making: Daily Temperature and the Pre-World War I Crisis,” International Interactions, 3, 2, 1977, pp. 97–108.

2.2 Journal Articles

75

“A Return Journey: Richardson, ‘Frontiers’ and War in the 1946–65 Era,” Journal of Conflict Resolution, 22, 1978, pp. 323–356 (Benjamin A. Most). “‘Opportunity’ and ‘Willingness’ as Ordering Concepts in the Study of War,” International Interactions, 4, 1978, pp. 363–387. Reprinted in Paul F. Diehl, ed., War, Vol.I, Concepts, Measurement and Patterns in the Study of War (Thousand Oaks, CA: Sage Publications, 2005), pp. 90–112. “Diffusion, Reinforcement, Geopolitics, and the Spread of War,” American Political Science Review, 74, December 1980, pp. 932–946 (with Benjamin A. Most). “The Kissinger Years: Studying Individuals and Foreign Policy,” International Studies Quarterly, 24, December 1980, pp. 465–496. – Reprinted in Dan Caldwell, ed., Henry Kissinger: His Personality and Politics (Durham, N.C.: Duke University Press, 1983), pp. 3–23. – Reprinted in Bruce Russett, Harvey Starr, and Richard Stoll, eds., Choices in World Politics: Sovereignty and Interdependence (New York: W.H. Freeman, 1989), pp. 181–205. “Kissinger’s Operational Code,” Korea and World Affairs, 4, Winter 1980, pp. 582–606. “Inherent Bad Faith Reconsidered: Dulles, Kennedy and Kissinger,” Political Psychology, 3, Fall-Winter 1981–82, pp. 1–33 (with Douglas Stuart). “Case Selection, Conceptualization and Basic Logic in the Study of War,” American Journal Of Political Science, 26, November 1982, pp. 834–856 (with Benjamin A. Most). “Conceptualizing ‘War’: Consequences for Theory and Research,” Journal Of Conflict Resolution, 27, 1983, pp. 137–159 (with Benjamin A. Most). “Contagion and Border Effects on Contemporary African Conflict,” Comparative Political Studies, 16, 1983, pp. 92–117 (with Benjamin A. Most). “International Relations Theory, Foreign Policy Substitutability, and ‘Nice’ Laws,” World Politics, 36, 1984, pp. 383–406 (with Benjamin A. Most). “The Relationship Between Defense Spending and Inflation,” Journal of Conflict Resolution, 28, 1984, pp. 103–122 (Francis Hoole, Jeffrey Hart, and John Freeman). “The Forms and Processes of War Diffusion: Research Update on Contagion in African Conflict,” Comparative Political Studies, 18, 1985, pp. 206–227 (with Benjamin A. Most). “Polarity, Preponderance, and Power Parity in the Generation of International Conflict,” International Interactions, 13, 1987, pp. 225–262 (with Benjamin A. Most). “Rosenau, Pre-Theories and the Evolution of the Comparative Study of Foreign Policy,” International Interactions, 14, 1988, pp. 3–16.

76

2 A Selected Bibliography of the Publications of Harvey Starr

“International Data as a National Resource,” International Interactions, 14, 1988, pp. 101–111 (with Patrick McGowan, Gretchen Hower, Richard Merritt, and Dina A. Zinnes. “Alliance and Border Effects on the War Behavior of States,” Conflict Management and Peace Science, 10, 2, 1989, pp. 21–46 (with Randolph Siverson). “Opportunity, Willingness and the Diffusion of War, 1816–1965,” American Political Science Review, 84, 1, 1990, pp. 47–67 (with Randolph Siverson). – Reprinted in Michael Don Ward, ed., The New Geopolitics (Philadelphia: Gordon and Breach, 1992), pp. 159–181. – Reprinted in Paul F. Diehl, ed., WAR, VOL. VI, The Moderation and Outcomes of Conflict (Thousand Oaks, CA: Sage Publications, 2005), pp. 1–25. “Alliances and Geopolitics,” Political Geography Quarterly, 9, 1990, pp. 232–248 (with Randolph Siverson). “Theoretical and Logical Issues in the Study of Diffusion,” Journal of Theoretical Politics, 2, 1990, pp. 391–412 (with Benjamin A. Most). “Democratic Dominoes: Diffusion Approaches to the Spread of Democracy in the International System,” Journal of Conflict Resolution, 35, 1991, pp. 356–381. “Joining Political and Geographic Perspectives: Geopolitics and International Relations,” International Interactions, 17, 1991, pp. 1–9. – Reprinted in Michael Don Ward, ed., The New Geopolitics (Philadelphia: Gordon and Breach, 1992), pp. 1–9. “Democracy and War: Choice, Learning, and Security Communities,” Journal of Peace Research, 29, 1992, pp. 207–213. “Why Don’t Democracies Fight Each Other? Evaluating the Theory-Findings Feedback Loop,” Jerusalem Journal of International Relations, 14, 4, 1992, pp. 41–59. “Regime Change and the Restructuring of Alliances,” American Journal of Political Science, 38, 1, 1994, pp. 145–16 (with Randolph Siverson). “Revolution and War: Rethinking the Linkage Between Internal and External Conflict,” Political Research Quarterly, 47, 2, 1994, pp. 481–507. “Opportunity, Willingness, and Political Uncertainty: Theoretical Foundations of Politics,” Journal of Theoretical Politics, 7, 4, 1995, pp. 447–476 (with Claudio Cioffi-Revilla). Reprinted in Gary Goertz and Harvey Starr, eds., Necessary Conditions: Theory, Methodology, and Applications (Lanham, MD: Rowman & Littlefield, 2003) “The Application of Geographic Information Systems (GIS) to International Studies,” International Studies Notes, 20, 2, 1995, pp. 1–8 (with Will Bain). “Extraction, Allocation, and the Rise and Decline of States,” Journal of Conflict Resolution, 40, 1, 1996, pp. 272–297 (with Marc V. Simon). “Democracy and Integration: Why Democracies Don’t Fight Each Other,” Journal of Peace Research, 34, 2, 1997, pp. 153–162.

2.2 Journal Articles

77

“Cumulation, Evaluation and the Research Process: Investigating the Diffusion of Conflict,” Journal of Peace Research, 35, 2, 1998, pp. 231–237 (with Randolph M. Siverson). “Substitutability in Foreign Policy: Theoretically Central, Empirically Elusive,” Journal of Conflict Resolution, 44, 1, 2000, pp. 128–138. “Two-Level Security Management and the Prospects for New Democracies: A Simulation Analysis,” International Studies Quarterly, 44, 3, 2000, pp. 391–422 (with Marc V. Simon). “Reconceptualizing International Borders Through the Application of GIS (Geographic Information Systems): A New Dataset on the ‘Nature’ of Borders,” International Studies Perspectives 2, 4, November 2001, pp. 441–43. “Uncertainty, Change, and War: Power Fluctuations and War in the Modern Elite Power System,” Journal of Peace Research, 38, 1, 2001 pp. 49–66 (with Jeffrey S. Morton). “Using Geographic Information Systems to Revisit Enduring Rivalries: The Case of Israel,” Geopolitics, 5, 1, 2001, pp. 37–56. “Opportunity, Willingness and Geographic Information Systems: Reconceptualizing Borders in International Relations,” Political Geography, 21, 2002, pp. 243–261. “The ‘Nature’ of Contiguous Borders: Ease of Interaction, Salience, and the Analysis of Crisis,” International Interactions, 28, 2002, pp. 213–235 (with G. Dale Thomas). “The Power of Place and the Future of Spatial Analysis in the Study of Conflict,” Conflict Management and Peace Science, 20, 1, 2003, pp. 1–20. “Democratic Dominoes Revisited: The Hazards of Governmental Transitions, 1974–96,” Journal of Conflict Resolution, 47, 4, 2003, pp. 490–519 (with Christina Lindborg). “The Nature of Borders and Conflict: Revisiting Hypotheses on Territory and War,” International Studies Quarterly, 49, 1, 2005, pp. 123–139 (with G. Dale Thomas). “Territory, Proximity, and Spatiality: The Geography of International Conflict,” International Studies Review, 7, 3, 2005, pp. 387–406. “International Borders: What They Are, What They Mean, and Why We Should Care,” SAIS Review, 26, 1, 2006, pp. 3–10. “Bad Neighbors: Failed States and Their Consequences,” Conflict Management and Peace Science, 25, 4, 2008, pp. 315–331 (with Zaryab Iqbal). – Reprinted in Harvey Starr, ed., Dealing With Failed States: Crossing Analytic Boundaries (London: Routledge, 2009), pp. 35–51. “On Geopolitics: Spaces and Places,” International Studies Quarterly, 57, 3, 2013, pp. 433–39. “Geopolitics and Conflict: Reconciling Spatiality, Borders, and Sovereignty in the Modern World System,” Journal of Territorial and Maritime Studies, 2, 1, 2015, pp. 139–148.

78

2 A Selected Bibliography of the Publications of Harvey Starr

“State Failure and the International Community,” Political Violence @a Glance: Expert Analysis on Violence and Its Alternatives, May 4, 2016 (with Zaryab Iqbal); at https://politicalviolenceataglance.org/ “Is language key to resolving the Israeli-Arab conflict?” The Conversation, January 16, 2018, https://theconversation.com/is-language-key-to-resolving-the-israeliarab-conflict-89215) (with Stanley Dubinsky).

2.3

Book Chapters

“Alliances and the Price of Primacy,” Chapter 4, in Bruce M. Russett, What Price Vigilance? The Burdens of National Defense (with Bruce Russett, New Yale University Press, 1970). – Reprinted in George Lopez and Michael Stohl, eds., International Relations: Contemporary Theory And Practice (Washington, D.C.: CQ Press, 1988). “An Appraisal of the Substantive Findings of the Correlates of War Project,” in Francis W. Hoole and Dina A. Zinnes, eds., Quantitative International Politics: An Appraisal (New York: Praeger, 1976), pp. 99–127. “Alliances: Tradition and Change in American Views on Foreign Military Entanglements,” in Ken Booth, and Moorhead Wright, eds., American Thinking About Peace And War (New York: Barnes & Noble, 1978), pp. 37–57. “Henry Kissinger’s Belief System and World Order: Perception and Policy,” in Henry H. Han, ed., World in Transition: Challenges To Human Rights, Development And World Order (Washington, D.C.: University Press of America, 1979), pp. 239–255. “‘Detente’ or ‘Two Against One’? The China Factor,” in Charles W. Kegley and Pat McGowan, eds., Foreign Policy: USA/USSR, Sage International Yearbook of Foreign Policy Studies, vol.7 (Beverly Hills: Sage, 1982), pp. 213–239. “Patterns of Conflict: Quantitative Analysis and the Comparative Lessons of Third World Wars,” in Robert E. Harkavy and Stephanie Neumann, eds., The Lessons Of Recent Wars In The Third World (with Benjamin A. Most, Lexington, Mass.: D.C. Heath, 1985), pp. 33–52. – Reprinted in George Lopez and Michael Stohl, eds., International Relations: Contemporary Theory And Practice (Washington, D.C.: CQ Press, 1988). “From War to War: Stochastic Models of Wartime Alignment Patterns,” in Michael Don Ward, ed., Theories, Models and Simulations In International Relations (with Barbara Hill and Stephen Portnoy Boulder: Westview Press, 1985), pp. 393–418. “Opportunity, Willingness and Small States: The Relationship Between Environment and Foreign Policy,” in Charles F. Hermann, Charles W. Kegley, and

2.3 Book Chapters

79

James N. Rosenau, eds., New Directions In The Study Of Foreign Policy (with Maria Papadakis, Boston: George Allen and Unwin, 1987), pp. 409–432. “Introduction,” in Bruce Russett, Harvey Starr, and Richard Stoll, eds., Choices In World Politics: Sovereignty and Interdependence (with Bruce Russett and Richard Stoll, New York: W.H. Freeman, 1989), pp. 1–5. “The Logic and Study of the Diffusion of International Conflict,” in Manus I. Midlarsky, ed., Handbook of War Studies (with Benjamin A. Most, and Randolph M. Siverson, Boston: Unwin Hyman, 1989), pp. 111–139. “Border and Alliance Effects in the Diffusion of Major Power Conflict, 1815– 1965,” in Charles S. Gochman and Alan Ned Sabrosky, eds., Prisoners of War? Nation-States in the Modern Era (with Benjamin A. Most, Philip A. Schrodt, and Randolph M. Siverson, Lexington, MA: D.C. Heath, 1990), pp. 209–229. “International Law and International Order,” in Charles W. Kegley, Jr., ed., Controversies in International Relations Theory: Realism and The Neoliberal Challenge (New York: St. Martin’s Press, 1995), pp. 299–315. “Advancing the Scientific Study of War: A Commentary,” in Stuart A. Bremer and Thomas R. Cusack, eds., The Process Of War: Advancing the Scientific Study of War (Philadelphia: Gordon and Breach, 1996), pp. 233–240. “How to Manage Interdependence? The State in a Multi-Centric/Transnational World,” in Brian Toyne And Douglas Nigh, Eds., International Business: An Emerging Vision (Columbia: University of South Carolina Press, 1997), pp. 274– 281. “A Two-Level Analysis of War and Revolution: A Dynamic Simulation of Response to Threat,” in Nehemia Geva and Alex Mintz, eds., Decision-Making on War and Peace: The Cognitive\ Rational Debate (with Marc V. Simon, Boulder, CO: Lynne Rienner Publishers, 1997), pp. 131–159). “Introduction: A Protracted Conflict Approach to the Study of Social Conflict,” in Harvey Starr, ed., The Understanding and Management of Global Violence (New York: St. Martin’s Press, 1999), pp. 1–20. “The Institutional Maintenance of 21st Century World Order,” in Charles W. Kegley, Jr and Eugene Wittkopf, eds., The Global Agenda, 6th ed. (Boston: McGraw-Hill, 2001), pp. 216–229. “From Democratic Peace to Kantian Peace: Democracy and Conflict in the International System,” in Manus I. Midlarsky, ed., Handbook of War Studies II (with Bruce Russett, Ann Arbor, MI: University of Michigan Press, 2000), pp. 93–128. “Visions of Global Politics as an Intellectual Enterprise: Three Questions Without Answers,” in Donald Puchala, ed., Visions Of International Relations: Assessing An Academic Field (Columbia, SC: University of South Carolina Press, 2002), pp. 42–61. “Cumulation, Synthesis and Research Design for the ‘Post-Fourth Wave’,” Harvey Starr, published simultaneously in: – Michael Brecher and Frank Harvey, eds., Millennium Reflections on International Studies (Ann Arbor: University of Michigan Press, 2002, pp. 361– 373).

80

2 A Selected Bibliography of the Publications of Harvey Starr

– Frank Harvey and Michael Brecher, eds., Evaluating Methodology in International Studies (Ann Arbor: University of Michigan Press, 2002, pp. 43– 58). “Introduction: Necessary Condition Logics, Research Design, and Theory,” in Gary Goertz and Harvey Starr, eds., Necessary Conditions: Theory, Methodology, and Applications, (with Gary Goertz, Lanham, MD: Rowman & Littlefield, 2003), pp. 1–23. “Introduction: The Future Study of International Relations, Two-Level Games, and Internal- External Linkages,” in Harvey Starr, ed., Approaches, Levels, and Methods of Analysis In International Politics: Crossing Boundaries (New York: Palgrave Macmillan, 2006), pp. 1–9. “Democratic Peace and Integration: Synergies Across Levels of Analysis,” in Harvey Starr, ed., Approaches, Levels, And Methods Of Analysis In International Politics: Crossing Boundaries (New York: Palgrave Macmillan, 2006), pp. 117– 138. “Conditions, Necessary and Sufficient,” in Bertrand Badie, Dirk Berg-Schlosser, and Leonardo Morlino, eds., International Encyclopedia Of Political Science (Sage Publications, 2011), pp. 384–88, Harvey Starr [Online version: http://www.sageereference.com/view/intlpoliticalscience/n92.xml] “Preface” and “A Scholarly Biography,” in Harvey Starr, ed., Bruce Russett: Pioneer In The Scientific And Normative Study of War, Peace, And Policy (Cham, Switzerland: Springer International Publishing AG, 2015), pp. vii–xiii; pp. 3–21. Preface, “Inquiry, Logic and International Politics 25 Years On: The Expanded Coverage of a Critical Logic,” in Benjamin A. Most and Harvey Starr, Inquiry, Logic And International Politics (Columbia, SC: University of South Carolina Press, 2015), pp. xi–xxviii. “The Geography of Conflict: Using GIS to Analyze Israel’s External and Internal Conflict Systems,” in Harvey Starr and Stanley Dubinsky, eds., The Israeli Conflict System: Analytic Approaches (with Roger Chi-Feng Liu, and G. Dale Thomas, London: Routledge, 2016), pp. 125–150. “Introduction: Crossing Disciplinary and Methodological Boundaries in Conflict System Analysis,” in Harvey Starr and Stanley Dubinsky, eds., The Israeli Conflict System: Analytic Approaches (with Stanley Dubinsky, London: Routledge, 2016), pp. 1–6. “The Dynamics and Processes of Conflict: Linkages Between Internal and External Conflict within Differing Contexts,” in, Michael Stohl, Mark I. Lichbach, and Peter Grabosky, eds., States and Peoples in Conflict: Transformations of Conflict Studies (with Marc V.Simon, London, and New York: Routledge, 2017), pp. 95–112. “Opportunity and Willingness: From ‘Ordering Concepts’ to an Analytical Perspective for the Study of Politics”, in William R. Thompson, ed., Oxford Encyclopedia of Empirical Theory of International Relations (New York: Oxford University Press, 2018).

Part II

Texts by Harvey Starr

At the Metropolitan Museum of Art, New York City, March 2011. Source From the author’s personal photo collection

In San Francisco for the 2015 annual meeting of the American Political Science Association. Source From the author’s personal photo collection

Chapter 3

Cumulation from Proper Specification: Theory, Logic, Research Design, and ‘Nice’ Laws

Abstract Jim Ray has set out a challenge to students of international conflict: how can we improve our research to do better in the cumulation of knowledge and understanding of international phenomena? While most of the commentary in this special issue revolves about more technical statistical issues, the questions that Ray raises must also be addressed not just on statistical/methods grounds, but within the context of broader theoretical concerns and theoretical specification. That is the aim of this article. Drawing on the concept of the “research triad” of logic, theory, and methodology presented by (Most and Starr in Inquiry, logic and international politics. University of South Carolina Press, Columbia, SC, 1989), this article will stress the relationship of theory to the development of research designs that will permit both the additive and integrative cumulation of knowledge. This article will discuss how statistical model specification can be informed by broader principles of research design, demonstrating how Ray’s basic points are supported by earlier, complementary, and converging lines of argument. Keywords Cumulation


Research triad
Research design

Jim Ray has set out a challenge to students of international conflict: how can we improve our research to do better in the cumulation of knowledge and understanding of international phenomena? While his specific concern has been with the ways in which we go about doing multivariate analysis, the challenge is a much broader one. While most of the commentary in this special issue revolves about more technical statistical issues, the questions that Ray raises must also be addressed not just on statistical/methods grounds, but within the context of broader theoretical concerns and theoretical specification. Simply, that is the aim of this article. Drawing on the concept of the “research triad” of logic, theory, and methodology presented by Most/Starr (1989: 9–17), this article will stress the This text was first published as: “Cumulation from Proper Specification: Theory, Logic, Research Design, and ‘Nice’ Laws,” Conflict Management and Peace Science, 22:353–363, 2005; Copyright © Peace Science Society (International). Republished with the permission of the Peace Science Society (International). © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7_3

83

84

3 Cumulation from Proper Specification …

relationship of theory to the development of research designs that will permit both the additive and integrative cumulation of knowledge. More specifically, this article will discuss how statistical model specification can be informed by broader principles of research design.1 In so doing, the article will show how Ray’s basic points are supported by earlier, complementary, and converging lines of argument, work that did not specifically look at the issues of control or the specific design of multivariate studies, but at the overall process of logic of inquiry and research design. The primacy of theory is a central theme in the work of Most and Starr, among many others. They call for a number of activities. Perhaps first and foremost is understanding the logical form of the relationship generated by a theory (which includes the logical causal structure such as necessity, sufficiency, being necessary and sufficient, etc.), which affects most of the other research design aspects as well, e.g., should we expect linear or curvilinear results, and what is the shape of the relationship between independent and dependent variables? For example, using logic, Most and Starr indicated how scholars had set up research hypotheses based on sufficient relationships, but then went on to test them with designs that could only deal with necessary relationships and vice versa. Their initial questioning of how researchers have tried to test or evaluate necessary relationships has been taken up more recently in collaborative work by Goertz and Braumoeller (see, e.g., Braumoeller/Goertz 2003; see also Goertz/Starr 2003a). One central issue presented in Goertz/Starr (2003b) and Goertz (2003) is that the use of ‘standard’ statistical techniques will not capture the effects of necessity.2 Logic, as developed in Most/Starr (1989), also dictates the need to understand more fully what (types of) cases are needed to test or evaluate theory. Case selection is thus intimately related to understanding the context in which the theory should be expected to hold (or, in which one could we expect to find evidence of the implications of the theory; see also Achen, in this issue). Understanding the broader concepts that are of concern in the theory, as well as understanding what phenomena are really under investigation and why (e.g., Most/Starr 1989, Chap. 4 on conceptualizing war) are also needed. Specifying this context involves “nice laws.” The broader understanding of concepts and process will also involve ‘substitutability.’ Each of these concepts will be addressed more fully below. As argued in Most/Starr (1989), all of these activities must take place for the proper specification of the theory, which is needed for the accumulation of knowledge. “Proper model specification” is here used in its broadest conceptual and theoretical sense. Nothing I say here will tell any individual scholar exactly how to 1

I would like to thank Kelly Kadera and Sara McLaughlin Mitchell, as well as two anonymous referees, for their comments on earlier drafts of this article, which was originally presented at the annual meeting of the Peace Science Society (International), November 2004, Houston. The responsibility for all errors and omissions, of course, is my own. 2 This is not to imply that sufficiency is any more easily captured, only that there are alternatives to what Goertz (2003) calls “correlational hypotheses.” See also Goertz (2005), as well as the responses to his piece in the “Necessary Conditions” Symposium in Political Analysis.

3 Cumulation from Proper Specification …

85

specify any model in any specific research project. As noted, this is not a technical/ methods discussion, directing the reader to specific techniques, but one trying to present broader implications. Here proper specification means asking any researcher to think through the nature and logic of the causal relationships involved in the theory being used/developed, to think ‘process,’ to understand the implications of nice laws and substitutability, etc. Indeed, the point here, as in Most and Starr, is to ask the researcher to think through the theoretical structure more fully and clearly, and to include those variables (and in their proper role, e.g., as independent, intervening, complementary, or control) that are relevant to the concepts in the theoretical process. This procedure stands in distinction from what Achen in this issue calls “garbage-can regressions.” Therefore, theoretical specification should be addressed before dealing with the more technical aspects of research design.3 The explicit argument in Most and Starr (and more implicitly in Ray's papers) is that such activities will have a positive impact on cumulation. Most and Starr (1989: 7), drawing on Zinnes (1976), note that “additive cumulation occurs when ‘one study adds some information to the existing literatures on the subject,’ through such activities as the citation of previous findings, using previously collected data, secondary or reanalysis of existing data, the incorporation of new cases or new variables into the analysis, or expanding the application of models, indices or techniques to new cases or research questions.” Integrative cumulation, however, goes further. It reflects instances where earlier studies are “‘crucial’ to the conceptual and theoretical components of the subsequent study’s research design.” Both forms of cumulation need to be pursued, as additive cumulation is an indispensable part of the process that leads to integrative cumulation. The arguments that Ray presents (2003a,b), can be easily translated to, or subsumed under, this broader enterprise of ‘cumulation.’ That is, how can we account for the lack of cumulation of knowledge in certain research areas, and what lessons concerning the logic of inquiry and research design could we apply to increase both the additive and integrative cumulation of knowledge in these areas. Most/Starr (1989) similarly used the area of international conflict to illustrate a set of research design problems that had hindered cumulation in the study of international relations. They also proposed solutions that had broader implications. Most and Starr were concerned with the failure to think through the many complicated aspects of research design, especially the multiple feedback loops among the components of research design, along with a rather rote application of the basic social science statistical cookbook. It is possible, then, to see the basic enterprise

3

For how more technical methodological choices may follow, see Most/Starr (1989, Chaps. 3–4) for the effects of logic and different forms of causality on case selection, and the analyses of those cases; see the chapters in Goertz/Starr (2003a) for the methodological impact of “necessary condition hypotheses,” and designs structured around dealing with necessity as the form of the relationship; or Cioffi-Revilla/Starr (1995) who use various logics to deal with the uncertainty of politics. Discussions of specific methods such as standard interaction terms, or Boolean methods, should be left to true methodologists, such as Braumoeller (2003).

3 Cumulation from Proper Specification …

86

and general thrust in Ray’s work and in the Most and Starr volume as one and the same: a critical view of how we, as scholars, design our research, and the problems we can fall into when we do not think through the full ramifications that our research design choices might have for the validity of our results.

3.1

Applying the Research Triad: Theory, Nice Laws, and Substitutability

Ray begins with a basic point that he uses to explain the increased (and indiscriminate) use of control variables. He takes one route to argue the primacy of theory in the selection and use of control variables, and I will take another. The basic point Ray develops (2003a: 2–4) is that over the past 15–20 years researchers have moved beyond searching for ‘ideal’ or ‘best fit’ general models. Instead, they have developed models “to evaluate the impact of one key factor” (2003a: 3), seeing if the impact of that factor can survive the addition of control variables. Even though its practitioners are almost certainly not thinking in these terms, the latter ‘key factor’ trend is clearly within the spirit of Most and Starr’s idea of “nice laws.” These are “sometimes true ‘domain specific’ laws” (1989: 98). Most and Starr remind us that while we should aim for generality, the “right type of law” is one which is clearly specified; that the relationships among variables that it proposes will work only under specified conditions.4 Most and Starr question whether social scientists will ever generate important ‘universal’ laws. They note, however, that it may be useful to recognize that there could very well be laws that are in some sense “good,” “domain specific,” or “nice” even though the relationships they imply are not necessarily very general empirically . . . it may be more productive to think of laws each of which is always true under certain conditions (or within certain domains) but which is only “sometimes true' empirically because those conditions do not always hold in the empirical world. (Most/Starr 1989: 117)

Again, it is likely that “key factor” analysts are not thinking in terms of nice laws and are still searching for universal laws when they add control variables. Yet, they are on the right track. Even if they do not know what they are doing, this more recent strategy can be viewed as a form of nice laws. It is one way to see under what conditions a key factor does not hold, through the use of control variables. In the spirit of additive cumulation, such studies indeed provide new information through 4

For example, a theory of third-party deterrence which holds in the 19th century might not hold under conditions which include nuclear warfare. The earlier debate over the effects of bipolarity versus multipolarity essentially argued that under one condition the balance of power (stability) would work well, but not under the other. Ray (2003b: 4), for example, discusses the different results of Bremer's research on the relationship between development and war in terms of different conditions, in that the structure and complexity of the models used were quite different. In physics almost all ‘universal’ laws provide conditions that must be met for the ‘law’ to hold.

3.1 Applying the Research Triad: Theory, Nice Laws …

87

the incorporation of new variables. For example, Ray (2003b: 5) discussing Bremer's work on dangerous dyads, stresses: “Again, the addition of one explanatory factor to the multivariate model originally constructed by Bremer (1992) has several substantial impacts on several of the other relationships that are analyzed. The inclusion of this one additional factor results in a fundamental revision of the rank orderings of the variables by the weight of their apparent causal impact.”5 The important question is whether or not studies of the impact of some key variable can be part of a process of integrative cumulation. The answer rests with theory, and the place of theory in the research process. Ray (2003a: 4) supports the “key factor” approach, using a theoretical critique of the ideal/best fit approach: We have no theory, or theoretical approach that will tell us what are the best six, seven, or eight predictor variables to put into a multivariate model aimed at accounting for interstate war or conflict. We have no sound theoretical basis for saying “these are the seven most important variables on which to base any explanation of interstate wars. It is these seven and only these seven.”

What Ray is pointing out is that researchers need theory. They need it first and they need it to be specified. In order for the theory to be specified, that is, to understand which factors, variables, etc. are important when, under what conditions, and in what form, researchers require a model, and a model which strongly incorporates some theoretical process (e.g., see Bremer/Cusack 1996; see also Starr 1996). In order for theory to be specified researchers require a procedure by which they can build process models. Lave/March (1975) provide such procedures for “conceptual modeling.” These procedures include thinking in terms of process as one of their three “rules of thumb.” Bremer (1996a: 11) helps us to capture the idea of process by contrasting Mohr’s “variance theory” and “process theory”: “Variance theories are deterministic, continuous, essentially static theories of being while process theories are stochastic, discrete, inherently dynamic theories of becoming.” Bremer’s (1996b: 267) description of “process theory” as “a set of interconnected rule-like statements that purport to explain how and under what conditions a system will undergo change” helps pull together the Lave and March model-building process and Most and Starr's notion of nice laws. Most/Starr (1989) rely heavily on the Lave and March model-building process, including their own variations such as the use of stylized facts for quick and dirty tests of the implications of the theory/model. The theoretical specification of a model or theory requires an understanding of nice laws. That is, the proposed relationships or effects of a model are seen to hold only under specified conditions, which derive from the process within the model-theory story. Most and Starr develop a strategy for honing in on such conditions.

5

Reflecting a point made below on substitutability, one question that must be asked is whether such results truly reflect the impact of nice laws, or simply statistical inadequacy in the face of complexity.

88

3 Cumulation from Proper Specification …

Rather than using elaborate or sophisticated statistical analyses,6 Most/Starr (1989: 136) look for “stylized facts”: “Any argument that might be developed must be capable of dealing with existing empirical evidence—stylized facts. While these observations are simple and intentionally impressionistic, they may not be entirely trivial.” The Most and Starr strategy for taking a first cut at a problem was to develop a process model, and try to think through “what the world would look like if” one or another of a set of hypotheses held; hypotheses were not seen as ‘contending,’ but rested on nice laws and proceeded from different conditions. These thought experiments were done rather quickly, generating rather crude and impressionistic predictions or derivations. Given these directional indicators, a second step was taken, a “quick and dirty,” intentionally nonsystematic consideration of some stylized facts and simple hypothetical cases. They note (1989: 182) that such procedures fit quite closely with Eckstein’s (1975: 118) typology of case studies, especially his notion of “must fit” cases. Thus, the use of stylized facts shows how case studies could be used to help delineate nice laws, or the conditions under which different models/theories might be expected to work.7 At the other end of the methodological spectrum is a more current approach also concerned with linking theory with methods, and linking theoretical process to empirical analysis: the EITM-related literature (e.g., see Empirical Implications of Theoretical Models, 2002). As Aldrich/Alt (2003: 309) note, “the problem of separation between good theorizing and good empirical work is common across many disciplines.”8

Again, let me make clear that the concept of “nice laws” does not presuppose any specific context, set of conditions, or form of relationship. The concept only asks the researcher to think about and hypothesize when and where—what spatial and temporal domains—and under what special conditions (or ‘treatments’ as in the diffusion studies of Most/Starr 1980 and Siverson/Starr 1991) the causal relationships should hold. Holsti (1991) provides a good example of the temporal dimension in his wide-ranging discussion of how the issues that lead states to war have changed over the centuries. The treatments/conditions involved might be very simple (e.g., bordering a state which is at war in regard to the spatial contagion of conflict, or the existence of ICBM delivery systems with nuclear warheads in a theory using the loss-of-strength gradient), or complex (the various conditions and their interaction in Deutschian models of integration), or somewhere in between (such as Holsti's study of the issues/stakes of war). 7 One illustration is Most/Starr (1989, Chap. 6), where the ‘contending’ hypotheses of power preponderance versus power parity were approached by quickly developing “an experiment that would produce conditions and conclusions regarding the logical properties of the hypotheses,” and also help delineate what might be sufficient or necessary conditions. 8 EITM has been specifically directed towards improving “technical-analytical proficiency in Political Science by bridging the divide between formal and empirical analysis” (EITM 2002: 1; see also the Autumn 2003 Special Issue of Political Analysis). The EITM movement has picked up on a number of issues raised by Most and Starr and stressed here. For example: “Specifying a model that links both formal and empirical approaches alerts researchers to outcomes when specific conditions are in place—and is also one of the best ways to determine an identified relationship” (emphasis in original; EITM 2002: 1–2). Discussing one aspect of EITM, Aldrich/Alt (2003: 310) note: “The result of this mismatch of testing strategic models with methods assuming no strategic interaction is, of course, inaccurate inference.” This echoes almost exactly analyses presented in 6

3.1 Applying the Research Triad: Theory, Nice Laws …

89

The theoretical specification of a model must take into account not only process and nice laws, but also the possibility of ‘substitutability’ (Most/Starr 1989; see also Cioffi-Revilla/Starr 1995). Substitutability may be seen as the flip side of nice laws and may also be approached through the use of stylized facts. Substitutability refers to the existence of a set of alternative modes of response by which decision makers could deal with the same situation. These behaviors have also been called “alternative modes of redundancy” (Cioffi-Revilla/Starr 1995: 456–457). Most/Starr (1989, Chap. 5) note that opportunity or willingness can operationally occur or be made available in a number of alternative, nonunique ways. Alternative possibilities or bases for choice produce ‘substitutability,’ which Most and Starr see as crucial for understanding the logic of causality, and thus, research design. Morgan/Palmer (2000) use a general meaning of substitutability where any single cause may have multiple effects, and that any single effect may have multiple causes. In the same way, McGinnis (2000) notes that different problems may lead to similar responses, or that a given problem may be dealt with in multiple ways. Bennett/Nordstrom (2000) observe that if policymakers can find multiple paths to success—actions that would all achieve the stated policy goals—then those policies may be thought of as substitutable. In their investigation of the interaction of opportunity and willingness with substitutability, Cioffi-Revilla/Starr (1995) distinguish between a first-order causality of world politics (opportunity and willingness) at the analytical level and a deeper second-order causality of substitutability at the operational level. The first-order (necessary) elements of opportunity and willingness are linked by the Boolean AND, while the range of possible modes of (sufficient) second-order substitutability are connected by the Boolean OR. Cioffi-Revilla/Starr (1995, 453) use their own version of stylized facts when they ask, “What would the world of politics be like if the first-order causality [of opportunity and willingness] were governed by disjunction (OR)? In such a hypothetical world behavior could occur when actors had political willingness to act, or when they had the opportunity to do so.” This demonstrates how stylized facts can provide a powerful way of looking at the logical and causal consequences of mis-specifying the form of the relationship. Theory, nice laws, and substitutability will all be used below in commenting on Ray’s ‘guidelines’ for the use of control variables. Can cumulation occur, and if so, do the relationships between logic, theory, and research design noted above (and inherent in Ray's arguments about the use of control variables) have an effect? Cumulation is a social process that occurs within a community of scholars. One can argue that it has occurred in areas such as the democratic peace, power transition theory, and the role of territory in conflict. Integrative cumulation, for Zinnes (1976), occurs “when earlier studies are ‘crucial’ to the conceptual and theoretical components of the subsequent study's research design” (Most/Starr 1989: 7). For example, Russett/Starr (2000) argue that it was

Most and Starr regarding problems studying the causes of war (a minimally dyadic phenomenon) with cross-national data on the characteristics of single states (1989, Chap. 3).

90

3 Cumulation from Proper Specification …

only after the actual phenomenon of the democratic peace was empirically established—that two ‘democracies’ do not fight ‘wars’ against each other—that intensive work on theories as to why this should be occurred. Such theoretical work involved intensive conceptual discussions of democracy and war, as well as their possible relationships. As theories were generated and evaluated, they were examined in a variety of different historical and regional contexts. The implications of these theories, which relied on process, were subsequently tested on a variety of different behaviors (dependent variables), both conflictual and cooperative. It was then, after a variety of theories proposed certain types of behaviors based on the nature and presence of democracy, that researchers expanded their dyadic investigations to monadic studies of democracies (with Ray being one of the first to argue strongly for such monadic effects). While Ray uses some examples from articles on the democratic peace for his critiques, this does not refute my assertion that there has been significant cumulation in the democratic peace literature. The additive cumulation, as noted, is obvious. The movement from one set of questions to another also indicates integrative cumulation. Chernoff (2004), drawing from discussions of ‘progress’ in the philosophy of science literature, argues that the study of the democratic peace exhibits the qualities of progressive science and cumulation. One important indicator, according to Chernoff (2004: 71), is the quality of the ‘debate’ or ‘dialogues’ between “defenders and opponents of democratic peace hypotheses,” demonstrating important aspects of consensus and progress. In discussing cumulation, it is important to delineate two types of consensus: it can occur with recognition of research findings as well as in agreement over theories that explain those results. Chernoff has focused more on the former. In regard to the latter, my view (based on Zinnes’ discussion) is that cumulation does not demand unanimity or strong consensus on a single theoretical position, but rather on the process of building towards fuller and deeper findings, and more fully specified theory that has developed along a variety of branches. It also involves more broadly accepted findings and theory. There is no doubt that this is currently the case regarding the basic findings of democratic peace researchers. But such cumulation also applies to the theory behind the democratic peace. Currently, very few (if any) scholars argue that either the normative/value theory or the structural/ constraint theory fully explains the democratic peace to the exclusion of the other. Rather, several related syntheses have been evolving, each based in what Russett/ Starr (2000) call the “strategic perspective” (for instance, the survival/selectorate model of Bueno de Mesquita et al. 2003). It is important to note, given the Oneal and Russett article in this issue, that Chernoff cites their work as an important element in the scientific ‘progress’ of the democratic peace research program. Indeed, the Russett and Oneal project has been driven very clearly by theory. It has moved theoretically from very simple models to the more complex model outlined in Triangulating Peace (2001, Chap. 1), using only a very small number of variables whose inclusion and relationship to one another have been theoretically and clearly specified. The spirit and message of the Achen article in this issue is to drop the garbage-can approach, where researchers

3.1 Applying the Research Triad: Theory, Nice Laws …

91

throw in all the variables they think might have an impact (including the use of large numbers of control variables). I see Achen’s Rule of Three within this context —not to be rigidly applied, but to force us to think through our theory and what variables we include. Russett and Oneal have done that quite well. With the democratic peace, then, integrative cumulation has been found, but only after hard theorizing, specification of theories and theoretical processes, the use of nice laws and substitutability, and substantial additive cumulation have taken place. It has not come through the use of wholesale, general, global models including some set of “best fit” variables.

3.2

Illustrations with Ray’s Guidelines

Ray (2003a) offers a set of five ‘guidelines’ that he believes could “improve the quality of evidence produced by multivariate analyses, as well as the credibility and intelligibility of that evidence.” Let me provide some brief comments applying my points to each of these guidelines. Guideline #1 states: “do not control for intervening variables.” Discussing an article that simply adds control variables “that previous research has shown to have an impact on war and dispute involvement,” Ray (2003a: 5) argues that research design should not treat intervening variables as control variables. The point of my arguments in the present discussion is that if one wants to move from additive cumulation to integrative cumulation, then researchers must recognize exactly what types of variables they are dealing with, and then build a ‘better’ model/theory. If the variables are truly intervening, then the model must be more closely specified to indicate in what ways the new variables are part of the story, what that means for the implications of the model, and under what conditions the model will hold. A good example is found in the Kadera and Mitchell article in this issue. In a study of one of the implications of theories explaining the democratic peace, Kadera and Mitchell investigate whether international institutions intervene in the relationship between the strength of democratic community and third party conflict management. Rather than just control for joint membership in international organizations, they look at alternative stories—or processes—of how these three variables could be related, and then devise research designs to evaluate these relationships. Kadera et al. (1964), note that we should not control for a factor that is part of the causality process. Thus, rather than simply throwing what could be an intervening variable into the hopper of control variables, we must specify (or re-specify) theories and theoretical processes, and look again at the logical form of the relationships implied. Kadera and Mitchell conclude that international organizations do not appear to be an intervening variable, and thus can move on to other considerations. Similar comments apply to the next three guidelines as well. Specifically, Guideline #2 states: “distinguish between complementary and competing explanatory factors.” Ray (2003a, 7) notes that in many cases, control variables are simply “alternative causes” and that “rival explanations” are simply alternative

92

3 Cumulation from Proper Specification …

causes. One major point developed in Most and Starr, and deriving from nice laws, is the position that very few models or theories are actually ‘contending’; the key question, proposed in Most/Starr (1989), is rather, “how can both be true?” That is, under what conditions does each hold? Thus, clearer specification and the logic of nice laws helps to support Ray’s guideline. One must also investigate whether ‘alternative’ causes are truly ‘alternative.’ That is, does substitutability come into play here? Are the alternatives simply substitutable, second-order factors by which the behavioral units involved satisfy the first-order factors of opportunity or willingness? (See Cioffi-Revilla/Starr 1995.) Similarly, all of these points apply also to Guideline #3: “do not introduce factors as control variables merely on the grounds that they have an impact on the dependent variable,” and #4 as well: “do not control for variables that are related to each other or the key explanatory factor by definition.” Especially regarding #4, substitutability can be used here to distinguish between truly substitutable factors or merely multiple measures of the same factor. Again, for #3 and #4, taking nice laws and substitutability into account drives the researcher to more clearly and fully specified theory. Ray (2003b) discusses the complexity of multivariate analyses more fully. However, Most and Starr reach many of the same conclusions as Ray by discussing the use of quasiexperiments, and looking at different factors as ‘treatments’ which are either present or absent, and the relationship of the treatment to the occurrence of some dependent variable (see Most/Starr 1980, and Siverson/Starr 1991, for the use of these ideas in the study of the diffusion of conflict and war). Ray (2003b: 11) notes, “Basically, my argument here is that multivariate models of interstate conflict should be simplified.” This can be done in many cases using the ‘treatment’ strategy. Nice laws would also ask for simplifying analyses by looking at how relationships stand up to different conditions, using methods as simple as contingency tables (a strategy noted by Achen in this issue; see also Ray 2003b: 5). For example, Ray quotes Achen’s (2002: 446) Rule of Three: “a statistical specification with more than three explanatory variables is meaningless.” As noted above, Achen critiques the inclusion of long lists of explanatory variables into equations in order to control for their effects, and specifically argues that linear link functions are not ‘self-justifying.’ He argues that linear links can make good equations go bad. Such variables should only be included (as I have argued here) for good theoretical reasons. Such variables should be included only after fairly extensive preliminary data analysis reveals what Most and Starr called “the form of the relationship” to the analyst. Indeed, as I have noted above, Achen explicitly notes that preliminary data analysis should include “careful graphical and crosstabular analysis.” While appearing in his discussion of guideline #4, the following example provided by Ray (2003a: 18) illustrates these points: Running two different analyses, one with contiguity as the control variable for geographic proximity and another with distance between capital cities as a control variable, is a perfectly legitimate strategy. Selecting the model with the “best” results from the point of view of the analyst is also justified, especially if the analyst can come up with some

3.2 Illustrations with Ray’s Guidelines

93

plausible theoretical conjecture that might account for the differences in the two analyses. This strategy is clearly preferable, in my view, to one involving the inclusion of both contiguity and distance as control variables in the same model or analysis.

Let me waffle here. I fully agree with the general principles in Ray’s comments in terms of how we treat two variables that are similar or measure the same concept. The specific example noted, however, is useful in that it shows how researchers can differ in their interpretations of concepts and indicators. Oneal and Russett in this issue have responded by indicating what they see as differences in the concepts of proximity and distance from that of contiguity, and how the measures may tap different things. While I might disagree with their specific positions (waffle again), the key point is that Oneal and Russett have presented reasonable theoretical/ conceptual arguments for using both variables. Ray's commentary is reasonable in questioning a design that (in his interpretation) uses two indicators of the same phenomenon. Guideline #5 states: “control for possible differences between across space and over time relationships.” Within the spirit of nice laws, #5 looks at the most frequently used and perhaps the most important of those conditions that would vary and affect theapplicability of our models/theories. Time and space are key elements in many studies that look to see “under what conditions” a model or theory might hold. Starr (2003) argues that, generally, time has taken precedence over spatial factors (and for a variety of understandable reasons). Most of our research has stressed the temporal dimension or temporal context. This is easily seen in the design of our research, where temporal patterns are central, the use of time series data and designs are standard, and the use of time to delineate of our units of analysis is standard. Scholars routinely divide time into historical eras or different international ‘systems’ with a temporal boundary. We do so because we think that theoretically (either explicitly or implicitly) some conditions have changed so much, or on some important dimension, that these changed conditions should alter the way that humans behave or how our theories should work (or cease to work). Changes in technology, especially, as well as the changes in geopolitical structure that often follow wars, also affect crucial spacetime relationships (e.g., Boulding’s loss-of-strength gradient). Simply put, guideline #5 is extraordinarily important, especially from a nice law perspective. These five guidelines for the use of control variables are presented by Ray as a response to one statistical problem that occurs in the research on international conflict. However, the problems that he identifies with the improper use of control variables represent broader issues in the relationship between theory, logic, and research design. Ray (2003a: 28) points to the broader issue of cumulation in his conclusion: Were these guidelines to be adopted in future research efforts, the results of empirical evaluations of multivariate models might be more readily interpretable and understood. Ideally, research on international conflict would then be more cumulative, as well as more productive of valuable insights into those processes leading to interstate conflict and war.

