Negotiation, Identity and Justice: Pathways to Agreement 2022050678, 2022050679, 9781032275741, 9781032275734, 9781003293361

Citation preview

‘This book tells the story of a highly productive Interdisciplinary research career. Though the main focus of Dan Druckman’s scholarship has always been on negotiation, he has also contributed to our understanding of many other topics including mediation, international diplomacy, representation, ingroup identity, and nationalism. His many theoretical contributions, most notably those on f lexibility and turning points in negotiation, have always been backed up by solid empirical data . . . and he has also made ground-breaking practical contributions in his work on aids for mediators and negotiator training. Informative and inspirational reading throughout.’ Dean G. Pruitt, Professor, State University of New York at Buffalo and Scholar in Residence, School for Conflict Analysis and Resolution, George Mason University, USA ‘An intellectual history from one of the leading conf lict management scholars of his generation and whose work has inf luenced multiple disciplines. This book comes with the bonus of twelve of Druckman’s original publications that helped define and extend the conf lict management field.’ Paul F. Diehl, Past President, International Studies Association and Past President, Peace Science Society (International) ‘The blend of intellectual curiosity, rigorous methodology and deep experience that has epitomized Professor Druckman’s career runs through this book. It is an invaluable resource for established scholars, one that enriches our understanding of negotiation and conf lict resolution. For young scholars, it serves a roadmap for crafting a meaningful and impactful career.’ Mara Olekalns, Professor of Management, Melbourne Business School, Australia ‘For nearly six decades, Dan Druckman has been a leading scholar and practitioner of such complex topics as conf lict analysis and resolution. This volume is a culmination of lessons learned from a distinguished career in the social sciences, and advice for future directions to advance scholarly understanding. Highly recommended, especially for young and emerging scholars who are building their own careers.’ Mark J. Rozell, Dean, Schar School of Policy and Government, George Mason University, USA ‘Dan Druckman has had a remarkable career that has resulted in many foundational concepts for the broad field of conf lict management. These concepts have grown from Dan’s experiences as both a practitioner and academic. As a result, the book not only provides a blueprint for how to build a successful career, but it also shows how an incredible research program can be created to provide many important contributions to social science. This book contributes a rich legacy that deserves to be on any scholar’s shelf.’ William Donohue, Professor, Department of Communication, Michigan State University, USA ‘This book comprehensively integrates and builds upon a prolific professional career that spans an impressive diversity of positions, research themes, sponsors, and methodological approaches. In doing so, Druckman offers scholarly insights into how the various research

programs to which he contributes have evolved, as well as where they might go next. In a rare move, he also pulls back the curtain for an honest discussion of what often remains unseen and unsaid: the stories, relationships, processes, and experiences that underlie the practice of social science itself. Anyone connected to the social science enterprise – scholars, non-profits, government agencies, consultants, sponsors, or otherwise – will therefore learn tremendously from this book.’ Andrew Owsiak, Josiah Meigs Teaching Professor & Professor of International Affairs, University of Georgia, Athens, GA, USA ‘This book is a goldmine for scholars and practitioners eager to understand and build cultures of peace. An extraordinary career driven by curiosity, rigor, generosity, and a constant quest for understanding. A model of determination, resilience, and creativity that will inspire readers to reach for the stars. This book is – and will remain – fascinating and moving for decades to come. A superb exercise of ref lexivity.’ Valerie Rosoux, Professor, University of Louvain, Director, Belgian National Fund for Scientific Research, Member of the Belgian Royal Academy, Beligum ‘Dan Druckman’s work on ‘Turning Points’ inf luenced much of my work (and those of many others) as it provides a lens to analyze a negotiation process. In this engaging volume, the turning point metaphor helps us appreciate the wide-ranging contributions Druckman has made to the field.’ Deborah M. Kolb, Deloitte Ellen Gabriel Professor for Women and Leadership (Emerita), Simmons School of Business, USA ‘This book is an impressive collection of carefully selected work by one of the true trailblazers of field of peace and conf lict studies. It brings together some of the most fundamental ideas and concepts that Daniel Druckman has pioneered throughout his outstanding research career. The volume juxtaposes themes that for a long time have been treated independently, offering a unique perspective of the intersections between processes of negotiation, identity formation and justice. In a world that is witnessing a resurgence of ethnonational political agendas, this volume provides a set of valuable reminders of the complexities of managing various forms of identity conflicts, how to time and maneuver those activities by better understanding key turning points, and how to address deep-seated grievances and perceptions of past injustices that drive the parties into conf lict. This book will prove to be mandatory reading for the new generation of scholars, who in one place may find the most consequential findings, insights and ideas in the field of conflict management.’ Siniša Vuković, Johns Hopkins University, School of Advanced International Studies, USA ‘Whether you are a seasoned negotiator or a newbee to the field; you will find useful information in this book about how to deal with difficult negotiations or conf licts. Dan Druckman is a productive researcher with a keen eye for theoretical insights and applications of these insights. If he hasn’t studied it, it is probably not worth studying. By writing this book, he shares his insights, knowledge and experience in the broad field of conf lict analysis and resolution. What a gift!’ Fieke Harinck, Leiden University, Netherlands

NEGOTIATION, IDENTITY AND JUSTICE

This volume presents contributions made by Daniel Druckman on the topics of negotiation, national identity, and justice. Containing research conducted and published over a half century, the volume is divided into seven thematic parts that cover: the multifaceted career, flexibility in negotiation, values and interests, turning points, national identity, and process and outcome justice. It rounds off with a reflective and forward-looking conclusion. Each part is prefaced with an introduction that highlights the chapters to follow. T he chapters comprise empirical, theoretical, and state-of-the-art articles. These essays offer an array of research approaches, which include experiments, simulations, and case studies, with topics ranging from boundary roles and turning points in negotiation to nationalism and war, and the way that research is used in skills training for diplomats and in the development of government policies. In addition, the book provides rare glimpses of behind-the-scenes networks, sponsors, and events, with personal stories that also make evident that there is more to a career than what appears in print. The articles chosen for inclusion are a small set of the total number of career publications by the author but are the ones that made a substantial impact in their respective fields. The concluding section looks back at how the author’s career connects to classical ideas and the value of an evidence-based approach to scholarship and practice. It also looks forward to directions for future research in six areas. This book will be of considerable interest to students of international negotiation, conf lict resolution, security studies, and international relations. Daniel Druckman is Professor Emeritus of Public and International Affairs at George Mason’s Schar School of Policy and Government and Honorary Professor at Macquarie University in Sydney and at the University of Queensland in Brisbane, Australia. He is the recipient of five Lifetime Achievement awards.

Routledge Studies in Security and Conflict Management Series Editors: Fen Osler Hampson, Carleton University, Canada Chester Crocker, Georgetown University, Washington DC Pamela Aall, United States Institute of Peace, Washington DC

This series will publish the best work in the field of security studies and conf lict management. In particular, it will promote leading-edge work that straddles the divides between conf lict management and security studies, between academics and practitioners, and between disciplines. Understanding Quality Peace Peacebuilding After Civil War Edited by Madhav Joshi and Peter Wallensteen International Negotiation and Mediation in Violent Conf licts The Changing Context of Peacemaking Chester A. Crocker, Fen Osler Hampson and Pamela Aall Conf lict and Collaboration For Better or Worse Edited by Louis Kriesberg and Catherine Gerard Multiparty Mediation in Violent Conf lict Peacemaking Diplomacy in the Tajikistan Civil War Testuro Iji International Negotiation and Political Narratives A Comparative Study Edited by Fen Osler Hampson and Amrita Narilkar Negotiation, Identity and Justice Pathways to Agreement Daniel Druckman For more information about this series, please visit: www.routledge.com/RoutledgeStudies-in-Security-and-Conf lict-Management/book-series/RSSCM

NEGOTIATION, IDENTITY AND JUSTICE Pathways to Agreement

Daniel Druckman

Cover image: Designed by Parker Wallman First published 2023 by Routledge 4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2023 Daniel Druckman The right of Daniel Druckman to be identified as author of this work has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Druckman, Daniel, 1939– author. Title: Negotiation, identity and justice : pathways to agreement / Daniel Druckman. Description: Abingdon, Oxon ; New York, NY : Routledge, 2023. | Includes bibliographical references. | Identifiers: LCCN 2022050678 (print) | LCCN 2022050679 (ebook) | ISBN 9781032275741 (hardback) | ISBN 9781032275734 (paperback) | ISBN 9781032275734 (ebook) Subjects: LCSH: Diplomatic negotiations in international disputes. | Mediation, International. | Nationalism. | Justice. Classification: LCC JZ6045 .D784 2023 (print) | LCC JZ6045 (ebook) | DDC 327.73—dc23/eng/20230118 LC record available at https://lccn.loc.gov/2022050678 LC ebook record available at https://lccn.loc.gov/2022050679 ISBN: 978-1-032-27574-1 (hbk) ISBN: 978-1-032-27573-4 (pbk) ISBN: 978-1-003-29336-1 (ebk) DOI: 10.4324/9781003293361 Typeset in Bembo by Apex CoVantage, LLC

To Marjorie with Love

CONTENTS

List of Figures List of Tables Preface Foreword by Fen Osler Hampson Acknowledgments Introduction

xii xiii xiv xvii xix 1

PART 1

A Career in Three Parts: Research Institutes, Consulting Firms, and Universities

7

Part Introduction

9

1 Building a Career

12

2 Learning About Bargaining

16

3 Tackling International Negotiation

20

4 Puzzles About National Identity

34

5 A Role for Justice

40

x

Contents

6 Behind the Scenes

44

Part Conclusion

50

PART 2

Flexibility in Negotiation Part Introduction

53 55

7 Prenegotiation Experience

57

8 Boundary Roles

76

9 Situational Levers

98

PART 3

Conflicting Values and Interests Part Introduction 10 Conf lict of Interest and Value Dissensus Daniel Druckman and Kathleen Zechmeister

119 121 122

PART 4

Turning Points Part Introduction

141 143

11 Base Rights Negotiation

144

12 Comparative Analysis

175

13 Thirty-Five Years and Counting

203

PART 5

National Identity Part Introduction 14 Ethnocentrism

219 221 223

Contents

xi

15 Nationalism and Patriotism

252

16 Scaling Up, Down, and Across Daniel Druckman and Steve Wood

285

PART 6

Process and Outcome Justice Part Introduction

309 311

17 Durable Peace Daniel Druckman and Lynn M. Wagner

313

18 Justice and Negotiation Daniel Druckman and Lynn M. Wagner

344

PART 7

Conclusions Part Introduction

381 383

19 Looking Back

384

20 Looking Forward

388

References Index

393 400

FIGURES

3.1 3.2 8.1 9.1 9.2 10.1

11.1 11.2 12.1 12.2 12.3 12.4 14.1 14.2 17.1 18.1

The Sawyer-Guetzkow Model of Negotiation Three Models of International Negotiations The Adjustment-Expectation-Decision Cycle U.S. Japanese Administrative Agreement U.S. Base Rights in Spain An Experimental Design in Which the Extent of Polarization on Ideologies is Orthogonal to Cross-Cutting Interests as an Example of a Mediational Mechanism Trend in Percent Hard Behavior for Both Teams Across the Negotiating Sessions Trends in Percent Hard Behaviors for the Spanish and American Teams Across the Negotiating Sessions Consequence of a Previous Escalation (t → t + 1) across Cases Paths for Security Cases Paths for Political Cases Paths for Trade Cases Mean Ratings of Members of Own Nation, Allies, and Enemies for Each Trait Combination Mean Ratings by Each Role of Members of Own Nation, Allies, and Enemies Serial Multiple Mediation. Adapted from Hayes (2013) Justice and Negotiation: A Framework.

21 22 84 113 114

135 164 166 192 193 193 194 239 242 321 355

TABLES

7.1 Means and Standard Deviations for Six Conditions of Prenegotiation Experience on Three Measures of Conf lict Resolution 66 7.2 Number of Dyads with 0, 1, 2, 3, or 4 Unresolved Issues for Four Conditions of Prenegotiation Experience 68 9.1 Comparing Conditions: Bargaining Experiment 107 11.1 Activities of Negotiators 152 11.2 Definitions of Categories Used to Code Negotiators’ Verbal Statements 154 11.3 The Spain Military Bases Negotiation: Processes and Influences 158 11.4 Correlations Among Indices of Negotiating Behavior 167 12.1 Cases Categorized by Type of Negotiation 183 12.2 Type of Negotiation by Precipitant Category (Proportions) 189 12.3 Type of Negotiation by Type of Precipitant (Frequencies) 189 12.4 Precipitant by Departure 190 12.5 Precipitant by Consequence 191 12.6 Departure by Consequence 191 14.1 Mean Ratings of Members of Own Nation, Allies, and Enemies, Arranged by Type of Trait and Set of Runs 234 14.2 Significant Effects and Size of F Ratios for Three Types of Analyses of Variance, Arranged by Type of Trait and Set of Runs 237 14.3 Means for Roles Rating Members of Own Nation, Allies, and Enemies 242 17.1 Descriptive Statistics 327 17.2 Procedural Justice and Control Variables Regressed on Reconciliation and DP. 332 17.3 PJ + DJ and Control Variables Regressed on Reconciliation and DP 334 18.1 Hypotheses for Future Studies on Justice and Negotiation 373

PREFACE

The story of my career as a social scientist is presented in this book. The career began in 1966 when I received a PhD from Northwestern University. It traveled widely across the globe and delved deeply into many areas. In these ways it is a unique career. What makes it interesting, and worth telling about, are the opportunities provided to contribute in many ways to the growth and further development of the social sciences. Three types of contributions are highlighted in the book: negotiation, identity, and justice. The reader will learn about a variety of facets on each of these themes. With regard to negotiation, both bargaining experiments and international negotiation case studies are discussed. On identity, both empirical and theoretical work are presented. The justice theme is investigated in different issue areas, including civil wars, trade, arms control, and environmental negotiation. My aim in all of this is to reveal what was accomplished and how it was done. Strap in for a ride that has some surprising and bumpy moments. It is less a matter of encouraging readers to model this career than taking away exemplums from it to fulfill their own aspirations. The career is unique in several ways. It has roamed around a landscape of puzzles asked about the most microscopic phenomena (micro-momentary facial expressions) and the most macroscopic topics (societal peace). Most of the research however has been in between these levels with a focus on groups of delegates in international negotiation. This “meso-scopic” focus illuminates a bridging role played by the social sciences. When studying groups or small communities, a researcher needs to understand both interpersonal and intergroup relations. The career has also bridged research and practice. Many of the projects were performed in a consulting milieu. With clients to satisfy, I needed to attend to their framing of the problems. This framing was usually oriented toward action and dealt with effectiveness. Alternatively, an academic focus is geared in

Preface

xv

the direction of understanding and embraces theory more than action. I wore both these hats during the consulting years. My “daytime job” focused on the clients’ problems. My “weekend job” altered the framing toward searching for more general implications suitable for academic contributions. There is more to the career that features less prominently in the book. I was also a teacher, trainer, administrator, and professional association leader. These activities were woven into the tapestry of the research and consulting career. For example my graduate teaching and training gigs were evidence-based. Students were fed a healthy dose of the most recent research on the topic of the class. They conducted both team and individual projects involving experiments, surveys, or archival sources. They learned how to write reports for publication. This teaching model was used in both the U.S. and Australia and guided the syllabi for each of the dozen courses that I taught in these and other countries. As director of a doctoral program, this approach became a staple model for the curricula. A benefit was the resulting quality of many doctoral dissertations. My leadership roles in the International Association for Conf lict Management (IACM) provided a support community and collaborators for the ongoing research and practices. A more intense version of the teaching approach is skills training. A variety of professionals from around the world have matriculated in our workshops. They leave the sessions with a grounding in research knowledge and methods. We leave with new insights and evaluations of training effectiveness. Thus, teaching, training, and administration are part of a holistic perspective on knowledge creation and dissemination. They are nourished by the research and consulting endeavors on display in this book. It is the tangling of these activities through time and space that make this a unique career. This comes through in the way the story is structured rather than by any particular event or publication. Looking back in the twilight of a long career, I recall the initial impetus for becoming a social scientist. That motivation resided in a quest for change toward a better world. My first-year university social-science teachers – Ann Olmsted and Charles Kenney – struck a chord that I have been playing ever since. The chord acquired embellishments along the way as I became immersed in various research literatures. The research on negotiating f lexibility provides paths to agreements. Reading and writing about national identity ignited a consciousness of the pernicious consequences of arrogant nationalisms. And the body of work on justice made me less cynical about making progress toward equality. The realization, a long time ago, that my progressive values could be subjected to evaluation has sustained the career presented in this book. The usual acknowledgments go to many unusual people who participated in and supported this career. My wife, Marjorie, was there at the beginning and remained steadfast throughout. The book dedication to her is a small token of my appreciation. Our children, Kathy and Jamie, lived through the career as they developed their own social science careers. They made it all worthwhile.

xvi

Preface

A large global community of colleagues through the years contributed in ways too numerous to document in a Preface. Many of them appear in the book. It is my fervent hope that the career described in this book is deemed worthy. It has been my privilege to share it with all of you. Thanks to Taylor & Frances/ Routledge and the series editors for providing the opportunity. Special gratitude goes to Devon Harvey at Routledge for her careful attention to detail and support throughout the editing process.

FOREWORD

Professor Daniel Druckman is one of the world’s leading conf lict management and international negotiation scholars. During his highly distinguished career that has spanned nearly six decades, he has written widely on such topics as nationalism, the role of identity in human conf lict, peacekeeping, and the role of justice in conf lict management and resolution. For many years, Professor Druckman ran the doctoral program at the Institute for Conf lict Analysis and Resolution at George Mason University, where he was the Vernon M. and Minnie I. Lynch Professor of Conf lict Resolution. He was also concurrently Principal Study Director at the National Academy of Science in Washington, D.C., where he spearheaded many pathbreaking projects on international relations and research methods in the social sciences. This book assembles in one volume many of the seminal essays that Professor Druckman has written during his impressive and highly productive career while including some of his latest work on peace processes and the ingredients for a durable peace. At a time in world history when great power rivalries have re-emerged with a brutal vengeance, and the forces of nationalism and identity politics are resurgent, Druckman offers a powerful antidote to the conventional wisdom that there is little political space for diplomacy and negotiation to address the world’s conf licts. The chapters in this book offer a combination of keen scientific insight and practical advice on how conf lict managers can address identity conf licts and resurgent ethnocentric political forces through their interventions. His seminal work on the importance of timing and how to manipulate key turning points in negotiation processes is especially apt as diplomats and students of conf lict resolution seek to find new pathways out of irridentist conf licts and promote the conditions for a durable peace. The book is also essential for students of diplomacy and conf lict resolution because of the stress it places on addressing deep-rooted perceptions of injustice

xviii

Foreword

and historical grievance at the root of many identity-based conf licts. As Druckman demonstrates, “justice” is not just about achieving a so-called fair outcome, but also designing a negotiation process that all parties perceive to be equitable, inclusive, and fair. Many scholars are now studying political narratives or memes in negotiation and the obstacles they pose to mediators and other third parties trying to move the parties from their entrenched positions to secure a political settlement. Druckman’s work on the distinction between values, ideologies, beliefs, and interests is important. How these concepts play into collective narratives and belief systems is especially relevant to this new and evolving body of scholarship, not least because of the keen insights it offers on how best to lay the foundations for a new political narrative that can bring parties together. As a study of one prominent scholar’s intellectual journey, this volume also underscores the importance of interdisciplinarity, scientific rigour, and careful attention to research design for those who aspire to become leading social scientists. It offers a spirited and robust defence of behavioralist and empiricist methodologies in the social sciences – one that deserves widespread attention at a time traditional methods are under attack by post-modernist scholars who question the validity of the scientific method and causal relationships in social phenomena. It is a volume rich with insight and critical self-ref lection that deserves to be widely read and discussed. As is evident throughout the book, he is not content to rest on the laurels of being a multi-faceted scholar. He is equally concerned with applications. His forays into the worlds of policy and skills training attest to a strong yearning for improving policy and achieving more effective training. Both these pursuits have been informed by evidence. It is fascinating to see how he translates findings from his own and others’ research into negotiating and mediating strategies. Countless diplomats and students around the world have benefitted from these training sessions. It is also encouraging to read about his contributions to policy in the forms of policy briefs and briefings as well as occasional op-eds. Most recently he offered a proposal, with Paul Meertz, on how diplomats can negotiate a way out of the war in Ukraine. This is a career of a socially responsible scholar, one who understands the value of social science for making the world a better place for his own and the generations to follow. Fen Osler Hampson Chancellor’s Professor & Professor of International Affairs Carleton University Ottawa, Canada

ACKNOWLEDGMENTS

Grateful acknowledgment is made to the following sources for permission to reproduce material for this book. Druckman, D. (1968). Prenegotiation experience and dyadic conf lict resolution in a bargaining situation. Journal of Experimental Social Psychology 4 (4): 367–383. © Elsevier. Reprinted with permission. Inc. Reprinted with permission. Druckman, D. (1977). Boundary role conf lict: Negotiation as dual responsiveness. Journal of Conflict Resolution 21 (4): 639–662. © Sage Publications, Inc. Reprinted with permission. Druckman, D. (1995). Situational levers of position change: Further explorations. The Annals of the American Academy of Political and Social Science 542: 61–80. © Sage Publications, Inc. Reprinted with permission. Druckman, D. and Zechmeister, K. (1973). Conf lict of interest and value dissensus: Propositions in the sociology of conf lict. Human Relations 26 (4): 449–466. © Sage Publications, Ltd. Reprinted with permission. Druckman, D. (1986). Stages, turning points, and crises: Negotiating military base rights, Spain and the United States. Journal of Conflict Resolution 30 (2): 327–360. © Sage Publication Inc. Reprinted with permission. Druckman, D. (2001). Turning points in international negotiation: A comparative analysis. Journal of Conflict Resolution 45 (4): 519–544. © Sage Publication Inc. Reprinted with permission. Druckman, D. (2020). Thirty-five years of research on turning points: Insights gained and gaps to be filled. Negotiation Journal 36 (2): 91–106.

xx

Acknowledgments

©John Wiley & Sons Inc. Reprinted with permission. Druckman, D. (1968). Ethnocentrism in the inter-nation simulation. Journal of Conflict Resolution XII (1): 45–68. © Sage Publication Inc. Reprinted with permission. Druckman, D. (1994). Nationalism, patriotism, and group loyalty: A social psychological perspective. Mershon International Studies Review 38 (Supplement 1): 43–68. © Oxford University Press. Reprinted with permission. Druckman, D. (2019). Justice matters: Peace negotiations, stable agreements, and durable peace. Journal of Conflict Resolution 63 (2): 287–316. © Sage Publication Inc. Reprinted with permission. Druckman, D. and Wagner, L. (2016). Justice and negotiation. Annual Review of Psychology 67: 387–413. © Annual Reviews. Reprinted with permission.

INTRODUCTION

This book shares the accomplishments of a 56-year career as a social scientist. Key themes of the career are divided into three broad topics – negotiation, identity, and justice. These themes provide a structure for the book. The career retrospective in the first part takes the reader on a journey through each of these themes. Studies on bargaining and international negotiation are followed by the research done of issues of identity and justice. I then go behind the scenes of the research by telling stories about the people, places, and institutions that made it possible. The next five parts present important articles on each theme. Parts 2, 3, and 4 cover topics about negotiation. The first article in the second part on f lexibility was published in 1968. The last article in the fourth part on turning points appeared in 2021. The chapters in Part 5 on national identity include three articles, one of which was written exclusively for this book. The chapters in Part 6 on process and outcome justice includes recently published articles on the topic. Although there is more coverage of negotiation topics, the same intensity and devotion went into the research done on each of the themes. The concluding Part 7 consists of two chapters. The first looks back on how classical ideas shaped this career and at the value of the evidence-based approach that is a hallmark of all the research. The second chapter in this part looks forward by offering concepts that may give definition to the scholarship in the decades to come. These concepts are organized into five topics discussed in the preceding six parts of the book. The aim of the book is to give readers a sense of how the career developed, turned in various directions, and came full circle to where I started. The development was marked by shifts in topical focus as I moved from one type of institution to another. Research was enriched by consulting experiences and teaching in several countries. These changes were also turning points in an evolving career. Each move took me from one to another intellectual culture, as discussed DOI: 10.4324/9781003293361-1

2

Introduction

in Part 1. The full cycle is ref lected in a connection between the earliest and most recent publications. I began with research on bargaining f lexibility of union and management representatives in simulated collective bargaining. I picked up again on those themes in current research on resource allocation bargaining by simulated congressional representatives.

Negotiation The three chapters in Part 2 on negotiation f lexibility include research on prenegotiation experience, boundary roles, and situational levers. Together they demonstrate how various aspects of the situation facing negotiators constrain or broaden their decisions. They are constrained by strategic preparation, by the inf luence of constituents or principals, and by audiences eager to be entertained by the give-and-take of the negotiation process. They are less hindered or unbridled when their preparation consists of studying the issues, when their constituents are empathic, and when the talks are held in private. This line of research is captured in the articles that appear in a special issue of the Annals of the American Academy of Political and Social Science ( Druckman and Mitchell, 1995). It resurfaces in recent research on impacts of group or party loyalty (Druckman et al., 2022). The introduction to Part 2 discusses the early fascination with this topic. The more general discussion of philosophical inf luences is discussed in Part 1 and again in Part 7 in conjunction with classical writing. The f lexibility theme emerges as well in the chapters in Part 3 on values and interests. I developed a laser-like focus on the role of values in conf lict. My earlier research showed that different values can be a source of conf lict and a constraining inf luence on attempts to resolve conf licts. I was encouraged in my readings about the sociology of conf lict to go beyond these findings by considering the way that values interact with interests through the course of conf lict encounters. The article in Chapter 10 contributes to our understanding of these dynamics. An application of the propositions in that article is found in the case study by Druckman and Green (1995). While working on these articles, I was struggling with distinctions between values as ideologies and as beliefs. Scholarly work on ideologies is largely in the realm of sociology, while beliefs and attitudes are primarily studied by social psychologists. This can be thought of as a levels of analysis issue where ideologies are understood as collective narratives while beliefs are studied as individual systems. Our work on values and interests has moved between these concepts. We focus on ideologies in the case studies on negotiations to resolve civil wars, while beliefs are front and center in the experimental research on dyadic bargaining. Values are used as an umbrella for encompassing both meanings. As we consider next steps in this line of work, we will give priority to sorting out the similar and different implications of conf licts over ideologies and over beliefs. A third area that also overlaps the f lexibility theme is the research on turning points in Part 4, where we focus on changing the course of events to resolve

Introduction

3

impasses. Indeed, changing the course has also been a hallmark of my career as will be evident in Part 1. An interesting story about how I discovered the importance of turning points is told in the Part 4 introduction. That discovery led to a long stream of studies on this concept. The base rights article in Chapter 11 was the product of a consulting assignment where the aim was to provide advice. We analyzed the process in real time, reading transcripts of discussions between the U.S. and Spain delegations within days of when they occurred. During this reading I realized that there were several critical moments that changed the course of the talks. Following earlier work on conf lict, I referred to these moments as turning points. My advice to the U.S. delegation was to be alert to opportunities and the actions that could consolidate them. This was also a scholarly opportunity for me. I developed definitions for the turning points, the crises that propelled them, and the stage transitions that resulted. This case study ignited the spate of research to follow across several decades. An analysis of turning points in the Intermediate Nuclear Forces talks followed the base-rights study. This case included the famous 1986 exchange between Reagan and Gorbachev in Reykjavik Iceland. My experiences with these cases were convincing. I was confident that this concept had legs. However, at that point we did not know whether its value was case specific or more general. To evaluate generality, we developed a framework and proceeded with a comparative analysis of 34 cases. The results of our efforts are reported in Chapter 12. Distinctions made among types of precipitating factors or triggers were useful. They were central in understanding patterns in different bilateral issue areas as well as in multilateral trade negotiations (e.g., Crump and Druckman, 2016). Moments for ref lection on the state of this art occurred with the convening of a set of meetings at the Harvard Program on Negotiation (Leary, 2004). We took stock of where we were and what remained to be done. The research picked up in a number of quarters. Young researchers wrote dissertations on turning points; veteran researchers seized opportunities for new research on this topic. Another pause to ref lect on progress came with the writing of a chapter in the Handbook of Research on Negotiation ( Druckman and Olekalns, 2013). Connections were made in that chapter to other topics, including synchronized interactions, expectancy violations, resilience, and temporal horizons and spillovers. These connections stimulated the next wave of research, which led to a second conference at the Program on Negotiation. Insights from that conference are collected in a 2020 special issue of the Negotiation Journal. The lead article for that issue appears as Chapter 13.

Identity Research on issues of national identity – discussed in Part 5 – ran parallel to the work on negotiation. I was drawn very early in graduate school to identity issues. My time as a research assistant on the cross-cultural ethnocentrism project (LeVine and Campbell, 1972) was a gateway to learning about theory

4

Introduction

development and empirical research. My first project resulted in the Chapter 14 article. This article captures the keystones of learning in graduate school: the idea of evaluating competing theoretical hypotheses, simulation methodology, and statistical analysis. The results from this early study are fascinating, particularly with regard to the evidence for the possible universality of ethnocentrism. I also developed an appreciation for the importance of context. Perceptions of individual role players were assessed in the setting of simulated international relations. I did not join the band of social identity researchers by performing laboratory experiments stripped of context. I did however use the findings from their studies in writing reviews of the group bias and nationalism literatures. Two of these syntheses are a chapter on regional politics (Druckman, 1980) and an article on nationalism and patriotism presented in Chapter 15. Both of these pieces promote the value of social-psychological perspectives for understanding collective behavior and national policies in regional and international systems. These treatments carry forward my early learning about the importance of context. Further writing for two Anniversary issues of International Negotiation expand on the collective behavior and representational negotiation themes (Druckman, 2006a, 2015). Another contextual theme that runs through much of my writing on negotiation and identity is connecting levels of analysis. This includes work on structures and processes (Druckman, 2003a; Irmer and Druckman, 2009), turning points (Druckman, 2004), and processes, identities, and situations (Druckman, 2003b). The broadest and most complex treatment of levels of analysis is the original paper presented in Chapter 16 of this volume. Steve Wood and I explore various pathways toward the development of theories of national identity. We do this by first summarizing the research at each of three levels (interpersonal, intergroup, and collective) and then use examples to show how the dynamics of interacting levels work. Those dynamics are captured by the idea of pathways that move in three directions – downward from macro to micro, upward from micro to macro, and across with simultaneous effects. This paper is a culmination of the earlier work presented in Chapters 14 and 15. The chapters in Part 5 sample my empirical, review, and theoretical work on national identity. They convey an engagement with various contexts, an appreciation of the way social-psychological factors inf luence collective decisionmaking, and a window into the knotty issues of interacting levels. My aim for this part is to share the experiences of a social scientist grappling with complex issues using and, when necessary, refining methodological tools for solving the puzzles that emerge from these issues.

Justice The justice theme in Part 6 of this collection is of more recent vintage. Although the research on justice was conducted within negotiation contexts, our primary interest in the earlier justice research was on implementing agreements and on

Introduction

5

durable peace. Our focus in this work has been on procedural (PJ) and distributive (DJ) justice. We were impressed by the earlier research on these concepts, particularly the work done by Lind and Tyler (1988) on PJ and by Deutsch (1985) on DJ. Our research, discussed in Part 6, added two important findings to this literature. One was that the key aspect of PJ resided in social conduct, which combined fair treatment with transparency. Another was that the key to DJ was equal outcomes. The former turned out to be the strongest predictors of durable peace and reconciliation as shown in Chapter 17. The latter did the best job ensuring favorable implementation of peace agreements as shown by Albin and Druckman (2012). Practical implications are also suggested in our articles. One is to design a process that encourages the key elements of PJ to surface in the discussions. Another is to include equality provisions in the agreements. The review in Chapter 18 expands our lens on the broad justice and negotiation literature. Of particular note about this article is the use of a framework – similar to Figure 3.1 – for organizing the review. Justice considerations are important at each stage of the negotiation, from prenegotiation planning to implementing an agreement. The outputs of the review are also valuable. We generated 22 hypotheses for future studies and raised a distinction between linear and systemic perspectives for building theories of the way justice operates in negotiation. We conclude that article with a suggestion about how these perspectives can be used together. More broadly, the work on justice calls attention to building cultures of peace. Such cultures depend on the development and sustenance of trusting relationships. Social conduct and equal outcomes are the conduits through which trust operates. The book concludes with two chapters, looking back on what has been accomplished (Chapter 19) and looking forward on what remains to be done (Chapter 20). It is interesting to observe how ideas from the early social science classics inf luenced a considerable amount of my research output. Examples are Lewin’s (1951) principle of contemporaneity, Simmel’s (1955) interplay of ideologies and interests, the Sherifs’ (1956) idea of developing reciprocities for social norm formation, and Schelling’s (1980) focal points. It is also interesting to ref lect on the impact of graduate training on my obsession with evidence-based approaches to research and training. The nuggets of wisdom that sprung from this research career open areas to be mined by future generations. Five areas are identified in the concluding chapter. These areas are familiar topics within the broad themes of negotiation, identity, and justice.

Putting It All Together In making a transition from this introduction to the rest of the book, I ask a question from the song Alfie: “What’s it all about?” The short answer is that it’s about philosophy, science, and substance. The philosophical part comes from early exposure to behaviorism and empirical approaches in social science. Whether observing university students in experiments or diplomats in complex

6

Introduction

negotiations, the quest for understanding is the same. The scientific part is the structure provided by theoretically informed hypotheses and the methodologies used to evaluate them. Tensions between deductive and inductive approaches and between quantitative and qualitative ways of knowing have pushed me to strive for integrated knowledge (Druckman, 2008a). These scientific tensions are evident throughout the book. The substantive part refers to relations among the three themes. My initial work on identity issues (Chapter 14) led to a focus on the f lexibility constraints posed by being a negotiating representative (Chapter 8). The work on justice has been informed by findings from earlier work on identity and negotiation. Strong group or national identities can undermine seeking to fulfill justice claims. Taken together the three themes are interwoven to make the fabric of a career. Examples of other tensions are between internal (reliability) and external (generality) validity in simulation design (Chapters 9 and 14), doing case studies or experiments in conf lict research (Druckman, 2010), preferences for self-report versus behavioral or observational data (Druckman, 2005), linear and non-linear perspectives (Chapter 18), and micro or macro levels of analysis (Chapter 16). With regard to training, the challenge has been finding ways to incorporate research-based knowledge as a basis for skill development (Druckman, 2006b). These have been creative tensions, motivating me to find solutions that combine the approaches. They define my career but also provide lessons that will be useful to readers that are or aspire to become social scientists.

PART 1

A Career in Three Parts Research Institutes, Consulting Firms, and Universities

PART INTRODUCTION

In this part I take the reader on a journey through the terrain, including hills and valleys, of a career as a social scientist. In so doing I am practicing a form of scholarship known as prosopography, more in the tradition of taking account of the contributions made in an historical context rather than doing a critical analysis of them.1 Following Driver’s (1982) career typology, my trajectory is best depicted as spiral, which has taken the form of thematic coherence but variation in the way the themes have been pursued. As will become evident, I have worked in several types of organizations (profit, nonprofit, and universities) where I managed the vicissitudes in a way that allowed the pursuit of many of my projects alongside the organization’s tasks. I take a retrospective look at a career and the contributions that resulted. Three broad themes – negotiation, identity, and justice – emerged from a quest to understand human behavior in its social contexts. Early articulations of this quest were presented to my first-year graduate school professors in the form of a term paper on social stratification and cognition. I argued that peoples’ internal lives ref lected the external world in which they lived. Although not a profound insight, this was a key understanding that permeates most of the research to be presented. It has guided my study of the impacts of situations and structures on negotiation processes, of the role of political institutions in forging national identities, and ways in which principles of justice inf luence governance and the emergence of cultures of peace. These are guiding themes in my attempt to organize a wide-ranging career. I begin this part with a discussion of how the career was built and maintained. This is followed by chapters on each of the several themes. The earlier research on bargaining is separated from the more recent work on international negotiation. The bargaining chapter highlights the research streams on boundary roles, f lexibility, and electronic mediation. The international negotiation DOI: 10.4324/9781003293361-3

10 A Career in Three Parts

chapter covers a wide array of case studies and comparative analyses. The cases included negotiations on base rights, conventional troop reductions in Europe, arms control, and shuttle diplomacy in the Middle East. They also include a tense negotiation that occurred just after Aquino came to power following the Peoples’ Revolution in the Philippines. The chapter concludes with an attempt to rein in the wide-ranging research by proposing lessons learned. The fourth chapter is on national identity. It starts with my graduate-school project on ethnocentrism in a cross-cultural context. The key article from that project appears later in this book. My interest in understanding group attachments stemmed from that project and was also a subject of my dissertation. A stream of studies on representing groups are reviewed. I then switch gears from small groups to international relations and discuss work done on regional politics. In a final sub-section I address the puzzle of developing pathways to a theory of national identity that prioritize linkages between micro and macro levels of analysis. The fifth chapter presents research on justice. It begins with research on the role of justice in attempts to implement peace agreements. The more ambitious work – in terms of sample size, measurement, and modeling – on lasting peace follows. Then I turn to three other issue areas – trade, arms control, and the environment – where we examine how justice inf luences the effectiveness of agreements. The chapter concludes with a discussion on how justice is linked to negotiation. Continuous support from three funding organizations sustained this research for nine years. In a final chapter I take a more relaxed approach to writing about this career. I talk about what went on behind the more visible scenes. It is about people, places, and opportunities. My hope is that readers will learn about the persons and organizations that enabled us to make these contributions. They may also learn that there is more to a career than what appears in print. The body of research and theory discussed in this book covers around a third of my published work. The largest swath of these publications (35) is on negotiating f lexibility and group representation. These encompass the earlier publications, starting with my first article appearing in a 1967 issue of the Journal of Personality and Social Psychology. The major article on f lexibility was my metaanalysis appearing in a 1994 issue of the Journal of Conflict Resolution. Ten articles on values and interests appeared primarily during the middle years, starting in 1970 and culminating in a 2019 article with Fieke Harinck. Twenty-two articles on turning points ref lect an abiding interest in the topic as well as recognition in citations. This stream began with a 1986 article on base rights and continued through to a 2022 paper on automotive industry mergers. Work on national identity began with my second publication in 1968 but continues to occupy my attention with a paper completed in 2022, a total of nine publications. The more recent stream on justice has produced 19 articles including one to appear in 2023 for a special issue on the topic.

Part Introduction

11

Missing from this collection are articles on peacekeeping, nonverbal communication, human performance, training applications, and methodology. I will however weave these topics into the discussions to follow.

Note 1 Dictionary definitions of prosopography refer to the collective study of the lives of groups of actors in history. I consider my career in a social context where contributions are made possible by the larger social science community. My work is nourished by and feeds into the life of this community.

1 BUILDING A CAREER

I have spent countless hours at the desk where I now sit surrounded by more than a thousand books mostly written by social scientists over the course of more than 70 years. The cloistered life of a scholar is rewarded by affording the time to read widely and write wisely. My academic career was shaped by this library, although some books have had more impact than others. These include recognized classics and less vaunted but highly regarded contributions to their fields. The classics include The Human Group by George Homans (1950), Behaviorism by John Watson (1925), Georg Simmel’s Conflict & The Web of Group Affiliations (1955, translated by Kurt H. Wolf and Reinhardt Bendix), The Social Psychology of George Herbert Mead (1956 edited by Anselm Strauss), Muzafer Sherif ’s The Psychology of Social Norms (1936), and Kurt Lewin’s Field Theory in Social Science (1951, edited by Dorwin Cartwright). No less inf luential has been John Thibaut and Harold Kelley’s The Social Psychology of Groups (1959), Lewis Coser’s The Functions of Social Conflict (1956), and Erving Goffman’s The Presentation of Self in Everyday Life (1959). These books opened my eyes to an intellectual world of authors trying to make sense of social life. It also revealed differences among them in how they rose to the theoretical challenges of sense-making. For those of us who came along later, the challenges were to find ways of navigating these mind fields as we developed our own careers. This book tells the story of how I carved out a career that built on this rich legacy. This body of work was the impetus for a career as a social scientist. Many of these books were read during my first year of graduate school, which included a course with John Thibaut, who used his new book with Harold Kelley as the textbook for the class. I learned about the emergence of norms and roles in groups as well as the push and pull of social interdependence and conf lict. It became clear that humans are shaped by their contexts but are also agents who could create or change those contexts. These early learnings contributed DOI: 10.4324/9781003293361-4

Building a Career

13

to a view of social behavior that includes both adaptation and agency or activism, the two poles of human plasticity. Issues of global climate change are an example. From one political perspective, people are encouraged to adjust their life styles to inevitable changes in climate. For example, people are encouraged to move from island countries threatened with the prospect of tsunamis. From another perspective, people are encouraged to be in the forefront of protesting government and corporate policies responsible for the environmental changes, which are not inevitable. Compare in this regard the current U.S. Republican party platform with the activities of the leading climate activist, Greta Thunberg. In both cases humans are shaped by their larger environments. In the latter example, they are attempting to change that environment – a choice between adaptation and agency. These early years provided both a broad vision and the motivation to pursue a career in social science. The following graduate school years, spent at another university, honed my analytical skills and identified the areas where contributions would be made. The skills were those needed for experimental and related forms of quantitative research. The areas were research on group attachments and negotiation. The former interests were developed as a graduate research assistant on the “Cooperative Cross-Cultural Research on Ethnocentrism” project at Northwestern University ( LeVine and Campbell, 1972). That project focused on macro-level dynamics and is represented in this book in the article on “Ethnocentrism in the Inter-Nation Simulation” (Chapter 14). During the course of my project reading, I also became aware that many of the group dynamics being studied were also characteristic of micro-level processes such as negotiation. With an interest in both levels of analysis, I sought to understand the linkages between them as discussed in the original paper with Steve Wood in this volume (Chapter 16). My interest in negotiation peaked with the discovery of two books written in the 1960s. One was a 1965 book by Richard Walton and Robert McKersie. The other was a 1960 book by Thomas Schelling. Both books alerted me to the importance of strategy and tactics. The former made the distinction between distributive and integrative bargaining (raised originally by Mary Parker Follett in 1942) as well as the dilemma of dual bargaining with the other party and one’s own team. The latter book is a wellspring of ideas, including the value of focal points as a solution to coordination problems, commitment tactics and casuistry, tacit bargaining, and the role of contextual detail in the discovery of stable solutions. A number of my contributions are imbued with many of these ideas: notably, empirical distinctions between distributive and integrative bargaining processes and outcomes, a model of the boundary role dilemma, focal points and turning points, and the value of including context in my experiments. These ideas surface in laboratory experiments, comparative case studies, mathematical modeling, and simulation design. The studies ref lect the multi-method approach that I learned in graduate school, with Donald Campbell and Harold Guetzkow, and write about in Doing Research (Druckman, 2005).

14 A Career in Three Parts

My training taught me to be f lexible with regard to topics and the methods I used to study them. This f lexibility came in handy for a career marked by change and transition. The course I followed was not linear. Starting my career at a basic research institute in Chicago (9 years), I benefited from opportunities to beef up my résumé with plenty of publications, mostly on experiments. Moving to a world of research consulting in Washington DC (11 years), I learned how to do case studies on important international negotiations and on political stability of national regimes. The following 12 years at the U.S. National Research Council exposed me to a wide variety of projects and to outstanding social scientists from many fields. Realizing that the time had come to be a member of an academic faculty, I seized an opportunity to join George Mason’s (GMU) Institute for Conf lict Analysis and Resolution, where I directed their doctoral program. During my 20 years of academic life, I discovered Australia and split my time between GMU and several Australian universities (University of Queensland, the University of Melbourne, the Australian National University, the University of Southern Queensland, and Macquarie University). The versatility required to survive these changes was astonishing as I adapted to new work cultures, projects, and people. But it may also have been the case that I contributed to changes in those cultures – both poles of plasticity – as a social psychologist among political scientists in the Schar School of Policy and Government at George Mason, as an American at several universities in the Australian higher education system, and as a quantitative researcher in settings with applied missions, such as the National Research Council and DC-area consulting firms, or those with a prevailing qualitative ethos as at Australian schools of political science. This spiral career took unexpected turns at several junctures, particularly during transitions from one organization to another. Uncertainty competed with my enthusiasm as I moved from the familiarity of one work culture to the unfamiliarity of another. The most jolting revelations were the learning challenges that I encountered during each move. These included learning about child development at the Institute for Juvenile Research; learning about government consulting at Mathtech and political stability at Booz Allen Hamilton; mastering unfamiliar literatures on human performance, environmental change, and neuropsychology at the National Research Council; becoming a first-time middleage member of a faculty at GMU; dealing with marginality in a foreign country and its academic system; and becoming a political scientist at the Schar School of Policy and Government. Little wonder then that I gravitated toward studying the relationship between turning points and negotiating crises. Despite the exhausting road traveled, retirement has not been easy. Unlike the “old soldier” aphorism, old researchers do not fade away. They plough on. And so I have with a mix of continuing and new projects. I continue performing experiments on group loyalty and negative partisanship with colleagues in San Francisco and Brisbane. The line of research on peacekeeping missions with Paul Diehl has recently ramped up with a new book manuscript. Mediation by humans and robots continue to fascinate me, with collaborators in the Netherlands,

Building a Career

15

Vienna, and Aarhus, Denmark. New projects on the role of civil society in peace processes with Esra Cuhadar in Turkey, the impacts of face-threatening acts with Bill Donohue, and theoretical work on national identity with Steve Wood in Sydney are stimulating. On the applied side, working with my Brisbane colleagues on conf lict coaching keeps me grounded in the everyday world of conf lict. So then, research continues with a zest for tackling the questions being asked and working with the colleagues who keep me on my toes. With this discussion as background, I am ready to dig into the substance of the book. The key themes that weave through the writing are about bargaining and international negotiation, national identity, and social justice. The chapters to follow provide an overview of each of these topics, not as literature reviews but as windows into why each is a fascinating area of study and what we know to date. These discussions are intended to set the stage for the more specialized treatments to follow. Those treatments include selections of articles that show how the career unfolded and the contributions that were made.

2 LEARNING ABOUT BARGAINING

In the years following graduate school I was immersed in doing gaming experiments on bargaining. These experiments followed the work of Siegel and Fouraker (1960) on bilateral monopoly but also included resource distribution tasks. The challenge for our experimental bargainers was to balance mutuality and opposition of interests, which is a fundamental feature of managing or resolving conf licts. We learned about constituency constraints on representatives, impacts of an opponent’s concession-rate strategy, effects of non-reciprocating opponents, audience effects, types of justification for moves, salient solutions, time pressure, culture, age, and gender. More generally, we learned about how a large variety of situations inf luence the way people bargain. These findings are summarized in a review of the bargaining literature by Druckman (1971b). The insights gained continued to inform us as we moved forward to study more complex negotiating situations, including coalition formation and maintenance ( Druckman et al., 1974b).

Boundary Roles and Sources of Conflict The early experimental research was refined and elaborated in later work. Particular attention was paid to theorizing about such group representation effects as boundary role conf licts (BRC) (Druckman, 1977), pinning down the key inf luences in a meta-analysis of 80 studies of compromising behavior (Druckman, 1994), synthesizing diverse approaches, including behavioral and cognitive psychology, for deeper insights into decisions made to take unilateral initiatives ( Druckman, 1990), and understanding the difference between focal and turning points as salient bargaining solutions (Druckman and Rosoux, 2016). The empirical theme that runs through these studies extends the pioneering ideas contributed by Walton and McKersie (1965) on the BRC, by Coddington (1968) DOI: 10.4324/9781003293361-5

Learning About Bargaining

17

on expectation-evaluation-adjustment models, on the monitoring function in negotiation by Druckman (1978), by Blake and Mouton on group loyalty, by Osgood (1962) on unilateral initiatives, and by Schelling (1980) on focal points. They also prepared me for a next career step, which was to analyze cases of international negotiation. The bargaining research consisted of studying how people attempted to settle conf licting interests, usually over preferred material outcomes. Less effort was put into studying other sources of conf lict also dealt with through negotiation. One of these sources is the way people think about the problem to be solved. Referred to as cognitive conf lict, this source has been found to be very difficult to resolve, even when the disputants are aware of the other’s approach (Brehmer and Hammond, 1977; Druckman, 1970). Another source of conf lict is values. Invoking value differences can be a source of intransigence or a tactic for getting the other party to concede. Both stem from the popular saying that “you cannot negotiate values.” One line of research seeks to find ways of getting around the values (Druckman et al., 1988). When used as a tactic, negotiators attempt to place the burden of concession-making on the other party (Schelling, 1980). In addition to demonstrating the negative effects of value differences on agreements (Zechmeister and Druckman, 1973), we extended the early theoretical work on an interplay of values and interests through time and repeated interactions among conf licting parties (Druckman and Zechmeister, 1973).

Flexibility To this point I have focused on factors that inf luence the bargaining process and outcomes, referred to as independent variables. I have said less about what is being inf luenced – the dependent variables. In all of the experiments reviewed, we attempted to measure bargaining f lexibility as perceptions, decisions, and outcomes. In some studies this means moving from initial stated positions toward agreement, which usually takes the form of compromise or settlement involving unequal concessions (Druckman, 1994). In other studies f lexibility refers to new ideas or creative thinking (Druckman et al., 2021), which gets bargainers closer to resolution. The key distinction is between distributive and integrative processes and outcomes as defined by Walton and McKersie (1965), but also in an application of these concepts to paradigms of international relations by Hopmann (1995). Viewed more holistically, f lexibility can be understood in terms of negotiating stages where different variables have more or less impact in the earlier or later periods, for example, more f lexibility for studying the issues in the earlier stages and for limited media attention in the later stages (Druckman, 1993). Viewed from a wider analytical lens, we recruited scholars from several disciplines to contribute to a special issue on f lexibility that takes into account the mix of parties, structures, processes, and issues that come into play in any complex conf lict ( Druckman and Mitchell, 1995).

18 A Career in Three Parts

Electronic Mediation The focus of the early research on bargaining was extended to study mediation. But rather than simply adding more experiments to that literature, we thought we could provide a novel approach in the form of an electronic mediator. This work began with a model of mediation that consisted of three parts: diagnosis, analysis, and advice. With support from the United States Institute of Peace (USIP), we incorporated this model into a computer screen system known as Negotiator Assistant (NA) (Druckman et al., 2002). System impacts were evaluated in a sequence of experiments, which included a comparison of the screen system with a human mediator who also provided a diagnosis, analysis, and advice. Data were collected by my collaborator James Druckman at the University of Minnesota and at the University of Maryland in Jonathon Wilkenfeld’s laboratory. We found that the screen system produced more agreements but fewer integrative agreements than the human mediator. We also discovered that negotiators were more comfortable with the human mediator (Druckman et al., 2004). Advances in technology enabled us to move from a synchronous (same point in time) to an asynchronous system where negotiations could be conducted with and without a mediator across time zones and locations. We called this system VienNA. This second generation of research showed strong effects for the e-mediator, including more reciprocation of offers, advantages from early use of the system, fairer solutions, and more discussion of the sources of the conf lict ( Druckman et al., 2014). We took a break from the e-mediation research for a few years. A call from a colleague, Johanna Seibt in Aarhus, Denmark, came in out of the blue. She asked if I would like to join her team in preparing a proposal on robotics. My curiosity got the better of me. I agreed to be part of the submission. The grant was approved by the Carlsburg Foundation for a project with wide international participation. My part was on robotic mediation. We put together a team consisting of both Vienna and Denmark colleagues. A classical experimental design consisted of a comparison of screen, human, and robot mediation conditions along with a no mediation control. The negotiating task was to resolve a de-merger of a pharmaceutical company. All seemed straightforward until we became aware of a variety of technical challenges around robot operators, gender-neutral voices, and scripting the human and robot mediators. After a number of false starts and considerable pilot work, we were ready to begin the data collection. Recruitment of participants (508 in all) and data collection in the Aarhus laboratory took more than a year to complete. Another year of analyses and group writing emanated in a 2021 publication with seven authors in Group Decision & Negotiation (GDN), the home journal for the electronic mediation stream of experiments (Druckman et al., 2021).1 The findings justified the labors of running the experiment. Negotiators with robot mediators got more and better agreements than those in the other conditions. They also had more positive views

Learning About Bargaining

19

about their mediator. Yet the most interesting finding concerned non-standard agreements (NSAs), by which we mean novel solutions to the demerger problem. A third of the agreements were NSAs with a significantly larger number of them occurring in the robot condition. In fact, practically all of the integrative NSAs were accomplished by the robot mediator. The NSA findings suggest two processes at play. One is divergent thinking, the other is creativity. NSAs are by definition an example of a divergent or nonlinear approach to the problem. Integrative agreements are, also by definition, more creative than distributive agreements. Our conjecture was that these are responses to a novel experience. None of our experimental negotiators, or anyone else, have experienced a robot mediator. Although there is some evidence on the effects of novelty on creative thinking, due perhaps to “mind wandering,” we are just beginning to probe this explanation in current analyses.2 The findings do however have the heuristic benefit of opening up new streams of research on mediation with and without robots. More generally, we can say that we have come full cycle with this research. We entered the field of mediation in the late 1990s with an impetus to contribute novel research findings or even paradigms. We conclude the stream, at least to this point in time, with an insight about the role played by novelty. The discussion in this chapter is a whirlwind tour of three decades of research. Each of the topics covered tells a story about the way that bargaining occurs and reveals the many factors that inf luence it. The focus on analysis needs to be complemented by synthesis in the form of a framework that ties the parts together. The chapter written by Sawyer and Guetzkow (1965) came to the rescue. Organized into five parts, their framework provides a holistic rendering of negotiation. It separates inf luences from background factors and processes. Most of all, the framework is adaptable to a wide range of topics, including the distinction between settlements and resolutions (Druckman, 2002), the role of justice in negotiation (Druckman and Wagner, 2016), and the relation between nationalism and war (Druckman, 2001b). I will return to this framework in the next section on international negotiation.

Notes 1 This article received the 2021 Gregory Kersten GDN Journal Best Paper award. 2 Our most recent analysis of the negotiation transcripts from the Druckman et al. (2021) study suggests preliminary support for an alternative explanation. Conversations during the talks showed significantly more statements indicating openness to new ideas and togetherness for negotiators in the robot than in the human mediation condition. Thus, flexibility and cooperation may be a more promising avenue than creativity for further research.

3 TACKLING INTERNATIONAL NEGOTIATION

The year was 1971. The big event for my family was awaiting the birth of our son Jamie. The career was on course with a f lurry of experiments being conducted at the Chicago-based research institute mentioned in the previous chapter. A career-changing invitation to attend a conference arrived. It was in Greenwich, Connecticut, on resolving international disputes sponsored by the Academy for Educational Development. I planned to attend the conference until it became apparent that our son would be born at the same time. Nonetheless, the organizers asked me to prepare a review of literature on the social-psychological aspects of international conf lict. This review was intended to update the Sawyer and Guetzkow (1965) chapter. I accepted the invitation with enthusiasm, put my experimental work on hold, and spent six months reviewing the literature and writing the paper. Published initially as the second Academy paper on Resolving International Disputes, the report went through a peer review process and became a monograph in the Sage International Studies Series (Druckman, 1973). This publication marked the beginning of a career that ran parallel to the experimental work on bargaining. It also prepared me for the research consulting jobs that followed nine years at the Chicago-based Institute.1 My early graduate school years exposed me to some facets of international relations through the prism of the Inter-Nation Simulation. These facets were alliance formation and change, role differentiation at the ministerial level, national identity, the limits of conceptualizing a nation as a small group, and simulation validity. I learned less about the negotiating process in the context of international relations. That learning occurred with my literature reviews in preparation for writing the 1973 Sage series monograph. The Sage monograph is notable for three contributions. One is the framework for organizing the review shown in Figure 3.1. As I noted previously, DOI: 10.4324/9781003293361-6

Tackling International Negotiation 21 Antecedent

Concurrent

Goals

Process

Communality of Goals Specificity of Goal

Background Factors Cultural and Cognitive Differences Attitudes Between Parties Relations Within a Party The Negotiator Role Personality Situation vs. Role vs. Personality

FIGURE 3.1

Preparation for Negotiation Bilateral Agenda Formation Prenegotiation Position Formation The Process: Models and Experiments Bargaining Type of Communication Opponent’s Concession Rate Influence Tactics: Offers, Threats, and Commitments A Cognitive-Cultural Model

Consequent Outcome Type and Magnitude of Incentive Range of Possible Outcomes Creating New Alternatives

Conditions Open vs. Secret Displomacy Number of Negotiating Parties Coalition Formation Mediational Mechanisms Information Concerning Another’s Values Stresses and Tensions Upon Negotiations

The Sawyer-Guetzkow Model of Negotiation

the framework provides a holistic perspective useful for organizing literature reviews. Another is the propositional format used for summarizing the review on each topic, a total of 70. The propositions serve two functions: summarizing research findings and raising hypotheses for further evaluation. For example, the more limited the scope of the issues discussed, the more likely that a settlement will be attained. A third contribution is the further development of Iklé’s (1964) models of negotiating objectives shown in Figure 3.2. Negotiating objectives set in motion particular types of processes, outcomes, and inf luences (conditions and background factors). The three models link negotiating behavior to the international context. In order to bring that context closer to the negotiating process, I added a section on a systems perspective, drawing mostly on work done by international relations scholars at a macro level. Much of my research ref lects the interdisciplinary training of my graduate programs: in sociology, social and experimental psychology, and international relations. The strong research methods training that I received in each of these fields honed my analytical skills. The immersion in different intellectual cultures led to a strong desire to integrate or synthesize. Both analysis and synthesis are represented in this book. My graduate research assistant work on the “Ethnocentrism project” gave me direction on what to analyze or synthesize. The transition from a focus on laboratory studies to a world of international politics occurred because of opportunities that landed in my lap – the AED project and the move to government consulting. And, yet, there was also a fascination in being part of activities that were on a world stage and meeting some of the actors in those dramas. The excitement came initially from reading books on international negotiation and then sky-rocketed with my consulting involvement in ongoing cases.

22 A Career in Three Parts Model I Objective

redistribution

Process

bargaining, influence tactics

Outcome

limited range including concession, compromise

Conditions

a wide range of situational variables as gleaned from experimental bargaining and influence studies Model II

Objective

innovation

Process

debate, communication, problem-solving

Outcome

expanded range of alternatives including creative, “integrative” solutions

Background Factors

culture, cognitive structures, ideology

Model IlI Objective

normalization, extension

Process

bargaining and debate

Outcome

expanded range of alternatives with some limitations imposed by previous situation

Conditions + Background both types of variable have an influence Factors

FIGURE 3.2

Three Models of International Negotiations

Base Rights, MBFR, and START I was struck initially by the untidiness of the negotiation cases I was assigned to analyze. This impression came from interviews with present and past civil servants and political appointees to important negotiations usually held in such international cities as Geneva, Vienna, Moscow, Madrid, Paris, and Washington, DC. The sensitivity of these negotiations prevented first-hand observation. As a consultant, however, I became more of an insider with access to transcripts of the U.S.–Spain base rights talks in Madrid and Washington, DC (1975–1976) and to the Mutual Balanced Force Reductions (MBFR) alliance negotiations in Vienna (1977–1980). The key benefit of that access was that I was able to pierce the veil. This was essential not only for the analyses I was asked to perform, but also to get a glimmer of some things that happened behind the scenes.