94

3 Cumulation from Proper Specification …

The present article has also attempted to indicate how the issue of control variables can be tied to a more general concern with cumulation. In so doing I have tried to indicate the utility of returning to a number of concepts and strategies to improve the cumulative research enterprise initially raised by Most/Starr (1989).

References Achen, C., 2002: “Toward a new political methodology: Microfoundations and ART”, in: Annual Review of Political Science, 5: 423–450. Aldrich, J., Alt, J., 2003: “Introduction to the Special Issue”, in: Political Analysis, 11, 4: 309–315. Bennett, D.; Nordstrom, T., 2000: “Foreign Policy Substitutability and Internal Economic Problems in Enduring Rivalries”, in: Journal of Conflict Resolution, 44, 1: 33–61. Blaylock, H. M., Jr., 1964: Causal Inferences in Nonexperimental Research (Chapel Hill: University of North Corolina Press). Braumoeller, B. F., 2003: “Causal Complexity and the Study of Politics”, in: Political Analysis, 11, 3: 209–233. Braumoeller, B. F.; Goertz, G., 2003: “The Statistical Methodology of Necessary Conditions”, in: Goertz, G.; Starr, H. (Eds.): Necessary Conditions: Theory, Methodology, and Applications (Lanham, MD: Rowman & Littlefield). Bremer, S. A., 1992: “Dangerous Dyads: Conditions Affecting the Likelihood of Interstate War, 1816–1965”, in: Journal of Conflict Resolution, 36, 2: 309–341. Bremer, S. A., 1996a: “Advancing the Scientific Study of War”, in: Bremer, S. A., Cusack, T. R. (Eds.): The Process of War: Advancing the Scientific Study of War (Philadelphia: Gordon and Breach). Bremer, S. A., 1996b: “Final Words”, in: Bremer, S. A.; Cusack, T. R. (Eds.): The Process of War: Advancing the Scientific Study of War (Philadelphia: Gordon and Breach). Bremer, S. A., Cusack, T. R., (Eds.), 1996: The Process of War (Philadelphia: Gordon and Breach). Bueno de Mesquita, B.; Smith, A.; Siverson, R. M.; Morrow, J. D., 2003: The Logic of Political Survival (Cambridge, MA: MIT Press). Chernoff, F., 2004: “The Study of Democratic Peace and Progress in International Relations”, in: International Studies Review, 6, 1: 49–77. Cioffi-Revilla, C.; Starr, H., 1995: “Opportunity, Willingness, and Political Uncertainty: Theoretical Foundations of Politics”, in: Journal of Theoretical Politics, 7, 4: 447–476. Eckstein, H. S., 1975: “Case Study and Theory in Political Science”, in: Greenstein, F., Polsby, N. (Eds.): Handbook of Political Science, vol. 7, Strategies of Inquiry (Reading, MA: Addison-Wesley). Empirical Implications of Theoretical Models Report. (2002). EITM Workshop, July 9–10, 2001. Goertz, G., 2003: “Cause, Correlation, and Necessary Conditions”, in: Goertz, G.; Starr, H. (Eds.): Necessary Conditions: Theory, Methodology, and Applications (Lanham, MD: Rowman & Littlefield). Goertz, G., 2005: “Necessary Condition Hypotheses as Deterministic or Probabilistic: Does it Matter?” in: Qualitative Methods, 3, 1: 22–27. Goertz, G.; Starr, H. (Eds.), 2003a: Necessary Conditions: Theory, Methodology, and Applications (Lanham, MD: Rowman & Littlefield). Goertz, G.; Starr, H., 2003b: “Introduction: Necessary Condition Logics, Research Design, and Theory”, in: Goertz, G.; Starr, H. (Eds.): Necessary Conditions: Theory, Methodology, and Applications (Lanham, MD: Rowman & Littlefield).

References

95

Holsti, K. J., 1991: Peace and War: Armed Conflicts and International Order, 1648–1989 (Cambridge, UK: Cambridge, University Press). Lave, C. A.; March, J. G., 1975: An Introduction to Models in the Social Sciences (New York: Harper and Row). McGinnis, M. D., 2000: “Policy Substitutability in Complex Humanitarian Emergencies: A Model of Individual Choice and International Response”, in: Journal of Conflict Resolution, 44, 1: 6289. Morgan, T. C.; Palmer, G., 2000: A Model of Foreign Policy Substitutability”, in: Journal of Conflict Resolution, 44, 1: 11–32. Most, B. A., Starr, H., 1980: “Diffusion, Reinforcement, Geopolitics, and the Spread of War”, in: American Political Science Review, 74, 4: 932–946. Most, B. A., Starr, H., 1989: Inquiry, Logic and International Politics (Columbia, SC: University of South Carolina Press). Palmer, G., (Ed.), 2000: Special issue on “Substitutability in foreign policy: Applications and advances”, in: Journal of Conflict Resolution, 44, 1. Ray, J. L., 2003a: “Explaining Interstate Conflict and War: What Should Be Controlled For?”, in: Conflict Management and Peace Science, 20, 1: 1–31. Ray, J. L., 2003b: “Constructing Multivariate Analyses (of Dangerous Dyads)”. Paper presented at the 2003 Peace Science Society (International) meeting, Ann Arbor, November. Russett, B.; Oneal, J., 2001: Triangulating Peace (New York: W.W. Norton). Russett, B., Starr, H., 2000: “From Democratic Peace to Kantian Peace: Democracy and Conflict in the International System”, in: Midlarsky, M. I. (Ed.), Handbook of War Studies II (Ann Arbor: University of Michigan Press). Siverson, R. M.; Starr, H., 1991: The Diffusion of War: A Study of Opportunity and Willingness (Ann Arbor: University of Michigan Press). Starr, H., 1996: “Advancing the Scientific Study of War: A Commentary”, in: Bremer, S. A., Cusack, T. R. (Eds.): The Process of War: Advancing the Scientific Study of War (Philadelphia: Gordon and Breach). Starr, H., 2003: “The Power of Place and the Future of Spatial Analysis in the Study of Conflict”, in: Conflict Management and Peace Science, 20, 1: 1–20. Zinnes, D. A., 1976: “The Problem of Cumulation”, in: Rosenau, J. N. (Ed.), In Search of Global Patterns (New York: Free Press).

Chapter 4

Opportunity, Willingness and the Diffusion of War, 1816–1965

Using borders and alliances as indicators of opportunity and willingness, respectively, we test the relationship between these and the diffusion of war during the 1816–1965 period.1 The impact of borders and alliances, individually and in combination, on the growth of ongoing war through ‘infectious’ diffusion is shown through the comparison of baseline cases to cases where states at peace were exposed to various ‘treatments’ comprised of warring border nations or warring alliance partners.2 The findings indicate that the probability of war diffusion is substantially increased as opportunities and willingness increase, particularly when such geographic and political factors are combined. The applicability of the opportunity and willingness framework to the study of war and diffusion is expanded and confirmed.3

A considerable amount of the early empirical research on war attempted to explain its onset by looking at the effects of one or several independent variables on a dependent indicator variable, such as, for example, the nation-months of war (Singer et al. 1972; Singer/Small 1968) or the number of nations at war (Singer/ Small 1974). The initially unrecognized problem with this procedure is the conflation of the onset of war (a dichotomous variable) with the size of a war (a This text was first published as: “Opportunity, Willingness and the Diffusion of War”, in: The American Political Science Review, Mar 1990, vol. 84, No. 1, pp. 47–67. The permission to republish it here was granted by Cambridge University Press. 2 We acknowledge with thanks the helpful research assistance of Sherry Lutz, Drew Froeliger, and Juliann Emmons in the generation and analysis of the data and the timely programming assistance of Brenda Gunn. The collection of the data was supported by the Institute for Governmental Affairs, University of California, Davis. 3 While diffusion has been given several meanings, and has a number of dimensions, we will be using it essentially to mean the growth of an ongoing conflict, the process by which states join an ongoing war and the scope of the conflict becomes enlarged. The concept of diffusion and its theoretical relevance to the study of international conflict is reviewed extensively in Most et al. (1989). For other reviews of the diffusion concept, see Most/Starr (1981), O’Loughlin (1984) and Welsh (1984). For an overview of the renewed interest in geopolitical approaches (including diffusion) to war, see Diehl (1988). For a classic statement on the contagion of political conflict, see Schattschneider (1960). 1

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7_4

97

98

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

continuous variable). The problem, however, has consequences significantly beyond what type of measurement is appropriate, since by using the size of a war while the theory under investigation specifies that onset is being measured, the distinct possibility of diffusion is overlooked. This means that the process by which the first two nations in a war begin fighting may be considerably different than the process by which subsequent participants join the war. This blurred distinction neglects what is usually referred to as Galton’s problem (Ross/Homer 1976). Under circumstances where diffusion is present but unrecognized, the resulting models are necessarily misspecified and most probably investigated with inappropriate methods. Because of the recognition of this problem, more recent research has moved in two new directions. One of these focuses on the behavior of the initial participants in wars and attempts to explain only their behavior. To a considerable extent this is accomplished by considering the onset of war as the end product of a dispute between nations (Bueno de Mesquita 1981; Bueno de Mesquita/Lalman 1986; Leng/Gochman 1982; Maoz 1982). A second line of research is based on the recognition that wars might diffuse, or be ‘contagious.’4 In general, research on the diffusion of war began with determining the extent to which wars were ‘infectious’ (Davis et al. 1978; Faber et al. 1984; Most/Starr 1981) and, after establishing that they were, seeking to uncover the factors responsible for variation in the diffusion processes. In general, two lines of investigation have been followed on the diffusion of war. The first of these is based upon borders as interaction opportunities (Most/Starr 1980; O’Loughlin 1984; Ward/Kirby 1987). The second centers on alliances as indicators of groups of states that share roughly the same international policies and may be willing to fight together for them (Altfeld/Bueno de Mesquita 1979; Siverson/King 1980). These two lines of research on the diffusion of war—borders and alliances— have only recently been connected in preliminary empirical analysis (Most et al. 1987; Siverson/Starr 1988). In the present research the data set has been expanded significantly over previous collections, new analytic tools have been added and greater attention is given to interaction of method, theory, and empirical findings. In particular, the research we present brings borders and alliances together within a theoretical framework based on the ideas of opportunity and willingness. The use of this framework permits us to examine ongoing wars as events that alter the incentive-constraint structures perceived by foreign policy decision makers, thus increasing the chances that nations will become involved in an ongoing war. The analyses presented here will be conceived in terms of a research design in which

4

We use the term treatment to emphasize the epidemiological nature of our research, not to suggest that we are performing an experiment in which we have the ability to randomize and control exposure. However, what we report is in many respects a historical experiment, similar in method to the procedures described by Singer (1974) and Holsti/North (1965).

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

99

particular conditions or sets of conditions are ‘treatments,’ which may or may not produce effects.5

4.1

Opportunity and Willingness in the Diffusion of War

Following the work of Most and Starr, the diffusion analyses presented here will be developed within the general framework provided by the concepts of opportunity and willingness (Most/Starr 1989, Chap. 2; Starr 1978a). As ordering concepts, they form the basis for linking environmental and systemic factors to the behavior of decision makers and governments that represent states. By opportunity we mean the possibilities that are available to any entity within any environment, representing the total set of environmental constraints and possibilities. While opportunity thus represents macro level (environmental and structural) factors, willingness represents the choice processes that occur on the micro level, that is, the selection of some behavioral option from a range of alternatives. This framework is derived from Sprout and Sprout’s (1969) “ecological triad” of the relationship among entity, environment, and entity-environment. In it their concepts of environmental possibilism, environmental probabilism, and cognitive behaviorism require the combination of both structure-environment and choice-decision process, captured in opportunity and willingness, respectively. Thus, opportunity and willingness are concerned with the relationships that nest decision makers within their surrounding environments. The central use of opportunity is as the degree of interaction. This conception of opportunity has been the primary impetus in Most and Starr’s work on the effects of borders as interaction opportunities in the diffusion of violent conflict. As in Sprout and Sprout’s environmental possibilism, this simply means that some activity must at base be physically, technologically or intellectually possible. Once the obstacle of possibility is crossed, however, opportunity is, in fact, a continuous phenomenon in which some nations have more or less of it with respect to other nations. The dual nature of opportunity-possibility (especially as it relates to capabilities) must be recognized. Initially, some capability (technology, ideology or religion, form of government, manner of organizing people to some task, etc.) must be created so as to be part of the range of possibilities available to at least some 5

Although research on this point is not clear (Garnham 1988), it is reasonable to suspect that under some circumstances geographical proximity to, or shared borders with, some other nation will in themselves influence a nation’s willingness to become involved in conflict. If so, opportunity and willingness are not independent, as we noted. Indeed, borders themselves can be seen in terms of environmental possibilism by providing possibilities for interaction. The warring border nation model, however, can be seen in terms of cognitive behaviorism or willingness, as activities on a state’s borders effect the perceptions of threat, uncertainty, and opportunity held by that state’s decision makers. Clearly, most nations are likely to be sensitive to the security concerns of neighboring nations or, when the capabilities are present, are likely to see their neighbors as the greatest potential threats to their own security.

100

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

members of the international system. However, there is then an important second dimension in the effects of opportunity-possibility—the distribution of such capabilities in the international system. At one level all international actors share the same menu of possibilities—for example, no nation could have had nuclear weapons to fight World War I; the United Nations could not have been used to separate the United States and Mexican forces in 1846; and any nation may avail itself of the International Court of Justice. At another level international actors may have very different menus of possibilities—for example, the wealth, technological talent, and resources needed to take advantage of the nuclear possibility are not evenly distributed across nations today. The concept of willingness is more familiar, being central to the study of decision making and choice. The dynamics of choice are embedded in a decision maker’s image of the world, or definition of the situation. Willingness is related to a decision maker’s calculations of advantage and disadvantage, cost and benefit, considered on both conscious and unconscious levels. It is through willingness that decision makers recognize opportunities and then translate those opportunities into alternatives that are weighed in some manner. As is implied above, opportunity and willingness are linked in a number of ways. They do not create mutually exclusive categories. Anything that affects the structural possibilities of the environment or environments within which decision makers must act also affects the incentive structures for those decision makers. Opportunity and willingness thus become more than organizing concepts. They take on theoretical characteristics when we understand that they describe the conditions that are necessary for the occurrence of events. They are necessary but not sufficient for the occurrence of international outcomes such as war. Opportunity, conceived of as interaction opportunity, has been central to the geopolitical study of positive spatial diffusion. This has been most explicitly and extensively developed by Most and Starr, who use borders as the interaction opportunity through which violent conflict would spread. Borders are investigated and conceptualized as “constraints on the interaction opportunities of nations” (Starr/Most 1976). Simple geographic proximity, as indicated by borders of various kinds, is related to both the amount of interaction and the probability of various types of interaction, such as war (see Starr/Most 1978). To some extent, borders represent a proximity between states that increases both the salience of the neighboring territories and (in general) the ease of interaction. Thus, because states are probably more likely to be more attuned to, and involved in, political situations on their borders, there is a tendency for opportunity to shape the range of willingness—although power certainly affords some nations the possibility of very wide openings for defining their borders. Briefly, the various works of Most and Starr argue that opportunity for interaction is a necessary condition for the positive spatial diffusion of violent conflict, that borders are one factor providing such an opportunity, and that the empirical evidence for the 1946–65 period indicates that positive spatial diffusion of violent conflict occurred along the easements provided by borders.

4.1 Opportunity and Willingness in the Diffusion of War

101

It needs to be emphasized that within such a warring border nation framework it is not reasoned that borders cause wars but rather that they contribute to the potential outbreak of violence because the more borders a nation has, the greater (1) the number of risks and opportunities confronting the nation, (2) the likelihood that the nation or its territories will be “conditionally viable” (Boulding 1962), and (3) the level of that nation’s uncertainty. Under these conditions, it is asserted, nations have a greater probability of going to war (see also Diehl/Goertz 1988). Additionally, Most and Starr argue that once a war starts on a nation’s border, that nation may find its environment changed in such a way that it either participates in a war it did not intend to join or forgoes fighting a war that it had intended to join. The warring border nation model thus analyzes whether states that experience the treatment of having a warring nation on their borders have their environment and decision calculus altered so as to alter the probability of their own subsequent war involvement, or not. (See Starr/Most 1985 for a full elaboration of the warring border nation model.) While borders can in general be conceptualized as an ‘agent’ of diffusion representing the effects of opportunity, another such agent, alliances, represents willingness6 (see Footnote 3). Numerous observers of alliances in international relations have commented on their entangling nature. Following this general line of argument, a second approach to the problem of war diffusion has thus focused on the extent to which alliance commitments could be responsible for the diffusion of wars. Siverson/King (1979,1980) explore the extent to which the independent effect of alliance memberships and the attributes of different types of alliances account for the extent and character of war diffusion. Recognizing the work of Most and Starr on diffusion and borders, Siverson and King argue that alliances, unlike geography, result from a deliberate process of policy choice. States clearly have much greater latitude in their choice of allies than the states on their borders.7 It is thus important to recognize alliances as manipulable interaction opportunities. More importantly, it is reasonable to look at alliances as a conscious choice among foreign policy behaviors or policy positions. The willingness to form alliances—and with specific partners—may be seen as an indicator of shared policy preference. Put simply, two (or n) nations forming an alliance are indicating, to 6

Of course, states may attempt to manipulate their immediate political geography by creating neighbors (e.g., Belgium in 1830) or eliminating them (e.g., Poland in 1939), but the costs of these activities are likely to be quite high. 7 Alliances, of course, are not a perfect indicator of willingness. For example, while there is general tendency for alliances to be reliable, it is clear that reliability is less than complete (Sabrosky 1980). In addition, some alliances are specific to certain issues or geographic areas (e.g., NATO), and their existence does not mean that one ally will join another no matter where the conflict. Also, alliances signed at one time may deteriorate while continuing to exist formally. In short, we do not construe alliances as indicating a general willingness to fight. However, alliances have long been recognized as the key means that nations have chosen to indicate their political position in the international system. Formal alliances are matters of serious concern to decision makers and in addition to their presumed rewards of enhanced security, they impose costs and risks both domestically and internationally (Sullivan 1974).

102

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

some degree, that they share policy preferences.8 This conception of alliances, implicit in the work of Siverson and King, is explicitly developed in Bueno de Mes-quita’s (1981) model of expected utility, an approach centered on the calculations that lead to willingness.9 The research we report here seeks to investigate the effects of both opportunity and willingness on the diffusion of war. This will be done by looking at the individual and combined effects of borders and alliances on the diffusion of war among members of the international system over the period 1816–1965.

4.2

Research Design and Data Generation

While several studies of conflict diffusion have relied on varieties of autocorrelation (Hill/Rothchild 1986; O’Loughlin 1984), the method we chose is considerably simpler. It has much in common with various epidemiological models in that it examines the magnitude of the effects of being exposed to various conditions as treatments. More specifically, it allows us to explore various changes in the probability that nations will enter an ongoing war if they have a warring border nation (WBN) or a warring alliance partner (WAP). In order to evaluate the impact of WBNs and WAPs on the diffusion of war we constructed a data set containing several types of information on national borders, national alliance commitments, and national war participation. Data were collected and coded in order to test both a WBN hypothesis and a WAP hypothesis. Thus, the WBN hypothesis and the WAP hypothesis may be tested separately, and the strength of the results compared. The data also permit us to look at the impact of a nation’s bordering states that are also allies. This will enable us to test the combined treatment of a nation’s being both a WBN and a WAP. Again, the results of these analyses may be compared to the results of the analyses of each factor singly. In order to make the appropriate comparisons, however, it is necessary to do more than simply record the relevant border and alliance information for the instances in which nations entered a war. Doing so would tell us something about the process of diffusion, but such a procedure would deal only with ‘successful’ cases in which diffusion took place. The cases in which a nation experienced some type of either a WBN or WAP but did not enter the war would be lost from view;

8 In fact, Altfeld/Bueno de Mesquita (1979) successfully apply expected utility theory to the problem of explaining which side nations will join in a war. While this is a problem similar to the one pursued here, it does not directly address the question of diffusion. For some other differences, see n. 16. 9 Following Small/Singer (1982), the major powers and the years of their inclusion in that group are Austria-Hungary (1816–1918), Italy (1861–1943), the United Kingdom (1816–1965), Russia* USSR (1816–1965), Japan (1895–1945), Prussia-Germany (1816–1945), the United States (1989– 1965), France (1816–1940, 1945–65) and China (1949–65).

4.2 Research Design and Data Generation

103

hence, no estimate could be made of the effect of the variable on the larger population within which these nations exist (see Most/Starr 1989, esp. Chaps. 3–5). In order to pursue our purpose of assessing the joint and individual effects of borders and alliances on the war experiences of states, it was necessary to bring together data on war participation, alliances, and borders of each state in the international system for each year between 1815 and 1965. Information on the set of states in the system, war participation, and national alliance commitments was relatively easy to acquire. The set of states was taken from the lists provided in Small/Singer (1982). We also noted the power status of the various nations, using a simple division of nations into either major power or minor power status.10 There are, of course, several generally available data sets on international war, including, most notably, those of Kende (1971, 1978), Richardson (1960), Small/ Singer (1982), and Wright (1965). Most and Starr’s initial diffusion analyses used a combination of all these sources, as those analyses, in actuality, investigated the diffusion of any organized violent conflict (whether it was civil war, internal conflict, or intervention, or relatively small-scale violence) as well as large-scale organized interstate violence. We, however, focus on interstate war for the entire state system over the post-Napoleonic era. We thus selected the larger-scale interstate war data set presented in Small and Singer 1982, which represents an updating and refinement of a well-established earlier effort by the same authors (Singer/Small 1972) to present the data of the Correlates of War Project. There were several other reasons to select this war data set. First, it incorporates a great deal of the information contained in the earlier studies by Wright (1965) and Richardson (1960). The data used here cover more contemporary events than the compilations by Wright and Richardson but also cover much earlier periods than the post-1945 data of others (e.g., Kende). Finally, utilizing Small and Singer’s data will make our findings compatible with the growing body of empirical work that has derived from the Correlates of War Project (Gochman/Sabrosky 1990). The alliance data also are a product of the Correlates of War Project. In this case we drew upon Sabrosky’s (1975) extensive revision of an earlier work by Singer/ Small (1968) that provided an initial listing of formal international alliances between states during the period 1815–1965. In the case of each alliance, we identified its class as coded by Singer and Small: defense, neutrality, or entente. These will be referred to as A1, A2, and A3, respectively. We view these types of alliance commitments as forming an ordinal index of willingness. Using Sabrosky’s discussion, Garnham (1988, 15) outlines these three types of alliances as developed by the Correlates of War Project: “The strongest alliance commitment is a defense pact... in which the signatories agree to intervene 10

This third category of data was the most difficult to obtain because of the number and almost unbelievable obscurity of some colonies. We were aided very considerably by Henige’s (1970) comprehensive list of colonial governors. This list, which begins with the fifteenth century, identifies all political units that were the colonial possession of some other nation. Finding them in this list then permitted relatively easy location in one of the historical atlases.

104

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

militarily in the event of an attack on one of their number. Next, insofar as the formal strength of the alliance is concerned, is the neutrality or nonaggression pact … which obligates the signatories to remain militarily neutral should one of them become involved in a war… Finally, the entente … merely required consultations or conversations if one of the signatories was attacked.” Thus, we expect that WAPs involving defensive alliances (Al) will have a greater likelihood of producing diffusion than nonaggression pacts (A2) or finally the ententes (A3). The most difficult data to gather were those indicating which states shared borders. However, through the use of several excellent historical atlases (Shepherd 1932 and Hammond Historical Atlas of the World), it was possible to ascertain the border network of states back to 1815. Specifically, in a modified version of the coding rules used by Starr/Most (1976), we recorded for each state the entities on its contiguous borders, those across less than two hundred miles of open water and those on the borders of its colonial possessions.11 We refer to these as B1, B2, and B3, respectively. As noted, we regard the divisions of the border and alliance variables as having an ordinal value. In terms of the opportunity and willingness concepts discussed we hypothesize that the greatest opportunity is present with contiguous borders, then cross-water borders, and finally, colonial borders (Starr/Most 1976). At that point considerable variation is introduced. Similarly, the greatest willingness should be present with defense alliances, then neutrality agreements, and then ententes. Using the nation-year as the basic unit of analysis, the completion of the data collection eventuated in a data set of 3929 cases, representing a coding of the borders, alliances, and war participation of all nations in all years between 1816 and 1965 in which a war either started or was ongoing. However, to make our test as rigorous as possible we removed a number of cases. First, in order to test for diffusion as the growth of war it was necessary to remove the initial two participants in a war. Second, once a nation entered a war it was deleted from the data set until the war was concluded or the nation left the war. These reductions produced a final data set of 3749 cases. Within this data set there were 94 cases of war diffusion whose existence we will attempt to explain on the basis of opportunity and willingness.12 11

Copies of the coding scheme employed in this study are available from Siverson. It should be noted that in general we followed the data given by Small/Singer (1982) with respect to war expansion. We did, however, depart from their delineation of what constituted war joining in a few cases. For example, they show Japan as joining World War II on 7 December 1941. While it is true that Japan began fighting the United States on that day, it is also true that the Japanese decision had relatively less to do with the European war that had been in progress since 1939 than it did Japan’s war with China that had been ongoing since 1937. Hence, we do not treat Japan as joining the war. Also, we do not treat as war joiners the nations that left World War II (albeit briefly) and then changed sides (i.e., Italy, Romania, and Bulgaria) to reenter it. It makes sense to count the initial participation only. It should be noted that including these cases would have favored the opportunity and willingness hypotheses. 12 Recall that Most/Starr (1980) distinguished between reinforcement, or addiction (where a state’s war behavior affects the probability of its own subsequent behavior) and diffusion (where a state’s

4.2 Research Design and Data Generation

105

In order to differentiate positive spatial diffusion from both reinforcement phenomena and negative spatial diffusion, a set of expectations was derived and tested. The procedure used here is derived from one used in an earlier study by Starr/Most (1983, pp. 110–11).13 In their design they began by “looking for all states at any given point in time which were at peace (to avoid complications with ‘reinforcement’ effects) and asked two questions: (1) at what point in time did they have any warring border nations? (2) Within the next five years did they have any new war participations?” There is one major deviation in the present study from the design set out above. Instead of looking at a treatment at time tO and subsequent behavior only during the t1–t5 period, we have looked at all years (tO) in which the international system was experiencing war (some 83 years between 1816 and 1965) and examined the extent to which nations not at war in t − 1, either were or were not exposed to various combinations of WBNs or WAPs and either did or did not go to war in year tO. Recall that once a nation enters a war, it is removed from the data set until the war is ended or the nation leaves the war. Under this procedure it is possible then to aggregate the results of all war years for each of the treatments. This procedure made it possible for us to estimate and compare the effects of various combinations of WBNs and WAPs. Moreover, by running identical tables for the conditions of any border or any alliance we were able to estimate the extent to which WBNs and WAPs made a difference over counterpart conditions where war as a condition or treatment was absent; that is, we could compare the effects of having a contiguous WBN to the effects of simply having contiguous border nations or compare the effects of having a defense pact WAP to the effects of simply having defense pact partners. Delineating the effects of the various types and combinations of borders and alliances without WBNs or WAPs thus gives us a set of baselines for comparing the war-joining rate of those states with WBNs or WAPs. The logic of our investigation of diffusion centers on the notion of treatment, that is, the notion that the environment of the decision makers of states will have been changed by the existence of war in a bordering nation or in one of its alliance partners. As developed by Starr and Most in regard to borders (1976), and in terms of foreign policy decisions in general (Most/Starr 1989), such treatments may alter the opportunities facing decision makers as well as their willingness to pursue certain behavioral alternatives, such as going to war. The argument also follows our notions that states’ interactions will tend to follow along the paths provided by interaction opportunities such as proximity (borders) and common policy interests-high value salience (alliances).

war behavior affects the probability of the subsequent war behavior of other states). They also recognized that either could have positive or negative effects (either increasing or decreasing the probability of war behavior). The reader must be alerted that the matrix employed in the present analyses follows that utilized in Most et al. (1987) and is not the same as the matrix used in Most/ Starr (1981) or Starr/Most (1983, 110–11). 13 The single case of war involvement by a nation with no bordering nations of any kind and no alliance involvement is New Zealand’s entry into World War II in 1939.

106

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

The theoretical basis of opportunity and willingness, as well as for interaction opportunity, has been developed at length by Most/Starr (1980, 1989). The WBN model derives from the notion of interaction opportunity—the physical possibilities for interaction, factors that increase the probability that such interaction will take place, and the perception of both possibility and probability by decision makers. This idea is rooted in the ecological and geopolitical concepts and formulations of Sprout/Sprout (1969), as later modified by Starr’s concepts of opportunity and willingness. It should be recognized that the concept of opportunity encompasses important components of the geographical theory of proximity and its effects (Boulding 1962; Gleditsch 1969; Zipf 1949). For example, Boulding’s loss-of-strength gradient links geographical components to threat perception and the role of uncertainty (Midlarsky 1975). Our expectations on the effects of such treatments also may be seen within a “loose necessity” framework; that is, these treatments are to be seen as loosely necessary but not sufficient for influencing the war-joining behavior of states. We are not arguing that such treatments always lead to joining ongoing wars but that war behavior is much more likely to occur if such treatments have occurred. This is the ‘loose’ aspect of the logically necessary relationship between treatment and consequence. The key to these expectations is cell d in Table 4.1, which sets out our basic treatment matrix. Cell d indicates that a state has joined a war subsequent to being exposed to the relevant treatment. The central expectation is that there should be a clear and strong difference in cell d between matrices where there is a treatment and matrices where there is no treatment; that is, analyses presented will contrast the results found in a matrix where the columns simply note the presence or absence of borders or alliances to the results of a matrix that looks like the one presented in Table 4.1, where warring border nations or warring alliance partners comprise the treatment columns. This is a simple idea: being exposed to a treatment will, using Alcock’s (1972) phrase, increase the chances of “catching the war disease.” We ask what the world should “look like” if a state has a bordering state at war or an alliance partner at war. Our expectation is that a WBN or WAP should substantially increase the probability of that state’s joining the ongoing war of its WBN or WAP. A further expectation is that this probability should increase with the combination of border and alliance treatments. Thus, we have taken one of several possible approaches to diffusion—the growth of ongoing wars. We have identified two possible (“loosely necessary”) agents by which those wars would grow. We have then posited a simple expectation of behavior given ‘treatments’ by those agents. The following section presents data that simply but directly test these expectations. Table 4.1 The border-alliance treatment matrix

War participation Absent Present Source The authors

Treatment Absent

Present

a c

b d

4.3 Findings

4.3

107

Findings

Initially, it may be noted that the overall rate of war diffusion is rather small. In the 3746 nation-years in our data set there were only 94 cases of war diffusion. The overall rate of diffusion is thus 2.51%. War diffusion is, to be sure, a relatively rare event, but rare does not mean ‘unimportant.’ Many statistically rare events are of considerable interest to scientists, particularly when their consequences are either highly lethal or very costly. For example, both lung cancer and earthquakes are both relatively rare, but are nonetheless the objects of attention for very large, nationally organized research efforts. War diffusion is both lethal and costly. Of the 240 cases of international war involvement in the Singer and Small data set, the 94 cases we studied represent 39% of the total. Additionally, a great deal of the severity of war (i.e., how many die) may be explained in terms of the expansion or growth of war. In fact, from the data in Small/Singer’s (1982) consideration of lethality in warfare, it is clear that most of the war-related deaths take place because of war expansion. Because war diffusion is a relatively rare event, it may be seen that under “loose necessity” a large number of treatments result in a much smaller number of cases of war diffusion. In this respect several points should be noted. First, we investigate here only one form of diffusion—infection, or the growth of an ongoing war. We do not deal with demonstration effects. Also if only the opportunity for interaction was of concern, we might expect ongoing wars to grow to include all those nations with opportunity. However, willingness is important in such decisions. Put differently, decision makers choose behavior within the constraints posed by the range of incentive structures within which they are imbedded. The effects of WBNs and WAPs are just one aspect of that structure. While other aspects of the structure may lead to the willingness to choose other foreign policy behaviors, it is impressive that the opportunity and willingness model of interaction opportunity is able to identify WBNs and WAPs as factors having a significant impact on war-joining behavior. The data analysis will take place in two stages. First, we will explore the extent to which the variables measuring opportunity and willingness have anything to do with war expansion. Put differently, we will show that there is a significant relationship between the various measures of opportunity and willingness and proclivities to join a war. That done, the second part of our data analysis will be to examine how various individual and combined treatments affect war-joining behavior.

4.4

Do Opportunity and Willingness Influence War Diffusion?

From the previous discussion of opportunity and willingness it is obvious that we expect nations that have been exposed to either WBNs or WAPs will have a higher propensity to join a war than those that have not. We begin our analysis by

108

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

Table 4.2 No treatment, treatments, and war involvement, 1816–1965 War involvement

No Yes Source The authors

Border or alliance (no treatment) No Yes Total

Warring border or warring alliance partner (treatment) No Yes Total

14 1

2320 8

3641 93

3655 94

1335 86

3655 94

examining the extent to which this expectation is borne out. Table 4.2 reports two initial parts of this analysis. The left half presents the cross tabulation of a nation’s having any border or any alliance and being a war joiner. Except for the 15 cases in the first column, this baseline is the same as the overall baseline. It is readily apparent that the table contains no relationship.14 Moving to the right half, it may be seen that when a nation is exposed to any of the treatments, the propensity to join a war increases substantially. Put simply, in the baseline case 2.4% of the nations participate, but under the treatments 6.1% join. While the left half of the table demonstrates that exposure to any treatment increases the likelihood of joining a war, we need to examine the effect of increasing numbers of treatments. Nations may at the same time have various types of warring borders, may be the members of several alliances of different types having members in the war or, more likely, have some combination of warring borders and alliances. Table 4.3 displays the cross tabulation of war involvement against the number of treatments to which a nation was exposed. It is clear from the data that as the amount of exposure increases, the rate of participation increases as well. The overall relationship (see Table 4.4) as measured by the correlation ratio (eta), derived from a one-way analysis of variance, is 0.452. The analysis has not thus far clarified which of the variables makes the greatest difference. There are, unfortunately, some difficulties in making such an assessment. Our dependent variable, war involvement, is dichotomous, which means that the usual techniques of regression and correlation are not appropriate. Normally, one could approach this type of a problem with log-linear analysis, but an analysis based upon dichotomies of the six possible independent variables and war involvement would produce a table of 128 cells. Given the large total number of observations we have, this would not ordinarily be a problem; but in this instance only 94 of the cases indicate war diffusion, and allocating these in a very large table would produce such a large number of empty cells that analysis would be questionable.

14

The weak results from A2 may be partly due to the fact that there were far fewer of this type of alliance in the data set than either of the other two types. When they are analyzed together, the more numerous alliances may overwhelm A2. In the analysis that follows it may be seen that A2 does not apply to a large number of cases but that when it does, a clear effect is evident.

4.4 Do Opportunity and Willingness Influence War Diffusion?

109

Table 4.3 War diffusion and number of treatments to which a nation was exposed War involvement No Yes Percentage yes Source The authors

Number of treatments 0 1 2

3

4

5

6

>6

2320 8 0.3

142 12 7.8

57 10 14.9

34 16 32.0

12 7 36.8

15 15 50.0

Table 4.4 Analysis of variance for treatments in Table 4.3

703 9 1.2

372 17 4.3

Source of variation

Sum of squares

df

Mean squares

1.441 Treatments 18.737 13a Error 72.906 3735 0.20 Note F = 73.839, p < 0.001, eta = 0.452 a The degrees of freedom are from the uncollapsed table Source The authors

Fortunately, it is possible to gain estimates of the effects of the various independent variables on war diffusion through probit, which, although sensitive to the highly skewed distribution of the data will allow us to make a preliminary comparison of the relative effect of the variables on war diffusion. Table 4.5 displays the results of the probit analysis when a nation’s war involvement is analyzed in relation to its number of (1) contiguous WBNs (Bl); (2) cross-water WBNs (B2); (3) colonial WBNs (B3); (4) defense WAPs (Al); (5) neutrality WAPs (A2); and (6) entente WAPs (A3). The estimates it gives for the independent variables are generally equivalent to the estimates of beta reported in regressions. The results

Table 4.5 Probit estimates of treatment variables on war involvement, 1816–1965

Parameter

Estimate

t-statistic

Constant −2.62 −31.80 B1 0.54* 9.97* B2 0.36** 3.47** B3 0.21* 3.77* A1 0.51* 5.68* A2 0.18**** 1.36**** A3 0.31*** 3.14*** Note Bl = contiguous borders, B2 = cross-water borders, B3 = colonial borders, A1 = defense alliances, A2 = neutrality agreements, and A3 = ententes *p < 0.01 **p < 0.02 ***p > 0.05 ****p > 0.10 Source The authors

110

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

given in Table 4.5 are generally consistent with the ideas of opportunity and willingness; that is, for the border variables BI has greater weight in influencing war joining than B2, and B2 has more than B3. The alliance variables are slightly different. While Al counts the most among these variables, A3 influences war joining much more than A2; in fact, from this analysis A2 seems to have little or no effect on war joining.15

4.5

How Much Do the Variables Contribute to War Diffusion?

Thus far, our analysis has indicated that any exposure to the treatment variables substantially increases the chances of a nation’s joining a war (Table 4.2), that more exposure increases such chances (Table 4.3), and that the strength of the impact of the treatment variables is ordered approximately as our rationale for opportunity and willingness suggested (Table 4.5). What the analysis has not revealed is the magnitude of the impact of the individual and combined indicators of opportunity and willingness. Let us begin with a brief, but necessary, description of the relatively straightforward analysis that follows. Recall from Table 4.2 that we compared the rate of war joining among all nations to only those nations that had been exposed to at least one treatment variable. We may now, in effect, take Table 4.1 apart, examine the impact of each of the variables individually on war diffusion, and then combine each of the border variables with each of the alliance variables to observe their combined impacts on war diffusion. The first 3 rows of Table 4.6 report the results, for all nations of the baseline, of being exposed to one of the three types of WBNs or WAPs and the percentage

15

Recall the treatment matrix from Table 4.1. In the single treatment tables that follow, the numbers being reported are those from cell d of that figure. Reporting all the tables would be cumbersome, so only the effects of the treatments are given. As an example of what a table looks like, we offer a full table for the treatment involving B1. Note that the numbers of cases reported in the tables refer to the total of the b-d column and that the percentages reported are for d of that total. War participation

Contiguous Warring bordering nation Absent Present

Absent Present Total

2950 27 2977

705 67 772

All nations Cell d percentage No Treatment treatment

Percentage change

Percentage change

17.5 143* (103) 24.4 216* (78) 17.9 90* (123) 40.0 304* (40) 27.3 362†† (11) 26.5 159* (49) B2 = cross-water borders, B3

Major powers Cell d percentage No Treatment treatment

2.4 8.7 262* 7.2 (3300) (772) (304) B2 3.1 9.2 196* 7.7 (1696) (295) (353) B3 2.6 5.6 115* 9.4 (1992) (659) (256) Al 5.3 28.6 439* 9.9 (808) (105) (191) A2 4.4 17.5 297* 5.9 (204) (40) (67) A3 4.2 11.2 166* 10.2 (453) (160) (137) Note Numbers of cases are in parentheses, B1 = contiguous borders, A2 = neutrality agreements, and A3 = ententes *p < 0.01 (difference of proportions) **p < 0.05 (difference of proportions) †p < 0.01 (Fisher’s exact test) ††p < 0.05 (Fisher’s exact test)

B1

Condition

Table 4.6 Effects of individual border and alliance treatments, 1816–1965 Percentage change

1.9 7.3 284* (2996) (669) 1.9 3.7 94 (1343) (217) 1.6 2.8 75** (1736) (536) 3.9 21.5 451* (617) (65) 3.6 13.8 283** (138) (29) 1.6 4.5 181* (316) (111) = colonial borders, A1 = defense alliances,

Minor powers Cell d percentage No Treatment treatment

4.5 How Much Do the Variables Contribute to War Diffusion? 111

112

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

change associated with the treatment.16 Beginning with borders, two things are apparent. First, the presence of a WBN has a significant impact on war diffusion. In each case the presence of the treatment produces a marked increase in the propensity for diffusion, ranging from a 262% increase for the contiguous borders to slightly over 100% for the colonial borders.17 Second, the strength of the results is ordered as the opportunity hypothesis predicts, with the closest borders producing the strongest result and the most distant borders producing the weakest result.18 Turning to the alliance variables shown in Table 4.6, a similar pattern may be seen, but with considerably stronger results. Defense alliances produce the substantial increase of 439% over the baseline. While the results for neutrality agreements and ententes are weaker, 297% and 166%, respectively, they are still strong. Again, the strength of the relationships is ordered from the strongest to the weakest commitments. As agents of the opportunity for interaction, both borders and alliances meet the expectation of increasing the probability of states’ joining ongoing wars. In addition to the increased ease of interaction that borders provide (opportunity), borders and alliances both individually increase the salience of the WAPs and WBNs and thus also the willingness to become involved in their conflicts.

16

What we mean by percentage increase is the magnitude by which involvement in a treatment situation is larger than the base. Thus, the first three columns of Table 4.6, while the B1 treatment percentage of 8.7 is more than three times the size of the baseline percentage of 2.4, the relevant measurement is the amount of increase from 2.4 to 8.7. Thus, the appropriate calculation is the treatment percentage minus the base percentage with the remainder divided by the base percentage. 17 The tests of significance used in Tables 4.6 and 4.7 require a brief comment. Because the percentage changes between the no-treatment and treatment columns are often based on relatively small numbers of cases, the skeptical reader may be curious as to (1) whether the observed changes are statistically significant and (2) whether there is any way that the method could fail to produce a change, i.e., whether the results are predetermined. With respect to the first question, we have tested the differences for statistical significance in two ways. First, when the number of cases in the treatment column is greater than 20, the p-value is derived from the Z-score in a test of a difference of proportions. When the number of cases is less than 20, we have determined the p-value from Fisher’s exact test. (Recall that the cases in the treatment column are also present in the baseline no-treatment column). Because the two results are therefore not independent, before we could test for statistical significance between the no-treatment baseline and the treatment, it was necessary to remove the treatment cases from their respective baseline. A full set of the resulting tables is available from Siverson. More important than each test of significance is the correspondence between our theoretical structure and the overall pattern of the data. With respect to the second question (whether the results are predetermined), if one suspects that the method will always produce an increase in the observed cases of war diffusion under the treatment condition, see the last three columns of Table 4.7, where failure does occur. 18 Altfeld/Bueno de Mesquita (1979) are interested only in those cases of war expansion that took place within the first two months of a war. Hence, their main data set contains only 40 cases of war joining. In addition, they exclude from their data set the participation of the four Commonwealth nations (i.e., Canada, Australia, New Zealand and South Africa) that joined World War II in 1939, on the grounds of missing data (that is, they had no alliance memberships). Those cases are included here. We also include the cases from the Korean War, which were totally excluded by Altfeld/Bueno de Mesquita (p. 94, n. 7).