Tackling International Negotiation 23

With regard to the base-rights talks, I worked with a frustrated U.S. delegation headed by Ambassador-at-Large Robert J. McCloskey. His Spanish counterpart was Hon. Juan Rovira. McCloskey’s team expected a routine two-round negotiation that essentially updated terms for renting the four U.S. military bases on Spanish territory. It turned out that these talks were anything but routine, with a roller-coaster ride of ten rounds through a year and a half. Rovira and his team caught the U.S. delegation off guard with near-outrageous demands. They held the base renewals hostage to NATO membership. McCloskey, in a desperate attempt to satisfy Spain’s demands, appealed to the NATO council in Brussels. The answer was no dice, at least until Spain reformed its human rights abuses record. The drama continued with more fits and starts over the course of a year and confusion between the Spanish delegation and their Ministry of Foreign Affairs and Defense. Franco’s death in January 1976 threw the Spanish government into chaos. Without direction from above, their delegation agreed to a Treaty of Friendship and Cooperation that largely favored the U.S. But all was not done. On the U.S. side, Congress entered the fray with the treaty ratification process. Human rights voices from Congressional representatives, notably representative Bella Abzug of New York, held the treaty in limbo. An exhausted Robert McCloskey pleaded that he “would move heaven and earth if needed to get the treaty ratified.” Apparently the planets moved, and the treaty was ratified on January 24, 1976, in Madrid. There was also a mix of domestic and international politics in the MBFR alliance negotiations (Hopmann, 1977). On the domestic sides of both alliances, there was the push to limit the other alliance’s conventional troops and weapons inventories and the pull to maintain the current ceilings on their own inventory. A way to satisfy both political desires was to negotiate mutual reductions in “good faith.” The talks went on without agreement for 15 years. But the long process had another advantage. We, as analysts, were able to learn about the unfolding dynamics of alliance negotiations. The primary insight from content analysis was leadership (Moscow and Washington) control of their respective allies and cooperation with each other, a process referred to as a bilateral condominium. The alliance leaders reciprocated each other’s cooperative moves. Members of both alliances reacted to their leader’s cooperative moves in a previous round with competitive moves in the next round. Less clear was whether this was a planned tactic or a reaction to a perceived sell out by the leader. Further analyses charted changes in alliance cohesion and areas where compromises were feasible. (For more on these analyses, see Druckman and Hopmann, 2002.) Yet the areas of possible compromise were overshadowed by shared domestic agendas, which I address later. In December of 1983, the Soviets abandoned the Strategic Arms Reductions Talks (START) protesting the deployment of intermediate-force nuclear weapons in Europe. The U.S. was eager to get these talks back on track. This event provided another consulting opportunity with the Arms Control and

24 A Career in Three Parts

Disarmament Agency (ACDA). I signed on for a six-month stint in the ACDA offices. With my content analysis skills honed by the base rights and MBFR experiences, I plunged into an analysis of the conversations between the delegations to the point of walk out. The coding categories were designed to uncover Soviet objectives and motivations. What I found was an interesting difference of approaches for the two national delegations: The Soviets took into account the balance of all forces, strategic and tactical; the U.S. compartmentalized strategic and tactical weapons, each considered in different negotiating venues. My recommendations to the ACDA director (Kenneth Adelman) were to engage the Soviet delegation in a more comprehensive discussion of arms control. My immediate reward was lunch in the State Department dining room near the Secretary’s office on the eighth f loor where I met Paul Nitze. My larger reward was that the Soviet delegation returned to the START table in 1985 to resume talks that eventually led to an agreement in 1991. Unlike MBFR, the SALT/START process was a serious negotiation intended to carve out a bilateral agreement ( Newhouse, 1973). The detailed accounts of what went on in specific cases opened my eyes to the roles played by domestic bureaucracies in setting agendas or even spoiling the progress made. Bureaucratic actors may have their own priorities that are pitched often with enthusiasm to the delegates. An overwhelmed delegation may be all too happy to defer to the bureaucrats with expertise in weapons systems or on force balances. This worked well at Reykjavik when Reagan proposed to Gorbachev a double zero solution for controlling the arms race. The State Department and Intelligence Agency civil servants called a “time out” during which they persuaded the President to retract that proposal. Thus, the bureaucracy can be a check on the whims of Presidents and political appointees with much less knowledge about the technical issues. Another process that helped with progress at SALT was the backchannel communications between Kissinger and Dobrynin. Although often thought to be critical in formal negotiations, we have had difficulty gauging with any precision the impacts of these informal discussions, largely because they are secretive. But it is also the case that many on-the-side meetings occur in parallel with the talks, particularly between legislators, such as Senator Henry Jackson in the SALT process, their staffs, and the delegation.

Arthur Lall and Fred Iklé Lall’s 1966 book on Modern International Negotiation provides a view of international negotiation from the perspective of a UN diplomat. As a result we learn about the roles of procedures and international law in the context of multilateral or global negotiations. These types of negotiations are more rule-driven and, in a sense, more predictable than bilateral talks that occur outside of international organizations. Yet despite the institutional overlay, these negotiations are neither

Tackling International Negotiation 25

hum drum nor of little consequence for the course of international relations. Indeed, because of their global reach, they are likely to have larger consequences than bilateral or trilateral talks initiated by governments. An example is the Eighteen-Nation Committee on Disarmament (ENCD). Formed in 1962 and subsumed by the Conference on the Committee on Disarmament (CCD) in 1969, this forum was effective in producing the Treaty on the Non-Proliferation of Nuclear Weapons. The ENCD was small enough to be efficient in producing results but too small to be representative of the larger organization. Thus, later renditions, leading to the current CCD, were much larger. And, yet, despite being bogged down in protocol and traditional diplomacy, Lall showed how these fora provided windows on the wrangles between the U.S. and USSR. Of particular interest was the conciliatory behavior of the Soviet delegates. Later research on summitry showed that Soviet, and later Russian, public cooperation was offset by their aggressive involvement in conf licts around the world (Druckman and Wallensteen, 2016). For a discussion on the summitry research, see the podcast with the authors and journal editor at the website of Global Summitry: Politics, Economics, and Law in International Governance. The early books made it clear that negotiation is another stage for playing out the games of international politics. My consulting assignments gave me firsthand experience with these machinations. Both sources buoyed my confidence about the role of social science as a source of research methods and practical insights. They also provided direction for the rest of my career. The following three examples show how empirical research demonstrates the value and extends our understanding of the earlier ideas. One example is a test of Iklé’s typology on negotiating objectives (Iklé, 1964). His five objectives include extension (continuation of normal), normalization (termination of abnormal), redistribution (new distribution in favor of the offensive side), innovation (new institutions), and side effects (continuation without agreement). They come to life in his examples of historical negotiations that exhibit the features of each objective. However, his approach falls short of a systematic evaluation of its usefulness. Realizing the value of the typology and the lack of empirical evaluation, we rose to the challenge. The question asked was this: Do the five objectives distinguish among a sampling of negotiation cases? We moved ahead to address the question. Thirty cases drawn from the set compiled by the Pew Case Studies in International Affairs were coded on six aspects of negotiation: parties, issues, timing, process, negotiating environment, and outcomes. The results of statistical scaling analyses answered our question. The cases were distinguished by their objective (see Figure 3.1 in Druckman et al., 1999). This means that, for example, the eight innovation cases had a different profile than the ten redistribution cases (see Table 4 in Druckman et al., 1999). Iklé’s framework is a useful way for organizing a sample of diverse and complex cases. Only four cases did not fit into his categories. These outliers had in common their focus on negotiating multilateral regimes, a new category that emerged since Iklé wrote his book. This study shows just what social science has

26 A Career in Three Parts

to offer. Fred Iklé agreed in a note to me dated May 12, 1999. He was pleased to see validation of his framework, and I was pleased to get his note. In addition to his typology of negotiating objectives, Iklé taught us about negotiating for side effects and about the dynamics of the continual three-fold choice. The side effects argument made clear to me that agreement may not be the primary objective. Negotiation may be a tool that serves such other purposes as maintaining contact, avoiding violence, collecting intelligence, and affecting decisions made by third parties. The most poignant example of side effects is the MBFR talks between NATO and the Warsaw Pact. These talks dragged on for 15 years without an agreement. It was replaced by the negotiations on Conventional Armed Forces in Europe, or CAFE, conducted within the framework of the Helsinki process. A consulting assignment brought me, along with my colleague Terry Hopmann, into the later stages of the MBFR process described earlier. We were asked to come up with a proposal that could resolve the long deadlock. Our suggestion for trading ground forces (a Warsaw Pact advantage) for tactical nuclear weapons (a NATO advantage) seemed plausible but fell on deaf ears. Little did we know at the time that neither alliance was negotiating for an agreement. Rather, they were using the talks to prevent unilateral troop and weapons reductions from either side. We were blindsided but learned about the importance of side effects (see Druckman and Hopmann, 2002). The continual three-fold choice idea recognizes that negotiators are evaluating and re-evaluating their prospects for agreement throughout the process. It is captured as well by the economist Coddington (1968) in his model of expectations, evaluations and adjustments. These ideas challenge simpler models of reciprocity in negotiation. They recognize the process of monitoring trends in offers and counter-offers as well as judging the relative size of concessions made by both (all) sides in a negotiation, and they inspired my work on modeling how international negotiators respond to each other over long periods of time. In this work I recognized that monitoring or keeping track was an important function in negotiation (Druckman, 1978) but that comparison was more important ( Druckman and Harris, 1990; Stoll and McAndrew, 1986). The best model, in terms of fit to the data, was referred to as comparative reciprocity, a more complex form of responding than tit-for-tat (Axelrod, 1980). More interesting perhaps is that we evaluated these models with data from a number of international negotiation cases. I will have more to say about this study in the later chapter on Turning Points. The anecdotal approach to understanding international negotiation is also a hallmark of Lall’s (1966) treatment of the subject. He draws on his participation in numerous international conferences, particularly during the period when he was India’s Permanent Representative and Ambassador to the United Nations (1954–1959). He frames his sweeping coverage of events with summary principles, noting that despite the contradictory postures of governments, international disputes are negotiable. This is due largely to a focus by governments on their

Tackling International Negotiation 27

interests rather than on their ideologies. I regarded these experiential observations as hypotheses. They encouraged thinking about the impacts of ideologies or beliefs on negotiation in many contexts, including political decision making, prison system reform, urban problems, ecumenical councils, Cyprus-type conf licts, bar owner–customer conf licts over price increases, and negotiations between the Aquino government in the Philippines and the National Democratic Front insurgency. In general the findings across these studies support Lall’s hypothesis that interests matter more than ideologies. But they also show that ideological differences can spoil a negotiation over interests by heightening the conf lict. This finding led us to design interventions that reduce the impact of ideologies in negotiations as shown by Harinck and Druckman (2017). The examples of projects in the previous paragraphs describe the social scientist’s craft. That craft consists of evaluating ideas from an earlier literature in a systematic manner for their more general value or relevance to theories. No doubt the ideas are interesting and even inspiring. Less clear is what qualifies as evidence supporting the ideas. Each of the books mentioned earlier – by Iklé, Lall, and Newhouse – uses diplomatic conversations or decisions as ways of illustrating a concept, such as objectives, shifting evaluations, or impacts of national interests.2 The problem is that the examples were chosen to elucidate the concept, a practice referred to as cherry-picking. Social science methodologies provide a correction on that practice. They do so in several ways. One is by separating the ideas, often stated as hypotheses, from the procedures constructed to evaluate them. Another is to perform the evaluation on randomly chosen cases that represent a larger population. A third is to place a particular study in the context of a broader literature, showing a need for the study and its contribution to future studies. This is referred to as cumulation. This approach to knowledge creation is demanding because it is rigorous and requires considerable patience to pull off. Thus there is a lag between the initiation of study and its publication. The lag has been reduced with the development of big-data computer technologies in recent years.

Shuttle Diplomacy Lall, Iklé, and Newhouse provided early inspiration for my focus on international negotiation. Other sources and opportunities sustained the interest. One opportunity was a project on the 1973 Kissinger Middle East shuttle diplomacy. My contribution with Terry Hopmann was part of a larger project initiated and implemented by Jeffrey Rubin in cooperation with the Society for the Psychological Study of Social Issues. Our key insight is that Kissinger’s attempts to negotiate a peace agreement in paired negotiations between Israel and Egypt and between Israel and Syria succeeded at a tactical but not at a strategic level. The agreement occurred, but durable peace in the region based on balance of power principles did not follow (Hopmann and Druckman, 1981).

28 A Career in Three Parts

Kissinger became enamored with his role as mediator and, as a result, overlooked the grand strategy he had formulated going into the negotiations. His step-by-step approach to mediation facilitated concession exchanges but also increased dependence on the United States. The lessons learned from this case are helpful for mediators as well as for the art of shuttle mediation. They also call attention to the dangers of decoupling tactics from overall strategies, a lesson that has informed my later work on the role of justice in durable peace. In that project, which I discuss later, the durable peace agreements were those that emerged from adherence to a larger framework where justice rather than a balance of power was the centerpiece.

Intermediate Nuclear Forces A move from the world of government consulting to the National Research Council (NRC) occurred in 1986. I directed a study on enhancing human performance and served as a senior staff officer for a committee on behavioral and social science contributions for the prevention of nuclear conf lict. In the latter role I had an opportunity to study the Intermediate-Range Nuclear Forces (INF) talks between the Soviet Union and the United States. The study, done in collaboration with Jo Husbands and Karen Johnston (Druckman et al., 1991), consisted of an analysis of the negotiating process that produced a treaty in 1987. The analysis was based on interviews with 15 members of the U.S. delegation and its supporting staff. The information provided by them led us to frame the negotiation in terms of a series of turning points (TPs). Three presidential summit meetings were the key TPs. Each removed an obstacle to an INF agreement. The first (Paris, 1983) led to a decision to remove French and British nuclear forces from the INF talks. The second (Geneva, 1985) separated the START talks on strategic from tactical nuclear weapons discussed in the INF talks. The third (Reykjavik, 1986) resulted in a principled agreement that set the stage for discussion of trade-offs to follow. Throughout the discussions, we see the negotiators struggling with Iklé’s (1964) three-fold choice. Each TP consisted of a decision that kept the talks in train until the final decision was made to agree. The treaty was signed in Washington in December 1987. This case heralds the value of summitry. But it is also likely that summitry worked because of changes in world events brought about by General Secretary Gorbachev’s unilateral moves to reduce force levels leading to the fall of the Soviet Union.

Foreign Service Institute In the early 1980s, before joining the NRC, I was involved in a Foreign Service Institute (FSI) project on lessons learned from cases studies of diplomatic negotiation. With the sponsorship of the FSI, Ambassador John

Tackling International Negotiation 29

McDonald convened a panel of academic scholars to analyze selected cases of negotiation. The format consisted of public seminars with members of the negotiation delegations followed, a week later, by half-day panel discussions intended to analyze the case from a variety of perspectives. The first set of cases consisted of the Panama Canal Treaties, the Falkland/Malvinas Islands Crisis, the Cyprus Dispute, and Zimbabwe’s Independence. The product from these seminars and analysis sessions was a book in which commentaries from the negotiators were complemented by interpretive essays and a lessons learned chapter from members of the academic panel (Bendahmane and McDonald, 1986). The lessons learned chapter was my attempt to provide key insights across the cases. I summarized a discussion of processes and inf luences in my accustomed form of propositions, 17 in all. The number and variety of lessons attested to the value of these exercises. A concluding discussion highlighted the new insights about formal and informal diplomacy, procedural f lexibility, conditions for negotiation, and tactics. Examples include the following: • • •

Formal negotiations encourage parties to adhere to official positions, serving to reinforce competing claims for a settlement. Shuttle diplomacy works best during prenegotiation phases and during crises; direct face-to-face talks work best when an agreement is being shaped. The decision to negotiate a treaty can help in sustaining the agreement; it confers legitimacy on the outcome.

These are insights recovered from past experience. As such they need to be evaluated as hypotheses with material generated from other cases. But they also alert practitioners to possible pitfalls and aid planning for future cases. More generally, this project led to the realization that those of us who study international diplomacy must paint on a very large canvass. It comes in many varieties and is particularly sensitive to the political context within which the talks are conducted. The project also introduced me to many of the most important diplomats and scholars in the world. My brief conversation with Ambassador Ellsworth Bunker, who began his diplomatic career in 1951 as Ambassador to Argentina (appointed by President Truman) and concluded it in 1977 as chief negotiator for the Panama Canal Treaties, was a highlight of my experience on the project. The four cases analyzed showed considerable variety. They come from different continents, dealt with different issues, and occurred during different time periods. The next FSI project dealt with three cases on the same issues occurring during the same time period. They were negotiations with the United States over base rights issues during the Cold War with Spain, Greece, and the Philippines (McDonald and Bendahmane, 1990). The same format of seminars and analyses was followed. Once again I wrote the lessons learned chapter in proposition form and benefited from my previous research on the Spain base rights talks.

30 A Career in Three Parts

These cases produced a number of similar and different lessons than those that emerged in the earlier FSI study. Here are examples of repeated lessons: •

•

Similar advantages were found for f lexible formats. Moving between a shuttle and direct format in the earlier cases; moving between working groups, private sessions, and direct formats in the base-rights cases. Both studies emphasize the importance of early moves, of preventing contentious issues and details from accumulating, and of developing credible alternatives to negotiating agreements.

Here are examples of new insights from the base-rights cases: • •

These cases highlight the tactical use of language, the decision dilemma in the face of a deadline, and the tactical use of parallel negotiations. Different tactics to avoid jeopardizing an agreement are emphasized: treaties and commitments in the earlier cases; Senate support and a good relationship between the sitting ambassador and the prime minister in the baserights cases.

Another important difference between the sets of cases is the magnitude of the issues for negotiating delegations. The earlier study chose cases more broadly construed as conf lict management. The base-rights cases were more focused on resolving very specific, mostly technical, issues. However, as noted previously, base-rights can be used as leverage to gain concessions on larger issues such as admission to NATO for Spain. More broadly, the unavoidable dilemma for negotiators in all these cases is the difficulty of separating negotiating challenges from the political context in which they are framed. As our panel worked on these cases, we were aware of the sensitivity of many of the issues that arose. A dilemma for researchers in this field is the availability of information. Despite the public discussions that were held, there were matters of tactical choices that governments preferred not to publicize. Not knowing what these issues were left us vulnerable to “whistle blowers” who winced at some revelations in our book. These sensitivities surfaced well after we went to press but were resolved. This sort of drama is uncomfortable territory for scholars who strongly advocate for open access. It no doubt remains a challenge for those who choose to analyze real-world cases of negotiation.

Aquino and the National Democratic Front An analysis of a two-week negotiation that took place shortly after the Philippines Peoples’ Revolution provided insights into the forces that bring disputing parties to the table and keep them there. The former analysis was based on the parties’ calculation of relative power and legitimacy (Druckman and Green, 1986). The latter was an evaluation of the propositions about conf licting values

Tackling International Negotiation 31

and interests developed by Druckman and Zechmeister (1973), and included in this book. Together, these analyses show how the Aquino regime and the National Democratic Front (NDF) played two games in this important internal negotiation (Druckman and Green, 1995). In the first game, the contending parties agreed to come to the table when the government’s power vis-à-vis the NDF was increasing, and their legitimacy was stable. We referred to this situation as a ripe moment. Although both parties continued to monitor the changing situation during the talks, they turned their attention to the issues that divided them. These issues consisted of differences in both ideologies and interests. The NDF’s proposals for fundamental societal changes were rejected by the government. The government’s proposal to use the draft constitution as a framework for the talks was rejected by the NDF. Issues of power sharing, an integrated military, and land reform were non-starters. With little progress being made on these issues, moderates on both sides moved away from the center, creating further polarization and a return to violence. This analysis demonstrated the relevance of several of the concepts discussed earlier in this section and in the broader literature on managing conf licts. The two-game idea captures a dynamic where parties assess power balances for readiness to negotiate and then judge their progress in promoting ideologies and interests. Both processes involve monitoring the other side for movement and one’s own side for signals on positioning (Druckman, 1977). This dynamic is captured as well by Iklé’s (1964) three-fold choice and by Coddington’s (1968) expectation-evaluation-adjustment cycles. While keeping track of developments and adjusting to changes, negotiators are switching from one negotiating objective to another. In the Aquino-NDF case, this consisted of seeking a redistribution of resources between welfare and military programs. Both sides also sought to be innovative by altering their visions for the Philippines. Although the negotiation ended abruptly, there were advantages in keeping the talks going to accomplish side effects: For the government, these were to reduce the threat and attract more left-of-center supporters; for the NDF, these were to increase its legitimacy and attract more supporters but also to have time to mobilize its troops for eventual offensives.

Lessons Learned My discussion of research on international negotiation raises a number of ideas that contribute to theory and practice. The cases studied have been wide ranging. I have done in-depth studies of 15 cases, including the base-rights set, the arms control set, the FSI series, the GATT and WTO trade conferences, and several internal and regional cases. The comparative studies were not in depth but added more generalizable insights about negotiations to terminate civil wars and U.S.–Soviet/Russian summitry. What emerges from these studies are larger insights about international negotiation that contribute as well to knowledge about managing conf licts. The insights also contribute to our understanding

32 A Career in Three Parts

of such central concepts in social science as interaction processes, power, legitimacy, and the role of ideology. Here is a sampling of some of the lessons gained from the research: •

•

•

•

•

A key challenge in any negotiation is to overcome constraints on being f lexible. These constraints include the pressures of representing teams or constituents, bringing values or ideologies into the talks, the other party’s willingness to move or make concession, the presence of audiences, and even locations for the talks. Progress in negotiation depends on taking advantage of moments that turn the talks in the direction of agreement. These turning points can be created by the negotiators with new ideas or procedural changes. They can also be precipitated by events that occur outside of the talks, such as policy changes and third-party interventions. External factors are especially important for progress in negotiations about security issues such as arms control. Types of issues play an important part in the way a negotiation unfolds. The course of a negotiation is inf luenced by an interplay between differences in values and interests. This interplay captures a dynamic that shifts from moving closer to getting further from agreements. Strategies found to be effective in dealing with value differences include delinking the values from the interests and affirming the other. Negotiations are tools for settling or resolving conf licts. Getting to the negotiating table is challenging. Disputing parties often look for ripe moments to enter a negotiation. These moments may occur when the parties calculate that their relative power is more symmetrical, and they perceive an uptick in their legitimacy. These calculations continue through the course of the talks as the parties deal with their conf licting values and interests. Tactics can divert attention from larger strategic goals in managing conf licts. This happens when the focus is on getting an agreement. The price paid is a short-term remedy rather than lasting peace. Third parties in particular react to their own incentives for getting the agreement. Reputation gains in the short term may f lip to reputation losses over time. Note in this regard Kissinger’s shuttle diplomacy in the Middle East, the mediations that occurred before the cease fire in Nagorno-Karabakh, and the short-term gains in cooperation achieved by conf lict resolution workshops.

International negotiations can also be understood through the prism of alternative metaphors. Game theorists approach the topic as puzzles to be solved. Laboratory experiments draw insights from bargaining games. Organizational theorists view negotiation as a process to be managed. Many international relations scholars share the perspective of negotiation as another setting for playing out the larger-stakes contests of diplomatic politics. Much like the parable of the six blind men and the elephant, what a disciplinary scholar imagines is shaped by the prism (or touch) through which it is viewed (Druckman, 2007).

Tackling International Negotiation 33

Training The connection between the bargaining and international negotiation research surfaced in my approach to skills training. In an attempt to bridge theory to practice, we developed 16 thematic narratives that captured the body of research on bargaining. These narratives were used in four exercises that represented roles of international negotiators: analyst, strategist, performer, and designer (Druckman and Robinson, 1998). The analyst and strategist roles were implemented with case studies, while the performer role was enacted in a multilateral simulation. The designer task encouraged participants to use their creative juices in constructing an exercise that embodied several of the research themes. This task was also the basis for a line of research on concept learning most recently reported by Druckman and Ebner (2018). This approach to training was well traveled. It was conducted on five continents, North and South America, Europe, Asia, and Australia. A large variety of groups experienced the approach, including diplomats, lawyers, business executives, labor union members, teachers, and students in a variety of graduate programs. A series of workshops was conducted with United Nations diplomats from all over the world, funded by the United States Institute of Peace and coordinated through the Institute of World Affairs in Washington, DC. Importantly, short- and long-term evaluations were conducted following each of the workshops and used to improve the approach for future applications. Of particular interest was the revelation that many participants used the insights gained in their every-day professional jobs. (For a summary of what was learned, see Druckman, 2006b.) These applications demonstrate the value of evidence-based learning.

Notes 1 Many years later, in 2004, I accepted an invitation to consult with a firm known as CENTRA Inc. That project was an opportunity to evaluate this framework with cases of international negotiation. The framework structured the analyses, which included several innovative statistical techniques. 2 Fred Iklé spent much of his career at RAND and other think tanks before becoming director of the Arms Control and Disarmament Agency and then Under Secretary of Defense for Policy. He had a thorough grounding in literatures on negotiation as shown in the Bibliographic Note in his 1964 book. Arthur Lall had first-hand experience with multilateral diplomacy at the United Nations. He draws on these experiences to develop more general insights about negotiating processes and the policies that guide them. John Newhouse was a journalist and author specializing in international affairs. He served on the staff of the U.S. Senate Committee on Foreign Relations for several years. He writes about non-fiction with a fictional pen, engaging the reader in the drama of international relations and its actors. Each of their books makes a conceptual contribution to the topic.

4 PUZZLES ABOUT NATIONAL IDENTITY

The year was 1962, the place was Evanston, Illinois. The Northwestern graduate programs in psychology and international relations were critical to my professional development. They provided the more specialized training needed to move forward as a social scientist. That training consisted of learning about conf lict in intergroup relations and a veritable platter of methodological choices for doing research. My mentors were social psychologists, anthropologists, and international relations professors in political science. These experiences provided a foundation for the wide-ranging career documented in this book. They also provided the confidence I needed to adapt to a variety of work settings and project challenges.

Ethnocentrism The LeVine-Campbell ethnocentrism project was expansive. It explored a diverse set of theoretical approaches to understanding intergroup relations. These approaches included both societal theories (e.g., realistic group conf lict, reference group) and those pitched at a micro level (e.g., frustration-aggression, transfer, and reinforcement). The final chapter in their 1972 book, on concord and discord among the theories, made evident the value of pitting competing hypotheses against each other.1 Yet, despite the richness of their analyses, they did not provide cross-cultural datasets to evaluate predictions from these theories. I addressed this gap in some ways with data collected from the Inter-Nation Simulation (INS). Three studies provided evidence for competing hypotheses derived from the literature organized by LeVine and Campbell and their collaborators. A 1968 article, reprinted in this book, addressed such puzzles as these: Is ethnocentrism a universal feature of groups or nations? Do strong outgroups evoke different DOI: 10.4324/9781003293361-7

Puzzles About National Identity

35

sentiments (a stronger bias favoring the ingroup) than weaker groups? Do roles that have more contact with outgroup members, such as Foreign Ministers, display weaker ethnocentric biases than those with fewer contacts, such as Defense Ministers? The answers were yes on the universal bias, yes on liking but no on respect ratings for stronger outgroups, and yes on contact roles. I will have more to say about this study and its methodology in Part 5 on National Identity. A second INS study, published in 1970 by Druckman and Ludwig, addressed another set of competing hypotheses about stereotypes. We asked if there is more variety in adjectives used to describe ingroups than outgroups. The answer was a qualified yes. A clearer stereotype (fewer adjectives) emerged for outgroups contingent on the type of contact as more or less friendly. These findings contribute to the body of work on contact theory. A third study on stereotyping, published in 1974 by Druckman, Ali, and Bagur, was a survey data collection from India, Argentina, and the U.S. We asked about the impacts of friendliness, familiarity, and type of adjective used to depict own and other nations. We found more agreement for friends on evaluative adjectives, more agreement for non-friends on descriptive adjectives, and several culture-specific results such as stronger agreement for unfriendly nations in Argentina. These findings provide further specification for predictions from ethnocentrism theory, particularly with regard to dimensions of evaluation and culture.

Group Attachments Emotional attachment to groups was also a subject of my dissertation research on simulated collective bargaining. I came across Blake and Mouton’s research during my time on the ethnocentrism project. They made a strong case for the importance of group attachments in attempts to resolve negotiating impasses ( Blake and Mouton, 1961). Looking closely at their design, I noticed that group membership was confounded with attitudes toward compromising and type of preparation for the negotiation. My study separated these variables and found stronger effects on outcomes for the attitude and preparation variables leading to my first publication (Druckman, 1967). A lively debate about my design with Neal Vidmar ensued. I “won” that debate and made a case for inadequacies in Vidmar’s study of representation effects (Druckman, 1971a). The experimental literature on representation effects took off and was a basis for a meta-analysis published 23 years later: Representation effects were weaker than a host of other situational variables (Druckman, 1994). Despite this finding, I received a nice letter from Robert Blake on November 2, 1994. He sung praises for the analyses. It is also the case that experimental manipulations of representation do not capture group attachment as it occurs outside the laboratory. Two anniversary issues of International Negotiation explore the group attachments issue in some depth. In the first article, appearing in the tenth anniversary issue, I focus attention on collective action (Druckman, 2006a). Group loyalties

36 A Career in Three Parts

are shown to connect to public opinion, representation, and the development of policies, norms, and mobilization for collective action. One puzzle that arises is how national loyalties encourage certain national policies and facilitate mobilization. Another puzzle is how more insulated loyalties inf luence a willingness to cooperate with other nations in security and trade alliances. The Trump presidency in the U.S. provides poignant examples of how a narrow conception of the nation discourages alliance cooperation. The second article, appearing in the 20th anniversary issue, focuses on the constraints and opportunities of representing groups (Druckman, 2015). A good deal of the article reviews research on constituent inf luence with attention paid to divided constituents along the lines of doves (soft-line) and hawks (hard-line). Interestingly, hawkish factions within constituencies have more inf luence on their representatives than dovish factions (Aaldering and De Dreu, 2012). This may be due to the perception that hawks are more loyal to the group, conferring higher status on them. Their inf luence decreases when their perceived loyalty is regarded as dysfunctional as in the case of Trump followers in the 2020 elections.

Regional Politics An invitation to contribute a chapter to a book on regional politics provided an opportunity to place group identities in a broader international context. I was particularly interested in relationships between structures and perceptions. Structures in regional politics include both intra- and inter-societal institutions, norms, and communication channels. Perceptions consist of views of own and other nations in the regions that stem from national identities. A puzzle is how regions are regarded as structured fields for perceptions. An answer to this puzzle is that “the more integrated the nations within the region, the more perceptions will be biased by the structural context of interactions; the more autonomous the regional actors, the more perceptions will ref lect ingroup needs, motives, and interests” (Druckman, 1980, pp. 25–26). An example of an integrated region is Europe due in large part to the European Union. We would expect that citizens’ perceptions of other nations are shaped by their cooperative relationships with the other EU members. An example of a relatively autonomous region is the Middle East due to the Israeli–Palestinian conf lict or to continuing conf licts between African neighbors. In those regions we would expect that perceptions of neighbors ref lect one’s own nation’s needs. Thus, we would expect a stronger sense of ingroup identity with the corresponding biases in the autonomous regions. The connection between structures and perceptions was part of a broad framework that weaves its way from perceptions to intra-regional relationships. The framework organized the chapter, beginning with a discussion of the formation, content, and accuracy of views about other nations in the region. Then I considered factors in regions that lead to changes in these types of perceptions with attention paid to international exchange programs (Merritt, 1972),

Puzzles About National Identity

37

problem-solving workshops (Kelman, 1972), and interactions between partisans and authorities within societies (Gamson, 1968). Structures come back into play with communication among representatives of nations within the region, which often take the form of negotiating values and interests. Those communications are inf luenced by the type of regional system characterized as cooperative (Western Europe, North America), conf lictual (Middle East, Africa), or transforming (former Soviet Republic, former Yugoslavia). The idea that regional dynamics are intertwined with perceptual dynamics is central to carving out a social science that strives to link micro and macro-level processes as discussed in the “Scaling Up, Down, and Across” paper in Part 5 on national identity (Chapter 16).

Identity With Groups and Nations My time on the NRC Committee on International Conf lict Resolution included a project on nationalism. In addition to organizing a workshop on the topic, I contributed a chapter for the 1995 volume Perspectives on Nationalism and War, edited by Comaroff and Stern (Druckman, 1995). In parallel, I prepared a 1994 article that appeared in the Mershon International Studies Review (MISR), which appears in this volume. These articles continued my early interest in bringing insights from the social psychology of groups to relations among nations. They also connected different approaches to the study of national identity. The approaches differ in scope and object. Face-to-face interactions with group members are useful for studying how tasks and the people who implement them inf luence group attachments. Large nations consisting of conglomerates of diverse populations are useful for studying mobilization for collective decisions and actions. The MISR article was an attempt to show how individuals build loyalty to groups that expand to larger collectivities.2 The small-group loyalties wend their way into norm formation and support for national policies as well as various forms of collective action. But the transfer of loyalties may also move in the opposite direction, from national to small group loyalties. The link between nationalism and war was treated in later writing of a chapter for a book edited by Daniel Christie and his colleagues on Peace, Conflict, and Violence ( Druckman, 2001b). In this piece I bring the framework inspired by Sawyer and Guetzkow (1965) back to work. Rather than to reduce the causes of war to one or a few factors, the framework places nationalist sentiments in a web of conditions and background factors that produce decisions to initiate or avoid civil or international war. In this framework, nationalism is defined by f luid or durable identities and responsiveness to myths usually created by decisionmaking elites. A current example of myth-making is the rationale used by the Russian leader, Vladimir Putin, for invading Ukraine. One insight was that new myths are easier to create among citizens with f luid identities, while older myths are more likely to be sustained when citizens have more durable identities that support the myths. Another insight is that citizens’

38 A Career in Three Parts

national identity may be less important in authoritarian than democratic systems. While authoritarian elites can more easily manipulate sentiments, they may have less reason to do so. This work was followed by an article that compared four North African societies on three dimensions, referred to as durability of identity, spread of identity, and the extent to which national identity is coerced as in the authoritarian–democratic distinction (Druckman, 2001a). Putin’s messaging about the invasion of Ukraine fell on mostly receptive Russian citizens’ ears, illustrating the effects of widespread durable identities in an authoritarian political system. Both the chapter and article, published in the same year, opened my eyes to the value of analyzing nationalism as a societal phenomenon. The NRC committee experience encouraged me to consider another issue concerning nationalism. This was the difference between nationalism doctrine and sentiments. The former is favored by historians and political scientists such as Gellner (1983) and Haas (1997), the latter by social psychologists such as Tajfel (1981). On the doctrine side of the subject, I was and continue to be impressed with a less well-known treatment of nationalism by Kedourie (1993). He conceived of nationalism as a doctrine or ideology invented in Europe at the beginning of the nineteenth century. For him the doctrine was a foundation for defining a nation as a cultural unit in the sense of shared beliefs and customs as distinguished from a state, which is a legitimate administrative entity. While agreeing with the earlier philosophers that a nation is no more than a collection of individuals, he steered clear of analyzing those individual sentimental attachments. Those attachments have been the stuff of the social psychology of intergroup relations. The puzzle is to find ways of connecting these silos as I have attempted in both early and recent articles in Part 5 on National Identity. The connection was also a guiding theme for a graduate course that I taught on “Conf lict and Isms.” This combination of empirical research and review of writings on national identity led to a realization that we needed to find pathways to theory. This means understanding the way identity is formed and changes through each of three levels of analysis – interpersonal, group, and collective. Interestingly, the literatures coalesce around these levels, which also divide roughly into social science disciplines. A way of construing this division is through a chronological sequence going from early socialization, involvement with peer groups, and becoming aware of a nation and relations among nations. As we shift our focus from one level to another, we do not shed the inf luences of earlier experiences. This then leads to addressing a puzzle about how the levels interact through time, for example, paths that go from early imitation of parents to conformity to group norms to a willingness to mobilize for collective action for or against current government policies. Further, this path moves up the levels, as in this example, but also down the levels as when mobilization reinforces or dismisses earlier identities. These problems have been understood since my graduate school days, and yet the puzzle remains. I return to these puzzles in Part 5.

Puzzles About National Identity

39

Notes 1 The evaluation of competing hypotheses permeates my research in several areas. It is in the falsification tradition of research (Popper, 1959), which has been a source of considerable debate in the social sciences. Though aware of the arguments on both sides, I have been attracted to the value of this approach. 2 This article received the 1995 Otto Klineberg Intercultural and International Relations award from the Society for the Psychology Study of Social Issues, a division of the American Psychological Association.

5 A ROLE FOR JUSTICE

A chance meeting with a former colleague, Cecilia Albin, at a conference in the UK in the early 2000s led to a 15-year project on justice. She asked if I would be interested in collaborating on a proposal for research on how justice principles inf luence negotiating processes and outcomes. Although my previous work did not prioritize these principles, I had followed the literature with particular interest in Morton Deutsch’s work on distributive justice (Deutsch, 1985), John Thibaut and Laurens Walker’s book on procedural justice (Thibaut and Walker, 1975), and the Lind and Tyler (1988) work on the social psychology of procedural justice. I accepted her invitation with enthusiasm, and we began work on a grant proposal. At the time, the Norwegian Foreign Ministry had created a unit to encourage research on peace processes. We prepared a working paper for an initial reaction from them. Their reaction was positive, and they asked for a budget. We in turn asked them if they would send guidelines for preparing a formal proposal, to which they responded that only a budget would be needed. Upon receiving the budget, they approved the project for 18 months of research. This was the easiest proposal either of us had ever prepared. I worked on the project from my perch at the University of Queensland; Cecilia worked from her office at Uppsala University. The focus was on the effects of distributive and procedural justice on implementing peace agreements.

Justice and Implementing Agreements The first study in this stream of research concentrated on distributive or outcome justice (DJ). In order to evaluate hypotheses about the relationship between DJ and the implementation of agreements, we needed a data set. The 16 peace agreement cases compiled by Downs and Stedman (2002) fit the bill. They also DOI: 10.4324/9781003293361-8

A Role for Justice

41

provided indicators of implementation (dependent variable) and the intensity of the conf lict. We added indicators of the justice variables. Our statistical analyses evaluated a hypothesized path from conf lict intensity (independent variable) to implementation (a dependent variable) through distributive justice principles (mediator). The DJ measure was an aggregate of four principles: equality, proportionality, compensation, and need. The most interesting finding was that adherence to the DJ principles reduced the negative impact of an intense conf lict on implementation success. The key DJ principle was equality. When outcomes were fair, defined as equal benefits, implementation worked well even in highly conf lictual environments. Thus, the study demonstrated an important role for justice in sustaining agreements through the implementation period (Druckman and Albin, 2011). The second study focused on procedural or process justice (PJ). Using the same data set, we extended the earlier results on DJ. Specifically, we found that the equal outcomes were encouraged by PJ in the negotiation process. Those outcomes in turn led to more successful implementation of the agreement provisions. Also interesting was the finding that power asymmetries between the government and rebels had no effect on PJ, equality or implementation. Taken together, the two studies highlight equal outcomes, first, as a mediating inf luence on the effects of conf lict intensity and, second, as a mediating inf luence on the effects of PJ. We dug deeper into the meaning of equality in this context. Three types of equality were identified: equal measures, equal treatment, and equal shares. Of the three, equal treatment made the largest difference perhaps because it deals more with relationships than with things (Albin and Druckman, 2012).

Justice and Durable Peace The largest study in this research stream, sponsored by the Swedish Research Council, was an analysis of a 50-case data set on peace agreements and durable peace. We moved from a focus on implementation to lasting societal peace. This move required, first, developing an understanding of the durable peace (DP) concept and, second, constructing indicators of its various parts. The four parts of the concept were reconciliation, security institutions, governance institutions, and economic growth. Each of these parts was further divided into components such as reception and reintegration (reconciliation), demobilization and disarmament (security), power-sharing (governance), and income equality (economic growth). Taken together, there were 16 components across the four parts of DP. The variety of measures provided options for evaluating the hypotheses. For example, we discovered that two of the four parts of PJ had the strongest impact on DP, that the justice variables had only a weak impact on economic growth, and that reconciliation was the key part of the DP index. More importantly, however, was the path showing that PJ led to fair outcomes, which produced successful implementation leading to DP (Druckman and Wagner, 2019).

42 A Career in Three Parts

Justice and Effectiveness Three studies, supported by a Swedish Bank Foundation, expanded the reach of our work on justice. We analyzed the role of justice in trade (Albin and Druckman, 2014b), arms control (Albin and Druckman, 2014a), and environmental (Albin and Druckman, 2017) negotiations. Since peace following war was not the aim of these cases, we switched our dependent variable from DP to effectiveness. Five measures were developed to assess effective outcomes in each of these areas: number of issues resolved, time to reach agreement, degree of agreement, type of agreement, and the balance among integrative and distributive elements. The analyses of trade cases showed that PJ strongly inf luenced effectiveness in both the bilateral and multilateral cases. In fact, PJ accounted for the relationships found between DJ and effective outcomes. We also dealt with methodological and explanatory issues. The former concerned matters of causation in case analysis. The latter encompassed the important role played by global norms that emphasize conduct (rather than distribution) in trade negotiations. The arms control cases produced different kinds of results. Whereas PJ principles inf luenced the effectiveness of the outcome in bilateral cases, they had little effect on those outcomes in the multilateral cases. Case-by-case analyses brought understanding to the different findings. We discovered that PJ exerted a stronger effect on outcomes when the incentives made alternatives to negotiating an agreement less attractive, when the negotiators shared a trusting political culture, when the scope of the issues was limited, and when there was an impending deadline. These findings bring negotiating variables to bear on the justice-outcome relationship. A different picture emerged with regard to the multilateral cases. Although these negotiators strongly adhered to PJ norms, the outcomes were not inf luenced by them. The lack of variation in PJ among the multilateral cases reduced the correlation between PJ and effective outcomes: Ineffective cases had the same general level of PJ adherence as did the effective cases. Add to this the risk-averse posturing of multilateral arms control negotiators, and it is not surprising to discover that PJ may take a back seat to other factors that move the talks towards agreements even if they are largely cosmetic. Interesting findings emerged as well with regard to the third issue area: environmental negotiations. In contrast to the arms control cases, PJ exerted stronger effects on outcomes in the multilateral than the bilateral cases. Three reasons were given for the multilateral case findings: belief by delegates in the importance of PJ principles, a more trusting environment, and confidence going forward. Although large arms control and environmental conferences are both regulated by similar cooperative norms, the latter delegates are less concerned with security consequences of outcomes. As a result, there is likely to be more variation on our measures of effectiveness. Looking more closely at the PJ-effectiveness relationship in the multilateral environmental cases, we found that the key PJ principle was fair representation, and the key aspect of effectiveness was the quality of the agreement. The cases that adhered most closely to this aspect of

A Role for Justice

43

PJ attained the best agreements. For the bilateral cases, power differences loom large and may reduce the impact of adhering to PJ principles. Across the four domains studied, PJ principles mattered. They increased the likelihood of peace following agreements, and they produced more effective bilateral and multilateral agreements in trade talks. Findings about justice in the arms control and environmental areas were contingent on the size of the negotiation. Bilateral arms control and multilateral environmental talks showed stronger relationships between PJ and effectiveness. An advantage of exploring the same hypotheses in several contexts is that we can evaluate the generality of the findings, pointing to directions for theory building. They also call attention to arenas where the PJ principles are more or less likely to have impacts. An Annual Review of Psychology article with Lynn Wagner connected two themes of this book: negotiation and justice. It also contributed a framework for organizing relevant research. This article, appearing in Part 6 (Chapter 17), opens the discourse, which is encouraged by a long list of hypotheses for further research with implications for practice.1

Note 1 The research on justice was presented in 2012 to the European Union’s Negotiation Club in Brussels just after they received the Nobel Peace Prize. Following the talk, a member of the audience asked whether I thought that the EU deserved the Peace Prize. Having given this question no forethought, I responded “without a doubt” and received an enthusiastic round of applause. On the train back to Paris I reflected on this answer and concluded that it was correct. Any organization that could keep Europe, and particularly France and Germany, from war was a worthy recipient.

6 BEHIND THE SCENES

Research of the magnitude discussed in this book is not simply a result of hard work and commitment by the authors whose names appear on the articles and chapters. It is that of course, but it is much more. People and organizations who may surface in acknowledgements often play a large part before and during the research process as well as after the research has been completed. These are participants in simulations and respondents to surveys, sponsors, liaisons between sponsoring and home organizations, interviewees and archivists, participants in the historical cases of interest and scholars who specialize on these cases, technical support, and those that help with dissemination of reports and articles.1 Much of what readers see is made possible by what they do not see.

Transitions: IJR, AED, ACDA My nine-year stint at the Chicago-based Institute for Juvenile Research (IJR) was research intensive. We depended on a large number of participants who contributed data for our analyses and the organizations that granted permission to implement them. These included university subject pools; children at a summer camp; school children in Illinois, Argentina, and India; and clinical professionals for a conf lict resolution workshop. A career turning point occurred in the early 1970s when I was invited by the Academy for Educational Development (AED) (sponsored by the Schweppe Research and Education Fund) to prepare a state-of-the-art review paper on international negotiation as part of its program on resolving international disputes. Following on the heels of this paper was a project requested by the U.S. Arms Control and Disarmament Agency (ACDA). The contract was given to Thomas Milburn at Ohio State University. Tom was on my dissertation committee. Our role consisted of preparing a compendium of propositions on international negotiation. These projects paved the way to DOI: 10.4324/9781003293361-9

Behind the Scenes

45

a research and teaching career that focused on negotiating in the international context. Thanks to the sponsors (AED and ACDA), I was ready to move from the Chicago-based institute to a Washington, DC consulting firm.

A New World The move to a consulting f irm introduced me to a world of secrecy and intrigue. The early projects were contracts for supporting ongoing negotiations that involved the U.S. government. The first challenge was to learn how intergovernmental negotiations work. This was aided by 25 interviews with delegates and support staff to such conferences as the Paris Peace talks, SALT, MBFR, the threshold test ban, the Conference of the Committee on Disarmament, and the Law of the Sea. A second challenge was to perform an analysis of the talks with Spain over base rights. To pull this off required coordination with delegates, including the U.S. delegation leader Robert McCloskey and his deputy, Lawrence Pezzulo. As I noted earlier, this project resulted in the first article on turning points, which needed to clear classification hurdles before being published ten years later. This work was the basis for the Foreign Service Institute (FSI) project on base rights negotiation spearheaded by Ambassador John McDonald (McDonald and Bendahmane, 1990), which also had classification sensitivities.

MBFR and START A more complex negotiation project followed the work on base rights. This was an analysis of a three-year period (1977–1980) of the alliance talks between NATO and the Warsaw Pact referred to as MBFR. It also required considerable coordination with delegates and support staff. Our team worked for a task force in the Pentagon. Our liaison was Richard Clarke, who later became famous with the release of his tell-all 2004 book, Against All Enemies: Inside America’s War on Terrorism. Dick was more of a collaborator than an administrator, but he also opened doors to key actors in this melodrama that ended without an agreement. We benefited immensely from John Dean’s near-perfect recording of the plenary conversions among the delegates in Vienna. Ambassador Dean became the chief delegate in the later years. The transcripts enabled us to perform content analyses that revealed alliance dynamics and Warsaw Pact objectives. Both Cold War projects, base rights and MBFR, taught me about the value of content analysis for understanding the meaning of conversations. They could not have been done without access to documents provided by government agencies. Invaluable feedback was provided by Brent Scowcroft (former U.S. National Security Advisor) and George Kennan (architect of the policy of containment) following project briefings at our Mathtech offices in Bethesda, Maryland. Content analyses of transcripts came into play in a project sponsored by ACDA in the early 1980s. The agency asked us to ascertain Soviet motivations for returning to the START table. To accomplish this task I needed to spend

46 A Career in Three Parts

three summer months in the State Department, without parking privileges. Our recommendation about incentives was briefed to the agency and read by its director, Kenneth Adelman. As noted earlier, our reward was that the Soviets did eventually return to the table.2

Elite Mobility Another consulting project during this period depended on support from the Library of Congress (LOC) and the Brazilian Embassy in Washington, DC. The data consisted of biographical profiles of generals and cabinet ministers in the military regimes that ruled the country from 1964–1975. Everett Larsen from the LOC Hispanic Division provided essential assistance by showing us how to use the Almanaque do Exercito, including translation from Portuguese to English. Staff at the Brazilian Embassy located “Who’s Who in Brazil” for English and Portuguese profiles of the ministers serving in the various cabinets. With these sources in hand, either at the LOC or on loan from the Embassy, we were able to code the variables suggested by our hypotheses (Druckman and Vaurio, 1983).

Nonverbal Communication Outside support was also needed to perform a project on nonverbal indicators of deception and related intentions. The sponsor, the Defense Advanced Research Projects Agency (DARPA), requested a program that combined research and training. This was done at the Hurlburt Field Air Force Base in Florida, where my colleague Richard Rozelle and I lived for ten days in the 24-hour whirl of C130 engines. Considerable coordination was needed to make the project work, from the Commander of the base to the co-directors of the Cross-Cultural Communications program through to the Lieutenants who aided in the data collection and lecturing. The result was an original experiment with practical findings about the value of several types of training to decode nonverbal cues. The best procedure was the most complex, which we referred to as inference training (Druckman et al., 1982: Figure 5.1). The elements of diagnosis and decision were used later in the work on e-mediation.3

Political Stability New consulting assignments came with a switch from one Bethesda, Maryland firm (Mathtech) to another (Booz Allen Hamilton). We had three years to analyze the stability of the Marcos regime in the Philippines. Considerable behind the scenes activity was needed to implement this complex project. First, an appropriate team was assembled. It included an academic expert on the Philippines, a professor of international relations, the former deputy director of the CIA, a colleague with experience in constructing large longitudinal data sets,

Behind the Scenes

47

and myself as director. Second, we assembled an advisory group of eminent scholars to oversee the project. They were Fred Fry, Jack Goldstone, and Gary Brewer. Third, we put together an advisory panel with expertise in Philippine politics. They helped with the task of expert-generated data on variables more difficult to measure from available sources, for example, aspects of legitimacy, group attributes, and social distance. Fourth, we searched for and found sources for measuring the many aspects of our model. These consisted of 97 sources organized into the four parts of the framework and can be found in Druckman and Green (1986: Table 5). This project posed both administrative and substantive hurdles. It could not have been accomplished without the support noted above. Add to this a difficult sponsor who, while providing the resources, was of little help with implementation and conveyed skepticism about whether we would produce anything useful.4 Nonetheless the project was successful,5 but the stresses of directing it sent me packing for a trip “downtown” to another Washington, DC organization where I began my next career adventure.

NRC Committees The earlier experiences at a research institute and in the Washington world of consulting prepared me for the next chapter in my career. Directing projects at the U.S. National Research Council (NRC) combined aspects of both think tanks and consulting firms. The think tank part is embodied in the collaborative process of doing studies and producing books. The consulting aspect consisted of attracting sponsors for the work and keeping them satisfied with its products. This was my dual challenge for 12 years, during which I directed eight committees. Career continuity was accomplished in my role for two committees on preventing nuclear war and on international conf lict resolution.6 In contrast to the earlier consulting experiences, the sponsor of both committees (Carnegie Corporation of New York) was unfailingly responsive to our requests for support. Collaborating with Nobel laureates such as Herbert Simon, Kenneth Arrow, and Thomas Schelling and reconnecting with Amos Tversky and Harold Kelley were rewarding. But so too were the opportunities to write several important articles on nationalism (in this collection), mutual security negotiations (with Terry Hopmann), and interventions in history (with Paul Stern). The more interesting challenges occurred in the shadows of another NRC committee where vested interests of various outside constituencies came into play. The committee evaluated unusual techniques for enhancing performance with possible military applications. The vested interests came from several quarters, some with strong financial stakes (subliminal and sleep learning, hemispheric synchronization), others with strong beliefs bordering on the fanatic (accelerated learning, meditation, extra-sensory perception [ESP]). Backers of these techniques lobbied the NRC for positive evaluations. Senator Claiborne

48 A Career in Three Parts

Pell of Rhode Island was eager to use the committee’s report to establish a National Science Foundation Program on ESP research. The reports were disappointing to these enthusiastic promoters, and some of them threatened law suits or cessation of support for the National Academies (NAS).7 Not since Abraham Lincoln signed the charter for the NAS had there been so much uproar over its reports, nor was there a more exciting time in my consulting career. A spin-off consulting assignment resulted from the human performance work at the NRC. The U.S. Federal Trade Commission (FTC) was investigating a California company purporting to improve foreign language learning by leaps and bounds in a short period of time. My job was to either substantiate or rebuke their claims for rapid improvement. Their techniques were within the scope of the accelerated learning approaches investigated by the NRC committee. I completed the review and briefed the Commission on the findings. A lawsuit was filed by the FTC. The company was dissolved for false advertising in interstate commerce. This is another example of how research can have direct (and costly) consequences. More generally these NRC projects are good examples of bridging research and practice. They show that evidence-based approaches can be used by people to improve themselves, their groups, and the organizations in which they work.

Australian Universities The next stop was academia. An offer from George Mason University (GMU) moved me from the NRC to the classroom. I had taught courses at the Institute for Conf lict Analysis and Resolution since 1984. I was spared the ugliness of academic politics until I took a full-time position in 1997, in my late 50s. The course of events that ensued was anything but cushy, even with an endowed chair and tenure. There is no need to revisit these events. Suffice it to say that they motivated me to switch departments at GMU and continue my career with a half-and-half arrangement divided between the U.S. (fall semester) and Australian universities (spring semester). Adapting to Australian higher education was challenging. On the surface I was dealing with a three-year undergraduate program, a research-only doctoral degree, a strong preference for theory and philosophy in political science, rigid course syllabi, and a single source of funding for research (the Australian Research Council). Behind the scenes I needed to negotiate funding, course loads, course content, and grading policy. Funding was never sufficient. Increased teaching loads and grants were needed to round off salaries and support research projects. I taught undergraduate (“Aggression and Cooperation”) and honors courses in psychology and political science at the University of Queensland (UQ) in Brisbane and a masters course on diplomacy at the Australian National University (ANU) for four years in Canberra. Grading policies at both these universities got me in trouble. The administrators demanded curved-grading; I judged the students’ work on the merits of their products. My penchant for intensive research training in

Behind the Scenes

49

the diplomacy course did not sit well with the students, who preferred tips on practice. I rose to the demand for outside funding only to be caught off guard by a 60% overhead rate at the UQ. Fortunately, Ren, an Australian academic colleague, helped me to find other Australian universities that welcomed my research grants with minimal overhead rates. I was able to continue the research on justice in the supportive environments of the University of Southern Queensland in Towoomba and at Macquarie University in Sydney. These were productive and happy years to close out my formal career.

Notes 1 Other behind the scenes stories are about collaborations that did not work. Often collaborative crises are not foreseen during the early days of a project. They occur for a variety of reasons, including loss of motivation; being side-tracked by new responsibilities that come with promotions, job changes, or personal life circumstances; and disputes about perspectives or credit in the form of author order on publications. I have experienced each of these types of project breakdowns. The most difficult collaboration crises are those where a co-author disappears for unknown reasons only to emerge after most of the work has been completed. 2 Shortly after the ACDA project I had a role in designing and analyzing war games held in the bowels of the Pentagon. This experience opened my eyes to the way gaming was used to develop military policy. The output from these games was used by Andrew Marshall and his Office of Net Assessment to make recommendations to the Office of the Secretary of Defense. The “wallpaper” in Marshall’s office consisted of a very detailed map of the world in which pins were placed in presumed hot spots. 3 We also studied micro-momentary facial expressions in Emanual Donchin’s laboratory at the University of Illinois. These expressions were part of an information-processing model that included changes in the bargaining opponent’s strategy and the amplitude of a neurophysiological wave-form shown on the EEG. The work contributed to further understanding of the monitoring function in negotiation (Druckman et al., 1983). Behind these scenes, this unusual collaboration had its moments as each of us brought different perspectives to the research. 4 To clarify this point, the sponsoring agency was funding a competing project that developed a utility theory model of regime stability. That model had the advantage of simplicity, which seemed to be attractive to the sponsor. Other conflicts of interest placed the sponsor in the role of spoiler rather than advocate for our project. 5 Additional validation came from the protagonist in our analyses, Imelda Marcos. Many years later, in 2002, my wife and I had an opportunity to interview her at her apartment in Manila. She endorsed our insights and the scenario that showed her as the heiress apparent to the throne. That scenario did not emerge due to the triumph of the Peoples’ Revolution in 1986. Imelda was skeptical of their tactics. The interview was held during a Senior Fulbright teaching stint that brought me to the Philippines. 6 I also directed four committees on enhancing human performance, a committee on brain and cognition, and a committee on the human dimensions of global environment change. 7 A wealthy donor to the NAS threatened to cut off support if the report was published. To his credit, Frank Press, president of the Academy, told this donor that the report would be published and he was free to withdraw his contributions as he saw fit.

PART CONCLUSION

The spiral career documented in this Part is an exercise in the art of prosopography. This art form provides little guidance on how to structure the narrative and what to include in telling the story. There is a temptation to garner the reader’s attention by highlighting the more dramatic episodes in a career. Indeed, I was smitten by this temptation particularly in the chapter on “Behind the Scenes.” But there is also a responsibility to contribute to the stores of knowledge in the relevant fields. I have been aware of this responsibility throughout the writing both in terms of topic coverage and research findings. Mixing drama with science – stories with stores of knowledge – without confusing them is challenging. It turns out that accomplishing the research goals depends to no small degree on managing events, taking advantage of opportunities when they appear, and staying the course, by which I mean learning how to be resilient. These are contributions made by this career to the craft of prosopography. My nine-year stint at the Chicago-based institute taught me the value of dedication to doing research projects and publishing the results. The discipline required to complete projects carried over to the consulting and academic worlds. My 11 years at for-profit consulting firms taught me to value f lexibility in the pursuit of projects that would interest sponsors. The following 12 years at the NRC taught me skills useful for turning sponsors into benefactors. But there were still lessons to be learned elsewhere. Nearly 20 years at a variety of universities made me appreciate the classroom, both as a way of sharpening communication skills and for furthering research goals with students. Immersing students in the discovery process, much like hands-on learning at children’s museums, is rewarding. In a nutshell, this is a career that is characterized by dedication, f lexibility, and inter-generational discovery. These features were developed and refined during DOI: 10.4324/9781003293361-10

Part Conclusion

51

the early, middle, and later career experiences. A sense of chronology is ref lected as well in the presentation of the fruits of labor in the chapters to follow. In an attempt to provide balance among the types of selections, I sought to cover an equal number of empirical, theoretical, and review contributions across the topics. Thus, there are three experimental and simulation articles appearing in the chapters on f lexibility and identity. There are three case and comparative case analyses in the chapters on turning points and justice. Three theoretical papers appear on f lexibility, interests and values, and identity. And three review articles are included across the chapters on turning points, identity, and justice. The selections within each of the parts are chronological, with the earliest article appearing first and the most recent last. An attempt to create a holistic conception of the career is made in the concluding part.

PART 2

Flexibility in Negotiation  

PART INTRODUCTION

My interest in f lexibility was stimulated by Milton Rokeach’s research on dogmatic belief systems (Rokeach, 1960). This work was part of the post–WWII authoritarian personality studies led by Theodor Adorno at the University of California, Berkeley, where Rokeach attended graduate school (Adorno et al., 1950). His dogmatism concept separated the content (right or left wing) from the structure (whether right or left) of belief systems. I found that a modification of his dogmatism scale was useful in my dissertation study on collective bargaining. Less useful, however, was the idea that these beliefs were aspects of personality. Flexibility takes a number of forms and is studied at each of several levels of analysis. It is often understood in relation to principled behavior. Showing a willingness to compromise, finding trade-offs, or searching for joint gains in negotiation are features of being f lexible. As well, it is manifest in being open to different views, asking questions, and being inventive or creative. The f lip side is a principled approach to negotiation. Developing a principle embodied in a doctrine or ideology is likely to lead to intransigence. Pruitt and Lewis’ (1975) idea of being firm on the goals while being f lexible on the means to achieve them captures the dual feature of negotiation. In bargaining research, this consists of maintaining high levels of aspiration in the quest for mutually acceptable agreements. The notion of f lexibility runs through all of the research on and practice of negotiation and mediation. The balance between commitment and f lexibility is the central challenge in the early bargaining studies during the 1950s and 1960s, in the consulting on arms control talks in the 1970s and 1980s, and in the work on peace agreements in the decade beginning in 2010 to the recent work on robot mediation. We learned about some conditions for making concessions in bilateral and ways to measure position change in complex multilateral talks. We also learned about how to infer a nation’s negotiating objectives from their DOI: 10.4324/9781003293361-12

56 Flexibility in Negotiation

statements and the way they adjust to changes in alliance dynamics. Most interesting perhaps is the finding that robot mediators are better at eliciting creative solutions than humans. Even more recently we discovered the way group loyalty constrains movement and the role played by trust in taking a move that signals eventual agreement or continued impasses. All of these findings support a more general viewpoint about the plasticity of human behavior (Ericsson and Pool, 2016). Rather than to continue to debate the enduring question of human potential to adapt and change, we try to unlock the secrets of how to bring that change about in a variety of experimental and case settings. The three articles chosen for this part unveil some of the secrets. They are early and mid-career pieces that display both empirical and theoretical work. The first article on f lexibility was my third publication. It grew out of the part of my dissertation study that focused on the difference between strategy and study as types of preparation for a bargaining encounter (Druckman, 1967). The study design explored several variants on the preparation theme. The second article, appearing in print nine years later, was an attempt to refine and extend the idea of boundary roles, suggested in the labor-management context by Walton and McKersie (1965). The refinement was a more precise, mathematical treatment of the concept. The extension was to place the idea in the context of international negotiations. The third article, appearing 16 years later, was an attempt to evaluate the impacts on f lexibility of a wide range of variables, occurring in different negotiation phases and referred to as situational levers. This entailed the invention of an approach to experimental research that could accommodate holistic and analytical approaches in the same study.

7 PRENEGOTIATION EXPERIENCE

I begin my introduction to the article with a back story about how I prepared for this study on preparation. My handwritten notes from a Sunday in 1967 show how I worked through the design. My first challenge was to separate the form of planning (as unilateral or bilateral) from the way the planning was done (as studying or strategizing). This resulted in adding three new experimental conditions to the Druckman (1967) experiment. The second challenge was to create a control group as a baseline to ascertain whether the conditions facilitated or restrained attempts to settle the conf lict. My notes reveal that I struggled with a possible confounding factor. Negotiators in all the conditions, including the control, needed to be equally familiar with the issues and stakes. I rose to this challenge by reserving an equal amount of time to gain familiarity in all of the conditions. The details are discussed in the article. Two contributions from this research stand out. One is the unravelling of the way collective-bargaining negotiators prepare for their task. Another is the discovery of processes and perceptions that help to explain the effects of type of preparation. With regard to the unravelling I evaluated the relative impacts of three factors: type of activity (strategizing or studying), who the bargaining representative is studying with (members of own or other team), and whether the other team’s study partners were the appointed bargainers or support staff. The findings were interesting. Developing strategies with teammates produce resistance to move but no more resistance than those in a control condition without preparation. Studying the issues produced more movement (and settlements) irrespective of whether it was done with teammates or with opponents. Thus, the baseline condition is resistance and the key to f lexibility is studying rather than strategizing. This pair of findings has carried through to more recent studies. DOI: 10.4324/9781003293361-13

58 Flexibility in Negotiation

The second contribution consisted of discovering three factors that helped to explain the findings. In an earlier age before sophisticated statistical mediation and moderation analyses, these were referred to as intervening variables. The two perceptual variables, commitment to positions and win-lose vs. problem solving, were strongly inf luenced by the conditions and related to the outcomes. Little doubt that these would have mediated the relationship between preparation condition (unilateral position-formation) and the various yielding measures. The third intervening variable was consensus among teammates on the ranked importance of the four issues. Considered to be an indicator of team cohesion, these ratings were inf luenced by condition – more consensus in the strategy condition – and in turn inf luenced the outcomes. Taken together these variables were considered to be a syndrome sparked by how preparation is done. In modern-day parlance they would be a double-mediation path (Hayes, 2022): from preparation to consensus on the issues through win-lose perceptions to outcomes. These findings were a forerunner to the large number of studies that followed. Two recent studies, one with Fieke Harinck on trust (Druckman and Harinck, 2022), another with Jennifer Palamis on party loyalty (Druckman et al., 2022), stem from these early efforts. Both these studies show strong mediation effects for win-lose perceptions, which we referred to as mindset. This article kicks off the collection because of the interesting findings as well as its role in the creation of a thread through the bargaining literature. Along with similar research by Bass (1966), this work opened a window on the importance of prenegotiation experience as a critical opportunity for framing negotiation.1

Note 1 Use of the male pronoun was the common practice for the early articles in this collection (1968 [2], 1973, 1977). This practice changed and evolved to include both/all and nonbinary genders as reflected in the remaining articles.

“This article was published in Journal of Experimental Social Psychology, 4, Daniel Druckman, Prenegotiation Experience and Dyadic Conflict Resolution in a Bargaining Situation, 367–383, Copyright Elsevier (1968).”  

PRENEGOTIATION EXPERIENCE AND DYADIC CONFLICT RESOLUTION IN A BARGAINING SITUATION Daniel Druckman

Several investigators in the area of intergroup conf lict resolution have highlighted the role of group commitment or group loyalty as an important source of constraint on the resolution of conf licts between groups (e.g., Blake and Mouton, 1962; Sherif and Sherif, 1965; Withey and Katz, 1965). However, Druckman (1967) claimed that a number of experiments on intergroup conf lict resolution confounded group representation with team position-planning before intergroup confrontation (e.g., Bass and Dunteman, 1963; Blake and Mouton, 1962; Sherif, Harvey, White, Hood, and Sherif, 1961). It was hypothesized that this experience before confrontation, rather than group loyalty per se, was responsible for the observed resistant behavior.1 In that experiment prenegotiation experience was separated from group representation in a complex bargaining situation. It was found that when opposing teams or individuals planned positions or strategies apart from one another before bargaining (unilateral strategy) they were more resistant to compromise than when opponents discussed the issues together before bargaining (bilateral study). However, group representatives were no more resistant to resolving the conf lict than were individuals who represented themselves. In fact, the prenegotiation experience main effect was large and consistent for all of the measures of conf lict resolution while the group-representation-self-representation main effect accounted for a negligible portion of the variance on each measure. This large prencgotiation effect merits further investigation. Teams (or individuals) in the unilateral-strategy condition created in Druckman (1967) were: (a) separated before bargaining, (b) focusing on their own positions, and (c) preparing rationale or strategies for bargaining. In the contrasting bilateral-study condition teams (or individuals) were: (a) together before bargaining, (b) focusing on both positions, and (c) not preparing strategies or

60 Flexibility in Negotiation

rationale for their own positions. Thus the two conditions differed in terms of the structural arrangement of the competing teams (apart versus together) as well as the type of activity engaged in during the session (plan strategies versus study and discuss). Since these factors were not separated, their relative contribution to the resulting differences in bargaining behavior could not be determined. The activity of strategy formation may not have provided any additional constraint on resolution attempts beyond that imposed by team separation and unilateral focus. That is, it remains to be determined whether unilateral focus versus bilateral focus contributed most of the variance between the conditions or whether the addition of strategy preparation to unilateral foctis accounted for the large differences obtained. The constraints on position change as a function of team separation and constraints due to strategy preparation (or the effort that goes into such preparation) are both documented in the literature. With reference to the Blake and Mouton (1962) experiments, Ferguson and Kelley (1964) claimed that as a result of unilateral discussion of a team product a person sees more of its positive features than one ·who has not had the same experience. The tendency to focus on the ingroup position prevents opponents from delineating the “region of validity” in the other’s stand or position (Hammond, Todd, Wilkins, and Mitchell, 1966; Rapoport, 1961). Rapoport (1961) suggested a conf lict-resolving technique that involves: (a) restating accurately the other person’s position, (b) exploring areas of agreement in the opponent’s stand, and (c) inducing assumptions of similarity. None of these activities is likely to be engaged in when bargaining opponents remain reticent. Bilateral discussion, before bargainers have a chance to focus on their own position, may be a necessary precondition for applying the “region of validity” technique. However, there is some evidence that indicates that unilateral focus without formal position-planning may not provide any more restraint on resolution attempts than does bilateral focus. Hammond et al. (1966) reported no differences in overt compromise or cognitive change between subjects in unilateral and bilateral focus (using the “region of validity” technique) conditions. However, in a similar experimental-task situation, when pairs of subjects were told to adopt a problem-solving, system-building orientation toward the problem (or were given a set to be analytical) they developed more initial conf lict and had less of a tendency to compromise during the conf lict session than subjects who were told to adopt a “hunch-playing” or intuitive orientation for dealing with the problem during training (Rappoport, 1965). Thus it appears that when unilateral focus is augmented with the preparation of rationale or strategies before debate, bargaining opponents are more resistant to reaching an agreement. Discussions of international negotiations among career diplomats have noted the cognitive constraint produced by position formation. For example, Rusk (1955) commented that “if negotiations between representatives were to occur before a fixed position was taken, the possibility of reaching an acceptable

Prenegotiation Experience

61

solution might be increased” (see also Blake, 1959, and Pearson, 1959). Implications for this problem may also be drawn from experimentation, stimulated by the dissonance-theory formulation, on the effects of effort on subsequent attitudes and behavior. Lewis (1965), in a review of work in this area, concluded that “the data on the effect of the expenditure of effort on the value of a stimulus event associated with effort indicates that value can be enhanced by effort” (see Aronson, 1961, and Festinger and Aronson, 1960). A hypothesized concomitant of high evaluation is commitment. The more committed one is to a position, the more dissonant it is to give in to an opposing position. Thus the successful use of techniques for inducing commitment should produce resistance to yielding. The intention of this study was to investigate the effects of several prenegotiation-experience conditions on the resistance to or facilitation of resolution in a conf lict-of-interest situation. The investigator sought to determine whether team consultation about an impending negotiation produces rigidity because positions are usually taken during these sessions, or simply because of the experience of meeting with a team. Two prenegotiation sessions that involved team separation were compared to a session in which team members studied the issues with a member of the opposing team before debate. In the latter condition team members consulted with a member of the opposing team who was not to be a future bargaining opponent. Thus the activity of bilateral study was not confused with communication with a future bargaining opponent. Also, in order to determine the effects of previous interaction per se, a condition was created in which bargainers had essentially no prenegotiation experience. A description of the prenegotiation conditions created for this study is followed by the task procedure.

Method Prenegotiation Conditions The task was a simulation of the collective bargaining process. In three of the conditions a prenegotiation session, which lasted for approximately 40 minutes, preceded the bargaining. Four types of prenegotiation sessions were designed for this study. In two conditions teams were separated before bargaining. The instructions for one of these conditions were specific with regard to the way in which positions were to be prepared, while in the other condition the instructions emphasized more informality and less effort in preparing for the debate. 2 The instructions for the unilateral position-formation condition were as follows: The next 40 minutes will be spent with your own team. You are to use this time to plan your bargaining strategy. First gain an understanding of

62 Flexibility in Negotiation

the issues. Then you are to plan a formal position for entering into bargaining. You are to write down, as a team, and so that each person has a copy, a list of points defining the rationale or arguments for your team’s position on each issue. This strategy should be prepared so as to guide the bargainers through the negotiations. You may decide on items that you think each person representing you should remain firm on. It may also include potential concessions, anticipations for trading, setting goals, and the like. After the bargaining session, the experimenter examined the written rationale to make sure that both teams complied with the instructions. A unilateral-discussion condition was created in order to separate unilateral experience from formal position preparation. The instructions were as follows: The next 40 minutes will be spent with your own team. We want you to use this time to gain an understanding of your situation and the issues. You are to discuss the issues among yourselves. While you may do some strategizing, you are not to plan a formal position to enter into bargaining with. In other words, it is not necessary to write anything down or to formulate a list of points defining the rationale or arguments for a team position. Just discuss the issues informally among yourselves. The purpose of this session is to give you a “frame of reference,” gained through familiarity with the issues, with which to enter the negotiations. In order to do this we feel that the most effective procedure is through informal discussion with one’s own team. Through observation of these sessions it appeared that participants were concentrating on supportive arguments for their team’s assigned positions, but formal position formation did not take place. In the bilateral-study condition, the bargainers were divided into three study pairs made up of a member from each team. The study pairs were separated into three corners of the laboratory. The composition of the dyads was determined randomly, with the restriction that the opposing representatives could not compete against each other in the bargaining session. (In the former study, bilateral study was not separated from communication with a future bargaining opponent. It is concrivable that this latter factor, apart from the activity of study, may have facilitated resolution.) The emphasis in this condition was on bilateral focus and mutual understanding. The instructions were as follows: The next 40 minutes will be spent in bilateral study. You are to use this time for learning as much as possible about Company and Union perspectives. We feel that the best way to achieve this is to split the teams up into study pairs made up of a member of each team. The pairs were randomly chosen before this session. Study the issues in order to gain

Prenegotiation Experience

63

understanding of both points of view as well as areas of greater or lesser agreement between proponents of either. Do not formulate or plan strategies for bargaining from either position. Do not take a position and argue its merits. The discussion should not break down into an inter-team competition. If it does the experimenter will stop it and remind participants of the goal. You should try to act bipartisan with respect to the issues during this session. There is no need to use any form of propaganda in order to obtain later gains. Openly discuss the issues. The emphasis is on informality and there is no one observing you or recoring data. Understanding is the goal! While the experimenter did not actively interfere with any study dyad, he maintained some degree of surveillance to make sure that the participants were “studying” and not bargaining or competing. A fourth condition was created in which bargainers had essentially no prenogotiation experience. All participants were given 10 minutes before the bargaining session to familiarize themselves, individually, with the issues and background information. No intra- or inter-team experience took place before bargaining. The purpose of this condition was to assess whether certain prenegotiation conditions facilitated resolution or whether other conditions produced a restraint on resolution attempts. This is referred to below as a control condition. Data from the unilateral-strategy condition and from the bilateral-study condition, reported in Druckman (1967), will be referred to. Bargaining outcomes from the two unilateral conditions created in this study were compared to the outcomes resulting from the unilateral-strategy condition in the former study. Also, bilateral study apart from a future bargaining opponent was compared to bilateral study with a future bargaining opponent, a condition created in the earlier study.

Selection of Participants In order to increase identification with the assigned positions, participants were chosen on the basis of their attitudes toward labor unious and management. Another selection criterion consisted of scores on a modification of the Dogmatism Scale. Items from the Labor-Management Scale and the Modified Dogmatism Scale were administered together to a large sample of Northwestern University undergraduate psychology students. Respondents were informed that we were interested in recruiting people with similar attitudes for participation in an interesting experiment. As soon as both scales were scored, eligible respondents (upper and lower quartiles on both scales) were recruited by phone and session times were arranged.3 Since dogmatic attitudes (or attitudes toward resistance to compromise) were found to relate to resolution behavior (Druckman, 1967), an equal number of high- and low-dogmatic sessions were conducted in each condition.

64 Flexibility in Negotiation

Also, an equal number of sessions with male and female participants were included in each condition. Each run of the simulation was homogeneous by attitude-disposition and by sex. There were twelve bargaining dyads per condition, three in each combination of sex and dogmatic attitudes. Since three dyads were included in each group session, there were two sessions with male participants, two sessions with female participants, two high-dogmatic sessions, and two low-dogmatic sessions, for each of the four prenegotiation conditions4 Across the four conditions, 96 subjects or 48 dyads participated. The mean score on the Labor-Management Attitudes Scale for the company representatives was 90.31 (V = .098), while the mean for the union representatives was 63.79 < V = .087).5

Procedure One intention of this investigation was to determine which factor was responsible for the large prenegotiation-experience effect found in the previous study. Thus it was necessary to maintain comparability by employing the same task, setting, subject population, and selection criteria. The following is a brief synopsis of the task, which is described in more detail in Druckman (1967). The prenegotiation condition used in this study necessitated the use of laboratory group representatives rather than individuals who represented themselves. Since the group-representation-self-representation variable accounted for a negligible portion of the variance on any measure of conf lict resolution, it is assumed that these findings have implications for bargainers representing themselves as well as for group representatives. The task was a non-zero-sum simulation of the collective bargaining process. Participants were assigned to a three-man union or company team of the hypothetical Acme Steel Corporation. They were told that this assignment was based on their attitudes as expressed on a questionnaire filled out some weeks earlier. Members assigned to the same team had similar attitudes. Earh team received four issues, each arranged on a 17-step continuum with the starting positions of each side at either end. Between the starting positions were possible compromises that might have been considered. The issues were wages, off-the-job training, hospital plan, and paid vacation. Below each position on the issue scale the estimated cost to the company for the duration of the contract was listed. In order to give bargainers a common perspective, a page of background information was provided. This contained a brief statement of union and company rationale for each of their demands, and “going rates” at four other similar hypothetical companies in the industry. However, the amount of information supplied was limited in an attempt to insure that the teams would be given sufficient latitude to be f lexible in deciding on a mutually agreeable position.

Prenegotiation Experience

65

The first phase consisted of the prenegotiation manipulation. Following this, members from both teams were brought together to rank, independently, the issues in terms of importance. Then the representatives from both teams were randomly divided into three bargaining dyads. In order to insure independence, the pairs were separated into three corners of the room and screens were placed around each table. Any given dyad was kept virtually unaware of the progress, situation, or completion of any other pair of bargainers. To make sure that team identification was maintained, the representatives were told that the results of all bargaining dyads would be combined to yield team totals. Thus each subject was bargaining for his team, and, perhaps, expected to be measured against how well his two colleagues fared. This competitive set was pitted against the advantages of a quick settlement. The latter pressure was created by informing the bargainers that each five minutes of debate time would result in a loss of $5,000 to each side in lost wages or profits due to the prolongation of the strike. Five minutes simulated one day of negotiation. The experimenter served as a “referee” and announced each five minutes of debate time. There was a 30-minute (or six-“day”) time limit on debate. If all issues were not settled in this period of time, deadlock ensued. Bargainers were provided with contracts. They were told to circle the position of settlement on any settled issue and to initial their contract beside the issue. If an issue was not settled at the deadline, they were asked to indicate, independently, how far they were willing to go in order to get an agreement on the issue. At the conclusion of bargaining all participants filled out postnegotiation questionnaires while the “simulation manager” combined the results across the dyads in order to provide feedback on team totals.

Results Conflict Resolution The means and standard deviations for six conditions of prenegotiation experience for each of three dyadic measures of conf lict resolution are presented in Table 7.1. The effect of the manipulated conditions was assessed across the two bargainers as well as across the four issues. The behavior of one bargainer was not independent of his opponent and the independence of the four issues during debate was questionable.

Speed of Resolution The harmonic mean time to complete the contract (the reciprocal of the mean of the reciprocals of the original values) was used as a measure of speed of resolution. A reciprocal transformation was performed on time to complete the contract in order to insure stability in the variances (Edwards, 1960) and so that deadlocks

66 Flexibility in Negotiation

could be taken into consideration. Deadlocks were assumed to be settled in an infinite amount of time. Since the reciprocal of infinity is zero, deadlocks could be treated as zero values and included in a distribution with the reciprocals of all the other obtained values. The harmonic mean times (see Table 7.1) are inf lated due to deadlocks. The differences in speed of resolution between the four conditions of this study are significant (F = 3.14, 3 and 44 df, p < .05). Unilateral position-formation resulted in slower resolution than did unilateral discussion (t = 2.19, 22 df, p < .05)6 or bilateral study apart from a future bargaining opponent (t = 2.58, 22 df, p < .02). The difference in speed of resolution between unilateral discussion and bilateral study apart from a bargaining opponent is small and nonsignificant. Dyads in the control condition were slower to resolve the conf lict than those in both the unilateral-discussion (t = 1.86, 22 df, .05 < p < .10) and bilateralstudy conditions (t = 2.13, 22 df, p < .05). However, the difference between the controls and the unilateral position-formation condition is nonsignificant. The hypothesized effect of the two unilateral conditions in producing both more (in the case of unilateral position-formation) and less (in the case of unilateral discussion) resistance to resolution than the unilateral strategy condition, reported in Druckman (1967), is supported. The harmonic mean for the latter condition fell between the means for the two conditions created in this study. The difference between the unilateral position-formation (N=12 dyads) and unilateral-strategy conditions (N = 30 dyads) approached significance (t = 1.77, 40 df, .05 < p < .10). Also, bilateral study with a future opponent resulted in slightly faster resolution than bilateral study apart from a future opponent, but the difference is nonsignificant (t = 1.01) .

TABLE 7.1 Means and Standard Deviations for Six Conditions of Prenegotiation Expe-

rience on Three Measures of Conf lict Resolution  

Speed of resolution (minutes)

Average distance Average apart yielding

Condition

Mean

S.D.

Mean

S.D.

Mean S.D.

.018 .019 .022 .030 .019

2.021 1.521 1.592 .458 .396

2.018 1.646 2.249 .714 .808

7.365 7.323 7.292 7.771 7.844

.042

.458

Unilateral position-formation 101.85 Control 78.74 Unilateral strategya 49.93 Unilateral discussion 31.56 Bilateral study apart from future 33.92 bargaining opponent Bilateral study with future bargaining 23.55 opponent a a

.909 .716 .982 .357 .366

.833 7.771 .416

These data are from Druckman (1967) and are based on 30 dyads in each condition.

Prenegotiation Experience

67

Average Distance Apart and Average Yielding These measures are represented in Table 7.1 as dyadic means taken across all four issues. Distance apart refers to the number of steps on an issue scale (each issue was arranged on a 17-step scale) separating the bargainers’ final positions at the conclusion of bargaining, while yielding refers to the average number of steps on the scale that a bargaining dyad moved from their respective starting positions. The differences between the four conditions of this study, on average distance apart, are significant (F = 3.89, 3 and 44 df, p < .05). Bargainers were further apart after unilateral position-formation than after either unilateral discussion (t = 2.53, 22 df, p < .02) or bilateral study apart from a future bargaining opponent (t = 2.59, 22 df, p < .02). Also, the control dyads were further apart than those in both the unilateral-discussion (t = 2.05, 22 df, .05 < p < .10) and bilateral study conditions (t = 2.13, 22 df, p < .05), but the difference between the controls and unilateral position-formation was nonsignificant (t = .67, p > .10). As with the speed-of-resolution measure, the average distance apart for the unilateral-strategy condition, reported in Druckman (1967), fell between the means for the two unilateral conditions created in this experiment. The difference between unilateral strategy (N= 30 dyads) and unilateral discussion (N = 12 dyads) approached significance (t = 1.70, 40 df, .05 < p < .10). The mean differences between unilateral discussion, bilateral study apart from a bargaining opponent, and bilateral study with a bargaining opponent are small and nonsignificant. A similar pattern of findings was obtained for the average-yielding measure. The control dyads were more resistant than the unilateral-discussion (t = 1.94, 22 df, p < .10) and bilateral-study conditions (t = 2.24, 22 df, p < .05), but did not differ from the unilateral position formation dyads. While the unilateral position-formation dyads were more resistant to yielding than dyads in the unilateral-discussion and bilateral-study conditions, the differences did not quite reach significance. Also, unilateral discussion did not differ from bilateral study in average yielding.

Number of Unresolved Issues Table 7.2 shows the number of contracts or dyads with 0, 1, 2, 3, or 4 unsettled issues for each condition. A Kruskel-Wallis one-way analysis of variance (see Siegel, 1956) showed the differences between the conditions to be highly significant (H = 13.60, 3 df, p < .01). The results of Mann-Whitney U tests, corrected for ties, indicate that unilateral position-formation led to a higher rate of unsettled issues per contract than did unilateral discussion (U=42, p .05), distance apart (r = -.33, p < .05), yielding (r = .20, p > .05), and number of unresolved issues (rho = -.35, p < .05). The extent to which the debate was considered as a “win-lose” contest was related to each measure as follows: speed (r = .34, p < .05), distance apart (r = -.60, p < .01), yielding (r =.50, p < .01), and number of unresolved issues (rho =.43, p < .01). It appears that the tendency to consider the bargaining task as a “win-lose” contest with both sides committed to positions was strongly related to resolution behavior, and was also characteristic of bargainers in the position-formation condition and, to a lesser extent, of those in the control condition. Perceived commitment to the assigned positions was also related to resolution behavior but the correlations were not so large.

Consensus Among Teammates on the Ranked Importance of the Issues Another hypothesized effect of formal position-formation is that it produces team solidarity, as ref lected in agreement among members with regard to stands on issues and the relative importance of issues. An index of the extent to which consensus is produced among team members as a result of prenegotiation experience is agreement on the ranked importance of the issues. After the prenegotiation session, members from both teams were brought together to rank, independently, the issues in terms of importance. Position-formation may be expected to produce more consensus on ranked importance than unilateral discussion. If, as a result of unilateral discussion, team members “focus on the ingroup positions,” they should be more consensual on the ranking of the issues than teams from the bilateral-study or control conditions, where

70 Flexibility in Negotiation

there was no ingroup experience prior to bargaining. That is, team experience per se may be expected to produce some degree of consensus with regard to the issues. Thus unilateral position-formation should have resulted in more agreement among team members than unilateral discussion and bilateral study in that order. The controls would not be expected to differ from bilateral study. The method for determining the extent of agreement among three members of a team (there were eight teams per condition) was to take the absolute differences between the ranks assigned each issue by each member and to sum the differences. The “agreement scores” were ranked, and Mann-Whitney U tests were performed (see Siegel, 1956). The hypothesis was confirmed. Unilateral position-formation before bargaining resulted in significantly more agreement among team members on the ranked importance of the issues than did unilateral discussion (U = 16.5, p < .052), bilateral study apart from a future bargaining opponent (U = 5.5, p < .002), and the control condition (U=1, p < .001). Unilateral discussion resulted in more agreement among team members than bilateral study (U = 17.5, p < .065) and the control condition (U=12, p < .02). The controls did not differ from bilateral study (U=26.5, p > .10). Thus, as predicted, the order of the conditions is as follows: UPF > UD > C= BS. Also, unilateral position-formation resulted in more agreement among team members than the unilateral-strategy condition reported in Druckman (1967) (U=24, N1=8, N2 =20, p < .01). Perhaps it is even more interesting to note that there was perfect agreement among the three members of almost every team in the unilateral position-formation condition. Clearly, this condition produced dramatic effects on team consensus as ref lected in private and independent ratings. Amount of consensus was not unrelated to other characteristics of the “winlose” syndrome. Amount of perceived commitment to positions and the extent to which the debate was considered to be a “win-lose” contest were averaged over the three members of each team and correlations7 were computed between these variables and amount of consensus. The average perceived commitment to the positions by team members was strongly related to the amount of consensus on the issue ranking (rho= .61, p < .01). The amount of team consensus was also related to the extent to which the debate was considered as a “win-lose” contest (rho=37, p < .10).

Discussion The immediately apparent contrast suggested by these findings is that between the unilateral position-formation and control conditions on the one hand and the unilateral-discussion and bilateral-study conditions on the other. When teams prepared strategies and attempted to justify their assigned positions before debate, their representatives found it difficult to achieve agreements. Nine out of twelve bargaining dyads ended in deadlock and four of these could not agree on any of

Prenegotiation Experience

71

the four issues. However, this high rate of resistant behavior did not differ significantly from the resolution behavior of bargaining dyads in the control condition. Representatives who bargained without a prenegotiation session were also highly resistant to conf lict resolution. Eight out of the twelve bargaining dyads in this condition ended in deadlock. The provision of certain types of prenegotiation experience apparently facilitated the achievement of resolution. Opponents who bargained after consultation or informal discussion with members of their own team or with members of the opposing team had little trouble in reaching agreements and were more cooperative than the control dyads. A very interesting and, perhaps, unexpected finding was that bargaining dyads, after informal discussion with teammates, were only slightly (and nonsignificantly) more rigid than those dyads whose prenegotiation experience involved consultation with a member of the opposing team using a bilateral-focus technique. An implication of these results is that informal discussion before debate that is concentrated on the issues is a technique for inducing cooperation. Whether the informal discussion consists of unilateral focus with teammates or bilateral focus with opponents seems to make little difference in bargaining outcomes. (This finding is attributable to the greater amount of cooperation in the unilateraldiscussion condition, rather than to the lesser amount of cooperation in the bilateral-study conditions.) These results do not, however, imply that the activity of generating supportive arguments and bargaining strategies predisposes bargainers to be more competitive. It is interesting to note the parallel between this pattern of results and that obtained by Harrison and McClintock (I965) in a study which employed a prisoner’s dilemma game. These investigators explored the relationship between reward for previous cooperative behavior between members of a dyad and cooperative choices on the prisoner’s dilemma task. They also manipulated the time period (immediate versus delayed) between the previous dyadic interaction and the game. Significantly more cooperative responses were made in the reward-immediate, reward-delay, and the nonreward-delay conditions than in the nonreward immediate condition. However, the nonrewardimmediate condition did not differ from a control group where members had no previous interaction experience. Thus, the nonreward-immediate condition did not increase competition but the former conditions served to induce cooperation. The unanticipated high level of cooperation demonstrated by dyads in the nonreward-delay condition of the Harrison and McClintock (1965) study parallels the obtained, and unanticipated, high level of cooperation by dyads in the unilateral-discussion condition of this investigation. Rapoport’s (1961) notion of bilateral focus as a conf lict-resolving mechanism is only partially supported. The bilateral-study conditions did produce more cooperation than the control and position-formation conditions. However, it appears that a Rapoport-type debate is not much better than unstructured discussion among teammates. As mentioned earlier, Hammond et al. (1966), investigating noncompetitive conf lict, also failed to obtain a difference between an

72 Flexibility in Negotiation

“Own Focus” and a “Region of Validity” condition. While there was clearly no difference between these conditions in this study that approached significance, on any measure of conf lict resolution, there was a slight tendency for bargainers in the unilateral-discussion condition to be somewhat more resistant to compromise. Perhaps if a “technique” for unilateral focus were more clearly spelled out in instructions for the unilateral-discussion condition, bargaining outcomes for dyads in this condition may have more closely resembled those obtained in the control condition. It is conceivable that the lack of difference between unilateral discussion and bilateral study ref lected the tendency of the simulated representatives, in both conditions, to engage in similar activity. There is evidence which indicates that the structural arrangement of groups (viz., unilateral versus bilateral) per se, apart from unilateral or bilateral focus, makes little difference in resolution outcomes. Bass (1966), in a recent study that employed a similar labor-management bargaining task, found no difference in speed of resolution between unilateral-study and bilateral-study conditions. Bargainers in both unilateral and bilateral arrangements were instructed to focus on and gain understanding of both positions. Thus it appears that unilateral focus among teammates without formal position-preparation facilitates resolution as much as unilateral or bilateral study. It is also interesting to note that in the Bass (1966) study, bargainers from a unilateral-strategy condition, similar to the one created in Druckman (1967), were slower in resolving the conf lict than those in three types of group-study conditions – with participants representing the same position, with members representing opposing positions, and bilateral-study groups including bargaining opponents. In this investigation a similar pattern obtained, with bargainers from the unilateral-strategy condition being somewhat slower in resolving the conf lict than those in the unilateral-discussion and bilateral-study conditions.8 A question may also be raised concerning the generalizability of these results to groups with a history of unilateral interaction (e.g., see Sherif et al., 1961). Perhaps, in more cohesive groups, simple group experience (with or without explicit unilateral focus) may be enough to produce loyalty to assigned group positions as ref lected in resistance to conf lict resolution. At the current stage of empirical development it is difficult to assess the relevance of findings from simulated groups for groups with their own developmental history. Several new unilateraldiscussion conditions may be established and compared to the prenegotiation conditions created in this design. For example, members of the same relatively permanent preexperimental group (e.g., members of the same fraternity, athletic teams, bunkmates, campus organizations), varying in terms of degree of familiarity among themselves, may be assigned to the same laboratory team. Another possibility is that of establishing laboratory groups that consist of members who have worked together for several laboratory sessions on a number of tasks unrelated to the bargaining situation. It would be interesting to determine whether these additional manipulations increase resistance to resolution more than does the unilateral-discussion condition created in this study.

Prenegotiation Experience

73

While the position-formation condition produced more perceived commitment to the assigned positions than did the control condition, bargainers in the former condition were only slightly more resistant to compromise. Perhaps the lack of difference in resolution behavior between these conditions may have been, to some extent, a function of the imposed 30-minute deadline. Conceivably, dyads in both conditions may have reached close to an upper limit of resistance within the confines of a time limit. The results of the postnegotiation question on perceived commitment perhaps indicated that the position-formation dyads had more “potential” for rigidity. (Also, four dyads in this condition did not settle any issue at the deadline as compared to only one control dyad.) Differences between these conditions may have emerged if there were no deadline. This might be explored. A prebargaining atmosphere which leads to a consideration of the debate as a collaborative, problem-solving venture in which neither side is committed to unilateral positions appears to facilitate conf lict resolution. Prenegotiation conditions which consist of open discussion of the issues before representatives have a chance to develop a rationale or a cognitive context for their own positions appear to constitute such an atmosphere. However, as indicated in the research of Blake and Mouton (1962), bilateral study or focus which occurs after position-taking is not effective in reducing rigidity. Also, it seems to make little difference whether this activity is carried out with a future bargaining opponent, opposing representatives who are not bargaining opponents, or teammates. (While communication with one’s opponent before bargaining resulted in somewhat faster resolution of the conf lict than bilateral study apart from a future opponent, the difference between these conditions was not strong enough to be significant on any measure.)

Notes 1 Blake and Mouton (1962) observed adult representatives negotiating over which group produced the best product while Sherif et al. (1961) manipulated competition between children’s camp groups. In the former study, deadlock was the invariable result. This finding has been replicated by others, notably by Bass and Dunteman (1963). 2 The latter instructions represented an attempt to separate unilateral experience from unilateral formal position-formation. The unilateral-strategy condition, created in Druckman (1967), was perhaps ambiguous with regard to the specific means to be used in strategy preparation. The nature of the activity of team members during the unilateral strategy session was not homogeneous across all sessions. It appeared that while some teams were putting in a great deal of effort in preparing formal positions for debate, other teams spent most of the allotted time discussing the issues from a unilateral perspective without preparing formal positions. Each of these activities is perhaps represented in the two unilateral conditions created in this study. Team members in the unilateral position-formation condition were to prepare formal positions, while informal discussion characterized those in the unilateral-discussion condition. Thus, it might be expected that the former condition would produce more rigidity while the latter condition would result in less rigidity than the unilateral-strategy condition. 3 The correlation between the scales was .01. There was thus no trouble in recruiting prolabor and pro-company respondents who scored at both ends of the Modified Dogmatism Scale.

74 Flexibility in Negotiation

4 Crowne (1966) presented data showing a relationship between family orientation and competitive choices in a prisoner’s dilemma game. In an attempt to determine whether this relationship held for a complex bargaining situation, dyads were also homogeneous with regard to family orientation. An equal number of “entrepreneurial” and “bureaucratic” pairs were placed in each condition and for each sex (i.e., this variable was not confounded with condition or sex). No relationship was found between this variable and any measure of resolut.ion behavior. 5 V is the coefficient of variability (V =s/x¯ ). These means are almost identical to those of the previous study (x¯ = 89.47 and x¯ = 62.73). Also, the meaus for the high dogmatic and low-dogmatic participants were very similar to those reported in the earlier study (145.13 versus 143.70 and 113.31 versus 112.96). 6 All of the reported probability levels are for two-tailed tests. 7 Since “amount of consensus” was considered to be an ordinal variable, Spearman rankcorrelation coefficients (rho) were computed. 8 In the Bass (1966) study the type of activity was the same for both unilateral and bilateral arrangements, while the unilateral-discussion condition of this study emphasized ownposition focus. The bilateral-study conditions in Bass (1966) consisted of eight-man study groups, while direct confrontation between a union and a company representative in study dyads was the procedure used in both study conditions of this investigation. Laboratory teams were not created in the former investigation.

References Aronson, E. The effect of effort on the attractiveness of rewarded and unrewarded stimuli. Journal of Abnormal and Social Psychology, 1961, 63, 375–380. Bass, B. M. Effects on the subsequent performance of negotiators of studying issues or planning strategies alone or in groups. Psychological Monographs, Whole No. 614, 1966. Bass, B. M., and Dunteman, G. Biases in the evaluation of one’s own group, its allies and opponents. Journal of Conflict Resolution, 1963, 7, 16–20. Blake, R. R. Psychology and the crisis of statesmanship. American Psychologist. 1959, 14, 87–94. Blake, R. R., and Mouton, Jane S. The intergroup dynamics of win-lose conf lict and problem-solving collaboration in union-management relations. In M. Sherif (Ed.), Intergroup relations and leadership. New York: Wiley, 1962. Pp. 94–140. Crowne, D. P. Family orientation, level of aspiration, and interpersonal bargaining. Journal of Personality and Social Psychology, 1966, 3, 641–645. Druckman, D. Dogmatism, prenegotiation experience and simulated group representation as determinants of dyadic behavior in a bargaining situation. Journal of Personality and Social Psychology, 1967, 6, 279–290. Edwards, A. L. Experimental design in psychological research. New York: Holt, 1960. Ferguson, C. K., and Kelley, H. H. Significant factors in overevaluation of own-group’s product. Journal of Abnormal and Social Psychology, 1964, 69, 223–228. Festinger, L., and Aronson, E. The arousal and reduction of dissonance in social contexts. In D. Cartwright and A. Zander (Eds.), Group dynamics. (2nd ed.) New York: Harper & Row, 1960. Pp. 214–231. Gladstone, A. I. Relationship orientation and processes leading toward war. Background, 1962, 6, 13–25. Hammond, K. R., Todd, F. J., Wilkins, Marilyn, and Mitchell, T. O. Cognitive conf lict between persons: Application of the “lens model” paradigm. Journal of Experimental Social Psychology, 1966, 2, 343–360.

Prenegotiation Experience

75

Harrison, H. A., and McClintock, C. G. Previous experience within the dyad and cooperative game behavior. Journal of Personality and Social Psychology, 1965, 1, 671–675. Lewis, M. Psychological effect of effort. Psychological Bulletin, 1965, 64, 183–190. Pearson, L. B. Diplomacy in the nuclear age. Cambridge, Massachusetts: Harvard Univ. Press, 1959. Rapoport, A. Fights, games, and debates. Ann Arbor, Michigan: Univ. of Michigan Press, 1961. Rappoport, L. H. Interpersonal conf lict in cooperative and uncertain situations. Journal of Experimental Social Psychology, 1965, 1, 323–333. Rusk, D. Parliamentary diplomacy - debate versus negotiation. World Affairs Interpreter, 1955, 26, 121–138. Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., and Sherif, Carolyn W. Intergroup conflict and cooperation: The robbers cave experiment. Norman, Oklahoma: University Book Exchange, 1961. Sherif, M., and Sherif, Carolyn W. Research on intergroup relations. In O. Klineberg and R. Christie (Eds.), Perspectives in social psychology. New York: Holt, 1965. Pp. 153–177. Siegel, S. Nonparametric statistics for the behavioral sciences. New York: McGraw Hill, 1956. Withey, S., and Katz, D. The social psychology of human conf lict. In E. B. McNeil (Ed.), The nature of human conflict. Englewood Cliffs. New Jersey: Prentice Hall, 1965. Pp. 64–90.

8 BOUNDARY ROLES1

The concept of a boundary role was introduced by Walton and McKersie (1965) as part of their intraorganizational model of labor-management bargaining. They define the concept in terms of two forces, a pull in the direction of the opponent’s expectations and a pull in the direction of the bargainer’s own party. They refer to these forces as conf licting expectations, which include objectives, priorities, aspirations, and behavior. Their broad treatment of these concepts is grounded in examples from diverse cases mostly involving unions. The article in this section extends and refines the idea of a boundary role. The extension places the idea in an international negotiation context. The refinement consists of focusing on the information-processing requirements of boundary roles, referred to as the monitoring function (Druckman, 1978). This function is expressed in the form of two mathematical models, the negotiator as bargainer and as representative. The bargainer model, with its focus on the opponent, consists of alternative concession-making algorithms. The representative model, with its focus on one’s own side, highlights two processes referred to as building a package and degree of agency.2 The concession-making dynamics captured in this article led to a number of studies on responsiveness. The most interesting finding from these analyses is the generality of a particular pattern similar to Version 2 in the boundary role article. This pattern was discovered in the study on base rights negotiations, reprinted below in the section on turning points, and is referred to as “thresholdadjustment.” When compared to nine other models of responsiveness, including tit-for-tat, this was the best predictor of measured concessions (Druckman and Harris, 1990). Bargainers in these international cases gauged the difference between their own and their opponent’s concessions in the previous round. Based on this assessment they matched the other’s move usually in a tougher direction. They did not make their moves based on a monitoring of progress over DOI: 10.4324/9781003293361-14

Boundary Roles 77

a long period of time. This pattern is in tune with Coddington’s (1968) model of holding expectations, evaluating those expectations and adjusting to maintain synchrony during the process. More generally, these findings have implications for equilibrium concepts in game theory (Guner and Druckman, 2000) and for situational effects in social psychology (Druckman, 2008). The representative dynamics discussed in the boundary role article were explored in many studies to follow. Bargainers sought the package that ref lected their preferences for the various issues and in the process weighed the relative importance of its parts. The research on agency takes into account the role played by a negotiator’s accountability to the group being represented. The finding that accountability pressures are more likely to cause impasses when interests are at stake but less likely to detract from agreement when values or ideologies are being contested is interesting. (Mittelmark et al., 1977). These findings call attention to the way conf lict issues are framed usually during prenegotiation phases. But there are other aspects of representation that play large parts in processes (as competitive or cooperative) and outcomes (as settlements or impasses). These include reputation costs for the representative, the composition of constituencies as hawks or doves, the stakes of the talks, and the time horizon of the relationship (Druckman, 1994). A more complex description of boundary roles is provided by Hopmann (1977). His study of alliance negotiations revealed that coordination was needed at several levels during the process. In addition to across and behind the table, the delegates had to work with each of the alliance member countries. These levels are characteristic of multilateral negotiations that are divided into blocs. Another tier of coordination, present also in bilateral talks, is interagency coordination within the countries being represented. Thus, you can see how the modeling can get very complex with multiple preferences to juggle. It is a wonder that agreements emerge from all the wrangling that goes on. When the conferences are global, a way out is to find the minimum consensus or the vaguest terms that satisfy the “committee of the whole” (Monheim, 2015). Similar coordination issues arise on the larger plane of foreign policy making. Putnam’s (1988) two-level game framework owes much to the boundary role literature in negotiation. His framework deals with an interplay between domestic and international politics which include, but are not limited to, negotiation processes. Negotiation may be regarded as a microcosm of policy making. The pressures on policy makers are similar to those experienced by diplomats. The policy-making bureaucracies are however much more complex than those that are responsible for conducting diplomatic negotiations. Boundary Role Conf lict: Negotiation as Dual Responsiveness, Daniel Druckman. Journal of Conflict Resolution 21 (4) (1977): 639–662. © Sage Publications, Inc. Reprinted with permission.    

BOUNDARY ROLE CONFLICT Negotiation as Dual Responsiveness Daniel Druckman

The modern-day international negotiator takes part in a process that has become the primary forum for articulating relationships among nations. He is a participant in an exercise that is designed to manage complexity in the international system, and he functions primarily as a problem solver. His contributions consist of synthesizing, or mediating among, the diverse claims made by other actors in the ongoing process of international relations. These functions and contributions are construed as role obligations. In the bilateral case, the negotiator is obligated to be responsive to the competing claims of his own and the other side. This dual responsiveness may be regarded as the defining feature of negotiation, referred to in the literature as a boundary-role conflict (BRC; see Walton and McKersie, 1965: 283 ff.). The BRC can be represented as psychological complexity – as a juxtaposition of images that the negotiator has of the interface between the opposite-number representatives, of the own-nation/other-nation relationship, and of the relationships among the various factions within his own party, and his own posture (Perry, 1957). Some commentators have described this problem as two-track negotiating – i.e., negotiating simultaneously with the bureaucracy and with the other side. The SALT process is perhaps the most illustrious example of such negotiations (see, e.g., Newhouse, 1973; Wolfe, 1975). The bidirectional responsiveness, ref lected in the juxtaposition of images is the monitoring function in international negotiation: where should we be? where are they? Negotiators monitor their opposite numbers; they also monitor their constituents (Winham, 1977a). The extent to which they are responsive to one or the other affects the course of a negotiation. The direction of responsiveness emphasized by a negotiator has implications for the way in which information

Boundary Roles 79

is processed and presented in international conferences. These implications are developed here in terms of two general models of negotiation in the bilateral case: bargaining and representation.

The Negotiator as Bargainer The bargaining model assumes responsiveness between opposite number negotiators. In its simplest form, the model assumes that each bargainer’s concession is a direct and calculated response to the other’s previous concession in an action-reaction cycle. The only information needed to make a concession calculation is the size of the other’s concession (see Druckman, forthcoming). This model can be subsumed under a more general model which assumes that a bargainer is responsive to both his own and the other’s previous concessions. According to this model, a bargainer’s response is a function of his own previous pattern of concession-making as well as the other’s concession rate. This two-parameter model is presented here as one version of the negotiator as bargainer. A second variant is a more complex version. It is based on Coddington’s conceptualization of an expectation/evaluation/adjustment process; it assumes, inter alia, indirect responsiveness. Each is discussed here in turn.

Version 1 Whether it pays for one to be tough or soft may depend both on the other’s response and on his (the other’s) own pattern of concession making. This variant claims that the other responds both to one’s previous demand-direct responsivenessand to his own previous demand-internal responsiveness. Put another way, a change in offers, a concession, is linear with the bargainer’s own last offer. A n i-1, and his opponent’s last offer Bn i-1. This can be expressed as A ni − A ni -1 = aA ni -1 + bBni −1 + ca,b > 0,

(1)

where n is the round number, i is the offer number, a and b are concession-rate parameters, and c is a constant. If terms are transposed, this equation can be written in simpler form: A ni = aA ni -2 + bBni −1 + c

(2)

Expression 2 has the advantage of being stated in a form that can be evaluated by multiple-regression procedures.

80 Flexibility in Negotiation

It should be noted that these expressions are similar to Bartos’ (1974) adaptation to negotiation of the Richardson arms-race model. (The main difference between his expressions and ours is that we refer to changes in offers while he focuses on changes in demands.) Instead of depicting the defense budgets of two enemy nations, Bartos refers to the negotiating demands made by two bargaining opponents. However, following Bartos, it is important to note two assumptions that are made when the model is applied to bargaining: (1) that a>0 – i.e., a concession will be made if the bargainer’s last offer A n i-2 is high, and (2) that b> 0 – i.e., a concession will be made if the other’s last offer Bn i-1 is low. Bartos goes on to say that: “Of the two, assumption (2) is more fundamental for two reasons: First, . . . it amounts to assuming that the process of negotiation is reciprocative. Second, it is obvious that once (2) is made, (1) must be made also if there is to be any negotiation at all” (1974: 77). He then presents a formal justification for the assumption that bargainers reciprocate concessions. For our purposes, the justification provides an hypothesis to be evaluated experimentally and in situ. We turn now to a consideration of this evidence. Bartos’ data only partially support the adapted Richardson model. The tendency to reciprocate concessions was weak and occurred only under certain conditions. Bargainers were just as likely to exploit each other’s concessions as they were to reciprocate: the tougher was a negotiator’s opponent, the softer did he tend to become. Yet neither the tendency to reciprocate nor the tendency to exploit accounted for a substantial portion of the variance in the bargainer’s concessionmaking behavior. The size of his concessions was as much a function of his own previous behavior (aA n i-2), taking both own and other parameters into account. This two-parameter model improves predictability over the single parameter of a simple action-reaction model (Druckman, forthcoming). The importance of a bargainer’s own previous behavior is highlighted by the series of within-bargainer correlations obtained in two other laboratory studies. Recent analyses of data collected for the Druckman et al. (1972) and Druckman and Bonoma (1976) bilateral monopoly experiments indicate that: (a) the sequential-phase correlations between early and late concessions ranged from .23 to .55; (b) earlier concessions accounted for a large portion of the variance in total concessions, with about 40% accounted for by phase 1 concessions (r = .63), and 64% accounted for by concessions in phases 2, 3, and 4 (r = .81; .82, and .80, respectively); and (c) a bargainer’s concession on the first trial, before the other’s programmed concession rate began, predicted concession-making in later phases as well as total amount of concessions, with r’s of -.62 between trial 1 and total, -.35 between trial 1 and phase 3, and -.42 between trial 1 and phase 4. (The more the concesson in trial l, the less the total concession as well as concessions in

Boundary Roles 81

later phases.) The third finding is especially interesting since trial 1 concessions were independent of the other’s bargaining behavior. Taken together, these results support the contention that bargainers establish their own concessionmaking pattern apart from the pattern established by their opposite number. The above evidence calls attention to the importance of an internal dynamic in bargaining. It does not, however, argue against the notion of mutual responsiveness in negotiation. Responsiveness is likely to occur in both directions. An interesting implication drawn from Bartos’ analysis is that eventually bargainers arrive at an equilibrium in concessions. It consists of a gradual discovery of the appropriate ratio of concessions, expressed by Bartos as C∗ ad + bd = C∗ ad + bd

(3)

where the C*s are the changes in demand for bargainer (C*) and other (C*), b is the concession-rate parameter from 2 above, and d is a new parameter added to the model in order to take time into account (referred to by Bartos as a modified Richardson model). While it may take some time before this ratio is established, once it is determined it remains unchanged throughout the remainder of the bargaining process. The determination of such a ratio is perhaps the strongest evidence in support of the notion of mutual responsiveness in negotiation. Some support for mutual responsiveness was obtained in the Bartos (1974) and Druckman et al. (1972) experiments. Sequential demand and concession correlations suggest that bargainers did respond to the immediately previous demand (concession) of their opponent. However, such responsiveness was weak: low correlations and limiting conditions (smaller bargaining groups in Bartos, 1974; early bargaining behavior in Druckman et al., 1972) indicate that factors different from the other’s demands and concessions affected concession-making. Stronger evidence was obtained in a recent dissertation study by Stech (1977). He found a tit-for-tat process such that the other’s competitive (intermediate, cooperative) bid at n - 1 was followed significantly more often by S’s competitive (intermediate, cooperative) bid at n. Moreover, a number of PDG studies reported that subjects are responsive to shifts in the other’s strategy (see the review by Druckman et al., 1972). However, none of these studies was an attempt to determine the relative amount of variance accounted for by own versus other responsiveness. A first attempt to do this is represented by the study recently completed by Love et al. (1977). Love et al. (1977) designed a scenario in which subjects represented groups with different preferences for the outcome of a negotiation over the allocation of resources to diverse programs. The size of the conf lict of interest between the groups and the extent of division over positions within the groups were manipulated. The effects of these variables on amount of concession-making, time to resolution, and type of settlement were assessed. The results were clear. The conf lict

82 Flexibility in Negotiation

of interest variable accounted for a significant portion of the variance on each measure of conf lict resolution; the within-group division variable accounted for a negligible amount of variance. Subjects were more responsive to the differences between them then to the extent of polarization within the groups they represented. Moreover, perceptions of the amount of conf lict, willingness to compromise, and so forth were inf luenced significantly by the extent of differences between them. This study is only a first step. The scenario will be used further to examine the conditions that affect the extent of external versus internal responsiveness. But, what of negotiations conducted in situ, or simulations of such situations? Here, there is evidence from several studies that bears on the question of responsiveness. The evidence is provided by analyses of negotiations as diverse as early superpower arms control deliberations toward a ban on testing nuclear weapons and efforts between Egypt and the West to cooperate in achieving a project that could aid economic development along the Nile, inter alia. Most interesting perhaps is the recent study by Hopmann and Smith (1977). These investigators attempted to determine the extent to which the Richardson model depicted the process of the Partial Nuclear Test Ban Treaty of 1963. The results generally support the interactive process depicted by the model. Both sides were responsive primarily to their opponents – the effects were strong and significant. However, responsiveness was not symmetrical – i.e., the two sides (the United States and the Soviet Union) were not equally responsive. Specifically, the U.S. negotiators were more responsive to their opposite-number (Soviet) delegation than were the Soviets to them. The Soviet negotiators, on the other hand, were somewhat responsive to their own prior behavior, while for the United States this effect was negligible. Other differences between the two delegations included the effects of external events and perceptions: the U.S. negotiators were affected more than the Soviets both by events outside the conference and by their perceptions of Soviet actions inside the conference. The latter finding has implications for the second variant of our bargaining model and will be discussed below. Analyses of other international negotiations suggest, however, that the titfor-tat exchange observed by Hopmann and Smith may not be a general pattern. Spector’s (l976a) analysis of the Aswan Dam negotiations indicates that both sides (the West and Egypt) were responsive primarily to their own objectives rather than to the other’s moves. High intraparty correlations indicated little change in Western or Egyptian positions or behavior (i.e., cooperative versus hostile postures) through the course of the negotiation. Bargaining strategies were strongly related to bargaining positions: the greater the tendency to stop negotiating without agreement, the greater the use of a hostile strategy. Correlations between Egypt’s behavior and the West’s subsequent behavior (and vice versa) were weak. Neither side responded to the other’s inf luence attempts. Egypt’s perceptions were independent of the West’s strategy, and the West’s subsequent behavior was unrelated to previous Egyptian strategy. This strong internal dynamic, namely, each side’s own

Boundary Roles 83

bargaining position inf luenced its perceptions which inf luenced its objectives, led the author to claim that the problem of the negotiation was communication or a “total lack of comprehension of the other side’s position and behavior” (Spector, 1976a: 370). However, evidence for an internal dynamic may also take the form of changes in strategy. The changes may be a function of altered objectives or external events rather than the other’s behavior, as noted in the author’s parallel (simulate/in situ) studies. In one study, negligible or inverse intraparty correlations between early and late behavior by each side (construed as toughness/ softness) indicated changed strategies (r1 = -.34, NS; r2 = -.07, NS). The f luctuations in strategy were different for each party to the negotiation: changes were not synchronized between the parties; when one side was tough, the other was conciliatory, as indicated by a high negative correlation between percent tough for one team and the other. Moreover, the activities of the two sides were unsynchronized: when one side attempted to bargain, the other was seeking instructions or debating the merits of general principles. Each side’s strategy was shown to ref lect a cultural style or to result from responsiveness to particular types of inf luences. The eventual agreement did not emerge from a process of mutually converging concessions. Rather, it occurred as a result of time pressures and external events. Yet, there was also some evidence for a pattern of reciprocation. The pattern was asymmetrical: lagged correlations between one side’s posture at n-1 and the others’ at n indicated that one side reciprocated the other’s moves to a large extent but not vice versa (r1-2 = -.20, NS; r2-1 = .86, 14 df, p < .01). This pattern may be referred to as lead-lag: one team followed the other as if in a leader-lagger relationship. While these results argue against mutual reciprocation, they do not refute the more general notion of responsiveness in negotiation. This study lends support to the two-parameter model proposed above. Both an internal negotiating dynamic and a lead-lag reciprocal pattern were obtained. The evidence reviewed above makes plausible a model that takes the form of version l. However, the extent to which negotiators were responsive to their own or other’s moves varied from study to study. The second version is a more sophisticated model of responsiveness. It entails more complex information-processing requirements. We turn now to a description of this model.

Version 2 The second version also assumes mutual reponsiveness between opposing bargainers; however, unlike the previous model, responsiveness in this case is indirect. It is a complex, nonlinear function of perceptions or expectations and evaluations, and can be represented schematically as shown in Figure 8.1. This type of model, stated originally by Coddington (1968) as a general conceptualization of two-person bargaining, ref lects well the dynamic of

84 Flexibility in Negotiation

Bargainer i decision

Bargainer j decision

expectation

expectation

adjustment

adjustment

FIGURE 8.1

The Adjustment-Expectation-Decision Cycle

international negotiating in the bilateral case. It is also similar to Iklé’s (1964) characterization of international negotiations in terms of “shifting evaluations.” He noted that agreement often comes about because changing evaluations lead to changing expectations of the other’s willingness to compromise, which in turn softens (or hardens) a negotiator’s position and produces a selffulfilling prophesy. But he also cautions that this cycle can lead to undesirable consequences, as when the West came to expect the Sovietization of Czechoslovakia (which came about) but remained firm on Soviet control of northern Iran (which the Soviets relinquished). Recent analyses of laboratory data and selected cases provide documentation for the process depicted by this model. It would be instructive to present the evidence before elaborating the process further. This pattern was obtained in the Druckman and Bonoma (1976) bilateralmonopoly experiment. The bargainer’s decisions were inf luenced by his expectations regarding the other’s behavior. Prebargaining information about the other and the other’s concession-making behavior created expectations for the size of his future concessions. These expectations then were evaluated against his behavior which presumably led the bargainer to adjust his own concessions and behavior. The results of such evaluation and adjustment were ref lected in the bargainer’s concession-making trends through time (e.g., bargainers made larger concessions when they expected larger concessions from the other) and in his reaction at the end of bargaining when the other’s pattern became apparent (e.g., disappointed expectations caused by a nonreciprocating opponent led soft bargainers to overreact by decreasing the size of their concessions). That nonlaboratory bargainers display a similar pattern is suggested by the results of three other studies conducted in various settings, including early arms control efforts, reactions to scenarios by a sample of diplomats whose combined experience included a wide range of types of domestic and international negotiations, and interactions within the United Nations context of multilateral negotiations, construed as parallel (simulate/in situ) analyses. The Hopmann-Smith (1977); Berman and Zartman (forthcoming); and the author’s studies provide additional evidence suggesting an indirect form of responsiveness between negotiators. In the first study the U.S. negotiators responded to their perceptions of the opponent’s actions toward them. When the United States

Boundary Roles 85

perceived the Soviets in more positive terms, they tended to toughen their negotiating stands; when they perceived the Soviets in relatively negative terms, they softened their posture. As defined in that study, perceptions could be construed as attributions of intent – attributions condition expectations which, according to our model, determine the decisions made with regard to posturing. (For an elaboration of the relationship between attributions and expectations in gaming situations, see Druckman [197lb] pp. 535–536.) The effect of expectations on behavior is illustrated by the Berman-Zartman (forthcoming) study on how diplomats negotiate. Diplomats responded to scenarios that varied in terms of the other’s bargaining strategy. The other’s opening offer created expectations which led the diplomats to adjust their response accordingly. Confronted with high opening bids, the diplomats expected high settlements and drawn-out negotiations. These expectations were confirmed as a self-fulfilling prophesy: the high opponent offers were reciprocated by low diplomat offers, resulting in stalemates. Even more interesting perhaps was the pattern observed when a diplomat, faced with intransigence, made a concession in order to indicate his willingness to negotiate. When the gesture was met by continued intransigence, the diplomat responded negatively by retracting his initial offer and standing firm. Thus, like the laboratory bargainers in the Druckman-Bonoma (1976) study, the Berman-Zartman diplomats were intransigent when they expected an unyielding position from the other, and disappointed expectations caused by a nonreciprocating opponent led generous negotiators to overreact by retracting their offers. A third study is notable for similar results obtained in a simulation and in situ. Both situations permit multiple measurements as the conference unfolds. Negotiators adjusted their behavior in response to their perceptions of the extent to which both sides’ behavior was synchronized. A large difference between the teams in percent tough led the softer team to increase its level of toughness, matching the level displayed by the harder team. The high level of mutual toughness led to an impasse. This pattern is suggested by the set of correlations obtained among indices of tough/conciliatory negotiating behavior (Total tough), differences between the teams in tough/conciliatory negotiating (Diff tough), and an index of the occurrence of deadlocks or impasses during the negotiations (Impasse). The contemporaneous correlations among these indices indicated that the difference between the teams, during impasses, in percent tough was small (rDiff tough/Impasse = – .35, NS), and both teams were tough (rTotal/Impasse = .63, 13 df, p < .01). This is in contrast to the lagged correlations, indicating a strong positive relationship between Diff tough and Impasse (r = .56, 13 df, p < .025) and a negative relationship between Total tough and Impasse (r = -.48, p < .05). The two sets of correlations suggest that the large difference in percent tough prior to an impasse decreased during the impasse and both sides were hard. The softer side became harder. The adjustment made by the softer team followed an evaluation of the perceived difference in negotiating behavior of the two teams. It appears as if the adjustment was made when the size of the difference between the teams approached a threshold value. Threshold recognition requires

86 Flexibility in Negotiation

a monitoring of the unfolding process as it moves through time. It is contended that such monitoring actually occurred, though at a general level. The results of these studies support the assumpton that responses in negotiation are mediated by expectations, which are adjusted through the course of the conference. The process can be viewed as a sequence of steps: (1) Initial expectations are formed on the basis of information about the other’s attitudes, beliefs, negotiating goals, and so forth. (2) The initial expectations are evaluated against the other’s early bargaining behavior. (3) Expectations are adjusted, if necessary, on the basis of the early evaluations. (4) The adjusted expectations are evaluated again, later in the sequence, when the other’s strategy becomes apparent. (5) Expectations are readjusted, if necessary, on the basis of the later evaluation. (6) The readjusted expectations are compared to one’s own negotiating pattern to ascertain where he is at and where I am at. Important changes in one’s own strategy (response, decision) may occur after step 4 and then again after step 6. The negotiator’s response to his perceptions of the other’s strategy (step 4) may be exploitative, as when he toughens his response to the other’s increasing softness (Hopmann and Smith, 1977), or reciprocative, as when the other’s unyielding insistence is met with intransigence (Berman and Zartman, forthcoming). His response to the comparison of own and other’s pattern (step 6) may lead to an impasse, as when he toughens his posture to match the other’s tough strategy (e.g., Druckman and Bonoma, 1976). Of special interest are the parallel results obtained in the laboratory and in situ. Some elaboration on this similarity seems appropriate. In both settings an adjustment occurred as a result of the comparison between own and other’s negotiating pattern (step 6). The adjustment took the form of reduced concessions in the laboratory or hardened rhetoric in situ. The resulting mutual toughness produced an impasse, defined as deadlock in the laboratory and as a breakoff of deliberations in situ. In both settings the impasse occurred when the size of the difference became apparent: in the laboratory, this occurred late in the session, at which point the other’s concession rate strategy could be discerned; in situ this occurred at that juncture where the difference between the negotiating teams was large enough to be noticeable. In both settings, the lagger-subjects in the laboratory or team 2 in situ – adjusted to the leader-programmed concession rate in the laboratory or team 1 in situ (see discussion above). The results of both types of studies illustrate the consequences of a comparison that leads to a judgment of unfair advantage. Other comparisons may suggest that neither side has benefited at the other’s expense, in which case agreements should be concluded. Indeed agreements were negotiated in other conditions of the laboratory experiment (namely, increasing concessions by the programmed

Boundary Roles 87

other) and at other junctures (to wit, turning points) in situ. Favorable comparisons took the form of judgments of equitable concessions or synchronized negotiating behavior. Some strategies or behavior that followed from the results of these comparisons (and other response combinations) are elucidated and demonstrated in Druckman and Bonoma (1976). In particular, a preliminary attempt is made to distinguish response combinations that may lead to impasses from those that are more likely to lead to agreements. More generally, the studies provide support for the sequence of steps outlined above. Corroboration of findings obtained in the two settings lends confidence to the contention that version 2 describes the way in which information about/ from the other is processed, and the implications of this processing for responsiveness to the other in negotiation. The preceding analyses concentrate on the basic process of responsiveness. These analyses can be extended to a consideration of the way in which a negotiator may use responsiveness. If, as both versions of the model assume, a negotiator is responsive to the positions and concessions of his opponent, the opponent can manipulate his own positions and concessions in order to inf luence the negotiator. Responsiveness renders a bargainer capable of shaping the other’s behavior, of causing the other to choose as he does. Such tactical maneuvering may take several forms, including concession-making strategies and such modes of communication as threats (negative consequences for noncompliance), promises (positive consequences of compliance), warnings, and commitment devices (e.g., linking one’s position to a more general principle). If effectively executed, the former strategies should enable the bargainer to effect a change in the opponent’s expectations for further concessions and evaluations concerning the likelihood of an agreement (see Druckman and Mahoney, 1977) while the latter tactics should enable him to place the burden of concession making on the opponent (see Schelling’s, 1960, discussion of the art of casuistry, and Druckman, 1976). Further development of the negotiator-as-bargainer model entails taking into account his arsenal of inf luence strategies.

Conclusion The versions presented in this section are considered as alternative construals on the negotiator-as-bargainer model. The primary dimension of difference between them is complexity, with the second version taking account of more variables and processes than the first. The increased elaboration of the versions also highlights a complementarity among them. Direct responsiveness between bargainers is likely to characterize early bargaining while own/other responsiveness occurs as they develop a concession-making history. The expectation/ evaluation/adjustment sequence of version 2 affects bargaining behavior at the point where the other’s strategy becomes apparent. Viewed this way, these are early, middle, and late processes. Viewed another way, version 2 could be regarded as the penultimate process, following from the earlier monitoring of the

88 Flexibility in Negotiation

other and preceding the dramatic gestures that lead to impasses or agreements. In either case, the conjunction of the alternative processes and different phases could be regarded as hypotheses to be evaluated experimentally. We turn now to the second general model of responsiveness: the negotiator as representative.