All nations Cell h percentage No Treatment treatment

Percentage change

Percentage change

37.0 302* (27) 40.0 577†† (5) 30.8 224 (26) 61.1 476† (18) 75.0 1019† (4) 489† 64.3 (14) 70.0 363† (20) 66.7 813†† (3) 40.0 238* (25) B2 = cross-water borders, B3

Major powers Cell h percentage No Treatment treatment

5.1 41.8 719* 9.2 (755) (55) (152) B1A2 4.4 22.2 404* 5.9 (204) (27) (67) B1A3 3.6 18.5 413* 9.5 (211) (65) (105) B2A1 6.8 47.8 602* 10.6 (355) (23) (161) B2A2 4.5 30.0 566†† 6.7 (112) (10) (60) B2A3 6.4 33.3 420* 10.9 (250) (27) (128) B3A1 6.5 34.1 424* 15.1 (352) (44) (106) B3A2 4.8 14.2 195 7.3 (83) (14) (41) B3A3 5.5 15.3 178* 11.8 (253) (72) (102) Note Numbers of cases are in parentheses, B1 = contiguous borders, A2 = neutrality agreements, and A3 = ententes *p < 0.01 (difference of proportions) **p < 0.05 (difference of proportions) †p < 0.01 (Fisher’s exact test) ††p < 0.05 (Fisher’s exact test)

B1A1

Condition

Table 4.7 Effects of combined border and alliance treatments, 1816–1965 Percentage change

3.9 46.4 1089* (603) (28) 3.6 18.2 405* (137) (22) 1.6 10.2 537* (306) (39) 3.6 0 −100 (194) (5) 1.9 0 −100 (52) (6) 1.6 0 −100 (122) (13) 2.8 4.2 50 (246) (24) 2.4 0 −100 (42) (11) 1.3 2.1 61 (151) (47) = colonial borders, A1 = defense alliances,

Minor powers Cell h percentage No Treatment treatment

4.5 How Much Do the Variables Contribute to War Diffusion? 113

114

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

How do the border and alliance variables interact with each other? The three border and three alliance variables combine, of course, to produce nine possibilities. The first three columns of Table 4.7, report the results of these combinations for all nations. (Because the tables reporting the combination of the variables contain eight cells, the cell corresponding to d in Table 4.1 is h.) It may readily be seen that there is a considerable amount of interaction between the border and alliance variables in influencing the diffusion of war. The most potent effect is with the combination of the contiguous WBNs and defensive WAPs. This should not be surprising, given our previous results and our hierarchy of importance within opportunity and willingness. The latter derives clearly from the international relations literature. Most geopolitical studies focus on direct contiguous borders because of their immediate impact and because they provide the most important opportunity for interaction. In turn, the form of alliance that, scholars theorize, creates the strongest bond between states is the defensive pact. Defense pacts provide the most important indicator of salience, commitment, and shared policy preference. The magnitude of the joint effect is, to say the least, considerable, with the combination of these two variables producing an increase of 719% over the baseline cases in producing war diffusion. While that result is the strongest, the other results on the several combined variables also generally reveal a considerable amount of interaction. Indeed, all contiguous border (B1) and all noncolony cross-water (B2) effects are improved when combined with alliance variables; and all defense and neutrality pact (Al and A2) effects are improved when combined with Bl and B2. Cross-water WBNs and defense WAPs, for example, combine to produce an increase of 602% over the baseline cases, while cross-water WBNs and neutrality WAPs combine to increase war diffusion 566% over the baseline cases. Even colonial borders (B3) and ententes (A3), the weakest of the individual effects, produce significant increases on war joining when combined with the other variables—indeed, even when combined with each other (178%). Given the various combinations of our variables it is, of course, difficult to produce an ordered prediction of their strength, as we did with the single variables. However, it is worth noting that when the class of the WBN is held constant and the WAP class varies, with one exception the magnitudes of the impact are ordered within the groups. The sole exception is that the combination of contiguous WBNs is stronger with entente WAPs (413%) than it is with neutrality WAPs (404%). Thus far, our consideration of the problem of war diffusion has treated all nations as if they were the same except for their exposure to the various treatments. Much of international relations theory, however, relies on the knowledge that all nations are not the same. A distinction among nations that traditionally has been central to an overwhelming part of the field of international relations is that between major and minor powers. For our purposes, too, this is an important distinction, since there is ample reason to believe that major and minor powers will differ in their behavior with respect to opportunity and willingness in joining wars. Specifically, we should expect the major powers to respond to opportunities that are cross-water or colonial because, in the first instance, they have the greater capabilities necessary to become involved and, in the second instance, they not only have the capabilities

4.5 How Much Do the Variables Contribute to War Diffusion?

115

but are much more likely to have colonial borders. Moreover, while minor powers certainly had many alliances, major powers are overrepresented among the nations with alliances, particularly defense agreements. The neatness of the analysis begins to break down when we analyze the data separately for the major and minor powers in the international system. We begin with the major powers. The effects of the individual border and alliance conditions for the major powers are given in the middle three columns of Table 4.6. Here it is evident that while each of the variables had a discernable effect on war diffusion, the results depart from the magnitudes and orderings given in the first three columns of the same table. Cross-water WBNs are the strongest border effect for the major powers as are neutrality WAPs. As we expect from our ideas about the strength of various forms of opportunity, colonial WBNs have only a weak effect. Note, however, that alliances are generally of more importance for major powers than are borders and that this relationship is stronger for major powers than for all states combined. By definition, major powers are the actors with global (or at least multiregional) interests and capabilities. They would be more likely to be able to dominate smaller neighbors. As such, the greater impact of alliances is not surprising. The middle three columns of Table 4.7 present the combinations of variables for the major powers. Again, there is clear interaction involved in these combinations, some of them astonishingly large. Indeed, every individual condition improves with the combined effects of borders and alliances. The combination of colonial WBNs and neutrality WAPs, for example, produces an increase over the baseline of 813%. As might be suspected in the case of such a sharp increase, the number of cases is small (N = 3). Similarly, the cross-water WBNs and neutrality WAPs produce a gain of 1019% for four cases. More surprising, perhaps, than the size of some of the gains observed is the fact that the combination that previously demonstrated the largest gain, the combination of contiguous WBN and defense WAP, for the major powers, shows only a modest increase. Here the gain is only 302%; and while such an increase is not to be ignored, its modest size relative to some of the other combinations of variables is not in accord with what we would expect from our interpretation of opportunity and willingness or the findings presented above. The significant increase found in the effects of B3 (colonial borders) in combination with all alliance types may be related to Most/Starr’s (1980) findings on colonial borders above. While the numbers of cases are rather small, it is clear that the combined effects of colonial borders and alliances—(far-flung) opportunity and willingness—now match the earlier findings and arguments about the interests and capabilities of major powers. These findings give us some anticipation of what will be found with the minor powers, to which we now turn. The last three columns of Table 4.6 report the individual effects of WBNs and WAPs for the minor powers. First, it may be noted that with two glaring exceptions (B2 and B3, where the increases are very low) the treatment variables had clear effects. The strongest of these was the presence of defense WAPs, where the increase was 451%. In addition, the results are ordered as our discussion of the

116

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

strength of the various indicators suggests. Comparing the results for major powers (the middle three columns) and minor powers, the latter are more strongly affected both by contiguous borders (Bl) and defense pacts (Al). The first finding is expected. Minor powers, virtually by definition, are those with local or regional concerns. Given the limited ability of minor powers to project power and thus their greater concern with viability in terms of immediate neighbors, direct borders should be of greatest importance. The Al results, however, are more intriguing. Previous work by Siverson/King (1979) has demonstrated that alliances are likely ‘agents’ of diffusion. Such research results directly address the war-alliance relationship that is of concern to many scholars and the related assertion that alliances act as conduits for the spread of international conflict (e.g., see Starr 1978b). Alliances might bring great power conflict into the regional subsystems of small allies. Conversely, alliances might drag major power allies into the local conflicts of their smaller alliance partners. Since the bulk of minor power defensive alliances are with major powers, the Al results appear to indicate in which direction the alliance conflict conduit has tended to work—minor powers being pulled into the ongoing wars of their major power allies. The last three columns of Table 4.7 report the results of the nine combinations of variables for the minor powers. Again, some of the increases are, to say the least, sharp. The situation where we expect the largest increase, contiguous WBNs and defense WAPs, does indeed show the strongest increase, with a magnitude of over 1000%. Moreover, this time, the number of cases is not small. The first four variable combinations all show marked increases, then something interesting takes place. Once the more distant forms of opportunity and willingness are encountered, the results drop substantially. In fact, for the last six combinations in the table the results are either meager or nonexistent. For all of these cases, the combined treatment results are less than the effects of the individual treatments; that is, the combined conditions make war diffusion less likely. For the first time the results show negative interaction. One may initially suspect that this result is nothing more than the artifact of variable combinations producing numbers of cases so small that they disappear. But, as shown in the last three columns of Table 4.7, this is not the case. While the number of cases does go down because of the combination, they are certainly too big simply to attribute the absence of an effect to that artifact. The more likely explanation is to be found in the fact that these combinations do indeed represent the more distant forms of opportunity and lesser types of willingness and that minor powers, because of their lesser capabilities, are either reluctant or unable to enter conflict. Wars, after all, do impose costs on participants; and minor powers may not occupy positions where the costs are tolerable.

4.6 A Problem?

4.6

117

A Problem?

Before drawing the results together there is a problem, or puzzle, that must be addressed. Throughout the data analysis it may be seen that there is a consistent pattern wherein the number of nations exposed to the various treatments is very large relative to the number of nations who actually join the war. Consider, for example, the data in the right half of Table 4.2, which offer the clearest display of this pattern. The data in that table show 1421 cases of nations exposed to any of the treatments; but only 86 instances of the exposure result in the nations joining a war. To be sure, the proportion of those joining under the treatment is considerably larger than under the baseline, but it is nonetheless perplexing that only 6.44% of the cases exposed to any form of the treatment joined a war. There are a number of possible approaches to evaluating the impact of these ‘error’ cases to the validity of our findings. First, recall that we explicitly stated that we are working within a framework of “loose necessity.” Under this circumstance we expect that there will be many ‘error’ cases but that such error is more apparent than real. The requirements of opportunity and willingness do not predict that a nation exposed to these conditions will join a war but only that its probability of joining will be substantially higher than where those conditions are not present. This situation is quite similar to that encountered by Bueno de Mesquita (1981) in his consideration of the initiation of war. In his analysis it may be seen that nations have positive expected utility for war far more than they actually go to war, while in this research there are far more WBN-WAP interaction opportunities than there are instances of diffusion. In this research the importance of the idea of necessity may be seen in Table 4.1. The concern with overprediction is reflected in the number of cases in cell b. However, for necessary relationships the number of cases in b is irrelevant. The relevant cell to compare to cell d is c, which should be empty. Given our “loose necessity” formulation, we should expect only a few cases in c, especially as a percentage of the total number of cases in the no-treatment column (i.e., the a + c column). Returning to the right half of Table 4.2, it is dear that although there are a very large number of cases in the no-treatment column (2320), there are only eight in cell c. Similar relationships occur in the matrices that provide most of our results. (In the example given in n. 13 reporting the results of contiguous WBNs there are only 27 cases in c out of a possible 2977 cases.) There is a second way to investigate the ‘error’ cases. It is instructive to examine once again the data in the right half of Table 4.2. There it may be seen that most of the ‘error’ is to be found in the cases in which a nation was exposed to only one of the treatments. The 703 ‘error’ cases under that condition represent 52.7% of the total ‘error.’ As exposure goes up, the rate of war joining goes up quite sharply. The question now is whether or not the 712 cases (i.e., the 703 cases of ‘error’ and the 9 cases of war joining) that received only single treatments are distributed evenly across the opportunity and willingness variables. Table 4.8 summarizes this

118

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

Table 4.8 Occurrence of single border and alliance treatments, all nations, 1816– 1965

Variable

Number of single treatments

Percentage of total

B1 328 46 B2 75 10.5 B3 259 36.4 A1 10 1.4 A2 6 0.8 A3 34 4.8 Note B1 = Contiguous borders, B2 = cross-water borders, B3 = colonial borders, A1 = defense alliances, A2 = neutrality agreements, and A3 = ententes Source The authors

distribution. Clearly, exposure to a single Bl or B3 happened fairly frequently but did not have much of an effect. There is at least one interesting implication that may be drawn from the above observations: decisions to join an ongoing war face considerable ‘friction.’ A single opportunity is rarely associated with war joining. It is only when the opportunities begin to accumulate, or, more importantly, are attached to the political affinity indicated by an alliance that the chances of joining an ongoing war begin to build significantly. The fact that wars are, in a sense, undersubscribed is not surprising. Theories of collective goods tell us that it is not rational for a nation to undertake a costly course of action if someone else will do it for them (Olson/Zeckhauser 1966) and wars can be among the most costly of all the courses of action a nation may choose.

4.7

Conclusion

This study began with the aim of investigating the diffusion of war—in terms of the expansion of ongoing conflict—between 1816 and 1965. This was done in the context of opportunity and willingness, as indicated by WBNs and WAPs. We have proposed a relatively simple expectation derived from the opportunity, willingness, and interaction opportunity concepts and investigated it through a set of comparisons between treatment and no-treatment groups of data. The data presented in the tables support the contention that war does expand among nations along the indicators that operationalize opportunity and willingness and that the strength of the relationship is ordered by the type of border or alliance, as hypothesized. We have added to the understanding of war diffusion in several ways. While one body of previous research has indicated an important geographic basis for diffusion and other research has indicated that alliances are important in spreading conflict, the present work has the virtue of combining geographic and political variables so that the strong interaction of these two elements may be observed.

4.7 Conclusion

119

The scope of the present analysis is also more extensive than previous studies of war diffusion. The data used are global (not limited to any single geographic region) and cover all international wars from 1816 to 1965 (see Footnote 16). We provide separate analyses of major and minor powers, revealing different patterns of diffusion and indicating how major and minor powers differentially relate to the elements of opportunity and willingness. Perhaps most useful to the theoretical development of the study of war diffusion, we have presented more evidence as to the applicability of research based on opportunity, willingness, and their interaction. We have shown here that this theoretical approach is applicable to large-scale war as delineated by the Correlates of War Project’s data set. The previous applications of this approach by Most and Starr focused violent conflict on a smaller scale, whether internal or external. We also have more confidence in the theoretical scope of this approach, since we now find that it can help explain two different conceptualizations of diffusion: (1) the linkage-penetration view taken in earlier research by Most and Starr (among others), where events elsewhere change a state’s disposition to behave similarly; and (2) the infection-contagion approach taken here, where diffusion is conceived as the growth of ongoing wars. The refinement of such analytic tools is vital in the design of future research. With multiple sources of evidence indicating that diffusion processes of various kinds do exist in international relations, we now require further work in the specification of the processes at work in various kinds of diffusion, including the conditions under which the diffusion of violent conflict takes place. The further specification of these processes might also help us understand the interaction between opportunity and willingness and various theories as to why wars are likely to be larger at particular times than at others.

References Alcock, Norman Z., 1972: The War Disease (Oakville, Ontario: Canadian Peace Research Institute Press). Altfeld, Michael; Bueno de Mesquita, Bruce, 1979: “Choosing Sides in Wars”, in: International Studies Quarterly, 23: 87–112. Boulding, Kenneth, 1962: Conflict and Defense (New York: Harper & Row). Bueno de Mesquita, Bruce, 1981: The War Trap (New Haven: Yale University Press). Bueno de Mesquita, Bruce; Lalman, David, 1986: “Reason and War”, in: American Political Science Review, 80: 1113–1129. Davis, William; Duncan, George; Siverson, Randolph, 1978: “The Dynamics of Warfare, 1816– 1965”, in: American Journal of Political Science, 22: 772–792. Diehl, Paul, 1988: “Geography and War: A Review and Assessment of the Empirical Literature”, Presented at the Annual Meeting of the International Studies Association, St. Louis. Diehl, Paul F.; Goertz, Gary, 1988: “Territorial Changes and Militarized Conflict”, in: Journal of Conflict Resolution, 32: 103–122. Faber, Jan; Houweling, Henk; Siccama, Jan, 1984: “Diffusion of War: Some Theoretical Considerations and Empirical Evidence”, in: Journal of Peace Research, 21: 277–288.

120

4 Opportunity, Willingness and the Diffusion of War, 1816–1965

Garnham, David, 1988: “Balancing or Band-Waggoning: Empirical Evidence from Nineteenthand Twentieth-Century Europe”, Presented at the International Studies Association, St. Louis. Gleditsch, Nils P., 1969: “The International Airline Network: A Test of the Zipf and Stouffer Hypotheses”, in: Peace Research Society Papers, 11: 123–153. Gochman, Charles S.; Sabrosky, Alan Ned (Eds.), 1990: Prisoners of War? Nation-States in the Modem Era (Lexington, MA: D.C. Heath). Henige, David P., 1970: Colonial Governors from the Fifteenth Century to the Present (Madison: University of Wisconsin Press). Hill, Stuart; Rothchild, Donald, 1986: “The Contagion of Political Conflict in Africa and the World”, in: Journal of Conflict Resolution, 30: 716–735. Holsti, Ole R.; North, Robert C., 1965: “The History of Human Conflict”, in: McNeil, Elton (Ed.): The Nature of Human Conflict (Englewood Cliffs, NJ: Prentice Hall). Kende, Istvan, 1971: “Twenty-Five Years of Local Wars”, in: Journal of Peace Research, 8: 5–27. Kende, Istvan, 1978: “Wars of Ten Years (1967–1976)”, in: Journal of Peace Research, 15: 227– 241. Leng, Russell J.; Gochman, Charles S., 1982: “Dangerous Disputes: A Study of Conflict Behavior and War”, in: American Journal of Political Science, 26: 664–687. Maoz, Zeev, 1982: Paths to Conflict (Boulder: Westview). Midlarsky, Manus, 1975: On War (New York: Free Press). Most, Benjamin A.; Starr, Harvey, 1980: “Diffusion, Reinforcement, Geo-politics and the Spread of War”, in: American Political Science Review, 74: 932–946. Most, Benjamin A.; Starr, Harvey, 1981: “Theoretical and Methodological Issues in the Study of Diffusion and Contagion: Examples from the Research on War”, Presented at the Annual Meeting of the International Studies Association, Philadelphia. Most, Benjamin A.; Starr, Harvey, 1989: Inquiry, Logic, and International Politics (Columbia: University of South Carolina Press). Most, Benjamin A.; Schrodt, P.; Siverson, Randolph M.; Starr, Harvey, 1987: “Border and Alliance Effects in the Diffusion of Major Power Conflict, 1815–1965”, Presented at the Annual Meeting of the International Studies Association, Washington. Most, Benjamin A.; Starr, Harvey; Siverson, Randolph, 1989: “The Logic and Study of the Diffusion of International Conflict”, in: Midlarsky, Manus (Ed.): Handbook of War Studies (Boston: Allen & Unwin). O’Loughlin, John, 1984: “Geographic Models of International Conflicts”, in: Taylor, Peter J.; House, John (Eds.): Political Geography: Recent Advances and Future Directions (London: Croom Helm). Olson, Mancur; Zeckhauser, Richard, 1966: “An Economic Theory of Alliances”, in: Review of Economics and Statistics, 46: 266–279. Richardson, Lewis F., 1960: The Statistics of Deadly Quarrels (Chicago: Quadrangle). Ross, Marc; Homer, Elizabeth, 1976: “Galton’s Problem in Cross-National Research”, in: World Politics, 29: 1–28. Sabrosky, Alan N., 1975: “The Utility of Interstate Alliances”, Presented at the Annual Meeting of the International Studies Association, Washington. Sabrosky, Alan N., 1980: “Interstate Alliances: Their Reliability and the Expansion of War”, in: Singer, J. David (Ed.): The Correlates of War, Vol. 2 (New York: Free Press). Schattschneider, Elmer E., 1960: The Semi-Sovereign People: A Realist’s View of Democracy in America (New York: Holt, Rinehart, and Winston). Shepherd, William R, 1932: Atlas of Modern History (New York: Henry Holt). Singer, J. David, 1974: “The Historical Experiment as a Research Strategy in the Study of World Politics”, in: Political Inquiry, 2: 23–42. Singer, J. David; Bremer, Stuart A.; Stuckey, John, 1972: “Capability Distribution, Uncertainty, and Major Power War, 1820–1965”, in: Russett, Bruce M. (Ed.): Peace, War, and Numbers (Beverly Hills: Sage). Singer, J. David; Small, Melvin, 1968: “Alliance Aggregation and the Onset of War”, in: Singer, J. David (Ed.): Quantitative International Politics (New York: Free Press).

References

121

Singer, J. David; Small, Melvin, 1972: The Wages of War, 1816–1965 (New York: Wiley). Singer, J. David; Small, Melvin, 1974: “Foreign Policy Indicators: Predictors of War in History and in the State of the World Message”, in: Policy Sciences, 5: 271–296. Siverson, Randolph M.; King, Joel, 1979: “Alliances and the Expansion of War”, in: Singer, J. David; Wallace, M. D. (Eds.): To Augur Well: Early Warning Indicators in World Politics (Beverly Hills: Sage). Siverson, Randolph M.; King, Joel, 1980: “Attributes of National Alliance Membership and War Participation, 1815–1965”, in: American Journal of Political Science, 24: 1–15. Siverson, Randloph; Starr, Harvey, 1988: “Alliance and Border Effects on the War Behavior of States: Refining the Interaction Opportunity Model of Diffusion”, Presented to the Third World Congress of the International Peace Science Society, College Park, MD. Small, Melvin; Singer, J. David, 1982: Resort to Arms: International and Civil Wars, 1816–1980 (Beverly Hills: Sage). Sprout, Harold; Sprout, Margaret, 1969: “Environmental Factors in the Study of International Politics”, in: Rosenau, James N. (Ed.): International Politics and Foreign Policy (New York: Free Press). Starr, Harvey, 1978a: “‘Opportunity’ and ‘Willingness’ as Ordering Concepts in the Study of War”, in: International Interactions, 4: 363–387. Starr, Harvey, 1978b: “Alliances: Tradition and Change in American Views on Foreign Military Entanglements”, in: Booth, Ken; Wright, Moorhead (Eds.): American Thinking About Peace and War (New York: Barnes & Noble). Starr, Harvey; Most, Benjamin A., 1976: “The Substance and Study of Borders in International Relations Research”, in: International Studies Quarterly, 20: 581–620. Starr, Harvey; Most, Benjamin A., 1978: “A Return Journey: Richardson, ‘Frontiers,’ and Wars in the 1946–1965 Era”, in: Journal of Conflict Resolution, 22: 441–467. Starr, Harvey; Most, Benjamin A., 1983: “Contagion and Border Effects on Contemporary African Conflict”, in: Comparative Political Studies, 16: 92–117. Starr, Harvey; Most, Benjamin A., 1985: “The Forms and Processes of War Diffusion: Research Update on Contagion in African Conflict”, in: Comparative Political Studies, 18: 206. Sullivan, John D., 1974: “International Alliances”, in: Haas, Michael (Ed.): International Systems (New York: Chandler). Ward, Michael; Kirby, Andrew, 1987: “Reexamining Spatial Models of International Conflicts”, in: Annals of the Association of American Geographers, 77: 279–283. Welsh, William A., 1984: “Inter-nation Interaction and Political Diffusion: Notes Toward a Conceptual Framework”, Presented at the Annual Meeting of the International Studies Association, Atlanta. Wright, Quincy, 1965: A Study of War, Rev. ed. (Chicago: University of Chicago Press). Zipf, George K., 1949: Human Behavior and the Principle of Least Effort (Cambridge, MA: Addison-Wesley).

Chapter 5

Democratic Dominoes: Diffusion Approaches to the Spread of Democracy in the International System

This article is an attempt to indicate how diffusion approaches, based on the concepts of linkage and interdependence, can be of help in our thinking about the spread of democracy.1 The analyses address the existence or absence of diffusion effects in regard to changes in the degree of freedom in the world’s governments, and whether or not there has been a more specific global movement towards democracy.2 The dependent variable is the set of “governmental transitions,” based on yearly Freedom House data. Diffusion analyses are at the global, regional, and neighbor-state levels (1974–1987). Thus analysis is limited to cues or prototypes from the external environment of states. Although neighbor effects are less than those found with the diffusion of war, all three levels support the proposition that there has been a diffusion of governmental transitions, including a movement towards democracy that provided a context for the dramatic events of 1988 and 1989.

5.1

Democracy, Dominoes, and Diffusion

For years analysts of various stripes have been concerned with the toppling of ‘dominoes’ somewhere in the world. The usual expectation, when this metaphor was employed, was that once a communist government had been imposed/installed

This text by the author was first published as: “Democratic Dominoes: Diffusion Approaches to the Spread of Democracy in the International System”, in: The Journal of Conflict Resolution, Vol. 35, No. 2, Democracy and Foreign Policy: Community and Constraint (Jun., 1991), pp. 356–381. The permission to republish it here was granted by Sage. 2 Author’s Note: I would like to thank Randy Siverson, William Jacoby, and Mac Rood for their comments and assistance on an earlier draft. Thanks also to those participating in the “Democracy and Foreign Policy” conference, especially Stuart Hill, Sung-Chull Lee, and Bruce Russett, for their reactions and comments. All errors, of course, are those of the author. 1

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7_5

123

124

5 Democratic Dominoes: Diffusion Approaches …

in a state, its neighboring democracies would, one by one, fall to communism in a manner resembling the chain reaction of a line of falling dominoes. This metaphor dominated American rhetoric during the coldest phases of the Cold War, usually accompanied by maps on which a red stain would be spreading inexorably from some locus of communist power. This view of the threat facing the United States, its Western allies, and neutrals, was based on an implicit model of positive spatial diffusion, whereby the communist ‘disease’ would be spread through contagious contact. After almost four decades of waiting for a domino effect regarding the successive changing of the form of government in a number of states to take place, it finally has. However, the striking set of governmental changes occurring in 1989 and 1990 has been in the successive ‘fall’ of communist and other authoritarian governments to some form of democratic, pluralistic, or polyarchic government. Clearly the dramatic changes in Eastern Europe include events that are dependent on one another. They are events where forms of emulation, demonstration, or modeling effects appear to have taken place. They also dramatically highlight the need to address a set of questions regarding the possibility of diffusion effects in the transition of governmental structures and processes, the existence of global or regional trends toward more democratic forms of government, and the thorny, but continuing, question regarding the linkages between internal and external phenomena. Thus this article will have two primary foci: (1) to present preliminary analyses that address the existence or absence of diffusion effects in regard to changes in the degree of freedom in the world’s governments, and (2) to investigate one specific substantive diffusion effect—whether or not there has been a global movement towards democracy. As will be discussed below, the time frame for my analysis excludes the significant governmental changes of 1989–1990 in Eastern Europe. This article will, however, investigate the broader context of the diffusion of governmental openness, placing the dramatic changes of 1989 to 1990 within an investigation of the prior fifteen-year period. The two research aims of the article must also be seen within a broad linkage politics framework. It is clear that when governments alter the level of political and civil rights allowed in society (through changes in structure or process), that domestic or societal factors are of major importance. However, states exist in an interdependent world, where the “hard shell” of the state, to use John Herz’s phrase, is permeable to many forces. These forces include the spread of norms throughout international society. The adoption of more democratic processes or forms as a result of global or regional demonstration effects is clearly outside of a realist framework. Instead, it conforms to analyses which have demonstrated the decline and abolition of slavery and dueling (see Ray 1989; Mueller 1989); analyses of the adherence to norms of alliance compliance (see Kegley/Raymond 1990); or even arguments concerning the delegitimization of the use of force (as most strongly argued by Mueller). Thus it is important to understand that, as demonstrated elsewhere (see Most/ Starr 1990), spatial diffusion may logically be seen as one subfield of linkage

5.1 Democracy, Dominoes, and Diffusion

125

politics, where policies or processes within a state are cued by some set of extrasocietal, prototypic behaviors. Indeed, in a sophisticated conceptual exercise directed towards diffusion, Welsh (1984: 1) notes that his paper is within “the tradition of scholarship that seeks to bridge the often-artificial gap between studies of interactions among nations … and comparative studies of the system and within-system attributes of nations.” This article will deal primarily with the effects that external factors (or context) have on the transition of governmental forms and constraints. A full model or theory would necessitate a detailed analysis of domestic factors as well. That task will be left for subsequent research. The present task is to demonstrate the extent of external effects and that the study of diffusion is a useful conceptual framework for linking the interaction of domestic and external factors (Most/Starr 1980, 1990).

5.2

Governmental Transitions

Models of the diffusion process have been used to study purely external/ foreign policy phenomena such as the spread of war (e.g., Siverson/Starr 1990), violent conflict—which could be foreign or domestic (e.g. Most and Starr 1980)—or purely domestic forms of conflict, such as coups d’etat (see, for example, Most 1973; Midlarsky 1970, 1975, Chap. 7; or Li/Thompson 1975). In the lattermost analyses diffusion was used to study internal changes and the processes by which certain types of change (coups) led to similar changes within other states. Here we will be concerned with the diffusion effects of another type of internal change—governmental transitions. The transition referred to here is a change in the status of a government’s degree of freedom as coded yearly by Freedom House (see, for example, Gastil 1989). Starting in 1973 every state in the system (as well as dependent territories) has been judged as ‘free’ (F), ‘partially free’ (PF), or ‘nonfree’ (NF), based on a 7-point scale each for political rights and civil liberties. With this typology of governmental freedom there are six possible transitions that could occur from one year to the next. The first three are called ‘positive’ in that they indicate movements towards democratic government; the latter three are ‘negative’ in that they indicate movements away from freedom or democratic forms and processes.

Looking at all states in the system each year, I analyze the 14 possible transition periods starting with the transition from 1973 to 1974 (which will be labeled 1974), and ending with the 1986 to 1987 transition (labeled 1987). I will investigate

126

5 Democratic Dominoes: Diffusion Approaches …

whether changes (transitions) in one country are independent of, or related to, changes (transitions) in other countries. I will be able to look both for patterns indicating general diffusion effects and for patterns indicating movements toward democracy.3

5.3

Diffusion Approaches to Governmental Changes

The nature, forms, and consequences of the diffusion of policies and behaviors at the international and transnational levels are complex subjects (see, for example, Most/Starr 1990; Welsh 1984). Diffusion may refer to several distinctive processes studied at several different levels of analysis. At its heart, however, we can apply the following definition to all diffusion phenomena: “Diffusion refers to the process by which institutions, practices, behaviors, or norms are transmitted between individuals and/or between social systems” (Welsh 1984: 3). In this broad sense, diffusion can, and in this study will, be employed as a conceptual or theoretical basis for empirical analysis. That is, diffusion, and methods developed to study it, involves a set of assumptions about the nature of systems, how units interact, and expectations of how the environmental context will affect the behavior of environed units. In disaggregating diffusion into external spatial diffusion effects and internal reinforcement effects, Most/Starr (1990), demonstrate that diffusion models may be conceived in a general linkage framework, where there are linkages between some state’s policy and other, previously occurring factors which are external to that state; (situations where the explanans and explanandums are on different sides of a state’s boundaries). The key issue in the study of diffusion is where the stimulus for emulation comes from. Midlarsky (1975: 156) notes that, “Perhaps the most important element of a theory of diffusion is the existence of a ‘model’ or prototype. The prototype exhibits certain behavior patterns and an observer matches these behaviors.” The source of ‘prototypes’ or ‘models’ could be internal or external. Internal cues/prototypes/ models would lead to reinforcement effects (either positive or negative) where the behavior within some polity increases or decreases the probability that such behavior will take place again. The source of cues/prototypes/models may also be external. Different models of diffusion again may be seen as looking at different sources of such stimuli. At a macro level, we can look to the general global environment as a source of information about certain types of activities (behaviors, innovations, etc.) taking place anywhere in the system. I will refer to the results of this source of cues, or level of 3

Note that another striking example of the spread of norms in international politics is the history of human rights in the post-World War II system. This example is particularly relevant here, because many aspects of the international law of human rights are congruent with Gastil’s definition of freedom through political and civil rights. The process of establishing human rights norms as part of the international societal context is integral to the growth of free states in the system.

5.3 Diffusion Approaches to Governmental Changes

127

analysis, as “global demonstration effects.” The effects here are based on high levels of global information flow and the sychological/political interdependence created by increasingly rapid and intrusive levels of communication technology (e.g., see Russett/Starr 1989: 489–91; Rosenau 1984: 292). International actors are inextricably linked through a variety of interdependent networks, but I am most concerned here with the cognitive and psychological interdependence promoted through the phenomenon Rosenau calls the “microelectronic revolution.” The key to “cascading interdependence” Rosenau argues (1988: 359) is “at the micro-level in the ever-expanding analytical skills and cathectic capacities of individuals ... citizens and leaders in all parts of the world are increasingly able to comprehend where they and their collectivities fit—and should fit—in the processes of global politics.” This point is also key to understanding how diffusion of “democratic innovation” could occur through general demonstration effects: constant and increasing information being electronically communicated on the nature of international society which acts to strengthen a conception of ‘society.’ Not only are individuals (and the political actors to which they belong) interdependent, but they are aware of their interdependence. There is an increasing cognition of interdependence. Thus, at the global level, I am concerned with the flow of information about changes in the openness of governments anywhere in the international system unconstrained by geographical considerations, as decision makers and peoples might use that information as cues and sources of emulation. The geographer Peter Gould (1969: 5) has noted that “simple geographic distance is not always the strongest influence in a diffusion process, for some ideas and innovations seem to leap over many intervening people and places.” Remember, however, that ‘negative’ movements away from freedom or democracy can serve as cues as well as ‘positive’ movements toward democracy. Thus, as noted, a first question will be directed toward discovering any diffusion effects changes in one government (anywhere in the system) might have on subsequent changes in government elsewhere in the system. At this level, we will employ ‘macro’ level methods (see Welsh 1984, Table 1) to detect nonrandomness of transitions in the system. Temporal patterns, the curve of innovations, and the Poisson distribution will be employed. Such macro level analyses cannot tell us many things (see Most/Starr 1990), but they can, as Welsh (1984: 15) notes, identify the general shape of diffusion processes; “an understanding of macro spatio-temporal distributions is an important first step toward developing explanatory process models.” The source of cues/prototypes/models might arise from a regional context rather than a global one. Such regional effects have been studied as “spatial heterogeneity” by Anselin/O’Laughlin (1989), in order to look for second order (and higher) effects, as distinguished from the first-order effects of contagion from neighbors. Regional factors have been delineated as operative in the diffusion of war (e.g., Bremer 1982; Houweling/Siccama 1985). Thus both global and regional effects will be conceived of as important parts of the external context. The changing levels of democracy in the system, the region, or in neighbors will be seen as changing the

128

5 Democratic Dominoes: Diffusion Approaches …

‘menu’ of states, that being the overall incentive structures within which decision makers and peoples consider foreign and domestic policy. In this sense, context is seen as providing both constraints and opportunities in a Sproutian conception currently being revived in the study of political geography (see Starr/Siverson 1990; O’Loughlin/van der Wusten 1986).

5.4

Neighbor Effects and the Interaction Opportunity Model

In research on the diffusion of violent conflict and war, I have been involved in the development and application of an approach and set of methods to tap neighbor effects through positive spatial diffusion and infection. Based on the opportunity and willingness framework, an “interaction opportunity” model was developed for studying borders and alliances as ‘agents’ of diffusion (see also Starr forthcoming). In such an approach, diffusion is seen to be facilitated by the salience and ease of interaction provided by proximity. Whereas borders approximated opportunity, alliances were surrogates for willingness, particularly as indicators of shared policy preferences and high mutual salience. Both borders and alliances were conceived of as ‘treatments’ through the identification of “warring border nations” (WBNs) and “warring alliance partners” (WAPs). States not at war under such treatments were investigated to determine if the treatments altered their subsequent probabilities of engaging in violent conflict or war. Here, instead of WAPs or WBNs, we will be concerned with “bordering governmental transitions” or BGTs. In the study of war, we found significant neighbor effects, whereby the activities in bordering states had contagious diffusion effects (both emulative and infectious). Transferring the logic and methods of the study of war to the study of governmental transitions is not exact, however. One dimension —that neighbors are accorded higher salience, importance, and concern, can be carried over to the study of governmental transition. The notion that decision makers are concerned with the uncertainty that changes in neighbors generate, and that such changes may also generate perceptions of threat can also be carried over. Both sorts of environmental conditions provide constraints and opportunities; both change the menu of incentives. The WBN treatment analysis, however, was also based on the ease of movement of military capability, making neighbors engaging in violent conflict particularly worrisome. This ease of interaction/threat relationship (based on ideas such as Boulding’s loss-of-strength gradient) is not central to the study of governmental transitions. In addition, although the decision to use violence or go to war is a policy decision within any single polity, the occurrence of a war is an interdependent outcome requiring the joint decisions of at least two parties. The behavior and changes that bring about governmental transitions take place solely within a single country. They are also not nearly as clear cut or neat as a war or no-war decision.

5.4 Neighbor Effects and the Interaction Opportunity Model

129

Thus we could propose that the positive spatial diffusion effects based on the interaction opportunity model might not be found here or might be much weaker than in the study of war. In that case, the analyses on the global and regional levels would be more important, as the study of governmental transitions bears a closer resemblance to the diffusion of innovations (e.g., Eyestone 1977; Walker 1969; Collier/Messick 1975) than the diffusion of war. It is more likely, however, that all three levels of analysis will be needed. Which states pick up on global or regional effects may depend strongly on which states are ‘ready’ for innovation in terms of their internal setting. That context certainly contains a number of necessary conditions for making a state ready for governmental transition. It is clear that the ultimate form of the study of governmental transition must involve a combination of external (diffusion) and internal (prerequites) factors (see Collier/Messick 1975).4

5.5

Data and Descriptives

The data from which the governmental transitions were developed were provided by the Freedom in the World series of publications (Gastil 1983, 1989). Clearly the concept of democracy is a complex one, subjected to a wide range of commentary by political scientists, sociologists, and philosophers of all stripes. As Gastil notes (1989: 3), “understanding freedom is not a simple task.” The purpose of this article is not to participate in the debate over what exactly constitutes democracy or freedom. For the Freedom House surveys, freedom simply means Western-style “liberal democracy.” Following from this view, freedom is defined along two dimensions, political rights and civil liberties: Political rights are rights to participate meaningfully in the political process. In a democracy this means the right of all adults to vote and compete for public office, and for elected representatives to have a decisive vote on public policies. Civil liberties are rights to free expression, to organize or demonstrate, as well as rights to a degree of autonomy such as is provided by freedom of religion, education, travel, and other personal rights. (Gastil 1989: 7)

Each dimension is rated along a 7-point scale, rating states comparatively rather than against some set of absolute criteria (Gastil 1989: 7). Because final ratings are then collapsed into the three categories—free, partially free, and nonfree, exact matching of codings is not required (Gastil 1989: 8).5

4

A brief discussion of domestic economic factors will be presented below. Gastil (1989: 9–17) provides an 11-point checklist for political rights, and a 14-point checklist for civil liberties. The former includes items such as “legislature recently elected by a meaningful process” or “significant opposition vote.” The latter checklist includes items such as “freedom of assembly and demonstration,” “free religious institutions,” or “free trade unions, peasant organizations, or equivalents.” 5

130

5 Democratic Dominoes: Diffusion Approaches …

The basic unit of analysis in this study is the state transition-year. Each state in the system was coded for the absence or presence (and type) of transition for each year it existed as an independent state in the international system during the period from 1973 to 1987. As noted above, there are six possible types of transitions: (1) PF to F; (2) NF to PF; (3) NF to F; (4) PF to NF; (5) F to PF; and (6) F to NF. For any state in this data set, 14 such transition-years are possible. The first possible transition was from 1973 to 1974; in all tables and analyses this transition is referred to as 1974. The last possible transition in the data set was from 1986 to 1987 (referred to as 1987). As noted, a significant movement toward democracy was evidenced in 1989 and 1990. If diffusion effects and a significant trend towards global democracy could be delineated in the period just prior to these dramatic recent events, then the results can be seen as being more robust. Including the latest data would increase the chances that such effects would appear.6 Thus the 1973/74 to 1986/87 period (particularly from 1984/85 onward) can be seen as the temporal or historical context within which the changes of 1989/90 have taken place (see also Russett 1990: 21). All independent states which existed in the system in 1973 generated fourteen transition-years. A number of states became independent subsequent to 1973, however. If a state had entered the system in 1975, then the first transition it could experience would be 1975 to 1976 (or simply 1976). Using this scheme, 2,141 units of analyses (or state transition-years) are produced. Eighty-six states were coded as having no governmental transitions during this period. A total of 131 governmental transitions, generated by 76 different states, were coded. Of the 131 governmental transitions, 77 (or 58.8%) were ‘positive,’ indicating the movement toward liberal democracy operationalized by the first three types of transitions. Thus 54 (41.2%) governmental transitions were ‘negative,’ indicating movement away from democracy (transition types 4, 5, and 6). The total numbers of each type of transition are: (1) PF to F = 25 (2) NF to PF = 51 (3) NF to F = 1

(4) PF to NF = 29 (5) F to PF = 21 (6) F to NF = 4

Only five transitions moved across two categories—only one transition moved directly from NF to F (Greece in 1975) and four cases moved from F to NF. On the basis of this limited sample it seems to be easier to drop democracy in one step than to pick it up. If the transitions of 1989 and 1990 were included, however, these numbers would be balanced. On the other hand, 51 transitions involved moving from NF to PF, and just 21 moved from F to PF. The movement from PF to either F or NF approximately offset

I should also note that the data for 1973 to 1982 came from Table 9, “ratings of nations since 1973” as presented in Gastil (1989). Data from 1983 to 1987 came from Table 6, “country ratings since 1973” as presented in Gastil (1989). The governmental transition data are available from the author on request. 6

5.5 Data and Descriptives

131

each other (25 and 29 respectively). Even with these simple descriptives, there is evidence that although the total set of transitions indicates greater movement towards democracy, there have been significant trends in each direction.

5.6 5.6.1

Analysis and Design: Global Effects Poisson Analyses

The temporal distribution of governmental transitions does not appear to be distributed randomly across time, as seen in Table 5.1. For instance, of the 131 transitions, 50 (or 38%) occur in the first four periods (1974 to 1977) including two of the three prominent clusters of negative transitions. Only 25 (or 19%) of the transitions occur during the 1984/87 period—a period dominated by positive transitions. There is a high of 18 transitions in 1977, and a low of two transitions in 1986. The decline in the rate of transitions begins about 1982. One approach to the study of diffusion, which Most/Starr (1980) have referred to as a “traditional approach,” is the application of the Poisson distribution. The Poisson is a rare event distribution that assumes randomness of occurrence. If an observed set of occurrences does not fit a Poisson distribution, researchers employing the Poisson have come to the conclusion that the events were not distributed randomly in time, and that diffusion effects could be posited as accounting for the lack of fit. We noted above that the universe of analysis involves 2,141 state transitionyears. Of those, 131, or 6%, involved actual transitions. It is clear that governmental transitions can be conceived of as “rare events.” The three assumptions of the Poisson that lead it to be used as a test for diffusion have been set out by Most/Starr (1976: 13): 1. randomness—occurrence of (a transition) in one nation does not alter the probability that the same nation or any other nation will experience a similar event; 2. homogeneity—the rate of proneness to the outcome of an event (lambda) is the same for all nations; and 3. constancy—the rate of proneness to the outbreak of an event (lambda) is constant through time. Discussing the utility of the Poisson, McGowan/Rood (1975: 861) note, “of all processes that generate random variables, this property of independence of past history is unique to the Poisson process. Poisson processes are also stationary stochastic processes, meaning that the origin of time is irrelevant.”

5 Democratic Dominoes: Diffusion Approaches …

132

Table 5.1 Types of governmental transition: by year Type Positive (1) (2) (3) Subtotal Negative (4) (5) (6) Subtotal Total Source The

Year 74 75

76

77

78

79

80

81

82

83

84

85

86

87

2 5 0 7

1 2 0 3

3 5 0 8

3 5 0 s

2 7 0 9

2 4 0 6

2 2 0 4

2 0 0 2

0 3 0 3

2 7 0 5

3 2 0 5

0 2 0 2

2 2 0 4

3 3 0 6 9

8 1 1 10 18

2 1 0 3 11

1 1 0 2 11

1 2 0 3 9

5 1 1 7 11

4 1 0 5 7

1 3 0 4 7

0 1 1 2 11

2 0 0 2 7

0 0 0 0 2

0 1 0 1 5

1 5 1 7

1 1 1 5 1 0 3 6 10 13 author

Table 5.2 presents the results of applying the Poisson distribution to the occurrence of governmental transitions.7 The actual number of years with specified numbers of transitions are matched against the number of years predicted by the Poisson. Following Richardson (1960: 129–30) and McGowan/ood (1975: 867), a chi-square test was employed. The significant chi-square demonstrates that the deviations from the Poisson did not occur by chance. Thus the hypothesis that governmental transitions are Poisson distributed is rejected. At the same time Table 5.2 indicates that we cannot reject the modified Poisson distribution, also called the contagious Poisson. Following Coleman (1964), the modified Poisson is a distribution based on assumptions that “the occurrence of an event does alter the probability that another outbreak will take place” (Most/Starr 1976: 53–54). For the modified Poisson the chi-square test was not significant, meaning that assumptions of nonindependence could not be rejected. Thus through the use of two forms of Poisson analysis, an initial case for the diffusion of transitions at the macro, or global, level is established.

5.7

Global Trends Toward Democracy

From Table 5.1 we saw a distinctive trend of positive governmental transitions over negative ones in the period from 1984 to 1987. There is further evidence for a global movement towards democracy. Based on Welsh (1984, Table 1), the macrolevel of analysis can provide information on different “dimensions of diffusion”

7

The expected values generated by the Poisson were determined by using both Kitigawa (1952) and Statistical Analysis System (SAS) software.