The Negotiator as Representative If the negotiator-as-bargainer model is of the genre referred to as unitary-rational actor models, then the negotiator-as-representative model is cut from the class of models referred to as bureaucratic (governmental) politics. Whereas the former emphasizes internation interactions, the latter focuses on intranational interactions. Together these two types of activities constitute international negotiation. Or, more broadly: “Applied to relations between nations, [the latter] model directs attention to intra-national games, the overlap of which constitutes international relations” (Allison, 1971: 149). The importance of internal bureaucratic politics in international negotiations has been well documented (e.g., Allison, 1971; Newhouse, 1973; Wolfe, 1975). These treatments concentrate on the interagency bargaining that occurs during (and inf luences the conduct of ) an international negotiation. In most nations, different departments or ministries are involved in the process of making decisions about the substance and tactics of negotiation, and they often have competing preferences. Particularly notable are the conf licts between defense and foreign ministries over arms-control issues: these conf licts have been characterized as classic domestic struggles that hinder attempts to develop an overall policy toward the negotiation. Other conf licts are likely to occur among different role occupants within the same agency as when those who negotiate with alliances place a higher emphasis on military factors than their counterparts who negotiate primarily with the other side or with the bureaucracy (Hopmann, 1977). These examples highlight the differing emphases of diverse agencies and roles. From the standpoint of the negotiator, these emphases take the form of competing claims made with regard to positions or postures that he ought to adopt. The following treatment depicts the way in which a negotiator attempts to reconcile these claims. The negotiating representative must arbitrate among diverse agency positions. He must also decide on the extent to which he will defend the “best” position at the conference. These decisions correspond to the two components of representation described in the literature as position commitment and group loyalty (see, e.g., Zechmeister and Druckman, 1973). Both can be construed in terms of a weighting (or balancing) process: The former is the weight assigned to each position in terms of its relative importance; the latter is the weight assigned to the positions in terms of “degree of agency.” The model is developed in two steps. First, alternative positions are construed as n-components of value. Second, the relative importance assigned to the n-components is weighted according to the negotiator’s “sense of agency.”

Boundary Roles 89

Building a Package As a negotiator, the representative attempts to build a package that will be acceptable both to the other side and to his bureaucracy. The process of building a package involves “monitoring certain feedback variables, such as domestic support for the negotiation, which indicate a willingness of the (negotiator’s) government and others to cooperate in a negotiated settlement” (Winham, 1977a: 353). The importance of a package is emphasized by the diplomats interviewed by Berman and Zartman (forthcoming). These respondents noted that concessions and redefinitions of bargaining norms were determined in reference to an overall package. Resources for trade were often evaluated in terms of impact on the larger package within which these items were contained. The package is a combination of offers made on a variety of items, e.g., numbers of ground forces to be redeployed, number and category of weapons systems to be removed from an area, ceilings to be placed on reinforcements. The offers can be construed as components of value, w1, w 2 , w3, . . . w n. The negotiating representative establishes priorities among the set of components by considering them in terms of relative importance or subjective worth. The subjective worth of the package to negotiator i can be regarded as the sum of the worths of each component of value associated with a particular package. Thus, in the two-component case, if u m is the manpower component, and u n is the nuclear-weapons component, the value of the package (U´) to negotiator i is

U i′ = um + un

(4)

It is assumed that the subjective worth of each component is a linear function of its objective worth (w m and w n respectively). Thus, um = aw m + b, and

(5)

un = a′w n + b′,

(6)

where a and a´ ref lect the relative importance of w m and w n, and b and b´ ref lect arbitrary differences in zero-point between objective and subjective worth – e.g., the point at which satisfaction turns into dissatisfaction. By substituting 5 and 6 for 4, we have, U′i = aw m + b′+a′w n + b′, or

(7)

U i′ = aw m + a′w n +b + b′.

(8)

By setting b˝ = b + b´, we have

90 Flexibility in Negotiation

U i′ = aw m + a′w n +b′′.

(9)

However, since we are concerned with maximization – i.e., building the best package – it is not the actual size of a and a´ that is important, but their relative sizes. To reflect relative importance, expression 9 is transformed into a new expression by a series of steps that involve multiplying and adding by a constant, substitution, and simplification. For our purposes here these steps are omitted. The result of the transformation is the expression: U i = θ i w m + (1 − θ i ) w n ,

(10)

where i is the extent to which negotiator i considers the manpower component (wm) to be relatively more important than the nuclear weapons component (wn). As i varies from zero to one, Ui will increasingly reflect wm more and wn less. (When i = 1, Ui will vary directly with wm and independently of wn.) According to this model, then, negotiators are assumed to seek that package which maximizes the value assigned to its components as: max j θ i w m + (1 − θ i ) w n .

(11)

The two-component version, presented above, provides the basis for a model that represents the process of weighting or balancing n-components. The latter is a generalized version of the former. It takes the form: U i′ =

∑u ,

(12)

k

k

where u k = a kwk + bk, so that U′i =

∑ (a

k

w k +bk ).

(13)

k

Then a series of steps that parallel the development of the two-component version lead to the expression, Ui =

∑θ

k

wk ,

(14)

k

which can also be expressed as a maximization function similar to 11 above. The essential notion of the generalized version is that each component of value is considered relative to all other components, such as 1/ 2 + 2 + 1 – (1 + 2) in the three-component case. If in the three component version any two of the three weights equal .33, then the remaining weight will also equal .33, and U i will be the simple sum of the three components, and so on.

Boundary Roles 91

As stated above, the model depicts a negotiator’s preferences for the components of a package. It does not describe his behavior. To assume that the one follows directly from the other is no more justified than to assume an isomorphism between attitudes (or values) and behavior. Both remain empirical issues. Some clarification on the nature of the relationship between preferences and actual decisions is provided by work in progress. An experiment designed to evaluate the predictive accuracy of the two-component version of the model can be summarized briefly. (Further procedural details and results are presented in Druckman, forthcoming). The experiment was a simulation of committee decision-making. Four committee members were to make decisions concerning the funding of alternative packages construed as programs. Members (subjects) represented different agencies, each with a stake in the outcome. Agency preferences for the programs were based on values assigned to the components wm and wn. Subjects represented their agency’s values. Estimates of  were provided by responses to a questionnaire item which asked subjects to indicate (on a seven-step scale) their own perception of the relative importance of the components, wm and wn. Following a role-induction session, committee members exchanged messages intended to influence other members to support or block the proposed programs. The outcome consisted of a winning coalition of members whose pooled resources enabled them to support or reject the program package. For each committee member an outcome index was constructed. The four values of this index included being in a successful blocking coalition (1), being out of a successful funding coalition (2), being out of a successful blocking coalition (3), and being in a successful funding coalition (4). The data provided by this experiment was used to evaluate the model. Each member’s final decision was predicted on the basis of the values of the components, wm, wn, and . Multiple correlations between these components and the outcome index were computed for each of the three program-packages to be decided by the committee. The correlations for the three programs were .61, .67, and .57, each accounting for a significant portion of the variance in the outcome index. Another set of multiple Rs were computed excluding the subjective component, , from the prediction equation. These correlations were .49, .58, and .51, respectively. A comparison of the two sets indicated that, in each case, adding  to the equation resulted in a significant increment in variance explained (F1/93 = 19.5, 18.6, and 8.1, respectively). These results suggest two general conclusions: (a) final decisions did reflect preferences, and (b) predictability is enhanced when the subjective component, , is included in the equation. Both conclusions support the contention that preferences, represented by the values of the model components, are indicators of actual decisions made with regard to the nature of the agreed package. However, while the model seems to capture the way a negotiator processes information toward building a

92 Flexibility in Negotiation

package, it does not take account of the extent to which he is responsive to his agency’s preferences. We now turn to a further extension of the model that takes account of degree of agency.

Degree of Agency Representation is defined in terms of the extent to which the negotiator represents his agency’s priorities. If these correspond to his own priorities, there is no problem. More often, however, individual priorities differ to some extent from agency preferences. When this occurs, the negotiator must decide which way to go. His sense of obligation is reflected in his willingness to forfeit his own subjective rankings in favor of the group positions, when these conflict. It is this conflict that can be reflected in our components of value model. This is done by letting i represent the degree of agency for negotiator i, such that when i is 1, the negotiator is a pure agent for the group; when i is 0, he acts strictly according to his own priorities; and as i varies from zero to one, he increasingly represents the group and decreasingly takes his own views into account. To begin, k can be divided into two parts: g,k – the group’s weight for component k – and i,k – the negotiator’s weight for component k. Each is assigned a weight of , which represents the extent to which one or the other is taken into account. Thus, θ ′ k = α iθ g, k + (1 − α i )θ i, k .

(15)

The generalized version, ref lecting all k components of the package (Ur), is Ur =

∑θ ′

k

wk (16)

k

By substituting expression 16 for k and shifting terms, we have Ur =

∑ (α θ ) w i g, k

k

k

+ ⎡(1− α i )θ i, k ⎤⎦ w k ,

(17)

which can be simplified by setting Ug =

∑θ

g,k w k

(18)

k

and Ui =

∑θ k

i,k w k

(19)

Boundary Roles 93

so that the negotiator attempts to achieve the package that maximizes the expression max jα i U g + (1 − α i ) U i ,

(20)

which ref lects the value of the package U to the representative r in terms of his sense of agency in representing group g, and compares with expression 11, a maximization function for the nonrepresentative dealing with only two components of value. Expression 20 is a preference function that can be evaluated by an experiment similar to the one performed above for evaluating expression 10. Such an experiment would require estimates for  similar to those obtained for . It would evaluate the extent to which the expanded version of the model predicts actual decisions. Or, put another way: does  add a significant increment to the variance explained, over and above that explained by the n-components of value version, represented by expression 14? While this particular experiment has not been designed, another implication of  has been explored. Degree of agency influences both relative emphasis on the components of a package and extent of commitment to those positions. The former affects decisions concerning the type of proposal to make to the other side. The latter affects the willingness to compromise positions in order to attain an agreement with the other side. The former can be evaluated by the type of experiment described above. The latter has been evaluated by a recent experiment on the effects of accountability on negotiating behavior (Mittelmark et al., 1977; see also Druckman, forthcoming for further details on procedures and results). This is summarized here briefly. In the recent experiment accountability was defined as the extent to which the representative was responsible and responsive to his group: highly accountable representatives were appointed and required to report regularly concerning their deliberations; low accountable representatives were elected as relatively free agents. This manipulation was embedded in a simulation of a policy-making council whose members represented different groups with particular interests in the outcome and ideological (value) stances. The extent of conf lict of interests and values among the groups were also manipulated: conf lict of interest was defined as vested interest in the final decision (namely, extent to which the decision would have implications for the control of resources); conf licting values were defined as extreme or moderate orientations underlying the positions taken on the issues in dispute. The effects of these factors on dyadic-negotiating behavior were assessed. A particularly intriguing finding from this experiment is that the accountability variable served to mediate the effects obtained for the conflicting interests and ideologies variables. The conflict-of-interest variable was significant when accountability pressures were high: high conlict-of-interest dyads had

94 Flexibility in Negotiation

more trouble resolving their differences than low-conflict dyads. The differences between the conditions under low-accountability pressures were nonsignificant. Interestingly, however, the ideological-conflict variable was significant when accountability pressures were low: high conflict-of-ideology dyads were less satisfied with the outcome, perceived compromise to be more like defeat, and perceived negotiations to be more futile than low-conflict dyads. Opposing decision makers were more responsive to their differences in ideology when they were not accountable directly to their group. Conversely, they were more responsive to their vested interests when they were accountable. Construed in somewhat different terms, the finding can be shown to have implications for our model. Accountability (degree of agency) can be regarded as a weight that determines the relative emphasis (responsiveness) placed on the components, interests and ideologies. When it was high, vested interests were emphasized; when it was low, conflicting ideologies were salient. An emphasis on interests resulted in resistance to compromise due, presumably, to a heightened sense of commitment to the group positions. The emphasis on ideologies under low accountability did not result in resistance to compromise, due, presumably, to a lessened sense of group commitment. (Rather, the effects produced concerned perceptions of the conflict and of the negotiations.) As noted above, this is one type of effect of the variable, degree of agency. Another effect is on the choice of the components of the package or the positions per se. This effect remains to be explored. Choice of components suggests another aspect of packaging not considered in this treatment. This is the notion of repackaging issues and programs as captured in Zartman’s (1977) analysis of formula and in Tracy’s (forthcoming) concept of structure. Their discussions highlight the fact that the issues or programs per se may be a variable. The practice of creating new packages involves reconceptualization, redefinition, disaggregation, and reintegration. These processes also contribute to the formulation of negotiating positions. For this reason, inter alia, it would seem appropriate to include them in a model. Such a model would be a more complex rendition of the concept of packaging. Attempts to add these processes to those construed above will be a forthcoming development.

Boundary Role Conflict: Toward a Reconceptualization As different cuts on negotiation, the models extend the concept of boundary role conflict. As co-occurring processes in negotiation, the models retain a defining feature of the concept. The extension renders the concept more complex than previously construed. The defining feature suggests an interplay between the

Boundary Roles 95

processes such that a negotiating dilemma is created. Each of these notions is an elaboration toward a reconceptualization of the boundary role conflict. As originally defined, the boundary role conf lict is the conf licting aspects of a negotiator’s role obligations: the expectations of his own side that emerge from an internal consensus on the positions he is to take and the expectations of the opponent that must be taken into account if an agreement is to be reached. The dilemma inherent in this definition has been captured in at least two experiments where effects on bargaining behavior were observed (Frey and Adams, 1972; Wall, 1975b). However, the models developed and the evidence reported above suggest that the components of the BRC are not simply competing demands but, perhaps, fundamentally different processes. Responsiveness to the other takes the form of a sequence of comparisons and adjustments leading to impasses or agreements. Responsiveness to one’s own side consists of ascertaining the relative emphasis to be placed on components of a package prepared as negotiating positions. The differences between these two monitoring functions are in terms of dynamics and influences: the former is in the tradition of dynamic-equilibrium models, the latter is a component of a variance model; the former emphasizes perceptions and concessions while the latter emphasizes group commitment and priorities. Preliminary relationships among these variables are demonstrated above. More important, however, is the heuristic value of the methodological strategy: observations made in situ served to elucidate the processes; the laboratory experiments served to confirm (or refute) the hypothesized relationships. The competing demands of the BRC are also interacting pressures on a negotiator. This aspect of the original definition is preserved in the conceptualization developed here. The differences between the models notwithstanding, the processes are not regarded as being mutually exclusive during the course of a negotiation. They are complementary and intertwined, the one inf luencing the other in a reciprocal manner. Whereas the negotiator-as-bargainer model (version 2) highlights the juncture where a change in posture is likely to occur, the negotiator-as-representative model suggests the probable nature of the modification. According to the former model, changes in posture occur when the other’s strategy is discerned and after a comparison between own and other’s pattern is made. Which concessions are made and the extent of the concessions depend on the relative importance of the various parts of the package and on the degree of agency. Both variables are estimated as components of the latter model. Together, the two models provide a framework for evaluating changes in postures or negotiating behavior. Adjustments result from dual responsiveness: cues from one’s own side are balanced against those from the other as the negotiator attempts to decide which way to go. And, it is this bidirectional monitoring function that defines the BRC. The interaction between the processes involved

96 Flexibility in Negotiation

in this function remains a topic for investigation. Further elucidation awaits the results of these studies.

Conclusions The boundary-role conf lict suggests two types of monitoring functions in negotiation: monitoring the other side for evidence of movement and monitoring one’s own side for evidence of preferences. Each of these functions was represented by a simple model. Differences between the functions made difficult the development of a more general model incorporating both functions. The differences are in terms of focus and information-processing. The negotiator-as-bargainer model assumes responsiveness between oppositenumber negotiators. Such responsiveness can take several forms. It can be a direct calculated response to the other’s concessions (see Druckman, forthcoming), it can be based on both one’s own previous concessions and the other’s concessions, or it can be a more complex function of expectations and evaluations. It was contended that the latter version-complex function-ref lects well the dynamic of international negotiation. According to that model, negotiators monitor the other’s concessions as a trend. A negotiator’s reponses are mediated by expectations which are adjusted through the course of the conference. Informationprocessing can be represented by a sequence of steps involving the formation of, evaluation of, and adjustment of expectations based on a comparison of where is he at now? and where am I at now? Evidence from the Hopmann-Smith, BermanZartman, and the author’s studies support this characterization. The negotiator-as-representative model assumes reponsiveness between the negotiator and his agency (agencies). Such responsiveness was depicted in the form of a utility model where the negotiator attempts to balance n-components of value in the process of building a package. He is concerned with maximizing the value of the package in terms of both his own and his agency’s priorities. This is essentially a weighting process. The cognitive processes involved consist of defining and ordering the various components of the package and ascertaining his agency’s preferences for this ordering. Both of these functions were ref lected in the model. However, as the number of components increase and as his agency’s positions diverge from his own, the information-processing requirements become rather complex. The former can benefit from technological aids. The latter requires reconciliation mechanisms that are part of the bureaucratic political process. Finally, it should be reemphasized that the two types of models are not presented as alternative conceptualizations of international negotiations. Both processes occur simultaneously in negotiations (see, for example, Wolfe’s 1975 treatment of SALT as a two-track negotiation). The models should be viewed as complementary, notwithstanding the differences between them noted above. However, complementarity would be enhanced if a more general model,

Boundary Roles 97

incorporating both processes, could be developed. Such a challenge will be a forthcoming consideration.

Notes 1 Note that the references for this article are part of the collected references assembled for the special 1977 issue of the Journal of Conflict Resolution edited by I.W. Zartman. 2 If the linear algebra used in this article is not sufficiently challenging, see the articles on modeling coalition formation (Druckman et al., 1974; Krause et al., 1975) or the game theory analyses of ancient diplomacy (Gunar and Druckman, 2000).

9 SITUATIONAL LEVERS

Fast forward to the early 1990s. I was a member of a research team at the International Institute for Applied Systems Analysis (IIASA) in Laxenburg Austria. At the time I was particularly interested in developing closer connections between frameworks and experiments. The negotiation framework discussed in the first chapter is an attempt to organize a variety of inf luences on processes, outcomes, and the implementation of agreements. They are especially useful for organizing reviews of research (Druckman, 1973, 1977; Druckman and Wagner, 2016) and provide a broad perspective. Experiments, on the other hand, focus on the nuts and bolts of the way a process unfolds and the drivers of those processes. They take a deep dive into what is going on and analyze only a few variables at a time. I wondered how to do both. What kind of research approach would be amenable to examine the impacts of combinations of variables while also unpacking the combinations to distinguish between stronger and weaker variables? The question was addressed in an article on situational levers (Druckman, 1993). The broad perspective was captured in design. Packages of variables were put together for each of four stages of a simulated international negotiation on environmental issues. The deep dive was captured in the way the collected data were analyzed. I called this an exploration into the situational levers of f lexibility. The mechanics are discussed in the article. The analyses showed differences between two samples (students and diplomats) and between the simulation and a comparable meta-analysis of earlier experiments. A follow-up article is included in this section. Building on the findings reported in the 1993 article, these analyses refine the list of important inf luences on f lexibility. They highlight the importance of negotiation privacy and the timing of firm and f lexible postures. They also show that situation design has stronger impacts on f lexibility than mediator suggestions. Taking these analyses further, the study promotes the relevance of the levers by showing DOI: 10.4324/9781003293361-15

Situational Levers

99

correspondences between diagnosed and actual outcomes of a set of historical cases. Three broader implications from these studies are interesting. One is that stages are important. For example, we showed that role as representative or delegation advisor inf luenced f lexibility in the early stages while media attention was important in the later stages of the talks. Another implication is that many of these aspects of the situation are easy to change as implied by the term “lever.” Examples are location, media coverage, type of agreement sought as more or less comprehensive and whether a mediator is present. A third is the connection between this approach and a behavioral perspective. Both share the idea and technology of manipulating aspects of the immediate environment to produce behavioral change. Although the popularity of behaviorism has dwindled in recent years, this study opens possibilities for change consistent with the early research on operant conditioning (Kimble, 1961; Ferster and Perrott, 1968). I return to a discussion of connections between the research featured in this book and many of the classical ideas in the book’s concluding chapter. These connections show thematic consistency through the decades of the career. Situational Levers of Position Change: Further Explorations, Daniel Druckman. The Annals of the American Academy of Political and Social Science 542 (1995): 61–80. © Sage Publications, Inc. Reprinted with permission.  

SITUATIONAL LEVERS OF POSITION CHANGE Further Explorations Daniel Druckman

This article examines the question, Under what conditions are negotiators likely to be more or less f lexible? This is a central question in the literature on negotiating behavior. Although a variety of measures of f lexible behavior are used in the various studies, many of them are concerned with movement toward an acceptable agreement. Focusing primarily on movement or change, the studies have identified aspects of the negotiation situation that inf luence f lexibility: issues, background factors, context, structure of the conference and negotiating teams, and the immediate situation. Building on this literature, this article explores the way that many of these aspects interact during several phases of an extended simulated international negotiation. The approach taken in this article has several features. One is that it uses data from experiments as the primary source of insights into f lexibility. Another is that the knowledge produced is neither case nor region specific but is more general, applying across a wide range of negotiating situations. A third is that it focuses on those aspects of the situation that can be controlled or altered by negotiators or third parties. A fourth is that the findings apply in particular to negotiating in an international, rather than domestic, setting. These features of the approach are elaborated further in the sections to follow, after which the results of a number of analyses will be presented to further our understanding of the conditions that inf luence a negotiator’s f lexibility.

Defining Flexibility Flexibility in this research is defined in behavioral terms, as movement from initial positions toward compromise positions. It may, however, be defined in other ways. From a cognitive perspective, we may ask about the negotiator’s intentions to be flexible or not, that is, some idea about why parties may or may not want to

Situational Levers

101

be f lexible at any given time during a negotiation. What considerations are entertained prior to making moves such as, for example, tabling new proposals? How do negotiators weigh alternative courses of action? How do they evaluate others’ moves in the negotiation? Coddington’s cyclical bargaining model emphasizing expectations, evaluations, and adjustments is suggestive. So, too, are the expanded versions of this model applied to international cases by Snyder and Diesing and by Druckman.1 Compromising is one indicator of f lexibility. It is achieved through a willingness by both or all parties to take some losses, and it is attractive because it is seen as fair in the sense that the parties are subject to equal losses. Agreements may, however, be achieved in other ways. One is through capitulation by one party. Another involves a search for solutions that satisfy both (all) parties’ interests without compromising. Referred to as integrative solutions, these outcomes do not entail the mutual losses incurred by compromise. They are achieved by strenuous cognitive activity on the part of negotiators or mediators as described recently by Kressel and his associates.2

An Experimental Simulation Approach A simulation approach has advantages for theory building, with contributions from both design and analysis. Unlike the simpler laboratory games used in experiments, simulation designers must articulate the features of a real-world setting that allows participants to enact roles that at the very least resemble the constraints and responsibilities of their counterparts in the setting being simulated. This process should, it would seem, further the designer’s understanding of the real-world setting, which is, in this case, an international conference on environmental decision making.3 The design process also has implications for analysis, however. Two issues in particular are emphasized: the extent to which the real-world setting must be simplified for interpretability of results, and the effects of role-players’ theories about the way they should enact their roles in the simulation. Regarding the former issue, efforts to achieve high fidelity may make it impossible to separate important from unimportant inf luences on decisions. Too much fidelity reduces the advantages of simulation as an analytical device; the gains in verisimilitude are balanced by costs in analytical rigor.4 Too little fidelity, on the other hand, reduces the value of the simulation approach, rendering it indistinguishable from simpler laboratory experiments.5 The challenge is to strike a balance between representing the setting of interest and isolating key factors hypothesized to inf luence the processes of interest.6 An attempt to achieve such a balance is made in the design of the simulation presented in this article. With regard to the latter issue, too much fidelity may lead role-players to enact their roles largely in terms of their knowledge or expectations of how they think the decision makers would behave in the actual situation. The results may,

102 Flexibility in Negotiation

therefore, ref lect their expectations about how others would react rather than their own interpretations of how to react to the situation. An attempt is made to limit information about particular nations or decision-making roles in the design of this simulation. Moreover, we ask a similar sample of respondents to indicate the extent of f lexibility or inf lexibility likely to result from each of the variables included in the simulation scenarios. The simulated negotiation reported on in this article focused on issues in the area of international decision making. Representatives were faced with a dilemma of giving up national (or international) authority over regulatory decisions and technology development for international (or national) authority over these matters. This is a problem faced by national decision makers in many environmental forums; it is especially salient in debates in the United States over participation in multilateral organizations. Two exercises were used in the experiments. In one, participants made decisions only in the context of scenarios representing each of four stages of the conference. In the other, participants engaged in bargaining during the stages referred to as “give-and-take” and the “endgame.” In the decision-making exercise, participants had the choice of sticking to their position of rejecting international control or moving in varying degrees toward positions that would yield authority to an international body. Flexibility was indicated by how far they were willing to go in the direction of international control – from their own position of giving an international commission no authority (step 1 on a 7-step scale), to a position of accepting general guidelines from the commission (step 2 on a 7-step scale), with gradations to a position at the other end of the scale of giving regulatory authority to the commission with strict policing for violations by nations (step 7 on the scale). Results are discussed in the next section. In the bargaining exercise, role-playing participants were confronted by opponents who took positions on the opposite ends of the 7-step scale on each of two issues: the issue of international control over regulatory decisions and the issue of participation in an international research and development (R&D) organization for developing new environmental technologies. In each bargaining pair, one representative was faced with the dilemma of giving up national control over decisions concerning regulatory standards and new technologies, while the other faced the dilemma of giving up a degree of international control for increased national autonomy over decisions in these areas. Flexibility was indicated by the willingness of both bargainers to move from their positions toward a compromise agreement. But it was also indicated by decisions on tactics and perceptions of the situation. The experimenters’ interest was to ascertain the relative importance of a number of possible inf luences on these indicators of f lexibility. In the sections to follow, I summarize the results obtained in the initial experiment and then present the findings from each of several analyses of data collected to address questions raised by those results.

Situational Levers

103

Situational Levers of Decisions The simulated conference was divided into four negotiating stages: prenegotiation planning, setting the stage, the give-and-take, and an endgame. For each of these stages, a scenario was constructed that included those five or six aspects of the situation thought to inf luence bargaining decisions at that juncture of the talks. For example, in the planning stage, decision makers were told that they either studied the issues or strategized, that their positions were either linked or not linked to broader national ideologies, and that they were either the primary representative or an adviser to the delegation. In the give-and-take, media coverage was either heavy or light, and the other parties made many or few concessions. In the endgame, decision makers were or were not faced with a deadline, and possible solutions were or were not suggested by a mediator. Two contrasting experimental conditions differed only in terms of the direction in which the variable was coded, as producing either f lexibility or inf lexibility. For example, in one version, the representative was his or her country’s chief delegate (hypothesized inf lexibility), and in the other, he or she was an adviser to the delegation (hypothesized f lexibility). A third condition consisted of the “inf lexible” scenarios for the first three stages followed by an endgame situation in which the variables were worded in a “f lexible” direction. Two samples participated in the simulation, environmental scientists and midcareer diplomats.7 For both samples, strong differences were obtained between the conditions on outcomes, willingness to move from initial positions, choice of strategy, and several perceptions of the situation. Decision makers in the condition hypothesized to induce f lexibility reached agreement more often, were willing to move further from their initial positions, chose more cooperative tactics, and viewed the situation more as a problem-solving debate, as opposed to a win-lose contest, than did those faced with scenarios intended to decrease f lexibility. Moreover, these effects occurred across all four stages of the simulated conference; there were few interactions between the three experimental conditions and the four stages. As hypothesized, those in the third condition became more f lexible in the endgame than they were in the earlier stages of the conference. More interesting, perhaps, are the results of analyses of the relative impacts of the variables embedded within each stage scenario. Using a pair-comparison procedure,8 decision makers judged each aspect relative to all other aspects of the situation highlighted in each stage of the conference. The procedure provides weights for each variable, allowing an analyst to rank them in terms of relative importance. The key variable or variables in each stage can be arranged on a trajectory toward an outcome in the endgame, which may be an agreement or a stalemate. For the diplomat sample, for example, the trajectory of key variables toward agreement (the decision makers in the “f lexible” condition) was as follows: being a delegate adviser and having amiable relations with the other parties (stage 1),

104 Flexibility in Negotiation

followed by being a delegate adviser and being at a power disadvantage (stage 2), followed by exposing the talks to limited media attention (stage 3), followed by exposing the talks to limited media attention and having unattractive alternatives (stage 4). The key variables in each stage leading to stalemate (for decision makers in the “inf lexible” condition) were as follows: strategy preparation (stage 1), followed by having a power advantage (stage 2), followed by wide media coverage and few concessions from others (stage 3), followed by having attractive alternatives (stage 4). These variables are regarded as situational levers of negotiating f lexibility. They are the factors that can be manipulated by negotiators or planners for impact. These findings, however, may be limited to the role-playing exercise used in this experiment. A question remains concerning the extent to which they apply also to actual negotiations. The progress made toward answering this question is reported in the section to follow.

Generality and Applications The simulation approach to studying f lexibility raises a number of questions unanswered by the initial experimentation. These include the following: 1

2

3

4

5

The approach assumes that negotiating f lexibility is strongly inf luenced by aspects of the situation that can be altered or engineered. This is different from suggestions made by a mediator about possible agreements. What would happen if an active mediator was inserted in each stage-scenario for the condition hypothesized to produce inf lexibility? This question is addressed in the first part of this section. The scenarios asked role-players only to make decisions about positions and tactics at each of four stages. They were not asked to actually bargain with other national representatives. Would the results be the same if participants actually engaged in bargaining? Does the bargaining process make a difference? Results described in the second part evaluate these questions. Faced with an actual opponent, the bargainer must respond to the opponent’s tactics. What impacts, in particular, do the timing of moves – proposals or concessions made either early or late – have on the bargainer’s decisions? These impacts are evaluated in the third part of this section. The use of simulation raises issues concerning the extent to which roleplayers’ theories or expectations affect their decisions as a consequence of acting as if they are diplomats. What are these theories? Are the results due primarily to the decision maker’s theories of behavior expected in these situations? The extent to which this may be the case is shown in the fourth part of the section. The simulation approach is presented as an alternative to simpler laboratory gaming experiments. Does the combining of variables, embedded in the scenarios, have a stronger, similar, or weaker impact on f lexibility as

Situational Levers

6

105

compared to when these variables are manipulated one at a time? How do the results obtained from the simulation approach, presumed to be closer to the real world, compare to those obtained from laboratory experiments that manipulate similar variables? Such comparisons are made in the fifth part. The research is considered to contribute primarily to the development of theories of negotiating behavior. How can the approach or findings be used in a practical way, as training tools or diagnostic aids? What are some prescriptive or practical implications of this research? A number of examples of tools are described in the final part of this section. We turn now to the analyses intended to provide answers.9

Mediator Effects It was suggested that the difficulties encountered by participants in the “hypothesized inf lexibility” condition could be reduced by the presence of an active mediator. A mediator was therefore inserted in each stage, and the results were compared to those obtained in the earlier simulation, where a mediator was included only during the endgame stage. During the planning stage, the mediator would gather information about each delegation’s views. In the next stage, he or she would help the parties decide about agendas, locations, and the kinds of agreement sought. In the give-and-take, he or she would shuttle between the parties in search of concessions. In the endgame, he or she suggested a solution that may be acceptable to all parties. Significant differences were obtained on only two of nine dependent variables. Role-players in the mediation condition moved further from their initial position on their desired outcome and indicated that they were more satisfied with a compromise outcome than were those without the assistance of a mediator in any stage except the endgame. They were also less satisfied with a compromise outcome in the first stage than in the later stages of the conference. Thus mediator effects were limited; the mediator did not offset those aspects of the situation in each stage that served to prevent compromise. Apparently what was needed were the sorts of situational changes engineered for the “hypothesized f lexibility” condition, namely, altering the situation rather than only giving advice. Further effects of such alterations are presented in the part to follow.

Bargaining Experiment The bargaining exercise was used to examine inf luences on position movement during and after actual bargaining. As noted earlier, opposing bargainers – scientists similar to the sample participating in the decision-making exercise – took different positions on the issues of establishing an international regulatory commission and an international R&D organization. These issues are distinguished in terms

106 Flexibility in Negotiation

of being either large (a regulatory commission) or small (an R&D organization). This distinction of the relative size of the issues was included in the design, as a repeated measure, along with the “f lexibility” condition (f lexible or inf lexible scenarios), issue position (for or against the commission or organization), and stages of the conference, also a repeated measure. In addition, messages were delivered to bargainers from constituents advising them to make a move either early or late in the negotiation; the results of this manipulation are discussed in the next section. The pattern of results was similar to that obtained from the earlier decisionmaking exercise. These are shown in Table 9.1. With regard to outcomes, bargainers in the “f lexible” condition moved further from their initial position than those in the “inf lexible” condition on the smaller issue but not on the larger issue: 92 percent of the bargaining dyads in the former condition settled the issue, compared to only 27 percent in the latter condition (χ2 = 10.56, df = 1, p < .01). Dyads in the f lexible-scenario condition moved further from their initial positions on both issues with respect to willingness to compromise, desired outcomes, and likely outcomes. They also viewed the conf lict as a problem to be solved, were less interested in having their own position prevail, and used less competitive tactics in the “f lexible” condition (see Table 9.1). A lack of interactions between the conditions and stages indicates that these condition effects occurred across the four stage scenarios. Results of the pair-comparison judgments indicate which variables in each stage most strongly inf luenced bargaining decisions. The following trajectory highlights the key variables (highest weightings) that produced f lexible decisions in each stage, resulting in agreements: friendly relations (stage 1), followed by peripheral location (stage 2), followed by limited media attention (stage 3), followed by limited media attention (stage 4). The following trajectory shows the key variables that led to relatively inf lexible decisions in each stage, often producing stalemates: lack of familiarity or being the primary representative (stage 1), followed by central location (stage 2), followed by wide media coverage (stage 3), followed by wide media coverage (stage 4). Whether the talks were held at central or peripheral locations and whether the coverage was limited or wide had strong and consistent effects on the bargainers. These variables had the strongest impacts on decisions made by both bargainers in each stage where they were included in the scenario. These results confirm and strengthen the findings obtained in the decisionmaking exercise. They highlight the importance of procedures or logistics used to arrange settings that may be conducive to getting agreements on issues that divide nations. As noted by Walton and McKersie, private settings serve to release negotiators from scripted positions by limiting the inf luence of constituents and audiences.10 It was just such effects that seemed instrumental to the progress that occurred toward an Arab-Israeli peace agreement, as noted by Norway’s Foreign Minister Johan Jørgen Holst.11

Situational Levers

107

TABLE 9.1 Comparing Conditions: Bargaining Experiment

 

Condition Means*

Measures

A

B

Significance †

Outcome as percentage agreeing Issue 1 Issue 2 Resistance point Desired outcome Likely outcome Tactics Win-lose or problem solving Own position prevails Satisfied with outcome

75 92 3.0 1.9 3.3 2.6 3.9 2.5 3.8

55 27 2.6 1.3 2.5 2.3 3.0 2.1 3.6

NS .01 .07 .04 .06 .10 .002 .02 NS

 

* Condition A is hypothesized f lexibility, B is hypothesized inf lexibility; resistance points and desired and likely outcomes were measured as movement from initial positions on 7-step scales; tactics were measured on a 4-step scale from most (lower score) to least (higher score) competitive, and the remaining variables were measured on 5-step scales. † The probability values shown in this column are for the F ratios calculated for condition main effects in an analysis of variance design.

These are factors in the situation surrounding the negotiations. Negotiators may also be inf luenced by aspects of the process itself. One of these aspects comprises the moves by the other negotiator. In the next part of this section, some possible effects of the other’s moves are explored.

Early Versus Late Moves A number of experiments examined impacts of alternative patterns of concession making on bargaining behavior. With regard to studies using the prisoner’s dilemma game, Rubin and Brown concluded that “a sequential change in the other’s behavior from low to high competitiveness induces greater cooperation than either a shift from high to low or a pattern of high unchanging cooperativeness.”12 Similarly, studies using variants of a bilateral monopoly bargaining game found that “a strategy of starting tough and then gradually making concessions is a more effective means of reaching an optimal division of resources than one in which a softer stance is maintained throughout.”13 When there are advantages in settling disputes, a relatively soft posture toward the end has benefits: it conveys to the other a willingness to settle after having established relatively high resistance points, that is, lack of movement from initial positions. This hypothesis was tested in this study. Half of the bargainers in each condition and dyad were randomly determined to receive a message before the negotiation began to make a move early, during the first 15 minutes; the other half were told to make a move late, during the

108 Flexibility in Negotiation

last 15 minutes of the negotiation. Bargainers who made late moves reached significantly more agreements than those who moved early: overall, 19 of 24 dyads in the late-move condition reached agreement compared to only 9 of 24 in the early-move condition (χ2 = 8.6, df = 1, p < .01); 10 of 12 dyads who moved late reached agreements on the regulatory commission issue as compared to only 4 of 12 in the early-move condition (χ2 = 6.2, df = 1, p < .02); with regard to the R&D organization issue, 9 of 12 of the late movers reached agreement compared to 5 of 12 early movers (χ2 = 2.8, df = 1, p < .10). These results support the earlier findings from gaming studies. They extend the evidence from simple gaming situations to complex simulations, bolstering the argument that concessions made after a period of no movement are more effective in producing agreements than early softness followed by a period of inf lexibility. However, it may be that the results reported to this point ref lect the role-players’ theories about how they should behave. This issue is taken up in the next part.

Role-players’ Theories Two samples of respondents were asked to indicate whether each of 32 aspects of a negotiating situation would, in general, increase, decrease, or have no appreciable effect on a negotiator’s willingness to change positions; these are the 16 variables embedded in the simulation scenarios worded in both directions, that is, corresponding to the “f lexible” and “inf lexible” conditions. One sample of 15 respondents closely resembled the simulation’s role-players. The other sample consisted of 42 French respondents, 25 of whom had firsthand negotiating experience, and the remaining 17 had participated in a program on international negotiation. Results were generally similar for the two samples. In both samples, wide agreement was obtained on whether positions were or were not linked to ideologies and whether the relationship between the nations was either friendly or antagonistic. Neither of these variables was judged to strongly inf luence decisions made in the decision-making or bargaining exercises. Strong effects were also obtained in both samples for media coverage and the best alternative to a negotiated agreement, both of which were judged to inf luence decisions in the decision-making exercise; media coverage inf luenced f lexibility in both the decision-making and bargaining exercises. Although agreement was also obtained on concessions made by others and on the frequency of informal meetings, these variables were not judged to strongly inf luence f lexibility in the simulation exercises. On the other hand, while peripheral or central location was judged to have a strong impact in stage 2 of the bargaining simulation, this variable was thought to have virtually no effect on f lexibility in theory. Overall, only modest correspondences were obtained between the simulation and theoretical judgments. It appears that the simulation results correspond to role-players’ theories with regard to media coverage and, to a lesser extent, best alternatives to negotiated agreements. Their theories about other variables

Situational Levers

109

seemed to have little inf luence on their decisions in the simulation exercises. Still unknown, however, are the theories or assumptions made by professional negotiators and the impacts of these theories on their negotiating behavior.

Meta-analysis The simulation approach has identified various aspects of the negotiating situation that inf luence f lexible or inf lexible behavior. These inf luences were shown to operate in combination at different junctures of the simulated conference. Their relative impacts as part of scenario packages of variables were also determined. As I noted previously, these variables were derived from earlier laboratory experiments on bargaining behavior. Unlike our simulation approach, the experiments explored the effects of only one or a few variables at a time; combined effects can be assessed only as statistical interactions between two or three variables. With the advent of meta-analytic procedures, however, it is now possible to assess effects on f lexible negotiating behavior of each of several variables across the many experiments in which they were included.14 The approach consists of statistical procedures that estimate the strength of each of the variables, expressed as an effect size.15 It also allows for a comparison of results obtained from the simulation described previously with those obtained from the earlier laboratory experiments. Several comparisons were made. First, the overall effect sizes (strength of relationship between inf luences and f lexible behavior) for the simulation (decisionmaking exercise) and the meta-analysis were compared. Second, effect sizes were computed for each stage of the simulation and compared to effects obtained for those variables in the earlier experiments. Third, the rank order of the variables obtained from the simulation (pair-comparison analysis) and from the metaanalysis were compared.16 The average size of effects on measures of compromising behavior was considerably larger for the simulation (.59) than for the laboratory experiments (.31, a difference of .28). This indicates that the combination of variables has a stronger inf luence on decisions, whether f lexible or inf lexible, than each variable one at a time. Differences regarding size of effects were also large when results were compared for each simulation stage separately. Stronger effects were obtained for the simulation than for the experiments in each stage, with particularly large differences for the give-and-take (difference of .28) and the endgame (.21) and smaller differences for the earlier prenegotiation (.12) and setting-the-stage (.12) phases. Again, combined effects are stronger than the separate effects of the variables included in each of the stages. With regard to the rank ordering of the variables, both similarities and differences between the simulation and experiments were found. In both settings, participants or subjects were inf luenced by the type of prenegotiation preparation, with strategizing reducing f lexibility, and by the opponent’s strategy, with tougher opponents eliciting more f lexible responses. In neither setting did being

110 Flexibility in Negotiation

a representative – as compared to a nonrepresentative or adviser – produce strong effects, nor did the definition of the issue as large or small. These similarities provide evidence for general effects occurring in different kinds of situations. Two variables produced different effects in the two settings. Although, as previously noted, media coverage had strong impacts in the simulation, visibility produced only modest effects in the experiments. On the other hand, time pressure in the form of deadlines or strike costs produced more f lexibility in the experiments than in the simulation, where participants were either under a deadline or not faced with an announced deadline during the endgame. These differences suggest that the effects for these variables may be specific to the situation, in this case, a multilateral conference versus bilateral bargaining. Taken together, the various analyses suggest some conclusions about the conditions that may produce intransigence in negotiation. For example, intransigence is particularly likely when negotiators prepare strategies in cohesive groups for talks in which only a few issues are being contested; when they perform before an audience that creates strong face-saving pressures; when they are accountable to constituents and do not expect future interactions with their opponents; when the differences between positions on important issues are derived from long-held opposed social attitudes or ideologies made explicit during bargaining; and when they are faced with a tough or exploitative opponent whose intentions are easy to discern. These are the aspects of a situation that can be altered as part of a tactical approach to negotiating beneficial agreements. They are also factors that can be used for diagnosing cases in terms of the likelihood that an agreement will occur, as illustrated in the discussion to follow.

Diagnosing Negotiating Flexibility The results reported to this point address theoretical issues concerning inf luences on negotiating f lexibility. In this part, some practical implications of the research will be illustrated by showing how the findings can be used to diagnose outcomes likely to be achieved in a negotiation. Based on answers to a series of questions, a computer program produces estimates of negotiating f lexibility for each party represented at the table. The estimates are then converted into likely outcomes ranging from deadlock to mutually satisfactory agreements. The questions are variables explored in the previously discussed work on situational levers and derive from the body of experimental studies included in the meta-analysis. They are grouped into five categories: issues; parties; delegation characteristics; situation, including external events; and process. For example, with regard to the issues, a user would be asked to indicate how important or how complex each issue is relative to the other issues on the table. With regard to the delegation, a user would be asked whether delegates had prepared for this negotiation by strategizing, studying the issues, or a mix of strategy and study. Answers to each question were coded in the direction of higher numbers for the

Situational Levers

111

“more f lexible” choice, for example, choosing a less important issue or studying the issues prior to negotiating. The coded answer for each question was then multiplied by a weight, ranging from zero to 1, assigned on the basis of the metaanalysis results discussed earlier: variables with stronger effect sizes received a higher weighting than those with weaker effect sizes. However, the weights could be altered by the user should he or she have another reason for ordering the variables (questions) in terms of their relative contribution to f lexibility. The weighted codes are aggregated both within and across the categories to provide an estimate of each party’s f lexibility (shown on a two-dimensional grid), which is then translated into a likely outcome.17 The program makes it possible to compare different cases of negotiation in terms of f lexibility and outcomes. Confidence in its diagnosed outcomes depends on the extent to which the diagnoses correspond to the actual outcomes obtained in the negotiation; a high correspondence suggests that the diagnoses are valid prognosticators of what happens. A test of correspondence was made on a number of completed negotiations, including negotiations on arms reduction, military base rights, internal security, territorial control (the Panama Canal), and on the environment (air pollution and acid rain), as well as such historical cases as the 1954 mutual security pact between the United States and the Republic of China, the trade talks between the United States and Turkey between 1937 and 1939, and the administrative agreement negotiated between Japan and the United States in 1951–52.18 The program’s questions were answered by scholars with specialized knowledge about one or another case. In 11 of the 13 cases coded, the diagnosed outcome – fair agreement, asymmetrical agreement, conclusion without an agreement, or impasse – corresponded to the outcome actually attained. In two cases, military base rights and the U.S.-Turkey trade agreement, the program produced a judgment of no agreement, when in fact both talks were characterized by long impasses preceding an eventual agreement. For the most cooperative negotiation, the U.S.-Japanese administrative agreement, the program produced the least ambiguous diagnosis of a fair agreement. These are impressive correspondences, lending confidence, at least to some extent, to projections for cases yet to be completed. The program has another interesting analytical feature. It allows an analyst to perform diagnoses based on the various parts of a negotiation: issues, parties, process, delegation activities, and situation. These parts can also be considered as different approaches to the study of negotiation. They ref lect the diversity of approaches found in the literature, for example, an emphasis on parties,19 process,20 bureaucratic politics,21 or situational factors22 as the primary inf luence on negotiating behavior. The program elucidates the impact on diagnosed outcomes of focusing on one or another of these factors. These impacts are now illustrated with both contemporary and historical cases. As noted, the U.S.-Japan administrative agreement was characterized by a cooperative process and produced an agreement that served both parties’

112 Flexibility in Negotiation

interests. This actual outcome corresponded to the diagnosed outcome of a fair agreement. It also corresponded to the diagnoses based on each of the factors, namely, the delegation’s activities, the situation, and the process (see Figure 9.1). Some deviation from a fair agreement was, however, found for the other historical cases. For the U.S.-Republic of China mutual security talks, the process indicators produced a diagnosis on the border between agreement and no agreement; both delegation and situational factors resulted in diagnoses of agreement. This corresponded to the actual talks, which consisted largely of competitive maneuvering despite the eventual agreement that satisfied both countries. The other two historical cases were difficult. This is reflected in our overall diagnosis, which was on the border between an agreement and no agreement. It is also reflected in the diagnoses based on the factors taken separately. For the U.S.Turkey trade agreement, an impasse was diagnosed for the situational factors. In fact, most of the deliberations were characterized by an impasse, and this may have been due more to aspects of the situation than to process or delegation activities. Similarly, for the U.S.-Saudi military assistance talks in 1950–51, the difficulties are reflected well in the diagnoses. Based only on the situation, the diagnosis suggested an impasse. Only the delegation factors suggested the possibility of an agreement. About a third of the way through these talks, the Saudi king reviewed the draft agreement and insisted on changes better suited to his country’s interests. In this case, the U.S. delegation had little latitude and followed instructions to work out an agreement. Indeed, it was the delegation factors that proved least problematic with respect to the outcome. In both cases, it was the situational factors that seemed to be most problematic. A more contemporary case shows high correspondence between diagnosis and actual outcome. The 1975–76 negotiation between Spain and the United States over base rights concluded with a treaty only after the parties could resolve a long impasse due largely to Spain’s ambitious demands. A reluctant Spanish delegation reached agreement shortly following Franco’s death in early 1976. Reflecting these events, the diagnosed outcome was on the border of cells E (fair agreement) and F (no agreement). The diagnosis also identifies the source of the problem, which was in process and in delegation activities: the delegations’ characteristics are located in a no-agreement cell; the process is located in an asymmetrical-agreement cell (see Figure 9.2). Indeed, the key problems of these talks were in the process, where asymmetrical demands and offers were made by the parties (resulting in an outcome that favored the United States), and in the delegations, where delays were caused by frequent consultations during and between rounds with the Spanish Foreign Ministry or the U.S. State Department. These problems are reflected quite accurately in the program’s diagnoses.23 Other contemporary cases show a high correspondence between the diagnoses and the course of events. The highly visible negotiations between the United States and the Soviet Union over intermediate nuclear force reductions

Situational Levers

more flexibility

U.S.

A

113

C

B

3 4 1 D

2

G

E

F

H

I

more flexibility Japan Key 1 2 3 4

FIGURE 9.1

Delegations' Activities Situation Process Total

A B C D E F G H I

Capitulation No agreement Fair agreement No agreement Fair agreement No agreement Impasse No agreement Capitulation

U.S. Japanese Administrative Agreement

NOTE: The figure is read as follows. Flexibility for the United States increases as one moves from the bottom to the top. Flexibility for Japan increases as one moves from left to right. Diagnoses of f lexibility were made separately for the factors of delegations’ activities, situation, and process. “Total” refers to a diagnosis made across all of the factors. The diagnostic results are indicated by the placement of the numbered factors in the cells. All the diagnosed factors are situated in the center cell, E, which, according to the key, indicates that a fair agreement is likely.

concluded in 1988 with an agreement considered by most observers to have been in the interests of both parties.24 Our diagnosis shows that the likely outcome is a fair agreement and that there are few problems with any of the parts, most of which are projected to result in an agreement. Interestingly, only the key issue, the extent of reductions, is shown to be problematic by its location in a no-agreement cell. A contrasting case was the 1986 negotiations between the Aquino regime in the Philippines and the National Democratic Front (NDF) over a cease-fire and reconciliation. These difficult talks ended in stalemate, due in large part to divisions with the NDF delegation and their unwillingness to consider various proposals put forward by a mediator.25 Again, these problems

114 Flexibility in Negotiation

more flexibility

U.S.

A

2 1

D

B

C

3 5 F

E

4 G

H

I

more flexibility Spain

1 2 3 4 5

FIGURE 9.2

Key Major Issue Delegations' Activities Situation Process Total

A B C D E F G H I

Capitulation No agreement Fair agreement No agreement Fair agreement No agreement Impasse No agreement Capitulation

U.S. Base Rights in Spain

NOTE: The figure is read as follows. Flexibility for the United States increases as one moves from the bottom to the top. Flexibility for Spain increases as one moves from left to right. The diagnoses of flexibility were made separately for the major issue in the talks as well as for the factors of delegations’ activities, situation, and process. “Total” refers to a diagnosis made across all of the factors. The diagnostic results are indicated by the placement of the numbered factors in the cells. The second factor, delegations’ activities, is in cell B, which, according to the key, indicates no agreement. The fourth factor, process, is found in cell 1, indicating capitulation by the Spanish delegation.

are identified well by the diagnosis, which shows projections of no agreement when either delegation characteristics or process is considered and of impasse when questions are answered only about the NDF party. Two contributions are made by these diagnostic procedures. One is that they pinpoint where problems are likely to be found – for example, in process, situation, or delegations – and therefore what might be done to improve the prospects for agreement. Another is that they shed light on some implications of emphasizing one or another factor as a primary inf luence on negotiating behavior.

Situational Levers

115

For example, focusing only on process could suggest a likely agreement for a particular case, while looking only at situational factors may indicate an impasse; taking both into account together could, on the other hand, suggest a tenuous agreement, located between agreement and impasse. More generally, the approach facilitates comparative analyses of cases, supplementing statistical analyses of data.26 It also demonstrates how experimental research can contribute to the analysis of complex cases of international negotiation.

Conclusion The research described in this article is multifaceted, including experiments, simulations, and the analysis of real-world cases of international negotiation. I began with questions about factors in the negotiation situation thought to inf luence a negotiator’s f lexibility. Using a simulation approach, I discovered that some aspects of the situation had stronger effects on f lexibility than others and that these effects varied with the stage of the negotiation. I also learned that some questions remained about the approach taken and the findings obtained in the initial simulation experiment. Further experimentation addressed these questions and provided evidence for the generality of the findings, the impact of the role-playing approach, and avenues for application. This evidence also has implications for the idea of f lexibility. Following is a summary of the key implications with respect to each of the questions. 1

2

3

4

Suggestions made by a mediator had less impact on decisions than other factors built into the design of the situation. These situational factors may override the impact on f lexibility of direct suggestions offered by a third party. Results obtained from the bargaining experiment replicated those obtained from the earlier decision-making experiment. Both studies highlighted the importance of media visibility as an inf luence on f lexible behavior. It appears that compromise is more likely if negotiations are held in private. Open agreements arrived at secretly support the less formal observations made in the context of both domestic and international case studies. Flexibility was also shown to be inf luenced by the timing of an opponent’s moves. Supporting earlier findings, negotiators reached agreements more often when their opponent exhibited f lexibility following a period of no movement. This finding adds the variable of timing to the idea of firm but f lexible behavior; early firmness followed by later f lexibility seems to work best.27 It may well be that the simulation results ref lect the theories held by roleplayers. Some correspondences were found between general theories and decisions made in the experiments. For example, media coverage inf luenced role-players’ decisions and was thought by respondents to have a strong impact on f lexible negotiating behavior. It may also be the case, however,

116 Flexibility in Negotiation

5

6

that decisions made by professionals also ref lect their theories about inf luences on their own behavior. Taken together, results obtained from laboratory experiments and the simulations suggest a short list of factors in the situation that can be altered for impact on flexible behavior. Such engineering of the situation to produce agreements is an alternative to the more traditional forms of mediation that rely on suggestions. The impacts are likely to be particularly strong when the factors are combined in the design of the negotiating situation. The key factors can also be used to diagnose ongoing cases for insights into the likelihood that agreement will occur. With the help of a computer program, both historical and relatively contemporary international cases have been diagnosed. The high correspondences obtained between the diagnosed and actual outcomes for many of these cases indicate that the factors influence diplomats’ flexibility in a manner similar to the way they influence the decisions made by our role-players.

The results obtained to date provide corroboration across settings and research approaches for certain relationships between aspects of the situation and negotiating f lexibility. Further work along these lines would investigate the extent to which these relationships generalize to other negotiating arenas.

Notes 1 Alan Coddington, Theories of the Bargaining Process (Chicago: Aldine, 1968); Glenn H. Snyder and Paul Diesing, Conflict among Nations: Bargaining, Decision-Making, and System Structure in International Crisis (Princeton, NJ: Princeton University Press, 1977); Daniel Druckman, “Boundary Role Conflict: Negotiation as Dual Responsiveness,” Journal of Conflict Resolution, 21(4):639–62 (Dec. 1977). 2 These authors distinguish between a settlement-oriented and a problem-solving style. Their results showed that the problem-solving style focused more on understanding the causes of conflict through persistent questioning. It was depicted by them as a more structured and vigorous approach to conflict resolution than the settlement-oriented approach used by some mediators. See Kenneth Kressel et al., “The Settlement-Orientation vs. the Problem-Solving Style in Custody Mediation,” Journal of Social Issues, 50(1):67–84 (1994). 3 For more discussion of design issues, see Daniel Druckman, “Tools for Discovery: Experimenting with Simulations,” Simulation & Gaming, 25(4):446–55 (Dec. 1994); idem, “Understanding the Operation of Complex Social Systems: Some Uses of Simulation Design,” Simulation & Games, 2(2):173–95 (June 1971). 4 A good example of this dilemma is the work done on simulating international processes reviewed in Harold Guetzkow and J. J. Valadez, Simulated International Processes: Theories and Research in Global Modeling (Beverly Hills, CA: Sage, 1981). 5 Examples are experimental simulations such as those reviewed in Druckman, “Tools for Discovery,” pp. 446–55. For a specific illustration, see Daniel Druckman, Benjamin J. Broome, and Susan H. Korper, “Value Differences and Conflict Resolution: Facilitation or Delinking?” Journal of Conflict Resolution, 32(3):489–510 (Sept. 1988). 6 For more on this issue, see Robert Mahoney and Daniel Druckman, “Simulation, Experimentation, and Context: Dimensions of Design and Inference,” Simulation & Games, 6(3):235–70 (Sept. 1975).

Situational Levers

117

7 For further details on design and procedures, see Daniel Druckman, “The Situational Levers of Negotiating Flexibility,” Journal of Conflict Resolution, 37(2):236–76 (June 1993). 8 For details on computations, see J. P. Guilford, Psychometric Methods (New York: McGrawHill, 1954), chap. 7. 9 Thanks go to James Druckman, Richard Harris, Anna Korula, Hubert Touzard, and Lynn Wagner for their contributions to these analyses. 10 Richard E. Walton and Robert B. McKersie, A Behavioral Theory of Labor Negotiations: An Analysis of a Social Interaction System (New York: McGraw-Hill, 1965). 11 Holst observed that “secrecy permitted the participants to perform as real negotiators rather than as spokesmen for opposing interests. . . . An attempt was made to provide a setting conducive to human contact, conviviality and solidarity of effort. . . . Hence, [the talks] took place in the country side, in small resorts rather than city hotels.” Johan Jørgen Holst, “Reflections on the Making of a Tenuous Peace, or the Transformation of Images of the Middle-East Conflict” (Paper presented to the School of International and Public Affairs, Columbia University, New York City, 28 Sept. 1993), p. 5. 12 Jeffrey Z. Rubin and Bert R. Brown, The Social Psychology of Bargaining and Negotiation (New York: Academic Press, 1975), p. 275. 13 Ibid. 14 For a discussion of the approach see Frederic M. Wolf, Meta-Analysis: Quantitative Methods for Research Synthesis, Sage University Paper series on Quantitative Applications in the Social Sciences, 07–059 (Beverly Hills, CA: Sage, 1986). 15 The effect size is a ratio that expresses the magnitude of a relationship between an independent variable (IV), such as the presence or absence of an audience during negotiation, and a dependent variable (DV), such as the extent to which negotiators move from their initial positions or compromise. Effect sizes are calculated for each experiment that explores a particular IV-DV relationship. The separate effects are then averaged for an estimate of the overall relationship across all experiments in that IV-DV category. The results summarized in this section are reported in terms of average effect sizes. For further discussion of the concept of effect size, see ibid., pp. 23–37. 16 For details, see Druckman, “Situational Levers of Negotiating Flexibility,” pp. 262–65. See also idem, “Determinants of Compromising Behavior in Negotiation: A MetaAnalysis,” Journal of Conflict Resolution, 38(3):507–56 (Sept. 1994). 17 A number of applications are illustrated in Daniel Druckman, “Statistical Analysis for Negotiation Support,” Theory and Decision, 34:215–33 (1993). 18 These historical cases were coded from declassified documents in the U.S. State Department’s Foreign Relations of the United States series deposited in the Library of Congress and a number of U.S. university libraries. The documents consist largely of correspondence and memoranda of conversations between principals in the negotiation as well as heads of state, foreign ministers, ambassadors, and other high-ranking officials of the countries involved in the talks. This is a valuable source of information for students of international negotiations. 19 For example, see Jeffrey Z. Rubin, “The Actors in Negotiation,” in International Negotiation: Analysis, Approaches, Issues, ed. Victor A. Kremenyuk (San Francisco: Jossey-Bass, 1991), pp. 90–99. 20 Examples include Otomar J. Bartos, Process and Outcome of Negotiations (New York: Columbia University Press, 1974); Christophe Dupont and Guy Olivier Faure, “The Negotiation Process,” in International Negotiation, ed. Kremenyuk, pp.  22–39; Charles Walcott and P. Terrence Hopmann, “Interaction Analysis and Bargaining Behavior,” in The Small Group in Political Science: The Last Two Decades of Development, ed. Robert T. Golembiewski (Athens: University of Georgia Press, 1978), pp. 251–61. 21 Examples include Robert L. Kahn, “Organizational Theory,” in International Negotiation, ed. Kremenyuk, pp. 148–63; John New-house, Cold Dawn: The Story of SALT (New York: Holt, 1973); Strobe Talbott, Endgame: The Inside Story of SALT II (New York: Harper &

118 Flexibility in Negotiation

22

23 24 25 26 27

Row, 1979); idem, Deadly Gambits: The Reagan Administration and the Stalemate in Nuclear Arms Control (New York: Knopf, 1984). Examples can be found in Daniel Druckman, “The Influence of the Situation in Inter-Party Conflict,” Journal of Conflict Resolution, 15(4):523–54 (Dec. 1971); idem, “Situational Levers of Negotiating Flexibility,” pp.  236–76; idem, “Determinants of Compromising Behavior,” pp. 507–56. For a description and analysis of the unfolding events in this case, see Daniel Druckman, “Stages, Turning Points, and Crises: Negotiating Military Base Rights, Spain and the United States,” Journal of Conflict Resolution, 30(2):327–60 (June 1986). For an analysis of this case, see Daniel Druckman, Jo L. Husbands, and Karin Johnston, “Turning Points in the INF Negotiations,” Negotiation Journal, 7(1):55–67 (Jan. 1991). For a description and analysis of these talks, see Daniel Druckman and Justin Green, “Playing Two Games: Internal Negotiations in the Philippines,” in Elusive Peace, ed. I. William Zartman (Washington, DC: Brookings Institution, 1995). For an application of statistics to the comparative analysis of international negotiations, see Daniel Druckman, “Dimensions of International Negotiations: Structures, Processes, and Outcomes,” Group Decision and Negotiation (in press). For a discussion of this idea, see Dean G. Pruitt, “Flexibility in Conflict Episodes,” this issue of The Annals of the American Academy of Political and Social Science.

PART 3

Conflicting Values and Interests

PART INTRODUCTION

The article in the chapter to follow deals with types of conf lict. This work was inspired by the literature on the sociology of conf lict. Among the earliest sources for understanding this approach to conf lict is the writing of Georg Simmel (see Simmel, 1955, translated by Wolfe) and its reformulation by Coser (1956). The contributions made by these sociologists are about sociation, group structure, and function. Yet the concept that I found most interesting was ideology. As an abstract worldview, different than personal attitudes or beliefs, an ideology is the substance of a group’s raison d’etre. Group conf licts are intensified when members believe that they are representing a larger cause. This sense renders differences over ideology to be non-negotiable. What is however subject to negotiation are conf licts over interests. It is when ideologies are infused with interests that things become particularly dicey. It was another sociologist, Vilhelm Aubert, who did the most incisive conceptual work on the way interests and ideologies interact through the course of trying to resolve conf licts. His 1963 article on competition and dissensus sets the stage for the article appearing in this section.

DOI: 10.4324/9781003293361-17

10 CONFLICT OF INTEREST AND VALUE DISSENSUS

A number of examples about the way that ideologies infuse conf licting interests are discussed in the Human Relations article in this chapter. The examples include the historical role played by intellectuals in radicalizing movements, foreign policy formulation, academic faculty disputes, and disputes within families. Another familiar example may further clarify the interplay between these types of conf lict. A group of chaplains from a university campus proposed the construction of an inter-faith chapel. They differed on such matters as governance, price, functions, and location. These were negotiable interests that could be resolved by compromise and trading off on different preferences. But there was another issue that slowed the decision-making process. In their discussion about the design of the chapel theological differences among them emerged. Apparently, there were scriptural reasons for preferring certain kinds of external and internal architectural designs. These preferences were sticking points because of the sanctity of their beliefs. Differences on the practical issues were intensified and the result was a stalemate that delayed the project. This example illustrates an interplay between ideologies and interests developed further in the article presented in this section. In the U.S. we are currently (as of June 2022) experiencing a bipartisan investigation of the January 6, 2021 insurrection of the Capitol building. The committee is revealing its findings. One committee member, a Republican representative from Wyoming, has publicly split from her party’s stance on the issue. In doing so, she has jeopardized her chances of re-election. This is a clear example of balancing values against interests. Interestingly our research has shown that decisions often come down on the side of interests (Druckman et al, 1974; Druckman et al., 1977). The Wyoming representative provides a counter example to the documented findings. DOI: 10.4324/9781003293361-18

Conflict of Interest and Value Dissensus 123

The article that appears in this Part builds on these examples. It highlights the dynamics of resolving conflicts with strong ideological and interest elements. The cycles of intensification and moderation captured by the set of propositions are applied to a case discussed in Chapter 1 above and published in Elusive Peace, edited by William Zartman (Druckman and Green, 1995). The discussion of the Philippines negotiation takes the reader though the two games of getting to the table and leaving the table with an agreement. The failed attempts to resolve the conflict between the Aquino government and the National Democratic Front was due in large part to irreconcilable differences in ideology with implications for the remainder of her time in office. This case study also complements earlier experimental work on this theoretical perspective (e.g., Zechmesiter and Druckman, 1973; Druckman et al., 1988). Another approach to the study of conflict types is to treat them separately and compare their impacts on conflict-resolving behavior. Referred to as a balancing perspective, we placed negotiators in simulated situations where they were asked to evaluate the relative importance of their interests and ideologies. Interestingly, the relative importance of these conflicts varied with the context being simulated: they were weighted equally in an interdenominational conflict similar to the chaplains example above (Krause et al., 1975); while interests prevailed in a negotiation over prison services, ideology was more important in debates over population policy (Druckman et al., 1977); values prevailed in the U.S. Congressional example described above. Taking the early work further, we asked about interventions that would reduce the intensity of these conf licts. Our question was answered with an easyto-implement intervention called affirming the other. Negotiators were asked to list five positive qualities of the person sitting on the other side of the table. These qualities were not shared. The intervention worked to increase joint outcomes in the value conf lict (Harinck and Druckman, 2017). A follow-up experiment showed that when the negotiators shared their lists, better outcomes occurred with a facilitative mediator (Harinck and Druckman, 2019). These experiments provide optimism about the possibilities of finding solutions to value conf licts. Conf lict of Interest and Value Dissensus: Propositions in the Sociology of Conf lict, Daniel Druckman and Kathleen Zechmeister. Human Relations 26 (4) (1973): 449–466. © Sage Publications, Ltd. Reprinted with permission.

CONFLICT OF INTEREST AND VALUE DISSENSUS Propositions in the Sociology of Conflict1 Daniel Druckman and Kathleen Zechmeister

Conf lict of interest between two parties is defined as a discrepancy between them in preferences for outcomes of decisions on the distribution of a scarce resource. The structure of conf lict of interest situations in terms of preferences for outcomes and the range of outcomes attainable has been analyzed extensively within the framework of game theory (e.g. Schelling, 1960). In addition, experimental social psychologists have examined some of the situational and personality factors that impede or facilitate the resolution of dyadic conf licts of interest (e.g. Terhune, 1970; Druckman, 1971a). A third perspective in the study of conf lict of interest is that of the sociology of conf lict. Conf lict of interest is considered in terms of the broader context in which it occurs. Included in this broader context are dimensions of difference between parties in addition to opposing preferences for outcomes (e.g. values, culture, ethnic background). One particular dimension of difference between contesting parties which has received some attention in this literature is that of differences in ideological orientation. In this paper we will consider the interplay between ideological or value differences and conf lict of interest and suggest some propositions which depict relationships between these processes and the intensity of conf licts. The interplay between ideological differences and conf lict of interest can be considered in terms of an extension of the experimental paradigm used to study relationships between manipulated variables and the intensity of inter-party conf licts of interest. The emphasis in this line of research is on factors which inf luence various operationalizations of conf lict intensity (e.g. ease with which conf licts are resolved in bargaining simulations, number of competitive choices in matrix games). Evidence has been presented which indicates that conf lict intensity varies with such attitudinal-dispositions as resistance to compromise (Druckman, 1967), need for power (Terhune, 1968) and Machiavellianism (Geis, 1964) and with such situational factors as type of communication opportunity

Conflict of Interest and Value Dissensus 125

(Krauss & Deutsch, 1966), prenegotiation experience (Druckman, 1968) and decision-latitude of representatives (Hermann & Kogan, 1968; Zechmeister & Druckman, 1973). The propositions presented in this paper are intended to introduce into this general paradigm other variables which are hypothesized to inf luence the intensity of conf licts. Before discussing the propositions, however, several examples of the phenomenon are presented.

Value Differences and Conflict of Interest Glenn and his collaborators (1970) distinguished between two types of conf lict: a conf lict of interests and a conf lict of understanding. A conf lict of interest occurs when each party, sharing the same understanding of the situation, prefers a different and somewhat incompatible solution to a problem involving either a distribution of scarce resources between them or a decision to share the work of solving it (the authors’ example). Game theory has been used as a framework for analyzing the relative efficacy of various compromise strategies in attaining a satisfactory resolution. However, many conf licts occur not because of competing interests but because the parties do not share the same conceptualization of the situation. This may occur because of divergent ideologies, values or cognitive structures. The mode of resolution for such conf licts is not joint compromise or concessions, making these conf licts less amenable to a game-theoretic analysis. An altered understanding of the situation by one or both parties is necessary. Glenn et al. (1970) discuss several ‘mediating structures’ for resolving these conf licts based on systems of ideas ‘either actually or potentially shared by the parties to the conf lict’ (p. 37). The approach taken in the present paper is to accept this distinction and to carry it forward by examining the interaction between these types of conf lict and the impact of this ‘interplay’ on the intensity of conf licts of interest. Similar distinctions between types of conf lict have been made by others. Aubert (1963) separated value dissensus from competition; Rapoport (1960) distinguished between ‘games’ (encounters designed for ‘playing out’ conf licts of interest) and ‘debates’ (encounters designed for discussion of ideological differences), and Hammond (1965) designed an experimental paradigm for the exploration of interpersonal ‘cognitive conf lict’ resolution or conf licts that occur between parties who have coincident interests or goals. A ‘finer’ distinction between types of conf lict was made recently by Marwell (1966) with respect to cleavages in legislative politics. He separated ‘competition’ (conf lict of interest) from ‘ideology’ (contestants disagree on goals for the group) and ‘expertise’ (contestants disagree on means for achieving a shared goal). His ideological conf lict is similar to Aubert’s value dissensus and Rapoport’s ideological differences, while his expertise conf lict is similar to Hammond’s cognitive conf lict. The ‘conf lict of understanding’ of Glenn et al. is more general and refers both to conf licts over means and conf licts over goals. In this paper, these various distinctions are combined to form the general class of ideological differences, and this is differentiated

126 Conflicting Values and Interests

from conf lict of interest or competition. However, in making this distinction we do not imply that the two types of conf lict are independent; conf licts of interest often derive from ideological differences. It is this particular type of conf lict, examples of which are presented below, that is considered in this paper. Examples of conf licts of interest that derive from ideological differences can be gleaned from several sources: Snow’s (1959) characterization of a dissensus in the academic-intellectual community; Rapoport’s (1964) discussion of foreign policy formulation; Druckman & Zechmeister’s (1970) simulation-model of urban political decision-making, and interpersonal relations within a family. Snow presented a caricature of two homogeneous ‘camps’ (humanists and scientists) polarized on ideological issues (e.g. the locus of culture is in the inheritance of man’s past or can be attained only in the future: humanists look to the past for knowledge while scientists believe that knowledge must be actively sought through novel discoveries) and characterized by contrasting modes of discourse (viz. the ‘game’ language of science versus the ‘myth’ language of the humanities: see Back, 1963). Competition between these communities in such areas as job availability, salaries, fellowships, grants and output can be seen to result from this dissensus which is perpetuated by structural differentiation in academic institutions and by a continued insulation between the sub-cultures. Conf licts of interest between hawks and doves in Congress over allocations for military spending can be seen as being derived from two polarized ideologies concerning the application of human values in foreign policy and international relations. These alternative modes of thought lead to disagreement over the relative efficacy of programs designed to prevent the outbreak of a nuclear war. (This is a conf lict over means or expertise conf lict as conceptualized by Marwell, 1966.) The strategic thinker (armer) accepts values as given and then proceeds to evaluate the effects of a given course of action on the environment (e.g. the capacity to destroy the opponent will lead to his backing down) while the conscience-driven thinker (disarmer) is concerned with the nature of the values themselves and conceives of actions primarily with regard to their effects upon the actor (e.g. the capacity to destroy an opponent will lead to escalation and war, bringing devastation to both parties) (see Rapoport, 1964: p. 187 ff.). In the area of domestic affairs, contrasting ideological orientations (e.g. the source of social problems is in the social structure which should be changed versus the source is in people who have not adjusted to the system) can be shown to lead to opposing preferences in the distribution of resources to programs that ref lect the contrasting ideologies (e.g. a guaranteed annual income versus tax credits to companies that agree to train unskilled workers). A simulation-model of political decision-making which highlights the interplay between the social change versus system maintenance ideologies and contrasting programs for the distribution of resources is presented in Druckman & Zechmeister (1970) and in Zechmeister & Druckman (1973). Finally, with regard to interpersonal relations in a family, competition between a spouse and a child over the receipt of such a scarce resource as attention from the other spouse may derive from opposing

Conflict of Interest and Value Dissensus 127

philosophies held by the two spouses (viz. a child-centered vs. a parent-centered approach to child-rearing).2 While these examples illustrate the diversity of settings in which conf licts of interest derive from ideological (or cognitive) differences, the processes that are depicted in the propositions (cleavages within and between groups, coalition formation, repeated interaction between the same groups, and cross-cutting loyalties) are seen to occur most vividly in the setting of political decision-making, either at the level of the small group conference (a policy-making committee) or at the system level (legislative politics). It is also in this setting that the pattern of conf lict is most likely to ref lect the interplay between ideological differences and conf lict of interest. Political organization, characterized by an interplay between these two types of conf lict, can be contrasted to the ‘sect-type’ of formal organization, characterized by ideological differences, and to the business ‘enterprisetype’ of organization, characterized by conf licts of interest (see van Doorn, 1966). The intensity of conf licts among political decision-makers is more likely to be affected by this interplay, depicted in the following propositions, than the intensity of conf licts that occur between religious sects or between representatives of different business enterprises. We will return to a consideration of the political decision-making process in a methodology section following a discussion of the propositionalized processes.3

Propositions The propositional format is used to highlight a series of interrelated processes that are presented as countervailing forces on the intensity of conf licts. Thus, Proposition 1 depicts an intensifying effect on conf licts that results from deriving conf lict of interest positions explicitly from a value dissensus. Proposition 2 takes into account the structure of ideological parties as a moderating inf luence on conf lict intensity while Proposition 3 describes a process of convergence of within-party differences that occurs through time resulting in conf licts that are more intense. Finally, Proposition 4 describes the moderating effects of crosscutting loyalties which serves to mitigate against a conf lict-intensifying spiral of ideological polarization and conf lict of interest. A brief section on methodological considerations for exploring the propositions follows a discussion of the propositionalized processes. Proposition 1: When contending positions in a conf lict of interest are derived explicitly from opposing ideological orientations, the conf lict will be more intense than when the link between competing positions and ideological orientations is not made explicit. (The more polarized are the parties in ideological positions, the more intense the conf lict.) The several narrative and empirical sources for this proposition include Coser’s (1956) reformulation of Simmel’s proposition concerning the intensification of conflicts between groups with contrasting ideologies, Schelling’s (1960) consideration of the bargaining strategy of linking a conflict of interest to a general principle,

128 Conflicting Values and Interests

Price’s (1965) characterization of factors that influence the stability of social systems, and empirical evidence presented by Druckman & Zechmeister (1970). It is the objectifying inf luence of ideological commitment on conf lict positions that serves to intensify conf licts of interest. According to Coser’s elaboration of Simmel’s proposition: ‘Conf licts in which participants feel that they are merely the representatives of collectivities and groups, fighting not for self but only for the ideals of the group they represent are likely to be more radical and merciless than those that are fought for personal reasons. Elimination of the personal element tends to make the conf lict sharper, in the absence of modifying elements which personal factors would normally introduce’. (p. 118) Commitment to the more abstract, and more idealized, interests of a group organized along ideological lines (we, not I), is seen as producing greater rigidity in the representative’s bargaining behavior, transforming the conf lict of interest into a conf lict of ‘truths’. Coser (1956) gives as an example of the intensifying effects of ideological commitment, a characterization of the American intellectual community as a force that has contributed to the ‘deepening and intensification of struggles by stripping them of their personal motivations and transforming them into struggles over “eternal truths”’ (p.  116). The extension of conf licts from immediate issues to political and ideological spheres is dramatically illustrated in the labor movement. Coser cites the work of Marx, Mannheim, Perlman and Schumpeter as sources for expositions on how ‘intellectuals’ have radicalized the movement. Discussions of bargaining strategies and international relations can also be cited as sources for this proposition. Schelling (1960, p.  34) describes the bargaining tactic of linking one’s position to a general principle as a way of communicating to one’s opponent that one is strongly committed to a position, and that he intends to remain adamant since capitulation would discredit the principle. The tactical process consists of, first, explicitly stating the principle and, secondly, indicating unambiguously to one’s opponent how it is related to the conf lict of interest position. Deadlocked deliberations are the likely result of a bilateral use of this strategy. In commenting on a means for reducing the intensity of inter-nation conf licts, Stagner (1967) recommends separating ideological from nationalistic (or group loyalty) energies. He posits that conf licts are ‘less intense when these powerful motives are not united in support of a single policy’ (p. 89). Following this line of reasoning, it is important to disentangle bargaining issues from the underlying, presuppositional base that supports the bargaining position – such separation would aid in the resolution of the conf lict of interest as well as facilitate a debate over the ‘real’ differences in ideology. Implications for this proposition can also be drawn from discussions of factors affecting the stability of social systems. The stability of a social system may

Conflict of Interest and Value Dissensus 129

depend in part on its ability to control the frequency with which opposing ideological ‘enclaves’ are drawm together to compete. Conf licts of interest between parties from opposing ideological ‘enclaves’ are more threatening to the stability of a social system than is competition between parties within an ‘enclave’. On this point, Price (1965), in his essay The Established Dissenters, remarks that: ‘A social system is stable as long as its disputes are not over fundamentals. On this principle, the present scientific establishment in the United States is healthy enough; its internal arguments are mainly over the ways in which its grants and contracts are distributed, and from the outside it is challenged mainly with respect to the amount of money it gets, or the specific uses in which the money is put. With one exception, there has never been in American politics a serious effort to present the arguments, from the point of view of a conservative philosophy, against the predominant point of view of the scientific establishment whose theories have so profoundly inf luenced our national policies’. (p. 95) In addition to the theoretical sources reviewed above, the proposition has been tested and confirmed experimentally. Using a simulation-model of political decision-making, Druckman & Zechmeister (1970) compared a condition in which opposing positions on resource allocation in a decision-making conference were derived explicitly from contrasting ideological orientations (social change vs. system maintenance) to a condition in which the conf lict of interest was not derived from contrasting ideologies. Differences between the conditions on conf lict resolution indices were significant. In the ‘Ideology’ condition, the decision-making dyads either deadlocked or one decision-maker conceded to his opposite number. In the ‘No Ideology’ condition, most of the dyads resolved the conf lict and the mode of resolution was usually mutual compromise. This finding was replicated in repeated runs of the experiment with male decision-makers and with those female decision-makers who were able to identify strongly with the role of ideological representative (see Druckman, 1971b). Extending this experimental paradigm further, the opposing parties can be considered to consist of several members or factions who have somewhat different ideological orientations and conf lict of interest positions. The structure of the parties is a variable which can be shown to have an intensifying or moderating effect on the conf lict. Competing parties are not homogeneous entities, either in ideological orientation or in conf lict of interest positions. Each party to the competition consists of representatives of similar, though not identical, orientation and positions. The nature of the units of competition often necessitate alliances between representatives of somewhat different ideological orientation. Also, there is a competitive advantage in reducing the number of groups in competition for scarce resources. The effect on conf lict intensity of such collusion for effective competition, and the concomitant within-group differences, is depicted in Proposition 2.

130 Conflicting Values and Interests

The dynamics of the process (changes that occur through time or repeated interactions between and within the competitive units) are depicted in Proposition 3. Finally, Proposition 4 describes the stabilizing inf luence of the structurallydetermined units of competition: cross-cutting interests among representatives of diverse ideological positions are forced by the need to collude occasionally in order to come out favorably in a competition. Proposition 2. Differences in orientation within each ideological group and/ or differences within groups in conf lict of interest positions will result in a less intense conf lict between the groups than when parties within each group are homogeneous in orientation or positions. The intensifying effects on conf lict of a link between ideological differences and conf lict of interest positions, depicted in Proposition 1, may be moderated somewhat by differences within the ‘camps’ on either (or both) ideological or conf lict of interest positions. Differences between parties within groups often lead to factionation, especially when these differences are on ideological positions. However, there is a countervailing force, based on social structural considerations, that mitigates against factionalization. In the competition for scarce resources, there is a competitive advantage in reducing the number of groups in contention. Structural unification through collusion has the advantage of reducing the number of parties. However, this process also has an effect on conf lict intensity. Schisms or cleavages within groups, organizations or institutions may have the effect of reducing the intensity of conf licts between these parties. This is especially likely to occur when the contending parties publicly air their internal divisions and has been referred to as the principle of dissent. Recently, Evan & MacDougall tested the proposition that ‘if the bargaining strategies of both negotiating teams are guided by the principle of dissent, there may well be a greater probability not only of a resolution of the conf lict but also of a “creative” or an “integrative” type of resolution than if negotiations are guided by the principle of unanimity’ (1967, p. 399). Using a simulation-model of collective bargaining, these investigators created three conditions which varied in terms of the nature of inter- and intra-team divisions in positions. A condition in which there was a moderate and an extremist on each team (‘bilateral dissensus’) was compared to a condition in which there were two extremists on each team (‘bilateral consensus’) and to a condition in which there was a moderate and an extremist on one team and two extremists on the other (‘unilateral dissensus’). The results confirmed the hypothesis that the ‘bilateral dissensus’ condition would result in more integrative terms of agreement. A process analysis of the debate and postnegotiation perceptions suggested that the mechanism which produced this result was that the moderates on each team in this condition acted as mediators in attempting to bring their more extremist teammates together. While the results of the Evan & MacDougall experiment can be considered in part as corroboration for this proposition, a separation of types of conf lict, as suggested in this paper, can serve to extend their experimental design to various

Conflict of Interest and Value Dissensus 131

combinations of intra- and inter-team differences in both conf lict of interest positions and the more general ideological orientations from which these positions are derived. The two types of conf lict were confounded in that experiment. The position differences between and within the teams combined both a conf lict of interest over the distribution of resources and presuppositional differences concerning the desirability of automation. Thus, examination of the data did not permit an assessment of the way in which the two types of differences interacted in producing a resolution through negotiations.4 According to this proposition, within-team differences on either ideological or conf lict of interest positions serves as a mediative mechanism, countervailing against the intensifying effects of ideological differences on conf licts of interest (Proposition 1). In its experimental form, a test of this proposition would consist of comparing a condition in which there are within-team differences in ideological positions (‘bilateral dissensus’) but only between-team differences in the conf lict of interest positions with a condition in which there are no within-team differences on either ideological positions (‘bilateral consensus’) or conf lict of interest positions. The various other combinations of between- and within-team differences on both types of conf lict positions can also be included in the design for comparison. Proposition 3: Through time and repeated conf lict encounters, the various factions within the contending groups converge on their ideological positions, producing further polarization between the groups which serves, in turn, to exacerbate the intensity of future conf licts of interest. In addition to the effects of ideological differences on the intensity of conf licts of interest, as depicted in the earlier propositions, confrontations between contending parties to resolve a conf lict of interest may have a polarizing effect on the ideological differences. It is this reciprocal relationship between the two types of conf lict (each type of conf lict is affected by the other) that, according to this proposition, affects the intensity of the conf lict. This process is documented by Aubert (1963). He claims that the nature of a confrontation to resolve a conf lict of interest that derives from opposing ideological positions will affect the dissensus. A resolution that satisfies both parties (i.e. an integrative resolution) may lead to a meaningful dialogue on their contrasting ideological positions, reducing the polarization at the ideological level, while an unfavorable resolution may lead to further isolation between the parties, increasing the polarization at the ideological level. However, often, in spite of the nature of the resolution of conf licts of interest, the structure of the institutions in which the parties are contained prevents a meaningful dialogue from taking place. An example of this phenomenon which serves to highlight the processes depicted in this proposition, is provided by Campbell (1969) with respect to academic departments in a university. The structure of most universities is such that competition for funds, students, space, equipment, etc., occurs between departments (e.g. psychology vs sociology) within colleges. These conf licts of interest lead to dialogue among members of the department for the purpose of agreeing on a unified bargaining position. Thus, a pot-pourri of subdisciplines (e.g. physiological psychology

132 Conflicting Values and Interests

and social psychology) will unite under one banner and representatives of the diverse theoretical orientations will be forced to give their allegiance to the institutionally-defined unit of competition, viz. the department. The necessity for successive collusions through time results in a convergence of theoretical orientations among the specialties that are represented, which leads to further polarization between the competing units. Institutional pressures create the necessity for successive collusions between members of the same department. According to Campbell, such pressures also discourage scholars from branching out into related specialties, and serves to reinforce ingroup cohesiveness and a xenophobic attitude toward other departments. Thus, academic departments come to resemble ideological ‘enclaves’ with a high degree of ingroup cohesiveness and an exaggeration of between-group ideological or theoretical differences (and, repeated competitions between the same units, forced by the structure of the university, will serve to increase the ideological distance between departments through time).5 However, neither Aubert nor Campbell considers the implications of this process for future conf licts of interest. According to Proposition 1, the increased ideological polarization will lead to an intensified conf lict of interest. Propositions 1 and 2 consider the interplay between the two types of conf lict in static terms. This proposition extends this conceptualization by considering the dynamic relationship between the two types of conf lict. When the outcome of a conf lict of interest, or other structural factors, serves to exaggerate the polarization between parties, future competitions between these parties will be intensified (Proposition 1). Successful collusions between factions within groups (e.g. area specialties within a larger discipline) will mitigate successively against the moderating effects of within-group divisions (Proposition 2). The convergence that takes place within groups is likely, over time, to produce a more effective competitive unit (e.g. the academic department), escalating the intensity of future conf licts (Proposition 3). Without a mediating mechanism, the interplay between the types of conf lict through time and repeated interactions, would lead to an increasing spiral of polarization and conf lict intensity. One such mediational mechanism is discussed in the next proposition. Proposition 4: As a result of previous collusions between representatives of different ideological orientations (as well as common reference group identifications) there are cross-cutting interests which serve to reduce the intensity of conf licts of interest between them. Cross-cutting interests are presented here as a type of mediational mechanism which serves to prevent the processes depicted in Proposition 3, viz. withingroup convergence and between-group polarization, from leading to an unending escalatory spiral of ideological polarization and conf lict intensity. This is a mechanism at the societal level which moderates the intensity of between-group conf lict irrespective of how far apart the groups are in ideological orientation (i.e. ‘ideological distance’) at any point in time. This mechanism for reducing the intensity of a conf lict can be contrasted both to the mechanism for achieving this through cleavages, as depicted in Proposition 2, and that involving third-party

Conflict of Interest and Value Dissensus 133

mediators or arbitrators. As illustrated earlier, cleavages within groups may serve to reduce the intensity of between-group conf licts through a process by which moderates in both groups serve as a link in bringing their more extremist teammates together. Cross-cutting interests are another type of link between groups, caused not by schisms within groups but by shared identifications between them. The mechanism here is previous collusion or potential future collusion rather than links between moderates. When opposite number parties to a conf lict are allies or potential allies on other issues, or when parties to the same coalition on one issue are antagonists on another issue (i.e. when interests are cross-cutting) the conf lict is likely to be less intense than when the same parties are allies or antagonists on many issues (i.e. when interests are overlapping). Cross-cutting interests are more likely to occur in a f luid social system where alliances can be entered into freely or without the constraint of ideological differences. In other words, the moderating effects of cross-cutting interests are likely to be especially prominent in a social system where competitive advantage, rather than ideological affinity, is the major consideration in entering into coalitions. Perhaps the clearest example of the operation of cross-cutting interests is in the arena of legislative politics where party identification and positions on issues often conf lict. The plight of liberal Republicans and of conservative Democrats is legend. Both of these factions want their parties to win in elections, but a sweep would put them in a disadvantageous position in the next Congress on many issues. The party structure itself is essentially an alliance of diverse ideological factions who join forces for a competitive advantage. According to this proposition, cross-cutting interests between parties occur in a social structure which permits shifting alliances among parties from diverse ideological factions. This source for cross-cutting interests differs somewhat from (although it is not completely independent of ) the source which is characterized in classical treatments of the phenomenon. Axelrod (1970, pp. 158–163) reviewed both early and recent conceptualizations, including Ross and Dahrendorf, in which cross-cutting interests derive from common reference group identifications. These earlier treatments attempted to characterize societies in terms of the extent to which cleavages are cross-cutting or overlapping. High conf lict is expected when cleavages overlap (i.e. the society is polarized when individuals identify with groups that take the same stand on several issues) while conf lict is likely to be moderate when cleavages are cross-cutting (i.e. there is a high degree of interdependence among individuals when they identify with groups that take competing stands on several issues).6 In keeping with these propositions, crosscutting and overlapping cleavages at the level of social structure are likely to be ref lected in small policy-making groups. For example, decision-makers representing diverse segments of a social structure in which there is a high degree of cross-cutting cleavages may share similar positions on a variety of issues, reducing the intensity of conf lict on any particular issue. On the other hand, conf lict may be more intense among representative decision-makers in a highly polarized society, although the camaraderie among high level representatives, as a function

134 Conflicting Values and Interests

of their similar status and background, noted by a number of investigators (e.g. Modelski, 1970; Turk and Lefcowitz, 1962), may mask the extent of societal polarization.

Summary Proposition The propositional format was used to depict a process which shifted from an intensifying factor to a mechanism which exerts a moderating inf luence to another intensifying factor and back to another mechanism which serves to moderate the intensity of the conf lict. Each proposition was shown to derive logically from a previous proposition, producing two sets of intensifying and moderating inf luences. The four propositions can be summarized in the simplified format of two countervailing processes on the intensity of conf licts. Essentially, Propositions 1 and 3 focus on the intensifying effects of ideological polarization while Propositions 2 and 4 focus on the moderating effects of two mediational mechanisms which are built into the structure of the conf lict situation. The intensity of conf licts of interest results from both the extent to which the parties are polarized ideologically and the extent to which mediational mechanisms are present and operative. The more polarized the parties are at any point in time, and the fewer mediational mechanisms operating, the more intense the conf lict and vice versa. According to this set of propositions, these two forces operate simultaneously, keeping the intensity of the conf lict at a tolerable level. This process is depicted in the form of the following summary proposition. Proposition: Ideological polarization and mediational mechanisms built into the situation (within-group cleavages or cross-cutting interests) have countervailing effects on the intensity of conf licts of interest; an increase in the polarization of an ideological dissensus between competing parties increases the intensity of the conf lict while an increase in the sources of mediation between parties decreases the intensity of the conf lict. This proposition takes the form of an hypothesis which can be explored experimentally. The distance between ideological orientations (extreme vs moderate) can be made orthogonal to cross-cutting interests between opposing parties (high vs low) or divisions within groups on conf lict of interest positions (division vs unanimity). The design of this experiment is illustrated in matrix form in Figure 10.1.7 According to this proposition, the predicated result in terms of the arrangement of the four conditions (cells) on conf lict intensity is: A>B=C>D This predicted result could also be depicted in statistical terms as two main effects with the ‘extreme polarization’ conditions (A and C) resulting in a more intense conf lict of interest than the corresponding ‘moderate polarization’ conditions (B and D) and the ‘high mediation’ conditions (C and D) resulting in a less intense conf lict of interest than the ‘low mediation’ conditions (A and B).

Conflict of Interest and Value Dissensus 135

Ideological Polarization

Crosscutting Interests

FIGURE 10.1

Extreme

Moderate

Low

A

B

High

C

D

An Experimental Design in Which the Extent of Polarization on Ideologies is Orthogonal to Cross-Cutting Interests as an Example of a Mediational Mechanism

(These effects would be represented graphically as two parallel diagonal lines.) An experimental demonstration of this and the other propositions could be made at either the dyadic (small group) or systems level of analysis. This issue is considered brief ly in the next section.

The Dyad Vs The System: Some Methodological Considerations The presentation of the propositions is intended, inter alia, to serve heuristic purposes both as researchable hypotheses and as an explanatory structure for the less formal observations of conf lict which are made outside of the laboratory. The summary proposition presented above is an example of an hypothesis which leads to experimental predictions. In this concluding section, some suggestions are made toward the construction of a paradigm which might serve as a heuristic device in guiding observations of the processes depicted in the propositions. Two issues in particular which are relevant to the construction of such a paradigm are considered: an appropriate setting for making observations and the unit of analysis. It was mentioned earlier that an appropriate model for the exploration of the propositions is political decision-making. It is in this arena where the interplay between value dissensus and conf lict of interest is most salient (see van Doorn, 1966). A simulation model of political decision-making in which contrasting ideologies lead directly to conf lict of interest positions has been constructed (see Druckman & Zechmeister, 1970; Zechmeister & Druckman, 1973). An exploration of the propositions would consist of representing the processes in terms of components in such an operating model of political decision-making.8 The

136 Conflicting Values and Interests

model is defined by its components and an assessment is made of the effects of variations in these components on the intensity of the conf lict. Some of the components for a model include a link between contrasting ideological orientations (e.g. Rapoport’s hawk-dove controversy or Snow’s two cultures) and competing preferences for resource distribution (e.g. competition for funds to foreign aid or to domestic urban renewal programs); cleavages within and between parties; various types of constituencies, including ideological cohorts, a geographical base, and a political party; time sequencing of events, and alliances with opposite number parties on issues related only partially to the issue under consideration at a particular point in time (i.e. cross-cutting interests). Strategies for operationalizing these components include symbolic representation (e.g. ideological constituencies), judicious selection of participants (i.e. selected because they identify with contrasting ideological orientations), or situational manipulation (e.g. cross-cutting vs overlapping interests). The way in which these components are used and the way in which the intensity of the conf lict is defined depends in part on the nature of the units of analysis that an investigator focuses upon. Political decision-making can be characterized in terms of small policymaking groups or in terms of a system such as legislative politics. Research on this phenomenon has focused on both of these levels. The intensity of a conf lict of interest can be measured as an attribute of a small group (e.g. two parties) or as an attribute of a system consisting of a plurality of interacting units. Research on inter-party conf lict has been conducted at both of these levels. These alternative units of analysis can be shown to be related both to contrasting perspectives on the operation of power in a political system and to alternative methodological strategies adopted by investigators who have used social simulations in the study of inter-party conf lict in the context of political decision-making. The contrasting theoretical perspectives and the methodological strategies are inextricably intertwined, with method f lowing out of theory and vice versa. A focus on the dyad vs a focus on the system is ref lected in the difference between an ‘inf luence’ perspective and a ‘social control’ perspective on the operation of power in a political system (see Gamson, 1968) as well as in the difference between experimental simulations (e.g. Evan and MacDougall, 1967) and systems simulations (e.g. Guetzkow, 1968) for representing and observing the processes depicted in the propositions. From the standpoint of an inf luence theory perspective, the social system is conceptualized in terms of competition for power between representatives of diverse interest or ideological groups. The focal process is the strategy of conf lict among participants who compete for a share in the distribution of such scarce system resources as positions of authority (see Gamson, 1968, pp. 2–10). A corresponding methodological approach is represented by the conf lict of interest, gaming paradigm used for behavioral analysis of strategies employed by participants in a laboratory setting. This paradigm is used to investigate the components of variance in players’ strategies accounted for by manipulated variables that are embedded in, for example, a conf lict simulation (e.g. see Evan & MacDougall, 1967). The procedure essentially consists of

Conflict of Interest and Value Dissensus 137

assessing the effects of variations in one component of the model, e.g. amount of ideological polarization, on a dependent variable, e.g. conf lict intensity, while holding other components of the model constant. (The experimental predictions which derived from the summary proposition discussed above are based on this paradigm.) The contrasting ‘social control’ perspective conceptualizes the social system in terms of the regulation of conf lict in order to mobilize resources for the attainment of societal goals. The focal process is social organization for sustaining a level of stability in the political system rather than attempts by individual parties to win or resolve conf licts of interest over resources necessary to obtain power (see Gamson, 1968 pp. 11–17). A corresponding methodological approach is represented by the construction of models of entire social systems in order to explore stabilities or changes in such system-processes as the amount of tension in the system (e.g. see Crow, 1966). A major feature of a systems simulation is that it allows one to explore the development of a process through time (e.g. the intensity of conf lict between alliances) as it changes concomitantly with other processes in the model (e.g. changes in coalitions) (see Proposition 3). This type of simulation is in the tradition of dynamic equilibrium models and is perhaps best represented in the work on such complex simulation models as the InterNation Simulation (see Guetzkow, 1968). The major concern of system simulation designers is to achieve a construction which is homomorphic to a referent system, i.e. a model that represents the central features of the system of interest. At this point, rather than to devolve into an extended discussion of modeling and unit of analysis considerations, let us return to the primary focus of the paper and, in summary fashion, indicate the points that have been made. The phenomenon of interest is factors that inf luence the intensity of a conf lict of interest. The thesis that is being advanced is that the interplay between a value or ideological dissensus and a conf lict of interest has an impact on the intensity of conf lict. This relationship was presented in the form of a series of interrelated propositions for which both theoretical and empirical sources were cited. Taken together the four propositions could be conceived of as a balancing mechanism, alternating between an intensifying inf luence and a moderating inf luence on conf lict intensity. In addition, by taking into account changes over time, this conceptualization can be conceived of as a dynamic equilibrium model. The set of propositions was summarized in the form of a general proposition in which two mediating mechanisms built into the structure of conf lict situations were shown to be countervailing forces against the intensifying effects of ideological polarization. This summary proposition was shown to lead to experimental predictions. In an overview of methodological considerations related to these processes, we indicated that the experimental paradigm that has been used to explore the inf luence of situational and attitudinal factors on the intensity of conf licts can also be used to demonstrate the effects of the factors depicted in the propositions on conf lict intensity. However, an alternative paradigm which is based on the

138 Conflicting Values and Interests

system level of analysis is also appropriate. Thus, the intensity of conf lict might be measured as the ease of resolving a conf lict of interest between two parties (e.g. Evan & MacDougall, 1967) or as the level of tension (or ethnocentrism) in the system (e.g. Crow, 1966). Also, cross-cutting interests can be seen to operate at both the micro-level (e.g. the small policy-making group) and at the macrolevel (e.g. the United Nations). At either level, a simulation (experimental or systems) of political decision-making is an appropriate model for making observations that derive from the propositions. It is hoped that this theoretical effort is a first step toward the realization of such a demonstration. A corresponding effort to construct a model designed to explore the propositions is under way.

Notes 1 Support for this work has been provided by a Grant-in-Aid awarded to the authors by the Society for the Psychological Study of Social Issues. 2 Numerous other examples of the phenomenon could be provided. For example, competition between departments in a research organization (or, between members of a grants committee) over priorities for funding projects may derive from opposing values concerning payoffs from basic vs applied research. Similarly, competition between academic departments for funds may be seen as being closely linked to disciplinary presuppositional differences (e.g. humanities vs sciences). The latter example is discussed below. 3 Van Doorn (1966) distinguishes between ideological differences as the primary source of conflict in ‘sect-type’ organizations and differences in interests as the primary source of conflict in ‘business-type’ organizations. The political arena, on the other hand, resembles both the religious sect, insofar as ideological differences between subgroups are salient, as well as the business enterprise, insofar as competing vested interests in controlling scarce resources are also present. A distinguishing feature between these various institutional forms is that in political decision-making, conflict is more likely to result from an interplay between these two forces. Kelley (1970) also characterizes multi-party political decision-making in terms of an interplay between conflicts over utilities and ideological differences. He presents an elaborate discussion of various operationalizations of the concepts of utility and ideology in the context of political coalitions. A contrasting model of political decision-making is presented by Iklé & Leites (1962). These authors characterize political negotiations between nations in terms of a utility theory model in which only scaled position differences in a conflict of interest are considered. 4 An interpretation of the Evan & MacDougall results in terms of the ‘moderating effects of within-team differences’ is not unambiguous. In that experiment, the average position differences between the teams and the proposed mediational mechanism of within-team differences were confounded (i.e. in the ‘bilateral dissensus’ condition the average distance between the teams in conflict positions was less than the distance in positions in the other two conditions). A determination of the relative importance of these two factors depends upon a design in which within-team position differences (i.e. schisms) and between-team position differences are made orthogonal. A separation of types of conflict, as suggested by these propositions, could serve to unravel these confounding factors. 5 The polarizing effects of repeated conflicts of interest are also documented by Coser (1967) with respect to Soviet Russia-Red China competition and by Snow (1959) with respect to competition between the humanities and the sciences. For example, Coser remarked that ‘the clash of national interests is so intertwined with the clash of creeds that coexistence within a common political or ideological framework seem practically unattainable’ (pp. 239–240).

Conflict of Interest and Value Dissensus 139

6 Axelrod (1970) presents as an example of cross-cutting vs overlapping interests, positions on the issue of ‘preferences for social security payments’. Catholics and wealthy people take different positions on this issue. If most Catholics in society are poor and most nonCatholics are rich the intensity of conflict on this issue would be high since overlapping interests exist. However, if many Catholics are rich and many non-Catholics are poor, the intensity of conflict on this issue would be low since cross-cutting pressures exist. In the latter case, many individuals identify with two reference groups that take competing positions. 7 This illustration treats the process in static rather than dynamic terms, as indicated in Proposition 3. An appropriate model for testing these predictions is political decision-making. This is discussed below. 8 An operating model is a construction that can be used for hypothesis-testing, or for a comparison of results (outputs) obtained with those obtained in a referent system. This may be an all-man, man-machine, or all-machine simulation. For a definition and discussion of these various distinctions, see Guetzkow (1968).

References Aubert, V. (1963). Competition and dissensus: Two types of conf lict and of conf lict resolution. Journal of Conflict Resolution 7, 26–42. Axelrod, R. (1970). Conflict of interest: A theory of divergent goals with applications to politics. Chicago: Markham. Back, K. W. (1963). The game and the myth as two languages of social science. Behavioral Science 8, 66–71. Campbell, D. T. (1969). Ethnocentrism of disciplines and the fishscale model of omniscience. In M. Sherif (ed.), Problems of interdisciplinary relationships in the social sciences. Chicago: Aldine. Coser, L. (1956). The functions of social conflict. Glencoe, Ill.: Free Press. Coser, L. (1967). Continuities in the study of social conflict. New York: Free Press. Crow, W. J. (1966). Simulation: The construction and use of functioning models in international relations. In K. R. Hammond (ed.), The psychology of Egon Brunswik. New York: Holt, Rinehart and Winston. Druckman, D. (1967). Dogmatism, prenegotiation experience, and simulated group representation as determinants of dyadic behavior in a bargaining situation. Journal of Personality and Social Psychology 6, 279–290. Druckman, D. (1968). Prenegotiation experience and dyadic conf lict resolution in a bargaining situation. Journal of Experimental Social Psychology 4, 367–383. Druckman, D. (1971a). The inf luence of the situation in inter-party conf lict. Journal of Conflict Resolution 15, 523–554. Druckman, D. (1971b). Understanding the operation of complex social systems: Some uses of simulation design. Simulation and Games 2, 173–195. Druckman, D. & Zechmeister, K. (1970). Conf lict of interest and value dissensus. Human Relations 23, 431–438. Evan, W. M. & Macdougall, J. A. (1967). Interorganizational conflict: A labor-management bargaining experiment. Journal of Conflict Resolution 11, 398–413. Gamson, W. A. (1968). Power and discontent. Homewood, Ill.: The Dorsey Press. Geis, F. (1964). Machiavellianism and the manipulation of one’s fellow man. Paper presented at the annual meeting of the American Psychological Association, Los Angeles.

140 Conflicting Values and Interests

Glenn, E. S., Johnson, R. H., Kimmel, P. R. & Wedge, B. (1970). A cognitive interaction model to analyze culture conf lict in international relations. Journal of Conflict Resolution 14, 35–48. Guetzkow, H. (1968). Some correspondences between simulations and ‘realities’ in international relations. In M. Kaplan (ed.), New approaches to international relations. New York: St. Martin’s Press. Hammond, K. R. (1965). New directions in research on conf lict resolution. Journal of Social Issues 11, 44–66. Hermann, M. G. & Kogan, N. (1968). Negotiations in leader and delegate groups. Journal of Conflict Resolution 12, 332–344. Iklé, F. C. & Leites, N. (1962). Political negotiation as a process of modifying utilities. Journal of Conflict Resolution 6, 19–28. Kelley, E. W. (1970). Utility theory and political coalitions: Problems of operationalization. In S. Groennings, E. W. Kelley & M. Leiserson (eds.), The study of coalition behavior. New York: Holt, Rinehart and Winston. Krauss, R. M. & Deutsch, M. (1966). Communication in interpersonal bargaining. Journal of Personality and Social Psychology 4, 572–577. Marwell, G. (1966). Conf lict over proposed group actions: A typology of cleavage. Journal of Conflict Resolution 4, 427–435. Modelski, G. (1970). The world’s foreign ministers: A political elite. Journal of Conflict Resolution 14, 135–175. Price, D. K. (1965). The established dissenters. Daedalus 94, 84–116. Rapoport, A. (1960). Flights, games, and debates. Ann Arbor, Michigan: University of Michigan Press. Rapoport, A. (1964). Strategy and conscience. New York: Harper and Row. Schelling, T. C. (1960). The strategy of conflict. Cambridge MA: Harvard University Press. Snow, C. P. (1959). The two cultures and the scientific revolution. Cambridge: Cambridge University Press. Stagner, R. (1967). Psychological aspects of international conflict. Belmont, California: Wadsworth. Terhune, K. W. (1968). Motives, situation, and interpersonal conf lict within Prisoner’s Dilemma. Journal of Personality and Social Psychology Monograph Supplement 8(3), part 2. Terhune, K. W. (1970). The effects of personality in cooperation and conf lict. In P. G. Swingle (ed.), The structure of conflict. New York: Academic Press. Turk, H. & Lefcowitz, M. J. (1962). Towards a theory of representation between groups. Social Forces 40, 337–341. Van Doorn, J. A. A. (1966). Conf lict in formal organizations. In CIBA Foundation [A. de Reuck & J. Knight (eds.)], Conflict in society. London: Churchill. Zechmeister, K. & Druckman, D. (1973). Determinants of resolving a conf lict of interest: A simulation of political decision-making. Journal of Conflict Resolution 17, 63–88.

PART 4

Turning Points

PART INTRODUCTION

This topic has generated my largest cache of publications. It was introduced to the negotiation literature in a study of base-rights talks between Spain and the U.S. (Druckman, 1986). That study was a time-series analysis of ten rounds of negotiation over a period of 14 months. As a consultant to the U.S. delegation I had access to the transcripts of the discussions. The TP concept emerged from a careful reading of these transcripts. Certain decisions or events stood out as being key moments that usually turned impasses into progress. During my reading time, I came across McClelland’s 1972 study of the Taiwan Straits Confrontations during the period of 1950–64. His time series analyses featured the idea of turning points, which were regarded as interruptions in the chronology of events. This concept struck me as a useful way to organize the f low of back-andforth conversations in the base-rights talks. It proved helpful to the delegation in two ways: making sense of the peaks and valleys they were experiencing and developing strategies for altering the course of the talks. It also provided the data needed for an academic contribution, which is presented in the first article of this section, and was the basis for continuing contributions to the literature as shown in the next two selections.

DOI: 10.4324/9781003293361-20

11 BASE RIGHTS NEGOTIATION

On the face of it, a bilateral negotiation about extending the contract for renting bases during the Cold War appears pedestrian. As it turned out – and as described earlier – this particular base rights negotiation was anything but boring. It was also a very good case study for applying the TP concept. The article does several things. One is to introduce a framework consisting of four parts: stages, turning points, activities, and verbal behavior. Another was to describe the base-rights case including its players. A third was to use the framework for performing a qualitative analysis of the case. That analysis revealed that TPs typically followed negotiating crises such as a delegation walking out of the talks either in Madrid or Washington DC. The fourth part consisted of a content analysis of the verbal statements made by the delegates as recorded in the transcripts. This analysis revealed a particular pattern of responding referred to as “threshold-adjustment.” The article concludes with lessons learned, most notably the functions played by crises and the importance of external events. The agreement would not have occurred if the Spanish leader, Francesco Franco, had not died in January 1976.1 In an attempt to evaluate the generality of the response pattern discovered in the base rights case, I teamed up with Richard Harris to perform a statistical analysis of seven arms control cases (Druckman and Harris, 1990). Ten models of responsiveness were evaluated: various types of directional, trend, and comparative responding. These models capture the key concession strategies used in negotiation. The directional model is a direct form of responding in the reinforcement learning tradition of psychology: reward a concession, punish standing still or retracting a concession. This tit-for-tat strategy has been shown to be effective over the long term (Axelrod, 1984). The trend model – in five variants – captures the idea of monitoring the other’s past moves for evidence of concession patterns. Moves are made to support or counter these trends (Druckman, 1978). The comparative model – in four versions – focuses on the most recent move DOI: 10.4324/9781003293361-21

Base Rights Negotiation

145

made by the other negotiator. (See the Stoll and McAndrew, 1986, analysis of the strategic arms control process from 1969–79.) This information is used to close any gap in relative concessions and orients the talks toward mutual toughness or stalemate. The results were clear. Support across all the cases was obtained for comparative reciprocity, a response pattern found in the base-rights study. These results also demonstrate the inf luence of the immediate situation on behavior. Stages, Turning Points, and Crises: Negotiating Military Base Rights, Spain and the United States, Daniel Druckman. Journal of Conflict Resolution 30 (2) (1986): 327–360. © Sage Publication Inc. Reprinted with permission. Stages, Turning Points, and Crises.

Note 1 Base rights issues have resurfaced in recent discussions (January 2023) between the US Secretary of Defense and the Philippines President. Both countries are considering a renewal of US bases to project US power in Asia, particularly in the wake of China’s aggression. Lessons may be learned from our early writing on US Bases Overseas (McDonald and Bendahmane, 1990).

STAGES, TURNING POINTS, AND CRISES: NEGOTIATING MILITARY BASE RIGHTS, SPAIN AND THE UNITED STATES Daniel Druckman

International negotiation is a process consisting of political, economic, and legal decisions, some of which have “side effects” (e.g., making comprehensible relationships among nations in the international system). It serves the preformulated interests of governments, as ref lected in manifest foreign policy, and the process and consequences of negotiations force governments to reconsider policy options and to act in ways contrary to previous policy directives (see Druckman and Mahoney, 1977). These considerations suggest that the negotiation conference is the primary forum for defining and redefining relationships among nations, whether they are friends or enemies. However, an explanation for the unfolding events in negotiation remains a challenge. Its importance demands that efforts be made to develop a framework that provides an explanation for developments, enables one to order and organize the f low of activities, and makes evident a characteristic rhythm that may prevail in “all” negotiations. This article is presented as one attempt to construct such a framework. The framework is an attempt to capture developments in the literature, including theoretical and case-study analyses, as well as insights gained through interviews with delegates to a variety of conferences.1 Various aspects are described in an earlier chapter by Druckman (1983). In the discussion to follow, framework concepts are interwoven with examples from various cases. Then the framework is used to capture the unfolding process of a bilateral negotiation, construed as “rhythms and patterns.” Similar to the approaches taken by Haskel (1974), Ramberg (1977), and Tracy (1978), among others, this analytical strategy consists of superimposing more general concepts on case-specific details.2

Base Rights Negotiation

147

This article is organized into four parts. First, a general description of the negotiation process is presented in terms of stages, turning points, crises, and activities. These concepts are construed as components of a framework; each is defined for analysis. Second, an overview of the negotiation between Spain and the United States concerning military base rights precedes an analysis of the case. Third, a series of qualitative and quantitative analyses demonstrates an application of the framework. The qualitative analyses consist of demonstrating the progression of the negotiation through stages, internal and external influences on the process, and the pattern of negotiator’s activities. The quantitative analyses are based on a content analysis of the discussions; indices are developed and correlations among the indices computed. Finally, some implications of the analyses are discussed and six lessons learned are advanced.

The Negotiation Process: A Framework International negotiation is a process. Clues as to the nature of the final settlement are contained in the process, in addition to the initial preferences of one party or the other. Negotiators proceed usually with more attention paid to the process of negotiating than to where the process will lead. Serious thought is given to what is acceptable only after deliberations have begun. What is acceptable is a function of what is possible, and this is demonstrated in the act of negotiation. Over time the negotiation accumulates a settlement from the bottom up – a phenomenon referred to as “building a package” (see Druckman, 1977; Winham, 1977). Generally this consists of four stages: defining the scope of, or agenda for, the negotiation; a search for formulas or principles; f lushing out the issues; and a search for the implementing details. Stages are one element of a framework that attempts to capture an “ebb and flow” marked by key events and depicted as trends. Focusing primarily on the small group of negotiators, I highlight interaction processes reflected in activities and verbal statements. Viewing the interactions as dyadic, the analysis probes complex patterns of mutual responsiveness as these are influenced by, and influence, key events referred to as “turning points” and “crises.” The insights generated by this analysis concern conditions for turning points, including the role played by external events and the causes and resolutions of impasses or crises. More broadly, the results underscore a need to explore further the conjunction between processes and events of international negotiation.

Stages Zartman (1976) cites the Vietnam peace negotiations as having passed through two distinct phases. The first concerned the U.S. attempt to find a formula acceptable to both sides: freezing the stalemate in place without removing either the North

148

Turning Points

Vietnamese or the Saigon regime from South Vietnam, and then selling the idea to Hanoi. The second concerned a search for the details that followed from the agreed image, “a process that was closer to discovery than to convergence.” On the other hand, he points out that there were a few cases of incremental decision making, such as the size of the International Control Commission, but these were rare. The SALT process has consisted largely of a search for referent principles and then for implementing details: The central problem has been one of shifting the parties’ conception of nuclear parity from one permitting deterrence to one foreclosing coercion. Once the principle is agreed on, the details of disarmament can be hung on, such as a search for numbers of launchers versus missiles or quantity of warheads, or dividing the strategic force structure into first-line and second-line forces (see Zartman, 1975).3 This distinction between phases suggests that certain conceptual problems must be resolved before bargaining on the details can take place. Problems at the conceptual stage are illustrated by negotiations as diverse as the Mutual and Balanced Force Reductions (MBFR) talks and the Philippines military base rights negotiation. MBFR is an example of a negotiation that has not resolved key conceptual problems: It is not clear how to conceive offensive force strength, a necessary preliminary agreement before the detailed adjustments in force size are entertained (Ruehl, 1976, 1982; Dean, 1983). In the case of the Philippines, a lack of agreement on a general framework of principles to serve as guidelines prevented the two sides from proceeding to a discussion of specific articles. The negotiation was stalled at this stage. The Philippine delegation, in particular, insisted that an agreement on principles must precede detailed consideration of issues (Washington Post, January 25, 1977). This process is set in motion around certain coordinating events referred to as “benchmarks.” Such events as the signing of a declaration of principles, a framework agreement (see Iklé, 1964: 215–216), or an interim agreement signal progress. Progress can be depicted as passage from one stage to another, such as an agenda debate, a search for guiding principles, defining the issues, bargaining for favorable concession exchanges, and a search for implementing details (Druckman, 1983: Figure 3.1). Although these stages are presumed to occur in any negotiation, their relative duration is likely to vary from one case to another. This can be illustrated. The extended agenda debate in the Seabed arms control negotiation (see Hopmann, 1974; Ramberg, 1977) is set off against the first-round agenda agreement obtained in the Spain military bases negotiation (see below). The strong inf luence of previous agreements serves to provide precedents for such negotiations as the renewal of cultural exchanges and the extension of status-of-forces agreements among allies. The previous agreements limit the area of dispute to one of merely updating those agreements. These negotiations move quickly to the bargaining stage (e.g., adjustments in compensation, taking account of inf lation) and then the provisions-implementing stage. Similarly, trade negotiations

Base Rights Negotiation

149

consist for the most part of bargaining to work out an acceptable exchange ratio. In sharp contrast to these types of forums are the various arms control negotiations. The examples of SALT and MBFR, suggested above, illustrate the tedious process of achieving an agreement in principle. Even when this stage is accomplished, the search for implementing details delays the attainment of an agreed package. In all of these cases, progression from one stage to another is accompanied by turning points, just as stalemates are accompanied by crises. Both turning points and crises are caused by identifiable events: One is regarded as a breakthrough; the other as an impasse that threatens the sustenance of the talks. The speed with which a negotiation progresses toward an agreed package depends largely on the incidence of turning points relative to crises. Anything that negotiators can do to increase the size of this “ratio” should move the talks along.4 Other renditions of the stage concept appear in the literature. Gulliver’s (1979) eight stages model is the most elaborate proposed to date. Pruitt’s (1981) distributive/coordinative distinction may be the most basic (see also Morley and Stephenson, 1977). Certain elements are found in each of the schemes, including the following: (1) sequential progression in chronological and conceptual time; (2) alternating antagonistic and coordinative behavior within and across stages; and (3) partially overlapping stages rather than rigid and precise steps toward an outcome. Less agreement exists on whether the stages are prescriptions for attaining outcomes or descriptions of processes observed in diverse cases. The stage concept can be a useful conceptual device for negotiators and analysts. It can be used to chart progress, to identify disjunctures in the process, and to provide a recognized pattern of expectations and behavior. Gulliver (1979) and Pruitt (1981) note the necessity of experiencing stages to know what is possible and to evaluate alternative agreements. According to Gulliver (1979: 177): “The parties need to experience the process and gain the experience of each other and of themselves so that they come to accept a particular outcome as satisfactory.” It is also a tool for managing cognitive complexity, a function of particular relevance to multilateral international conferences (Haas, 1980).

Turning Points The process of moving from formulas to details can be characterized in terms of “turning points.”5 Taking the process through several turning points reveals the buildup of a momentum toward agreement (see Zartman, 1975). The first consists of the point at which both (all) sides agree that negotiations are realistic because they perceive each other’s expectations to be within range. This may take the form of a declaration of principles, reinforced by repeated statements of agreement to

150

Turning Points

adhere to those principles or to continue to negotiate in good faith. This stage is more problematic in innovative negotiations in which there is no precedent for renegotiating or continuing previous agreements (compare MBFR or SALT with military bases). Agreement in principle leads to the preparation of an agenda for the talks. This achievement is less problematic in talks between friendly nations who seek to reinforce and extend their relationship. It is more of a problem in negotiations between nations seeking to redefine their relationship. A second turning point occurs when an agreement has been reached on the interpretation of the problem. This leads to an attempt to negotiate a framework from which details can be deduced. Agreement on a framework is a third turning point. At this point, each side estimates the range of concessions needed to reach an agreement and acknowledges a commitment to see the negotiations through to a final document. This is when the parties proceed point by point to reconcile their different positions; this process is more like incremental convergence as each party adjusts its expectations in favor of what is needed to get an agreement. In the Spain talks, this stage occurred during the process of drafting the documents, including the articles of agreement and the supplementals. It occurred during the Kennedy round of the trade negotiations among 82 nations, between 1963 and 1967, when the delegates developed calculation formulas based on the percentage tariff cut and the volume of imports. Such measures as the weighted-average depth of cut and the 50% equivalence ratio facilitated comparisons of tariff reductions leading to necessary compromises (Winham, 1977). It is contended that these turning points occur in all negotiations; precisely when they occur varies from case to case. Two types of turning points are those that occur after a period of no progress and those that occur after a threat to the sustenance of the talks. The former, a period of no progress, is defined as an impasse; the latter, a threat to talks, is regarded here as a crisis. Both types of turning points are inflections in a trend, or “upturns” that represent either sudden progress or a return to a period of stability. The progress that occurs after an impasse often signals passage to a new stage of the negotiation. However, the recovery that follows a crisis or threatened breakdown usually does not signal a new stage. The new stage occurs during the period of stability after the recovery. The case described below provides an opportunity to examine periods of recovery following crises. A smooth negotiation process is one that moves through the stages with few impasses and no crises. Whether smooth or difficult, however, most negotiations reach a critical juncture during the period of stability: The talks must progress to the next stage, or an impasse will occur. The subsequent analysis provides indicators that can be used to chart progress and to forecast impasses or impending crises. This is regarded as a first step toward suggesting efficacious strategies.

Activities As the negotiations proceed through the “turning points,” negotiators engage in two general types of activities, monitoring and strategizing. Monitoring consists

Base Rights Negotiation

151

of determining the extent to which various parties maintain an interest in the negotiation process. The parties consist of the opponent, the executive department that is being represented, legislative committees that have to ratify the agreement, and other nations with a stake in the outcome. The interest of these agencies in the negotiation might be considered feedback variables. This is an assessment problem that emphasizes the representational aspects of negotiating: The negotiator must weigh the interests, but must also attempt to have an impact on those interests by the way he or she communicates the process of negotiating. This is a matter of strategy. Strategic variables involve the tactics used in moving the negotiations along. These consist of actions taken to influence the course, including tabling a position, decomposing and aggregating relevant information, packaging issues, deciding on the order of discussion of issues (simultaneous, successive, and so on), and proceeding point by point to reconcile the different positions of the parties. These activities occur predominantly during different stages and are inf luenced by different types of variables (see Table 11.1). Although the monitoring and coordinating functions occur throughout, these are the primary activities during the early rounds of the negotiation. Negotiators are less spontaneous and more cautious in the earlier rounds than in the later sessions; each side searches for indications from the other’s behavior that negotiations are being conducted in good faith, and both sides are reluctant to entertain new proposals without instructions from the home office. As the negotiation moves on, the teams engage one another in polemical discourse intended to give shape to a framework from which details can be derived. This conceptual stage of the negotiation culminates in an agreement in principle. Having attained such agreement, the parties are ready to consider the shape of the eventual package. Persuasive debating is interspersed with bargaining as the negotiating teams begin to define and refine the issues. At this stage, details are worked out by exchanging concessions on tangible items and by haggling over the wording of the agreement. Posturing considerations for exchanging concessions are entertained. The above discussion suggests stages of a negotiation, turning points, and activities engaged in by negotiators. These are components for a framework. Each can be defined for analysis. There are, however, other aspects of international negotiation that should be included. Two in particular are (a) factors that inf luence the course of developments, including those contextual, structural, social psychological, and bureaucratic variables that “drive” a conference, and (b) the behavior of the parties as revealed in the rhetoric of the deliberations, including such verbal expressions as threats, commitments, promises, and accommodations.

Influences The forces that drive a negotiation are construed as inf luences. Inf luences are of two types: facilitating effects and interferences on the negotiating process.