5.7 Global Trends Toward Democracy

133

Table 5.2 Poisson analysis of governmental transitions Number of transitions per year

Observed number of years

0 0 1 0 2 1 3 0 4 0 5 1 6 0 7 3 8 0 9 2 10 1 11 4 12 0 13 1 14 0 15 0 16 0 17 0 18 1 131 transitions divided by 14 years = 9.36 rate Poisson v2 = 44.61 Df = 18 probability = 0.001 Modified Poisson v2 = 21.8 df = 18 probability = < 0.30 Source The author

Predicted by Poisson

Predicted by modified Poisson

0.0014 0.00112 0.0532 0.165 0.385 0.721 1.13 1.5 1.76 1,83 1.71 1.46 1.14 0.82 0.547 0.342 0.20 0.11 0.057

0.006 0.039 0.133 0.314 0.582 0.898 1.2 1.43 1.54 1.53 1.42 1.23 1.015 0.795 0.595 0.428 0.297 0.200 0.130

such as spatial patterns, temporal rate, and extent. The temporal rate involves looking at the “accumulated curve” of adoptions, or occurrences, over time. The extent of diffusion involves the “success rate,” or “adopters and proportion of potential adopters.” Whereas the data in Table 5.1 indicated the temporal distribution of different types of governmental transitions, we may also look at the results of those transitions. Based on the transitions that have occurred, the number (and percentage) of free, partially free, and nonfree states in the international system can be calculated for each year. From this data one can informally derive the “success rate” of democracy—democratic ‘adoptions’—as well as their accumulation. After two successive decreases in the number of free states in 1975 and 1976, the number steadily increases from 38 in 1976 to 56 in 1987 (with one drop in 1983).

134

5 Democratic Dominoes: Diffusion Approaches …

Thus there is a steady, if unspectacular, accumulation of free states—from 27% of the international system in 1976 to 34.6% of the system in 1987. Perhaps as important is the combination of free and partially free states. While PFs fluctuate (as one might expect, being the intermediate stage for both the ‘positive’ and ‘negative’ movement of states), the 57 PFs in 1987 constitute both the largest absolute number and percentage of PFs. In 1974 the combined total of F and PF states was 78, or 56.5% of the international system. By 1987 the combined total was 113 states, or 69.8% of the system. Again, the accumulation of ‘adopters’ of ‘positive’ transitions is substantial—from just over half of the system to almost 70% of the system. At the same time, the absolute number of nonfree states fell to 49 in 1987, the lowest yearly total for the entire time period (down from 60 in 1974, or 65 in 1977). NF states constituted only 30.2% of the system, again its lowest mark, down from 43.5% in 1976. It is clear that there is a steady temporal trend towards greater democracy in the system during the period from 1974 to 1987, which accelerated somewhat during the last four years of that period. Given that states could adopt then reject democratic forms, many types of analyses based on adoption cannot be employed. Nevertheless, we can still approximate some of those analyses. In 1974 the Freedom House data indicate 138 system members. Of these there were 42 free states, meaning there were 96 others. These 96, plus the 24 new states that entered the system before 1987, constitute the pool of potential adopters. By 1987 14 new Fs were added, or 12% of this potential total; a substantial if not overwhelming success rate. Using the same calculations for F + PF states, the 1974 total was 78, leaving 60 NFs. These 60 plus the 24 new members constitute a potential pool of 84. The 1987 total of F + PF states was 113, or an addition of 35. These 35 states constitute 41.2% of the potential adopters, a much more impressive success rate. As for NF states, the proportion of such states in the system dropped by almost one third during this time period. In sum, we can see the steady absolute and percentage increase of free states, and the long term trend towards smaller numbers and percentages of nonfree states. The “accumulated curve” of F and PF states is a clear indicator, moving from just over half to over two-thirds of the system. Even without complex forms of analyses we can still demonstrate that there has been a global trend towards democracy in the system—and that this movement formed the context for the events of 1989 and 1990. For example, note that from 1984 onward the movement from PF to F (1) occurred in states such as Argentina, Brazil, Uruguay, Honduras, Malta, and the Philippines. The movement from nonfree states to partially free states (2) took place in countries such as Hungary, Poland, Haiti, Guatemala, Nigeria, Sudan, South Africa, and Iran.

5.8 Regional Effects

5.8

135

Regional Effects

We have already seen evidence of a global demonstration effect. Governmental transitions are not randomly distributed across time and there is a global trend towards democracy. Are there regional effects which reinforce these global trends? Can we find both positive and negative regional diffusion effects—promoting movements either toward or away from democratic forms? Table 5.3 provides a geographic regional breakdown of governmental transitions, dividing the time period into two equal groups of seven transitions. The shorter time frames help to isolate regional diffusion effects, or spatial heterogeneity. A few broad regional patterns can be discerned. Europe has a strong positive trend (11 positive transitions to 2 negative ones) with similar patterns in each subperiod; sub-Saharan Africa exhibits a clear, but less strong positive pattern (16 transitions to 8). Over the whole time period, only the Indian Ocean island states (not including Sri Lanka), exhibit a negative pattern (7 negative transitions to 2 positive). There is a quite strong positive regional pattern in Latin America. After a set of negative transitions in 1982 and none in 1983, from 1984 to 1986 there are eight transitions, all positive. We may also look at smaller regional groupings, (although this begins to move into possible neighbor effects). As part of the Latin American pattern Brazil, Argentina, and Uruguay all move from PF to F status in 1984 and 1985. Thus Latin America, between 1984 and 1986, appears to have created a powerful regional context for positive transitions, which is reinforced by global patterns and is working through some clearcut neighbor effects. Other indications of regional effects are difficult to disentangle from neighbor effects. Clusters of negative PF to NF, and F to PF transitions occurred with Argentina, Paraguay, and Uruguay (1976/77); with Bangladesh, India, and Sri Lanka (1976/77), and Syria, Iran, Pakistan, Turkey, and Cyprus (1980/82). Clusters of positive transitions, NF to PF took place in the 1977 to 1980 period in Sudan, Syria, Iran, Kuwait, Qatar, and the UAE; as well as in Chile, Paraguay, Bolivia, and Ecuador.8 Geographic regional effects are also difficult to disentangle from other, nongeographic ‘regions.’ Over twenty years ago Russett (1967) demonstrated that the political ecology of the international system contains a variety of ‘regions,’ based

Both positive and negative transitions were “plotted” on global maps. The trends noted above can be seen easily, especially in Latin America and sub-Saharan Africa. What emerges from looking at the maps of positive transitions, however, is an arc of positive transitions from Poland, through Hungary, Yugoslavia, Greece, Turkey, and Iran that loops around the eastern and southeastern border zones of the Soviet Union. Given the scarcity of negative transitions in the latter years, the plot of negative transitions from 1980 to 1987 highlights their presence in the Middle East/South Asian arc—Turkey, Syria, Iran, and Pakistan—as noted.

8

5 Democratic Dominoes: Diffusion Approaches …

136

Table 5.3 Regional breakdown of governmental transitions Region

Western hemisphere North America Central America/Caribbean South America Total Europe East Europe West Europe Total Asia North Asia South Asia South-East Asia Total North Africa Middle East Total Sub-Saharan Africa Indian Ocean Islands Oceania Source The author

Number of transitions 1974 to 1980 Positive Negative

1981 to 1987 Positive Negative

0 2 6 8

0 4 7 11

0 5 6 11

0 2 2 4

1 5 6

0 1 1

3 2 5

0 1 1

3 4 3 10 3 8 11 10 0 1

1 5 3 9 0 4 4 3 5 0

1 0 2 3 1 3 4 6 1 0

1 1 0 2 0 4 4 5 2 2

on factors such as sociocultural homogeneity, trade patterns, networks of international organization membership, and United Nations voting blocs (see also Russett 1990). This complexity of regional ties is highlighted by looking at the Organization of Petroleum Exporting Countries (OPEC) and the Organization for Economic Cooperation and Development (OECD). The Arab members of OPEC are connected by organizational and economic ties, by geographic regionalism, and by common cultural ties. Of the nine Arab members, four experienced NF to PF transitions in the 1977 to 1979 period. Only one of the 24 members of the OECD— linked by organizational and sociocultural ties, especially democratic government —experienced a transition away from free status. This was Turkey in 1981, arguably one of the members with the most tenuous sociocultural ties to the others. Many other analyses could and should be performed, especially as this conception of regions links domestic factors to external relations. New data on the groupings identified by Russett need to be gathered and analyzed to indicate links of salience and interdependence, and thus provide linkage between domestic and international conditions.

5.8 Regional Effects

137

In sum, the regional phenomena noted here support the global analyses. It is clear that many governmental transitions were not random nor free from the regional context—geographic or otherwise—within which they took place. Although we might posit emulation via awareness of events throughout the global system, those same diffusion effects were taking place at a regional level for some parts of the world system.

5.9

Bordering Governmental Transitions: Neighbor Effects

Because the diffusion of governmental transitions appears to be more similar to the diffusion of innovations rather than the contagious spread of war, we have looked for general demonstration effects of emulation at the global and regional levels. At this point however, we will look specifically for first-order neighbor effects that indicate spatial dependency, or positive spatial diffusion based on proximity. To do this, I will use a variant of the interaction opportunity model used by Most/Starr (1980) and Siverson/Starr (1990). As noted, having neighbors or alliance partners at war (WBNs or WAPs) were seen to constitute ‘treatments.’ The treatment was either present or absent with the presence or absence of violent conflict (or war) in the specific study of borders or alliances. Here the treatments are not so simple. We need to create a set of expectations as to what conditions should hold for a BGT (bordering governmental transition) to have a positive spatial diffusion effect. It is important to remember that we will be looking for the relationship between BGTs and those states which: (a) did not experience governmental transitions, and (b) which did experience governmental transitions. I will make two cuts at this level of analysis. The first looks for spatial diffusion effects by including all 2,141 units of analysis (see Table 5.4). Then I will look at a matrix of the 131 actual transitions to analyze BGT treatments in terms of same and opposite direction effects (see Table 5.5). The presence of a BGT treatment in Table 5.4, Part A, requires some discussion. First, I have coded BGTs in the same year (t0) and in the previous two years (thus the treatment period is from t−2 to t0). The central concern is that a BGT treatment be one that could have had an effect on the state in question. Thus a set of coding rules had to be established. If a state did not have a transition in some year, the first step was to identify the condition of that state in that year: was it free, partially free, or nonfree. If there was no BGT then it would be coded into cell (a). If there were a BGT then I matched the BGT to the condition of the state. If a state were free, then bordering transitions of the types (1), (2), or (3) would have no effect (positive diffusion effect) on the state undergoing the treatment. The same logic holds for nonfree states with BGTs of type (4), (5), or (6). In a sense, such states would be ‘immune’ to the effects of the BGT; there would be no place for the effect to spread (see Bremer 1982 for a discussion of ‘immunity’). For example, for

138

5 Democratic Dominoes: Diffusion Approaches …

Table 5.4 Name is missing Part A Presence of a transition in t0

Part B Presence of a transition in t0

Row proportion of cell (b) = 24.8% Row proportion of cell (d) = 41.2% Column proportion cell (d) = 9.8% Column proportion cell (c) = 4.8% Source The author

Neighbour effects: The BGT treatment matrix Presence of a BGT treatment (from t−2 > t0) No Yes No (a) (b) Yes (c) (d) BGT treatment analyses: 1947–1987 Presence of a BGT treatment (from t−2 to t0) No Yes Total No 1512 498 2010 Yes 77 54 131 Total 1589 552 Chi-square = 1.74 df = 1 Probability > 0.20

Table 5.5 BGT analysis: governmental transitions only Presence of a BGT Treatment (from t−2 Positive Negative No RGTs 25 8 46 9 12 31 34 20 77

Positive Negative Total Chi-square for 6-cell table = 5.825 df = 2 Probability  0.05 Phi = 0.21 Chi-square for 4-cell table (without ‘No BGTs’) = 5.1 df = 1 Probability = 0.025 Phi = 0.31 ‘Positive’ transitions are toward democracy: types (1) (2) (3) ‘Negative’ transitions are away from democracy: types (4) (5) (6) Source The author Presence of a transition at t0

to t0) Total 79 52

(1), (2), or (3) BGTs, if the bordered states were already ‘free,’ no governmental transitions in response would be expected; the same holds for nonfree states and the (4), (5), or (6) BGTs. These sorts of states—either free or nonfree and with BGTs for which they were already immune and with no transitions of their own at t0—would also be coded into cell (a).

5.9 Bordering Governmental Transitions: Neighbor Effects

139

For states without transitions, which at t0 were categorized as partially free, any BGT could have an effect, and they were coded into cell (b). Also coded into cell (b) were those states not having transitions, categorized as free and having (4), (5), or (6) BGTs; states categorized as nonfree and having (1), (2), or (3) BGTs.9 For states without transitions, same year treatments took precedence over those from the previous two years. One hundred thirty-one governmental transitions did occur during this period. If states experienced no BGT treatments in the years when such transitions occurred, they were coded into cell (c). All states experiencing a governmental transition that were also treated by a BGT, were coded into cell (d).

5.10

Neighbor Effects: A First Cut

Table 5.4, Part B provides the data for a first set of analyses of interaction opportunity or neighbor effects. As discussed by Siverson/Starr (1989), the search here is for “loose necessity,” and a set of expectations that indicates a necessary relationship between BGT treatments and governmental transitions. Following Siverson and Starr, a search for necessary relationships means we should be unconcerned with large numbers of cases in cell (b), look for no or few cases in cell (c), and expect that the following hold in regard to cell (d): Row proportion of (d) > row proportion of (b) Column proportion (d) > column proportion (c) Column proportion (d) > column proportion (b) This last expectation was seen as a ‘strong’ expectation of necessity. Looking at Table 5.4, Part B, we see that the first two expectations hold, and that cell (c) contains only 3.6% of all the cases, as we would expect from a loose necessity standpoint. States that undergo treatments show some tendency to have governmental transitions of their own. However, the strong expectation is not met, and the chi-square is not statistically significant. Whereas some effects are found, they are not, as I suggested above, as strong as those found for the positive spatial diffusion of war. Nevertheless, the effects are nontrivial. Siverson/tarr (1989: 32–33) address the question of the ‘vacuousness’ of predictions from a necessary condition. They adopt a methodology employed by Anderson/McKeown (1987) in an analysis of the vacuity of a series of ‘necessary’ steps in predicting war. Anderson and McKeown develop two measures which I will use here to evaluate models of ‘loose’ necessity (Siverson/Starr 1989: 33): “The first is del, a measure of

9

For free or nonfree states at t0 without transitions this still holds even if they had same-type BGTs. For example, a free state could have a (1) and a (5) BGT in t0—it would still be coded into cell (b).

140

5 Democratic Dominoes: Diffusion Approaches …

the proportional reduction in error which evaluates the accuracy of the prediction (varying from +1.0 and minus infinity). The second is scope, which measures the vacuousness of the prediction, by indicating the ‘extent to which our conditions are at risk of making a prediction error.’” A nonvacuous necessary treatment would thus have a high scope, or strong risk of making an error, as well as a high del indicating that the treatment substantially reduced the probability of making an error (see Anderson/McKeown 1987: 8). Both conditions hold. As seen in Table 5.4, Part B, del = 0.947 and scope = 0.742, indicators that the effects of the BGT treatments (or ‘filters’ in Anderson and McKeown’s terminology) are nontrivial. Whereas the effects of the BGT treatments do not explain much variance in the standard sense, they do indicate that undergoing a BGT treatment has some effect (as a necessary condition) on the diffusion of governmental transition at the neighbor level.

5.11

Neighbor Effects: A Second Cut

We turn now to an interaction opportunity analysis of those states that did have governmental transitions. The matrix of results is set out in Table 5.5. Some descriptives are in order. Of the 131 transitions, 54 experienced some type of BGT treatment, 77, a majority, did not.10 Even with this large number of nontreatment transitions, the chi-square is still statistically significant at just over the 0.05 level (with a phi of 0.21)—demonstrating a positive spatial diffusion effect. States with transitions toward democracy (positive transitions) are associated with positive BGTs; states with negative governmental transitions are associated with negative BGTs. There is other evidence to support a conclusion that neighbor effects exist. If one substracts the 131 transitions from the 2,141 possible, and divides 131 by the remaining 2,010, the result is 6.5%—a conservative general probability of having a bordering state which is experiencing a governmental transition. In contrast, 37 of the transitions (28.2%) have a same direction BGT treatment (from t–2 to t0). There were 20 transitions (15.3%) with same direction BGT treatments in the same year, t0.11 Both figures indicate that in comparison to all possible state transition-years, states undergoing governmental transitions were clearly more likely to have BGT treatments. Returning to Table 5.5, we can look only at the first four cells (ignoring the last column, “no BGTs”). The truncated matrix is a BGT matrix, looking at the relationships that exist for the 54 governmental transitions having BGT treatments. The 10

An interesting point to note is that states which are islands, archipelagoes, or peninsulas with only one border, account for a full 31 of the 77 transitions without BGT treatments. 11 The contrast is even stronger if we look at governmental transitions with any type of BGT treatment, not just same direction treatments. The proportion of 54/131 equals 41%, compared to the 6.5% of the global population.

5.11

Neighbor Effects: A Second Cut

141

chi-square is statistically significant, just over the 0.02 level, and with a stronger phi =0.31. Again, positive transitions are related to positive BGTs; negative transitions are related to negative BGTs. In comparison to the baseline number of 6.5% we can again see the impact of BGTs. For positive governmental transitions seen in the top row of the matrix, 25 out of 79—or 32%—indicate a diffusion effect. For negative governmental transitions seen in the bottom row of the matrix, 12 out of 52—or 23%—indicate a diffusion effect. It is also clear that although bordering changes are associated with governmental transitions, the effect is greater for movements toward democracy. In Gould’s treatment of spatial diffusion (1969: 11–23), he presents a section on ‘barriers’ to diffusion which act to absorb, reflect, or otherwise block the wave-like spread of some phenomenon as it diffuses. The notion of barriers may be used as a metaphor regarding the number of the states which experienced BGT treatments but were apparently unaffected in terms of their own governmental transitions. It was shown above that a number of states were exposed to a contagious treatment, but were not infected themselves. It must be stressed again that a full explanation of governmental transitions will require detailed considerations of domestic factors as well. Because the thrust of this piece has been on external stimuli which could be modeled through the diffusion concept, the diffusion models presented here are clearly underspecified. Returning to the original Most/Starr (1980) framework, we need to look at internal reinforcement effects as well as the possible effects from spatial diffusion. Such prior experiences would be part of a combination of internal conditions that make a state ‘ready’ for diffusion, or, in contrast, impermeable to outside effects. In conjunction with the delineation of nongeographic regions, one avenue of future research will be to investigate the cultural, linguistic, religious, psychological, or ideological barriers (Gould 1969: 16, 18) which deflect or ease diffusion effects. The array of possible domestic factors is vast, and would require a consideration of the literatures on political change, legitimacy, the conditions of democracy, and so forth. For the specific analysis of governmental transitions, factors such as the history of political coalitions, collective violence, or the development of technology for coercion and control would be relevant. One factor which strikes at the heart of legitimacy, discontent, and vulnerability to external leverage is the economic performance of states.12 Even here, the possible effects of economic performance (and distribution) on internal stability and/or external conflict are many and complex (see, for example, Lichbach 1989; Russett 1987). Based on the relative deprivation tradition in the study of revolution and collective violence, it can be proposed that economic 12

Bruce Russett (in a private communication) has argued that economic conditions (especially deteriorating conditions) need to be taken into account. He notes that it is possible that the Latin Americans’ change first to authoritarian rule in the early 1970s, and then to democracy a decade later, was due less to diffusion effects than to the fact that those governments were unable to handle their economies successfully in two rather different economic phases. In each phase the economic conditions were common to most of these states, and they responded in a similar manner.

5 Democratic Dominoes: Diffusion Approaches …

142

downturns would be related to domestic discontent, affect domestic assessment of governmental legitimacy, domestic perceptions of stability, and ultimately the need to change how the government works. A second set of relationships would relate economic performance to growing economic needs. If needs cannot be met internally then the search for resources would make states more vulnerable to the demands (explicit or implicit) of those states in the system able to provide those resources (Russett 1990). In the period under study here, the states able to provide economic resources were the economically advanced Western democracies. The norms of democracy might be spread through a perceived need to make a recipient more attractive to prospective democratic suppliers. Clearly, a systematic analysis of the relationship between economic performance and governmental transitions needs to be undertaken.13

5.12

Conclusion

These last comments help to bring us back to basic issues regarding diffusion and internal-external linkages. The evidence presented in this article supports a hypothesis that the hard shell of the nation-state is permeable; that it has been penetrated by external demonstration or diffusion effects, and that these effects are generated by the global context, the regional context, and neighbor effects. However, we must understand not only the external context, but the internal one as well, and then move on to the “complex linkages between national and international systems” (Rosenau 1969: 3). Rosenau’s call for “systematic conceptual exploration of the flow of influence across the changing boundaries of national and international systems” (1969, 3) is as relevant today as it was two decades ago— perhaps more so in an era of “cascading interdependence.” The concept of internal-external linkages through a series of models of the diffusion process is one attempt to meet the gap in theory that Rosenau also identified. A key element in the application of diffusion approaches and in the study of internal-external linkages is that of interdependence. One set of hypotheses about which states are more or less ready to be penetrated by diffusion effects, or which will retain their position as ‘barriers,’ must be based on the levels and types of interdependence any state has with its external environment. Any notion that events in the external context alter the willingness of governmental elites, nongovernmental elites, or broader segments of the population toward

13

A preliminary investigation of economic performance (as conducted through change in GNP per capita) was undertaken. In brief, those states moving away from democracy, from F to PF (5), exhibit the worst economic performance. At the same time (1) transitions, which move from partially free to free status, do the best. Although incomplete, these simple analyses indicate that economic conditions are associated with governmental transitions; with economic improvements related to movements toward democracy and the worse downturns tending to be related to movements away from democracy.

5.12

Conclusion

143

governmental transitions must be based on how the external events alter incentive structures for the actors involved. Ties of interdependence affect perceptions of incentive structures and the costs and benefits of governmental change based on the sensitivity and vulnerability (Keohane/Nye 1989) to outside forces, especially the vulnerability to various forms of leverage that external actors can bring to bear (see North/Choucri 1983). All this, of course, is dependent on internal conditions. One crude example can be drawn from events in Eastern Europe. In this region calculations of governmental control over the populace across the Cold War period were in large part based upon the interdependence of the Warsaw Treaty Organization (WTO) states and the specific dependence on the Soviet Union for the military support of the separate Communist regimes. Calculations of control thus included the ability to draw upon Soviet and WTO forces to repress any political or violent movements toward more open government or challenge to the supremacy of the Communist party. These external factors were important in that none of the East European countries could depend upon their own military forces to use violence against internal political movements. Past events—Hungary in 1956 and Czechoslovakia in 1968— indicated the willingness of the Soviet Union and the WTO to use force against challenges to Communist party authority in East Europe where the local military was not a viable tool for repression. A second form of interdependence, in terms of economic relations with the rest of Europe, was also a factor in the calculations regarding governmental change. The movement of the European Community toward 1992 and the concomitant troubles within the Council for Mutual Economic Assistance (CMEA) created considerable economic pressures for change—pressures for the East European states to adapt to the changing economic environment or face highly negative economic conditions. Gorbachev’s policies, in effect, cut the linkages of interdependence/dependence on WTO and Soviet military support, and opened the way for even greater sensitivity and vulnerability to the economic changes and policies of the European Community.14 Diffusion approaches, based as they are on linkage and interdependence concepts, should be of help in our thinking about the spread of democracy. We have demonstrated a global movement towards democracy even prior to 1989/90. The penetration of the hard shell of the state by communications technologies and by global psychological and cognitive interdependence, is captured in models of the diffusion of government transitions. Such models must also alert us to the impact of other interdependence networks.

14

Such conclusions would suggest the importance of continued investigation of the interaction between democracy and economic growth in regard to integration, peace, and security communities, (e.g., Russett/Starr 1989, Chap. 14).

144

5 Democratic Dominoes: Diffusion Approaches …

References Anselin, L.; O’Loughlin, J., 1989: “Spatial Econometric Analysis of International Conflict”, in: Chatterji, M.; Kuenne, K. (Ed.): Dynamics and Conflict in Regional Structural Change (London: Macmillan). Anderson, P. A.; McKeown, T.J., 1987: “Changing Aspirations, Limited Attention, and War”, in: World Politics, 40: 1-29. Bremer, S. A., 1982: “The Contagiousness of Coercion: The Spread of Serious International Disputes, 1900–1976”, in: International Interactions, 9: 29–55. Coleman, J. S., 1964: Introduction to mathematical sociology (Glencoe, IL: Free Press). Collier, D.; Messick, R. E., 1975: “Functional Requisites Versus Diffusion: Testing Alternative Explanations of Social Security Adoption”, in: American Political Science Review, 69: 1299– 1315. Eyestone, R., 1977: “Confusion, Diffusion, and Innovation”, in: American Political Science Review, 71: 441–47. Gastil, R. 1983: Freedom in the world: Political rights and civil liberties 1982–83 (New York: Freedom House). Gastil, R. 1989: Freedom in the world: Political rights and civil liberties 1988–89 (New York: Freedom House). Gould, P., 1969: Spatial Diffusion. Washington, DC: Association of American Geographers, Resource Paper No. 4. Houweling, H.; Siccama, J., 1985: “The Epidemiology of War, 1960-1980”, in: Journal of Conflict Resolution, 29: 641–63. Kegley, C. W.; Raymond, G. A., 1990: When Trust Breaks Down (Columbia: University of South Carolina Press). Keohane, R. O.; Nye, J. S., 1989: Power and Interdependence, 2d edn. (Boston: Scott, Foresman). Kitigawa, T., 1952: Tables of Poisson Distribution (Tokyo: Baifukan). Li, R. P. Y.; Thompson, W. R., 1975: “The ‘Coup Contagion’ Hypothesis”, in: Journal of Conflict Resolution, 19: 63-88. Lichbach, M. I., 1989: “An Evaluation of ‘Does Economic Inequality Breed Political Conflict?’ Studies”, in: World Politics, 41: 431–70. McGowan, P. J.; Rood, R. M., 1975: “Alliance Behavior in Balance of Power Systems: Applying a Poisson Model to Nineteenth-Century Europe”, in: American Political Science Review, 69: 859–70. Midlarsky, M. I., 1970: “Mathematical Models of Instability and a Theory of Diffusion”, in: International Studies Quarterly, 14: 60–84. Midlarsky, M. I., 1975: On War (New York: Free Press). Most, B. A., 1973: “Testing for Geographic Contagion: A Cross-National Study of Nine Types of Violent Events, Latin America, 1946–1965” (Master’s thesis, Department of Political Science, Indiana University, Bloomington). Most, B. A.; Starr, H., 1976: “Techniques for the Detection of Diffusion: Geopolitical Considerations in the Spread of War”. Paper presented at the annual meeting of the International Studies Association, Toronto, February. Most, B. A.; Starr, H., 1980: “Diffusion, Reinforcement, Geopolitics, and the Spread of War”, in: American Political Science Review, 74: 932–46. Most, B. A.; Starr, H., 1990: “Theoretical and Logical Issues in the Study of International Diffusion”, in: Journal of Theoretical Politics, 2: 391–412. Mueller, J. 1989: Retreat from Doomsday (New York: Basic Books). North, R. C.; Choucri, N., 1983: “Economic and Political Factors in International Conflict and Integration”, in: International Studies Quarterly, 27: 443–61. O’Loughlin, J.; van der Wusten, H., 1986: “Geography, War and Peace: Notes for a Contribution to a Revived Political Geography”, in: Progress in Human Geography, 10: 326–52.

References

145

Ray, J. L., 1989: “The Abolition of Slavery and the End of International War”, in: International Organization, 43: 405–39. Richardson, L. F. 1960. Statistics of Deadly Quarrels (Chicago: Quadrangle Books). Rosenau, J. N., 1969: “Introduction: Political Science in a Shrinking World”, in: Rosenau, J. N. (Ed.): Linkage Politics (New York: Free Press): 1–17. Rosenau, J. N., 1984: “A Pre-theory Revisited: World Politics in an Era of Cascading Interdependence”, in: International Studies Quarterly, 28: 245–305. Rosenau, J. N., 1988: “Patterned chaos in Global Life: Structure and Process in the Two Worlds of World Politics”. International Political Science Review, 9: 327–64. Russett, B. M., 1967: International Regions and the International System (Chicago: Rand McNally). Russett, B. M., 1987: “Economic Change as a Cause of International Conflict”, in: Schmidt, Christian; Blackaby, Frank (Ed.): Peace, Defence, and Economic Analysis (London: Macmillan): 185–203. Russett, B. M., 1990: “Politics and Alternative Security: Toward a More Democratic, Therefore more Peaceful, World”, in: Weston, B. H. (Eds.): Alternative Security: Living Without Nuclear Deterrence (Boulder, CO: Westview): 107–36. Russett, B. M., Starr, H., 1989: World Politics: The Menu for Choice, 3d edn. (New York: Freeman). Siverson, R. M., Starr, H., 1989: “Alliance and Border Effects on the War Behavior of States: Refining the Interaction Opportunity Model”, in: Conflict Management and Peace Science, 10: 21–46. Siverson, R. M., Starr, H., 1990: “Opportunity, Willingness and the Diffusion of War, 1816–1965”, in: American Political Science Review, 84: 47–67. Starr, H., Forthcoming: “Joining Political and Geographic Perspectives: Geopolitics and International Relations”, in: International Interactions. Starr, H., Siverson, R. M., 1990: “Alliances and Geopolitics”, in: Political Geography Quarterly 9: 232–48. Walker, J. L., 1969: “The Diffusion of Innovations Among American States”, in: American Political Science Review, 63: 880–99. Welsh, W. A., 1984: “Inter-nation Interaction and Political Diffusion: Notes Toward a Conceptual Framework”. Paper presented at the annual meeting of the International Studies Association, Atlanta, March.

Chapter 6

On Geopolitics: Spaces and Places

The study of international relations sits at the convergence of human inquiry that crosses both time and space.1 The aim here is to elaborate on the spatial context of international relations, to contrast it to the temporal context, and to indicate broadly the continuing importance of the geopolitical spatial context to the study of international relations. I briefly demonstrate how this relationship is based not on an earlier approach based on geographic determinism, but rather possibilism—the possibilities presented by the spatial, geographic, and geopolitical context. In elaborating on space and place, I return to the central research focus of my career: the dynamism and importance of the spatial context for understanding international relations, along with the need to take both time and space into account, the need to appreciate both a locational view and the perceptual/symbolic/constructed view of space and place, and to do so within an increasingly globalized, interdependent, and transnational world system.2

The study of international relations (or international politics or world politics) sits at the convergence of human inquiry that crosses both time and space. Much of my scholarly career has been devoted to the study of space, as reflected in both the theme of the 2014 annual meeting of the ISA, “Spaces and Places: Geopolitics in an Era of Globalization,” and my recent book, On Geopolitics: Space, Place, and International Relations. In a much more cursory overview, I want here to touch on themes central not only to my book, but to a set of issues that form a central research focus of my career: the dynamism and importance of the spatial context for

This text was first published as: “On Geopolitics: Spaces and Places,” International Studies Quarterly (2013) 57, 433–439. The text is republished with the permission of the International Studies Association. 2 This article is based on the Presidential Address delivered at the Annual Meeting of the International Studies Association, April 4, 2013, San Francisco. Much of the paper draws upon my book, On Geopolitics: Space, Place, and International Relations (Boulder, CO: Paradigm Publishers, 2013). I would like to thank Jennifer Knerr of Paradigm for all of her help and advice on this book. A great number of colleagues have influenced my work on geopolitics over the years, but for this specific piece I would particularly like to thank Zaryab Iqbal and Paul Diehl for their assistance in thinking about the contours and direction of my Presidential Address. Paul deserves a special mention for suggesting “spaces and places” as part of the title and focus. 1

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7_6

147

148

6

On Geopolitics: Spaces and Places

understanding international relations, along with the need to take both time and space into account, the need to appreciate both a locational view and the perceptual/ symbolic/constructed view of space and place (and to do so within an increasingly globalized, interdependent, and transnational world system); and to demonstrate how the relationship between geopolitics and international relations covers state and nonstate actors alike. Why be concerned with space and spatiality? Whether attributed to Confucius, to Buddhist precepts, to Buckaroo Banzai, or to generations of geographers, we have all been reminded: “no matter where you go, there you are.” Much of my research has been devoted to what this means. What does it mean—“there you are”? Location? Absolute? Relative? Related to what? Space? Distance? What is the meaning of distance? Of space? Of place? In distinction from a number of scholars of geopolitics (usually ‘realists’ studying national security policy in some form) who see the geopolitical context at least as enduring—if not immutable or deterministic—I have been concerned with the dynamism of that context. Foucault once commented that scholars saw space as “the dead, fixed … immobile” (Agnew/Duncan 1989: 1). I have, however, argued that space or the spatial dimension was dynamic and changing. The realist Gray (1977: 1) once asserted that “Geography is the most fundamental factor in the foreign policy of states because it is the most permanent.” I argue that Gray’s assertion tells only part of the story. Geography is important not just because of its relative stability, but also because of its role in shaping the dynamics of opportunities and risks. Geography affects changing perceptions of the possibilities and probabilities provided by the geographic environment. Although geography—in terms of topography or the absolute distance between two points, for example—is relatively stable, technological change or political change (such as those brought about by the creation or dissolution of alliances) alters the meaning and impact of geography on interaction opportunity and the structure of incentives and risks.

6.1

Context: Time, Space, and Determinism

Time and space are two of the primary ways in which we contextualize social behavior and interactions. But, in much of our work, time—the temporal dimension—seems to be privileged. Abler et al. (1971: 10), in the classic geography text Spatial Organization: The Geographer’s View of the World, succinctly indicate the importance of these two dimensions: “Time and space are obvious and immediate aspects of human existence…. Time and space are the fundamental contexts of all experience… Experience must be located in time and space before we can begin to process it further… Locating an event in the spatio-temporal continuum is our first step in ordering our experience of it.” Although this is rather obvious, it is key to understanding that analyses structured solely (or almost entirely) around time are only telling us half of the story. So we have the question: Why is it that social scientists have been primarily concerned with time and only more recently turned to consider space?

6.1 Context: Time, Space, and Determinism

149

There are some powerful differences in the ways in which time and space have been perceived and understood. These differences have enormous effects on how we think about phenomena and how we design our research to study international relations. The quotation from Foucault starkly outlines some of these differences— space as dead, fixed, immobile, whereas time was seen as rich and dialectic. Abler et al. provide one way of contrasting time and space, and how humans have considered them. Their argument allows us to turn Foucault’s observation on its head. That is, in addition to the perceptual and analytical reasons found in discussing relative space, there are deeper reasons to argue that it is space that is richer and more complex and thus deserves far more attention. Time is simply ‘easier’ to deal with. Abler et al. observe: Because we cannot control our movement through time and because it is divided into formal units like days, lunar periods, years, we are more aware of time than space. Our consciousness of existing in time produces three regions along the temporal dimension: the past, the present, and the future…. At each moment we occupy a point in time. At that point, experience is very intense and immediate. Intensity and immediacy diminish as experience moves further into the past or as events we probably will experience move into an increasingly distant future. (Abler et al. 1971: 8)

I think this is a powerful statement of why we are more aware of time. However, the ‘Western’ linear temporal ordering of the past, the present, and the future— measured by standard units of time such as days or years—echoes the character and advantages geographers give to absolute space or absolute distance when compared to relative space or distance: an ease of measurement and simplicity of conception. In contrast to this view of time, Abler et al. point out how we occupy space: Spatially, we also occupy a nodal (central) position, although we have no widely used categorization of space analogous to the division of time into past, present, and future. Individuals and groups have spatial ranges of various sizes, but these have not been formalized in any generally accepted way…. All of us have territories and ranges and crudely formed conceptions thereof… Perhaps our insensitivity to space is related to the fact that movement in space is voluntary whereas movement in time is wholly involuntary…. (Abler et al. 1971: 9; emphasis added)

What this means is that even absolute space with its standard units (miles, kilometers, latitude and longitude) has no equivalent to the universal human experience with past, present, and future. All humans must be located physically somewhere (as noted, “no matter where you go, there you are”), but there are no common reference points. All location is somehow relative and non-formalized. The notion of movement through space being voluntary whereas movement through time is involuntary and crucial (if stunningly simple). Being more universal, more linear, and invariant (in the sense of it being involuntary), the temporal context is easier to understand and work into our research designs. Just because it is voluntary, the spatial context should be more important to many aspects of our research designs. However, there is no doubt that for other parts of research design, it makes our tasks much more difficult. This discussion points to the dynamism I have thought about and studied over my career. It stands in stark contrast to geographic or geopolitical determinism. It

150

6

On Geopolitics: Spaces and Places

would not be unfair to characterize a major thrust of the work of Harold and Margaret Sprout (for example, Sprout/Sprout 1965) as an attempt to counter previous deterministic views and uses of geography. Although not the only perspective on geopolitics, nor the only scholars to present that perspective, the Sprouts’ version of possibilism has held a central place in the study of international relations. Fox (1985: 27), in elaborating on the Sprouts’ possibilism, once noted that “Harold and Margaret Sprout are the American political scientists with the most sustained interest in and influence on geopolitical thinking from the 1930s to the 1970s.” The Sproutian “ecological triad” is the mechanism by which we join politics and geography. This triad is composed of an entity, its environment, and the entity-environment relationship. The advantages of this framework derive from its applicability to any number of levels of analysis, thus assisting the analyst to cross levels of analysis. That is, whether the focus is on a single decision maker, a small group of decision makers, a foreign policy organization, a government as a whole, or the state as an international actor, the concept of the ecological triad argues that we need to look at the ongoing policy/choice processes within that entity, its context or environment, and then the interaction between the entity and the environment. Determinism is only one form of the entity-environment relationship that could be hypothesized. However, it is a form in which the full causal force flows from the geographic environment to the human or institutional environed entities. In response to this model of the entity-environment relationship, the Sprouts argued the existence of alternatives where decision makers would be capable of making choices. One alternative was their construct of environmental possibilism, the central tenet of which: is that the initiative lies with man, not with the milieu which encompasses him. Possibilism rejects the idea of controls, or influences, pressing man along a road set by Nature or any other environing conditions. The milieu, in the possibilist doctrine, does not compel or direct man to do anything. The milieu is simply there… In the possibilist doctrine, the milieu is conceived as a set of opportunities and limitations. (Sprout/Sprout 1965: 83)

In this view, the environment is seen as a number of factors that limit human opportunities, constraining the types of action that can be taken as well as the consequences of that action. Although the limits set by the environment may be wide or narrow, it is assumed that the limitations are discoverable. Once these limitations are known to some degree, another form of entity-environment relationship comes into play: environmental probabilism. As the humans in the decision units of any entity view their environment, the characteristics of that environment provide cues as to the probability of certain outcomes. The environment presents the entity not only with what is possible, but with what choices would be more or less likely under those particular circumstances. If possibilism rests on the most basic notions of choice, elaborated by probabilism, in that some choices will be made more or less likely, then there is one more component that is essential to the Sproutian alternative to determinism: cognitive behaviorism. This is “the simple and familiar principle that a person reacts to his milieu as he apperceives it—that is, as he perceives and interprets it in light of past

6.1 Context: Time, Space, and Determinism

151

experience” (Sprout/Sprout 1969: 45). This “psychological milieu” is about how humans see the environment and their images of the environmental context. Thus, the entity-environment relationship depends on the perceptions of the entity—a conception of the entity-environment relationship as far removed from determinism as possible (and quite congenial with recent ‘constructivist’ approaches). The ‘real’ world has an impact only after choices are made and an implementation attempt is sent out into that real world. Note, however, that even the ‘feedback’ from the real world somehow must be perceived in order to be learned and to affect future choices. It should now be clear how the set of entity-environment relationships proposed by the Sprouts provided the basis of my own agent-structure framework of opportunity and willingness (see Starr 1978; Most/Starr 1989). I have argued that both opportunity (possibilism) and willingness (probabilism and cognitive behaviorism) are necessary for understanding behavior: The environment must be permissive, and the acting unit must choose. The opportunity-willingness framework forces an analyst to take all three components of the ecological triad into account. The spatial and geographic components of the environment of any international actor are thus essential to understanding choice in foreign policy and international relations. The geopolitical environment includes the effects of space, topography, position, and climate (Osterud 1988). Sprout (1963) asserted that geopolitical hypotheses deal with the configuration and layout of lands and seas, climate, and the distribution of natural resources, as well as those dealing with the distribution of people, social institutions, or behavioral patterns. Thus, all international actors or entities may be located spatially—in some geopolitical arrangement of topography and/or some distribution of people, behavior, and resources. Geopolitical factors in the environment thus provide a structure of opportunities and constraints. The geopolitical structure, including the geographic structure, is “always both enabling and constraining,” as Giddens (1984: 169) has observed about structure in general. Opportunity thus consists of both the possibilities that exist in the international system at any point in history (for example, technology, ideology, religion, and social inventions such as new forms of government) and how those possibilities are distributed in the system. Thus, there are two dimensions to my version of possibilism. First, the phenomenon must already exist somewhere in the world system. The phenomenon—be it nuclear weapons, telecommunications satellites, Protestantism, Marxism, railroads, or financial markets—must have been ‘invented’ so that it is available as a possibility to at least some actors in the system. The second dimension centers on this possibility’s distribution in the system. For example, nuclear weapons do exist; however, most states cannot “take advantage” of them, because they have neither the wealth nor the expertise to produce their own. Though a possibility may exist, limits on resources and choice affect the ability to make use of it. This second dimension is analogous to Harold Sprout’s concern for the distribution of resources, people, or behavior and also derives from the Sprouts’ discussion of “capability analysis” (for example, 1969: 53). The possibilities or opportunities that exist and their distribution help us understand both how costly, or risky, certain options appear to decision makers and how they might calculate the

152

6

On Geopolitics: Spaces and Places

expected utility of those actions. That is, the geographic/geopolitical structure of opportunities and constraints (possibilism) are translated through environmental probabilism into the incentive structures of human beings who have to make choices. The geopolitical environment, then, has an impact on both opportunity and willingness, as people perceive the environmental opportunities and constraints, plug them into a structure of incentives that make choices more or less likely, and, through some form of utility calculations, have them affect the willingness to behave. The geopolitical approach of the Sprouts can, then, be summarized in terms of the existence of interaction opportunities and their effects on choice. Geopolitical factors or environments can affect interaction, processes such as diffusion, or the occurrence of various types of events based on the interdependent choices of states. An interaction opportunity approach also directs our attention to those phenomena or mechanisms that can change the ease of interaction or the importance of spatial relations. For example, technology has been studied for the ways in which it has been used to overcome the spatial and temporal constraints of topography. Technology or anything else in the environment that affects ease of interaction (and thus also the salience or importance of space) is able to change the meaning of the geographic/geopolitical context of environment. New technology permits humans to overcome physical barriers to the movement of ideas or things (especially military things) or to overcome the spatial distribution of resources through the creation of man-made synthetic alternatives. Alliances have a similar ability to change the impact and meaning of time and space, by ‘leapfrogging’ physical barriers, such as mountains, oceans, or simply distance (see Starr/Siverson 1990). These ideas are important in that they support the view that the geopolitical environment is dynamic in nature. The meaning of the geopolitical context can be changed rapidly through such mechanisms as technological innovation, the formation or dissolution of alliances, or the integration or the disintegration of states. Far from a determinist view of the immutability of the earth’s physical environment, the constraints of the environmental context (which would ‘determine’ human choice and behavior) are being overcome by human technological invention or political innovation—thus changing what the physical environment means to decision makers in terms of the opportunities it presents and the probabilities of behavioral choices. As noted, an important component of cognitive behaviorism is the set of geographic/geopolitical images that decision makers hold and how those affect their other images and calculations of choice. Decision makers’ “mental maps” derive from geographic maps to form images of global and regional environments and the risks, threats, and opportunities in those environments (see Nijman 1991). Again, this follows on earlier work of Sprout/Sprout (1971: 248), who note that “maps are geographic models” and, as such, are simplifications that affect our images of the geographic/geopolitical environment (on the nature of cartography and maps, see Monmonier 1991; Akerman 2009).

6.2 Context: Space and Spatiality

6.2

153

Context: Space and Spatiality

But what is space? As with all important concepts, space is multidimensional. As made explicit in the Sprouts’ idea of “cognitive behaviorism,” space is also a concept that takes on meaning only as it is perceived by individuals or groups of individuals. We must now introduce the idea of ‘place.’ Agnew and Duncan provide three approaches to the idea of place. They note that geographers usually look at these separately, but that all three meanings are “complementary dimensions” of place. By looking at these three approaches, we can begin to highlight the meaning of space or spatiality: Approaches to defining a geographical concept of place have tended to stress one or another of three elements rather than their complementarity. Firstly, economics and economic geographers have emphasized location, or space sui generis, the spatial distribution of social and economic activities resulting from between-place factor cost and market price differentials. Secondly, microsociologists and humanistic geographers have concerned themselves with locale, the settings for everyday routine social interaction provided in a place.3 Thirdly, anthropologists and cultural geographers have shown interest in the sense of place or identification with a place engendered by living in it. (Agnew/Duncan 1989: 2)

It should be noted that the first two approaches to place are clearly related to ‘opportunity.’ The third approach is related to ‘willingness,’ as part of identity and self-identification, of how people locate themselves in the universe, as well as how they value things.4 The first approach—location or space—is the one with which I think students of international relations are most familiar, emphasizing the location of things in relationship to other things and how things are distributed. This idea of spatial contingency is picked up in Kirby/Ward’s (1987: 3) definition of ‘spatiality’ as “a contingent factor within the operation of any social formation,” in which society’s “components are themselves dependent upon their spatial setting.”5 This view of place-as-location matches the two basic ways to think about location, as presented by Abler et al. (1971: 59): “absolute location” and “relative location.” According to Abler et al., “Absolute location is position in relation to a conventional grid system designed solely for locative purposes.” In this view, location is provided by such things as latitude and longitude or a street address. The concept becomes much richer, but also much trickier, in the second way to think

3 O’Loughlin/Anselin (1992: 16) note that “Giddens prefers the term locale over place because place suggests a spatial container while locale is the setting of interaction and the contextuality of social life.” In earlier work, I have also used Giddens in this way, preferring the broader idea of context over more narrow meanings such as “container.” 4 See, for instance, the work of John Agnew or that of David Newman (for example, 1991, 1996, 1999). 5 A good example of the function of space and spatiality in the “operation of any social organization” is found in the fascinating ‘Sugarscape’ simulation presented in Epstein/Axtell (1996). The “agent-based” computer modeling methodology presented by Epstein and Axtell puts spatial distribution at the heart of a multidisciplinary model of how societies develop, grow, become more complex, and die.

154

6

On Geopolitics: Spaces and Places

about location: “Relative location is position with respect to other locations.” This can be expressed in terms of a variety of factors, such as distance or travel times from other locations, or the cost of such travel. We have noted above how technology changes “relative location”—places that were once weeks apart in time, are now only hours apart—and other mechanisms, such as alliances, can do likewise. As Abler et al. (1971: 82) note, “Any activity we undertake which makes it easier or more difficult for people, ideas, or objects to move through space has significant effects on spatial processes and the structures they produce” (emphasis added). Abler et al. (1971: 72) note that prior to 1950, geographers generally dealt with space (and distance) in the ‘absolute’ mode. Since then, however, in most research “relative location and relative distance has been used to define new kinds of stretchable, shrinkable spaces.” Contrasting the two views, they note: “There are a large number of ways of describing distance and location in a relative context, but in the absolute context we are restricted to customary and unchanging units such as miles, kilometers, or degrees of latitude or longitude to measure distance” (1971: 59). In a view similar to the one I have expressed in my work, they note (in a way that would be at least congenial to Foucault) that “human decisions constantly alter and restructure relative spaces. It has taken geographers a long time to challenge the pervasive tyranny of absolute space” (1971: 82). Not only do human decisions ‘alter’ relative spaces, but various types of relative space explicitly take time into account, so that relative space and relative distance, and the meaning of relative space or distance, are heavily dependent on perceptions: “The spaces in which people live are much more psychological than absolute. If we are concerned, as we often are, with explaining spatial interaction, what is important is not how far two interacting places are from each other in absolute space, but rather how far the people at the two places think they are apart” (Abler et al. 1971: 75). This view captures the Sprouts’ cognitive behaviorism, along with the extant set of ‘constructivist’ approaches and models (see also Snyder et al. (1954) “definition of the situation”). In my own work, this involves how perception of space or distance affects willingness. Although in recent decades the overwhelming focus has been on relative location and distance, geographers have not abandoned the traditional questions of geography: “‘Where?’ and ‘What is where?’” These questions—especially the latter —become more complex and interesting when dealing with relative distance. For example, in absolute space, using unchanging measures such as miles or kilometers —conceived of as Euclidian space—the shortest distance between two points is a straight line. Geographers, looking at relative space, have provided other ways to think about distances between two points; for example, a road that winds its way around a mountain. Similarly, cities where streets and avenues form grids that channel or constrain movement between the corners generated by the grid can be viewed as “Manhattan space.” Manhattan space is “a variant of Euclidian space, in which the shortest distance between two points is a path consisting of line segments which meet at right angles” (Abler et al. 1971: 73). Thus, a straight line as the

6.2 Context: Space and Spatiality

155

shortest distance between two points, or “as the crow flies” distance, loses its meaning in a setting where a straight line is impossible. Manhattan space is a useful example of how any number of constraints can, and do, change the meaning of relative location or relative distance.

6.3

Relative Spaces

Earlier we introduced place as dependent on perception, identity, and values. We can now elaborate on that, using additional ways to look at relative location or relative distance. In Fig. 6.1, we can see four additional different ways in which space can be presented and thus contrast relative locations in relative spaces to the ‘reality’ of absolute space. The figure indicates how five separate points can change in their distances/location to each other based on absolute space, time-space, cost-space, and social-space. Using absolute space, each location—from A to E—is located six kilometers from its next neighbor. Location A is situated in absolute space six kilometers from location B, which is located six kilometers from location C, and so on, based on “a conventional grid system designed solely for locative purposes.” The other three types of space are relative spaces. ‘Time-space’ is based on the amount of time it takes to go from one location to another (or move some object from one location to another). Whereas deterministic geopolitics is based on absolute location/distance, most of the questions we study in international relations regarding security, conflict, international political economy, or cooperation are based on time-space (and, as we shall see, cost-space). This is where technology has an important impact on how the meaning of space can change. In Fig. 6.1, note that location C is now substantially closer to location A than location B. Such locations may flip for a number of reasons; for example, A may be closer to C in time-space because they are linked by an expressway, which shortens the time to drive between them, in contrast to the other locations, which are connected by winding, two-lane roads (or no roads at all!). Time-space is reflected in international relations’ discussions of a “shrinking world.” For example, we can track the pace of civilian intercontinental travel since the early nineteenth century. Around 1820, the earliest steamships traveled about 5 miles per hour. However, the transoceanic liner RMS Lusitania (which figured so prominently in the US entry into World War I), which was launched in 1907, could travel approximately 30 miles per hour. The DC-3 airliner, introduced in 1936, had a maximum speed of approximately 230 miles per hour. But, only forty years later, following continually faster jet airliners, the Concorde supersonic airliner began flying commercially in 1976 with a speed just over Mach 2, or around 1350 miles per hour! If we graphed these speeds, we would see a sharply accelerating exponential curve. Although the absolute location and distance between New York and London, for example, remained the same, the time-distance between them fell from about 2 months for the early trans-Atlantic steamships to about three and a half

156

6

On Geopolitics: Spaces and Places

Fig. 6.1 Relative locations in relative spaces. Source The author based on Abler et al. (1971)

hours for the Concorde.6 In the national security arena, we can find a parallel reduction of time-space with the development of intercontinental bombers and the subsequent development of ICBMs. In any of these types of examples, it is the meaning of distance that has changed in regard to human perception and choice. Similarly, the cost of moving across distance has changed as well. Whereas time-space captures much of what Boulding (1962) meant by the “loss-of-strength gradient,” cost-space captures the rest. Simply put, cost-space deals with the costs of moving objects from one location to another. Cost-space is critical to what Agnew and Duncan meant when they stressed economics in discussing ‘location,’

Regarding time-space, Abler et al. (1971: 83) note, “A peculiarity of time-space convergence is that distant places converge on each other at a greater rate than close places.” This reflects the impact that transportation technology has on spanning large distances more quickly and the much more marginal improvements in moving across a city than across a continent or an ocean. One example they present is looking at the time distance between Edinburgh and London in 1776 and then in 1966. The change that they found was based on the comparison of stage coach (5760 min) to airplane (180 min). This yields a time-space convergence of 29.4 min per year. That is, for each year between 1776 and 1966, the two cities became closer by an average of almost half an hour.

6

6.3 Relative Spaces

157

dealing with “economic activities resulting from between-place factor cost.” In Fig. 6.1, cost-space is represented by the bus fare between locations. In the figure, each location is equidistant from its neighbors—the bus ride costs $0.50 from location A to location B, and $0.50 from location B to location C, and so on. However, we also know that some public transportation systems are based on fare-stages, where the cost per absolute distance does not stay the same, so that fares are not proportionate to absolute distance: for example, the British bus system. Taxicab fares are based on cost-space, but also include the time it takes to get from location A to location B as well as the distance traveled (and sometimes within broad fare areas), so that, again, price is not based on the absolute distance traveled. Just as changes in technology reduce the time it takes to move from one point to another, they also affect the costs of moving people and things. Not only must railroads be invented to become part of the menu of opportunities, but they must be available to any specific area if they are to reduce time and cost (consider the cost of renting an apartment, or buying a house, near a commuter train station compared with one that is a thirty-minute drive away). Although closer in terms of time-space if moved by plane, some types of goods are closer (that is ‘cheaper’) in cost-space if they move by rail, truck, or ship, owing to size or the amount that must be shipped to be economically worthwhile. Although Boulding’s loss-of-strength gradient focuses on the time required to move military capabilities, cost-distance also highlights the economic and human costs of moving military capabilities. The cost of moving large numbers of troops has historically been high in terms of money and lives, with armies often losing more lives to disease, accidents, and hunger in long marches or sieges than in actual combat. The “projection of force,” especially by major powers using blue water navies, aircraft carriers, or long-distance bombers, has often been much less costly than the movement of massive numbers of troops. The use of US air power in the first Gulf War is an example. Again, a nuclear-armed ICBM also illustrates how technology can alter cost-space. To indicate the variety of other possible relative dimensions, Fig. 6.1 also includes “social-space,” measured in terms of interactions: here, the number of social contacts per week. Although locations might be close in terms of absolute space/distance, people may not interact proportionately with those who are closest. This is contrary to one of the basic tenets of geography, or Zipf’s “principle of least effort,” which argues that there will be more interaction between entities that are close than those farther apart (Zipf 1949). Social-space might reflect the different ethnic makeup of groups at specific locations or major class differences (which could be reflected not only in local prices but the types of stores in the different areas). At such a microlevel of analysis, we see this view of social-distance in the study of protracted social conflicts. As developed in the work of Edward Azar (for example, Azar 1984), protracted social conflict was a special form of social conflict that was long term, ongoing, and permeated all aspects of the two societies involved. It was seen as highly intractable and apparently unresolvable to normal conflict resolution mechanisms because of the extensive linkages between development, violence, and identity that permeated the two actors. Classic cases of protracted conflict, the Israeli-Palestinian conflict

158

6

On Geopolitics: Spaces and Places

and the conflict in Northern Ireland, also indicate that a key feature was conflict between the government of a state and at least one nonstate actor on that state’s territory. Friedman (2002: 11) has noted that the typical protracted conflict situation finds the “geodemographic integration of rival nations,” that is, the intermingling of peoples from different ethnic-national groups on the same territory. Thus, this constant opportunity for conflict is also embedded within a context of constant willingness. At the same time, individuals and groups that literally share the same territory have very low levels of interaction: Although absolute distance is low, social-distance is quite high in terms of locale—“the settings for everyday routine social interaction provided in a place.” Social-space distances can also be found when looking at the international system in terms of a ‘feudal,’ or hierarchical, model. Studies of former colonial areas in the 1960s and 1970s revealed that they continued patterns of interaction that found former colonies interacting most with the former imperial power, next with other former colonies of that same imperial power, and far less with other states, including their neighbors and near regions. In the 1960s and 1970s, for many Francophone African states, the quickest way to fly from one to another would be through Paris. The same relationships applied to former British colonies in Africa, where the quickest way to travel from one to another would be through London (for example, see Gleditsch 1969). International relations scholars have also used social-space in looking for ‘distances’ between states or groups based on any number of socio-political-economic factors. For example in Russett’s (1967) study of regions, his analyses revealed homogeneous groupings of states that were ‘close’—that is, were similar and clustered together—in terms of voting in the UN General Assembly, their networks of memberships in international organizations, trade interdependence, and sociocultural similarity. Looking across these analyses, we clearly find evidence that countries that were similar on one or more of these dimensions did interact more (see also Wright’s (1942) version of social-space in his different approaches to ‘distance’). As we can see, the various forms of relative space are ways of viewing space, place, and distance that help make the argument that geography and spatial approaches are dynamic and complex. One way this happens is because “Individuals and groups of people live at intersections of numerous relative spaces” (Abler et al. 1971: 82; emphasis added). By noting the types and number of such intersections, we can see the different ways space is dynamic: “The number of dimensions we use and the way we measure distance along them determine the nature of any space we construct. By choosing different distance measures, we can change space” (Abler et al. 1971: 73). Briefly returning to my own work, this task of developing “different distance measures” in order to reconceptualize space is exactly the aim of my research using GIS to remeasure and reoperationalize opportunity and willingness in regard to contiguous borders (for example, Starr 2002). In an absolute distance sense, contiguity indicates that two states touch. This is an on/off measure of contiguity that is relatively static unless the border is altered by war or diplomacy (for example, integration, decolonization). By using GIS to develop indexes of ease of interaction

6.3 Relative Spaces

159

(opportunity) and salience (willingness) that characterize any single border or any area along each border, variance has been added—variance that changes the nature and meaning of the border.7 The elements from the ARC/INFO GIS used to develop these indexes identified the location of key aspects of a state’s transportation, communication, energy production, industrial, agricultural, and security infrastructures. This approach demonstrates how human activity and its creations can change borders and the meanings of borders beyond on/off touching—and might do so relatively quickly.

6.4

To Conclude

We see, then, why relative location/distance is so strongly focused on time-space; why it is so oriented to how location and space relate to time, and vice-versa. And this is why technology and the range of other human invention are so important— they change the meaning of space, location, and distance. Time-space and cost-space are about how long it takes to move objects across some distance, from some location to another, and the costs of doing so. They are crucial in our study of security, of military affairs, and thus to realist models of international politics. They are crucial in our study of economic, social, and political transactions: to our study of trade, diplomacy, and integration, and thus to liberal, neoliberal, and pluralist models of international politics. And, social-space captures much of the social construction of meaning and identity central to a broad set of constructivist approaches. The opening sentence of this article argued that space and time are the central components of the context of international relations. Thus, space and geography are inextricably intertwined with the study of international relations. Very simply, all humans live (in permanent abodes) on the territory of some state. As with Mark Twain, the reports of the death of the territorial state have been greatly exaggerated. Territory and the borders that separate states from each other provide key elements in the structure of the global system—mapping the number and arrangement of the territorial units upon which all humans live. Territory and borders permit a spatial approach to international or global politics by setting out the location of states and their absolute and relative distances from each other. Borders represent important legal boundaries between states. Such boundaries are just as important (or perhaps even more so) between democracies at peace with one another, who rely on a multitude of economic and social transactions, as they are to adversaries who ‘securitize’ their borders with fortifications and military capabilities.

The new dataset also permitted new analyses and findings concerning the relationships between borders and both conflict and cooperation. See Starr/Thomas (2002, 2005), and Starr (2013: Chaps. 6, 7, 8).

7

160

6

On Geopolitics: Spaces and Places

Borders continue to act as factors of constraint on human interaction, as well as factors that facilitate human interaction. Territory and borders have significant effects on international relations because of their meaning to humans (scholars, policymakers, or peoples). There is a deep connection between individuals and groups to the territory on which they live and the territory that their ancestors lived on. What is so important to the identity and cohesion of groups, thus, also can become one of the major obstacles in managing and resolving conflict between those groups. I have argued that geography is ‘dynamic’ in that the meaning of space, distance, territory, and borders can change in the perceptions of peoples and foreign policy-making elites. This can be done through technology or through alliances. Both the internal and external politics of peoples, substate organizations, and states affect the creation, dissolution, and meaning of borders. In sum, even in the world of growing interdependence and globalization, geography, territory, and borders have important roles to play in the reality of, and study of international relations. I hope I have helped us understand a bit better the importance of geography, border, space, and place—even in the contemporary globalized and ‘borderless’ world, and the attendant claim of de-territorialization. It is important to stress that despite living in the interdependent, transnational, and globalized world of the twenty-first century, geographic factors such as territory and borders are still integral and meaningful elements of world politics.

References Abler, Ronald; Adams, John S.; Gould, Peter, 1971: Spatial Organization: The Geographer’s View of the World (Englewood Cliffs, NJ: Prentice-Hall). Agnew, John A.; Duncan, James S., 1989: “Introduction”, in: Agnew, John A.; Duncan, James S. (Eds.): The Power of Place (Boston: Unwin Hyman). Akerman, James R. (Ed.), 2009: The Imperial Map (Chicago: University of Chicago Press). Azar, Edward E., 1984: “The Theory of Protracted Social Conflict and the Challenge of Transforming Conflict Situations”, in: Zinnes, Dina A. (Ed.): Conflict Processes and the Breakdown of Social Systems (Denver: Monograph Series in World Affairs, University of Denver). Boulding, Kenneth E., 1962: Conflict and Defense (New York: Harper and Row). Epstein, Joshua M.; Axtell, Robert, 1996: Growing Artificial Societies (Cambridge, MA: MIT Press). Fox, William T. R., 1985: “Geopolitics and International Relations”, in: Zoppo, C. E., Zorgbibe, C. (Eds.): On Geopolitics: Classical and Nuclear (Boston: Martinus Nijhoff). Friedman, Gil, 2002: “Toward a Spatial Model of Protracted Conflict Management: The Palestinian Case” (Unpublished doctoral dissertation, University of South Carolina). Giddens, Anthony, 1984: The Constitution of Society (Berkeley: University of California Press). Gleditsch, Nils Petter, 1969: “The International Airline Network: A Test of the Zipf and Stouffer Hypotheses”, in: Peace Research Society: Papers, 11: 123–153. Gray, Colin, 1977: The Geopolitics of the Nuclear Era (New York: Crane Russak). Kirby, Andrew; Ward, Michael D., 1987: “Space, Spatiality, Geography, Territoriality, Context, Locale—And Conflict”. Paper Presented at the Annual Meeting of the American Political Science Association, Chicago, September. Monmonier, Mark, 1991: How to Lie with Maps (Chicago: University of Chicago Press).

References

161

Most, Benjamin A.; Starr, Harvey, 1989: Inquiry, Logic and International Politics (Columbia, SC: University of South Carolina Press). Newman, David, 1991: “On Writing “Involved” Political Geography”. Political Geography Quarterly, 10: 195–199. Newman, David, 1996: “Writing Together Separately: Critical Discourse and the Problems of Cross-Ethnic Co-Authorship”, in: Area, 28: 1–12. Newman, David, 1999: “Real Spaces, Symbolic Spaces: Interrelated Notions of Territory in the Arab-Israeli Conflict”, in: Diehl, Paul (Ed.): A Road Map to War (Nashville, TN: Vanderbilt University Press). Nijman, Jan, 1991: “The Dynamics of Superpower Spheres of Influence: U.S. and Soviet Military Activities, 1948-1978”, in: International Interactions, 17: 63–91. O’Loughlin, John; Anselin, Luc, 1992: “Geography of International Conflict and Cooperation: Theory and Methods”, in: Ward, Michael D. (Ed.): The New Geopolitics (Philadelphia: Gordon and Breach). Osterud, Oyvind, 1988: “The Uses and Abuses of Geopolitics”, in: Journal of Peace Research, 25: 191–199. Russett, Bruce M., 1967: International Regions and the International System (Chicago: Rand McNally). Snyder, Richard C.; Bruck, H. W.; Sapin, Burton, 1954: Decision Making as an Approach to the Study of International Politics. Princeton University: Foreign Policy Analysis Project. Sprout, Harold, 1963: “Geopolitical Hypotheses in Technological Perspective”, in: World Politics, 15: 187–212. Sprout, Harold; Sprout, Margaret, 1965: The Ecological Perspective on Human Affairs (Princeton: Princeton University Press). Sprout, Harold; Sprout, Margaret, 1969: “Environmental Factors in the Study of International Politics”, in: Rosenau, James N. (Ed.): International Politics and Foreign Policy (New York: Free Press). Sprout, Harold; Sprout, Margaret, 1971: Towards a Politics of the Planet Earth (New York: Van Nostrand Reinhold). Starr, Harvey, 1978: “‘Opportunity’ and ‘Willingness’ as Ordering Concepts in the Study of War”, in: International Interactions, 4: 363–387. Starr, Harvey, 2002: “Opportunity, Willingness and Geographic Information Systems: Reconceptualizing Borders in International Relations”, in: Political Geography, 21: 243–261. Starr, Harvey, 2013: On Geopolitics: Space, Place, and International Relations (Boulder, CO: Paradigm Publishers). Starr, Harvey; Dale Thomas, G., 2002: “The ‘Nature’ of Contiguous Borders: Ease of Interaction, Salience, and the Analysis of Crisis”, in: International Interactions, 28: 213–235. Starr, Harvey; Dale Thomas, G., 2005: “The Nature of Borders and Conflict: Revisiting Hypotheses on Territory and War”, in: International Studies Quarterly, 49: 123–139. Starr, Harvey; Siverson, Randolph M., 1990: “Alliances and Geopolitics”, in: Political Geography Quarterly, 9: 232–248. Wright, Quincy, 1942: A Study of War (Chicago: University of Chicago Press). Zipf, G. K., 1949: Human Behavior and the Principle of Least Effort (Cambridge, MA: Addison-Wesley).

Chapter 7

Opportunity, Willingness and Geographic Information Systems: Reconceptualizing Borders in International Relations

Abstract This article reports on a continuing project which has developed a major reconceptualization and revision of how borders may be seen and measured through the use of GIS. Using the data layers of the ARC/INFO GIS system, a new dataset has been developed which allows analysts to talk about the specific qualities of borders in terms of opportunity and willingness, that is, the ease of interaction and salience, respectively (This text was first published as: “Opportunity, Willingness and Geographic Information Systems: Reconceptualizing Borders in International Relations,” Political Geography, 21 (2002) 243–261. The permission to republish this text was granted by Elsevier). The theory and method behind this reconceptualization is described. The results are represented both in visual terms—maps— and through the use of a quantitative dataset which lets us go beyond simply observing the number of borders a state possesses, whether or not a border existed between two states, or the length of that border. The dataset is presented and discussed, as well as preliminary analyses of the borders of conflict dyads from three separate conflict datasets. The basic ‘interaction opportunity’ model that underlies the opportunity and willingness framework is supported.

7.1

Geopolitics, Borders and IR: A Brief Introduction

A review of the literature on war, militarized disputes, enduring rivalries and alliances, indicates that over the past 20 years there has been a renewed attention to the role and impact of geography in the study of international relations. Much of this is related to the ‘new geopolitics’ which treats geography as an essential part of the context of possibilities and constraints that face foreign policy decision makers (see Goertz 1994; Goertz/Diehl 1992; Starr 1991a; and even Ward 1992). Studies of the diffusion of behavioral phenomena as well as the investigation of the relationships between proximity, contiguity, location and territory to international interactions, have burgeoned. These activities by students of international relations have paralleled those of geographers who are focusing on a ‘new geopolitics’ based on the possibilities that the geopolitical environment provides to human decision makers. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7_7

163

164

7 Opportunity, Willingness and Geographic Information …

There has been a good deal of work from an agent-structure perspective which has addressed those questions that fit within this broad geopolitical perspective. The development of the opportunity and willingness framework by Starr (1978) and subsequent work with collaborators, has sought both to refine the theoretical basis of these concepts as well as to specify the relationships between them (Cioffi-Revilla/Starr 1995; Friedman/Starr 1997; Most/Starr 1989). These concepts have also been clarified through efforts to operationalize them in the study of international political phenomena. The latter investigations include the study of the diffusion of international phenomena: violent conflict (Most/Starr 1980; Siverson/ Starr 1991), and democracy (Starr 1991b, 1999). The central feature of the analyses has been a focus on the nature and effects of spatial proximity as operationalized by international borders. One major aim of the present project can be seen as a concept clarification exercise—to revise and reconceptualize how we think about borders and how they should be measured. While there are many facets to a geopolitical approach to international relations, borders have been of primary interest to and the main focus of, geopolitical scholarship. For example, as one important component in the mapping of factors related to the onset of war, the Correlates of War project has developed the most extensive and complete dataset on borders available for international systemic actors since 1816 (see Gochman 1992, for a description of the COW measurement rules; see also Starr/Most 1976). Borders have been studied as part of the analysis of many of the central concerns of international relations. A brief list of these concerns would include: the number and types of interactions among states; interdependence among states, within regional groupings and the level of interdependence within the international system as a whole; regional integration; the probability of war among states; the diffusion of war and other forms of international conflict; the diffusion of additional international phenomena, such as the spread of democracy; understanding why and how territory affects the onset of war; the processes underlying the lack of war between pairs of democracies; the effects of alliances; the question of international regions and the structure of the international system.

7.2

Conceptualizing (and Re-conceptualizing) Borders

Thus, the location of states, their proximity to one another, and especially whether or not they share ‘borders’, emerges time and again as key variables in studies of international conflict phenomena: from major power general war, to the diffusion of international conflict, to the analysis of peace between pairs of democracies. From Boulding’s (1962) ideas of ‘behavior space’, ‘loss-of-strength gradient’ and ‘critical boundary’ to the simple but profound concern of geographers that humans interact most with those to whom they are closest, there are powerful theoretical reasons to be interested in borders, and how they affect international relations. But how exactly do borders affect international interaction?

7.2 Conceptualizing (and Re-conceptualizing) Borders

165

Clearly, a key dimension for many researchers is proximity (see for example, Gochman’s (1992) discussion of borders as they relate to the overall COW project). The diffusion research of Starr and colleagues (Most/Starr 1980) moved to the study of borders after concluding that the diffusion of certain phenomena could only be studied by looking at units that were ‘relevant’ to one another—and that such relevance could be indicated by geographical proximity (see also the work of Lemke 1995, 1996). Proximity, in turn, could be operationalized through ‘borders’. Borders were seen as important indicators of proximity because they had important relationships to both the opportunity and willingness of state actors as conceptualized by Most/Starr (1976). One key aspect of borders is that they affect the interaction opportunities of states, constraining or expanding the possibilities of interaction that are available to them. States that share borders will tend to have a greater ease of interaction with one another, and thus will tend to have greater number of interactions. This idea developed from multidisciplinary sources, such as economist Boulding’s (1962) concept of the loss-of-strength gradient; or geographer Zipf’s (1949) ‘law of least effort’. The important issue raised here is that borders create the opportunity for interaction (see Siverson/Starr (1991) and Starr/ Most (1976) for a full discussion of geographic opportunity). Such opportunity might be seen in terms of the number of other countries with which any single state has interaction opportunities. It might also be seen in the degree to which such opportunity exists between any particular pair of states. So, for example, Wesley (1962) argues that the length of a common border between two countries is a better measure of ‘geographic opportunity’ than simply the number of borders. And, presaging the GIS discussion to be presented, Wesley goes on to suggest than length should be measured not in ‘actual physical length’ but in terms of population units. Starr and colleagues have argued that the opportunities for interaction view of borders gets at the important conceptual core of proximity in a way that other measures of ‘distance’ do not. Such measures have included the use of the air mileage between the capitals of states to measure distance (Garnham 1976; Gleditsch/Singer 1975). One purpose of the present project would be to test the utility of such different conceptions of proximity as interaction opportunities. Secondly, borders also have an impact on the willingness of decision makers to choose certain policy options, in that they act as indicators of areas of great importance or salience. Because other states are close, having greater ease of interaction and the ability to bring military capabilities to bear, they are also key areas of external cues (or diffusion). Accordingly, activities in these areas are particularly worrisome, can create uncertainty, and thus deserve attention. The notion that changes in bordering areas create uncertainty because of their proximity was based on arguments developed by Midlarsky (1970, 1975), and applied in Most/Starr (1980). Starr/Most (1976: 10) were also particularly concerned with the ‘roles that different types of borders appear to play’ in war involvement. Different types of border might have differential impacts on both opportunity and willingness. Thus, borders were differentiated in terms of homeland borders and borders generated by colonial territories. This differentiation allowed the testing of whether all territory was seen

166

7 Opportunity, Willingness and Geographic Information …

as equally important, or whether homeland territory generated greater willingness than more distantly held colonial/imperial territories. Implicitly tested in such analyses was the notion that it was homeland territory per se, that was important: that the proximity of any homeland territory of one state to any homeland territory of another state was the important factor. While some of the Most and Starr diffusion analyses indicated that this was probably the case, other analyses also demonstrated the strong impact of colonial territorial borders on the diffusion of war. Simply, colonial territories were responsible for creating a greater number of opportunities for conflictual interaction (in a way that is usefully complementary to Choucri/North’s (1975) notion of lateral pressure ‘intersections’). Starr/Most (1976) also distinguished between land-based contiguity and across-water proximity. Again, such a distinction implicitly dealt with possible variations in ease of interaction and salience. The present project seeks to build upon these two dimensions of borders as indicators of proximity, to revise and reconceptualize how borders may be seen and measured. The use of GIS (Geographic Information Systems) will permit a much fuller and clearer specification of borders by allowing us to talk about the specific qualities of borders in terms of opportunity and willingness. The reconceptualization will permit us to go beyond simply observing the number of borders a state possesses, whether or not a border existed between two states, or the length of that border. By so doing, a number of questions raised in studying the issues noted above can now be more fully addressed.

7.3

Methodology: ARC/INFO and the Conceptualization —Operationalization of Borders

Geographic information systems, developed through the early to mid-1960s, are now the focus of a large amount of literature produced by geographers and regional scientists. It is not my intent to review that literature here, nor cite the most recent contributions to it. As could be expected, there are many approaches and perspectives on GIS. It is important, however, to understand that a GIS is a tool, founded on a variety of computer technologies, that permits the integration of data about the spatiality of phenomena along with data about other characteristics of those phenomena.1 It is important to note that GIS is more than mere computer mapping. According to Cowen (1990: 57), the heart of a GIS system is its ability to 1

According to Marble (1990: 10), to be considered a true GIS a system must include the following four major components: (a) a data input subsystem which collects and/or processes spatial data derived from existing maps, remote sensors, etc; (b) a data storage and retrieval subsystem which organizes the spatial data in a form which permits it to be quickly retrieved by the user for subsequent analysis; (c) a data manipulation and analysis subsystem which performs a variety of tasks such as changing the form of the data through user-defined aggregation rules or producing estimates of parameters and constraints for various space-time organization or simulation models;

7.3 Methodology: ARC/INFO and the Conceptualization—Operationalization …

167

overlay various layers or coverages of data so that the GIS would have, “created new information rather than just have retrieved previously encoded information” (emphasis added). The reconceptualization of borders using GIS derives from the ability to generate new measures—or new information—about the nature of borders. The GIS system utilized in this study is ARC/INFO, developed and supplied by Environmental Systems Research Institute, Inc. (Environmental Systems Research Institute 1991). It is currently one of the most widespread commercial GISs in use globally. Its strengths, in part, reside in its ability to integrate many kinds of data, as well as, “an open architecture which allows it to be linked to a number of relational database management systems” (Peuquet/Marble 1990: 91). ARC/INFO employs a ‘georelational’ approach, which abstracts “geographic information into a series of independently defined layers or coverages, each representing a selected set of closely associated geographic features (e.g., roads, streams, and forest stands)” (ESRI 1992: 14).2 The large scale database consists of the 16 layers of data. These layers contain data ranging from physical characteristics, such as drainage networks, hypsography (elevation and topographic relief), and land cover, to man-made features such as road networks, railroad networks, and aeronautical data.3 The GIS methodology, however, must be driven by theoretical considerations. The various layers of the ARC/INFO GIS contain a great number of variables and the key question must be which of these variables should be selected to create valid indexes to represent opportunity and willingness (ease of interaction and salience). and (d) a data reporting subsystem which is capable of displaying all or part of the original database as well as manipulated data and the output from spatial models in tabular or map form. 2 Data for this project is from the Digital Chart of the World (DCW), produced by ESRI for the Defense Mapping Agency in 1992. The data contained in the DCW was derived primarily from maps in the Defense Mapping Agency Operational Navigation Chart series that were used to generate a 1:1,000,000-scale vector database covering the entire surface of the earth. One major (perhaps ‘heroic’) assumption of the project, is that the border data generated by the 1992 DCW can be usefully applied backward for 20–25 years, and forward for at least a decade. That is, the data retain validity as a rough surrogate for the ease of interaction and salience of areas for this time frame. 3 List of data layers in the ARC/INFO Digital Chart of the World: Political and Oceans; Populated Place; Railroads; Roads; Utilities; Drainage; Drainage Supplemental; Hypsography; Hypsography Supplemental; Ocean Features; Physiography; Aeronautical; Cultural Landmark; Transportation Structure; Vegetation; Land Cover. A 17th layer, the Data Quality layer, provides information on the particular source of data for a given tile and when that source was last updated. A detailed description of the GIS programming has been set out in the Technical Appendix to the National Science Foundation grant (generated from the ongoing technical logs of the three research assistants). Readers interested in obtaining a copy of the Appendix should contact the author. Starr/Bain (1995) describe how various coverages within the ARC/INFO system were used to revisit the opportunity and willingness dimensions of borders. In subsequent work (with the research assistance of Deb Thomas and Richard Deal), the specific methodology used for extracting and combining the border data from the various ARC/INFO coverages has changed substantially. However, the data sources, coverages, and overall criteria for determining the opportunity for interaction and salience did not vary.

168

7 Opportunity, Willingness and Geographic Information …

There is a large literature on the nature of boundaries produced by geographers; (and by geopoliticians writing earlier this century). This literature often focuses on the meaning of boundaries for group formation, group identity, and group maintenance (see Falah/Newman 1995, as well as such seminal works as Prescott 1987 or Glassner 1992). As noted, the mere existence of a border can tell us many things and generate research hypotheses. Yet, ‘a border is not a border is not a border’. Borders serve many functions, and obviously take on different meanings in different specific contexts [e.g. see such discussions as Giddens (1984), or Goertz (1994)]. No single operationalization of border, and no dataset based on such an operationalization, will be able to provide the total historical-political context for all pairs of states. Nevertheless, drawing on two decades of attention to these ideas, the argument here is that the opportunity and willingness conceptualization does tap key elements of the proximity-border concept, and constitutes a progressive step in the more general conceptualization of borders, context, and the analysis of international interaction.

7.3.1

Opportunity for Interaction (Ease of Interaction)

Regarding opportunity, the variable selection decisions were based on theoretical considerations presented by both geographers and IR scholars. The notion of ease of interaction derives from Boulding’s (1962) concern with the “loss-of-strength gradient”, and the ability to project conventional military power. Out of the welter of possible variables (and taking various technical/analytic constraints into account), three central factors for the movement of land-based military capability were selected—the existence of roads, railroads, and the steepness of terrain. This project draws from Wesley (1962), who suggests that scholars should use cross-border roads to operationalize geographic opportunity. Similarly, Lemke (1996), using Bueno de Mesquita’s (1981) operationalization of the loss-of-strength gradient, is concerned with the distance over which military forces can move in specific periods of time and how this relates to the propensity for wars or MIDs (militarized disputes). Lemke is specifically concerned with paved roads and railroads in attempting to estimate the loss-of-strength gradient, and those countries that make up ‘relevant’ dyads/neighborhoods for African states. Based on the work of Boulding, Bueno de Mesquita, and Lemke, an index created from these three factors both reflects ease of interaction, and is applicable (valid) across a large set of international dyadic boundaries. Using data available in the DCW’s data layers an index was constructed which aggregates values generated from ARC/INFO. It can be used to characterize any border or border segment on the globe. After reviewing all the variables within all of the layers, the following were selected to create an ‘ease of interaction’ index. The first variable looked for the presence or absence of roads within the locations being studied. Roads included multi-lane divided roads, as well as primary and secondary roads (Layer 4, Road Layer, RDLINE). The second variable was the presence or absence of railroads

7.3 Methodology: ARC/INFO and the Conceptualization—Operationalization …

169

(Layer 3, Railroad Layer, RRLINE). The third variable involved the slope of an area, which was based on the elevation values of contour lines (in mean feet above sea level; Layer 8, Hypsography Layer, HYNET), and was derived from a digital terrain model by converting the hypsography into a triangulated irregular network. Each of these values was investigated for a buffer area of 10,000 m on each side of all international borders.4 The methodology used by Starr/Bain (1995) involved the use of hexagons and the combined density of the different coverages for roads, railroads and hypsography produced by a polygon attribute table for each hexagon. This generated a metric (an ease of interaction index running from 0 to 373) whereby high ‘saturation’ or density indicated the greatest ease of interaction. The least saturated hexagons are those which are most difficult to move across; the most heavily saturated hexagons are the most easily traversed. However, despite several advantages to the use of hexagons, this methodology increased both the computing capacity and time required for generating the data from ARC/INFO. A methodology using the vector data directly was developed which appeared to simplify the generation of the data considerably, while producing roughly analogous results.5 The revised color map of opportunity for interaction for Israel is seen in Fig. 7.1. With this formulation we simply note the presence or absence of roads and railroads (rather than the saturation of a hexagon), and represented the hypsography or slope as follows: coded 1 if the slope was 05°, coded 2 if the slope was 5–20°, and coded 3 if it was greater than 20. This created a simple combined 1–4 index, with 4 (red in the color maps) representing the greatest ease of interaction, and 1 (dark green in the colormaps) the most difficult areas to move across.6 Note that the revised procedures described here were specifically developed to generate a four-category scheme. This was purposely done in order to facilitate the translation of the color maps into black and white representations. While maps are 4 One of the more useful aspects of GIS systems is the ability to create any size ‘buffer’ on either side of a border. It should be noted that the original tests and analyses were performed on Israel. Given Israel’s size, a 10,000 m buffer on each side of the border seemed appropriate. However, after other test-case analyses, it was decided to use a 50,000 m buffer. 5 For example, it is estimated that the final data for a map of Israel’s Vital Border measures required only one-tenth the computer space as the data required for Israel’s Vital Borders using the hexagon methodology. All of the information, at all the stages required to construct the Vital Borders (see discussion below) required 2.5 to 3 megs. Using the earlier hexagon methodology, however, the figure was closer to 15 megs. 6 The Opportunity for Interaction/Ease of Interaction index was developed as follows: • 4 = presence of a road and the presence of railroad, and low slope • 3 = a road or a railroad, and low slope • 3 = a road and a railroad, and medium slope • 3 = no road, no railroad, and low slope • 2 = a road or a railroad, and medium slope • 2 = a road and a railroad, and high slope • 2 = a road or a railroad, and high slope • 1 = no road, no railroad, and medium slope • 1 = no road, no railroad, and high slope

170

7 Opportunity, Willingness and Geographic Information …

Fig. 7.1 Opportunity for interaction along Israel’s borders. Source The author

an important medium for the presentation of results, it must be stressed that any section of any border can now be represented by a value from 1 to 4: values that can be used in data analyses within the GIS, both with other GIS variables or any other data sets that are imported into the ARC/INFO GIS. Such a dataset will be described below. Figure 7.1 is important in demonstrating that the ease of interaction can vary along any single border that a state might have with a contiguous neighbor (the values, as represented by different colors, vary). The opportunity for interaction variable can be used to indicate this variation along any single border (for example

7.3 Methodology: ARC/INFO and the Conceptualization—Operationalization …

171

Israel’s border with Lebanon). This would capture the variation that might occur on a very long border—any particular portion of a border can be thus be characterized as to its degree of permeability. Thus, we are now able to go beyond the simple idea that in some way contiguity provides the possibility for interaction—while some parts of some borders would make this highly likely or possible, other parts would make interaction much less likely. We could make such judgments irregardless of the length of a border, or the number of different borders that a state might have.

7.3.2

Salience (Willingness)

The salience dimension of proximity/borders is concerned with the importance or value of territory along or behind a border. Again, the question is how importance/ value is to be measured. Below I will discuss an alternative hypothesis that it is territory per se which is important. However, here we must be concerned with indicators which would discriminate the level of value or concern over territory (as Most and Starr did by differentiating between homeland territory and colonial possessions; and between contiguous land borders and across-water borders). Drawing once more on geographers, demographics are seen as important: the territory on which a state’s population lives. This was to be operationalized by areas of population concentration. A capital city, the locus of governmental activity and the symbol of the state, should also be used to indicate the importance of territory. While I disagree that distance between capital cities should be considered a primary indicator of proximity, such studies (Gleditsch/Singer 1975) do highlight the central importance of capital cities. Note that in selecting areas of population concentration and the seats of government we have now captured all three of the central elements of the state found in the international relations literature: territory, population, and government. Areas of urban concentration including urbanized areas and capital cities were extracted from the Populated Place Layer, Layer 2 (PPPOLY and PPPOINT). Other coverages provided the location of items that would indicate the importance of an area. For instance, from Layer 13, the Aeronautical Layer, active civil and military airports were identified (AEPOINT). The Cultural Landmark Layer (Layer 14) provides a catalogue of such items, including: military camps, forts, oil wells and refineries, power plants of various kinds, water tanks, factories, industrial complexes, hospitals, telecommunications stations, etc. The wide variety of items found in Layer 14 was used because the substantive importance of any single type of installation could vary considerably across states. By identifying the location of key aspects of a state’s transportation, communication, energy production, industrial, agricultural, and security infrastructures, we have items that tap ‘importance’ in a manner generally relevant to all states. The salience index was developed in much the same fashion as the index for opportunity for interaction. After reviewing the various coverages, the salience or importance of a border area was determined by places of population concentration, state capitals, airfields, and selected cultural features located within a 50,000 m buffer of the region’s borders. One rule that was

172

7 Opportunity, Willingness and Geographic Information …

used across methodologies, was that a capital city was automatically coded with the highest value found in any of the units of analysis. The revised, vector approach to salience is described thus: “The willingness for interaction component of the study provided more of a challenge in using a vector approach because the hexagons were used to ‘count’ the number of cultural features, number of population centers, and number of airports within each one. Using the POINTDISTANCE command in combination with a FREQUENCY command allowed each feature to be given a value based on the number of other features that were within 4 km. These could then be mapped based on the value, showing where clusters arose” (NSF Technical Report Section 2 by Thomas; see also Deal, Technical Report Section 3 on how the salience indexes were constructed). The results of these revised analyses are shown in Fig. 7.2, which displays the clustering of point coverages indicating the importance of an area, with graphics representing the numbers of points that overlap within 4 km ranges. Note that while reducing the time and capacity requirements for analysis, the vector-based datasets/methods also produce more graphically pleasing and interpretable results. While the data-generating strengths of GIS have been stressed, it would be useful here to emphasize as well the visual utility of GIS—its ability to generate maps which help investigators to look at data differently, and provide heuristics for generating hypotheses and model specification. Yet, while color map representations of salience are useful, remember that any area in a buffer around a border can now be characterized by a value from 1 to 4, which can be utilized in data analyses. A four-value scale was created to indicate areas with one important feature up through those with four or more. The size of the circles also helps indicate the level of salience. And, again, Fig. 7.2 is important by demonstrating how borders may differ in their importance—in terms of where people live, where the capital city is located, where significant elements of the transportation, military, or economic systems are situated. Portions of borders where more of these items are located, (within a 50,000 m buffer of the border), could be seen as more important or salient (the areas in red or orange) than segments without population centers, or economic, military or transportation facilities. As with opportunity for interaction, this representation of the salience of borders permits us to differentiate whole borders, to differentiate portions of long borders, and to make sense as to why some borders might be seen as more important than others; why changes or events across some borders might generate more uncertainty than occurrences across other borders. The use of a GIS dataset, then, gives us a new mechanism for operationalizing a state’s borders. A GIS system has been used to create new data. Through the indexes generated, we can attach values to a single dyadic border or border segment. These values will indicate the ease of interaction provided by that border, and/or the importance of any particular border or border segment. These two dimensions can be used separately or combined.7

7

Recall that Most/Starr (1989) argue that opportunity and willingness are jointly necessary conditions for certain types of behavior, and that they are related to each other in complex ways (see

7.3 Methodology: ARC/INFO and the Conceptualization—Operationalization …

173

Fig. 7.2 Salience of Israeli borders. Source The author

also Cioffi-Revilla/Starr 1995). For want of a better term, I have suggested that a border with high values on both could be considered a ‘Vital Border’. The core of this concept is that an arc or a border segment may combine high or low values reflecting both opportunity and willingness. Vital borders thus represent areas that are both highly permeable—easy to cross—and also encompass population centers and/or features of economic, political or social importance. Once more, they can be represented by values that can be used in statistical analyses or represented on maps.