152

Turning Points

TABLE 11.1 Activities of Negotiators

Activity

Process

Stage

Occurs Seeking feedback: Is throughout but performance in line with emphasized in instructions? early rounds Are they negotiating in good faith? Stages 2 and 3: Persuasive debate Resolving conceptual middle rounds differences toward on acceptable definition of purpose or an agreement in principle Bargaining Building a balanced package Stage 4: later rounds by trading on preferences or utilities; involves deducing the details from the agreed principle Monitoring and coordinating

Critical Influences Team composition Channels of communication

Issue priorities and objectivees Cultural and ideological difference Relative power Audiences External events

The facilitating effects promote coordination and move a negotiation through turning points to an agreement. The interferences cause stalemates and crises, delaying progress toward an agreement. Both types of variables are found in each of the four categories of inf luences – contextual, structural, social psychological, and bureaucratic. Probable effects based in part on extrapolations from the negotiation literature are discussed in Druckman (1983). It is evident that different factors affect different aspects of the process, to wit, transition to a new stage, activities of negotiators, tendencies to cooperate or compete. Some examples (see Druckman, 1983: Table 3.3) serve to illustrate relationships. Studies of international negotiation have paid particular attention to contextual and bureaucratic inf luences. External events and the international climate are two contextual factors of interest: For example, the level of soft or hard bargaining in the test-ban talks was shown to vary with the level of tension in the international atmosphere (Hopmann and Smith, 1978). Problems of interagency coordination may be ref lected in negotiating behavior, as when internal bargaining causes delays and when unresolved internal differences are exploited by the other side (see the experiment described by Evan and MacDougall, 1967). Other relevant factors are differences in culture, bargaining power, and issue priorities or objectives. A number of experiments on cultural styles have shown strong effects on such aspects of the process as trends in cooperative and competitive behavior and on bargaining strategies (see the Druckman et al., 1976, study for a review). Effects have also been observed for differences in bargaining power, defined in terms of consequences of deadlock; the side that can least afford a stalemate is more likely to make a concession in order to get an agreement (Midgaard and Underdal, 1977). And differences

Base Rights Negotiation

153

between the teams in issue priorities were shown by Bonham (1971) to be a problem in attempts to make progress in the early talks on disarmament. However, most of the results obtained to date are more suggestive than conclusive. It is not clear just how these variables relate to one another, at what rate they are likely to change, and the precise nature of their inf luence on negotiating behavior. Effects on various aspects of the process are demonstrated in the subsequent case analysis.

Verbal Behavior The range of statements made during the course of a conference can be categorized in terms of apparent intent. One type of categorization is presented in Table 11.2. This system derives from earlier work on the analysis of small-group interaction processes (e.g., McGrath and Julian, 1963). It consists of categories that take account of the unique aspects of international negotiation as one form of small-group interaction (e.g., Hopmann, 1974). Some of the categories are based on the Bargaining Process Analysis system developed by Walcott and Hopmann (1975, 1978; see also Hopmann and Walcott, 1977). Others are similar to codes developed by McClelland (1972) for the World Events Interaction Survey. An elaboration of the rationale for certain aspects of the system can be found in the earlier articles. The categories can be combined in various ways to form indices. One particular aggregation is the distinction between “hard” (retractions, commitments, threats, accusations) and “soft” (initiations, accommodations, promises, praise) behaviors. The extent to which negotiations are hard is indicated by the ratio of hard to soft statements or by the percentage of all statements coded as hard. The decision to use this particular index is based on the assumption that soft behavior moves a negotiation toward an agreement, whereas hard behavior reduces the chances of attaining an agreement. Preliminary work suggests the usefulness of this distinction for depicting movement in a negotiation from one stage to another (Hopmann, 1974). The application reported below provides further evidence for the usefulness of the system. Other indices can be constructed from the categories. One of these is the difference(s) between the negotiating teams in percent hard. The trend of the differences would highlight the extent to which the teams are making the same kinds of statements at different junctures. Another index is based on the number of categories used per negotiating session. The trend of “variety” would highlight the extent to which negotiating behavior is characterized, at different junctures, by few or many categories. These indices could also be coordinated to various types of benchmarks, including transitions of different stages, turning points, and crises. The verbal codes can be used to probe fundamental patterns of interaction between negotiating teams. A basic issue in this domain is the extent to which opposing representatives are interdependent. In the parlance of negotiation

154

Turning Points

theory, interdependence is referred to as “responsiveness,” to wit, a negotiator’s behavior is contingent on what his or her opponent does, and vice versa. The extent to which opponents are mutually responsive is an empirical question of some importance. Implications for inf luence turn on the issue of benefits to be derived from matching or mismatching. This issue has received considerable attention in both empirical and theoretical literatures. Patterns of responsiveness have been examined in explicit and implicit bargaining contexts, including laboratory bargaining (Bartos, 1974; Druckman and Bonoma, 1976), arms control talks (Hopmann and Smith, 1978), American and Soviet foreign policy rhetoric (Tetlock, 1985), and counseling situations (Mercier and Johnson, 1984). Trends have been charted for offers

TABLE 11.2 Definitions of Categories Used to Code Negotiators’ Verbal Statements

Initiation:

Advances new proposal or states own position for first time (.67)a Accommodation: Expression of willingness to compromise a stated position or retract a proposal in the face of resistance (.66) Promise: Prediction of positive consequences if other behaves in the stated manner (.86) Praise: Commendation or gratuitous statement oriented toward other (.92)b Retraction: Taking back a previously made initiation or accommodation, or modification of previously stated positions to make them less agreeable (.70) Commitment: Reiteration of positions with statements that they will not change or declares own position nonnegotiable (.86) Threat: Prediction of negative consequences if other does not behave in the stated manner (involves sanctions or the withholding of potential reward) (.95) Accusation: Assign responsibility or blame to other (.96)b Agreement or revealed Expressions of agreement with the other’s statement, similarity of positions: whether or not the statement was a retraction or an accommodation (.82) Disagreement or revealed Expressions of disagreement with other’s statement, dissimilarity of positions: irrespective of type of statement (.98) Procedural statement: Behavior designed to move the discussion along, to organize the debate, or to suggest procedural innovations such as a suggestion to move to another agenda item (.68) b Question/answer (.98/.93) Restatement of positionc

a

b

c

Intercoder reliabilities as computed by Walcott and Hopmann (1978); all significant below the .001 level. Variants on the Walcott-Hopman categories of positive affect, negative affect, and subject change, respectively. Not included in the earlier versions of the bargaining process analysis system.

Base Rights Negotiation

155

and concessions (Druckman and Bonoma, 1976), as well as for specific words (Mercier and Johnson, 1984), linguistic structures (Tetlock, 1985), and inf luence tactics (Hopmann and King, 1976). The findings make evident a process that is more complex than action-reaction and may, in fact, take different forms at different stages of a negotiation. Those forms seem to be the outcome of complex delayed adjustment processes involving expectations and evaluations (Coddington, 1968; Druckman, 1978). It is these processes that lead to either impasses or acceptable agreements. The following analysis addresses the issue of responsiveness, focusing in particular on the ways in which observed patterns relate to crises and turning points.

Military Base Rights: Spain and the United States Background In 1970, the U.S. government (USG) and the government of Spain (GOS) signed an Agreement of Friendship and Cooperation (AFC). Among other things, this agreement defined for both nations the use of four military bases located within the Spanish borders, the amount of compensation paid, over a five-year period, by the USG for the operation of these facilities, personnel, weapons, equipment, and other related matters, including customs and taxes. (See Tracy, 1978, for background and analysis of these talks.) The AFC expired on September 26, 1975, at which time a one-year transitional period began. In preparation for the renegotiation of this agreement, a Joint Declaration of Principles was issued on July 9, 1974 (Department of State release no. 291). The Declaration consisted of 10 articles that reaffirmed the “existing cooperation-based on firm friendship” between the two nations. Shortly thereafter, a USG delegation, led by Ambassador-at-Large Robert J. McCloskey, was assigned the task of negotiating an extension of the 1970 agreement. Other permanent members of the USG delegation included representatives of various agencies in the Departments of State and Defense. The GOS delegation was led by the Hon. Juan Rovira, representing Foreign Minister Cortina. Other permanent members of the delegation included representatives from the Ministries of Foreign Affairs and Defense. The USG delegation entered the talks expecting that they would be concluded rather quickly. They were prepared merely to carry out the business of extending the provisions of the 1970 agreement. An extension of the earlier agreement would be consonant with the long and short-term objectives of USG policy regarding the bilateral relationship. The first round began on November 4, 1974, in Madrid.

Overview of the Negotiation Expectations that the negotiation would be brief were frustrated. The deliberations were long and difficult, lasting 14 months. During this period there were l0

156

Turning Points

plenary sessions, referred to as rounds. Each lasted for about four days, and there was usually a one-month interval between rounds. The business of the talks was either transacted or ref lected in these sessions. The rounds alternated between Madrid and Washington. During the interval, on the U.S. side, internal working committees were organized to prepare for the next session. The plenary was the essential forum for the talks. This format was supplemented during the summer rounds (Rounds 7–9) with bilateral working committees, which were organized for the intensive discussions of separate issues. The five committees were concerned with military matters, science and technology, implementation, cultural matters, and taxes and customs, and the committee members were experts in these fields. In each case, the committee chairman submitted recommendations to the plenary sessions. The work of the committees paved the way for a framework agreement, which was drafted in private meetings between Secretary Kissinger and Foreign Minister Cortina, during the period of September 15 to October 4 (New York Times, September 24, 1975). The final plenary session (Round 10) was convened in October 1975 to consider this document. The deliberations culminated in the five-year Treaty of Friendship and Cooperation, which was signed in Madrid on January 24, 1976 (U.S. Goverment Printing Office, Washington, DC, 1976; see also Washington Post, January 25, 1976) and ratified by the U.S. Senate on June 21, 1976. This seemingly smooth progression belies its difficulties. The negotiation was beset with ultimatums, stalemates, and crises, and the f low of events could be characterized in terms of alternating valleys and peaks. Just as one crisis was resolved, a new one would begin to surface. A threatened stalemate during Round 4 led to an impasse at the end of Round 5 (Washington Post, May 7, 1975). The progress achieved in Round 6, following an extended recess between Rounds 5 and 6, was threatened by a digression in Round 7 (Washington Post, June 2, 1975). The return to progress, achieved in Round 8, was threatened by a stalemate in the middle of Round 9 (Washington Post, September 15, 30, 1975). The ultimate crisis occurred in Round 10 following the major breakthrough of the talks: The GOS delegation refused to acknowledge the framework agreement authorized by both Secretary Kissinger and Foreign Minister Cortina (New York Times, December 18, 1975). Presumably, it was the drama of events that followed, culminating in Franco’s death, that forced Spain to reconsider this posture. With a new course decided on, the talks moved quickly to the drafting of the articles of agreement. The major problem of the talks was a difference in objectives. The USG wanted to extend the previous agreement, making minor adjustments to take into account inf lation and perhaps some relocation of equipment. The GOS aimed at redefining its role in the Western Community. It sought to do this either by gaining admission to NATO or by obtaining security guarantees on a bilateral basis from the USG. The talks were seen by the GOS as a vehicle for achieving these goals. The conf lict over objectives surfaced early and permeated the entire substance of the talks (Washington Post, November 8, 1974). The basic

Base Rights Negotiation

157

difference was overriding: The GOS held other issues hostage to its resolution. From this brief overview we now turn to an analysis of the negotiation in terms of the framework presented above.

Analysis of the Negotiation Stages, Turning Points, and Crises The negotiation passed through the stages outlined above. Passage from one stage to another was signaled by turning points, and the progression of the talks through the turning points was slowed by intermittent crises. These stages, turning points, and crises are depicted in Table 11.3. Staged movement from formula to detail, discussed earlier, characterized this negotiation. Agreement on the purpose of the negotiation had to be reached before bargaining on details could take place. This was a difficult transition. The U.S. image of an extension negotiation (namely, keeping the bases and adjusting the payments) was in sharp contrast to the Spanish image of redefining their role in the European security community by admitting them to NATO or attaining security guarantees from the United States (Washington Post, May 7, 24, 27, 1975). Because an agreement on this definition eluded the negotiators, they were not ready to work out the details: Even a change to a working committee structure, in order to hammer out agreements on the specific issues, did not lead to an overall agreement. A number of pressures, to be discussed below in the section on inf luences, finally forced Spain to accept the U.S. definition of purpose, and the United States also converged toward the Spanish image, paving the way to working out the details of a package that correlated facilities with the size of the payments. Key turning points are those that aided the transition from agreement on formula to bargaining over details. One of these occurred halfway through the negotiation when the United States was willing to consider a new defense relationship and Spain was willing to move to a working committee structure. Spain had trusted the United States to make the necessary arrangements for redefining the nature of the bilateral defense relationship, including the possibility of entry into NATO. The United States expected progress toward reaching agreements in the working committees. Progress did not occur until it became obvious to Spain that their own demands on a new defense relationship were non-negotiable. This then led to another turning point, consisting of a framework agreement from which bargaining points could be deduced (New York Times, September 24, 1975; see Table 11.3). Formulas and details, conceptual issues and bargaining, weaved through the four stages of this negotiation. The talks moved quickly through the first-stage agenda debate, and acceptance of the agenda indicated a willingness by both sides to proceed with an orderly discussion of the issues. The extended debate at the second stage was at times incisive, at times vague. It made apparent the basic conflict

TABLE 11.3 The Spain Military Bases Negotiation: Processes and Inf luences 158

Stages of Negotiation  Flushing out issues

Chronology Rounds 1–2 Crises

Rounds 3–6 Round 4: threats/counterthreats; commitments

Turning points

Agenda accepted early Round 2

Aftter Round 10 Rounds 7–9 Round 9: stalemate as GOS Round 10: breakdown as GOS refuese rejects each point of USG to acknowledge proposal framework agreeement Executions Franco’s Framework agreement death after signed between Rounds Round 10 9–10

Inf luences

Differences in objectives Opening speeches: disagreement over interpreting Declaration of Principles surfaces

Joint accommodation middle Round 6: USG accepts GOS definition; GOS accepts a move to working committees Differences in objectives; cultural Bureaucratic coordination; differences cultural differences Debate:steering by GOS to Monitoring of GOS maneuver USG into accepting bureaucracy by their definition of negotiation; delegation to determonitoring of GOS by USG to mine what positions it is determine what is necessary to to take; USG attempts move GOS to accept a working to construe issues in committee structure bargainable form Proposals, disagreements, Offers/counteroffers; restatement of positions rejections; amended (Round 3); threats/counterthreats offers/counteroffers; and commitments (Round 4); threats/counterthreats tabling of positions (Round 5); (Round 9) accommodations (Round 6)

Activities

Behaviors

Agenda Debate

Procedural statements; initiations; questions and answers

Bargaining

Internal coordination (GOS) Monitoring by GOS delegation to ascertain its own posture

Accommodations (USG)-accusations (GOS)

 

External events; bargaining power Bargaining

Turning Points

Formula Search

 

Base Rights Negotiation

159

of objectives that threatened in Round 4 to halt the deliberations. Partial reconciliation of the conflict indicated a willingness by both sides to try to make the issues negotiable. The committee structure of the third stage was designed to flush out the issues for bargaining, but disagreement over the framing of the issues (Round 9) threatened to deadlock the deliberations. Interim activity produced a framework agreement that indicated a willingness by both sides to work out the details of an acceptable package. This document made it possible, for the first time, to project the shape of an agreed package, with the details to be decided on by the delegations. The bargaining over details of the fourth stage was interrupted by another crisis caused by problems of coordination between the GOS delegation and the foreign minister’s office. A reconciliation of this problem, combined with external events, signaled a move to reconvene in order to work out the details and to prepare provisions for implementing the agreement. The ensuing activity led quickly to a final document. The model suggested here reveals a pattern for the analysis of this negotiation. The key elements are the stages, turning points, and crises. Turning points are signs indicating passage to the next stage. Crises disrupt an orderly progression from stage to stage. Between these turning points and crises, and within the stages, are activities. Different activities were emphasized at different stages, as depicted in Table 11.3 and discussed below.

Influences on the Negotiating Process In this negotiation, facilitating effects included the international climate, external events, and interim activities between officials who were at a higher political level than the delegates. Interferences included the issue priorities and objectives of the two parties, cultural differences, intrateam coordination, and delegation/ bureaucracy coordination. These factors produced different types of effects: Each affected the process at a particular juncture and had an impact on a particular aspect of the negotiation (see Table 11.3). These effects are summarized in terms of the crises and turning points. Crisis 1, Round 4. Apparently irreconcilable differences in issue priorities presented problems of synchronization of activities and behaviors leading to the threatened stalemate (Washington Post, May 7, 1975). These differences in objectives were aggravated by stylistic differences: The GOS preferred to argue in terms of broad generalities, whereas the USG preferred to focus on specific, tangible items. Stylistic preference may be due to a more basic difference in culture between the parties (see Young, 1968). Turning Point 1, Round 6. Interim activity included a visit to Madrid by President Ford and Secretary Kissinger (Washington Post, June 1, 1975), and their discussions led to a joint accommodation: The USG accepted the GOS objectives by promoting NATO membership in Brussels; and the GOS expressed a willingness to move to a committee structure that would facilitate working on the details of specific issues.

160

Turning Points

Crisis 2, Round 9. Lack of coordination within the GOS prevented the delegation from responding affirmatively to the concrete proposals of the USG. And the coordination problem was aggravated by disagreements within the GOS delegation between the diplomatic and the military délegates over posture. The resulting confusion was ref lected in the synchronization problem depicted above and led to the impasse at the end of round 9 (Washington Post, September 15, 1975). Turning Point 2, Round 10. The impasse was resolved by the framework agreement, which was produced as a result of high-level interim meetings between Secretaries Kissinger and Cortina (New York Times, September 24, 1975). The GOS accommodation could be regarded as an acquiescence to the “realities” of their situation. External events and relative bargaining power prevailed. Prospects for admission to NATO were dim. The cooling off of détente and the impending change of government in Spain forced Cortina to strive for a new agreement. The alternative for GOS would have been to pass up opportunities for modernizing its military and to forfeit the security provided by the U.S. bases. The USG bargaining position prevailed. Crisis 3, Round 10. This crisis was caused by a lack of coordination between the GOS delegation and the foreign minister’s office. The negotiation breakdown warranted private discussions between McCloskey and Cortina. (McCloskey briefed the Senate Foreign Relations Committee on these developments; New York Times, October 23, 1975.) Another plenary did not take place. Turning Point 3, January 1976 Meetings. The death of Franco and reactions by Western Europe to the executions in Spain precipitated a return to the table (New York Times, December 18, 1975); bargaining ensued. Differences in bargaining power ensured an agreement that was favorable to the USG (Washington Post, January 24, 1976). Conceivably, the GOS could have attained a better agreement earlier in the negotiation, but concessions that the USG was willing to make earlier were unnecessary at the end. The pattern suggested by these inf luences is one of countervailing effects. Through the course of the negotiation, the interfering effects of one type of variable were offset by the facilitating effects of another. The differences in issue priorities or objectives of the earlier rounds were moderated, after an impasse, by the joint accommodation achieved during the interim. The disjointed or “outof-phase” negotiation behavior in later rounds was offset by accommodations made during private high-level political discussions. The facilitating effects of the high-level interim meetings were aided by external events that pressured the sides into compromising their interests and goals. This pattern resembles a balancing mechanism, alternating between an intensifying inf luence and a moderating inf luence on the conf lict. When changes over time are taken into account, the pattern takes the form of a dynamic equilibrium model. Such a model has been proposed elsewhere as a conceptualization of the interplay between interests and ideologies in political decision making (Druckman and Zechmeister, 1973; Hopmann and Druckman, 1981).

Base Rights Negotiation

161

Activities of Negotiators The negotiators’ activities conformed generally to the pattern described in an earlier section. Aside from speeches and security briefings, the predominant activities in this negotiation were persuasive debate and monitoring. The bargaining that occurred was limited, for the most part, to private discussions that led to a framework agreement and to the January meetings, during which attempts were made to draft the articles of agreement. The contrasting USG and GOS postures produced a pattern of activity that could be considered “out of phase.” Debate and monitoring were used at different times by the two delegations. The USG posture consisted essentially of attempts to move the negotiation to a format that would facilitate bargaining. In the early rounds, this took the form of monitoring the GOS delegation to determine how they could be persuaded to accept a working-committee structure. In the middle rounds, the posture took the form of persuasive arguments designed to define the issues in such a way as to permit bargaining. The GOS posture consisted largely of efforts to move the negotiation in the direction suggested by their primary concern. In the early rounds, this posture took the form of persuasive appeals designed to maneuver the USG into accepting the GOS definition of the negotiation. In the middle rounds, the posture involved monitoring the Foreign Ministry and Defense Ministry bureaucracies in order to coordinate government policy with negotiating positions. This activity continued through the later rounds as confusion developed over just what posture the GOS delegation was to assume (see Table 11.3). This “out-of-phase” pattern of activity precipitated the three crises indicated in Table 11.3. The first crisis (Round 4) resulted from a lack of coordination between the delegations: The USG did not accede to the GOS steering attempts, and the GOS resisted the USG attempts to ascertain what was needed (e.g., the gestures) to move the negotiation to a working-committee structure. A similar lack of coordination precipitated the second crisis (Round 9). The USG offered a concrete proposal for consideration; but the lack of consistent GOS policy caused confusion within the delegation, leading to a rejection of each point in the proposal. The third crisis (Round 10) occurred as the USG delegation prepared to bargain over the details of the framework agreement. Communication problems within the GOS prevented their delegation from engaging in bargaining. Yet in spite of these crises, the negotiation eventually did move to a bargaining format. Each time a crisis occurred, interim activities or external events precipitated turning points (see Table 11.3), which kept the deliberations on course. Bargaining presumably occurred during the critical intervals. Clearly, there was bargaining over the wording of the agreement during the January meetings. The above appraisal suggests that progress in negotiation depends in part on the synchronization of activities within and between teams. But progress is also likely to depend on how these activities are executed. This is revealed in the rhetoric of negotiation – the types of statements made by negotiators

162

Turning Points

at different junctures. Persuasive appeals can consist of threats and commitments, or promises and accommodations. Tough bargainers are unyielding and make demands; soft bargainers are compromising and willing to consider the other’s point of view; and so on. The stance taken may reveal a negotiator’s intentions. Negotiators react to their perceptions of the other’s intentions (Druckman and Bonoma, 1976). And the course of deliberations seems to be influenced by this kind of action-reaction dynamic. It should be possible to project, on the basis of the kinds of statements made by negotiators, how a negotiation will progress through stages, turning points, and crises. First, however, it will be necessary to demonstrate a correspondence between statements and the course of events.

Negotiating Rhetoric: Content Analysis of Statements The coding system described in Table 11.2 was applied to the records of the plenary sessions.6 These consisted of verbatim statements, or elaborate summaries of statements, made by the principal negotiators on both sides. The records are relatively complete and accurate. Moreover, the delegates who were interviewed claimed that informal discussion and interim activities were reflected in the formal proceedings. From the coded statements a number of indices were constructed. These include the following: (1) Percent Soft or Percent Hard: the number of statements coded as soft or hard relative to the total number of statements coded in these categories for a defined time period, that is, a session or a phase. (2) Percent Agree or Percent Disagree: the number of statements per time period in the agree or disagree category relative to the total across the two categories. (3) Difference in Percent Hard: the absolute difference between the teams in the percentage of statements per time period that are coded as hard. (4) Variety Index: a measure, based on information theory, used to determine the way that responses fall into a fixed number of alternative categories. It is calculated according to the following formula: n

H ( p) =

∑ p log p i

i

i=1

where pi is the proportion of statements coded in each category relative to the total number of statements coded for that time period. H(p) is a measure of uncertainty and has been referred to as the entropy of the distribution (pi, p2, . . . pn) or of the variate X having pi as the probability of its being equal to i (see Morse et al., 1959: chapter 9). Put in another way, if there are ten categories, and events fall equally into those categories so that each occurs 10% of the time, the score is 1. This condition is one of complete uncertainty or randomness,

Base Rights Negotiation

163

in which no category is any more predictable than any other. Conversely, if all events occur in only one category and the other nine are empty sets, then the score is 0. Here, one has complete certainty about which category will occur and thus perfect predictability. The variety index is similar to Hrel, a ratio that was introduced to the international relations literature by McClelland (1972: 92), who describes the index as follows: A common sense way to view a series of Hrel numbers is to think in terms of a “fanning out” toward equality of the distribution across the category system with the larger figures and a “channeling in” of the distribution toward relatively frequent occurrences in fewer categories with the smaller figures. As the ratio approaches 1.000, it suggests not only that almost everything that could happen has been occurring but also that behaviors have shown increasing signs of disorderliness. The information measures do not tell us what the particular lack of ordering is, but they do give us a technical indication of a large amount of “variety” in the emissions. It is the “variety” notion that renders the index appropriate for the purposes of this project.7 (5) Crisis/ Turning Point Index: the occurrence (per time period) of a crisis, turning point, or neither, scored as crisis (1), period of stability (.5), and turning point (0). The index is based on the assumption that a given time period (session) can be characterized as primarily one or another of these three events. The difficulty of construing the variable as “amount of crisis/turning point” renders the index as one of three points separated by the same interval. The events are identifiable. A crisis results in a deadlock: Both sides acknowledge it by calling for a recess or by refusing to proceed to the next item of business. A turning point is a recovery from a crisis or a breakthrough: Both sides acknowledge the turning point by reconvening after a resolution, obtained during the recess, or by signing a framework agreement or other document that indicates progress toward a final settlement. A period of stability is one that is characterized by neither an identifiable crisis nor a turning point. For this case, seven of the sessions were classified as crises, four were turning points, and eleven were periods of stability.8

Trends in Soft/Hard Behavior Overall Trend Figure 11.1 shows the values of percent hard over the course of the negotiation. The trend is charted for the 22 coded sessions of Rounds 3 through 10. (The decision to exclude Rounds 1 and 2 from the analysis was based on incomplete data for

164

Turning Points

Round 1 and the fact that Round 2 consisted mostly of a lengthy security briefing.) For this analysis, all statements coded in the “soft” category (initiations, promises, accommodations, and so on) for both the GOS and the USG negotiators were aggregated for each session. Similarly, all statements coded in the “hard” category (e.g., retractions, threats, and commitments) for both teams were aggregated for each session. Each point represents a percentage calculated for each session within a round. The higher the score, the harder the behavior. The trend does correspond to the course of events. In particular, the following observations may be made: (1) The upward trend from the first session of Round 3 to the middle of Round 4 indicates increasingly hard behavior, culminating in the crisis of Round 4. This crisis threatened to stalemate the negotiation. (2) The recovery is ref lected in increased soft behavior, which continued until the third session of Round 5. The increase at that point preceded recognition, by both sides, of an impasse. (3) Following the two-month recess between Rounds 5 and 6, there was a dramatic decrease (increase) in hard (soft) behavior, which reached a low point at the second session of Round 6. (4) The digression that occurred during Round 7 is reflected in the sudden increase in hard behavior indicated for the second and third sessions of that round. (5) The return to progress is ref lected in increased soft behavior comparable to the low point attained in Round 6.

C

100 90

PERCENT HARD

80

C C

C

C TP

C

70 C 60 50

TP

40

TP TP

30 C = Crisis Session TP Turning Point

20 10 0 Session 0 Round

FIGURE 11.1

1 3

2 3 4 4

5 6 7 8 9 5

10 11 12 13 14 15 16 17 18 19 20 21 6

7

8

9

22 10

Trend in Percent Hard Behavior for Both Teams Across the Negotiating Sessions

Base Rights Negotiation

165

(6) The upward trend reached a high point at the end of Round 9. This increasingly hard behavior culminated in another impasse. (7) The hard behavior of Round 10 reflects the crisis caused by the GOS delegation’s unwillingness to acknowledge the authenticity of the framework agreement. Additional findings related to the overall trend are based on correlations computed among the various indices. The set of correlations is discussed below. These observations suggest that the types of statements made by the negotiators correspond to the course of events. The behavioral trend either signaled a crisis or ref lected a turning point: An upward trend always preceded a crisis. A downward trend usually preceded a turning point or a recovery. These observations are bolstered by the strong correlation between hard behavior and events (r = .64). Harder behavior occurred during crises than at other times (see Figure 11.1). Further insights into the process are derived from an analysis of the separate trends for the GOS and the USG teams. Separate Trends The separate trends are presented in terms of the percentage of total statements per session that are coded as hard. The USG and the GOS trends are illustrated in Figure 11.2. A comparison of the two trends suggests the following: (1) Overall, one pattern could be characterized as the inverse of the other. USG soft behavior was accompanied, for the most part, by GOS hard behavior, and vice versa. A correlation of -.59 between USG and GOS percent hard substantiates this observation. (2) Similar trends appear only for the later sessions of Round 4 and the first session of Round 5. (3) The inversion is strongest during the later rounds of the negotiation, being most apparent at the middle of Round 8. (4) Overall, the GOS delegation had a higher percentage of statements coded as hard. Moreover, the GOS team made more statements in most categories: They made three times as many threats, accommodations, and agree and disagree statements as the USG team. The two teams were approximately equal in number of commitments, with most of these statements coming during the early rounds (Rounds 3 and 4 = 35; Rounds 5–9 = 18). Finally, the teams did not differ in frequency of “questions,” “answers,” or “procedural” statements.9 Additional findings related to the separate trends are based on correlations computed among the various indices. These include the following: (5) A stylistic difference between the sides is suggested by the correlations between percent hard and the variety index. USG hard behavior was

166

Turning Points

100 Spanish Team United States Team C Crisis Session Refers to Points for Both Teams TP Turning Point

90

}

PERCENT HARD

80 70

C C

60 50

TP

C

C

TP C C

40 30

TP

20 TP

C

10 0 Session 0 Round

FIGURE 11.2

1 2 3 4 3

4

5 6 7

8 9 5

10 11 12 13 14 15 16 17 18 19 20 21 6

7

8

9

22 10

Trends in Percent Hard Behaviors for the Spanish and American Teams Across the Negotiating Sessions

accompanied by a small number of statements (primarily commitments); GOS hard behavior was accompanied by a large number of statements (e.g., threats, commitments, and retractions). The percent hard/variety index correlation for USG is -.41; the GOS correlation is .32 (see Table 11.4). (6) The correlation between the difference in percent hard and the variety index, reported in Table 11.4, is -.45: The larger the difference between the teams in percent hard, the fewer the variety of categories emitted by both sides.

Crises, Turning Points, and Negotiating Behavior An attempt was made to predict crises and turning points on the basis of negotiating behavior that occurred in the period preceding the event. To do this, a crisis/turning point variable was constructed. Each session was coded for the occurrence of a crisis (1), a period of stability (.5), or a turning point (0). Lagged correlations were computed between this index and each of the five indices of negotiating behavior, also coded for each session. These are reported in Table 11.4. The crisis/turning point index (CRISIS) at time t+ 1 was correlated with each of the indices of negotiating behavior at time t. The behavioral indices were (a) total percent hard (TOTAL), (b) USG percent hard (USG HARD), (c) GOS percent hard (GOS HARD), (d) the difference between USG percent hard and GOS percent hard (DIFF HARD), and (e) the variety

167

Base Rights Negotiation TABLE 11.4 Correlations Among Indices of Negotiating Behavior

 

CRISIS CRISIS VARIETY USG GOS TOTAL DIFF (t +1) (t) HARD HARD HARD HARD

CRISIS (t+1) CRISIS (t) .06 VARIETY .05 USG HARD -.37* GOS HARD .27 TOTAL HARD -.07 DIFF HARD .45*

.30 .30 .29 .64** -.17

-.41* .32 -.19 -.45*

.59** .64** .13

.25 .12

.04

-

*p .05). Figure 14.2 shows that the difference in rating ally and enemy

.22 External Decision-Maker

.20

Central Decision-Maker

.18 Decision-Maker for Force

.16

Aspiring Central Decision-Maker

.14

MEAN EVALUATION

.12 .10 -.08 -.06 -.04 -.02 -.00 -.02 -.04 -.06 -.08 -.10 -.12

OWN NATION

ALLY

ENEMY

NATION OF MEMBER BEING RATED

FIGURE 14.2

Mean Ratings by Each Role of Members of Own Nation, Allies, and Enemies.

TABLE 14.3 Means for Roles Rating Members of Own Nation, Allies, and Enemies*

 

Own Nation

Allies

Enemies

“Head of Government” (CDM) “Foreign Minister” (EDM) “Defense Minister” (DMF) “Aspiring Head of overnment” (CDMa)

-.022 +.212 -.080 -.112

+.012 -.097 +.028 +.056

+.009 -.115 +.050 +.055

* The higher the mean in a negative direction, the more favorable the evaluation.

Ethnocentrism

243

members is very slight and the order of the roles, in terms of mean favorability, is the same for both “targets.” However, the role ordering is reversed for own group ratings. This pattern supports the major proposition of “ethnocentrism theory” which posits a correlational relationship between ingroup adulation and outgroup derogation (Campbell and LeVine, 1965). An opposing hypothesis, from a “disposition theory” framework, predicts that the more favorable the ingroup sentiment, the more favorable the sentiment toward outgroup members. This suggests the presence of a philanthropic or misanthropic disposition evidenced by consistent sentiment toward both ingroup and outgroup members (e.g., see Sullivan and Adelson, 1954). This hypothesis is not supported. The results indicate that the more favorable the ingroup ratings, the less favorable the outgroup (ally and enemy) ratings. Additional support for this basic proposition was obtained through the correlation between ratings of own nation members and ratings of enemies. A correlation of -.48 (p < .001) indicates that the more favorable the ingroup ratings, the less favorable the enemy ratings. The role with the most international contacts, the EDM, was least ethnocentric, while the lowest status role, the CDMa, was the most ethnocentric rater.10 Application of Duncan’s multiple range test (Edwards, 1960) to the data in Figure 14.2 indicated that the mean difference between each pair of roles is significant except that between the DMF and CDMa enemy ratings. The EDM is least favorable to the ingroup and most favorably disposed toward all outgroups, allies and enemies, followed by the CDM. On the other hand, the CDMa is most favorable toward the ingroup and least favorably disposed toward all outgroups, allies and enemies, preceded by the DMF. The evidence clearly supports the “equal status contact” hypothesis and Pruitt’s (1964–65) interpretation as well as those theorists who claim that lower status roles are most ethnocentric (e.g., Sherif and Sherif, 1953). Since the CDMa was the role most concerned with status mobility, i.e., with replacing the CDM, it is probable that this, rather than compensatory satisfaction or fear, is the best explanation for the findings.

Discussion and Implications Using a disguised measure of evaluative bias, strong support for hypotheses from the ethnocentrism literature was obtained. The results of this study may be considered to have major implications for “ethnocentrism theory” as well as for the Inter-Nation Simulation, both as a representation of international relations and as a viable “site” for testing hypotheses generated from other settings (e.g., small group laboratory, field observations of nonlaboratory processes). The provision of eleven replications of a simulation with identical built-in properties made it possible to explore the central hypothesis of “universal” ethnocentrism. When alliance changes were taken into account, the basic proposition of the concomitance of ingroup-and-ally adulation and outgroup derogation was found to be generalizable across all of the runs and for all nation-raters. Also, the proposed correlational relationship between ingroup adulation and outgroup derogation

244

National Identity

was supported. A methodology and technique designed for exploring hypotheses in the Inter-Nation Simulation enabled this setting to serve as an experimental arbiter between competing predictions. A search of the intergroup relations literature uncovered competing predictions with regard to strong and weak enemies as targets of ethnocentric hostility and with regard to the ethnocentrism of decision-making roles. Correspondences between Inter-Nation Simulation processes or functions and processes in other settings were demonstrated. A three-way comparison between findings from the laboratory, from simulation, and from the field is a strategy which might provide information concerning the convergent validity of the construct of ethnocentrism as well as evidence for the validity of the Inter-Nation Simulation as a simulate of international relations. Such comparisons, using the Inter-Nation Simulation as the focal point, provide information about the generality of those social-psychological processes which are hypothesized to be specific to international systems as well as those processes which are hypothesized to be present in other settings. However, the outputs to be compared must be derived from the application of a rigorous methodology appropriate to the setting in question. This investigation provided information about processes occurring in the Inter-Nation Simulation through the utilization of a systematic methodology. The technique of interviewing informants from randomly selected systems of contiguous cultures throughout the world is a methodological strategy adapted to the study of ethnocentrism in the field setting (see Campbell and LeVine, 1961). Propositions similar to those tested in this study, derived from an amorphous literature on ethnocentrism, are being subjected to test in this setting (see Campbell and LeVine, 1965). It is this sort of systematic endeavor that can provide useful evidence for comparison of simulation processes and processes in the system being simulated; it can also demonstrate the ubiquity of ethnocentric phenomena. Another implication of this investigation concerns the potential use of the Inter-Nation Simulation for exploring hypotheses about general social processes which may not be specific to the system of international relations. For example, the occurrence of alliance shifts in the Inter-Nation Simulation permitted the testing of a prediction derived from the principle of belief congruence (Rokeach and Rothman, 1965) as well as from other experimental findings (e.g., Aronson and Linder, 1965; Singer et al., 1963). The phenomenon of more intense negative sentiment toward renegades and more intense positive sentiment toward converts to “the cause,” as compared to continuing members, was supported. Also, results were presented which were relevant to the general issue of selective bias (Wilson et al., 1965) versus ingroup-outgroup bias across all dimensions (e.g., Osgood, 1961). The own nation-ally-enemy bias was not as strong on “respect” traits or on “potency” traits as it was on the “liking” traits. I shall now discuss brief ly the relevance of these findings for the construct of ethnocentrism, the ability to extrapolate to international systems, and certain methodological features of the setting that make it attractive for exploring hypotheses.

Ethnocentrism

245

To interpret these highly confirmatory results, we must ask whether the results are a function of general laws which operate in real international systems. Implicit in this statement are two issues. First, was the ethnocentrism produced a function of basic laws operating in the simulation? It might be argued that just as alliances, nation strength, and roles were built into the structure, the ethnocentric bias may have also been simulated. In other words, while important features (e.g., ethnocentrism) of international systems may have been produced, it is important to know whether these are there because they have been built into the system or whether they are there because certain basic laws are operating. It is quite probable that participants come into the simulation with their own “theories” of international relations and that these notions play a part in decision-making (e.g., see Waters, 1964). Guetzkow (1959, p. 188) commented that, “because we are using human beings as decision-makers, they bring both their personal characteristics into the model and their own implicit theories of the way in which nations should behave.” Also, Orne’s (1959) notion of “demand characteristics” highlights another potential source for participant behavior insofar as the simulation laboratory resembles an experiment. He has shown that much behavior in experiments can be accounted for by suspicions, on the part of subjects, of the investigator’s hypotheses. Even in well-disguised experiments, the task may elicit cues from which subjects infer the experimenter’s intent. This might be an even more serious problem in a simulation in which subjects are told to play unfamiliar roles. Much behavior in the simulation may thus be largely a function of expectations, on the part of the participant, of how the decision-maker that he is simulating would behave in a similar situation. However, the important question for this research is to what extent the ratings were a function of the subjects’ “theories.” An interpretation of these results in terms of simulated bias implies that the participants were aware of the hypotheses and that the task was sufficiently transparent so that they knew what the investigator was looking for. Neither of these hypotheses was checked. An independent check on whether the participants were aware of the hypotheses could be made by having a group, similar to the naval recruits used in this study, predict differential sentiment from and toward nations and roles. This would provide information concerning the subjects’ own theories. It would not, however, solve the more subtle problem of the effect of the particular task environment on participants’ notions (see Orne, 1959). Successful attempts at demonstrating similar ethnocentrism phenomena in other settings would minimize the likelihood that these results are largely a function of task “demand characteristics.” However, even if the participants were aware of the hypotheses from ethnocentrism theory, an attempt was made to disguise the purpose of the task sufficiently so that any resulting bias would be inadvertent. But this too will have to remain conjectural, since only indirect evidence (e.g., comparison with nation ratings, subtle biases found) was obtained in this set of runs. This might be checked more directly in a future simulation by having the participants state what they believe the purpose of the rating task to be.

246

National Identity

The second issue concerns whether these results have direct implications for real international systems with their own developmental history. That is, to what degree is the structure of the simulation similar to the system being simulated? The features of high emotional involvement and motivations of a real-life quality make simulations particularly attractive as tasks for investigators who are not primarily interested in generalizing to the situation being simulated. However, the primary interest of this research is in the extent to which hypotheses from ethnocentrism theory are characteristic of nations in an international system. From one particular epistemological orientation, Campbell (1967) argues that a “complex simulation is a better base for generalizing to a specific natural situation than a simple experiment if the greater complexity provides greater similarity to the natural situation in question. It seems indubitable that the Inter-Nation Simulation does so.” He goes on to suggest that, once we have obtained striking findings in lower cost simulations that we hope to generalize, we should build more complexity into the simulation or “explore degrees of similarity” in an attempt to see what happens to the “laws” as complexity increases. The InterNation Simulation is so constructed that there is similar variability to “real world” counterparts on some variables (e.g., nation strength, role) while others are held constant. Some investigators have found ethnocentric hostility to be a function of such variables as geographical location (Mitchell, 1959), ideological or value differences (Rokeach, 1960; Williams, 1947), and physical similarity (Triandis, 1961). Controls for these variables were built into the simulation used for this study. Simulated nations did not differ on these variables. Thus the source of the ethnocentrism produced in this setting was in the competitive decision-making process. The effect of other potential sources, or of an increase in the number of sources, on ethnocentrism may be investigated through the use of simulation. It is probable that the model, as a simulation, is unnecessarily limited by the disregard for the above-mentioned components. However, from the standpoint of interpretability of results, this reduction in the number of parameters input gives simulations an advantage over field studies in natural settings. Verba (1964) suggested that, if the parameters that are input are indeed homomorphic, then confirmation of a hypothesis in the simulation may be less ambiguous than in the “real world,” because of tighter control of extraneous variation. The more variables that are built into the setting, the more difficult it is to make them orthogonal to one another. In order to avoid hopelessly confounded results the number of participants and the number of nations would have to be greatly increased for each additional variable input. (An example of a confounding factor in this simulation was that between nation strength and allies.) On the other hand, Verba (1964) notes that disconfirmation is more ambiguous in the laboratory setting because of the paucity of rival hypotheses that are available (the parameters of differences are reduced). Also, a reduced number of variables, or an increased number of controls, incurs the “cost” of not simulating interaction effects which might operate among them in the setting being simulated. Thus, in designing simulations,

Ethnocentrism

247

one must decide on criteria for establishing the proper balance between capturing the complexity of the setting being simulated and insuring an unambiguous interpretation of the results (or reducing potential confounding factors). Perhaps the construction of several sets of simulation runs, differing in terms of which variables are controlled and which are manipulated, is a feasible strategy for approaching theoretical sophistication in the area of international relations. The results of this study demonstrated strong confirmation for ethnocentrism as a function of the parameters of this setting. It seems to be of special significance that the ethnocentrism was produced among “nations” with a reduced number of dimensions of difference and at the level of the perception of individuals. It is probable that an increase in the number of dimensions of difference between “nations” would result in increased ethnocentrism.

Summary After a week of participation in each of eleven runs of an Inter-Nation Simulation, naval recruits, simulating decision-makers in contrived nations, were asked to act as judges in assessing the degree to which certain personality traits characterized each of the participants. The investigator was actually not interested in obtaining information about the personality characteristics of ratees, but wanted instead to assess biasing tendencies of raters who had functioned as decisionmakers in simulated nations. The social structure of the simulation was found to be predictive of respondent attitudes. The major findings were as follows: (1) Ethnocentric bias, evidenced in more favorable ratings of own nation members followed by allies and enemies in that order, was found to be generalizable across all runs and nation-raters for all types of traits. This finding supports the central proposition of ethnocentrism theory, i.e., that ethnocentrism is a universal sociopsychological syndrome. (2) When alliances shifted in the course of some runs, the bias shifted correspondingly. In fact, in support of the principle of belief congruence (Rokeach and Rothman, 1965) and the experimental findings of Aronson and Linder (1965) and Singer et al. (1963), renegade “allies” were rated significantly less favorably than members of continuing enemy nations for all trait combinations, while former enemies that converted to the alliance were rated more favorably than continuing ally members. (3) While the biasing tendency was significant across all traits, it was strongest on “liking” traits. This supports the selective bias hypothesis. (4) Competing predictions in the literature concerning differential sentiment toward strong and weak enemies were resolved. When sentiment was broken down into “liking” and “respect” a significant Trait × Strong vs. Weak Enemy interaction was obtained. Strong enemies were “respected” more than members of weak enemy nations, while weak enemies were “liked” more than strong enemy members.

248

National Identity

(5) The role with the most equal status international contacts, the “Foreign Minister” (EDM), was the least ethnocentric role, rating its ingroup colleagues least favorably while rating outgroup members most favorably. The lowest status role, the “aspiring Head of Government” (CDMa), was most ethnocentric. These findings are in keeping with several theoretical sources. (6) Ratings of own nation, ally, and enemy members by role and the correlation between own nation and enemy ratings indicated support for the major proposition of “ethnocentrism theory” as opposed to a “disposition theory” interpretation. The latter posits a positive relationship between sentiment toward ingroup and outgroup members. The results showed that the more favorable the own nation ratings, the less favorable the outgroup (ally and enemy) ratings.

Notes 1 This study was carried out as part of the Cooperative Cross-Cultural Study of Ethnocentrism sponsored by a grant from the Carnegie Corporation of New York to Northwestern University. (Professors Donald T. Campbell and Robert A. LeVine are the principal investigators.) The study benefited greatly from the guidance and significant contributions of Professor Campbell. Additional support came from Professor Harold Guetzkow’s Simulated International Processes project (ICAF/JWGA/ ARPA/NU Project [Advanced Research Projects Agency] SD260), which financed most of the analysis computations on which this paper is based. Special gratitude is extended to Dr. Wayman Crow and his staff at the Western Behavioral Sciences Institute for providing the setting and resources necessary for this data collection. Special thanks are also extended to Larry Ludwig for aid on the data analyses and to Marjorie Druckman for aid throughout the course of the project. 2 This term was used by Palmore (1962) in a study in which he found a relationship between number of ethnophaulisms used for an outgroup and the social distance at which the group was held. 3 The literature on ethnocentrism in intergroup relations suggests hypotheses which are relevant to several sources of ethnocentrism (e.g., value differences, physical differences, cultural differences, etc.). Hypotheses derived from this literature that can be tested in the setting of the Inter-Nation Simulation probably have most relevance for that source of ethnocentrism inherent in the competitive decision making process. However, for the sake of brevity the source of the hypotheses tested in this study is referred to as “ethnocentrism theory.” More will be said of this in the discussion section. 4 My thanks go to John R. Raser for pointing out the nature of these structural changes. 5 In the context of this investigation, this term does not imply random error or artifact. It is an index of the systematic tendency to overvalue or undervalue a particular ratee or “target,” relative to other raters and “targets,” for any reason. “They are the systematic contributions to the ratings within the context of these raters, traits and ratees” (Guilford, 1954). The methodology of constant errors in rating scales was found to be appropriate to the investigation of inadvertent ethnocentric bias and so was adopted in this study. Also, since the “halo” errors are a function of the matrix of means per se and not the scores on which each mean is based, the phrase raw mean is used instead of the term mean. 6 “Halo” errors were scored directionally. The higher the score in a negative direction, the more the overevaluation; the higher the score in a positive direction, the more the underevaluation of a particular “target.”

Ethnocentrism

249

7 The degrees of freedom for the Runs effect differed for each of these analyses. In order to maintain orthogonality in the final structure analyses, one run was excluded (9 df for the Runs effect). 8 The higher the mean “halo” score in a positive direction, the less favorable the evaluation. 9 It was mentioned earlier that lack of orthogonality prevented analyses with strong vs. weak allies as “targets.” 10 The “halo” errors, rather than the raw means, are depicted in Figure 14.2. Inter-rater comparisons in ethnocentric bias are summarized by the “halo” error (i.e., the difference between ingroup and outgroup ratings for one rater relative to this difference for other raters). Figure 14.2 does not indicate that the EDM is not ethnocentric or is negatively ethnocentric (see Swartz, 1961) but only that it is the least ethnocentric role relative to other role-raters. All roles were ethnocentric. When the raw means, rather than the “halo” errors, were plotted there was a linear relationship between “target” being rated (i.e., own nation, allies, enemies), and sentiment, going from favorable to unfavorable, for each rolerater, with the steepest slope being represented by the CDMa ratings (most ethnocentric role) and the flattest slope being represented by the EDM ratings (least ethnocentric role). The “halo” errors, as depicted in Figure 14.2 and Table 14.3, essentially summarize this comparison of slopes.

References Allport, G. A. The Nature of Prejudice. Garden City, N.Y.: Doubleday, 1958. Aronson, E., and D. Linder. “Gain and Loss of Esteem as Determinants of Interpersonal Attractiveness,” Journal of Experimental Social Psychology, 1 (1965), 156–71. Bass, B., and G. Dunteman. “Bias in the Evaluation of One’s Own Group, Its Allies and Opponents,” Journal of Conflict Resolution, 7, 1 (March 1963), 16–20. Berkowitz, L. Aggression: A Social Psychological Analysis. New York: McGraw-Hill, 1962. Blake, R. R., and Jane S. Mouton. “Competition, Communication, and Conformity.” In I. A. Berg and B. M. Bass (eds.), Conformity and Deviation. New York: Harper, 1961. ———,———. “The Intergroup Dynamics of Win-Lose Conf lict and Problem-Solving Collaboration in Union-Management Relations.” In M. Sherif (ed.), Intergroup Relations and Leadership. New York: Wiley, 1962. Brody, R. A. “Some Systemic Effects of the Spread of Nuclear Weapons Technology: A Study Through Simulation of a Multinuclear Future,” Journal of Conflict Resolution, 1, 4 (Dec. 1963), 663–753. Buchanan, W., and H. Cantril. How Nations See Each Other. Urbana: University of Illinois Press, 1953. Campbell, D. T. “The Principle of Proximal Similarity in the Application of Science.” Unpublished manuscript, Northwestern University, 1967. ———, and R. A. LeVine. “A Proposal for Cooperative Cross-Cultural Research on Ethnocentrism,” Journal of Conflict Resolution, 5, 1 (March 1961), 82–108. ———, ———. “Propositions About Ethnocentrism from Social Science Theories.” A Working Paper of the Cooperative Cross-Cultural Study of Ethnocentrism, Northwestern University, 1965. Catton, W. R., Jr., and S. G. Hong. “The Relation of Apparent Minority Ethnocentrism to Majority Antipathy,” American Sociological Review, 27 (1962), 179–91. Cherryholmes, C. H., and H. Guetzkow. Inter-Nation Simulation Kit. Chicago: Science Research Associates, Inc., 1966. Coser, L. A. The Functions of Social Conflict. Glencoe, Ill.: Free Press, 1956.

250

National Identity

Druckman, D. “Ethnocentric Bias in the Inter-Nation Simulation.” Unpublished master’s thesis, Northwestern University, 1965. Edwards, A. L. Experimental Design in Psychological Research. New York: Holt, Rinehart and Winston, 1960. Emerson, R. M. “Power-Dependence Relations,” American Sociological Review, 27 (1962), 31–41. Ferguson, C. K., and H. H. Kelley. “Significant Factors in Overevaluation of Own Group’s Product,” Journal of Abnormal and Social Psychology, 69: (1964), 223–28: Gough, H. G. “The Adjective Check List as a Personality Assessment Research Technique,” Psychological Reports, 6 (1960), 107–22. Gregor, A. J. “Ethnocentrism Among the Australian Aborigines: Some Preliminary Notes,” Sociological Quarterly, (1963), 162–67. Guetzkow, H. Multiple Loyalties: Theoretical Approach to a Problem in International Organization. Princeton, N. J.: Princeton University Press, 1955. ———. “A Use of Simulation in the Study of Inter-Nation Relations,” Behavioral Science, 4 (1959), 183–91. ———. “The Use of Simulation in the Study of Inter-Nation Relations.” In H. Guetzkow, C. F. Alger, R. A. Brody, R. Noel, and R. C. Snyder, Simulation in International Relations: Developments for Research and Teaching. Englewood Cliffs, N.J.: Prentice-Hall, 1963. ———. “Correspondences Between Some Analyses of World Processes and Simulations of International Relations.” In M. A. Kaplan (ed.), New Approaches to International Relations. New York: St. Martin Press, 1967. ———. C. G. Alger, R. A. Brody, R. Noel, and R. C. Snyder. Simulation in International Relations: Developments for Research and Teaching. Englewood Cliffs, N. J.: PrenticeHall, 1963. Guilford, J. P. Psychometric Methods. New York: McGraw-Hill, 1954. Gumplowicz, L. Der Rassenkampf. Innsbruck: Wagner, 1883. Hamblin, R. L. “The Dynamics of Racial Discrimination,” Social Problems, 10: (1962), 103–21. Harvey, O. J. “An Experimental Investigation of Negative and Positive Relations Between Small Groups through Judgmental Indices,” Sociometry, 19 (1956), 201–209. Hermann, C. F., and Margaret G. Hermann. “An Attempt to Simulate the Outbreak of World War I,” American Political Science Review, 61 (1967), 400–16. Homans, G. C. The Human Group. New York: Harcourt, Brace, 1950. ———. Social Behavior: Its Elementary Forms. New York: Harcourt, Brace and World, 1961. Jennings, H. H. Leadership and Isolation. New York: Longmans Green, 1950. Katz, D., and K. Braly. “Racial Stereotypes of 100 College Students,” Journal of Abnormal and Social Psychology, 28 (1933), 280-90. Lambert, W. E., and O. Klineberg. “A Pilot Study of the Origin and Development of National Stereotypes,” International Social Science Journal, 11 (1959), 221–38. Mitchell, J. C. The Kalela Dance. Manchester, England: Manchester University Press, 1959. Orne, M. T. “On the Social Psychology of the Psychological Experiment: With Particular Reference to Demand Characteristics and Their Implications.” In C. W. Backman and P. F. Secord (eds.), Problems in Social Psychology. New York: McGraw-Hill, 1966. Osgood, C. E. “An Analysis of Cold War Mentality,” Journal of Social Issues, 17 (1961), 12–19. ———, G. J. Suci, and P. H. Tannenbaum. The Measurement of Meaning. Urbana: University of Illinois Press, 1957. Palmore, E. B. “Ethnophaulisms and Ethnocentrism,” American Journal of Sociology, 67 (1962), 442–45.

Ethnocentrism

251

Pruitt, D. G. Problem Solving in the Department of State. Denver: University of Denver, 1964–65. Raser, J, R., and W. J, Crow. “An Inter-Nation Simulation Study of Deterrence Postures Embodying Capacity to Delay Response.” Contract N123(60530)35639A, Project Michelson, United States Naval Ordnance Test Station, China Lake, California, 1964.

Reigrotski, F., and N. Anderson. “National Stereotypes and Foreign Contacts,” Public Opinion Quarterly (1959–1960), 515–28. Rieken, H. W., and G. C. Homans. “Psychological Aspects of Social Structure.” In G. Lindzey (ed.), Handbook of Social Psychology. Reading, Mass.: Addison-Wesley, 1954. Robinson, C., and W. Wilson. “Intergroup Attitudes and Strategies in Non Zero Sum Dilemma Games. II. Selective Bias in Both Authoritarians and Non-Authoritarians.” Unpublished manuscript, University of Hawaii, 1965. Rokeach, M. The Open and Closed Mind. New York: Basic Books, 1960. ———, and G. Rothman. “The Principle of Belief Congruence and the Congruity Principle as Models of Cognitive Interaction,” Psychological Review, 72 (1965), 128–42. Rosenblatt, P. C. “Origins and Effects of Group Ethnocentrism and Nationalism,” Journal of Conflict Resolution, 8, 2 ( June 1964), 131–46. Sherif, M., and Carolyn W. Sherif. Groups in Harmony and Tension. New York: Harper, 1953. Sherif, M., O. J. Harvey, B. J. White, W. R. Hood, and Carolyn W. Sherif. Intergroup Conflict and Cooperation: The Robbers Cave Experiment. Norman, Okla.: University Book Exchange, 1961. Simpson, G. E., and H. M. Yinger. Racial and Cultural Minorities: An Analysis of Prejudice and Discrimination. New York: Harper, 1953. Singer, J. E., Lenore S. Radloff, and D. M. Wark. “Renegades, Heretics, and Changes in Sentiment,” Sociometry, 26 (1963), 178-89. Sullivan, P. L., and J, Adelson. “Ethnocentrism and Misanthropy,” Journal of Abnormal and Social Psychology, 49 (1954), 246–50. Sumner, W. G. Folkways. Boston: Ginn, 1906. Swartz, M. J. “Negative Ethnocentrism,” Journal of Conflict Resolution, 5, 1 (March 1961), 75–81. Thibaut, J. W., and H. H. Kelley. The Social Psychology of Groups. New York: Wiley, 1959. Triandis, H. C. “A Note on Rokeach’s Theory of Prejudice,” Journal of Abnormal and Social Psychology, 62 (1961), 184–86. Verba, S. “Simulation, Reality and Theory in International Relations,” World Politics, 16 (1964), 490–519. Waters, M. “Concepts and Methodology in the Teaching of International Relations: A Review,” Journal of Conflict Resolution, 8, 3 (Sept. 1964), 306–13. Williams, R. M., Jr. The Reduction of Intergroup Tensions. New York: Social Science Research Council, 1947. Wilson, W., N. Chun, and M. Kayatani. “Projection, Attraction, and Strategy in Intergroup Competition,” Journal of Personality and Social Psychology, 2 (1965), 432–35.

15 NATIONALISM AND PATRIOTISM

The literature on nationalism and its synonyms developed over the next quarter of a century. The time was ripe in the early 1990s to take stock of what we learned. The stock-taking is discussed in the next article. Here the question of a universal bias is approached from a different angle. I ask about whether loyalties to small groups transfer upward to large collectivities such as nations. The answers are yes with regard to the basic phenomenon but no with respect to the way these sentiments spread through a population and their consequences. This means that the root causes of various types of identities are similar: small group identities are the basis for collective identities. However, the way that identities are used to mobilize populations for collective action is context-dependent. As well, there is a difference in complexity as we move up the ladder from membership groups to a national citizenry. In addition to the differences between faceto-face interactions and the imagined communities of nations, national identity is understood in relation to changing societal norms, support for particular government policies, and electoral/representation dynamics. This larger context is discussed in the article and updated in an essay prepared twelve years later for an Anniversary issue of International Negotiation ( Druckman 2006a). As of this writing we are witnessing a devastating invasion by Russia in Ukraine. News reporting applauds the valiant Ukrainian citizen-fighters for standing down the Russian onslaught. The apparent commitment to defend is in no small part due to a durable national identity. Many of these citizens are willing to pay the ultimate price to defend their country. The trade-offs being considered are conceding the eastern part of their homeland and declaring neutrality in return for stopping the war. Little negotiation progress has been made in securing these trades. The intensity of national identities in this case has reduced cognitive complexity to an “us vs. them” perception. The causes and consequences of these stereotypical images are also discussed in the article. DOI: 10.4324/9781003293361-27

Nationalism and Patriotism

253

Nationalism, Patriotism, and Group Loyalty: A Social Psychological Perspective, Daniel Druckman. Mershon International Studies Review 38 (Supplement 1) (1994): 43–68. © Oxford University Press. Reprinted with permission.

NATIONALISM, PATRIOTISM, AND GROUP LOYALTY A Social Psychological Perspective Daniel Druckman

At a time when ethnic nationalism seems insurgent and capable of pushing much of the world into chaos and war, there is increased need both to understand and to learn how to cope with the conditions that promote such extreme group loyalty. While each of the social sciences has something to say about nationalism, social psychologists have, over the years, contributed, in often neglected ways, to our knowledge about the roots of nationalism. Specifically, they have explored the factors that arouse feelings of group loyalty when such group loyalty promotes hostility toward other groups; how cross-cutting or multiple loyalties can change the face of nationalism; and how individual group loyalties inf luence and shape collective behavior. It is the purpose of this article to discuss this literature and show its relevance to what is happening in the postCold War world. Focusing their attention primarily on individuals and small interacting groups, social psychologists have sought basic knowledge about the ways in which people relate to groups and nations. Central to this focus is the role played by feelings of loyalty to groups and the conditions that arouse or reduce attachments. While relying largely on data from laboratory experiments and surveys of college students, the results are relevant to a wide variety of situations and populations. Whether or not the findings have such broad implications depends on the conditions under which we can reasonably draw conclusions from them about the behavior of national aggregates. It may be that the phenomena do, in fact, aggregate directly from the individual to the collective much as votes can be aggregated. Or, we may be able to make a strong analogy between the behavior of individuals and small groups, on the one hand, and that of leaders, social movements, and whole national populations on the other. We will use both approaches and consider some of the implications of such issues as we review the research.

Nationalism and Patriotism

255

Although granting that nationalism is a political, economic, and sociological phenomenon, it becomes a social-psychological phenomenon to the extent that individuals develop attitudes about their own and other nations. Such attitudes ref lect the feelings that persons have toward these objects and their sense of loyalty to them. These feelings of attachment are at the heart of nationalism. In the review to follow, we will focus on how these feelings of attachment are formed and take root. In particular, we will explore how such loyalty on the part of individuals: (1) can lead to hostile reactions toward other groups and become translated into stereotypes that are shared across individuals, (2) can shape the collective behavior of groups, and (3) can help differentiate among the multiple groups that define any political environment.

National Loyalty and Inter-Group Conflict Even the casual observer of the international scene today is haunted by the impact that loyalties have on what people are doing politically. Strong, even extreme, attachments to ethnic, religious, national, and clan identities have appeared to push individuals and groups to engage in what often seem to be inhumane and improbable acts toward those perceived to be the enemy. Where do such feelings come from? Why do they arise? Why do feelings of loyalty to one group generate negative or hostile feelings toward other groups? In order to understand how individual loyalties can affect inter-group conf lict, we must first consider the importance of group loyalty to the individuals in the group. The feelings of attachment that comprise loyalty for many are not whimsical but are generally basic to the individuals’ definitions of themselves. Loyalty to a group strengthens one’s identity and sense of belonging. Let us explore what social psychology has to say about the bases for loyalty.

The Bases of Group Loyalty The bases for group and national loyalty are widely assumed to be lodged in human needs: “Groups in general are organized to meet human needs, their structures and processes are in part molded by these needs” (Guetzkow, 1957:47). At the level of the nation, the group fulfills economic, sociocultural, and political needs, giving individuals a sense of security, a feeling of belonging, and prestige. While these needs are regarded as universal, their strength appears to vary in different nations and in different individuals (see Terhune, 1964; DeLamater et al., 1969). These needs are not limited to national identifications but have been found to be the basis for group identification in general. “The ways by which an individual relates to his nation have aspects in common with the ways an individual relates to any group of which he is a member” (Terhune, 1964:258). The underlying needs for attachment take several forms. In general these needs tend to arise out of the affective and instrumental functions that nations

256

National Identity

serve for their citizens. As Terhune (1964) has observed, the nation achieves personal relevance for individuals when they become sentimentally attached to the homeland (affectively involved), motivated to help their country (goal oriented), and gain a sense of identity and self-esteem through their national identification (ego involved). DeLamater et al. (1969) added to this triad a normative involvement which occurs when individuals internalize the norms and role expectations of the nation. Parallel concepts are found in the literatures on individual motivation and small groups. Much of social behavior appears to be motivated by the need for affiliation (affective involvement), the need for achievement (goal involvement), and the need for power (ego involvement) (McClelland, 1975; Winter, 1973). Groups function because they are attractive to members (affective involvement), accomplish things and solve problems (goal involvement), and provide status for members (ego involvement) (Bass, 1981; Stogdill, 1974). Kelman (1988) has recently proposed that these needs form the basis for the individual’s need for self-protection and self-transcendence. People see the nation as providing them and their progeny with security and safety as well as status and prestige in return for their loyalty and commitment. Although these needs, in general, characterize the bases for nationalism, individuals often combine and emphasize them in different ways. Terhune (1964) found a group of non-American students more focused on nations as problem solvers while American students perceived nations more as symbols and objects for sentimental attachment. DeLamater, et al. (1969) reported stronger affective attachment to nations by people who saw the nation as a symbol for what was important in their lives. These same individuals were more traditional in their nationalism and more negative toward all supra-national organizations. Theories of cognitive development such as that of Piaget (1965) suggest that children typically move from a focus on themselves to identifying with those who are important to them in their surroundings. Thus, building attachments to groups is part of the normal socialization process as individuals move toward adulthood. It is the way we learn to understand and function in the world around us. There is a progression from being egocentric to increasingly sociocentric (Gladstone, 1962). As a result, as children grow older, they become less focused on themselves per se and more focused on themselves as part of a larger social setting. These theories highlight the importance of self-definition and emotional identification with objects in early learning. Gradually, individuals develop a sensitivity to the needs and interests of others. This movement from self- to other-orientation is evidenced in increased prosocial behavior that includes helping others, showing sympathy for them, and even altruism (Wispe, 1972). Such forms of social behavior have adaptive advantages. It has been argued from the perspective of evolutionary psychology that cooperative behavior promotes individual survival (Corporeal, et al., 1989), and that groups composed of members who are cooperative are more effective than those with members who are less cooperative (Brewer and Kramer, 1986). Such behavior also contributes to a person’s sense of identity by distinguishing them from those

Nationalism and Patriotism

257

who are like them and those who are not, between friends and foes (Volkan, 1988). The cooperative behavior displayed between members of one’s own group, strengthened by pressures of conformity to group norms, is rarely seen in relations between members of different groups. It is in this sense, as Ross (1991:177) notes, that “sociality promotes ethnocentric conf lict, furnishing the critical building block for in-group amity and out-group hostility.” As individuals move from a self- to other-orientation, they also begin to distinguish among the others, becoming more attached and sympathetic to some and more critical and detached from others. The groups they belong to through birth or through early experience have an impact on which they deem to be ingroups and which outgroups. Do these feelings, which developed in children for groups in their immediate environment, transfer to the larger system in which those groups are embedded? In other words, is loyalty similar if the group consists of interacting individuals that one can actually talk with or involves abstract entities like the nation? Feelings expressed toward small groups such as the family or clan need not precede those expressed toward such larger entities as nations and may, in fact, be contradictory. Much of social psychological experimentation is guided by the assumption that similar processes can be invoked simultaneously; that is, in the present case, national loyalties do not depend on the prior development of communal loyalties. But Etzioni (1968) has called this assumption into question, proposing that loyalties transfer from smaller to larger entities. He argues that the feelings expressed toward small groups or communities form the basis for those expressed toward nations. Such a perspective requires an upward transfer of loyalties from smaller units to the center of the system of which they are a part. This transfer may be facilitated by gradually enlarging the group that the individual perceives himself to represent and building in accountability to that bigger group. The construction of the Egyptian pyramids as a case of nation-building exemplifies Etzioni’s point. During the Nile’s f lood season workers from throughout the country were brought to work on the pyramids. Working in teams composed of neighbors and family members, they competed in bringing in the biggest and best stones. While working on the pyramids, however, they were rewarded as a larger unit for cooperating with teams from other villages in putting the stones together to create the pyramids. These men had loyalty originally to those in their village, but transferred that loyalty to the larger community by participating in a process whereby they enlarged their definition of the group to include all countrymen.

From In-Group Loyalty to Inter-Group Conflict Our discussion to this point suggests the intrinsic importance of group loyalty to individuals. But it begs some basic questions. Do national feelings once developed necessarily lead to negative feelings toward other groups? And can feelings of loyalty, once they emerge, be changed or broadened? Two strands of social psychological studies are relevant here. The first explores the differences

258

National Identity

between nationalism and patriotism, suggesting that not all loyalty is associated with hostile feelings toward others. The second shows how endemic in-group bias is to the human species and suggests that loyalty toward one’s own group leads to the denigration of other groups.

Nationalism and Patriotism Feshbach and his colleagues (Feshbach, 1987, 1990; Kosterman and Feshbach, 1989) have wrestled with the relationship between positive feelings toward one’s own group and negative feelings toward others. They wondered if negative feelings toward other groups was an automatic consequence of loyalty to one’s own group. Or are there different types of attachments to one’s own group that may or may not include consideration of other groups? In their research, a factor analysis of responses to items in a questionnaire about attitudes toward one’s own and other countries revealed two relatively independent factors. One focused on feelings about one’s own country. Strong loadings were obtained for such items as “I love my country,” “I am proud to be an American,” and “In a sense, I am emotionally attached to my country and emotionally affected by its actions.” This factor was labeled “patriotism” by Feshbach and his associates. The second factor involved feelings of national superiority and a need for national power and dominance. Strong loadings here were found for such items as “In view of America’s moral and material superiority, it is only right that we should have the biggest say in deciding UN policy”, and “Other countries should try to make their governments as much like ours as possible.” This factor was called “nationalism.” Correlations between these factors and such variables as early familial attachments, attitudes toward nuclear arms, and readiness to go to war suggest distinct patterns. Nationalists indicated stronger support for nuclear armament policies and were ready to go to war but less willing to risk their lives than patriots. Patriots showed a stronger early attachment to their father than did nationalists. The pattern of relationships between nationalism and hawkish attitudes suggested by the data did not, however, include individual aggressive activities as part of the “syndrome.” Feshbach found that his respondents’ own reported aggressiveness was only weakly related to both nationalism and to attitudes toward war. While supporting more aggressive postures for their nation, nationalists were not more aggressive in their personal activities than were patriots. The patriots indicated a willingness to subordinate personal interests to national interests but were not particularly supportive of war. Both sets of findings suggest that nationalism is associated more with a competitive or militaristic approach to the world, patriotism with a more cooperative or peaceful approach to the world. The strategies that each would advocate appear to differ. In effect, the nationalists would constitute the hawks, the patriots the doves in any debate over policy.

Nationalism and Patriotism

259

This distinction between nationalism and patriotism suggests that attitudes toward militarism or peace are rooted in basic underlying dispositions which may prove difficult to change. Studies of children’s development of attitudes, reviewed by Sears (1969), indicate that nationalism, at least in the U.S., arises first as highly favorable affect without supporting cognitive content. Children experience positive feelings about their country before they can put a label on such feelings. When content becomes attached to the feelings, it is as a rationalization for the feelings which, in turn, may linger even after other cognitive components of the attitudes have changed. The nature of the rationalization leads to the differentiation between nationalism and patriotism. Are the feelings justified in terms of being superior to others or because one’s nation is good? Since the feelings develop earlier than the content, they are likely to be more resistant to extinction. It is these feelings, ref lected in the distinction between patriotism and nationalism, that render debates between hawks and doves so vociferous and difficult to resolve. Attempts to mediate the differences are made difficult by the deep-rooted needs served by the attitudes. Policy consensus is more likely to be the result of compromises in positions than changes in the underlying feelings. The findings and recommendations made by Feshbach and his colleagues are consistent with the argument that we should be able to distinguish among kinds of ingroup orientations and identifications. Certain types of ingroup orientations are associated with a tendency to denigrate outgroups, while others are not (Berry, 1984). This distinction is also made in the well-known work of Adorno, et al. (1950) on the authoritarian personality. They note a difference between a healthy patriotic love of one’s own country, not associated with prejudice against outgroups, and an ethnocentric patriotism (like Feshbach’s nationalism) which was associated with such prejudice. More recently, Duckitt (1989) indicated that ethnocentric patriotism was associated with insecure group identifications, and patriotism was related to secure group identifications. The more secure individuals felt in the groups to which they belonged, the more healthy their relationship to the group and the lower their need for distancing their group from others. This set of studies calls into question the overall relationship between ingroup and outgroup attitudes, that is, the notion that attitudes toward ingroups explain attitudes toward outgroups. But Feshbach’s interpretation is only one among a number of theories that can be applied to these data. Another interpretation is that nationalism is merely a more complex form of patriotism. Thus, patriotism is commitment – a readiness to sacrifice for the nation – while nationalism is commitment plus exclusion of others, a readiness to sacrifice bolstered by hostility toward others. Patriotism is the simpler relationship between the individual and group, in this case, the nation. Nationalism requires a more elaborate matrix embedding one’s own nation into a set of nations and differentiating among those nations. In this interpretation, patriotism would be the orientation acquired earlier in the socialization process and, as a consequence, the stronger feeling.

260

National Identity

Moreover, Feshbach’s view assumes that these orientations are stable attitudinal dispositions which are difficult to change. Conceivably, however, the orientations are not stable across situations but are aroused under certain well-defined conditions. For example, the patriotic orientation may occur more frequently in non-competitive situations while the nationalistic attitudes are expressed more strongly in competitive intergroup situations. There is little need for persons to consider outgroups in situations that are not competitive whereas they loom large when the situation becomes more competitive. At issue for the current discussion is whether it is possible to arouse nationalistic attitudes in cooperative situations? In other words, must the situation already be conf lictual before nationalism becomes a force or do nationalistic feelings overlay the situation and lead people to be more prone to seeing situations involving certain others as laden with conf lict and competition? The research literature on ingroup bias provides us with some answers to this question.

Ingroup Bias Results from a wide variety of experiments leave little doubt that the mere classification of people into groups evokes biases in favor of one’s own group. Just by being told that one belongs to a particular group as opposed to another  – even if one has never seen or met any other members of that group – is enough to make the individual prefer the group over others. This group is perceived as better, friendlier, more competent, and stronger than other groups. And classification in these experiments has been on such things as whether one is an over- or under-estimator of certain patterns of dots or favors one type of painter over another. Reviews by Brewer (1979), Tajfel (1982), Brewer and Kramer (1985), and, most recently, by Messick and Mackie (1989), document the extent of this bias toward one’s own group. Moreover, this ingroup bias is found even under conditions of cooperative interdependence when no group gains an advantage by such a bias and often risks losing as a result (Brewer and Silver, 1978). This research challenges the theory that favoring one’s own group over other groups is aroused by the task or situation and is more prevalent when groups are in competition with one another or in conf lict as suggested by early field and laboratory experiments (Blake and Mouton, 1962; Bass and Dunteman, 1963; Ferguson and Kelley, 1964; Sherif and Sherif, 1965; Druckman, 1968b). The more recent studies (and an earlier one by Hatt, 1950) suggest that competition and conf lict are not necessary conditions for ingroup bias, although such situations may enhance such bias. Membership in a group appears to lead people to favor that group and see others as less worthy in comparison. Tajfel (1981, 1982) has advanced “social identity theory” to explain ingroup bias. According to this theory, people’s self-evaluations are shaped in part by their group memberships so that viewing their group in positive terms enhances their self-esteem. An individual’s self-esteem is further increased by making a

Nationalism and Patriotism

261

favorable comparison between his or her own and another group. Not only are they part of a “good” group, but it is “better” than another group. The person’s social identity is tied to the importance of the groups to which he or she belongs. In effect, nationalism links individuals’ self-esteem to the esteem in which the nation is held. Loyalty and identification with the nation become tied to one’s own sense of self. If this theory is accurate, we should see an increase in self-esteem with being able to discriminate one’s own group from others. Oakes and Turner (1980), and Lemyre and Smith (1985), indeed, did find support for this hypothesized relationship. When people were divided into groups, they showed higher self-esteem scores when they were given the opportunity to make allocation decisions favoring their group. These researchers, however, reported that the mere act of dividing individuals into groups without allowing for discrimination among the groups led to lower self-esteem. Being able to discriminate one’s own group from others appeared most important to enhancing self-esteem and loyalty. The discrimination process, in effect, provided a rationale for why one belongs to a particular group and made one feel good about oneself. Another theory proposed to explain ingroup bias, initially advocated by Turner (1987), is referred to as self-categorization theory. This theory, in contrast to social identity theory, places greater emphasis on the nature of the categorization process that occurs when people identify with groups. Self-categorization theory contains two elements: perceptual accentuation and positive regard for the ingroup. Perceptual accentuation is the process whereby objects in the same category appear more similar to one another than to objects in different categories. Nations that have freely elected governments seem more similar to one another than they do to nations whose leaders came to power through coups or involve hand-picked successors. This theory posits that this perceptual bias leads to an evaluative preference for those individuals and groups that are similar to oneself. Without the categorizations that distinguish people in terms of similarities and differences, the evaluative bias would not be possible. Brown and Adams (1986) found that people strive to find differences between their group and others even when they appear highly similar. Individuals in groups that appeared similar in status and attitudes differentiated among themselves by evaluating the other group’s performance more negatively. “We should not assume that just because two groups enjoy friendly or cooperative relations that they will necessarily not seek ways to derogate each other by making ingroup-favoring-judgments on other subjectively important dimensions of intergroup comparison” (Brown and Adams, 1986: 89; see also Tesser and Campbell, 1980). Insko and his colleagues (1988) have posed a theory to explain ingroup bias and the increased competition and hostility that appear to follow. They found that when members of a group reached consensus on the strategy (goals, priorities) they were going to follow, these groups became more hostile and competitive toward other groups. A consensual strategy within the group may shape the perceptual lens through which other groups are discriminated and

262

National Identity

evaluated. Druckman (1968a) also found that when members of groups agreed on the relative importance of the issues under discussion in a prenegotiation session, the group became more competitive. The consensus facilitates group members working in tandem in defining their environment and the other groups that can hinder or help them. Such consensus establishes the parameters that are important to discriminating and evaluating groups. These theories of ingroup bias show how group membership becomes entangled with the way individuals perceive themselves in relation to their world. Ingroup bias helps individuals organize their world and place themselves in that world. In turn, such bias enhances their feelings about themselves and those in their group. Membership in a clan, religious group, or ethnic group, becomes part of the individual’s self identity and critical to a sense of self-worth. The self is threatened by information that calls into question the groups to which one belongs. People learn to react based on their loyalties; they defend those groups that are important to their definition of who they are. Moreover, these loyalties differentiate whom in their environment is appropriate to support and whom to avoid. And such loyalties can foster a consensus among members that becomes self-fulfilling and difficult to change. The stronger the loyalty, the more likely members of a group are to hold similar views and endorse similar strategies. They approach the world in lockstep, perceiving and defining others in the world similarly. There is little, if any, chance for discrepant information to filter through or for reasons to change to be considered.

From Individual to Collective Images This discussion of nationalism, patriotism, and ingroup bias suggests that loyalty not only has feelings associated with it but also images of what one’s own and other groups are like. In other words, there is a cognitive as well as an affective component to loyalty. Images provide individuals with maps of the groups in their environment on which to act. But these image-derived maps only really become critical to politics when they are held by larger collectivities and help define the world for those larger collectivities. When these individual images become shared within a group, they become stereotypes. Stereotypes represent widespread agreement among members of a particular group about the nature of a specific image. We are moving here to consider how individual loyalty becomes translated into a more collective phenomenon that can inf luence what groups of people do. Let us explore some of the research that has a bearing on how these collective images come about.

Descriptive and Evaluative Aspects of Images In their study of children’s views of foreign peoples, Lambert and Klineberg (1967) tried to distinguish between two aspects of images: the content versus

Nationalism and Patriotism

263

evaluative emphases of images, and the diversity in terms used in describing images. These researchers were interested in separating considerations of content (references to the nature of the political or economic system, the nation’s demography, its physical and social attributes, its value orientations, for example) from evaluative concerns (descriptions of the people as kind, naive, talkative, arrogant, competitive). They found that children from a national group with a positive or friendly orientation toward another nation tended to describe that other country with a wider variety of content-oriented terms while using a minimum number of evaluative terms. There was more agreement among the children on how to evaluate the other people than on how to describe them. In contrast, if the children from the same national group had a negative or unfriendly orientation toward another nation, they were in agreement about how to describe the other country while at the same time exhibiting a proliferation of evaluative references. In other words, the children agreed on how to describe the other nation but not on how to evaluate it. Druckman, Ali, and Bagur (1974) report a similar set of findings in studying stereotyping in three cultures: India, Argentina, and the United States. In each culture, there was more stereotyping in terms of the descriptive categories used for other nations that were considered unfriendly and about which little was known than when the other countries were viewed as friendly and a lot was known about them. In other words, the people in these cultures were more likely to stereotype disliked (and less familiar) nations than those that were liked (and familiar). However, just the reverse was found for how people evaluated other countries. There was more stereotyping in the evaluative terms used to talk about liked than disliked nations and familiar than unfamiliar. (See also Druckman and Ludwig, 1970.) In effect, cognitive factors appear to underlie judgments based on familiarity with another country. The more familiar the nation, the less likely people are to describe it in a stereotyped manner. Motivational factors, on the other hand, seem to underlie judgments based on friendliness and evaluative stereotypes. Friendly nations were evaluated in a more stereotypic fashion than unfriendly countries. The mechanisms responsible for these findings differ. The lack of agreement on appropriate categories to use in describing nations with which one is unfamiliar is likely due to a lack of knowledge. The lack of agreement obtained for unfriendly nations on evaluative categories is probably the result of a lack of desire to become informed. In both cases, information is important to the explanation. The cognitive explanation focuses on the possession of knowledge while the motivational explanation emphasizes the desire to seek any information at all about the other nation. What appears to happen is that people only seek information about nations they like. If they are willing to gain information and can do so, the other country becomes more familiar. There appears to be a motivational screen in place that limits people to wanting to learn more about the countries they like rather than those they dislike.

264

National Identity

This screen makes changing negative stereotypes difficult as there is little desire to gain information about disliked nations and, thus, the tendency to keep them at a distance (unfamiliar) and easy to describe as well as classify. One wonders what impact the global media, such as CNN, will have on these relationships as it exposes people of most nations to information they are not actively seeking. It may have the effect of making nations and peoples familiar who were unfamiliar even if they are disliked. At issue is whether the motivational screen will merely block out information on disliked nations or selectively filter the information to reinforce the stereotype and, in turn, the negative evaluation of the group.

Interaction Between Descriptive and Evaluative Aspects of Images This distinction between motivational and cognitive aspects of stereotypes is not meant to imply a single factor explanation for the findings on the extent of agreement about images of one’s own and other nations. Any stereotype or image is the result of an interplay among cognitive and affective factors. Cottam (1977, 1987; see also Herrmann, 1985, 1988) has illustrated this interplay in his theoretical work on the relationship between motives and the content of images. Governments and peoples, he argues, develop their images of other nations based on their perceptions of the threat or opportunity the other nation poses for them, their perceptions of the relative power distance between their country and the other nation, and perceptions of the differences between their culture and that of the other nation. Cottam’s description of the “barbarian” and “degenerate” images demonstrates how governments and people use these perceptions to form images. The “barbarian” image results from the perception that the other nation is threatening, superior to one’s own country in terms of military and economic capabilities, but inferior in terms of culture. People holding the image describe the other nation in diabolical terms similar to the Reagan administration’s view of the Soviet Union as an “evil empire.” The content of their stereotype includes: (1) perceiving the other nation as a simple, single-minded aggressive enemy, (2) that has a monolithic decision structure, (3) and a decision style oriented toward elaborate conspiracies, (4) as well as a determination that the enemy’s advantage in capabilities is due to one’s own inability to use one’s capabilities to optimal advantage in opposing the “morally inferior enemy,” (5) with any citizens in one’s own nation who disagree with this portrayal being “at best dupes, at worst traitors.” The other nation is viewed as dangerous, a view that tempers hasty aggressive actions against them. The contrasting “degenerate” image results from a perception that the other nation offers one’s own country an opportunity coupled with a view that the target nation is comparable in strength yet vulnerable culturally. The content of this stereotype consists of the perceptions: (1) that the other nation is uncertain, confused, and inconsistent, (2) has a highly diffuse and uncoordinated decision structure, (3) as well as a decision style without a strategic framework for action,

Nationalism and Patriotism

265

(4) and a lack of will that prevents effective use of capabilities, (5) with any citizens in one’s own nation who disagree with this view being “effete and weak,” much like those in the target nation. The other nation is viewed as being “ripe for the picking,” a view that encourages and/or rationalizes aggressive actions against them. These stereotypes can be inferred from the statements and policies articulated by a nation’s elite. Statements that embody these images are simplistic and potentially dangerous stereotypes of other nations. They ref lect motivations that lead one to perceive a threatening enemy or that prompt one to be opportunistic and exploitative in one’s orientation toward the “enemy.” Statements that do not embody these images are more likely to ref lect a complex view of other nations – as neither consistently threatening or “ripe for the picking.” Unlike much of the literature on national stereotypes, Cottam relates the images to strategies that governments and leaders can adopt. He proposes that a containment strategy is appropriate to counter an opportunistic aggressor’s “degenerate” image; a detente strategy is useful in countering the enemy stereotype of nations that perceive one as threatening. Cottam’s theory, however, does not take into account the apparent rapid swings in feelings that can occur as events and leadership change. We have observed how rapidly the Chinese became “good guys” after Richard Nixon went to China, how fast Saddam Hussein became “a devil” after his invasion of Kuwait, and how feelings expressed by Americans toward the U.S. military changed dramatically following the war in the Persian Gulf.

Images Embedded in Schemas Cottam’s images illustrate the sort of stereotyped thinking psychologists consider simplistic and undifferentiated (Holt, 1984; Tetlock, 1985). His theory can be compared with one posited by Silverstein and Holt (1989) that focuses on images of the enemy as part of schemas or theories of war that people have and maintain in order to account for their world and the nations and groups they perceive to define that world. These researchers posit three theories of war that people use to organize their images of the enemy: (1) the folk or Rambo theory, (2) the realpolitik theory, and (3) the scientific or systems theory. The folk theory posits that politics is a contest between good and evil, that the world is composed of “dichotomous certainty without ambiguity.” Thus, evil is the work of one’s enemies. The folk theory, in effect, contains a demonic image of the enemy which includes within it both of Cottam’s images – the “barbarian” and “degenerate.” This theory is closest to the traditional meaning of stereotypes as oversimplified images of one’s own and other nations. When the peoples of two nations both hold this demonic view of the other, researchers have talked about them as holding “mirror images” of one another (Bronfenbrenner, 1961). Such polarized thinking on the part of both parties generally reduces interest

266

National Identity

in interaction, increases perceptions of threat, leads to distrust, and creates a win-lose orientation in attempts at conf lict resolution (see Druckman, 1977a; Druckman et al., 1988). In effect, as Osgood (1962), Milburn (1972), Mueller (1973) and others have shown, people lapse into simplified stereotypes of the enemy. Deutsch (1983) has shown how this folk theory of war can precipitate a malignant social conf lict that becomes a self-fulfilling prophecy as it spirals out of control, each side perceiving only the worst of the other. Holsti (1972) and Stuart and Starr (1982) have shown how some American presidents and their advisors have built an inherent bad faith image of the Soviet Union growing out of this folk theory. They perceived that all behaviors of the Soviet Union were motivated by expansionary intentions and hostile feelings. The realpolitik theory of war is based on the notion of power as the focus of politics-power that is defined on the basis of self interests as translated into national interests (Morgenthau and Thompson, 1985; Hoffmann, 1985). In an anarchic world, people and governments must seek to preserve themselves by adding to their strengths. Cooperation with others is only viable if it is done on one’s own terms and is assured to benefit one’s own nation. One’s own power is taken into account when defining self or national interests and in weighing the consequences of alternative political actions. Power here is relative vis-avis other peoples and nations. Other nations and peoples are defined in terms of their power and what they can do or provide to enhance one’s own power and, thus, ensure that one can achieve and maintain what is in one’s own self or national interest. As Hoffmann (1985: 134) has observed, governments with this theory guiding their perceptions and behavior start with a representation of their ends and “then calculate the amount of power needed to reach them, distinguish among the very different sorts of power appropriate to different ends, and decide, if necessary, how to increase one’s power of the kinds required for the absolutely indispensable ends.” Whereas in the folk theory of war, the enemy was the focus of attention; in the realpolitik theory, one’s own nation is the focus of attention. In the folk theory, conf lict is perceived as under the control of the enemy; in realpolitik, conf lict is perceived to be more under one’s own control and as potentially good if one gains one’s ends and more power in the process. Silverstein and Holt (1989) argue that the systems (or scientific) theory of war is a more complex way of organizing images that allows individuals to maintain their attachment to their own nation while using the political context to differentiate among other nations. Other nations can play different roles in the system depending on what is happening at the moment. One’s own actions need to be guided by what the current organization of the system is perceived to be and the problem that is the focus of attention. Other nations do not consistently get placed into a particular category. Rather there is a search for which others should be included to solve this specific problem in this political context. The situation becomes the focal point and others are judged by their positions in relation to the particular situation. This theory makes stereotyped images

Nationalism and Patriotism

267

more difficult unless nations are perceived to assume the same role all the time. There are usually other countries or entities out there who share one’s interests or concerns to whom one can turn for support. Since as the adage goes “your friend one day may be your enemy the next,” giving other nations fixed labels may defeat one’s purposes and make it more, rather than less, difficult to reach one’s goals. The Silverstein and Holt images suggest that stereotypes can focus on different aspects of the environment and attribute control to these different arenas. The folk theory image assigns control to the enemy. One’s stereotype about the enemy becomes all important in determining strategies and tactics as Cottam has proposed. In the realpolitik image, control is retained by the actor who anticipates what others will do given the nature of the political system. Collective stereotypes about one’s own power vis-á-vis others lead to action. In the systems image, the current situation and problem help to determine how one’s own and other nations are viewed and defined. Standard operating procedures as well as lessons from the past concerning particular problems and issues take over as the stereotypes of the moment to guide behavior.

Group Loyalty and Collective Behavior Our discussion has begun to turn to how these stereotyped images of one’s own group and other groups inf luence collective behavior. Of interest is the nature of the relationship between group loyalty and collective action. This question is at the interdisciplinary juncture between psychology, sociology, and political science. Moreover, it crosses the boundaries between levels of analysis, from individuals and small interacting groups to collectivities and nations. In what follows, we will focus on several possible connections between people’s loyalties and what collectivities do. We will consider several instances where the nature of peoples’ loyalties can have an impact on collective behavior at the institutional level: (1) through action as a representative of a group; (2) through support for particular policies; (3) through defining group norms; and (4) through the decision-making process. In each of these cases there is a nexus where images and actions at the individual level can help to shape what happens at the collective level. These effects get played out through the inf luence they have on the expectations of leaders, the shaping of public opinion, or creation of norms. Let us explore each of these points of connection further.

Role as Representative When people are told that they are representatives of a group and are perceived to stand for a group, research suggests their loyalty constrains behavior. Blake and Mouton (1962) began this series of studies by noting the effects of group pressures on the compromising behavior of representatives. In effect, individuals designated as representatives of groups reached fewer agreements

268

National Identity

and were more competitive than individuals acting on their own behalf. These researchers argued that group loyalty served to limit persons known to be representing a group. As representatives, these individuals were relatively inf lexible in their search for solutions to problems shared with members of other groups because they would be accountable for what they knew their group wanted. In particular, representatives were likely to approach intergroup negotiations over conf licts of interest with a “win-lose” rather than problem-solving orientation. Druckman and his associates (Druckman, 1967, 1971a, 1971b, 1977b; Druck man et al., 1972, 1976, 1977, 1988; Solomon and Druckman, 1972; Zechmeister and Druckman, 1973; Benton and Druckman, 1974) explored a set of variab1es that increased the representative’s responsibility for his or her own group’s outcome and found more constraint as responsibility increased. The studies compared individuals negotiating on behalf of a group with those negotiating on behalf of themselves. As the individual’s accountability to the group increased, as the outcome became more important to the group, and as the chances that the group would know what happened increased, the representative felt more and more tied to the goals, norms, and values of the group. In a sense, these individuals perceived themselves more as agents of the group than representatives, with little room to maneuver. Their loyalty was always in question and they could do hardly anything without undergoing scrutiny. They had to prove that the group was important to them through their behavior. In the world of international relations these variables can be manipulated by groups or their leaders to produce the “desired” degree of loyalty. They can also be manipulated by third parties who are asked to mediate conf licts between groups or nations. Note that to prove their allegiance members of terrorist groups are often required to kill, kidnap, or rob one of the enemy (see Crenshaw, 1986). They risk death themselves should they fail to carry out the mission. In effect, these findings draw attention to ways of increasing or decreasing actions taken by individuals on behalf of a group. As the demands for loyalty and commitment increase, the tendency is to defend the group’s position and to gain something for the group in any negotiation. These effects are not only found in the representatives’ behavior but in their perceptions of what is happening. They report stronger loyalty and commitment as the pressures on them mount. Their perceptions and feelings mediate between the demands of the group and their behavior (see Druckman, 1968a; Zechmeister and Druckman, 1973). The perceptions and feelings help to “justify” the behavior.

Support of Particular Policies In discussing ingroup loyalty, we observed that nationalist feelings magnify the propensity for misperception. These feelings inf luence how others are defined

Nationalism and Patriotism

269

and the nature of desirable action. They lead to “bad faith” images (Larson, 1986), “ethnocentric bias” (Druckman, 1968b), or “partisan bias” (Fisher and Brown, 1988) in which others’ actions are evaluated on whether or not they are a friend or an enemy. As a result of this bias, people – leaders included – often overestimate the relative strength of their own group vis-á-vis other groups. Moreover, the same actions are viewed differently if carried out by an ally or an enemy. In effect, a double standard is applied and similar behaviors are perceived as different. While regarded in much of the psychological literature as a “universal” phenomenon, this biasing tendency of loyalty can be enhanced by events that cause tension or are perceived as threatening to one’s view of one’s own nation. These examples of misperceptions – precipitated by exaggerated national self images – can lead to miscalculations of likely outcomes from collective actions. With regard to strategic policy decisions, an overestimation of a nation’s strength compared to that of the enemy can lead to aggressive postures culminating in wars that are lost. When people overvalue a nation’s worth vis-á-vis other countries, such perceptions can restrict leaders’ behavior, forcing them to become committed to possibly misguided courses of action. Lebow (1981) cites the example of American foreign policy toward China in the 1950s to illustrate these effects. The view of China as a weak nation was a self-serving stereotype that contributed to American miscalculations of the risk involved in taking more aggressive actions toward China. These linkages involve attempts to understand policy in terms of broader national self images. A similar phenomenon is found in case studies reported by Stoessinger (1978) on SinoSoviet-American relations, by Brecher (1975) on Israel’s foreign policy, and by Lebow (1976) on British rule in Ireland. Of these latter three studies, only Brecher’s extends the causal sequence to an actual decision to go to war. He makes a strong case, based on interviews with policy makers, for the connection between Israeli self perceptions as “victims” and the decision to go to war in 1967. The Arab threat was seen as yet another attempt to impose a “final solution” on Israel – to drive them into the sea. In effect, national self images and the strong feelings often attached to such images constrain the range of options that policymakers have in dealing with other nations. In some instances they can lead to overly aggressive actions when such are precipitous; in other instances they can prevent action when such may be relevant. These images and feelings set the parameters within which leaders can work. If leaders, themselves, have similar images and feelings, they can enhance and exacerbate what the public is feeling. An illustration of the latter is Slobadan Milosevic of Serbia. He helped to intensify the growing feelings of nationalism within Serbia by launching the movement toward “ethnic cleansing.” Building on the bias already present in the Serbia people, he heightened the contrast with the enemy and clarified the action that needed to be taken. In other words, he gave direction to feelings that were already intense and ready to explode.

270

National Identity

Defining Group Norms Another connection between group loyalty and collective action lies in how group norms are defined. Of interest here is the social psychological literature on group and organizational cohesion. Considered to result from interactions among group members, cohesion is an attribute of groups that may mediate the relationship between the loyalties or attachments of individuals to groups and collective action or performance. Group loyalty is usually regarded as a defining aspect of cohesion. As Zander (1979:433) notes, the most commonly accepted definition of cohesion is “the desire of members to remain as members of a group.” Researchers have observed that group loyalty may lead to cohesion but it also appears to be a result of cohesion as well (Terborg et al., 1976; Landers et al., 1982; Stokes, 1983). There is a reinforcing effect of loyalty and cohesive groups whose members are loyal perform better, leading the group to become more cohesive and the members more loyal. This relationship between loyalty and cohesion can increase to the point that we get the phenomenon Janis (1989; see also, t’Hart, 1990) has called “groupthink.” Here members of the group become excessively protective of the group as an entity, restricting membership to those also loyal to the group, isolating themselves from any information counter to their own image of the group, and viewing threats to their policies as threats to the group. Because of the insular quality of such groups, they can take extreme actions without realizing the impact more information would have had on such behaviors. Moreover, these groups tend to define the outside world in such a way as to engage in self fulfilling prophecies-they perceive others as working against them, so they engage in aggressive behavior against these others, and thus ensure that the others will be hostile in return. Terrorist groups, military juntas, authoritarian leaders with their close advisors often give evidence of groupthink. But it also occurs in governmental cabinets, policy planning staffs, and national security councils. Moskos (1970), Lynn (1984), and Henderson (1985) have commented on how prevalent this phenomenon is in combat troops, particularly those in front-line positions. These ideas have implications for groups, organizations, and nations that are fairly homogeneous. The more alike people are, the easier it is to generate loyalty and cohesion. There are common customs, values, and socialization to build upon. It becomes less difficult to develop a shared image and to define what are standard operating procedures toward certain other groups. Leaders and representatives have a sense of where people stand and can take actions with some assurance of support. Moreover, the population is more easily mobilized by certain cries to arms. As cohesion and loyalty reinforce one another, such groups can become insulated from the rest of the world and appear to be “marching to their own drummer.” Hitler’s Germany, Castro’s Cuba, and Khomeini’s Iran provide some relevant examples of these phenomena.

Nationalism and Patriotism

271

Decision-Making Processes In the above three ways of connecting individual loyalties and collective behavior, the focus is on the impact of the attitudinal climate on collective actions. Loyalties toward a group condition the perceptions and images people have of themselves and other groups and impose certain constraints on the leadership through support for particular policies or through shared stereotypes. Individual and group factors, however, probably have their most direct link to collective action in the policymaking process itself. Since policy decisions are often made by single, powerful individuals or in small, face-to-face groups (Hermann and Hermann, 1989; Hermann, Hermann, and Hagan, 1987), the perceptions and biases induced by loyalty can inf luence both how the decision making is structured and the nature of the process. As an example, consider Ronald Reagan’s image of the Soviet Union as “the evil empire,” and of communism as intent on dominating the world. In his own experience, he had faced an attempted takeover of the Screen Actor’s Guild by a communist leaning faction and knew how dangerous “they” could be (Cannon, 1982). He believed that communists were behind most anti-American activity around the world and, as a consequence, made staffers and members of his administration prove they were not supporting such actions rather than the more usual procedure of proving that they were. He viewed the communists as guilty unless proven innocent (Glad, 1983; Hermann, 1983). They were the enemy, and we know what enemies do. His image and bias made it difficult for people who did not share a similar tendency to remain in the administration and, indeed, led to persons who were like him to become part of the administration (Bennis, 1973). We noted above the influence of group loyalty on the creation of groupthink, which can affect decision making in small, interacting political groups. It is also important to note the impact of the images and self-serving stereotypes that individuals bring to such groups. Are the missiles the Soviet Union placed in Cuba in 1963 perceived as offensive or defensive? They are more likely to be judged offensive if the Soviet Union is seen as an enemy. Are Japanese-Americans perceived by American policymakers as “one of us” or as untrustworthy? They are generally seen as “just one of us” unless Japan has launched a war against us, at which time they become potential enemies too. Are the Muslim fundamentalists seen as more or less of a threat by moderate Arab regimes? After Khomeini came to power in Iran and called such fundamentalists to action, they were seen as a growing threat – they had a sponsor who had been successful in his own revolution. Loyalties and attachments inf luence and help to shape the lens through which leaders and policymakers interpret their world. The resulting images narrow the range of options that are likely to be considered and provide rationalizations for actions.

Modifying Group Loyalty In previous sections we have described the inf luence that group loyalty can have on perceptions of one’s own group and other groups and how such perceptions

272

National Identity

shape stereotypes and collective action. Much of the discussion has focused on the way loyalty increases the potential for intergroup conf lict and self-fulfilling prophecies regarding the correctness of one’s own behavior and the lack of good faith on the part of others. It seems appropriate to ask if there is any hope for the resolution of the clan rivalries in Somalia or the ethnic conf licts in the former Yugoslavia given the portrait we have painted here. Similarly, is a European community feasible? What does the social psychological perspective have to say about our ability to modify, create, or extend loyalties to new, larger, or more inclusive groups? Exploring the literature on multiple loyalties can give us some perspective on this question. Most of the studies reviewed to this point have concentrated on assessing group loyalty in terms of two groups – an ingroup and outgroup. They have engaged individuals in distinguishing between the group to which they belong, and another to which they do not belong. In effect, the ingroups have been their membership group and the outgroups have been non-membership groups. Largely ignored in this research is how people deal with multiple group loyalties and the attendant outgroups such loyalties invoke. Overlapping and crosscutting group memberships are part of the landscape of intergroup relations, particularly within and between nations. When faced with multiple groups with which one can identify and become committed to, how do people choose among them and how does such choice inf luence their images and collective action? Does the presence of multiple groups dampen the impact of loyalty or condition the nature of its inf luence?

The Scaling Hypothesis One proposal advanced to explain how people cope with the multiple groups in their environment is called the “scaling hypothesis.” According to this hypothesis, people rate the various groups along a continuum from positive to negative. They also rate such groups in terms of the intensity – from weak to strong – with which they hold these feelings. Of interest are the characteristics of groups that determine where they are placed on these dimensions and the impact of the ratings, and concomitant feelings, on behavior. Four studies speak to this so-called “scaling hypothesis,” and the inf luence of this intuitive scaling on the actions of citizens. Using an inter-nation simulation, Druckman (1968b) studied the effects of nationalism on the individuals playing the roles of policymakers for a set of nations. As expected, the findings showed that these decision makers overvalued members of their own government and undervalued members of other governments. There was a distinct bias and loyalty toward the ingroup as compared to the outgroups. But each rating was inf luenced by the relationships among the “nations” in the system. Outgroups were differentially undervalued depending on whether they were allies or enemies, and on how conf lictual or cooperative their behavior was in general. Moreover, members of one’s own government

Nationalism and Patriotism

273

differentially over- or undervalued other governments based on the roles they played vis-á-vis other governments .  .  . whether they favored or pushed for another government’s position, like a foreign minister, or were they highly loyal to their own consensual government position, like a head of state. Strong support was found for “ethnocentrism theory” in the ratings. The governments were arrayed along a continuum with members of one’s own government being rated more positively when compared to one’s allies and enemies being viewed as least similar and most negative. Among allies, “renegade allies” were considered more dissimilar and more negative than continuing enemies and former “enemies” converted to members of the alliance. Individual members of governments were able to make subtle distinctions among the various other governments based on the perceived nature of the relationships they had with such governments and their recent behavior toward them. There was also some elasticity to the concept of ingroup to include those other governments that were currently friendly and cooperative. At least such governments were perceived as closer to one’s own government than all others. Singer, Radloff, and Wark (1963) examined the roles of heretic and renegade in exploring feelings toward the members of one’s own group and other groups. These investigators reported that renegades were liked even less than members of other groups who were, in turn, liked less than members of one’s own group. Feelings expressed toward the heretic role, however, were more positive than those expressed toward renegades, and only somewhat less positive than those directed toward regular ingroup members. Because heretics never disavow their membership in the group, their deviation can apparently be tolerated better than that of renegades, who renounce their membership and the group. Disagreements with heretics are usually over the preferred actions to take rather than over the group’s central values which the renegades are questioning. This study suggests that members of groups not only differentiate among other groups but among roles played by members of their own group. Generally, members value loyalty to the group and favorable attachment to the group. Thus, individuals who question or challenge the group’s core values are considered deviant and not loyal. They are no longer good members and are assigned roles that befit their expressions of dissent. The feelings of other group members toward them indicate their marginality. Fishbein (1963) elaborated on the above findings by exploring differences in feelings toward non-group members based on their interest in joining the group (positive or negative), their eligibility to join the group (eligible or ineligible), and their current membership status (continuing non-group member or ex-group members). He was interested in which non-members were perceived as most threatening as well as least liked. The findings showed that candidates for membership who were eligible and interested were viewed as least threatening and most favorable. The potential candidate for membership, however, who was not interested in the group and, indeed, saw the group in a negative light, was perceived by the group as most threatening and least favorable. Ex-group

274

National Identity

members who viewed the group negatively were seen by members of the group as more threatening than continuing non-members (members of outgroups). In both instances the more negative the orientation toward the group, the more concerned group members became about what the non-member could do to undermine group confidence and loyalty. Parallels to organizations like the Palestinian Liberation Organization come to mind, especially the attempts of its leadership to constrain the more militant members when negotiations with the Israelis and other Arab states seem imminent. The PLO leadership perceives that what they want to do can be limited by what others see in these nonconforming members; the actions of these extremist members become viewed by the rest of the world as representative of the position of the PLO. Brewer (1968) built on the above studies, examining the criteria on which people base their feelings about various types of outgroups. What was it about these groups and individuals that triggered negative or positive feelings? In gathering data, she examined the 30 African tribes in Kenya, Uganda, and Tanzania. She was interested in the kinds of variables that helped determine when the various tribes would be willing to work with, to have as a neighbor, to share a meal with, and to become related to the other tribes. The more a tribe was interested in engaging in these behaviors with members of another tribe, the closer the social distance between the two tribes was perceived to be. Among the variables she associated with social distance were perceptions of similarity, physical or geographic proximity, and educational and economic advancement. Members of the tribes were willing to be socially closer to other tribes they perceived to be more similar to them, geographically adjacent, and more advanced. Judged similarity had the strongest impact, followed in order by geographic proximity, and advancement. There was an interaction between similarity and advancement such that being educationally and economically advanced was more important to judgments of social distance if the tribe was considered dissimilar rather than similar. In effect, perceived similarity appears to be the most important dimension for scaling other groups. “We only want to interact with groups like us.” Visible and successful outgroups are also worthy of attention and emulation unless they are perceived to be different from one’s own group. If different, they may have a bad influence on the members of the ingroup and force change; thus, they are to be avoided. Together, these studies reveal some dimensions along which multiple groups are perceived and regarded in the decision making of ingroups. Feelings expressed in favor of one’s own group members or against another group’s members depend on how each is perceived in relation to one’s own group. Toward some groups – those that are perceived to be allies, similar to one’s own group, advanced, or having persons eligible and interested in membership in one’s own group – there are more favorable leanings and a willingness to interact on a regular basis. However, toward other groups – those that are perceived to be enemies or contain renegades from one’s own group, dissimilar and “backward” or distant, as well as having individuals not eligible or disinterested in membership in one’s own group – there is a sense of threat and vigilance and a wariness about

Nationalism and Patriotism

275

interaction unless absolutely necessary. When, and if, interaction occurs with these latter groups, it should be on one’s own terms.

Reference Groups These studies do not, however, help us answer the question about how people decide among potential ingroups. If one is born into a clan with a particular religious orientation in a specific region of the country, to which of these groups does loyalty develop? Or, does the individual build loyalty to all or integrate them in forming his or her identity? At issue is when do other groups become ingroups and when do they become outgroups? Under what conditions do people see a group as one to be emulated or, at the least, positively regarded? And under what conditions do they perceive another group as threatening and, thus, to be hostilely regarded as an outgroup? Merton (1957) among others (Kelley, 1952; Hyman, 1960) proposed reference group theory as a possible means of answering these questions. In reference group theory, nationalism does not necessitate that other nations be viewed with hostility. Having negative feelings toward another nation may be merely a special case of nationalism because intense loyalty to one nation does not necessarily have to lead to hostility toward another. The distinction between patriotism and nationalism made earlier is relevant here. It should be possible to have pride, patriotism, in one’s nation while still recognizing its shortcomings and being willing to cooperate with and, perhaps, even include other nations in one’s ingroup. Having positive feelings toward groups is a necessary, but not sufficient, condition for designating such groups as reference groups. Positive feelings for another group or nation does not automatically indicate identification or loyalty to that group or nation. More is required for it to become a reference group, such as adopting its values or aspiring to membership or citizenship. How do we know when a person has moved from positive feelings toward another nation to identification with that other nation? Varying degrees of identification with groups or nations can be scaled in terms of the way one relates to the group. The following is an example of such a scale: (1) Motivated toward becoming a member (2) Assuming the group’s norms and values (3) Using the group’s standards for evaluating performance --------------------------------------------------------------------------(4) Taking a positive orientation toward the group (5) Understanding the group’s norms and values (6) Recognizing the group’s existence Another group becomes a reference group as one moves from step four to step three. The closer to step one, the greater the identification the individual has with the reference group. To the extent that the reference group meets the

276

National Identity

individual’s needs or enhances his or her self-esteem more than, or at least in the same way as, current membership groups, negative feelings are reduced. An example of this process can be seen in the transference of loyalties to the European Community from its member nations. The European Community probably has reference group properties for many Europeans. When it begins to fulfill the needs of individuals in its various member countries better than their national governments, loyalty toward one entity may be decreased and increased toward the other. An extreme form of identification with groups of which one is not a member is depicted by Swartz’s (1961) concept of negative ethnocentrism or by Kent and Burnight’s (1951) term, xenocentrism. The former involves attributing one’s own values to another group and at the same time evaluating that group as being more successful than one’s own group in being able to achieve them. There is a bias not for one’s own group but for the other group. The latter concept refers to using another group as the center of everything, and all others, including one’s own group, are scaled and rated in reference to it. Recent studies in Latin America (see Montero, 1986) suggest that xenocentrism – or altercentrism as they refer to the phenomenon – is characteristic of countries that have a dependent relation with a powerful country. Researchers (Quieros de Ramos, 1979; Salazar, 1983) have found that individuals in several Latin American countries evaluated themselves in relation to the United States, undervaluing themselves and overvaluing Americans. “People in countries such as Venezuela, who engage in self-deprecation, tend to exhibit a United States-dependent ideology – the United States being the country with the highest evaluation” (Montero, 1986: 421). Dependency is not only externally imposed but internally it reproduces the norms and values of the center of power. The people in the dependent country begin to “compare themselves to those in the centers of power and, in turn, to devalue themselves” (Montero, 1986: 421; see also Salazar, 1983). Altercentrism leads to a negative national identity because the reference group is outside one’s own country. These researchers argue that this negative self-identity can be inhibiting as people deny their possibilities and blame themselves for their condition, always looking outside themselves for the defining characteristics of what is good and important. The studies we have just reviewed suggest that where people place their loyalty depends on how they perceive and array the various groups in their environment. The first cut at differentiating among these groups appears to depend on whether one has positive or negative feelings toward the group. Groups toward whom one has positive feelings may become reference groups – that is, potential ingroups; those toward whom one has negative feelings become outgroups and potential targets for bolstering positive feelings toward one’s ingroup. Which groups an individual will build a strong identification with and, in turn, develop loyalty to become those perceived to have similar norms, values, and customs to one’s own preferences, those that inculcate what one holds dear or desires. Contact with other groups can result in positive feelings if the groups are viewed as similar to one’s own. Contact with groups seen as dissimilar from one’s own

Nationalism and Patriotism

277

will lead to negative feelings unless members of those groups are perceived as advanced or more powerful than one’s own group. With a more powerful group that is perceived as dissimilar, individuals may refocus their attention and begin to devalue their own group, holding the more powerful group up as a model for emulation-particularly if the more powerful group can inf luence what happens to one’s own group. The research on the scaling hypothesis and reference groups gives us some hints about how we might modify group loyalty. Getting people to develop positive feelings toward another group appears to facilitate paying attention to that group and a willingness to learn about the group. Positive feelings seem more likely to occur toward groups that are viewed as similar, as friends, as advanced or having status, and as interested in one’s own group. Contacts, information, and image-building need to center around interactions that indicate how the other group fits one of these criteria for being viewed positively. Contacts, for example, should be between high status or, at least, equal status, representatives of the two groups with some interest or curiosity about the other group. The purpose of such contacts would be to help members of the group begin to perceive outgroups as reference groups and, thus, potential ingroups, by seeing how these outgroups are similar to them. The scaling hypothesis would suggest that such movement will be more feasible the closer the outgroup is to the ingroup to start with. It is unlikely that members of a group will perceive an outgroup they place very far away from them and have strong negative feelings toward as worthy of interaction. It is interesting to conjecture what happens when people have multiple reference groups which, in turn, each have a set of outgroups. Say an individual was involved with an ethnic group, religious group, and nation as his or her reference groups. If the reference groups all espoused the same outgroups using similar criteria, we might hypothesize that the reinforcing qualities of the groups would increase the individual’s hostility toward those outgroups and facilitate strong loyalty to all three ingroups. There is little contradiction among what is seen as good and desirable and what bad and undesirable. But what if the reference groups differ in whom they see as outgroups and in the criteria they use for defining themselves vis-á-vis others? How do individuals resolve these contradictions? Do they focus on the most proximate group, the ethnic or religious group, and follow their dictates; or, do they create an amalgam of the groups, picking and choosing what they will use to define their ingroup and, in turn, outgroups? The above studies raise these questions but provide us with few answers.

Conclusions The purpose of this essay was to introduce the reader to a social psychological perspective on the roots of nationalism. At its heart is the description of how individuals develop feelings about and attachments to groups – how they build loyalty to groups. The nation is viewed here as one type of group that

278

National Identity

fosters loyalty. What have we learned in this process and what questions form the research agenda for the future? Studies suggest that people see groups as providing them with security and safety as well as status and prestige in return for their loyalty and commitment. Nations, in particular, achieve personal relevance for individuals when they become sentimentally attached to the homeland (affectively involved), motivated to help their country (goal-oriented), and gain a sense of identity and self esteem through their national identification (ego involved). This sense of loyalty builds through the socialization process, as people become less focused on themselves and learn to take into account the needs and interests of others. And they begin to distinguish among these others, becoming more attached and sympathetic to some and more critical and detached from others. The feelings transfer across the socialization process from smaller to ever larger groups as people perceive themselves to represent and be accountable to these larger entities. The bias toward one’s own group that comes with loyalty appears critical to defining who one is and occurs in both cooperative and competitive intergroup situations. In fact, people have been shown to favor their own group even when they could lose substantially in the process. This ingroup bias, however, generally is more extreme in competitive situations where there is incentive to favor one’s own group. It also appears to increase when there is consensus among group members about their goals and strategies – not only are they loyal to the group but they all define the problem, what they want to do, and how they want to do it in a similar manner. There is little chance for discrepant information to filter into the group or for them to consider reasons for change. Loyalty includes both emotional and cognitive aspects that interrelate to form images of groups. When these images are shared among members of a group, they become stereotypes. The emotional aspects of an image indicate whether there is any desire to seek information about another group based on one’s evaluation of the other as friend or foe; the cognitive aspects denote the information one possesses about the other group. As a result, stereotypes become a means for maintaining one’s images because there is little desire to gain information about disliked nations and, thus, the tendency to keep them at a distance, easy to describe and classify. Stereotyped images can be very simplistic with a focus on a dichotomous world made up of us and them, or they can become more differentiated based on the situation in which people find themselves at the moment. Moreover, other groups can be viewed in static, never-changing terms or in very dynamic terms based on the political context. Stereotyped images inf luence how political entities act. They do so by defining the climate in which action takes place – public opinion, norms, values – and by affecting the decision-making process. In each instance, these images limit and constrain the options leaders can choose. Much of the time individuals define their ingroups and outgroups in the context of a set of groups. In this process, they seem to scale the groups intuitively according to their feelings and their perceptions of the status, preferences, norms, values, and power of the groups. Those that are seen in more positive

Nationalism and Patriotism

279

terms and as more similar may become reference groups; those that are seen in more negative terms and as more dissimilar become outgroups. Some scholars have raised the question of whether or not feelings toward ingroups can be decoupled from those toward outgroups. There appears to be some distinctions that can be made between “patriotism” and “nationalism” that facilitate such a decoupling. Patriotism seems to lead to strong attachments and loyalty to one’s own group without the corresponding hostility toward other groups while nationalism encourages an orientation involving liking for one’s own group and disliking of certain other groups. While those strong in patriotism are willing to risk their lives for their country, they are not as prone to war as those strong in nationalism who have an enemy built into their attachments to their nation. As the above paragraphs attest, the existing social psychological literature can contribute in important ways to our understanding of the foundations of national loyalty and its relation to intergroup conf lict. Moreover, the literature suggests insights into the challenges of transforming and modifying such loyalty. The extant research, however, also triggers a great many questions begging for further exploration. The following is but a sample: (a) Assuming that group loyalty is developed in small familial or community groups, how do these early attachments transfer to such larger units as ethnic groups, nations, or regions of the world? What happens in the socialization process that leads individuals to expand what they are loyal to and how do they link early and late attachments? (b) Under what kinds of conditions is it possible to limit loyalty to positive feelings toward one’s own nation without generating negative feelings toward other nations? In other words, how can we decouple ingroup amity from outgroup enmity? How do we build patriotism instead of nationalism? (c) Taken together, the experiments on the inf luence of loyalty on ingroup bias show support for the phenomenon under a wide variety of circumstances. There is much less consensus on why this occurs. Why do individuals’ self-esteem and identity become entangled with loyalty? Are there ways of being loyal without placing one’s sense of self on the line? (d) Much is still to be learned about the roles played by cognitive and emotional factors in stereotyped images. Of particular interest is whether loyalty, and its accompanying biasing tendencies, hinder the development of complex images of other groups and their policies. Does loyalty predispose individuals to screen out information and selectively impose a certain vision on the environment? (e) What is the impact of multiple loyalties on how people act? Do multiple loyalties keep group conf lict in check, or do they merely reinforce a particular view of the world, exacerbating conf lict? These issues fall at the juncture between disciplines and cry for interdisciplinary exploration. The challenge for future research is to bring diverse perspectives and methodologies to bear on the topic. The current piece is but a start in that direction.

280

National Identity

References Adorno, Theodore W., Elise Frenkel-Brunswik, Daniel J. Levinson, and R.N. Sanford. (1950) The Authoritarian Personality. New York: Harper and Row. Bass, Bernard M. (1981) Stogdill’s Handbook of Leadership. New York: Free Press. Bass, Bernard M., and George Dunteman. (1963) Bias in the Evaluation of One’s Own Group, Its Allies and Opponents. Journal of Conflict Resolution. 7:16–20. Bennis, Warren. (1973) The Doppelganger Effect. Newsweek, September 17:13. Benton, Alan A., and Daniel Druckman. (1974) Constituent’s Bargaining Orientation and Intergroup Negotiations. Journal of Applied Social Psychology. 4:141–150. Berry, Jack W. (1984) Multicultural Policy in Canada: A Social Psychological Analysis. Canadian Journal of Behavioral Science. 16:353–370. Blake, Robert R., and Jane S. Mouton. (1962) “The Intergroup Dynamics of Win-Lose Conf lict and Problem-Solving Collaboration in Union-Management Relations.” In Intergroup Relations and Leadership, edited by Muzafer Sherif. New York: Wiley. Brecher, Michael. (1975) Decisions in Israel’s Foreign Policy. New Haven: Yale University Press. Brewer, Marilynn B. (1979) In-Group Bias in the Minimal Intergroup Situation: A Cognitive Motivational Analysis. Psychological Bulletin. 86:307–324. Brewer, Marilynn B. (1968) Determinants of Social Distance among East African Tribal Groups. Journal of Personality and Social Psychology. 10:279–289. Brewer, Marilyn B., and Roderick M. Kramer. (1986) Choice Behavior in Social Dilemmas: Effects of Social Identity, Group Size, and Decision Framing. Journal of Personality and Social Psychology. 50:543–549. Brewer, Marilynn B., and Roderick M. Kramer. (1985) The Psychology of Intergroup Attitudes and Behavior. Annual Review of Psychology. 36:219–243. Brewer, Marilynn B., and Madelyn Silver. (1978) Ingroup Bias as a Function of Task Characteristics. European Journal of Social Psychology. 8:393–400. Bronfenbrenner, Urie. (1961) The Mirror Image in Soviet-American Relations: A Social Psychologist’s Report. Journal of Social Issues. 17:45–56. Brown, Rupert, and Dominic Adams. (1986) The Effects of Intergroup Similarity and Goal Interdependence on Intergroup Attitudes and Task Performance. Journal of Experimental Social Psychology. 22:78–92. Cannon, Lou. (1982) Reagan. New York: G. P. Putnam’s Sons. Caporael, Linda R., Robyn M. Dawes, John M. Orbell, and Alphons J. Van De Kagt. (1989) Selfishness Examined: Cooperation in the Absence of Egoistic Incentives. Behavior and Brain Sciences. 12:683–739. Cottam, Richard W. (1987) State Motivation and Political Realism. Paper presented at the Mershon Center, Ohio State University, October 8, 1987. Cottam, Richard W. (1977) Foreign Policy Motivations: A General Theory and a Case Study. Pittsburgh: University of Pittsburgh Press. Crenshaw, Martha. (1986) The Psychology of Political Terrorism. In Political Psychology, edited by Margaret G. Hermann. San Francisco: Jossey-Bass. Deutsch, Morton. (1983) The Prevention of World War III: A Psychological Perspective. Political Psychology. 4:3–31. DeLamater, John, Daniel Katz, and Herbert C. Kelman. (1969) On the Nature of National Involvement: A Preliminary Study. Journal of Conflict Resolution. 13:320–357. Druckman, Daniel, Ed. (1977a) Negotiations: Social-Psychological Perspectives. Beverly Hills: Sage.

Nationalism and Patriotism

281

Druckman, Daniel. (1977b) The Person, Role, and Situation in International Negotiations. In A Psychological Examination of Political Leaders, edited by Margaret G. Hermann. New York: The Free Press. Druckman, Daniel. (1971a) On the Effects of Group Representation. Journal of Personality and Social Psychology. 18:273–274. Druckman, Daniel. (1971b) The Inf luence of the Situation in Interparty Conf lict. Journal of Conflict Resolution. 15:523–554. Druckman, Daniel. (1968a) Prenegotiation Experience and Dyadic Conf lict Resolution in a Bargaining Situation. Journal of Experimental Social Psychology. 4:367–383. Druckman, Daniel. (1968b) Ethnocentrism in the Inter-Nation Simulation.Journal of Conflict Resolution. 12:45–68. Druckman, Daniel. (1967) Dogmatism, Prenegotiation Experience, and Simulated Group Representation as Determinants of Dyadic Behavior in a Bargaining Situation. Journal of Personality and Social Psychology. 6:279–290. Druckman, Daniel, and Lawrence D. Ludwig. (1970) Consensus on Evaluative Descriptions of One’s Own Nation, Its Allies, and Its Enemies. Journal of Social Psychology. 81:223–234. Druckman, Daniel, Faisunisa Ali, and J. Susana Bagur. (1974) Determinants of Stereotypy in Three Cultures. International Journal of Psychology. 9:293–302. Druckman, Daniel, Alan A. Benton, Faisunisa Ali, and J. Susana Bagur. (1976) Cultural Differences in Bargaining Behavior: India, Argentina, and the United States. Journal of Conflict Resolution. 20:413–452. Druckman, Daniel, Benjamin J. Broome, and Susan H. Korper. (1988) Value Differences and Conf lict Resolution: Facilitation or Delinking? Journal of Conflict Resolution. 32:489–510. Druckman, Daniel, Richard M. Rozelle, and Kathleen Zechmeister. (1977) “Conf lict of Interest and Value Dissensus: Two Perspectives.” In Negotiations: Social-Psychological Perspectives, edited by Daniel Druckman. Beverly Hills: Sage. Druckman, Daniel, Daniel Solomon, and Kathleen Zechmeister. (1972) Effects of Representational Role Obligations on the Process of Childrens’ Distribution of Resources. Sociometry. 35:387–410. Duckitt, John (1989) Authoritarianism and Group Identification: A New View of an Old Construct. Political Psychology. 10:63–84. Etzioni, Amitai. (1968) “Social-Psychological Aspects of International Relations.” In Handbook of Social Psychology. Volume Five, edited by Gardner Lindzey and Eliot Aronson. Reading, MA: Addison-Wesley. Ferguson, Charles K., and Harold H. Kelley. (1964) Significant Factors in Overevaluation of Own-Group’s Product. Journal of Abnormal and Social Psychology. 69:223–228. Feshbach, Seymour. (1990) Psychology, Human Violence and the Search for Peace: Issues in Science and Social Values. Journal of Social Issues. 46:183–198. Feshbach, Seymour. (1987) Individual Aggression, National Attachment, and the Search for Peace: Psychological Perspectives. Aggressive Behavior. 13:315–325. Fishbein, Martin. (1963) The Perception of Non-Members: A Test of Merton’s Reference Group Theory. Sociometry. 26:271–286. Fisher, Roger, and Scott Brown. (1988) Getting Together. Boston: Houghton Miff lin. Glad, Betty. (1983) Black and White Thinking: Ronald Reagan’s Approach to Foreign Policy. Political Psychology. 4:33–75. Gladstone, Arthur L. (1962) Relationship Orientation and Processes Leading Toward War. Background. 6:13–25.