7 Opportunity, Willingness and Geographic Information …

174

Table 7.1 Components of a new dataset: the example of Israel’s borders Israel’s border with Length (km) Area (km2) Percentage ease of interaction category 1 2 3 4 Percent salience category 1 2 3 4 Percent vital border category 1 2 3 4 Weighted average ease of interaction Weighted average of salience Weighted average of vital border Source The author

7.3.3

Egypt

Jordan

Lebanon

Syria

220 21,108

410 39,282

110 8944

92 10,184

7.41 4.44 86.03 2.12

28.97 18.13 50.24 2.67

29.58 24.78 42.56 3.08

19.06 12.43 61.55 6.96

91.62 6.18 2.20 0.00

92.34 6.31 1.35 0.00

81.50 11.48 6.13 0.88

90.23 7.22 2.56 0.00

10.80 80.50 8.56 0.90 2.83 1.11 1.98

45.60 46.70 7.49 0.22 2.27 1.09 1.62

49.30 36.80 11.90 2.03 2.19 1.26 1.67

29.90 56.80 12.90 3.76 2.56 1.12 1.84

A New Dataset on the ‘Nature’ of Borders

As shown in Table 7.1, the GIS-generated maps can be reduced to a relatively compact dataset useful for quantitative analysis. For each of the 301 contiguous land borders between states 17 variables have been developed, which can be transformed into a variety of nominal, ordinal, and interval measures. For any dyad border (see the example of Israel’s borders used in Table 7.1) we can present the length of that border in kilometers, and the area under the buffers created from that border. From these two variables we can present the percentage of each border that falls into categories 4 through 1 (or, red through dark green in the color figures). This can be done for ease of interaction or saliency (or vitalness). Knowing the length of the arc, the area under the buffer along it, and the percentage of each category, permits the analyst to use interval data (as noted below) or broadly based categories such as high-salience or low-salience. Note also that Table 7.1 provides a weighted average for each border in terms of ease of interaction, salience, or vitalness, showing the average value across the whole border.8 8

This dataset is available from the author.

7.3 Methodology: ARC/INFO and the Conceptualization—Operationalization …

175

Table 7.2 GIS-based dataset: summary statisticsa Ease of interaction

Salience

Minimum 1.195 1.000 Maximum 3.296 1.369 Median 2.800 1.013 Mean 2.597 1.044 Standard deviation 0.500 0.071 a N = 301 cases; values are weighted averages (except for length) Source The author

Vitalness

Length

1.097 2.462 1.918 1.818 0.299

3.0 6900.0 520.0 792.8 863.8

This dataset includes 151 states with land borders, which generate 301 separate contiguous land borders between states. The states in this group thus average almost four borders each. Table 7.2 provides descriptive data on the total set of borders, using the weighted averages. For example, we see that the average salience is quite low, barely getting above 1.000 (with a maximum value of 1.369 on the 4.000 scale). This means that although we find many ‘red’ areas (scale of 4.000) on the maps, they constitute only very small portions of the total border. The values for ease of interaction are much higher (thus, so are the values for vitalness). Interestingly, the border with the highest salience score exists between Moldova and the Ukraine. In many ways this should not be surprising since until less than a decade ago, this was only an internal border, or the equivalent of the border between Connecticut and Massachusetts. With a weighted average of 1.342, the German-Dutch border is the next highest. A cluster of relatively high salience borders are found among the original members of the EEC. And, because of a high density of road and rail facilities, they also have among the highest weighted averages in terms of ease of interaction. That is—the borders of the original core countries of the EU also have borders that look like the internal jurisdictional boundaries of states.

7.4

The Scientific Utility of a GIS-Generated Border Data Base

One basic point raised in Most/Starr (1989) was that researchers needed to be much clearer as to the broader concepts which were really under investigation, so that their models and the resulting research designs could be more logically and fully specified. Perhaps ‘borders’ can be used in some research for reasons that are innate to ‘borderness’—that they separate entities from one another. However, as discussed above, most uses of borders involve their representation of proximity—that is, entities are close to one another, important to one another, and have an enhanced ability to interact with one another. But, does the existence of a border actually

176

7 Opportunity, Willingness and Geographic Information …

represent these notions? Borders that are difficult to traverse, either commercially or militarily may not fit this idea of proximity. Borders which are ‘buffered’ by empty and meaningless spaces may not fit this idea of proximity. Conversely, legal borders in the contemporary world may be meaningless in terms of full permeability and high levels of transactions—as in the European Union.

7.4.1

Revisiting IR Hypotheses with More Fully Specified Borders

A wide array of research questions based on the assumption that borders indicate proximity, salience, and ease of interaction may be addressed by a vital borders dataset. For example, which types of borders are most or least related to spatial diffusion? In the Most and Starr analyses of the diffusion of conflict (Most/Starr 1980), they investigated whether states which were subjected to the ‘treatment’ of having a Warring Border Nation (WBN) were more likely to become involved in conflict than those without such a treatment. Is it the nature of the border rather than numbers of borders that affect conflict behavior? Boulding (1962) also introduced the concept of critical boundaries, as part of his concern with the viability of states in regard to neighbors.9 One suggestive analysis in Starr/Bain (1995) was that the maps of salience and opportunity for interaction indicated that in 1967 Israel had changed its legal borders with Jordan and Syria to match what could be considered to be its critical boundaries. What, then, is the relationship between vital borders and critical boundaries? Are attempts to match the two behind the events listed in the MIDs dataset? causes of war in general? causes of wars over territory in particular? One finding of Siverson/Starr (1991: 54–55) is that the relationship between joining an ongoing war and being subjected to warring border nations (WBNs) and/ or warring alliance partners (WAPs) is one of ‘loose necessity’. Many states have treatments, but do not join wars. In fact, having only one treatment (only one WBN and/or WAP) appeared to have almost no effect on the behavior of states. Thus, one potential research project using the new dataset would be to investigate the nature of the borders that separate the state from its WBN or WAP. If, for example, the border is a vital border, does it increase the probability of the diffusion of war/ conflict? That is, both sets of studies were concerned with ‘treatments’. A new null hypothesis could be that any WBN treatment enhances the probability of war diffusion. More fully specified hypotheses would propose that vital border WBNs Boulding (1962: 265) notes: “The legal boundary of a nation, however, is not always its most significant boundary. We need to develop a concept of a critical boundary, which may be the same as the legal boundary but which may lie either inside it or outside it…The penetration of an alien organization inside this critical boundary will produce grave disorganization... War, therefore is only useful as a defense of the national organism if it is carried on outside the critical boundary” (emphasis in original).

9

7.4 The Scientific Utility of a GIS-Generated Border Data Base

177

have a greater probability of enhancing diffusion; (or that a border needs to score high on either salience or opportunity for interaction). Is it simply ‘borderness’ in some vague sense, or these more specific qualities that are involved? Lemke (1995, 1996) raises the same question. As with Most and Starr, Lemke is concerned with the key question of identifying ‘relevant dyads’. Indeed, he goes beyond this to try to identify ‘relevant neighborhoods’. Using this project’s newly created opportunity for interaction index, Lemke’s questions can be specified even more closely. Another research project could involve the comparison of Lemke’s findings to those based on GIS generated border data on opportunity for interaction. Further refinement of the procedures used (particularly the size of the buffers employed) are clearly needed, as is an independent operationalization for critical boundary. But even setting aside the question of critical boundaries—do states (especially those with large numbers of borders) demonstrate a higher incidence of conflict along salient borders? along borders with the greatest ease of movement?

7.4.2

Revisiting Borders as Part of the War Puzzle

Part of the considerable research devoted to borders and territory (e.g. Vasquez 1993) suggests that territorial contiguity is a major determinant of whether or not a state will go to war with another state. Indeed, as a significant piece of the war puzzle, Vasquez suggests that territorial contiguity is the source of conflict most likely to result in war (Vasquez 1993: 307). However, as noted above, perhaps simple contiguity may not be the critical factor. Dropping one level of analysis lower, Vasquez also hypothesizes that the nature of the border between two states also affects the probability that states will go to war. Specifically, he hypothesizes that borders that coincide with natural frontiers or that traverse uninhabited regions or are seen as having little value, are much less likely to provoke wars than dissimilar borders and border areas (this is a theme picked up later by Lemke). The dataset generated by the GIS project provides an ideal way to test this latter hypothesis regarding the nature of borders. Three different groups of conflict dyads were selected for this analysis: a set of 22 enduring rivalries dyads (Goertz/Diehl 1993, 1995); a set of 27 territorial disputes dyads (Huth 1996); and a set of 61 militarized interstate disputes dyads (using the 1996 version of the MIDS dataset). These cases were selected for pairs of states that shared a contiguous land border, and where the conflict (or series of conflicts) involved fell into the broad temporal band covered by the GIS data (see note 2). In order to test the null hypothesis that borders coinciding with natural frontiers (a greater difficulty in interaction) have either no effect on the likelihood that states will go to war with neighbors or make states more likely to go to war, one must compare the nature of borders where conflict has occurred with those where conflict has not occurred. The same is true for the effect of the perceived importance (salience) of a border area. One strategy would be to compare the nature of conflict borders with all other borders in the system. However, this strategy is flawed since

178

7 Opportunity, Willingness and Geographic Information …

it fails to account for differences in government and for differences in the propensity of individual states to enter into wars. Instead, a more conservative strategy is employed that can take these differences into account; testing for statistically significant differences between conflict dyad borders (the shared contiguous homeland border between two states) and the remaining borders of the two states which form the conflict dyad. An alternative hypothesis is also examined; it posits that rather than the nature of the border being important, the length of the border is the primary distinguishing factor between conflict borders and non-conflict borders. The longer a border is, the greater the opportunity for interaction and, therefore, conflict. As stated earlier, each border has been measured in terms of ease of interaction, salience, and a joint measure of the two, vitalness. Border areas have been assigned to one of four categories, category one being the most difficult for interaction or least important and category four being the easiest for interaction and the most important. Similarly, vitalness is also measured in four categories with category one being the least vital and category four being the most vital. These measures are then given as percentages of the total border length (ease of interaction) and area (salience). They have also been calculated in absolute terms of kilometers and kilometers squared for each of the four categories. Finally, a weighted average for ease of interaction, salience, and vitalness is given for each border. The means for enduring rivalry borders are compared to the remaining borders of states in the enduring rivalry. Specifically, a test of independent means has been conducted for each category both in relative terms (percentage) and absolute terms (kilometers or kilometers squared). The choice of a test of independent means is appropriate since the nature of any state border is independent of the nature of the remaining borders for that state. Of the 23 comparisons made (nine each for ease of interaction and salience, and five for vitalness), surprisingly, not a single category measuring the nature of the border shows a statistically significant (a = 0.10) difference in means. On this basis, one can conclude that the nature of the borders of a state engaged in an enduring rivalry does not significantly improve our chances of predicting which neighboring state will be the enduring rival. On the other hand, the alternative null hypothesis that no statistical difference exists between the mean border length for enduring rivalry conflict dyads and the mean length of remaining enduring rival borders can be rejected. The probability of seeing a sample difference this large (520.81 km), if the difference in the population equals zero is p = 0.09. This suggests that based solely on the knowledge of the length of the borders of a state engaged in an enduring rivalry, we can confidently predict which state will be the enduring rivalry conflict dyad partner. Conflict dyads taken from Huth’s (1996) territorial disputes dataset show similar results. Whereas with the enduring rivalry data set, no statistically significant differences existed between the conflict dyad borders and the other borders of enduring rivals in terms of ease of interaction, salience, or vitalness, two different categories of salience show statistically significant differences. These are the number of square kilometers ranked category one and the percentage of the border ranked category three. Oddly, the difference between the territorial dispute dyads and the remaining borders of the disputants has the opposite sign from which the

7.4 The Scientific Utility of a GIS-Generated Border Data Base

179

original hypothesis would suggest. The conflict dyads from Huth’s territorial dispute data set actually have more kilometers of low importance border area than do their corresponding non-conflict border areas. Similarly, the hypothesis would lead one to expect that conflict dyads would have a larger percentage of high importance areas than corresponding non conflict dyads; however, this is also shown to be false. Nonetheless, again the null hypothesis associated with border length can be easily rejected—the probability of seeing this large of a mean difference (465.19 km) in the sample if the population difference is actually zero is very small (p = 0.03). Thus, support for the argument that the longer a border is the more likely that concerned states will go to war appears to be growing stronger. The final set of analyses uses the militarized interstate dispute (MIDS) data. Similar tests regarding the differences between the means of conflict dyads and the remaining borders of conflict parties have been conducted. The MIDS data is much more inclusive than that of either Goertz/Diehl (1993, 1995) or Huth (1996). Consequently, 61 usable conflict dyads have been identified as opposed to the 27 for Huth and 22 for Goertz and Diehl. Nevertheless, a phenomenon similar to that found in the Huth territorial dispute data exists in the MIDS data. Several categories show statistically significant differences (a = 0.05), but in every case the sign of the difference is opposite of that suggested by the hypotheses. One is forced once again to accept the null hypotheses that more difficult to cross borders and less important border areas do not lower the likelihood that states will go to war with that neighbor. However, differences in the mean length of borders for conflict dyads and nonconflict borders for conflict parties remain statistically significant, with a difference of 295.40 km (p = 0.03).

7.5

Conclusion

These initial analyses do not support Vasquez’s notion that dropping to lower levels of ‘interaction opportunity’ increases the ability to explain war. The alternative hypothesis, that the nature of the border doesn’t matter but that what does matter is the length of the border, does receive broad support from an analysis of three major conflict data sets. Yet, the finding that the length of the border matters does suggest that the general concern by Most and Starr with ‘opportunities for interaction’ (Most/Starr 1980) was generally correct in terms of territorial contiguity; and, indeed, that Wesley (1962) was on the right track. It also supports other analyses by Vasquez (1993) in which he demonstrates the importance that any territory has to states. Territory appears to be important; the opportunities for territory to become part of conflict are increased by the length of contiguous territory, and not by more specific measures of opportunity and willingness. These analyses provide some indication of the utility of the GIS-based conceptualization and dataset. They demonstrate that such a dataset can be used to investigate a number of related questions, for example: What sorts of borders can be found between states in enduring rivalries? What is the nature of the territory over

180

7 Opportunity, Willingness and Geographic Information …

which conflicts arise? Goertz/Diehl (1992), Holsti (1991), Huth (1996), for example, focus on territory per se as a cause of war; as both the issue over which war breaks out, and as a factor which increases the stakes of a war. Such analyses provide us with a very important alternative hypothesis: it is territory—any territory —which creates an opportunity for conflict, which serves as the issue for war, and which makes the stakes worth fighting over. The GIS-based dataset now permits analysts to test these competing hypotheses. In sum, both theoretically and substantively geopolitical factors are important to a wide range of issues in the study of international relations. One key aspect of borders is that they affect the interaction opportunities of states, constraining or expanding the possibilities of interaction that are available to states. States that share borders will tend to have a greater ease of interaction with one another. Secondly, borders also have an impact on the willingness of decision makers to choose certain policy options, in that they act as indicators of areas of great importance or salience. The ARC/INFO GIS permits us to operationalize and investigate these two dimensions—opportunity as ease of interaction, and willingness as salience/importance. Using data available in the different data layers found in ARC/INFO’s Digital Chart of the World, we have constructed indexes of both ease of interaction and of salience. They can be used to characterize any border (or arc) or border segment on the globe. The use of a GIS dataset permits a new mechanism for operationalizing a state’s borders. We can now go beyond simply noting the existence of a border, or its length. Through the indexes generated, we can attach values to the ease of interaction a border or border segment provides, and/or the importance of any particular border or border segment. These two dimensions can be used separately or combined. The GIS generated indexes permit us to tap both dimensions, and to use them singly or combined given the research question under consideration. Acknowledgements This research project has been supported by a University of South Carolina Research and Productive Scholarship Award (No. 13570-E120), which in turn was instrumental in securing a National Science Foundation grant (SBR-9731056) to continue the project. I would also like to acknowledge the contribution of the research assistants who have so substantially contributed to this project through the excellence of their GIS programming: Will Bain, Deb Thomas, Richard Deal, and Guojing Shou. In addition, I would like to thank Dale Thomas for help in analyzing the data generated from this project.

References Boulding, K. E., 1962: Conflict and Defense (New York: Harper & Row). Bueno de Mesquita, B., 1981: The War Trap (New Haven: Yale University Press). Choucri, N.; North, R. C., 1975: Nations in Conflict (San Francisco: W.H. Freeman). Cioffi-Revilla, C.; Starr, H., 1995: “Opportunity, Willingness, and Political Uncertainty: Theoretical Foundations of Politics”, in: Journal of Theoretical Politics, 7(January): 447–476.

References

181

Cowen, D. J., 1990: “GIS Versus CAD Versus DBMS: What Are the Differences”, in: Peuquet, D. J.; Marble, D. F. (Eds.): Introductory Readings in Geographic Information Systems (London: Taylor and Francis): 52–61. Environmental Systems Research Institute, 1992: ARC/INFO: GIS Today and Tomorrow (New York: ESRI). Falah, G.; Newman, D., 1995: “The Spatial Manifestation of Threat: Israelis and Palestinians Seek a ‘Good’ Border”, in: Political Geography, 14: 689–706. Friedman, G.; Starr, H., 1997: Agency, Structure and International Politics: From Ontology to Empirical Inquiry (London: Routledge). Garnham, D., 1976: “Dyadic International War: 1816-1965: The Role of Power Parity and Geographic Proximity”, in: Western Political Quarterly, 29: 231–242. Giddens, A., 1984: The Constitution of Society (Berkeley: University of California Press). Glassner, M. J., 1992: Political Geography (New York: Wiley). Gleditsch, N. P.; Singer, J. D., 1975: “Distance and International War, 1816-1965”, in: Proceedings of the IPRA 5th Conference IPRA Studies in Peace Research (Oslo: International Peace Research Association). Gochman, C. S., 1992: “Interstate Metrics: Conceptualizing, Operationalizing and Measuring the Geographic Proximity of States Since the Congress of Vienna”, in: Ward, M. D. (Ed.): The New Geopolitics (Philadelphia: Gordon and Breach): 139–158. Goertz, G., 1994: Contexts of International Politics (Cambridge: Cambridge University Press). Goertz, G.; Diehl, P. F., 1992: Territorial Changes and International Conflict (London: Routledge). Goertz, G.; Diehl, P. F., 1993: “Enduring Rivalries: Theoretical Constructs and Empirical Patterns”, in: International Studies Quarterly, 37: 147–171. Goertz, G.; Diehl, P. F., 1995: “Taking ‘Enduring’ Out of Enduring Rivalry: The Rivalry Approach to War and Peace”, in: International Interactions, 21: 291–308. Holsti, K. J., 1991: Peace and War: Armed Conflicts and International Order 1648-1989 (Cambridge: Cambridge University Press). Huth, P. K., 1996: Standing Your Ground: Territorial Disputes and International Conflict (Ann Arbor: University of Michigan Press). Lemke, D., 1995: “The Tyranny of Distance: Redefining Relevant Dyads”, in: International Interactions, 17: 113–126. Lemke, D., 1996: “Why So Much Peace? Interstate War in Africa”. Presented at the Annual Meeting of the American Political Science Association, San Francisco, CA. Marble, D. F., 1990: “Geographic Information Systems: An Overview”, in: Peuquet, D. J.; Marble, D. F. (Eds.): Introductory Readings in Geographic Information Systems (London: Taylor and Francis): 8–17. Midlarsky, M. I., 1970: “Mathematical Models of Instability and a Theory of Diffusion”, in: International Studies Quarterly, 14: 60–84. Midlarsky, M. I., 1975: On War (New York: Free Press). Most, B. A.; Starr, H., 1980: “Diffusion, Reinforcement, Geopolitics, and the Spread of War”, in: American Political Science Review, 74: 932–946. Most, B. A.; Starr, H., 1989: Inquiry, Logic and International Politics (Columbia, SC: University of South Carolina Press). Peuquet, D. J.; Marble, D. F., 1990: “ARC/INFO: An Example of a Contemporary Geographic Information System”, in: Marble, D. J.; Marble, D. F. (Eds.): Introductory Readings in Geographic Information Systems (London: Taylor and Francis): 90–99. Prescott, J. R. V., 1987: Political Frontiers and Boundaries (London: Allen and Unwin). Siverson, R. M.; Starr, H., 1991: The Diffusion of War (Ann Arbor: University of Michigan Press). Starr, H., 1978: “‘Opportunity’ and ‘Willingness’ as Ordering Concepts in the Study of War”, in: International Interactions, 4: 363. Starr, H., 1991a: “Joining Political and Geographic Perspectives: Geopolitics and International Relations”, in: International Interactions, 17: 1–9.

182

7 Opportunity, Willingness and Geographic Information …

Starr, H., 1991b: “Democratic Dominoes: Diffusion Approaches to the Spread of Democracy in the International System”, in: Journal of Conflict Resolution, 35: 356–381. Starr, H., 1999: “Democratic Dominoes Revisited: Democratic Diffusion, 1974-1996”. Presented at the Annual Meeting of the International Studies Association, Washington, DC. Starr, H.; Bain, W., 1995: “The Application of Geographic Information Systems (GIS) to International Studies”, in: International Studies Notes, 20: 1–8. Starr, H.; Most, B. A., 1976: “The Substance and Study of Borders in International Relations Research”, in: International Studies Quarterly, 20: 581–620. Vasquez, J., 1993: The War Puzzle (Cambridge: Cambridge University Press). Ward, M. D. (Ed.), 1992: The New Geopolitics (Philadelphia: Gordon and Breach). Wesley, J. P., 1962: “Frequency of Wars and Geographical Opportunity”, in: Journal of Conflict Resolution, 6: 387–389. Zipf, G. K., 1949: Human Behavior and the Principle of Least Effort (Cambridge, MA: Addison-Wesley).

Chapter 8

Democracy and Integration: Why Democracies Don’t Fight Each Other

While a large and growing literature has emerged which investigates the impact of the expansion of democracy on foreign policy and international politics, much of it has been characterized by insufficient attention to theoretical and conceptual clarity.1 To address such problems, this article is an exercise in concept clarification.2 It stresses that the democratic peace is a subset of the processes and results of integration, that it fits within an integration framework, and that it works according to processes already identified by integration theory which permits the synthesis of a number of ‘contending’ explanations for the democratic peace. As part of this argument, the article also stresses the transparent nature of democracy, emphasizing the importance of the mutual perceptions of two democracies, that the other is clearly a democracy. Finally, this article reminds scholars that the focus of the democratic peace proposition is on war. While it is important to explore our theories in terms of their extension to other conflictual phenomena, we must be careful in specifying exactly what these relationships should look like.

8.1

What Is to Be Explained?

A large and growing literature has emerged that investigates the impact of the expansion of democracy on foreign policy and international politics. Much of this literature has been characterized by insufficient attention to theoretical and conceptual clarity and the logical development of its argument. This essay identifies some of these problems explicitly and suggests how to address them but will not present any new empirical data or data analysis. It is not intended to be a synoptic

This text was first published as: “Democracy and Integration: Why Democracies Don’t Fight Each Other”, Journal of Peace Research, Vol. 34, No. 2 (May, 1997), pp. 153–162. This text is republished with permission from Sage Publications, Ltd. 2 I acknowledge with gratitude the helpful comments provided by Bruce Russett, Charles Kegley, Steve Chan, Nils Petter Gleditsch and three anonymous referees on earlier drafts of this paper. I also wish to thank the Department of International Relations, the Australian National University, which provided me a home as a Visiting Fellow during the revision of this article. 1

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7_8

183

184

8 Democracy and Integration: Why Democracies Don’t Fight Each Other

literature review of theory and findings on democracy and war, conflict and cooperation, or even a partial attempt to order and synthesize the most recent work in the area (see Starr 1992a).3 This essay is an exercise in concept clarification and reconceptualization. By linking the democratic peace to integration theory, for example, my aim is to contribute to our understanding of democracy and the democratic peace. Over the past half-decade there has been a renewed interest in these topics. Almost every issue of the major journals in the field of international relations contains at least one article devoted to the empirical analysis of the democratic peace, a theoretical or analytic discussion of why the democratic peace has occurred, or a critique of the democratic peace literature. Such critiques will question either the existence of the democratic peace, whether or not it truly challenges realist models of world politics, and/or whether or not it is trivial in terms of prevailing theories of international politics. However, much of this writing on the democratic peace – whether devoted to analysis, proof, or critique – loses track of exactly what is under discussion or what is to be explained. That is, the author(s) often forget that the democratic peace proposition (or hypothesis, or law, or whatever) is a statement that claims the following: there is a virtual absence of war among dyads of democratic polities. Thus, the democratic peace proposition (to be noted as DPprop) is about the absence of war. The DPprop is meant to explain the lack of war between democracies. It is about a set of conditions that explains the variance in a specific dependent variable – war.4 This comment does not mean, however, that we must look at war between pairs of democracies and then stop. Much of the debate, as will be discussed below, is about the nature of generalizability of the theories developed to explain the DPprop. One way to evaluate theories and to compare their relative explanatory power is to see what else they can do for us, and how far their logic can take us in understanding other phenomena; hear what other behaviors, outcomes or events are implied by the theories that explain the DPprop. If we have a theory that we think works, then what else should we expect to see? As suggested in Most/Starr (1989) it would be more useful not to ask which theory is ‘correct’ or works ‘best’, but which explanation works best under what conditions. Each theory or model should be looked at as a ‘story’ which not only produces some particular outcome (peace in

3

Note also that a number of points made here have been raised elsewhere in the literature on the democratic peace. Since I have not attempted a full survey of the literature, I apologize in advance to those not cited in this paper. 4 This somewhat tendentious statement of the central concern of the DPprop has been made consciously so. The DPprop is specifically about the escalation of conflict to the level of large-scale organized violence between state actors. Given that war has been perhaps the single most central concern to students of international relations across history, uncovering one factor, variable, or set of conditions that can be considered a sufficient condition for the absence of war (or almost complete absence of war) is a stunning achievement.

8.1 What Is to Be Explained?

185

democratic dyads) but under certain conditions other outcomes as well.5 Therefore, if some explanation is set forward for the DPprop, what else should we expect to follow from that explanation or model? And, does empirical research support such expectations? It is here, regarding Olson’s (1982, p. 13) ‘consilience’, that we can move beyond simply ‘war’.6 War is not simply conflict. It is not the same as escalation, minor incidents, interventions, or even covert operations (all of which have been investigated as ‘tests’ of the DPprop). War is sustained violent conflict fought by organized armed forces which are directed by a governmental authority (Most/Starr 1989, Chap. 4, see also Bremer/Cusack 1993). While war can be subsumed under more general forms of social conflict,7 and while war may also be seen as the ultimate stage of a conflictual escalatory process, war is simply not the same phenomenon as either social conflict or escalation. While the exact levels of casualties and/or duration and/or capabilities committed needed to identify a ‘war’ can be debated, war is more than a militarized incident, a minor incursion, or the like. War’s defining feature is a large-scale, organized violence that either imposes heavy costs on the participants, or creates the potential for the participants to suffer heavy costs. War is considered the ultima ratio by realists; that is, the ultimate form of self-help to which states may resort in the anarchic Westphalian system. Indeed, throughout much of the history of this system one of the defining elements of the sovereign state was its right under international law to wage war. As distinct from the post-Second World War legal regime that has made aggressive war illegal (and which has placed strong constraints on any use of force beyond that of self-defence), war had simply been a special legal condition that could exist between states. Thus war was, and was perceived as, an integral component of the anarchic Westphalian state system. Yet, the DPprop makes a simple empirical claim – that war does not occur in democratic dyads (or that it has occurred in only a very few, extraordinarily rare, cases). To realist critics this appears to be a radical claim. If, however, the DPprop is placed within the context of the peace created by processes of integration, the position may not appear to be so ‘radical’. After all, both the theory and findings reported in the integration literature present exactly the same type of challenge to realism – identifying state behavior which, according to realists, was not supposed to occur in the international system.8 It is obvious that integration theories/studies

5 Such an approach to theory, of course, follows the Lave/March (1975, pp. 19–20) procedure for the development of disciplined speculation and models in the social sciences. 6 Olson (1982, p. 13) asserts that ‘If a theory explains facts of quite diverse kinds it has what William Whewell, a nineteeth-century writer on scientific method called ‘consilience’.’ 7 This is one of the main themes of the long term Two-Level Security Management Project I have developed to investigate various commonalities across different forms of social conflict, see Starr (1994). 8 For analysis and interpretation of Deutschian integration theory (e.g. Deutsch et al. 1957) see Puchala (1981) and Lijphart (1981). For a more general discussion of this argument, see Starr (1995).

186

8 Democracy and Integration: Why Democracies Don’t Fight Each Other

are not under attack by realist critics in a manner similar to the current response to the democratic peace literature (see Spiro 1994; Layne 1994). Integration in world politics cannot be ignored in the commentary and critique of the democratic peace. A key theme to be developed below, is that the phenomenon called the DPprop is a subset of the processes and results of integration (especially with reference to the Deutschian model of pluralistic security communities).

8.2

The Confusion

The DPprop is about war – the empirical fact that democracies never, or almost never, go to war against one another. A variety of statistical and case studies (e.g. Doyle 1986; Ray 1993, 1995; Russett 1993) have pretty well proved the DPprop: that it is very difficult, if not impossible, to find war (clearcut, large scale organized, sustained, violent conflict) between two democratic (clearcut, readily recognizable as democracy) states. The critiques of this work are not critiques of the DPprop (as a correlation). The enterprise engaging the attention of scholars has been the problem of developing theories which explains why this should be the case. Thus, the preponderance of the extant critiques are directed at the theories which have been developed to explain this behavioral phenomenon. To be fair both proponents and now opponents of the DPprop are engaging in some version of the Lave and March process noted above. They have observed some phenomenon in the world – lack of cases in which democracies go to war against democracies – and then have developed ‘stories’ that would account for that phenomenon. The next step in the Lave and March process is to ask what else would one expect to see or to follow if the story were valid. To do this in regard to the DPprop it is essential to have a full understanding of government and government-societal arrangements – before one could usefully produce a story that fits, and that generates additional propositions about what the world is like. This point was raised in Starr (1992a), which also reviewed some ‘stylized facts’ to see if the DPprop could simply be extended to whether or not democracies treat each other ‘better’ across a number of behavioral phenomena (such as the distribution of postwar payoffs). On the basis of a small sample of research results the answer was ‘no’. However, it is important to understand that ‘treating each other better’ is different from not going to war. Many of the discussions of the ‘limitations’ of the DPprop investigated such international behavior as interventions, militarized disputes, escalations, etc., and similarly concluded that democracies may not treat each other ‘better’. To follow the argument here, these exercises do not disprove or weaken the DPprop. What they do say is that some specific theory of why the DPprop occurs, when extended at least in part through its own internal logic, is not adequate by itself to explain related behaviors. That is, for a more complete explanation which is also consistent with extensions, the theory/model must be more clearly or fully specified. It would be valuable in assessing competing explanations to see which explanations of the

8.2 The Confusion

187

DPprop could be extended to other behavior, to assess comparative power as well as the limits of such explanations. But, if theories/models fail in these other areas, or do not work as expected, it does not necessarily cast doubt on the DPprop itself, only some single, specific model which purports to explain how and why the DPprop works.9

8.3

Integration: Security Communities and the Democratic Peace

Such observations about specifying and evaluating theory may be illustrated by returning to theories of integration. The end product of the process of integration according to Deutschian theory is the ‘security community’: A security community is a group of people which has become ‘integrated’. By integration we mean the attainment, within a territory, of a ‘sense of community’ and of institutions and practices strong enough and widespread enough to assure … dependable expectations of ‘peaceful change’ among its population. By sense of community we mean a belief … that common social problems must and can be resolved by processes of peaceful change. (Deutsch et al. 1957, p. 5)

As with the DPprop, the Deutschian definition of integration focuses on peace, and the conditions for peace; but it does more. A security community involves not only the absence of war, but more importantly, the absence of the military option in the interactions of the states within the security community.10 Additionally, Deutsch allows for the existence of pluarlistic security communities – where such conditions can hold even among a set of independent, nonamalgamated, states. It should be clear then, that the Deutschian pluralistic security community is an outcome which is broader than, but overarches, the democratic peace phenomenon.11 Thus, an investigation of integration may provide some clues in the explanation of the DPprop.

It is always possible that following the logic of ‘nice laws’ (Most/Starr 1989; Cioffi-Revilla/Starr 1995) explanations of the DPprop are limited to the condition of war and peace. Yet, the search for extensions is part of the process of looking for general explanations, of asking ‘of what is this an instance’ (Rosenau 1980). While it is possible that explanations of the DPprop might not be very good at explaining other behavior, the search for extensions is an important part of the process of looking for such broader phenomena. 10 For a particularly good description of this phenomenon in action, see Archer’s (1996) discussion of the ‘Nordic Zone of Peace’ (a Scandinavian security community). 11 With such prerequisites for the formation of security communities as compatibility of values, the extension of political elites and multiple forms of transactions and communication (Deutsch et al. 1957) it is not surprising that the only historically identified security communities—either amalgamated or pluralistic – are composed of democracies. 9

188

8 Democracy and Integration: Why Democracies Don’t Fight Each Other

When we look closely at the components of the Deutschian socialcommunication model of the integration process as well as the neo-functional process model of Ernst Haas, we find all the primary components of the two central theories used by scholars to explain the DPprop’. (1) the structural constraints model involves the constraints of organizations and formal laws or constitutions; (2) the democratic culture argument involves the presence of community, responsiveness, shared values, and norms. Thus key components of the two basic explanations of how the DPprop works are found in the two basic theories of integration. Additionally, the theories of integration stress the role of learning in the development of norms of cooperation and a sense of community, they stress the need for mutual benefits and the positive impact of the interdependence on the management of interdependent relations. Again, these are key components of theories attempting to explain the DPprop. From an integration perspective, what does it mean when one uncovers evidence that democracies might intervene in other democracies, or escalate a conflict with another democracy to the point where the military option is considered? From this perspective, all it means is that there has been, to that point in time, incomplete integration in terms of ‘dependable expectations of peaceful change’. That is, the democracies involved exist within a not yet fully formed, or imperfectly formed, pluralistic security community. Here, under certain dispute conditions, the military option may indeed be raised – but still does not escalate to actual war. Does such a situation invalidate theories of integration? Is there not a difference between claiming that integration does not exist or is a failure, and merely indicating that the processes involved have not yet produced the final end-product of integration – the pluralistic security community as Deutsch envisioned it? The behaviors that are (wrongly) claimed to invalidate or disprove the DPprop, would not be used to argue against the reality and explanations of integration; they may simply indicate that the processes creating integration had not finished. If this is the case, then we should look at those conditions which are specified as necessary and/or sufficient for integration. The ‘confusions’ highlighted in this section are about the nature of the theories scholars have used in attempting to explain the DPprop. I have suggested that this can, in part, be clarified through the use of Deutschian integration theory. Before returning to further correspondence in conditions and processes between pluralistic security communities and the democratic peace, it is necessary to discuss one of the basic features of democracy that is central to any theory attempting to explain the DPprop – the notion of transparency.

8.4 What Does Democracy Mean?

8.4 8.4.1

189

What Does Democracy Mean? Democracy and Transparency

It was argued previously that analysts need to have ‘a full understanding of democracy – its nature, the processes involved, what makes it different from other forms of government and government-societal arrangements’. One possible factor which distinguishes democracy from other governmental forms is its ‘transparency’. Democractic transparency – the openness of its political processes and the vast amount of economic, political and social information which is public and generally available – is a prerequisite for democracy as conceptualized in terms of the contestation for political leadership, regardless of the specific definition used (see Dahl 1971, 1989). Such a conceptualization sees democracy as providing an environment within which oppositions can effectively challenge incumbent governmental office holders for power in a legal, legitimate manner through prescribed procedures. In order to do so, the range of political and civil liberties commonly understood as those embodied in the American Bill of Rights, must obtain – freedom of speech and the press, freedom of assembly, freedom from a range of techniques of repression available to a government. It is only through transparency that a society can monitor and know of abuses of political and civil liberties. It is only through transparency that a government would fear the repercussions of such abuses. Thus, only transparency can provide the safe environment for effective governmental opposition that is the core of democracy as conceptualized in terms of the contestation for political leadership. The concept of transparency is also a useful point of departure for investigating how the nature of democracy could lead us to the DPprop. Why should transparency be important? One way to approach this question is to look at the arguments presented by Bueno de Mesquita/Lalman (1992) regarding the policy choices of ‘hawks’ and ‘doves’ and how these two types of states are related to democratic and non-democratic states.12 Bueno de Mesquita/Lalman (1992) develop the extensive form ‘international interaction game’ to investigate the behavior of ‘instrumentally rational’ foreign policy decision makers. They use this game to analyze how the sequence of actions taken by decision makers may lead to war or one of several forms of peaceful resolutions. What is key to us here is that they present the central problem facing decision makers as that of separation’, how to distinguish what type of opponent one is facing in terms of its preferences and preference ordering. Under what conditions will the opponent prefer the status quo?, prefer negotiations over other alternatives?, prefer capitulation to war?, prefer war it initiates over war initiated by the other party? They analyze the outcomes that will be produced with states that 12

Note that much of the material to follow is drawn from Starr (1992b) which should be consulted for a more complete treatment.

190

8 Democracy and Integration: Why Democracies Don’t Fight Each Other

might be characterized as ‘doves’ (states with preference orderings that make them generally averse to the use of force, depending of course on the nature of the opponent they face) as they interact with other doves or non-doves. They present a proof demonstrating that if both states are doves, and both know that the other is a dove, then war outcomes are impossible. As they note at several points, a crucial assumption is that there is ‘common knowledge’ by each side of the other’s dovishness.13 Bueno de Mesquita/Lalman (1992, Chap. 5) apply those results to the puzzle of why democracies rarely fight one another. In the real world, they argue, decision makers can never be sure what type of state they are facing – doves or non-doves. How can both sides in the real world attain the common knowledge regarding dove/ non-dove that is simply assumed in the game model? While decision makers cannot know who is a dove/non-dove, Bueno de Mesquita and Lalman argue that there is fairly common knowledge whether or not the opponent is a liberal democracy. Basing their logic on a combination of the theories used to explain the DPprop, if a certain country is a liberal democracy then decision makers know that the leaders of the state will be under greater constraints not to use force, and will bear heavier costs than the leaders of non-democracies if they do. All of this is known because of the various elements of transparency – the free movement of information in liberal democracies, the existence of opposition groups and knowledge of internal politics, institutions and debates.14 The transparency of democracy means that outside observers can see into such states, scrutinize the activity that occurs within the society and political system and recognize that the political behaviors conform to some broadly accepted notion of democracy and are robust enough to cross some threshold in order to be called democracy. Such transparency is inherent in truly open societies. Transparency that reveals the democratic nature of a polity is crucial for Bueno de Mesquita and Lalman’s use of the democracy as an indicator of dovishness. The Bueno de Mesquita and Lalman formulation only works when each party can be clearly recognized as a democracy by the other party. The mutual recognition of democracy (which equates to dovishness) is crucial to the use of the international interaction game to explain the DPprop. It is also central to how most other theories of 13

Common knowledge means there is no hidden information. Common knowledge has been defined as: any information that each player knows and the fact that each player has such knowledge is known to every other player. This assumption is crucial, because Beuno de Mesquita and Lalman also show that under imperfect information there exist conditions whereby doves could fight other doves. Fearon (1994) makes a somewhat similar argument when he asserts that democracies are able to signal their intentions—clearly—to one another. 14 I have summarized their argument as follows: (1) two states who are doves, and know that each of the pair are doves will not go to war; (2) this requires common knowledge, which cannot be assumed in the real world; (3) various aspects of liberal democracies can be seen as making them averse to the use of force, by the higher costs (constraints) imposed on leaders; (4) most often, the indications that a state is a liberal democracy are known and can be used as prior information by decision makers in helping them separate opponents into types (doves and non-doves); (5) the higher the belief that a state is dove-like, the lower the probability that a dove will use force against it.

8.4 What Does Democracy Mean?

191

the DPprop ultimately work. Transparency (including the degree or level of transparency) is thus crucial to how we measure democracy and which cases we look at to test our theories. Transparency means that leaders and populations of other states can see that a country provides for the political and civil liberties which permit the regularized and legal contestation for political power. In democratic dyads this means both sides can see into each other. On one simple level, this makes war between democracies much more difficult than war between a democracy and an authoritarian regime. One mechanism by which the leaders of states create a willingness for societal masses (and elites) to support and prosecute a war, is the creation of an enemy image which involves the dehumanization of the opponent. A number of studies indicate that this enemy image is used to portray the opponent as evil and/or non-human in some way – thus justifying the use of violence against such an enemy and warranting the costs of war.15 Images of the ‘Hun’ on British posters during World War I, or images of Japanese in American films during World War II exemplify this phenomenon (e.g. see Dower 1986). With two democracies, and the amount of information flowing in and out of each, it is almost impossible to create such an image. If both countries are democracies, it is likely that they share a broad range of transactions, and that the levels of transactions are high enough that each society knows a great deal about the other; such a multiplicity of transactions and communications is also strongly consistent with Deutschian integration theory (Dougherty/Pfalzgraff 1990, p. 436). Such transparency means that each party has too much information about the other to create convincing enemy images, for either elites or masses. It means that one important component often used to mobilize societal support, and willingness to bear the costs of war, is unavailable to democratic leaders when facing one another. Not only does transparency mean that leaders cannot use the evil/non-human image of the enemy, but the reason for that unavailability lies in seeing that the opponent respects the political and civil liberties within its domestic political arena – including the free press which provides much of this information.16

As one example, Kelman/Hamilton (1989, p. 163) notes that ‘When victims are dehumanized … the moral restraints against killing or harming them become less effective. Groups of people who are systematically demonized, assigned to inferior or dangerous categories, and identified by derogatory labels are readily excluded from the bonds of human empathy and the protection of moral and legal percepts’. 16 And, as Van Belle (1995, p. 2) notes, ‘when two democracies come into conflict, the domestic news media on both sides share common norms of reporting responsibility, accuracy and accountability. They accept each other as legitimate sources of information and reports travel relatively freely between the news gathering institutions in both states’. 15

192

8.4.2

8 Democracy and Integration: Why Democracies Don’t Fight Each Other

Lesser Forms of Conflict

The point immediately above follows from a Kantian argument as expanded by Russett (1993) – that the political processes within democracies may be used to structure relations between democratic states. Both this argument and transparency may be useful in looking at other forms of conflict between pairs of democracies. Many discussions of the DPprop are based on findings which allegedly indicate that democracies have been involved in interventions against other democracies (e.g. Kegley/Hermann 1995a). Some have seen as even more damning, those studies which indicate the use by some democracies of covert intelligence/intervention operations against other democracies (e.g. Forsythe 1992; Stedman 1993). A careful analysis of covert operations within democracies, however, provides evidence in support of the DPprop, not a critique. Recall the findings of the experimental work being done by Alex Mintz and associates, which tests a ‘political incentive’ explanation for the DPprop (e.g. Geva et al. 1993). This work argues that the leaders of democracies do not pursue war against other democracies because they have no political incentive to do so. The results of their experiments indicate that the use of force by one democracy against another is perceived by the public as incompetent leadership – incompetent in part because the use of force is not seen as being worth the costs or risks, especially in interactions with states where peaceful conflict resolution is expected (see also Fearon 1994).17 The results of Mintz’s experiments are also quite consistent with the structure and preference orderings of states in the international interaction game of Bueno de Mesquita/Lalman (1992, Chap. 4). Bueno de Mesquita and Lalman argues that if two states are doves, and the leaders of each state knows that the other is a dove, then based on expected utility calculations, ‘negotiation or the status quo are the only possible equilibrium outcomes of the game’. With the work of Mintz and Bueno de Mesquita and Lalman in mind, it is clear that decision makers leading democracies could expect different levels of domestic opposition and support for the use of covert force against different types of targets. While American leaders generally have not received and would not expect to receive, widespread and strident criticism when revealing covert operations against countries such as Cuba or Iraq or Libya, they would expect such a reaction if the targets were democratic allies (or even marginal democracies in newly liberalizing states). Indeed, the fact that covert operations against democracies would be roundly denounced across the political spectrum, would lead democratic leaders to hide

17

An interesting example may be found in Archer (1996) where he discusses the 1905 separation of Norway from Sweden. After the referendum in Norway showing overwhelming support for the Norwegian government, the Swedish Social Democrat leader Branting noted in speeches that any attempt by Sweden to use force in Norway would cause a general strike in Sweden. Indeed, King Oscar told the Swedish parliament, ‘the Union (between Norway and Sweden) is not anything, if it is to be upheld by force’.

8.4 What Does Democracy Mean?

193

such activities. In the terms of Putnam’s (1988) two-level games, covert action against another democracy would not be included in the domestic (Level II) win-set: such a policy would lose because it would be perceived as illegitimate. Thus such activities are covert because they would generate high levels of opposition, and leaders wish to keep them out of the open democratic political process. In large part policy makers attempt to keep covert operations secret because the public does not support such activities against democracies (see also Russett 1993, pp. 12–24). The use of covert activities against other democracies simply confirms the conclusions of Mintz and Bueno de Mesquita and Lalman (and by extension some combination of both structural and cultural explanations for the DPprop) – that there is no political benefit attending such activities. It also follows that policy makers keep operations low key – at low levels of violence and levels of forces employed – because higher levels (as they escalate towards war) will bring major societal and political opposition. It should be clear that the forms, needs and traditions of democratic transparency are also why covert activities within democracies are usually difficult to pursue and why they will raise opposition when uncovered: such behavior is inherently anti-democratic. Thus, there will be opposition to almost any covert operation.18 It will become especially vociferous when employed against another democracy. As Mintz indicates, the public will expect that there are non-violent ways to manage disputes or conflicts with other democracies. And, in fact, research has indicated that democracies are more likely to utilize international organizations and third party mechanisms to resolve their conflicts with one another (see for example, Dixon 1993, 1994; Raymond 1994).

8.5

Conclusion: Democracy and Integration

Most discussions of the DPprop present two basic explanations for the behavior of democracies – democratic political culture and political structure/constraints. Many discussions have also concluded either that neither of these two explanations are adequate, or that they have to be combined in some way (e.g. Farber/Gowa 1995; Starr 1992b). If one looks at the DPprop through the lenses of integration theory then these two broad explanations should be used in combination. For the fullest conception of the integration process both Deutshcian social communication and Haasian neo-functional models must be considered. In combining these two models

18

Despite these normative and structural constraints and clear evidence that the cost-benefit calculation will usually work against either covert operations or the overt use of force against other democracies, we know that some elected leaders of democracies do attempt such behavior. The question then becomes what types of individuals – personalities, decision making styles, risk orientations – will push the envelope of public acceptance of violent conflict with other democracies? These questions are central to the ongoing project of Kegley/Hermann (e.g. 1995b); see also Hermann/Kegley (1995).

194

8 Democracy and Integration: Why Democracies Don’t Fight Each Other

one includes an approach that begins with, and builds upon, the growth of common norms, expectations and identities with an approach that begins with, and builds upon, the development of institutions that constrain behavior both through learning and spillover (and increasing ties of interdependence) and formal agreements on common interests, responsibilities and obligations. Neither culture nor structure are individually sufficient to explain democracy or the behavior of democracies, because they interact with one another and each contributes to a larger syndrome that makes up democracy. Each is an important component to the legitimacy of the political system - a legitimacy which rests upon the same conditions that undergird Deutschian security communities. Also drawing upon the work of Deutsch, look at Jackman’s (1993, p. 98) definition of legitimacy: A regime is thus legitimate to the extent that it can induce a measure of compliance from most people without resort to the use of physical force. The compliance need not be total, but it does need to be extensive.

As with security communities, in a legitimate system there are expectations of compliance and expectations of the non-use of force. With such legitimacy, democracies can look at other democracies and see systems where norms against the use of force are important, and where governments are constrained by the societies they represent (through both norms and institutions). These conditions obtain because, as in security communities, there is a ‘compatibility of political values associated with common political institutions ... and links of social communication that reflect a sense of community and shared identity (‘we-feeling’) among the members, including mutual sympathy and loyalties’ (Kacowicz 1995, p. 2). The legitimacy of democracies also derives from the expectation of joint economic reward, one of the Deutschian preconditions for integration (Deutsch et al. 1957). The legitimacy of integrated societies also includes the expectation that there will be a certain equity in the distribution of economic benefits, and thus the narrowing of gaps between the richest and poorest in society (see Deutsch 1997). The transparency of democracies means that people inside and outside of the society can see how the political and economic systems work, that they can participate in those systems and additionally expect some ‘fair’ mutual participation in the payoffs. Transparency involves, in part, the open movement of large amounts of information. It would be impossible for any form of free market system to work without very large amounts of information available to all participants, and potential participants, about supply, demand, prices, performance etc. Of necessity, the information required by free market moves unobstructedly in and out of society; it also moves freely within the society across its political and social systems, as well as the economic system. The transparency required in a free market system thus reinforces the transparency inherent in a democratic political system. This synergy has domestic (and international) consequences in that individuals and groups within a society can understand in what ways the economic system distributes its rewards/wealth/payoffs. When an economic (and political) system can provide positive outcomes for most participants – again, a condition of Deutschian

8.5 Conclusion: Democracy and Integration

195

integration – this too promotes societal legitimacy and community through the expectation of mutual rewards. This synergy of political and economic transparency also creates conditions whereby democracies can draw upon ‘the shadow of the past’ – and can tell that other polities are democracies. It permits and encourages transactions across state boundaries, promoting integration processes between democratic societies and states. If we are concerned with Kantian arguments of democratic norms, or the notion of common culture and values – then the values associated with legitimacy, and how they affect integration are important. If members of one democratic society with effective levels of legitimacy see another society with similar characteristics, then it would make no sense to fight a war. War would make no sense (returning to Mintz’s arguments) either as a mechanism to settle conflicts or as a mechanism to allocate values authoritatively (politics) or scarce resources (economics).19 In an attempt to move the theory and analysis of the DPprop forward, I have tried in this essay to clarify a number of issues found in the writing on the democratic peace. It is important to understand that the focus of the DPprop is on war, and that while it is important to explore the theories behind the DPprop in terms of their extension to other political phenomena, we must be careful in specifying exactly what these relationships should look like. Following earlier admonishments, I have attempted to clarify the concept of democracy and its relationships to theories of the DPprop by stressing the transparent nature of democracy. This aspect of democracy is of particular importance in regard to the need for mutual perceptions within a pair of democracies that the other is clearly a democracy. This essay has also reconceptualized the democratic peace by indicating the various linkages the study of DPprop has to integration theory. Firstly, this is a useful way to demonstrate that although the DPprop presents a strong challenge to realist (and especially neorealist) models of international politics, it is not unique or radical in its challenge. Secondly, this perspective also allows us to understand that the DPprop fits within an integration framework, that it works according to processes already identified by integration theory and that integration theory would permit us to synthesize rather easily a number of the ‘contending’ explanations of the DPprop. Thirdly, regarding the question of how to extend explanations of the DPprop and in which direction should analysts move, the linkages among integrations theory, integration processes, and disintegration with the DPprop promise to be fertile areas. Russett, in discussions not of ‘the democratic peace’ but of a broader ‘Kantian peace’, has argued that the DPprop is only one leg of a triad that includes economic interdependence as well as international law and organization. All three of the Kantian dimensions can be explored under a broader integration perspective. In a communication with the author, Steve Chan argues that it is only with the ‘shadow of the past’, that is, long and positive experience with each other, ‘that leaders are willing to accept each other as doves, thereby running the risk of type-1 error, mistaking an aggressive power as a status-quo one’. 19

196

8 Democracy and Integration: Why Democracies Don’t Fight Each Other

References Archer, Clive, 1996: “The Nordic Area as a Zone of Peace”, in: Journal of Peace Research, 33,4: 451–467. Bremer, Stuart A.; Cusack, Thomas R. (Eds.), 1993: “Special Issue: Advancing the Scientific Study of War”, in: International Interactions, 19,1–2. Bueno de Mesquita, Bruce; Lalman, David, 1992: War and Reason (New Haven, CT: Yale University Press). Cioffi-Revilla, Claudio; Starr, Harvey, 1995: “Opportunity, Willingness, and Political Uncertainty: Theoretical Foundations of Politics”, in: Journal of Theoretical Politics, 7,4: 447–476. Dahl, Robert A., 1971: Polyarchy (New Haven, CT: Yale University Press). Dahl, Robert A., 1989: Democracy and Its Critics (New Haven, CT: Yale University Press). Deutsch, Karl W., 1977: “National Integration: Some Concepts and Research Approaches”, in: Jerusalem Journal of International Relations, 2,1: 1–29. Deutsch, Karl W., et al., 1957: Political Community and the North Atlantic Area (Princeton, NJ: Princeton University Press). Dixon, William J., 1993: “Democracy and the Management of International Conflict”, in: Journal of Conflict Resolution, 37,1: 42–68. Dixon, William J., 1994: “Democracy and the Peaceful Settlement of International Conflict”, in: American Political Science Review, 88,1: 14–32. Dougherty, James E.; Pfaltzgraff, Robert L., 1990: Contending Theories of International Relations, 3rd ed. (New York: Harper & Row). Dower, John W., 1986: War Without Mercy: Race and Power in the Pacific War (New York: Pantheon). Doyle, Michael W., 1986: “Liberalism and World Politics”, in: American Political Science Review, 80,4: 1151–1169. Farber, Henry S.; Gowa, Joanne, 1995: “Politics and Peace”, in: International Security, 20,2: 123–146. Fearon, James D., 1994: “Domestic Political Audiences and the Escalation of International Disputes”, in: American Political Science Review, 88,3: 577–592. Forsythe, David P., 1992: “Democracy, War, and Covert Action”, in: Journal of Peace Research, 29,4: 385–395. Geva, Nehemia; De Rouen, Karl; Mintz, Alex, 1993: “The Political Incentive Explanation of “Democratic Peace”: Evidence from Experimental Research”, in: International Interactions, 18,3: 215–229. Hermann, Margaret G.; Kegley, Charles W., Jr., 1995: “Rethinking Democracy and International Peace: Perspectives from Political Psychology”, in: International Studies Quarterly, 39,4: 511–533. Jackman, Robert W., 1993: Power Without Force (Ann Arbor, MI: University of Michigan Press). Kacowicz, Arie M., 1995: “Pluralistic Security Communities in the Third World? The Intriguing Cases of South America and West Africa”, Presented at the Annual Meeting of the International Studies Association. Kegley, Charles W., Jr.; Hermann, Margaret G., 1995a: “Military Intervention and the Democratic Peace”, in: International Interactions, 21,1: 1–21. Kegley, Charles W., Jr.; Hermann, Margaret G., 1995b: “The Political Psychology of Peace Through Democratization”, in: Cooperation and Conflict, 30,1: 5–30. Kelman, Herbert C.; Lee Hamilton, V., 1989: Crimes of Obedience (New Haven, CT: Yale University Press). Lave, Charles A.; March, James G., 1975: An Introduction to Models in the Social Sciences (New York: Harper & Row). Layne, Christopher, 1994: “Kant or Cant: The Myth of Democratic Peace”, in: International Security, 19,1(Summer): 5–49.

References

197

Lijphart, Arend, 1981: “Karl W. Deutsch and the New Paradigm in International Relations”, in: Merritt, Richard L.; Russett, Bruce M. (Eds.): From National Development to Global Community (London: Allen & Unwin): 233–251. Most, Benjamin A.; Starr, Harvey, 1989: Inquiry, Logic and International Politics (Columbia, SC: University of South Carolina). Olson, Mancur, 1982: The Rise and Decline of Nations (New Haven, CT: Yale University Press). Puchala, Donald, 1981: “Integration Theory and the Study of International Relations”, in: Merritt, Richard L.; Russett, Bruce M. (Eds.): From National Development to Global Community (London: Allen & Unwin): 145–164. Putnam, Robert, 1988: “Diplomacy and Domestic Politics: The Logic of Two-Level Games”, in: International Organization, 42,3(Winter): 427–462. Ray, James Lee, 1993: “Wars Between Democracies: Rare, or Nonexistent?” in: International Interactions, 18,3: 251–276. Ray, James Lee, 1995: Democracy and International Conflict (Columbia, SC: University of South Carolina Press). Raymond, Gregory A., 1994: ‘Democracy, Disputes and Third-Party Intermediaries”, in: Journal of Conflict Resolution, 38,1: 24–42. Rosenau, James N., 1980: “Thinking Theory Thoroughly”, in: Rosenau, James N. (Ed.): The Scientific Study of Foreign Policy, Rev. edn. (London: Frances Pinter): 19–31. Russett, Bruce, 1993: Grasping the Democratic Peace (Princeton, NJ: Princeton University Press). Spiro, David, 1994: “The Insignificance of the Liberal Peace”, in: International Security, 19,1 (Summer): 50–86. Starr, Harvey, 1992a: “Why Don’t Democracies Fight Each Other? Evaluating the Theory-Findings Feedback Loop”, in: Jerusalem Journal of International Relations, 14,1: 41–59. Starr, Harvey, 1992b: “Democracy and War: Choice, Learning, and Security Communities”, in: Journal of Peace Research, 29,2: 207–213. Starr, Harvey, 1994: “Revolution and War: Rethinking the Linkage Between Internal and External Conflict”, in: Political Research Quarterly, 47,2: 481–507. Starr, Harvey, 1995: “International Law and International Order”, in: Kegley, Charles W., Jr. (Ed.): Controversies in International Relations Theory: Realism and the Neoliberal Challenge (New York: St. Martin’s): 299–315. Stedman, Stephen John, 1993: “The New Interventionists”, in: Foreign Affairs, 72,1(America and the World 1992/93): 1–16. Van Belle, Douglas A., 1995: “Kant, Cronkite, and Conflict: Democracy and the Political Role of Independent Domestic News Media in International Conflict”, Presented at the Annual Meeting of the International Studies Association, forthcoming in the Journal of Peace Research.

Chapter 9

The Kissinger Years: Studying Individuals and Foreign Policy

Henry Kissinger’s memoirs, White House Years, constitute an important addition to the extant literature on the former Secretary of State.1 This article is concerned with the study of individuals and the ways which can be used to gain ‘access’ to foreign policy makers.2 Memoirs provide a potentially useful source of data and means of access to high-level decision makers.3 Kissinger’s memoirs are discussed and evaluated as a research tool within the context of previous biographical and psychological/cognitive studies of Kissinger. Three specific areas of comparison— Kissinger’s views of Nixon the individual and statesman, Kissinger’s approach to bureaucracy, and Kissinger’s policy for dealing with the Soviet Union—are used to illustrate the continuity of his belief system and the consistency between his memoirs and previous research on Kissinger.

This text was first published as: “The Kissinger Years Studying Individuals and Foreign Policy”, International Studies Quarterly, Vol. 24 No. 4, December 1980 465–496. The permission to republish this text was granted by ISA. 2 Author’s Note: Research reported here has been supported by grants from the Center for International Policy Studies, Indiana University (under Ford Grant 750-0514), and by a Summer Faculty Fellowship, Indiana University. 3 The present author has spent the last several years both studying and adding to that literature. This essay is an outgrowth of a long-standing interest in Henry Kissinger and a research project which has included secondary analyses of biography and psychobiography, as well as operational code and content analyses of Kissinger the scholar and policy maker. The present article draws on a number of papers (Starr 1976, 1979, 1980a, b) resulting from that research project, as well as a number of unpublished analyses. 1

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7_9

199

200

9.1

9 The Kissinger Years: Studying Individuals and Foreign Policy

Kissinger, His Memoirs, and the Kissinger Industry

A vast literature has been produced on Henry Kissinger, the man and policy maker. To this has been added one massive volume of memoirs by Kissinger himself (1979). These writings may be seen as relating to a complex of questions concerning the study of individuals in the foreign policy process and how scholars can gain ‘access’ to such individuals. This review will take a brief look at the question of access. It asks where Kissinger’s memoirs fit into the study of individuals, cognitive and psychological approaches to the analysis of foreign policy, and other scholarly writings about Kissinger. Do Kissinger’s memoirs contribute to systematic theory about foreign policy? One of the first questions that one must ask when looking at individual, idiosyncratic, or psychological approaches to the study of foreign policy is, Do individuals count? Can and do individuals make a difference? Is there, in Greenstein’s words (1969), “actor indispensability” (as opposed to “action indispensability”)? Henry Kissinger is clearly one of the major foreign policy phenomena of our time. His background, style of behavior, foreign policy positions, relationships with Presidents Nixon and Ford, and preeminence in American foreign policy for eight years have fascinated the man in the street, journalists and academics alike. Much of this fascination stems not only from the contrast between Kissinger and Nixon (or the rest of Nixon’s advisers), but from Kissinger’s ‘preeminence’ in Nixonian foreign policy. As suggested by the bureaucratic politics literature, Kissinger’s dominance derived from both his personal relations with the president and his position within the formal policy-making process. Many observers (such as Brandon 1973: 23; Eldridge 1976: 20) have pointed out the similarity in foreign policy views between Nixon and Kissinger. In his memoirs Kissinger indicates areas of broad agreement with Nixon regarding the structure of the foreign policy process and notes his approval of Nixon. Describing his November 25, 1968 meeting with Nixon at the Hotel Pierre, Kissinger says, “Nixon outlined some of his foreign policy views. I was struck by his perceptiveness and knowledge so at variance with my previous image of him” (1979: 11–12). Along with a similarity of views came the development of Kissinger’s central position within the foreign policy process. Kissinger, as National Security Adviser, controlled the apparatus of the National Security Council and its staff. He became both the conduit and screen for the great bulk of information about foreign policy and foreign policy alternatives that moved from the bureaucracy upward to the president. As National Security Adviser, Kissinger came to chair the five major interagency committees that supervised foreign policy: the Washington Special Actions Group (WSAG) dealing with crises, the Defense Programs Review Committee, the Viet Nam Special Studies Group, the Forty Committee dealing with covert intelligence operations, and the Verification Panel dealing with the SALT negotiations. The many and intense “bureaucratic politics” battles that occurred in the Nixon administration are vividly described by Kissinger time and time again in White House Years. In the period covered (up to 1973, before Kissinger assumed the

9.1 Kissinger, His Memoirs, and the Kissinger Industry

201

position of Secretary of State), the major battles were waged between Kissinger and the nominal Secretary of State, William Rogers. Great attention and resources were devoted by Nixon and Kissinger to end-running (‘backchanneling’) Rogers, the State Department, and the foreign service (see, for example, pp. 28–31). As noted, Kissinger won more than his share of bureaucratic victories not only from his position, but also because of presidential support. The memoirs often make it clear that both Kissinger and Nixon recognized the hierarchy of power and responsibility between the two men (pp. 40–41 or p. 717). Kissinger succinctly describes their working relationship (p. 805): By the end of 1970 I had worked with Nixon for nearly two years: we had talked at length almost every day; we had gone through all crises in closest cooperation. He tended more and more to delegate the tactical management of foreign policy to me. During the first year or so I would submit for Nixon’s approval an outline of what I proposed to say to Dobrynin or the North Vietnamese, for instance, before every meeting. He rarely changed it, though he rarely failed to add tough-sounding exhortations. By the end of 1970 Nixon no longer required these memoranda. He would approve the strategy, usually orally; he would almost never intervene in its day-to-day implementation.

It seems that, in the case of Henry Kissinger, individuals did count. Kissinger had the official position, the unofficial ‘clout,’ and the opportunity to shape and execute policy in an important era of American foreign policy. This individual impact was also popularly recognized. In 1972 Kissinger placed fourth in the Gallup Poll’s “Most Admired Man Index,” and in 1973 he was first. Never before had a secretary of state or any presidential adviser even been placed on the list. In May 1973, 78% of Americans were able to identify Kissinger, a number unmatched except for presidents, presidential candidates, and major sports and entertainment figures (Eldridge 1976: 31). Given that an individual could count, we can move on to the question, How can we study what differences specific individuals make? Over the years a number of observers have referred to the ‘mysterious’ Kissinger; a man who was inexplicable, unpredictable, full of surprises, or all of the above (see Dickson 1978: 17). Is Kissinger any more ‘mysterious’ than other individuals, or are there clues to why he behaved as he did? In his Harvard College senior thesis, The Meaning of History, Kissinger observed: “Everybody is a product of an age, a nation, and environment. But beyond that, he constitutes what is essentially unapproachable by analysis, the form of the form, the creative essence of history, the moral personality” (Kissinger 1951: 127). It is hoped that foreign policy makers are not “essentially unapproachable,” though the task of studying them is a difficult one.

9.2

The Problem of Access

Let us begin with Brody’s (1969: 116) famous question: “How can we give a Taylor Manifest Anxiety Scale to Khrushchev during the Hungarian revolt, a Semantic Differential to Chiang Kai-shek while Quemoy is being shelled, or simply

202

9 The Kissinger Years: Studying Individuals and Foreign Policy

interview Kennedy during the Cuban missile crisis?” Obviously, we cannot. The question and answer highlight the problems both of access (see, for example, Holsti et al. 1968) and the crucial need for ways to “get at” foreign policy decision makers while in office and afterward (and, perhaps, even before they enter office). In this instance we have Henry Kissinger, an example of the crucial individual in foreign policy during a dramatic era of world politics and dramatic American foreign policy moves—from the “Nixon Doctrine” through the opening to China, ‘detente,’ and “shuttle diplomacy.” What sources of data and techniques for acquiring and analyzing those data do we have? The study of decision making and decision makers presents the researcher with a number of problems. One of the central problems is access to decision makers—how we go about studying them while they are ‘embedded’ within the vast and overlapping organizational structure of the government of the modern state. Within the broad cognitive and psychological approach to the study of foreign policy there are a number of techniques which can be used to gain access to the belief systems, personalities, and decision-making processes of individuals, some of which specifically attempt to gain access by the study of how individuals use words and symbols: using secondary sources for traditional biographical analysis and more unconventional (and controversial) ‘psychohistorical’ approaches; operational code analysis; formal content analysis; and events-data analysis. A growing literature on cognitive and psychological approaches reviews how these approaches can be used, their strengths and weaknesses, promised payoffs and research costs.4 These approaches depend upon some basic assumptions which, if accepted, permit scholars to attack the problem of access. One set of assumptions is the view that decision makers are subject to the same psychological processes that affect all humans, and that they, like all humans, can act only in terms of their images of the world. As Holsti et al., among many, have pointed out, “the essential point is that the actor’s response will be shaped by his perception of the stimulus and not necessarily by the qualities objectively inherent in it” (1968: 129). One view found throughout the cognitive literature is that the decision maker’s belief system, personality, values, and so on are in some way related to decisions that are made and actions that are taken. Where do Kissinger’s memoirs fit into all of this? A decision maker’s recollection and reconstruction of his time in office is a potentially important source of data about his beliefs, values, attitudes, and the various ‘whys’ behind his decisions and behavior. Memoirs, as an additional source of data, provide one further means of access to the decision maker. The question of the actual utility of memoirs, however, is an open one. What can we gain from any specific set of memoirs? With what confidence can we use and apply the information contained within them (given the base assumption that they will be, to some extent, self-serving)? The

4

A short list of relevant research and review articles, and their accompanying bibliographies, would include Holsti (1975, 1976a, b), Holsti/George (1975), Shapiro/Bonham (1973), Brodin (1972), Falkowski (1979a, b), Sullivan (1976: Chap. 2), Axelrod (1976), and Jervis (1976).

9.2 The Problem of Access

203

remainder of this essay will be devoted to these questions. First we will review some of the sources of information, and the methodologies used to gather that information, that have been used to study Henry Kissinger. Then, several specific substantive topics covered in the memoirs will be contrasted to data, analysis, and conclusions from the earlier research to help appraise the utility of White House Years in the continuing study of Henry Kissinger and the Nixon-Kissinger era of American foreign policy.

9.3

Psychological Approaches to the Study of Kissinger

Biographical studies are a traditional means for attempting to understand the nature of public officials as individuals and their policies. The typical goal is to draw a ‘portrait’ of the individual through history, chronology, and descriptive detail. Again, typically, data sources involve interviews, material written by the subject (especially ‘private’ documents such as letters and ‘secret’ memos), secondary material about the subject and his or her times, chronologies of historical events including journalistic accounts, and sometimes interviews with the subject. Several biographies and analyses of Kissinger exist which do all of the above. They are useful in setting forth the public record of policy while Kissinger was in office and the facts about Kissinger’s life before he became a high-level decision maker. Such works attempt to describe more fully who Kissinger is—his background, family data, education, and ‘significant’ events in his life. For Henry Kissinger we have not only a straightforward (bland?) account of his life in the biography by Blumenfeld et al. (1974) based on over 400 interviews, but also a most flattering account by Kalb/Kalb (1974) and a most critical one by Landau (1974). Note that all three arrived in print the same year, after Kissinger’s clear public emergence as a major foreign policy figure (in 1971–1972) and after he assumed the title of Secretary of State in September 1973. Since then, most works on Kissinger have concentrated on his foreign policy style and decisions. Stoessinger’s (1976) treatment of Kissinger also contains some biographical material. The more traditional biographical approach, especially in regard to its picture of the pre-decision-making Kissinger, may also be supplemented by the psychobiographical or psychohistorical approach. In this approach, using approximately the same types of data sources, the analyst employs psychoanalytic techniques to draw additional inferences about the personality, psychological makeup, and style of the subject and to link these influences to the subject’s behavior. A lively debate over the utility of ‘armchair’ analysis from a distance began with the publication of the study by George and George of Woodrow Wilson (1956) and accompanies each of the new works of Mazlish, who has written about Nixon, Kissinger, and, most

204

9 The Kissinger Years: Studying Individuals and Foreign Policy

recently, Jimmy Carter.5 Mazlish’s work on Kissinger (1976) is the main source for the psychobiographical complement to the biographical approach; it attempts to ‘interpret’ Kissinger’s life, style, scholarly writings, and policies from the perspective of his personality and the early life experiences.6 A second source is Ward’s (1975) lengthy article (which opens with the observation that “it is indeed curious that political scientists have been so hostile to psychohistorians who attempt to incorporate personality into their political analysis” [1975: 287]). Biography and psychobiography are methods that can be used to study the preoffice lives of foreign policy decision makers. For some, such as Kissinger, there is an additional and important preoffice data base. Kissinger wrote about the world in his position as a faculty member of the Department of Government at Harvard. Thus, we have not only the broad outlines of Kissinger’s personal life, but a mine of scholarly writings about international relations and foreign policy by Kissinger before he entered public office. Several observers have surveyed and reviewed these writings in order to summarize Kissinger’s basic intellectual themes (such as Graubard 1973) or to trace the philosophical basis for his intellectual positions (Dickson 1978). Whether the academic writings reveal clues to a broader belief system or can be used as tools to predict behavior are issues not taken up in the standard discussions of Kissinger’s writings. There is, however, a form of content analysis which may be used to perform these tasks. Applications of the “operational code” approach have become more numerous and refined in recent years, mostly through the work of George and Holsti.7 The operational code was designed explicitly to study an individual’s belief system, through a set of ten questions concerning ‘philosophic’ and ‘instrumental’ beliefs. These ten questions have been elaborated into an extensive “coding manual” by Holsti (1977). The operational code has been utilized simply as a “guide to research,” as a set of independent variables, or at least as a way to ‘narrow’ the range of choices that an individual might use—by understanding the belief system with which the individual views the world. It is a methodology designed to gain access to decision makers through the content analysis of a broad set of writings. Following the lead set by Holsti in the study of John Foster Dulles (1962a, 1970), operational code analysis has been performed on Kissinger’s preoffice academic writings (Walker 5

See, for example, Weinstein et al. (1978) or Tucker (1977) for a reappraisal of the work by George and George on Wilson and commentary on the psychohistorical approach. For a preliminary response by the Georges to Weinstein et al. see George/George (1979). 6 Mazlish (1976: 8) states: “My approach throughout is interpretive. This fact must be underlined from the very beginning. By now there are a number of acceptable and sometimes fine studies and articles on Kissinger’s thought, policies, or actual negotiations. Often weakened by an uncritical adulatory or condemnatory bias, such works are nonetheless fundamental in helping to establish some of the ‘facts’ about Kissinger in his many guises. They rarely, however, center on Kissinger’s personality in the sense I undertake here.” 7 See, for example, George (1979) and Holsti (1977). Both of those works, as well as the editor’s introduction in Hermann (1977), provide lists of the operational code studies performed on American foreign policy decision makers and a number of foreign leaders.

9.3 Psychological Approaches to the Study of Kissinger

205

1975, 1977; Starr 1979, 1980a). The objective here is the delineation of the belief system that the decision maker held upon taking office. In Kissinger’s case, this method has been used to study preoffice writings in order to analyze, explain, compare, and predict behavior in office. Kissinger’s words can also be studied by more rigorous and quantitative content analysis techniques. Holsti (1962a, b) demonstrated that “evaluative assertion analysis” was a useful methodology for studying the belief system of a high-level decision maker when he used that technique to analyze John Foster Dulles’ perceptions of the Soviet Union. In a partial attempt to replicate Holsti’s study, this author has also used evaluative assertion analysis to study Kissinger’s images of the Soviet Union and the People’s Republic of China. By studying an individual’s “image of the enemy,” research is being focused on what a number of scholars feel is a “master belief” that plays a central role in the construction of an individual’s operational code (see, for example, George 1969, 1979). Thus, in addition to analyses of Kissinger’s preoffice words, there is also this author’s study of all of Kissinger’s public statements as a high-level decision maker in regard to his images of the Soviet Union and China. Studies of both Kissinger’s general belief system concerning international affairs and a specific aspect of that system—the image of the enemy—are thus available.8 The evaluative assertion analysis study is particularly relevant to the memoirs, which spend a great deal of time on relations with the Soviet Union and China. Much of Kissinger’s interest in ‘triangular’ diplomacy (1979: 763) is reflected in the author’s study where Kissinger’s perceptions of the USSR and PRC are analyzed in terms of the triadic patterns of behavior using events-data. In general, the content analysis provides broad empirical data against which to compare the specific statements Kissinger makes in the memoirs. Thus, we have works concerned with Kissinger’s early life, his academic career, and his period as a foreign policy decision maker. There have been only a few attempts to pull all of these strands together. In addition to the author’s work, Stoessinger (1976) comes closest to merging these approaches successfully. There are, as we are well aware, a number of articles and books which analyze American foreign policy during the Kissinger tenure, as well as specific policy issue areas or decisions (for example, Shawcross 1979). The Kissinger industry is indeed a large one. This brief review will not specifically deal with the policy-oriented works, as much of those are covered in Stoessinger, in events-data analyses, and in the chronologies found in biographies. Since most of these works do not make the linkages between perceptions, beliefs, and style on the one hand and his political behavior on the other, they will not be central to an essay concerned with ‘access’ and the assessment of one possible source of that access—Kissinger’s memoirs.

8

An early outline and discussion of the author’s research project can be found in Starr (1976). Most of the content analysis remains unpublished (but see Starr 1980b). Another content analysis of Kissinger is Eldridge (1975).

206

9.4

9 The Kissinger Years: Studying Individuals and Foreign Policy

Memoirs, Access, and Confidence: Some Specific Areas of Comparison

To the analyses of Kissinger’s preoffice and official words, we now add 1500 pages of Kissinger’s postoffice words. There are a number of ways by which one could approach and discuss Kissinger’s memoirs. As one reviewer of the memoirs has noted, some reviewers, overwhelmed with the amount of material, have fallen back on preconceptions and previous opinions of the Nixon-Kissinger foreign policy (Watt 1979: 59). Rather than reargue the ‘morality’ or ‘success’ of these policies, this essay will examine the utility of the memoirs as a research tool. Given the author’s own set of preconceptions—based on the types of research noted above— the idea was to gauge how well White House Years reflects Kissinger’s views of international relations, how well it helps us understand what he thought he was trying to do, and why. One way would be to see how closely the memoirs reflect Kissinger’s view of the world, a style based on deeply rooted personality traits, and previous views on such issues as the role and impact of bureaucracies. Recognizing that there may always be changes in beliefs (especially during years in high office), the search will be for consistency between the memoirs and previous research on Kissinger. Within limits, high consistency will provide greater confidence in the memoirs as a data source and as a means of access. It is hoped that this treatment of the memoirs will also demonstrate a process by which other memoirs could be appraised. ‘Consistency’ can be used to summarize several criteria by which an individual can be judged a good subject for operational code analysis as well as content analysis in general (Walker 1975: 7–8). The issue of Kissinger as a useful subject for operational code analysis has been addressed elsewhere (Starr 1980a: 8–11). First, on subjects such as European diplomatic history, the nature and philosophy of international relations in general, contemporary international problems (for example, nuclear weapons policy and strategy), alliance politics, and American foreign policy, Kissinger’s academic writings are ‘broad’ enough to reveal a belief system for operational code and other analyses. Second, judging by secondary sources, discussions with acquaintances, and the content analysis research noted earlier, these academic writings can be considered ‘honest’ in that they appear to reflect Kissinger’s ‘real’ beliefs. For instance, in discussing Kissinger’s senior honors thesis, The Meaning of History, Graubard (1973: 7) observes, “Much of what he learned at Harvard was incorporated into a thesis that pretended to deal with selected philosophies of history since the 18th century; it was in fact, a kind of personal testament” (italics added). Finally, Kissinger’s writings reflect an important ‘continuity,’ in that his philosophy and political style appear to be essentially the same for the decision maker and the academic. Such ‘continuity’ is an important criterion for using an operational code approach. One Kissinger observer has noted, for instance, that Kissinger “turned scholarship into projective biography” (Mazlish 1976: 151). That is, in his writings on international relations, Kissinger was stating quite clearly what he

9.4 Memoirs, Access, and Confidence: Some Specific …

207

would do as a policy maker under certain circumstances. As Walker and others have noted, a distinctive characteristic of Kissinger’s writing is the tendency to generalize and hypothesize lessons and conclusions from the analysis of historical cases. Through these generalizations Kissinger developed, articulated, and outlined his operational code and provided a guide to his future behavior. From this we can identify the continuity in Kissinger’s thought as reflected in his academic works. When asked, in 1976, about how Professor Kissinger of Harvard was appraised by Secretary of State Kissinger, he replied: “I have tried—with what success historians will have to judge—to have an overriding concept. It can be found in innumerable, maybe pedantic, speeches I have given over the years” (U.S. Department of State 1976a: 124–125). In 1975, when asked if he had changed his basic conception of international politics over the past twenty years, his reply was consistent (if somewhat facetious): “I think it is possible—at least I leave open the theoretical possibility—that I might have changed my mind on something in my life, but don’t press me too hard” (U.S. Department of State 1975: 25–29). Many examples of this continuity can be found in Kissinger’s academic works and subsequent statements as a foreign policy maker. One of the basic points about world order which Kissinger makes in his Ph.D. thesis, published as A World Restored: Castlereagh, Metternich and the Restoration of Peace, 1812–1822, is about the nature of peace: Whenever peace—conceived as the avoidance of war—has been the primary objective of a power or a group of powers, the international system has been at the mercy of the most ruthless member of the international community. Whenever the international order has acknowledged that certain principles could not be compromised even for the sake of peace, stability based on equilibrium of force was at least conceivable. [1957: 1]

These sentiments were echoed in 1974, at hearings before the Senate Foreign Relations Committee: “If peace is pursued to the exclusion of any other goal, other values will be compromised and perhaps lost” (U.S. Senate 1974). Other sections and passages from A World Restored have returned time and again in Kissinger’s statements as a decision maker. As Stoessinger and Walker have pointed out, Kissinger was virtually quoting from several passages in chapter 8 of his dissertation when he explained the Indochina peace agreement to a news conference on January 24, 1973: “It was always clear that a lasting peace could come about only if neither side sought to achieve everything it wanted; indeed that stability depended on the relative satisfaction and therefore the relative dissatisfaction of all the parties concerned” (Stoessinger 1976: 74). Examples of continuity are not limited to Kissinger’s dissertation. In an article in Foreign Affairs (1956: 349), he discussed the relationship between Kant and contemporary international politics: In his whimsical essay “Perpetual Peace” written in 1795, the German philosopher Kant predicted that world peace could be attained in one of two ways: by a moral consensus which he identified with a republican form of government, or by a cycle of wars of ever-increasing violence which would reduce the major powers to impotence.

208

9 The Kissinger Years: Studying Individuals and Foreign Policy

Kissinger then applies this to the impact of nuclear weapons, and quotes Eisenhower’s observation that “there is no alternative to peace.” In Kissinger’s first speech as Secretary of State, at the United Nations on September 24, 1973, he said: Two centuries ago, the philosopher Kant predicted that perpetual peace would come about eventually—either as the creation of man’s moral aspirations or as the consequence of physical necessity. What seemed utopian then looms as tomorrow’s reality; soon there will be no alternative. (Dickson 1978: 51)

The continuity of both these themes is demonstrated in a single passage from White House Years (p. 70). Talking about the era of international politics in which he was soon to become a major participant, Kissinger reflects: In the late eighteenth century the philosopher Immanuel Kant in his essay Perpetual Peace, had written that world peace was inevitable; it would come about either because all nations shared the same sense of justice or because a cycle of wars of ever increasing violence would teach men the futility of conflict…. But the root dilemma of our time is that if the quest for peace turns into the sole objective of policy, the fear of war becomes a weapon in the hands of the most ruthless.

Stoessinger, a graduate student with Kissinger at Harvard in the early 1950s and an acquaintance ever since, provides us with a useful summary concerning continuity (1976: 7, 37): His diplomacy as Secretary of State is deeply rooted in the insights of the young doctoral student at Harvard a quarter century ago. It is, in fact, a virtual transplant from the world of thought into the world of power. … We are witness here to a unique experiment in the application of scholarship to statesmanship, of history to statecraft.

As one way to further test this consistency, the discussion will now turn to three substantive issue areas that arise in the White House Years, and contrast Kissinger’s views, ideas and policies in those areas to previous analyses based on the methods discussed above.

9.5

Kissinger and Richard M. Nixon: Nixon the Individual/Nixon the Statesman

One aspect of White House Years and Kissinger’s career has been the issue of Kissinger’s relationship to Richard Nixon. Critics have often pointed to Kissinger’s ‘duplicity’ in disparaging Nixon the person and yet working for Nixon the President, and even lauding various aspects of Nixon’s foreign policy performance. Explanation of this theme is useful because it permits us to link biographical and operational code findings; to get into some of the most basic concepts in Kissinger’s operational code; and to demonstrate that the different perspectives on Nixon are consistent with psychological and intellectual components of Kissinger which have been identified by a number of observers over the years. References in White House Years to Nixon both as man and ‘statesman’ support these earlier analyses.

9.5 Kissinger and Richard M. Nixon: Nixon …

209

Biographical and psychobiographical material may be used to help us understand the basis for the key concepts in Kissinger’s operational code. All of the major works cited above discuss the impact on Kissinger of his family’s experiences in Germany—his father’s loss of his job, the various acts of persecution against Jews, the forced flight from one’s home and native land. Similarly, they all quote Kissinger’s conscious denial that those days had any effect upon him (a disclaimer which is refuted in the memoirs on pages 228–229). Fritz Kraemer best sums up the experience. Kraemer was Kissinger’s first ‘patron,’ an older man of the ‘heroic’ tradition who helped to guide and direct the young Kissinger. The two met in the army during World War II when the Prussian-born, self-exiled Kraemer appealed to Kissinger as a man of principles and action. Kraemer, as all the biographers and analysts have noted, provided Kissinger with his first sense of self-worth and direction and his first role-model. It is ironic that Kraemer was the catalyst for Kissinger’s headlong ‘Americanization’ (a theme Mazlish dwells upon which the others also pick up) during his time in the army. Kraemer is quoted as saying: What the Nazis did to these people is unspeakable. You can do damage to the soul of a man and never touch his body. For five years, the most formative years [10–15], Henry had to undergo this horror. And the real horror is the breakdown of the world. Imagine what it means when your father, who is your authority, the father you admire, is suddenly transformed into a frightened little mouse. (Blumenfeld 1974: 6)

Kraemer notes later that Kissinger had confided to him the ‘psychological’ nature of the Nazi horror (Blumenfeld 1974: 59). A general theme in analyses of Kissinger is that this early experience forged his basic philosophic belief that the world is a place where the forces of chaos constantly battle the forces of order (this is especially well developed in Mazlish, Stoessinger, and Ward). Kissinger’s passion (some say obsession) for stability, balance, and order is supposedly derived from his firsthand experience with the tragedy of upheaval and the desire to prevent it from recurring. Hoffmann (in Blumenfeld 1974: 68–69) summarizes this argument: These experiences have to have an effect. But the lessons one derives are not predictable. I think he came out of it with a kind of burning need for order … this whole emphasis on structures of peace and world order. People in these experiences have a real memory of chaos, of violence and brutality, like the whole world is collapsing under them. Kissinger’s whole search has been to impose a stable world order, a moderate order which would not humiliate anybody. Kissinger, more than most, would probably agree with Goethe’s statement that disorder is worse than injustice.

Indeed, Stoessinger reports that under his questioning during their days at Harvard, Kissinger used Goethe’s words to choose order over justice (Stoessinger 1976: 14). Hoffmann’s quote touches on Kissinger’s most central belief; on the establishment of limits and constraint in international relations; on his distinctions between states (‘legitimate’ and ‘revolutionary’) and between leaders (‘statesmen,’ ‘prophets,’ and ‘conquerors’) based on their acceptance of limits; and on diplomatic

210

9 The Kissinger Years: Studying Individuals and Foreign Policy

behavior designed to moderate, limit, and order the international environment. The prophet and conqueror were both ‘revolutionaries.’ In A World Restored, Napoleon the conqueror and Czar Alexander the prophet were revolutionaries “because both strove to identify the organization of Europe with their will” (Kissinger 1957: 316). The prophet is also viewed as a revolutionary because of the ideologically timeless aspect of his leadership. The prophet’s ‘vision’ is used as the standard for order, truth, and reality. In contrast, Kissinger saw Castlereagh and Metternich as representing ‘statesmen.’ Statesmen “live in time”—they recognize the limits and the possibilities of the present; they conduct themselves on the basis of the recognition of limits which include the survival of the states in the system. Statesmen employ diplomacy and negotiation to recreate and restore international order. In A World Restored, the statesman is characterized as follows: The test of a statesman, then, is his ability to recognize the real relationship of forces and to make this knowledge serve his ends… His instrument is diplomacy, the art of relating states to each other by agreement rather than the exercise of force, by the representation of a ground of action which reconciles particular aspirations with a general consensus … diplomacy depends upon persuasion and not imposition. (Kissinger 1957: 326)9 (see Footnote 6)

In his tendency toward “projective biography,” Kissinger saw himself as a statesman. In relationship to other operational code questions on chance, optimism, pessimism, and mastery over history, this basic conception of political life and how it should be ordered is crucial. Kissinger believed that individuals could make a difference, especially if they understood the basic nature of international relations, the historical context, and were courageous enough to grasp opportunities as they arose. The important point here is that Kissinger, for all his criticism of Nixon’s personal style and personality (to be developed below), also saw Richard Nixon as a statesman and applauded his performance as a statesman. In White House Years, Kissinger gives Nixon high marks for his courage, purpose, and principle in creating, carrying out, and sticking to his foreign policy principles and decisions. Stoessinger (1976: 209) notes: “In Kissinger’s hierarchy of values, courage and decisiveness came first.” Ward (1975: 329) also observes that Kissinger’s modern heroes were statesmen who could act alone in facing and overcoming adversity, and could overcome incredible obstacles. Examples abound. While in Moscow during the 1972 summit, faced with powerful bureaucratic pressures from all sides, Nixon agreed to let Kissinger carry on the SALT discussions on the basis of the unequal number of launch vehicles. This conversation was held while Nixon was having his back treated: “Lying naked on the rubbing table, Nixon made one of the more courageous decisions of his Presidency…. Nixon took a heroic position from a decidedly unheroic posture” (Kissinger 1979:

9

In his memoirs Kissinger again returns explicitly to his intellectual framework. Here he describes the nature of a 1971 interchange with four dedicated anti-war activists: “Ours was the perpetually inconclusive dialogue between statesmen and prophets, between those who operate in time and through attainable stages and those who are concerned with truth and the eternal” (p. 1016).

9.5 Kissinger and Richard M. Nixon: Nixon …

211

1223). In discussing the lattermost stages of the Indochina agreements, Kissinger similarly observes that “maddening as Nixon’s conduct could be in calm times, it verged on the heroic when really critical issues were at stake” (p. 1352). Much of this was due to another strength of the statesman which Kissinger saw in Nixon: “No political leader I have met understood the dynamics of negotiations better than Nixon” (p. 1422). Of course, in large measure, Nixon was being praised for holding the ‘correct’ principles—those that Kissinger believed the statesman should hold, and that Kissinger himself possessed. These beliefs centered on the creation and maintenance of a “stable structure of peace” and world order. For example, Kissinger has repeated throughout his writings and diplomacy the point noted above, that peace at any price is an invitation to disaster. In describing a meeting between Nixon and Prime Minister Heath in December 1971 in which they reviewed the world situation, Nixon repeated, “part of our reason for conducting our Vietnam withdrawal so slowly is to give some message that we are not prepared to pay any price for ending a war; we must now ask ourselves what we are willing to pay to avert war. If we are not, we have tough days ahead.” Kissinger’s response to Nixon’s full review of U.S. and Atlantic relationships was: “Nixon’s later fate must not obscure the importance of this passage. He [Nixon] was right; he had diagnosed both the American and Atlantic problem correctly” (p. 965). Similarly, Kissinger praises Nixon for understanding ‘geopolitical’ structures and balance, of balancing interests, goals, and the capabilities of competing states: It was a hard lesson to convey to a people who rarely read about the balance of power without seeing the adjective ‘outdated’ precede it. It was not one of the least ironies of the period that it was a flawed man, so ungenerous in some of his human impulses, who took the initiative in leading America toward a concept of peace compatible with its new realities and the awful perils of the nuclear age, and that the foreign leaders who best understood this were the two grizzled veterans of the Long March, Mao and Chou, who openly expressed their preference for Richard Nixon over the wayward representatives of American liberalism. … For sovereign nations, predictability is more crucial than spasmodic brilliance or idiosyncratic moralistic rhetoric. They must gear their actions to the performance of others over extended periods of time; their domestic survival and international security alike may depend upon it. And it was on this level of shared geopolitical interest transcending philosophies and history that the former Red-baiter and the crusaders for world revolution found each other. [p. 1089]

Significantly, Kissinger ends his memoirs trying to deal with the contradictions between Nixon the individual and Nixon the statesman: What extraordinary vehicles destiny selects to accomplish its design. This man, so lonely in his hour of triumph, so ungenerous in some of his motivations, had navigated our nation through one of the most anguishing periods in its history. Not by nature courageous, he had steeled himself to conspicuous acts of rare courage. Not normally outgoing, he had forced himself to rally his people to its challenge. He had striven for a revolution in American foreign policy so that it would overcome the disastrous oscillations between over-commitment and isolation. Despised by the Establishment, ambiguous in his human perceptions, he had yet held fast to a sense of national honor and responsibility. [pp. 1475– 1476]

212

9 The Kissinger Years: Studying Individuals and Foreign Policy

Nixon the individual held little charm for Kissinger. He was not the ‘typical’ patron who had figured so conspicuously in Kissinger’s personal and professional development. Three men became important patrons in Kissinger’s life—Kraemer, William Yandell Elliott (Kissinger’s mentor at Harvard), and Nelson Rockefeller. As Mazlish (1976: 77) and others have pointed out, Kissinger’s patrons fit a basic mold, “Renaissance types, men of seigneurial presence.” Not only were these men “great gentlemen,” they were courageous individuals willing to take on the forces of history (Ward 1975: 320). Additionally, Rockefeller had been Kissinger’s final and “quintessential patron: a renaissance man who was also a Prince” (Mazlish 1976: 128). Both Ward and Mazlish point out that as the years passed, Kissinger also sought out men whose intellectual positions fit his own. Ward comments that Kissinger “would commit himself to the ideas of Nixon, a man for whom he has no personal respect” (1975: 320). Mazlish notes that the relationship between the two was formal and intellectual; that the two men never became friends. Mazlish contends that Kissinger’s relationship with Nixon supported Mazlish’s proposition that Kissinger had a unique ability to ‘fit’ with different kinds of men (1976: 212). Ward echoes this idea by suggesting, “Perhaps because of his growing confidence in his own ability and his own ideas, Kissinger became able and willing to accept the challenge of trying to put his ideas into action even if the man who would make the final decision was not a man who Kissinger particularly admired” (1975: 320). What is clear, however, is that Nixon was not of the mold of former patrons. It is obvious from other interviews, and from numerous passages in White House Years, that Nixon as an individual did not hold Kissinger’s esteem. Kissinger uses phrases such as “this lonely, tortured and insecure man” (p. 1086); a “vulnerable and austere” personality, “eager for acceptance but incapable of the act of grace” that could have brought him such acceptance (p. 933); ‘shy’ and “ill at ease with strangers” (p. 760). Nixon is often portrayed as petty (as in the case of forbidding any channel of communication with China being opened through Canada due to his thorough dislike for Trudeau—p. 736); overly concerned with image and public relations (for example p. 734); and paranoid—pursued by the “same liberal conspiracy that had sought to destroy him ever since the Alger Hiss case” (p. 299). The interesting aspect of all this is a certain parallel with criticisms that have been made of Kissinger’s own personality. Landau in particular comes down on Kissinger’s paranoia while a faculty member at Harvard. Ward’s description of Kissinger as a ‘dysmutal’ or depressive personality mirrors many of the observations that Kissinger makes of Nixon. Mazlish presents a clear example of Kissinger’s use of projection as a psychological defense mechanism. He discusses Kissinger’s dependency on his patrons, yet his total rejection of those who were, in turn, dependent upon him (1976: 96–97). Landau, Mazlish, and even the Kalbs discuss Kissinger’s dislike and harsh treatment of graduate students, secretarial help, and his first wife, Ann Fleischer Kissinger. Mazlish observes, “Thus, his dislike of dependency in others mirrored his feelings about himself” (1976: 96). Another argument supporting Kissinger’s use of projection in regard to Nixon is evidence that both men could be classified as “active-negatives” in Barber’s

9.5 Kissinger and Richard M. Nixon: Nixon …

213

classification system (1972). Ward describes Kissinger as a “depressive personality,” which has many of the characteristics of the “active-negative.” Kissinger graphically describes Nixon’s active-negative personality. In a number of references in his memoirs, Kissinger discusses how Nixon would be most withdrawn, unhappy, and mean at points when he should have been the most exultant (for example, in discussing the first inauguration, p. 1). Concerning Nixon’s ruthless handling of his staff after his landslide victory in 1972, (pp. 1406–1408), Kissinger observers: “He did not seem at all elated. Rather, he was grim and remote as if the more fateful period of his life lay ahead.” Indeed, describing Nixon’s first presidential visit to Great Britain, in February 1969, with all its pomp, ceremony, and history, Kissinger writes: “It all produced one of the few occasions of nearly spontaneous joy I witnessed in my acquaintance with this withdrawn and elusive man” (p. 93). The discussion of Kissinger’s personality and ‘projection’ by Mazlish and by Ward (1975: 312) gives us an understanding of Kissinger that is consistent with his observations of the ‘personal’ Nixon. A knowledge of Kissinger’s writings and operational code brings us an understanding of his concept of the ‘statesman,’ the prominent, positive place it holds in his belief system, and his application of the concept to Nixon. His descriptions in White House Years of Richard Nixon are neither petty nor inconsistent, but derive from background and a long-standing intellectual concept.

9.6

Kissinger, Nixon, and the ‘Obstacle’ of the Bureaucracy

In constructing Kissinger’s operational code, especially the instrumental beliefs, one encounters an ongoing concern throughout Kissinger’s writings with the constraints that the statesman faces. As statesmen, both Kissinger and Nixon were confronted by the various international and domestic constraints that Kissinger first described in A World Restored, to which he returned time after time in his writings. These constraints are summarized by Kissinger (1957: 326): The statesman is inevitably confronted by the inertia of his material, by the fact that other powers are not factors to be manipulated but forces to be reconciled; that the requirements of security differ with the geographic location and the domestic structure of the powers. … The acid test of a policy … is its ability to obtain domestic support. This has two aspects: the problem of legitimizing a policy within the governmental apparatus, which is a problem of bureaucratic rationality; and that of harmonizing it with the national experience, which is a problem of historical development.

The point here is that in discussing the domestic and international constraints on the statesman and the various shortcomings of both Castlereagh and Metternich, Kissinger made it clear that he believed that certain statesmen, standing at history’s fateful moments, could influence history through acts of vision and courage. Such

214

9 The Kissinger Years: Studying Individuals and Foreign Policy

statesmen had, first, to recognize such moments. Second, they had to overcome the constraints of domestic opinion and governmental bureaucracy; and in this the statesman had to take on the role of ‘educator’ for both the mass and the governmental elites. A leader who thus ‘understands’ the world (as Kissinger, with his explicit and strongly held concept of world order, would) at a crucial time in history could, through the exercise of will and skillful personal diplomacy, affect history. Kissinger’s view was that the ‘lonely’ and ‘heroic’ leader (as reflected in his famous ‘cowboy’ quote to journalist Oriana Fallaci)10 could be effective only if the statesman was not encumbered by the webs and weight of bureaucracy. This view was in accord both with his academic writings and with his personality and style (see Mazlish 1976: 96, 152, 167). Both Nixon and Kissinger were, to varying degrees, ‘loners,’ and Kissinger’s view of the bureaucracy fit perfectly with Nixon’s desire to tame the bureaucratic system and turn it into his own ‘commissar’ style system. The Kalbs sum up the point, made by many other observers: “In Kissinger, Nixon found an enthusiastic disciple, equally elitist in orientation, equally distrustful of the bureaucracy” (1974: 98). All descriptions of “Henry’s Wonderful Machine” (to use the Kalbs’ phrase) point out the mutual delight of both men in creating a national security system that could be controlled from the White House and that would circumvent and ignore the State Department. Kissinger sketches an identical picture in reporting the meeting at which Nixon actually offered him the job of national security adviser: The President-elect repeated essentially the same arguments he had made two days earlier, emphasizing more strongly his view of the incompetence of the CIA and the untrustworthiness of the State Department. The position of security adviser was therefore crucial to him and his plan to run foreign policy from the White House. [pp. 14–15]

Kissinger’s antipathy to bureaucratic constraints on the statesman and diplomacy go back to his discussion of the statesman in A World Restored (see Graubard 1973: 50–51, 20–21). The theme of the bureaucracy’s quest for ‘safety’ and conservatism which first appeared in that book reappeared in Nuclear Weapons and Foreign Policy (summarized by Graubard 1973: 101) and subsequent writings. As Mazlish and Stoessinger, among others, have noted, Kissinger’s experiences with both the Kennedy and Johnson administrations only served to reinforce his views on bureaucratic constraints. The best summary of these views is to be found in “Domestic Structure and Foreign Policy,” published in 1966, and is repeated and elaborated to varying degrees in “The Policymaker and the Intellectual” (1959) and “Bureaucracy and Policy Making: The Effect of Insiders and Outsiders on the Policy Process” (1968). It is interesting that in his memoirs Kissinger refers to the Fallaci interview as “without doubt the single most disastrous conversation I ever had with any member of the press” (p. 1409). Kissinger notes that this interview was influential in Nixon’s developing sense of competition with Kissinger for public attention. Kissinger says that “what drove him [Nixon] up the wall was a quotation Fallaci put in my mouth: ‘Americans like the cowboy … who rides all alone into the town, the village, with his horse and nothing else… This amazing, romantic character suits me precisely because to be alone has always been part of my style or, if you like, my technique’” (p. 1410).

10

9.6 Kissinger, Nixon, and the ‘Obstacle’ of the Bureaucracy

215

A number of observations from “Domestic Structure and Foreign Policy” illustrate the position Kissinger took into government three years later: The vast bureaucratic mechanisms that emerge develop a momentum and a vested interest of their own…. There is a trend toward autarky. The purpose of the bureaucracy is to devise a standard operating procedure which can cope effectively with most problems…. Bureaucracy becomes an obstacle when what it defines as routine does not address the most significant range of issues or when its prescribed mode of action proves irrelevant to the problem….When this occurs, the bureaucracy absorbs the energies of top executives…. Serving the machine becomes a more absorbing occupation than defining its purpose. [1966: 507]

These observations are reflected in Kissinger’s views of the State Department in White House Years. There are numerous references to its inertia, lack of imagination, stagnation, and general obstructionism to the necessarily ‘innovative,’ ‘bold,’ and ‘heroic’ policies of the true statesman. Kissinger talks of the department’s ‘vacillation’ (p. 603); its inertia and conservatism in regard to opening relations with China (p. 690). The bureaucratic battles between Kissinger and Rogers (as well as Laird at Defense) most often arose after the State Department found out about the ‘real’ foreign policy being conducted from the White House. The main aim of Kissinger and Nixon was to keep information on their activities from the department, to keep it from “dragging its feet” on whatever issue was under consideration (p. 694).11 Indeed, the Kissinger-Nixon style followed the blueprint Kissinger set out in “Domestic Structure and Foreign Policy.” Kissinger predicted that the executive would be forced into ‘extra-bureaucratic’ mechanisms in order to avoid the problems of the “administrative machine”: Faced with an administrative machine which is both elaborate and fragmented, the executive is forced into essentially lateral means of control. … All of this drives the executive in the direction of extra-bureaucratic means of decision. The practice of relying on special emissaries or personal envoys is an example; their status outside the bureaucracy frees them from some of its restraints. [1966: 510–511]

Kissinger, in another example of “projective biography,” became just such a special emissary and relied on ‘lateral’ means of control. Kissinger comments on Nixon’s “penchant for operating through his Assistants rather than Cabinet” (p. 589) in general, and later notes that in deciding on whom to send to Peking for the first major meeting, Nixon decided that “undoubtedly of all the potential emissaries I was the most subject to his control” (p. 717). Not only did Nixon get his special extra-bureaucratic emissary, but he also set up elaborate communications procedures by which to end-run the State Department. Some of the most fascinating material in White House Years is found in Kissinger’s

11

Critics of the Nixon-Kissinger policy process point out that this style was very costly in State Department morale and performance in this underutilization of talent, and in creating a Kissinger “bottleneck” in the foreign policy process. Kissinger does not explore the risks and costs of by-passing the bureaucracy in his memoirs.

216

9 The Kissinger Years: Studying Individuals and Foreign Policy

description of the various ‘backchannels’ developed to conduct foreign relations without using the State Department or without its knowledge: As time went by, the President, or I on his behalf, in order to avoid these endless confrontations, came to deal increasingly with key foreign leaders through channels that directly linked the White House Situation Room to the field without going through the State Department—the so-called backchannels. This process started on the day after Inauguration. [p. 29]

The establishment of the backchannel, how it circumvented the State Department, the dual process of carrying out foreign policy negotiations in public (State Department) and private tracks, all came to play a central role in the Nixon-Kissinger foreign policy. The effort was enormous. The purpose was a foreign policy of statesmen unconstrained by the bureaucratic politics, leaks, and the time-consuming nature of working through the State Department. Kissinger sums up the results of this effort: In May 1971 the Secretary of State did not know of the negotiations in the White House-Kremlin channels that led to the breakthrough in the SALT talks until seventy-two hours before there was a formal announcement. In July 1971 Rogers was told of my secret trip to China only after I was on the way. In April 1972, the President gave Rogers such a convoluted explanation for my trip to Moscow—which had been arranged secretly and which Rogers opposed when he was told at the last minute—that it complicated negotiations. Such examples could be endlessly multiplied. [p. 30]

Whatever Nixon’s motives, Kissinger’s style as National Security Adviser adhered to that of the statesman and his continuous battle against the constraints of the bureaucracy. This style derived from Kissinger’s idiosyncratic background and an intellectual belief that went back to his dissertation. This belief reemerged in Kissinger the policy maker and is repeated almost verbatim in his memoirs. Kissinger recalls that when he was appointed National Security Adviser, My major concern was that a large bureaucracy, however organized, tends to stifle creativity.... In the modern state bureaucracies become so large that too often more time is spent on running them than in defining their purposes.... It seemed to me no accident that most great statesmen had been locked in permanent struggle with experts in their foreign offices. [p. 39]

9.7

Dealing with the Soviet Union

The greatest part of Kissinger’s diplomacy was centered on the Soviet Union and U.S.-Soviet relations. In the post-Afghanistan invasion period, when debate over the manner in which the United States should treat the Soviet Union has been renewed, it is interesting to investigate Kissinger’s views on the subject. Kissinger (and others) would probably concur that there are numerous parallels between today’s situation and those he faced. Kissinger’s comment on the situation after India attacked Pakistan in 1971 sounds familiar to those who have asked about

9.7 Dealing with the Soviet Union

217

Afghanistan, how we will get the Soviets out, or what effect American policy might have: “Though it was now too late to prevent war, we still had an opportunity— through the intensity of our reaction—to make the Soviets pause before they undertook another adventure somewhere else” (p. 886). Observers of the Kissinger era foreign policy, those who have analyzed his preoffice writing as well as his speeches as policy maker, have commented that American policy was centered on the Soviet Union. All other policy issues were considered only as they impinged on this central relationship. Partial support for this assertion has come from the author’s study of Kissinger’s perceptions of the Soviet Union and China, as well as their triadic interactions. For example, in terms of a Soviet-Chinese comparison, the saliency of the Soviet Union in all of Kissinger’s public statements from 1969 to 1976 is clear. Whether one looks at numbers of documents, numbers of evaluative assertations, or percentage of documents, the Soviet Union is by far the more salient attitude object: over four times as many evaluative assertions and more than double the percentage of documents in which one or both of the Communist powers are discussed. Why is this the case in terms particular to Henry Kissinger? The stable world order that is the core of Kissinger’s belief system requires that the major opponent in the system be central to one’s policy. This is particularly true if that power, along with oneself, has the military capability to destroy the world. The Soviet-American capability for mass destruction is a common theme in many of Kissinger’s public statements while in office.12 In addition, both Mazlish and Ward comment on Kissinger’s fascination with the ‘opponent.’ Mazlish discusses how Kissinger’s early experiences promoted a style of interaction which led him to identify with and to be very effective in dealing with opponents (1976: 44–46). Ward quotes Kissinger from the press in 1974: “Allies can be tiresome necessities, but adversaries pose a challenge that can be capped with the sweet triumph of a hard bargained deal” (1975: 340). As discussed in Starr (1980a), Kissinger’s instrumental beliefs revolve around the central theme of negotiations as the tool of the statesman and how the statesman uses force and diplomacy. The tools chosen and the strategy for their use depend upon whether the opponent is a legitimate or revolutionary state. Ideas that appeared in The Meaning of History jelled in A World Restored into conceptions of the ‘legitimate’ and ‘revolutionary’ state which have been significant in Kissinger’s analyses ever since. The legitimate state used diplomacy for achieving limited objectives; it accepted the international context as legitimate and agreed to negotiate its differences within the rules and constraints of that context. The ‘revolutionary’ state, as the ‘revolutionary’ statesman, was ideological in its pursuit of unlimited objectives and thus did not accept as axiomatic the survival of other states. 12

The range of these statements while in office is quite broad. They appear in news conferences in Moscow (on May 27 and 29, 1972) during the summit, Senate hearings on Kissinger’s appointment to Secretary of State in September 1973, and numerous speeches during the Ford Administration (for example, in Miami, August 1974). During Ford’s presidential campaign Kissinger made numerous speeches, and many employed this theme.

218

9 The Kissinger Years: Studying Individuals and Foreign Policy

The design that Kissinger constructed for a stable international order and which he details in White House Years was initially developed in 1968 while he wrote foreign policy papers for Rockefeller. It was based on Kissinger’s conviction that the Soviet Union and the People’s Republic of China were no longer ‘revolutionary’ states (Stoessinger 1976: 43). Hoffmann (1978) asserts that Kissinger came to the conclusion that the USSR and PRC were no longer revolutionary states through his analysis of their perceived needs and self-interest. The change in weapons technology and the mutual balance of terror, as well as the Sino-Soviet split, brought about the primacy of prudence over ideology, Kissinger concluded (1978: 44–45). Hoffmann’s assertion may be supported by the following sequence. The USSR was clearly identified as ‘revolutionary’ by Kissinger in Nuclear Weapons and Foreign Policy (1957). The changeover began in Necessity for Choice (1961), where Kissinger was developing a number of his ideas on arms control. He saw the Soviet Union as willing to enter into arms control agreements through the self-interest of survival. The changeover becomes more explicit in The Troubled Partnership (1965), when he speaks of both the necessity of nuclear survival and the Sino-Soviet split. The final step comes in early 1968 during Rockefeller’s unsuccessful bid for the Republican presidential nomination. Graubard chronicles Rockefeller’s public statements (based on Kissinger’s policy papers) calling for new “understandings being reached with the Soviet Union and Communist China” (see Graubard 1973: 243, 248, 252, 253). Thus, the Soviet Union was the central and salient foreign policy object, and it could be dealt with as a ‘legitimate’ state. As opposed to Dulles’ closed or frozen image of the opponent (see Holsti 1962b), a similar evaluative assertion analysis of Kissinger’s perceptions of the Soviet Union reveals an open and flexible image.13 Kissinger rejected the classic psychological image of the enemy as an omniscient and omnipotent actor, who left nothing to chance and who pursued a coherent “grand strategy” (see Gladstone 1959). Indeed, in his memoirs Kissinger describes an analysis of Soviet policy sent to Nixon: “It began by rejecting the proposition that Soviet policy necessarily followed a master plan.” The document went on: It is always tempting to arrange diverse Soviet moves into a grand design. The more esoteric brands of Kremlinology often purport to see each and every move as part of a carefully orchestrated score in which events inexorably move to the grand finale. Experience has shown that this has rarely been the case. [pp. 161–62]

Given all this, how was the Soviet Union to be fit into a stable structure of peace? Kissinger’s views have always taken two tracks (the “dual policy” he was to For example, Holsti looked at the relationship between Dulles’ “General Evaluation” of the Soviet Union and his perceptions of Soviet ‘Hostility.’ For three-month and six-month periods, Holsti’s Spearman correlations were −0.10 and −0.03—no relationships at all. For Kissinger, the results are 0.32 and 0.49, positive and statistically significant at the 0.10 level. For product moment correlations, the figures are 0.42 and 0.63 (the latter, significant at the 0.05 level). This means that Kissinger’s overall “General Evaluation” of the Soviet Union did change; as his perceptions of Soviet ‘Hostility’ became more positive, so did his “General Evaluation.” 13

9.7 Dealing with the Soviet Union

219

follow as a policy maker): (1) increasing the economic, political and cultural ties of interdependence, to give any state a major stake in the system, and thus to impose costs on it should it break the rules of the system; and (2) to utilize a balance of power process, based on the costs that countervailing power would impose on states that broke the ‘rules’ of the system. Again, during his time in office Kissinger often spoke of entangling the Soviet Union within the ‘webs’ of economic and political relations that existed in the system: We have an historic obligation to mankind to engage the Soviet Union in settlements of concrete problems and to push back the shadow of nuclear catastrophe.... And we have begun to construct a network of cooperative agreements in a variety of functional areas— economic, scientific, medical, environmental and others—which promise concrete benefits if political conditions permit. … It has been our belief that, with patience, a pattern of restraints and a network of vested interests can develop which will give coexistence a more hopeful dimension and make both sides conscious of what they would stand to lose by reverting to the politics of pressure, confrontation and crisis. [U.S. State Department 1976b: 201–212].

This logic formed the basis for the Kissinger-Nixon ideas of ‘linkage,’ where limits (central to Kissinger’s operational code) were imposed by self-restraint. This restraint came from understanding that costs would be imposed in other areas for lack of restraint; “because in an interdependent world the actions of a major power are inevitably related and have consequences beyond the issue or region immediately concerned” (p. 129). Restraint would not come from simply demonstrating one’s “good will.” In a 1962 article on the Cuban missile crisis, Kissinger questioned what sorts of lessons the Soviets learned from U.S. behavior at Suez, in Lebanon, and in the Bay of Pigs. Kissinger’s view was Kennanesque in the assumption that unilateral acts of goodwill (in the mold of Osgood’s GRIT process) would not be enough. Soviet behavior would best be influenced by demonstrating that costs would be imposed. These would come from the loss of cooperative arrangements, or the possibility of the use of military force. Walker (1975) presents an excellent analysis of Kissinger’s instrumental beliefs concerning the use and interplay of negotiations and force. While he uses his analysis mostly on the Indochinese negotiations, it is also relevant to this “dual track” policy toward the Soviet Union. In the memoirs, Kissinger presents a strategy similar to the one Walker has extracted from preoffice writings: “In dealing with the Soviets a point is inevitably reached where it is important to make clear brutally that the limits of flexibility have been reached, that the time has come either to settle or to end the negotiation. This is a more complex matter than simply getting ‘tough’” (p. 818). To Kissinger, “getting tough” was simply part of a larger process of positive incentives and negative costs which one used to help an opponent see his own interest in restraint, and to impose restraint if necessary. He makes this clear in a 1976 speech: We therefore face the necessity for a dual policy: on the one hand, we are determined to prevent Soviet military power from being used for political expansion; we will firmly discourage and resist adventurist policies. But at the same time, we cannot escalate every political dispute into a central crisis; nor can we rest on identifying foreign policy with

220

9 The Kissinger Years: Studying Individuals and Foreign Policy

crisis management. We have an obligation to work for a more positive future. We must couple opposition to pressure and irresponsibilty with concerned efforts to build a more positive world. [U.S. State Department 1976c]

Thus, Kissinger’s policy toward the Soviet Union was one part ‘détente’ (the use of positive incentives of mutual benefit and increasing webs of interdependence) and one part Kennan (firm response to adventurist, expansionist, and destabilizing behavior). This dual response comes through in numerous discussions of detente in White House Years. For example, he notes, “Our strategy of detente always depended on a firm application of psychological and physical restraints and determined resistance to challenge” (p. 1143). Specifically, “All experience teaches that Soviet military moves, which usually begin as tentative, must be resisted early, unequivocally, and in a fashion that gives Soviet leaders a justification for withdrawal” (p. 569). Early in the memoirs, discussing SALT and U.S. defense policy, Kissinger noted, “It was the perennial debate whether Soviet interest in compromise was best elicited by making unilateral American gestures or by presenting the Kremlin with risks and programs they were eager to stop” (p. 535). In a later discussion on the development of Indochina policies in 1972, Kissinger makes clear how Moscow should be treated: “It was essential to react strongly, if necessary violently, in the early stages of Soviet expansion—I advocated this over Cienfuegos, Jordan, India-Pakistan, and was to do so over Angola. In May I was prepared to risk the Moscow summit for whatever was necessary to break Hanoi’s offensive” (p. 1158). Kissinger’s insistence that this dual path be followed is a major component of his most recent critiques of the Carter administration’s foreign policy—first for ignoring the use of stern restraints, and then for jumping to ‘tough’ measures without continuing positive measures. His critiques include poor timing on Carter’s part, an issue crucial to the use of negotiations and force (see Walker 1975). Timing is especially relevant to the USSR: “Success in negotiations is a matter of timing, and nowhere more so than with Moscow” (p. 818). Part of the poor timing is a lack of consistency and clear signals. Interestingly, these critiques reflect criticisms of his own policy toward the Soviet Union14 (see Footnote 11). Many observers never understood how both sides of the dual policy could run together. Critics on the right, of a ‘hawkish’ persuasion, were dubious and mistrusted any cooperative moves toward the Soviet Union, especially those connected with arms control; critics on the left, of the more ‘dovish’ sort, attacked the continuing use of or threat of the use of force and the periodic ‘tough’ confrontationary measures of the Nixon administration. Such activities appeared to be merely a throwback to the days of the Cold War. And yet, in policy making up through 1973, Kissinger felt that the dual track had been

14

See Starr (1980a) for several examples where Kissinger’s criticisms of statesmen such as Metternich and Castlereagh are mirrored closely in contemporary criticism of Kissinger himself.

9.7 Dealing with the Soviet Union

221

successful—in SALT, in Vietnam, in central Europe, even in the Middle East and on the subcontinent.15 The point, once again, is that Kissinger’s views on dealing with opponents in the system, the Soviet Union in particular, had been presented in his earlier scholarly writings, and had a strong basis in previously analyzed idiosyncratic factors.

9.8

Conclusion

Henry Kissinger’s memoirs, White House Years, provide us with one source of data on an important era in American foreign policy; on the processes of the Nixon White House; and on Henry Kissinger the policy maker, intellectual, and individual. How useful is this data source to us, with regard to its general accuracy, its consistency with Kissinger’s previous thought and earlier analysis of Kissinger, and its reflection of how Kissinger viewed his policy and behavior? The issue areas selected here are representative of a number of substantive and stylistic areas dealt with in the memoirs which are indeed consistent with earlier biographical, operational code, content analytic, and policy research on Henry Kissinger. As noted at the outset, many of these previous studies are psychological or cognitive. In many ways this is quite fitting for analysis of Henry Kissinger. While Kissinger insisted that it was foolish to try to understand him using psychological techniques (Mazlish 1976: 196), he himself has always used them. His army career introduced him to psychological warfare activities, which he appeared to master easily. A Harvard roommate recalls (Mazlish 1976: 58) that at one point Kissinger considered becoming a psychiatrist. Graubard’s discussion of A World Restored continually highlights Kissinger’s concern with knowing and understanding how one’s opponents perceive an issue and what they believed (1973: 11). Part of Metternich’s ‘genius’ was in his handling of negotiations with Napoleon—“He had to know Napoleon’s mind” (1973: 22); “For Napoleon to have acted other than he did . .. would have meant that he had ceased to be Napoleon” (1973: 27). In the end, Mazlish terms Kissinger a “psychiatrist manque, a psychological historian” (1976: 197). Just as Kissinger searched for the clues to the personality and world view of the leaders with whom he would be dealing (as reported by the Kalbs), he had a clear and consistent world view or belief system of his own, and he acted on it. This link between belief system and behavior has been made by Walker (1977) in regard to

An illustrative comment on detente is provided on page 949: “As I explained earlier, we considered a detente based on strict reciprocity in the national interest But whatever one’s views about detente in the abstract, in the context of 1971 and 1972 the carefully calibrated measures of the Administration toward the Soviet Union were imperative to prevent a headlong rush toward abdication of responsibility in America and among our allies. Our willingness to discuss detente had lured Brezhnev into an initiative about mutual force reductions that saved our whole European defense structure from Congressional savaging. It was a classic example of why such a policy was needed to maintain the essential elements of national security.”

15

222

9 The Kissinger Years: Studying Individuals and Foreign Policy

Kissinger’s Indochina policies. Walker has provided an excellent analysis of Kissinger’s positions and strategies on the basis of an operational code analysis, especially Kissinger’s instrumental values on the use of force and negotiation. This Schelling-like analysis clearly indicated that Kissinger’s decision-making behavior followed the outlines set forth in his academic writings. Walker’s analyses are reconfirmed in a number of places in the memoirs (see pp. 308, 436, 478, 498, 1315–1316, 1328, 1457). The point is that Kissinger had a well-defined belief system, based on personal and intellectual factors, and he acted on it. His memoirs indicate that he also framed his retrospective in terms of that belief system. He both recalled and described his behavior as policy-maker in a manner consistent with his personality and belief system. Kissinger is understandable. He held a coherent and plausible view of the world —concerned with certain values, and not with others; concerned with order, limits, and restraint, and a psychological perspective on how one forged agreements with opponents. These views had been set out in detail before he entered office. While in office, he pursued the goals of the ‘statesman’ with the tools of the statesman. This review is not intended to be a judgment on how accurate that view of the world was or ultimately how successful it was. Such judgments will be left to the policy analysis commentaries that were excluded at the beginning of this article. The point is that Kissinger’s view is an intellectually respectable and coherent view, and it appears from a variety of analyses that he attempted to follow it in good faith. White House Years reinforces the idea of continuity in Kissinger’s world view. There is also a remarkable congruence between the self-analysis of the memoirs and previous analyses. In sum, White House Years is a volume rich in the details of the Nixon-Kissinger foreign policy. It adds new context and sequence to many facts, and provides additional data and arguments as to what was occurring and how it was being perceived from the White House.

References Axelrod, R., 1976: The Structure of Decision (Princeton, NJ: Princeton University Press). Barber, J. D., 1972: The Presidential Character (Englewood Cliffs, NJ: Prentice-Hall). Blumenfeld, R., et al., 1974: Henry Kissinger (New York: New American Library). Brandon, H., 1973: The Retreat of American Power (Garden City, NY: Doubleday). Brodin, K., 1972: “Belief Systems, Doctrines, and Foreign Policy”, in: Cooperation and Conflict, 7: 97–112. Brody, R., 1969: “The Study of International Politics Qua Science”, in: Knorr, K.; Rosenau, J. N. (Eds.): Contending Approaches to International Politics (Princeton, NJ: Princeton University Press). Dickson, P., 1978: Kissinger and the Meaning of History (Cambridge, England: Cambridge University Press). Eldridge, A., 1975: “Pondering Intangibles: A Value Analysis of Henry Kissinger.” Prepared for a Symposium on the Theory and Practice of Henry Kissinger’s Statesmanship, College Park, Maryland.

References

223

Eldridge, A., 1976: “The Crisis of Authority: The President, Kissinger and Congress (1969-1974)”, Presented at the International Studies Association Annual Meeting, Toronto. Falkowski, L. S., 1979a: “Introduction: Evaluating Psychological Models,” in: Falkowski, L. S. (Ed.): Psychological Models in International Politics (Boulder, CO: Westview Press). Falkowski, L. S., 1979b: “Psychological Models and Systemic Outcomes”, in: Falkowski, L. S. (Ed.): Psychological Models in International Politics (Boulder, CO: Westview Press). George, A. L., 1956: Woodrow Wilson and Colonel House: A Personality Study (New York: John Day). George, A. L., 1969: “The ‘Operational Code’: A Neglected Approach to the Study of Political Leaders and Decision Making”, in: International Studies Quarterly, 13: 190–222. George, A. L., 1979: “The Causal Nexus Between Cognitive Beliefs and Decision-Making Behavior: The ‘Operational Code’ Belief System”, in: Falkowski, L. S. (Ed.): Psychological Models in International Politics (Boulder, CO: Westview Press). George, A. L.; George, J. L., 1979: “Dr. Weinstein’s Interpretation of Woodrow Wilson: Some Preliminary Observations”, in: Psychohistory Review, 8: 71–72. Gladstone, A. L., 1959: “The Conception of the Enemy”, in: Journal of Conflict Resolution, 3: 132–137. Graubard, S. R., 1973: Kissinger, Portrait of a Mind (New York: W. W. Norton). Greenstein, F. I., 1969: Personality and Politics (Chicago: Markham). Hermann, M. G. (Ed.), 1977: A Psychological Examination of Political Leaders (New York: Free Press). Hoffmann, S., 1978: Primacy or World Order (New York: McGraw-Hill). Holsti, O. R., 1962a: “The Belief System and National Images: John Foster Dulles and the Soviet Union” (Doctoral thesis, Stanford University) (unpublished). Holsti, O. R., 1962b: “The Belief System and National Images: A Case Study”, in: Journal of Conflict Resolution, 6: 244–252. Holsti, O. R., 1970: “The ‘Operational Code’ Approach to the Study of Political Leaders: John Foster Dulles’ Philosophical and Instrumental Beliefs”, in: Canadian Journal of Political Science, 3: 123–157. Holsti, O. R., 1975: “Foreign Policy Decision Makers Viewed Psychologically: ‘Cognitive Process’ Approaches.” Duke University (mimeo). Holsti, O. R., 1976a: “Foreign Policy Decision Makers Viewed Psychologically: ‘Cognitive Process’ Approaches”, in: Rosenau, J. N. (Ed.): In Search of Global Patterns (New York: Free Press). Holsti, O. R., 1976b: “Foreign Policy Formation Viewed Cognitively”, in: Axelrod, R. (Ed.): The Structure of Decision (Princeton, NJ: Princeton University Press). Holsti, O. R., 1977: “The ‘Operational Code’ as an Approach to the Analysis of Belief Systems: Final Report of the National Science Foundation, Grant No. SOC 75-15368.” Duke University. Holsti, O. R.; George, A. L., 1975: “The Effects of Stress on the Performance of Foreign Policy-Makers”, in: Cotter, C. P. (Ed.): Political Science Annual (Indianapolis: BobbsMerrill). Holsti, O. R.; North, R. C.; Brody, R. A., 1968: “Perception and Action in the 1914 Crisis”, in: Singer, J. D. (Ed.): Quantitative International Politics (New York: Free Press). Jervis, R., 1976: Perception and Misperception in International Politics (Princeton, NJ: Princeton University Press). Kalb, M.; Kalb, B., 1974: Kissinger (New York: Dell). Kaplan, M. A., 1957: System and Process in International Politics (New York: Wiley). Kissinger, H. A., 1951: The Meaning of History: Reflections on Spengler, Toynbee and Kant (Bachelor’s thesis, Harvard University). Kissinger, H. A., 1956: “Force and Diplomacy in the Nuclear Age”, in: Foreign Affairs, 34: 347– 366. Kissinger, H. A., 1957: A World Restored: Castlereagh, Metternich and the Restoration of Peace, 1812–1822 (Boston: Houghton-Mifflin). Kissinger, H. A., 1959: “The Policy Maker and the Intellectual”, in: The Reporter, 5: 3–35.

224

9 The Kissinger Years: Studying Individuals and Foreign Policy

Kissinger, H. A., 1961: The Necessity for Choice (New York: Harper & Row). Kissinger, H. A., 1965: The Troubled Partnership (New York: McGraw-Hill). Kissinger, H. A.,1966: “Domestic Structure and Foreign Policy”, in: Daedalus, 95: 503–527. Kissinger, H. A., 1968: “Bureaucracy and Policymaking: The Effect of Insiders and Outsiders on the Policy Process”, in: Brodie, B. (Ed.): Bureaucracy, Politics, and Strategy (Los Angeles: University of California). Kissinger, H. A., 1979: White House Years (Boston: Little, Brown). Landau, D., 1973: Kissinger: The Uses of Power (New York: W. W. Norton). Mazlish, B., 1976: Kissinger, The European Mind in American Policy (New York: Basic Books). Shapiro, M. J.; Bonham, G. M., 1973: “Cognitive Processes and Foreign Policy Decision-Making”, in: International Studies Quarterly, 17: 147–174. Shawcross, W., 1979: Sideshow (New York: Simon and Schuster). Starr, H., 1976: “Images and Belief Systems of Decision Makers: A Study of Henry Kissinger”, Presented at the Peace Science Society (International). Midwest Section Annual Meeting, Chicago. Starr, H., 1979: “Henry Kissinger’s Belief System and World Order: Perception and Policy”, in: Han, H. H. (Ed.): World in Transition (Washington, DC: University Press of America). Starr, H., 1980a: “Kissinger’s Operational Code”, Presented at the International Studies Association Annual Meeting, Los Angeles. Starr, H., 1980b: “The Operational Code and Other Forms of Content Analysis: Analyses of Henry Kissinger”, Presented at the International Society of Political Psychology Annual Meeting, Boston. Stoessinger, J. G., 1976: Henry Kissinger: The Anguish of Power (New York: W. W. Norton). Sullivan, M. P., 1976: International Relations: Theories and Evidence (Englewood Cliffs, NJ: Prentice-Hall). Tucker, R. C., 1977: “The Georges’ Wilson Reexamined: An Essay on Psychobiography”, in: American Political Science Review, 71: 606–618. U.S. Department of State, 1975: Bulletin, July 7, 1975. Washington, D.C. U.S. Department of State, 1976a: Bulletin, July 26, 1976. Washington, D.C. U.S. Department of State, 1976b: Bulletin, February 23, 1976. Washington, D.C. U.S. Department of State, 1976c: Bulletin, April 5, 1976. Washington, D.C. U.S. Senate, Foreign Relations Committee, 1974: Hearings, September 19, 1974. Walker, S. G., 1975: “Cognitive Maps and International Realities: Henry A. Kissinger’s Operational Code”, Presented at the American Political Science Association Annual Meeting, San Francisco. Walker, S. G., 1977: “The Interface Between Beliefs and Behavior: Henry Kissinger’s Operational Code and the Vietnam War”, in: Journal of Conflict Resolution, 21: 129–168. Ward, D., 1975: “Kissinger: A Psychohistory”, in: History of Childhood Quarterly, 2: 287–348. Watt, D., 1979: “Kissinger’s Track Back”, in: Foreign Policy, 37: 59–66. Weinstein, E. A.; Anderson, J. W.; Link, A. S., 1978: “Woodrow Wilson’s Political Personality: A Reappraisal”, in: Political Science Quarterly, 93: 585–598.

About the University of South Carolina

The Palmetto State established South Carolina College on December 19, 1801, as part of an effort to unite South Carolinians in the wake of the American Revolution, promoting “the good order and harmony” of the state. The founding of South Carolina’s state college was also part of the Southern public college movement spurred by Thomas Jefferson. Within 20 years of one another, Georgia, North Carolina, South Carolina, and Virginia established state-supported colleges. In the antebellum era, the Palmetto State generously supported South Carolina College. Its faculty included noted European scholars such as Francis Lieber and Thomas Cooper, as well as renowned American scholars John and Joseph LeConte. Offering a classical curriculum, South Carolina College became one of the South’s most influential colleges, earning a reputation as the training ground for South Carolina’s antebellum elite. But South Carolina’s secession from the Union unleashed the devastation of the American Civil War, and the state and South Carolina College paid dearly. The institution closed for want of students, and in the ensuing decades it struggled to regain its former status. As Reconstruction from the Civil War proceeded, the state’s General Assembly chose the first African-Americans to serve on the University’s Board of Trustees in 1868, and in 1873 the first black students enrolled. While politically controversial, this development was an extraordinary opportunity at a time when opportunities for higher education were rare. The University of South Carolina became the only Southern state university to admit and grant degrees to African-American students during Reconstruction. Following the end of Reconstruction in 1877, South © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7

225

226

About the University of South Carolina

Carolina’s conservative leaders closed the University. They reopened it in 1880 as an all-white agricultural college, and during the next 25 years the institution became enmeshed in the state’s political upheaval. In 1906, the institution was re-chartered for the final time as the University of South Carolina. In 1917 it became South Carolina’s first state-supported college or university to earn regional accreditation, and the 1920s brought the introduction of new colleges and degree programs, including the doctorate. The outbreak of World War II launched an era that transformed the University. Carolina hosted naval training programs during the war, and enrollment more than doubled in the postwar era as veterans took advantage of the G.I. Bill. In the 1950s, the University began recruiting national-caliber faculty members and extended its presence with the establishment of campuses across South Carolina. In 1963 the University of South Carolina became the university of all the people of the state. As the result of a federal court order, Henrie D. Monteith, Robert Anderson, and James Solomon became the first African-American students to enroll at the University in the 20th century. Minority enrolment would continue to grow in their wake and was complemented by a substantial international student population in subsequent decades. At the same time, enrollment was 5,660 in 1960, but nearly quadrupled 20 years later. That increase in the student body was accompanied by the introduction of many new programs, including three that would prove to be momentous. University 101 became a national model for cultivating freshman-year success; the accompanying National Resource Center for the First-Year Experience and Students in Transition is based at the University. The international master’s in business administration program, also launched in the 1970s, has consistently ranked among the country’s top three such programs. The honors program blossomed into the South Carolina Honors College and is now hailed among the nation’s finest. As the result of concerted efforts to expand its research capabilities, the University in 2006 was designated by the Carnegie Foundation for the Advancement of Teaching as a research institution of “very high research activity,” the foundation’s highest classification. At that time the University was South Carolina’s only institution to have earned that distinction.

University of South Carolina, Department of Political Science Founded in 1937, the Department of Political Science is home to active scholars who span the major fields of the discipline, drawing on a range of methodological approaches, and whose research has appeared in the leading journals. One of the largest departments in the College of Arts & Sciences with more than 800 undergraduate students, the Department of Political Science offers two bachelor’s degrees – one in political science, the other in international studies – as well as the

About the University of South Carolina

227

Master of Public Administration (M.P.A.), the Master of Arts in International Studies (M.A.I.S.), and the Ph.D. Programs include American Politics, International Relations, Comparative Politics, Public Administration, and Public Law. The MPA Program is the leading program in the state and among the best in the Southeast. Begun in 1968, it was South Carolina’s first fully accredited autonomous professional public administration program. The program has graduated hundreds of professional administrators and public leaders. The MAIS Program is designed for students interested in a professionally-oriented program geared toward service in governmental, nonprofit, or private sectors – both domestic and international. The doctoral program, designed to produce leading scholars in the discipline, has produced many distinguished academics and numerous public leaders, both nationally and abroad. The department is home to a diverse and distinguished faculty. Faculty members play prominent roles in many professional organizations and on numerous Editorial Boards, serve as resources for the news media and community organizations, and participate in and direct major University initiatives. More at: http://artsandsciences. sc.edu/poli/.

About the Author

Harvey Starr (Ph.D., Yale University 1971) is the Dag Hammarskjöld Professor in International Affairs Emeritus in the Department of Political Science. He joined the faculty at the University of South Carolina in 1989 after 17 years at Indiana University, and has served as chair of the Political Science Department at both institutions. He is author or co-author of 19 books and monographs, and over one hundred journal articles and book chapters. The recipient of multiple grants from the National Science Foundation, he has served as President of the International Studies Association (2013–14), President of the Peace Science Society (2000–01), and Vice-President of the American Political Science Association (1995–96). In 2015 he received the Lifetime Achievement Award from the Conflict Processes Section of the American Political Science Association, “in recognition of scholarly contributions that have fundamentally improved the study of conflict processes.” He also received the International Studies Association Section on Political Demography and Geography’s 2016–17 Distinguished Scholar/Myron Weiner Award. Harvey Starr’s research and teaching interests include theories and methods in the study of international relations, war and international conflict, geopolitics and diffusion analyses, and domestic influences on foreign policy (revolution; democracy), with current research interests in the causes and consequences of failed states and the theory and methods of necessary conditions. In November 2016 he brought to a successful conclusion the final Ph.D. committee on which he served as chair; the twenty-first such committee at South Carolina. Selected Publications: World Politics: the Menu for Choice 10th edition (Wadsworth Cengage, 2012; with David Kinsella and Bruce Russett); On Geopolitics: Space, Place, and International Relations (Paradigm, 2013); Bruce Russett: Pioneer in the Scientific And Normative Study Of War, Peace, and Policy © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 H. Starr, Harvey Starr: Pioneer in the Study of Conflict Processes and International Relations, Pioneers in Arts, Humanities, Science, Engineering, Practice 29, https://doi.org/10.1007/978-3-030-78907-7

229

230

About the Author

(Springer International Publishing AG, 2015); The Israeli Conflict System: Analytic Approaches (Routledge, 2016; ed. with Stanley Dubinsky); State Failure in the Modern World (Stanford University Press, 2016; with Zaryab Iqbal). E-mail: Webpage:

April 2017 at Cologne, on the Viking River Cruise up the Rhine. Source From the author’s personal photo collection