282

National Identity

Guetzkow, Harold. (1957) Isolation and Collaboration: A Partial Theory of Internation Relations. Journal of Conflict Resolution. 1:46–68. Haas, Ernst. (1986) What Is Nationalism and Why Should We Study It? International Organization. 40:707–746. t’Hart, Paul. (1990) Groupthink in Government: A Study of Small Groups and Policy Failures. The Netherlands: Swets & Zeitlinger, Lisse. Hatt, Paul K. (1950) Occupations and Social Stratification. American Journal of Sociology. 55:533-543. Henderson, W. Darryl. (1985) Cohesion: The Human Element in Combat. Washington, D.C.: National Defense University Press. Hermann, Margaret G. (1983) Assessing Personality at a Distance: A Profile of Ronald Reagan. Mershon Center Quarterly Report. 7:No. 6. Hermann, Margaret G., and Charles F. Hermann. (1989) Who Makes Foreign Policy Decisions and How: An Empirical Inquiry. International Studies Quarterly. 33:361–387. Hermann, Margaret G., Charles F. Hermann, and Joe D. Hagan. (1987) “How Decision Units Shape Foreign Policy Behavior.” In New Directions in the Study of Foreign Policy, edited by Charles F. Hermann, Charles W. Kegley, Jr., and James N. Rosenau. London:Allen and Unwin. Herrmann, Richard. (1985) Perceptions and Behavior in Soviet Foreign Policy. Pittsburgh: University of Pittsburgh Press. Herrmann, Richard. (1988) The Empirical Challenge of the Cognitive Revolution: A Strategy for Drawing Inferences about Perceptions. International Studies Quarterly. 32:175–203. Hoffmann, Stanley. (1985) Realism and Its Discontents. The Atlantic, November:131–136. Holsti, Ole R. (1972) Crisis, Escalation, and War. Montreal: McGill-Queen’s University Press. Holt, Robert R. (1984) Can Psychology Meet Einstein’s Challenge? Political Psychology. 5:307–313. Hyman, Herbert H. (1960) Ref lections on Reference Groups. Public Opinion Quarterly. 24:383–396. Insko, Chester A., Rick H. Hoyle, Robin L. Pinkley, Gvi-Young Hong, and R. M. Slim. (1988) Individual-Group Discontinuity: The Role of a Consensus Rule. Journal of Experimental Social Psychology. 24:505–519. Janis, Irving. (1989) Crucial Decisions: Leadership in Policymaking and Crisis Management. New York: Free Press. Kelley, Harold H. (1952) “Two Functions of Reference Groups.” In Readings in Social Psychology, edited by G. E. Swanson, T. M. Newcomb, and E. C. Hartley. New York: Holt. Kelman, Herbert C. (1988) “Nationalism, National Sentiments, and International Conf lict.” Paper prepared for the Workshop on Nationalism and International Conf lict, National Academy of Sciences, Washington, D.C. Kent, Donald P., and Robert G. Burnight. (1951) Group Centrism in Complex Societies. American Journal of Sociology. 57: 256–259. Kosterman, Rick and Seymour Feshbach. (1989) Toward a Measure of Patriotic and Nationalistic Attitudes. Political Psychology. 10:257–274. Lambert, Wallace E. and Otto Klineberg. (1967) Children’s Views of Foreign Peoples: A Cross-National Study. New York: Appleton-Century-Crofts. Landers, Daniel M., M. O. Wilkinson, B. D. Hatfield, and H. Barber. (1982) Causality and the Cohesion-Performance Relationship. Journal of Sport Psychology. 4:170–183. Larson, Deborah W. (1986) “Game Theory and the Psychology of Reciprocity.” Paper presented at the American Political Science Association meeting, Washington, D.C.

Nationalism and Patriotism

283

Lebow, Richard Ned. (1981) Between Peace and War: The Nature of International Crisis. Baltimore: Johns Hopkins University Press. Lebow, Richard Ned. (1976) White Britain and Black Ireland: The Influence of Stereotypes on Colonial Policy. Philadelphia: Institute for the Study of Human Issues. Lemyre, Louise, and Philip M. Smith. (1985) Intergroup Discrimination and SelfEsteem in the Minimal Group Paradigm. Journal of Personality and Social Psychology. 49:660–670. Lynn, John A. (1984) The Bayonets of the Republic. Urbana, Illinois: University of Illinois Press. McClelland, David C. (1975) Power: The Inner Experience. New York: Irvington. Merton, Robert K. (1957) Social Theory and Social Structure. New York: Free Press. Messick, David M., and Diane M. Mackie. (1989) Intergroup Relations. Annual Review of Psychology. 40:45–81. Milburn, Thomas W. (1972) “The Management of Crisis.” In International Crisis: Insights from Behavioral Research, edited by Charles F. Hermann. New York: Free Press. Montero, Maritza. (1986) “Political Psychology in Latin America.” In Political Psychology, edited by Margaret G. Hermann. San Francisco: Jossey-Bass. Morgenthau, Hans J., and Kenneth W. Thompson. (1985) Politics Among Nations: The Struggle for Power and Peace. New York: Alfred A. Knopf. Moskos, Charles C. (1970) The American Enlisted Man. New York: Russell Sage. Mueller, John E. (1973) War. Presidents. and Public Opinion. New York: Wiley. Oakes, P., and John C. Turner. (1980) Social Categorization and Intergroup Behavior: Does Minimal Intergroup Discrimination Make Social Identity More Positive? European Journal of Social Psychology. 10:295–301. Osgood, Charles E. (1962) An Alternative to War or Surrender. Urbana, Ill.: University of Illinois Press. Piaget, Jean. (1965) The Moral Judgment of the Child. New York: Free Press. Quieros de Ramos, A. (1979) Relation Entre Minusvallia Social y Externalidad [Relationship Between Externality and National Underevaluation]. Caracas: Venezuela Central University, School of Psychology. Ross, Marc H. (1991) The Role of Evolution in Ethnocentric Conf lict and Its Management. Journal of Social Issues. 47:167–185. Salazar, Jose M. (1983) Bases Psicologicas del Nacionalismo [Psychological Bases of Nationalism]. Mexico: Trillas. Sears, David O. (1969) “Political Behavior.” In The Handbook of Social Psychology. Volume Five, edited by Gardner Lindzey and Eliot Aronson. Reading, MA: Addison-Wesley. Sherif, Muzafer, and Carolyn W. Sherif. (1965) “Research on Intergroup Relations.” In Perspectives in Social Psychology, edited by Otto Klineberg and Richard Christie. New York: Holt, Rinehart and Winston. Silverstein, Brett, and Robert R. Holt. (1989) Research on Enemy Images: Present Status and Future Prospects. Journal of Social Issues. 45:159–175. Singer, Jerome E., Lenore S. Radloff, and David M. Wark. (1963) Renegades, Heretics, and Changes in Sentiment. Sociometry. 26:178–189. Solomon, Daniel, and Daniel Druckman. (1972) Age, Representatives’ Prior Performance, and the Distribution of Winnings with Teammates. Human Development. 15:244–252. Stoessinger, John G. (1978) Nations in Darkness: China, Russia, and America. 3rd ed. New York: Random House. Stogdill, Ralph M. (1974) Handbook of Leadership Research. New York: Free Press. Stokes, Joseph P. (1983) Components of Group Cohesion: Inter-member Attraction, Instrumental Value, and Risk Taking. Small Group Behavior. 14:163–173.

284

National Identity

Stuart, Douglas, and Harvey Starr. (1982) The ‘Inherent Bad Faith Model’ Reconsidered: Dulles, Kennedy, and Kissinger. Political Psychology. 3:1–33. Swartz, Marc J. (1961) Negative Ethnocentrism. Journal of Conflict Resolution. 5:75–81. Tajfel, Henri. (1982) Social Psychology of Intergroup Relations. Annual Review of Psychology. 33:1–39. Tajfel, Henri. (1981) Human Groups and Social Categories: Studies in Social Psychology. Cambridge: Cambridge University Press. Terborg, James R., Carl Castore, and John A. DeNinno. (1976) A Longitudinal Field Investigation of the Impact of Group Composition on Group Performance and Cohesion. Journal of Personality and Social Psychology. 34:782–790. Terhune, Kenneth W. (1964) Nationalism among Foreign and American Students: An Exploratory Study. Journal of Conflict Resolution. 8:256–270. Tesser, Abraham, and Jennifer Campbell. (1980) Self-Definition: The Impact of Relative Performance and Similarity of Others. Social Psychology Quarterly. 43:341–347. Tetlock, Philip E. (1985) Integrative Complexity of American and Soviet Foreign Policy Rhetoric: A Time-Series Analysis. Journal of Personality and Social Psychology. 49:1565–1585. Turner, John C. (1987) Rediscovering the Social Group: A Self-Categorization Theory. New York: Basil Blackwell. Volkan, Vamik D. (1988) The Need to Have Enemies and Allies: From Clinical Practice to International Relationships. New York: Jason Aronson. White, Ralph K. (1970) Nobody Wanted War: Misperceptions in Vietnam and Other Wars. Garden City, NY: Doubleday. Winter, David G. (1973) The Power Motive. New York: Free Press. Wispe, Loren G. (1972) Positive Forms of Social Behavior: An Overview. Journal of Social Issues. 28:1–19. Zander, Alvin. (1979) The Psychology of Group Processes. Annual Review of Psychology. 30:417–451. Zechmeister, Kathleen, and Daniel Druckman. (1973) Determinants of Resolving a Conf lict of Interest: A Simulation of Political Decision Making. Journal of Conflict Resolution. 27:63–88.

16 SCALING UP, DOWN, AND ACROSS

I returned to the national identity topic 15 years later with my colleague Steve Wood. As discussed earlier, we picked up and expanded on the levels of analysis theme in the third chapter of this Part. In the paper, prepared exclusively for this book, we review literatures at each of three levels of analysis – interpersonal, intergroup, collective – and then provide examples of how these levels interact as pathways to theory development. The pathways move up, down, and across. But in addition to theoretical contributions made by the chapter, we bring the micro (psychology) and macro (international relations) social science fields into contact. Three insights stand out in this chapter: (a) The foundations for national identity are developed in the earliest attachments within the family, (b) the link between early attachments and national identity is bridged by group or community-level processes, and (c) collective-level attitudes may not be predicted from attitudes expressed by individuals or small groups. Issues about the causes and consequences of national identity are raised but not resolved. We struggle with the distinction between linear pathways from one to another level of analysis and a simultaneous interplay among these levels as illustrated by the EU example. These questions will no doubt keep us busy for a long time to come.  

DOI: 10.4324/9781003293361-28

SCALING UP, DOWN, AND ACROSS Pathways to a Theory of National Identity Daniel Druckman and Steve Wood

Introduction Scholars of national identity and nationalism approach their subject matter largely from disciplinary perspectives and conduct research within the silos of social psychology, sociology, history, anthropology, political science, or international relations. An exception is Malešević’s (2013) treatment, which progresses toward an interdisciplinary analysis. His discussion of intertwined micro-macro relationships attempts to the show how the nation-state emerges from a variety of small group interactions. Malešević contributes to efforts to develop an integrative theory of nationalism. He emphasizes the importance of understanding the phenomenon of nationalism in relational terms, as a blending of an overarching or umbrella level with many local instances. Noting that “the relationship between the micro- and macrouniverse is filled with tension” (2013, p.  14), he provides a platform for interdisciplinary research that cuts across the various psychological, social science, and humanities research silos. Addressing the puzzle of individual-society integration, Jaspal et al. (2016) claim that “there has been little attempt to synthesize the three prime levels of analysis” (2016, p.  2; see also Ashforth, 2016). We perceive an opportunity for progress at these levels of analysis and to advance toward an interdisciplinary theory on national identity within the context of more general group identities. Our aim is to develop pathways to a theoretical framework conducive to understanding the formation and spread of national identities. The idea of pathways refers to directions for constructing a theory. One direction is a linear path from micro through meso to macro level interactions. It is illustrated by the statistical path found in the Druckman and Wagner (2019) analyses of peace agreements: adhering to procedural justice principles in a negotiation produced

Scaling Up, Down, and Across

287

fair agreements that led to communal cooperation that facilitated a durable societal peace. This is an example of an upward path from early micro to later macro interactions. In this chapter we also develop downward (early macro to later micro interactions) and simultaneous paths where inf luences from one level to another occur at the same time. We argue that the three directions are critical in developing a theory of national identity. Issues of forming identities are addressed at the interpersonal level where such processes as mimicry and matching come into play. The research reviewed in this section provides insights into how these processes operate. Issues of spread are treated at a macro level where such processes as public opinion, opinion cascades, and social networking come into play. The section on collective dynamics provides insights into these processes. Smaller groups and organizations can be considered as mediating the micro–macro relationship. Becoming aware of group membership and its consequences for intergroup relations connects identity formation at the interpersonal level to spread at the collective level. Insights from research at this level pave the way toward an understanding of how collective identities emerge. This section is placed between the reviews of research at the micro and macro levels of analysis. The next challenge on the path to a theory is to conceptualize interactions among the three levels. We do this by using examples from several spheres of activity. These include the way that micro and macro identities are connected in conf lict resolution workshops, how we move from the macro to micro levels in military mobilization and combat on the ground, and how simultaneous interactions occur among levels when civil society actors participate in peace processes. A more complex example of multilevel interactions and shifting dynamics is the European Union (EU). These examples suggest pathways to theory that encompass vertical and simultaneous perspectives. A key is the bridging role played by group processes, such as how community cooperation links previous negotiation processes to durable societal peace. The questions raised open new lines of inquiry, including possible disconnects from individual to collective opinions, changes in the strength of identities as processes move from the interpersonal to the collective, and mechanisms at the group level that link the micro and macro levels. The chapter concludes with suggestions for further development of these ideas in concert with empirical research.

Interpersonal Dynamics: Mimicry, Matching, and Synchrony There is little doubt that emotional attachments to others are acquired in infancy (Bowlby, 1988). A feature of these attachments is spontaneous or implicit mimicry, a pattern of copying the behavior and expressions of another person, ref lecting their inf luence. Kavanagh and Winkielman (2016) show that people mimic others they like and avoid copying people they dislike. Individuals preferentially mimic members of their own group, a trend that begins with the family unit. This is a basis for group identity. Research across disciplines has provided insights

288

National Identity

into the way mimicry operates and the functions it serves (Mead, 1934; Zentall, 2006; Whiten et al., 2009). Mimicry is selective. We are more likely to mimic others who share our values and backgrounds (Over and Carpenter, 2012). But it is also subject to change, as children grow and meet others outside the family, as relationships change, or when a liked model reveals exploitative intentions. Thus, observed mimicry may be regarded as a diagnostic of the state of interpersonal relationships. When the mimicry is high in fidelity, the relationship between the copier and the model is likely to be strong (McIntosh, 2006). Individuals mimic outgroup members less or even display the opposite behavior, such as scowls in response to smiles (Carr et al., 2014). These effects are cyclical in the sense of being mutually reinforcing where mimicked behaviors strengthen a relationship or group ties, which in turn increase the fidelity of the mimicking. These cycles also facilitate interpersonal and intra-group coordination. Related concepts are matching, cooperative, and competitive tactical responding, and synchrony, moving together over time. Mimicry is central to both concepts and precedes them in the learning or developmental process. A difference is that these processes are neither spontaneous nor automatic. Rather they are intentional choices made typically in game-like contests that capture elements of negotiation. By building cooperation or instigating competition, matching strategies contribute to creating or reinforcing positive and negative identities. When cooperative moves are reciprocated, as in tit-for-tat gaming strategies, a trusting relationship develops, and a shared identity results. When competitive moves are reciprocated, a distrusting relationship ensues as parties pull apart. A way out of competitive or negative reciprocation spirals is by taking unilateral cooperative moves proposed originally by Osgood (1962) and pursued further by Adelman (1984) and Ramberg (1993). These interpersonal dynamics extend the research on mimicry by documenting the way cooperation evolves in repeated interactions (Axelrod, 1984). Implications for identity derive from the connection between cooperative responding and attraction or liking, which are key emotions for a durable identity to emerge.1 A question of interest is whether compatriots imitate because they are similar or because they share attachment to the same nation? When considered over long periods and a variety of occasions, matching may take the form of interactional synchrony. In an analysis of a negotiation over military base rights, Druckman (1986) showed how delegations matched and mismatched their verbal statements, as hard or soft rhetoric, through a year and a half of talks. Impasses occurred when the parties were out of sync; progress was made when they were in sync. In a study on eight bilateral international negotiations, Druckman and Harris (1990) showed that parties adjusted their moves or language in the direction of reciprocity. In all cases agreements depended largely on discovering a synchronous pattern of responding. More generally, Taylor (2014) showed how synchronous linguistic phases and cycles escalate and de-escalate conflicts. Hove and Risen (2009) showed that interpersonal synchrony increases affiliation or likability. The key was the timing of movements

Scaling Up, Down, and Across

289

between participant and experimenter, indicating that this effect was due to interpersonal synchrony rather than mimicry. One implication is that people infer closeness when they notice synchrony. Another is that “synchronized movements act as cues to a communal sharing relationship” (2009, p.  958). The second implication, noted also by Fiske (2004), draws attention to the sort of micro to macro connections that create and reinforce larger attachments, including national identities. Findings from an experiment conducted by Druckman, Parlamis, and Burns (2022) show that party loyalty constrains the extent to which political representatives are f lexible in negotiations. The negotiators represented their party in Congressional talks about the allocation of funds to four national programs (social security, labor and unemployment, veterans’ benefits, and agriculture). Those representatives whose voting record was party line (higher loyalty) moved significantly less on their preferred issues than those who had a mixed voting record (lower loyalty). They were especially reluctant to move on their most preferred issues. The inf lexibility observed by these experimental negotiators is similar to the strong in-group attachments central to nationalism. These findings suggest a link between the interpersonal and group levels of analysis. Concession dynamics in bilateral negotiations is an example of interpersonal compromise. The representational roles taken by negotiators are understood at the group level where they must balance their constituents’ demands against those made by their opponents. To summarize, there are three important forms of micro-level imitation: mimicry (shown in familial and tribal attachments), matching (building cooperation), and interpersonal synchrony (creating or reinforcing attraction or affiliation). Each of these forms of imitation is a social inf luence process that provides a basis for the spread of group identities. They are also precursors (when scaling up) to meso- and macro-level identities and are essential for theoretical paths to an integrated whole. We now move to the meso level and consider literature and issues associated with group dynamics.

Group Dynamics Few issues have received more attention in the social sciences than how individuals are inf luenced by the groups they belong to or aspire to joining. Although widely regarded as an empirical issue, our understanding of the group-individual nexus is hampered by a lack of scientific consensus about how to define a group. The debate is framed at the extremes as whether the group is “merely” an assembly of its members or is an emergent property separate from its members. The impetus for this debate comes from Durkheim’s (1895) assertion that groups are a separate level of analysis, referred to as social facts, not understood by the psychology of individual behavior. The debate gained steam with the advent of social psychology during the early part of the twentieth century (Allport, 1985). The field was divided into sociological (groups) and psychological (individuals

290

National Identity

in groups) factions. The crux of the issue lies in coming to terms with how to perform empirical research on groups. Building on Homans (1950), Sherif (1966) defined the essential properties of groups as entities with an organization or structure and a set of values or norms. He was keen to point out that a norm is not to be regarded as the statistical average of the behavior or attitudes of its members. It emerges from member interactions and is thus considered to be a sociological designation. Back (1979) described several conceptual difficulties with the group concept, noting that a group is both real or visible and an abstraction or nominalist concept. Herein lies the difficulty in using groups as a unit of analysis for research. As “something new arising from individual components” (Back, 1979, p.  286), the group concept retains an element of elusiveness that confounds empiricists searching for operational definitions. So vexing was this dilemma that social psychology f loundered as a separate academic discipline in the decades following the 1980s. 2 Despite these trends, progress toward developing empirical indicators of groups was being made from earlier work receiving less attention. Campbell (1958) addressed the matter of evaluating the status of social groups as empirical entities. His indices of common fate (moving in the same direction), similarity (boundaries), and proximity (spatial contiguity) strengthened the argument that groups were entities independent of the individuals in them. Drawing on statistical analyses, he developed coefficients for each criterion that provided a basis for hypothesis testing. A decade later Druckman (1968) developed a method for evaluating hypotheses about a system of interacting simulated nations. Recently, Druckman, Mueller and Diehl (2022) used indicators of similarity and proximity to evaluate the compatibility of multiple missions within peacekeeping operations. These three articles moved the debate about the concept of group from an axiomatic, a priori discussion to an evidence-based exchange about methods for evaluating hypotheses about group behavior. They join the conversations about “groupness” stimulated by Brubaker and Cooper (2000), who distinguished between a strong bounded and a weaker f luid sense of connectedness, and Tilly (1978), who combined categories and networks in his concept of “catnet.” A substantial literature has developed on the way that members of a group perceive other groups. Most of these studies, conducted primarily by social psychologists, consisted of analyzing ratings of own and other group members.3 The research demonstrates a robust partisan bias: more favorable rating of ingroup than out-group members. Most impressive perhaps are the minimal group (MGP) studies where laboratory subjects are assigned to ad hoc groups based on such features as hair or eye color. Tasks consisted of deciding on point allocations for solving puzzles or other group products. Ratings almost invariably favored their own temporary group (Messick and Mackie, 1989). In-group bias may be stronger when members are appointed as representatives of their group or party. This is illustrated by the loyalty–f lexibility relationship

Scaling Up, Down, and Across

291

shown by Druckman, Parlamis and Burns (2022). In that study, strong constraints on movement in negotiation were demonstrated for role-playing members of the US Congress. These attachments to political parties, based on emotions or incentives, may encourage the loyalty–f lexibility relationship obtained in the experiment. They also add a political identity dimension to the interpersonal dynamics that occur in bilateral negotiations or in the MGP experiments. Of course, if the existence of groups is temporary, the loyalty associated with them is demonstrated as transitory.4 A variety of explanations have been proposed for the ubiquity of loyaltybased in-group biases. These include, at the level of individuals, enhanced self-esteem (Tajfel, 1981), categorization in the service of cognitive economy (Turner, 1987), uncertainty reduction (Hogg and Mullin, 1999), and reduced threats to one’s identity (Hornsey and Hogg, 2000). Taking social processes within groups into account, Insko et al. (1988) showed that cohesion, fostered by interactions among group members, strengthens identity and corresponding aspersions cast on other groups. Another source for in-group bias may be found in negative partisanship. This literature is updated by the Bohm et al. (2020) review. Abramowitz and Webster (2016) showed that partisanship in US politics derives from negative feelings toward the other party. It may also be the case that strong dislike for other nations strengthens national identities. These various explanations for an inward-looking bias call attention to a dynamic interplay among emotional, cognitive, and social processes that anchor individuals in the world around them. They do not, however, come to terms with the issue of whether ingroup amity and outgroup enmity are driven by similar or different mechanisms, nor do they call attention to the impact of such structural factors as power, legitimacy, or access to resources on ingroup bias (see Scheepers et al., 2006). Part of that dynamic interplay is captured by social network analysis (SNA). Based on advances in statistical analysis, SNA provides a sophisticated understanding of social behavior within and between groups (Wolfer et al., 2015). Of particular interest to national identity researchers is the focus on both friendship (referred to as positive interdependencies) and hostility (negative relationships) among a population of respondents. The analyses of nomination data (of the form “who are your friends”) provide graphic representations of networks that can be distinguished in terms of tightly or loosely connected groups. A next step, yet to be taken by network researchers, is to use this structural distinction to evaluate the intensity of group identity and the strength of the ingroup bias. For nationalism scholars, these analyses provide a window into how and why individuals favor their own groups over other groups (Tarrant, 2002). By scaling up from the interpersonal to the group level, we add structural dimensions to our understanding of the role played by imitation in the formation identities. A question is whether the insights gained at the group level have relevance for larger collectivities, including those represented in the relatively new context of social media, a topic that we will take up in the next section.

292

National Identity

Collective Dynamics Moving from analyses of interpersonal and small group processes to collectives, we note similarities and differences. A similarity is the durability of attachments to interacting people and to abstract nations. Important differences turn on size, population diversity, and institutions. For nations, the spread of loyalties across sub-populations and the type of leadership count. These dimensions have implications for emotions, particularly trust and mistrust, and cognitions, regarding the spread of worldviews. They inf luence the ease or difficulty of mobilizing populations for action. Problems encountered by social psychologists dealt primarily with small laboratory groups. These are compounded when the units are organizations or nations. As we move up the size scale, the entities become fuzzier and thus less amenable to empirical analysis. For public opinion scholars, the nation is represented by the results of large probability samples of bounded populations. These polls provide aggregated responses that ref lect attitudes, beliefs, or anticipated behavior of the respondents. The data ref lect the extent to which a nation is united or divided around certain public issues, such as media coverage of the Olympics (Billings et al., 2013). They also include sentiments toward a nation, defined as a statistical aggregation of its citizens’ attitudes. As in small group research, the unit of analysis in these surveys is the individual. The Durkheimian proposition about social facts is not addressed, nor do they address the Thibaut and Kelley (1959, p. 147) assertion that “it is possible to describe much of the complex patterning in the interactions of an entire society in terms of a system of such reciprocal roles.” Like Durkheim, we, in the 21st century, struggle to come to terms with the difference between statistical aggregations and a separate level of analysis that transcends individuals and small groups, where the sum is larger than its parts. What then are some directions for defining the nation as an entity at the systemic or societal level of analysis, separate from its citizens? Insights from research using SNA can contribute to our understanding of collective dynamics. As we note earlier, the group membership networks uncovered by SNA distinguish between in-groups and out-groups. A question is whether the approach can be scaled up to capture fault-lines or cleavages in the larger society. For nationalism scholarship, a key division is party identification and the corresponding political ideologies. SNA can provide a window into connections among networks in the private and public sectors that engender support for policies and candidates. By tracing the evolution and change in these networks, we may be able to predict the direction of policies from a, more or less, nationalist (or internationalist) perspective. An example is the emergence of the nationalist president Donald J. Trump following a relatively internationalist Barack Obama administration. Particularly interesting in this example are the shifts in allegiance (group membership) that occurred among independents who joined Republican networks.

Scaling Up, Down, and Across

293

Another direction is to account for societal variation. A set of pertinent dimensions consists of durability, spread, and coercion (Druckman, 2001). Durability or f luidity is defined by the extent to which national identities change with changed circumstances. Durable national identities are more likely in traditional religious societies. Spread refers to how widely an identity is shared through a national population.5 Diaspora nation minorities, for which a traditional identity may be stronger than the majority “at home,” serve to increase these identities’ durability. They would not inf luence the spread of an identity through a single foreign host society, though by definition they do spread an identity outside a “homeland.” Spread may be indicated by polarization or the frequency of dissident protests. It has implications for the ease or difficulty of mobilizing for collective action: Fractionated or divided societies are more difficult to mobilize for collective action; they are also less vulnerable to mythmaking by regimes in power (Van Evera, 1995). Coerced identities, ref lecting demands for loyalty to authoritarian regimes, can be outwardly strong at the collective level and less so at other levels. These nationalism dimensions were used to facilitate case analyses in class exercises. Students were asked to analyze current international negotiations in terms of these dimensions. Cases recently used were Brexit and the Copenhagen Climate talks. These were understood, in a first exercise, through the lens of durability of national identities in delegates’ societies, the spread of identities through the negotiating delegations’ societies, and the extent to which identities were coerced or voluntary in their respective societies. A fourth dimension was whether the various national delegations shared or were divided on their identities. They were also asked to project an outcome of the talks, which was compared to the actual outcome of the completed talks. A second exercise asked students to analyze the same cases through the lens of a set of situational variables, including time pressure, negotiating alternatives, accountability to principals or constituents, and media exposure. A final exercise consisted of a comparison of the two analyses. The questions included the following: What insights emerged from each of the analyses? Which analysis, nationalism dimensions or situational variables, produced more interesting or deeper insights? Which analysis was more accurate in predicting the outcome? In the most recent class, students thought that the nationalism lens produced more compelling insights than the situational lens. These class exercises illustrate the relevance of collective dynamics in analyses of international interactions or negotiations among nations. Although the focus was on small groups of delegates, several dimensions ref lected aspects of the delegates’ societies. A third direction is suggested by the research on collective opinion formation and cascades. The social inf luence simulations run by Moussaïd and colleagues (2013) identify two attractors of opinion that they call the expert and majority effect. Of particular interest is the discovery of a positive reinforcement loop where a majority of the population converge on a similar opinion in the manner

294

National Identity

of groupthink or social “echo chambers.” This loop is a feature of complex collective dynamics where the collective outcome is strongly path dependent on the initial conditions (see also Valori et al., 2012). Those conditions consist of the two attractors, expert and majority opinion, acting together but not necessarily beneficial to the group, which may be either relatively small or very large. Their probes go further by identifying what happens when the two attractors disagree with one another. These investigators evaluated their opinion formation model at both the individual/small group and collective levels of analysis. The former consisted of experiments; the latter was explored with simulations. The simulations enabled them to scale up from individual behavior to collective dynamics. Interestingly, similar insights about opinion formation and change were generated at the two levels, but the study also calls attention to difficulties in doing empirical research at the collective level.6 A similar challenge exists when we scale up from studies of individual identities to population-level nationalism, including evaluating how collective opinion changes over time. A key question is whether national identities have similar properties to opinion formation and change. A related body of work is on opinion cascades. Of particular interest is the Macy et al. (2019) finding on political polarization. They show that collective dynamics produce counter-intuitive findings on how inf luence operates. Specifically, the strong correlation between substantive political opinions and party identification found in repeated studies conducted with sample survey respondents was overturned at a collective level of analysis. For these authors, the collective level emerged from repeated runs, referred to as cascades, of persuasive messages, similar with how Monte Carlo simulations are conducted (Carsey and Harden, 2015). They compare responses to 20 questions in an inf luence condition with an independence condition: the former reinforces party identification by displaying opinions of previous participants in blue (Democrats) or red (Republicans) fonts; participants in the latter condition always saw the statement in a purple or neutral font. Party alignment was stronger in the inf luence than in the independence condition. However, alignment is not predictable based on identification where Democrats and Republicans align with their own party’s positions. Rather it is a result of the opinions expressed by first movers. The distribution of opinions across ten repeated runs or worlds are sensitive to chance variation in the opinions of early movers, who may align with either party.7 The Macy et al. (2019) findings challenge assertions that partisan divisions are based on deep-rooted opinions or core values. They are more likely to result from a path dependence on chance events in the initial conditions. The opinions that divide national populations may be more subject to situational f luctuations than to durable belief structures. The extent to which these findings apply to national identity, which encompasses a cluster of related opinions, remains unpacked. They do, however, raise the issue of durability of identities and illuminate the factor of contingency. They also call attention to the importance of a collective level of analysis separate from the interpersonal and group levels.

Scaling Up, Down, and Across

295

Additional insights come from holistic dynamical systems analysis. The causal loop diagrams and mathematical models constructed by Liebovitch et al. (2020) reveal that there is no single factor that can explain the way that outcomes emerge from system processes. Rather small effects of many interacting factors produce collective outcomes like sustainable peace or contribute to the transitions from negative (war prevention) to positive (sustained relationships) peace. Similar dynamics may occur in the transition from nationalism (where ingroup amity coincides with outgroup enmity) to patriotism (where ingroup amity is independent of outgroup enmity). The research challenge is to develop models that demonstrate these transitions at a collective level. An implied logic is that as ever more issues and choices are introduced into a polity/society, felt group identities will gradually disintegrate, even if they initially stimulate new groups. Another impetus for changed collective identifies is socio-political change. Todd’s (2005) typology of responses to social change is an attempt to explain the connection between individual choice and changes in institutional structures. These kinds of system analyses add complexity to our understanding of identity developed from imitation processes and group structures. They set the stage for a discussion of interactions among the levels to which we now turn.

Integrating the Levels The previous sections review research at each level of analysis. We now discuss interactions among the levels. Our aim is to describe and demonstrate how interpersonal and group experiences connect to larger contexts in inf luencing the way that national identities emerge and change through time. One way to conceptualize connections among levels is a path that scales up from micro through meso to macro processes. Research on justice in peace agreements by Druckman and Wagner (2019) shows how this is done. The authors measured processes at each level. Peace negotiator interactions were coded in terms of indicators of procedural justice during the negotiation process and by indicators of distributive justice in the outcome. The implementation period following agreement was assessed in terms of indicators of adherence by former combatants to the terms of agreement over a five-year period. The collective level was captured by an index of durable peace in the larger society. Statistical analyses showed that each level, from interpersonal to collective, provided the necessary conditions for the next stage. The quality of negotiator interactions was essential for societal change to occur. Another example from the realm of peace agreements, but more relevant to issues of national identity, comes from research on civil society. Cuhadar and Druckman (2022) examined the role played by different types of civil society groups in the way peace processes unfolded in 50 societies. The three levels of analysis were engaged in each society. Interpersonal dynamics were ref lected in interactions among negotiators representing one of the disputing groups. Group dynamics occurred among the civil society actors who represented women, minorities,

296

National Identity

labor unions, and other groups. Collective dynamics were captured by the larger civil society with a stake in the outcomes and their implementation. This study provides an example of both competing and overlapping identities, for example, preferences for different outcomes among the various civil society groups and between certain civil society groups and the negotiators. This matrix of conf licting identities served to slow progress in talks and cause problems for implementing the agreements (Nilsson, 2012). Rather than a path that scales up or down, these interactions occur simultaneously among multiple levels, during and after reaching agreement. A third example of interactions among levels is the 2018 Prespa agreement between Greece and the Former Yugoslav Republic of Macedonia (FYROM). The agreement changed the latter’s official name to the Republic of Northern Macedonia. This decision incited a nationalistic backlash in Greece, including increases in hanging national f lags in cities and support for right-wing political parties (Martinez et al., 2022). These reactions were interpreted by the authors in terms of identity threat posed by sharing one’s cultural heritage via nomenclature. From the vantage point of our levels framework, these reactions can be interpreted as a scaling down from official symbolic decisions to popular attitudes that spread through mimicking in neighborhoods and communities. A similar scaled-down re-invigoration of nationalist movements and populist parties, construed as group dynamics, occurred within some states that became members of the EU. A stronger sense of national identity in the face of perceived threats from, or dissension with, a supranational organization was evident. We elaborate on this case of competing identities in a later section. A fourth example, also from the field of conf lict and peace studies, is found in the literature on interactive conf lict resolution (ICR). This type of workshop intervention was designed with the avowed purpose of transferring changes in attitudes that occur during face-to-face interactions to the level of societal change. It provides another example of scaling up. Identity issues are central in the design and implementation of the ICR workshops. Rouhana (2000, Figure 8.1) presents a theoretical path that travels from micro- to macro-level components of the ICR technology. His micro-level activities include training and analysis exercises engaged in by workshop participants. The participants are selected on the basis of criteria that define them as inf luence-agents in their own society (e.g., Israel and Palestine, Greeks and Turks in Cyprus). These activities are intended to serve conf lict-resolution objectives: increasing the differentiation of the other side, changing the enemy image, reducing mutual stereotypes, achieving deeper understandings of the psychological bases of the conf lict. These objectives serve the larger goal of societal change, including changing beliefs about the adversary, inf luencing decision makers, and changing the dynamics of the conf lict. Evaluations have shown that attitude change usually occurs in the workshops but rarely translates into societal change (Fisher, 2007). The ICR interventions suggest that identities may be more f luid at the interpersonal than the societal level. This could be due to the difference

Scaling Up, Down, and Across

297

between the settings of direct face-to-face interactions and less direct attempts by the “converts” to inf luence decision-makers who did not have the workshop experience. Although the workshop participants were inf luential in their societies, they may be viewed as heretical by friends and colleagues who did not have the workshop experience (Doob and Foltz, 1974).8 It could also be due to a strategic misstep at the group level of analysis. Insufficient attention to networking may pose a barrier to attempts to persuade elite decision makers. Rouhana’s (2000) framework skips this intermediate level of analysis. It connects micro (interpersonal) to macro (collective) without the intermediate step of changing group or community perceptions of out-groups. Our scaling up argument emphasizes graduated changes through the three levels, accounting for specifics of the context in which the changes occur. We take this up in our case analysis of the EU. Interaction dynamics among the three levels also occur between national, sub-national, and supra-national identities. Levendusky (2018) illuminates how context inf luences the relative salience of party and national loyalties. His example of ardent US Democrats and Republicans meeting in a foreign country conveys how context heightens or relativizes one or the other or both identities. They related to each other in terms of shared national identity rather than party differences. Levendusky (2018, p.  59) notes that “when respondents’ sense of American national identity is heightened, they come to see those from the other party as fellow Americans more than members of an opposing political tribe.”9 Similar with the function served by superordinate goals, shared collective identity may be an antidote to the emotion-driven politicization and aligned dislike of the opposition party and its supporters. This scaling up ref lects an interplay between group or party and collective dynamics. Interplay in this example is a result of context changes. The shift may also work in the other direction. For example, national identity is salient when citizens are recruited or volunteer for military service. However, that may give way to primary (sub-national) group attachments on the battlefield. Loyalty shifts from the nation to combat buddies when the going gets tough (Lynn, 1984; Daley, 2018). These arguments and examples raise questions about the conditionality or plasticity of identities (Legro, 2009). Grommé and Scheel (2020) show how bureaucratic-political agencies combine performativity and statistics to create identity categories and distinguish a majority (ethnic) national category from various work-related categories. More generally, identities are part of a dynamic social field, as described by Sztompka (1994) and Delanty and O’Mahony (2002). Examples come from the arena of US Congressional politics where collective outcomes in the form of electoral victories or defeats emerge from the loyalties of individual voters. Most representatives display an unwavering loyalty to their parties. Conf licting narratives about issues like presidential bribery are framed to sway public opinion with the goal of garnering votes for their candidates in upcoming elections. The key micro-level variable is the party loyalty of individual voters. The key macro-level variable is aggregate

298

National Identity

votes in the election – a form of collective action. The relationship between these variables is mediated by narratives espoused during the public committee hearings and by social media networks, regarded as a group or meso-level dynamic. Election results are indicators of performance, serving to weaken or strengthen loyalties depending on the outcome. The US politics example illustrates an intertwining of loyalties or attachments at the individual level and party cohesion at the group level (Rosenberg and Beattie, 2019). Effects move in both directions, from loyalty to cohesion and vice versa. Through time and repeated interactions among party members, a set of norms in the form of consensual narratives emerge. When these norms are shared widely, they can harden into ideologies that rationalize and reinforce pernicious stereotypes of competing groups, as Fyfe (1992) observed regarding racial examples. A critical step in this process is the transition from words in the form of narratives to collective actions in the form of voting or recruitment for fomenting regime change or warfare. More broadly, the competing narratives espoused by the contending parties have implications for a political culture. Democratic and Republican Party candidates in the US often point out that the next election is about who we are as Americans. This rhetoric invokes core beliefs often contested in cultural conf licts: for example, different definitions of honesty in interpreting evidence of wrongdoing. Both political parties are challenged to be sufficiently persuasive to retain loyalty from members and to attract potential supporters from independents or converts from the other party. They often appeal to national identity, with Republicans espousing America first rhetoric and Democrats urging their adherents to recover the essence of American values. Each party accuses the other of selling America short by forfeiting what it means to be an American. A research theme suggested here is the relative impacts of group loyalty and incentives on political actions. For example, are congressional representatives motivated primarily by party loyalty or the power accrued from winning elections? The prevailing narratives (words) of both parties emphasize the importance of loyalty to country, which is a necessary accompaniment to their primary goal of obtaining or retaining power by winning (deeds). The examples presented in this section show how the three levels interact in practice. They illustrate the interdisciplinary analyses on nationalism advanced by Malešević (2013). Sometimes these processes scale up through time from interpersonal to intergroup to collective level dynamics. An example is when meso-level civil society agents interact with the larger macro-level civil society. Other processes scale across, such as when small combat units deal with primary group and national identities at the same time, or they may move from macro-level agreements to many popular (though separate) expressions of a national identity. Dilemmas that occur when trying to transition from one level to another are evident in the ICR interventions, in the conf licts between national and sub-national or small group identities, and from words in the form

Scaling Up, Down, and Across

299

of political narratives and deeds in the form of voting or collective action. Two questions emerge: How would a multi-level analysis of national identity be performed in complex, multi-layered organizations? What are the pathways to the development of an integrated theory? These questions are taken up in the sections to follow.

Multilevel Identities in the European Union – Conflicting, Compatible, Contingent? The examples discussed earlier deal mostly with a few nations. The EU presents a more challenging case for the levels framework. The interplay among supranational institutions, states, nations, and other entities has implications for group and interpersonal processes. Other features are shifting dynamics and diffusion or mutation through time. These themes raise questions about the application of the three-level framework. If we conceive of the EU as an organization, Ashforth’s (2016) model can assist our search for theoretical pathways. He conceptualizes an “I think, we think, it is” progression, analogous to phased integration. Collective identities “tend to emerge from individual identities, or at least individual conceptions of what the collective is or can be, and collective identities in turn both enable and constrain identities nested within them” (Ashforth, 2016, p. 81). Later, after politics has entered the discussion, Ashforth (2016, p. 82) suggests that “espoused identities tend not to be literal descriptions of the organization’s essence, but somewhat idealized and aspirational depictions.” These and other observations are insightful and useful for our aims.

National Identities and Supranationalism Overcoming nationalism was a principal motivation for the ideas and institutions that developed into the EU, which some see as a threat to nations and their states. There are several ways that national identities could be reinvigorated in the context of an organization or system that encompasses a supranational element. One manifestation could be confrontational, as opposition to a perceived rival for macro-level authority. Such friction existed throughout the European integration process and intermittently intensified. A second possibility is that the EU’s forerunners rescued the nation-state, a unique form of principal-agent arrangement (Milward, 1992). Identity is less crucial in this account, which emphasizes material interests and pragmatic accommodation. Tension resides in the coexistence of both threat from and accommodation to the supranational concept. A third possibility is that which represents the de facto predominant EU elite consensus. National identities revive in partnership with supranational institutions, which do not make claims on the sentimental or cultural dimensions of individual, group, or collective identities. Such a condition implies a division of responsibilities and a sharing of macro-level status, in functional terms. This

300

National Identity

arrangement emerged because integrationist elites realized that the idea of the nation – present in individuals, families, and villages (micro); cities and regions (meso); and constitutions, parliaments, and electoral systems (macro) – was too powerful to attempt its dissolution. Acknowledgement of nations and their myriad local or regional expressions was conveyed in principles of devolution and the allocation of competences. These principles eventually found residence in supranational EU laws, budgets, and policies.10 Despite the presence of alternative identities, neither nations nor identification with them have expired. Nations co-exist, in the minds of individuals and at a collective level, with a networked and semi-integrated EU. Can the process of integration itself function as a social emotion, as it does in small groups and nations? Does drawing closer as a society/ polity, or as a family of cultures, as well as a single economy or security community, presume emotional attachment? An alternative conception, recalling the dynamic social field noted earlier, is provided by Delanty (2016). Focusing mainly at the meso level, he argues that networks, rather than coherence around a national idea, is the basis for social formation in Europe. Whereas Karl Deutsch (1953) regarded social communication as the glue of national identity, Delanty perceives the same as providing connectivity for networks. Contrary to the assumption of society as a nation, defined by a state, the social is a field of relationships, which do not need to be understood through a national or state prism.

Shifting Dynamics Among Levels The EU is a useful case to draw on as an illustration of how politics operate up, down, and across multiple levels of governance. However, the three-level concept may not capture multiple identities in the EU’s institutional context. If we retain three and make the EU the macro or collective, and interpersonal relations the micro, then the national becomes the meso or group level. There is overlap between meso and micro, or meso and macro, such that interim levels might be required. One would be needed to accommodate regions, or we divide the meso into upper (nations) and lower (regions) layers. If nations are elevated to share the macro level, the meso (group) level still contains many different types of entities (regions, interest groups, cities, small and medium-sized enterprises, clubs, networks), some of which are transnational. We could also note a ‘meta-macro’ level of exogenous inf luences and global conditions. Thus, supranational–national interaction is only part of the story. Observance of these developments, manifesting in the emergence of a multilevel polity, precipitated the first studies on multilevel governance (Marks and Hooghe, 1996; Hooghe and Marks, 2001). This literature recognized the importance of horizontal as well as vertical interactions. It also recognized a meso level, principally regions, embedded as a key part: the EU cooperating with member states, responding and appealing down to local communities, up to national governments, or circumventing the last to deal directly with

Scaling Up, Down, and Across

301

Commission Directorates or the Committee of the Regions. Regional policy is one of the largest EU budget items, and regions can be regarded as mediating bodies between supranational (macro) and local (micro) levels that, if not entirely bypassing nation-states, reduce their inf luence as gatekeepers or ultimate decision-makers. In some EU member states, regions have greater formal authority than in others. They are an “unextractable and important tier of government within the EU multi-level polity,” a defining component in a Europe “of ” or “with regions” (Schakel, 2020, p. 771). Regional governments and civil groups deploy subnational patriotisms in competition among themselves and sometimes with the national level. Alliance with supranational actors can exploit subnational identities, potentially to offset national identities.11 The previous sections also call attention to a more dynamical conception of complex organizations. As we noted earlier, these models depict how many interacting factors produce collective outcomes. They also trace the paths that lead to transitions such as from a pre- to post-Brexit EU environment. The idea of paths or path-dependencies is also highlighted by models of social inf luence processes. Of particular interest in these is the emergence of unpredictable collective outcomes from interpersonal and group dynamics. The first-mover finding from the Macy et al. (2019) research may apply to certain EU processes. For example, policy initiatives, often taken by more powerful member states such as Germany or France, have moved upward to the supranational level and resulted in new EU laws and regulations. It is also the case that EU decisions can cascade downward in unpredictable ways, with different interpretations and implementations by various member states, networked partners, and regions. Meso-level actors may have mediating or constitutive roles in these processes. Power and authority dynamics shifted and diffused as multiple actors and levels entered the EU context. It is not so clear how identity accompanies the politics through all the levels of governance, from supranational institutions, sharing or competing for macro-level authority with nation-states, down to the meso level and the micro level. A strong connection between governance and identity through all these levels is patchy or underdeveloped: it may be the missing link.

Pathways to Theory An encompassing and plausible theory will be dynamic and multi-level, addressing fundamental group attachments and emergent properties in the world. Progress in accomplishing this goal depends on understanding how the levels interact. The examples provided in our earlier sections are first steps in this direction. They call attention to both vertical and horizontal interactions: downwards from macro to micro, upwards from micro to macro, and across with simultaneous interactions. In this section we raise questions about the interplay among levels that promote theory development.

302

National Identity

Moving from micro- to macro-level processes, we ask: What motivated Greek citizens to display f lags as an expression of protest to the Prespa agreement? How did changes in the intensity of identity occur among participants in the ICR workshops? Why did a majority of British citizens opt out of the EU, viewing membership as a threat to their sovereignty? These examples illustrate a ramping up (Prespa, Brexit) and reducing (ICR) of the strength of national identities. Rather than trying to understand these responses through a micro-level or disciplinary lens, we suggest a focus on the paths that followed at meso and macro levels. The Macedonia name change occurred at an official level. Brexit was made official through long and difficult negotiations. In contrast, the ICR experience did not lead to societal changes and, in fact, caused some participants to regret their experiences. Explaining these paths is a basis for multi-level theory development: Why do some changes at the micro level lead to positive and others to negative changes at the macro level? Moving in the opposite direction, we posit other questions: Why are shared national identities heightened in foreign contexts? How do aggregate opinions or election results strengthen or weaken national (or party) identities? Why do national identities weaken in the transition from military mobilization to combat? These questions ask about the relationship between context or aggregate events and the strength of national identities. A related idea is that micro-level processes are a microcosm of macro-level decisions. This idea gains relevance in the literature on diplomatic negotiations where negotiated decisions ref lect policy-level processes. Explaining these downward paths is another challenge for developing multi-level theory. A further consideration is simultaneous movement up and down the levels ladder: individual loyalties spiral upward, as in opinion cascades, and downward, as when collective norms surface as shared attitudes. In the ICR workshops, the clash between upward (inf luencing society) and downward (inf luenced by society) pressure was evident. In building a multilevel theory, it is important to identify mechanisms that explain how microlevel processes inf luence macro-level processes and vice-versa. Little is written about these connecting factors. One element of a new approach, consistent with our multi-level concept, is to examine the roles played by meso-level group dynamics. The implementation stage in peace agreements provides an example. Druckman and Wagner (2019) showed that micro-level interactions during the negotiation process inf luenced durable peace at the societal level through cooperative community activities during the implementation of the agreements. The failure of ICR workshops to induce change at the societal level may have been due to overlooking the inf luence of networks operating in both the Israeli and Palestinian communities. Instead, the workshop organizers assumed that their inf luential participants would elicit change at the highest levels of government. The concept of bridging people was undeveloped in this conf lict resolution context. The bridge was more likely to reside in larger networks with a stake in social change. Summitry provides a different type of example of multi-level processes. Improved interpersonal relations between political leaders at bilateral

Scaling Up, Down, and Across

303

or multilateral meetings must translate into negotiated agreements at a meso level, which then lead to policy change at a macro level.12 A recent example of both types of summitry is President Biden’s participation in the G7 in England followed by his meeting in Geneva with Vladimir Putin. Van Wijk et al. (2019) argue that social innovators, usually regarded as emerging from the micro level, are unavoidably inf luenced by the entirety of social orders, which pattern their behavior. Micro and macro levels are in a symbiotic relationship. It is, however, the meso level, where “actors’ interactions and framing produce the frictions, highlight the tensions, and identify or create the cracks behind the new opportunities for social innovation” (Van Wijk et al., 2019, p. 890). It is at this level that actors might “begin to jointly (re)negotiate the structures, patterns, and beliefs that constitute their social worlds” (Van Wijk et al., 2019, pp. 890–891). These examples highlight the importance of mesolevel processes in developing a pathways theory. A theoretical understanding of the levels concepts would be incomplete if we did not also consider simultaneous interactions, by which we mean two or more levels interacting at the same time. The Summit example mentioned is a case in point. Interpersonal relations between two or a few leaders merge with official government decisions at a macro level. The relationships developed between leaders during their face-to-face discussions have direct policy consequences. Another example comes from international exchange programs. Pettigrew and Tropp’s (2006) meta-analysis of contact studies shows that the key to cooperative relations between hosts and visitors is institutional support for the program. The macro-level endorsement underwrites the program and contributes to the quality of the micro-level exchange experience. A more complex example of simultaneous interactions is the EU case. Shifting identities between a supranational entity and nations incite a push and pull in citizens’ attachments. A question is whether the EU attachment weakens the strength of earlier socialized national identities. Or do these identities serve different functions, with the EUropean more instrumental and the national more sentimental? Our focus on pathways emphasizes directions through which national identities travel. It also raises questions about how the strength of those identities may change as they move around the levels landscape. This contribution is a step towards a theoretical understanding of the concept. Next steps include discovering the mechanisms that help to explain changes and their consequences. Central to our quest is: What happens to individual identities when we focus attention on groups or nations? From a scaling up perspective, insights come from research on cascade dynamics. From a scaling down perspective, the negotiation microcosm idea provides an interesting pathway. From a horizontal perspective, the merging of levels idea from summitry and the simultaneous supranational and national identities from the EU example are promising. Each path may depend on activities that bridge the individual to the collective. They occur largely at the communal or meso level of analysis. These are possible routes to an integrative theory of national identity.

304

National Identity

Conclusion In this chapter we build on earlier interdisciplinary analyses. These include studies of nationalism in (international) sociology (Malešević, 2013), of individualsociety linkages in social psychology (Jaspal et al., 2016), and on identities in organizational studies (Ashforth, 2016). Each of these treatments emphasizes the importance of cross-level connections. We also recognize the value of these connections and extend previous work in several directions. One is to define and review relevant research on identity from each of three levels, referred to as interpersonal, group, and collective dynamics. Another is to integrate the levels by providing a variety of examples that illuminate interactions among two or more. The examples raise a distinction between vertical (up and down the levels ladder) and horizontal (across levels at the same time) linkages. A third direction is to reveal complexity using a case where supra and national identities intersect and shift through time. A fourth direction is to suggest pathways toward theory development. The pathways include possible disconnects from individual to collective opinions, changes in the strength of identities as processes move from the interpersonal to the collective, and the importance of group-level processes as bridges between the micro and macro levels. These contributions open new lines of inquiry on the mechanisms that link the levels of national identity. One type of mechanism focuses on the way that group or community dynamics connect citizens’ attitudes to societal norms. A second deals with process-context relationships: for example, how do exchange programs and other interactions reflect and influence international relations? A third type aims to understand how interpersonal relations among leaders translate into national foreign policies. These lines of inquiry can be investigated with process tracing methodologies (George and Bennett, 2005). They encompass scaling up, scaling down, and scaling across modes of analyses.

Notes 1 The relationship between behavioral matching and influence has been exploited by entrepreneurs eager to make profits. The most notable example is Neuro-Linguistic Programming (NLP) based on modeling the way that family therapists influence their patients (see Druckman and Swets, 1988). 2 Witness in this regard the demise of social psychology departments in the US, notably at Michigan and Harvard, and the migration of social psychologists to business schools and communication departments. 3 The earliest studies in this tradition were conducted by Blake and Mouton (1962), Bass and Dunteman (1963), and Singer et al. (1963). 4 An important difference between political and experimental group identities is that the former are likely to last longer than the latter. This difference raises a question about the durability of identities in laboratory groups. Those may be temporary, reflecting the situation constructed by the designers of the experiments. This does not, however, render them as less important than political or organizational identities. It suggests that identities are strongly influenced by situations (see Meyer et al., 2010, for a review of the situational strength literature). 5 Note the difference between durability and spread. The former refers to strength of identity (intensity), the latter to sharing of an identity (extension) in a population.

Scaling Up, Down, and Across

305

6 We distinguish between aggregate analysis of individual opinions and collective behavior. The former are valuable for studies of mobilization such as protests, but do not capture the idea of a collective level of analysis. 7 An international example is norm cascades or norm band-wagoning (Finnemore and Sikkink, 1998). 8 The attitude changes that occur from participation in the workshop can be problematic in another way. A softening of national identities can lead to a larger re-evaluation of lifelong identities with the attendant psychological problems. A way around these possible consequences is to retain one’s pride in country while working with adversaries on cooperative projects (see Deschamps and Brown, 1983, for a way to do this). 9 An example of recognition of common identity at a supranational or civilizational level is the chance meeting of English and French expeditionary personnel at Botany Bay in 1788. The French commander, Count Lapérouse, wrote in his journal that “Europeans are all compatriots at such a great distance.” Similar connections among individuals who do not share a national identity but do share a supra identity, in the sense that it can incorporate many and scale up to another level, are common today. The situation is crucial. 10 Further complicating the EU system is the range of opt-outs and other variations across policy and legal domains, such that in those instances not all member states or regions operate at the same level. 11 Another complication for a three-level schema is that majorities in some regions consider themselves to be ‘nations,’ for example, Catalunya or Euskal Herria in Spain. 12 See our earlier discussion of public Congressional committee hearings as a meso-level process.

References Abramowitz, A.I. and Webster, S. (2016). The rise of negative partisanship and the nationalization of U.S. elections in the 21st century. Electoral Studies 41: 12–22. Adelman, K. (1984). Arms control with and without agreements. Foreign Affairs 63 (2): 240–263. Allport, G. (1985). The historical background of modern social psychology. In G. Lindzay and E. Aronson (eds) Handbook of Social Psychology (3rd ed.). Volume 1: Theory and Methods. New York: Random House. Ashforth, B. (2016). Organizational, subunit, and individual identities: Multilevel linkages. In M. Pratt, M. Schultz, B. Ashforth, and D. Ravasi (eds) The Oxford Handbook of Organizational Identity. Oxford, UK: Oxford University Press. Axelrod, R. (1984). The Evolution of Cooperation. New York: Basic Books. Back, K. (1979). The small group – Tightrope between sociology and personality. The Journal of Applied Behavioral Science 15 (3): 283–294. Bass, B. and Dunteman. G. (1963). Bias in the evaluation of one’s own group, its allies and opponents. Journal of Conflict Resolution 7 (1): 16–20. Billings, A.C., Brown, N.A., Brown, K.A. and Neth, H. (2013). From pride to smugness and the nationalism between: Olympic media consumption effects on nationalism across the globe. Mass Communication and Society 16 (6): 910–932. Blake, R.R. and Mouton, J.S. (1962). The intergroup dynamics of win-lose conf lict and problem-solving collaboration in union-management relations. In M. Sherif (ed) Intergroup Relations and Leadership. New York: Wiley. Bohm, R., Rusch, H. and Baron, J. (2020). The psychology of intergroup conflict: A review of theories and measures. Journal of Economic Behaviour & Organizations 178: 947–962. Bowlby, J. (1988). A Secure Base: Parent-Child Attachment and Healthy Human Development. New York: Basic Books. Brubaker, R. and Cooper, F. (2000). Beyond “identity.” Theory and Society 29 (1): 1–47.

306

National Identity

Campbell, D.T. (1958). Common fate, similarity, and other indices of the status of aggregates. Behavioral Science 3 (1): 14–25. Carr, E.W., Winkielman, P. and Oveis, C. (2014). Transforming the mirror: Power fundamentally changes facial responding to emotional expressions. Journal of Experimental Psychology General 143: 997–1003. Carsey, T. and Harden, J. (2015). Can you repeat that please? Using Monte Carlo simulation in graduate quantitative research methods classes. Journal of Political Science Education 11 (1): 94–107. Cuhadar, E. and Druckman, D. (2022). Let the people speak: What kind of civil society inclusion is best for durable peace? Paper presented at the annual meeting of the International Studies Association, Nashville TN. Daley, P. (2018). On Patriotism. Melbourne: Melbourne University Press. Delanty, G. (2016). The making of European society: Contesting methodological nationalism. Innovation: The European Journal of Social Science Research 29 (1): 3–15. Delanty, G. and O’Mahony, P. (2002). Nationalism and Social Theory: Modernity and the Recalcitrance of the Nation. London: Sage. Deschamps, J.-C. and Brown, R. (1983). Superordinate goals and intergroup conf lict. British Journal of Social Psychology 22 (3): 189–195. Deutsch, K. (1953). Nationalism and Social Communication. Cambridge, MA: MIT Press. Doob, L.W. and Foltz, W.J. (1974). The impact of a workshop upon grass-roots leaders in Belfast. Journal of Conflict Resolution 18 (2): 237–256. Druckman, D. (1968). Ethnocentrism in the Inter-Nation Simulation. Journal of Conflict Resolution 12 (1): 45–68. Druckman, D. (1986). Stages, turning points and crises: Negotiating military base rights, Spain and the United States. Journal of Conflict Resolution 30 (2): 327–360. Druckman, D. (2001). Negotiation and identity: Implications for negotiation theory. International Negotiation 6 (2): 281–291. Druckman, D. and Swets, J.A. (1988). Enhancing Human Performance. Washington, DC: National Academy Press. Druckman, D. and Wagner, L. (2019). Justice matters: Peace negotiations, stable agreements, and durable peace. Journal of Conflict Resolution 63 (2): 287–316. Druckman, D., Mueller, G. and Diehl, P.F. (2022). Exploring the compatibility of multiple missions in UN peace operations. International Peacekeeping 29 (1): 85–114. Druckman, D., Parlamis, J. and Burns, Z. (2022). Can loyal party members be f lexible negotiators? Impacts, of constituent support term limits, and bi-partisan roles. International Journal of Conflict Management 33 (5): 741–761. Durkheim, E. (1895). Les Règles de la Méthode Sociologique. Paris: Felix Alcan. Finnemore, M. and Sikkink, K. (1998). International norm dynamics and political change. International Organization 52 (4): 887–917. Fisher, R. (2007). Interactive conf lict resolution. In I.W. Zartman (ed) Peacemaking in International Conflict: Methods & Techniques. (Revised Edition). Washington, DC: United States Institute of Peace Press. Fiske, S.T. (2004). What’s in a category? Responsibility, intent, and the avoidability of bias against outgroups. In A.G. Miller (ed) The Social Psychology of Good and Evil. New York: The Guilford Press. Fyfe, C. (1992). Race, empire and the historians. Race and Class 33 (4): 15–30. George, A.L. and Bennett, A. (2005). Case Studies and Theory Development in the Social Sciences. Cambridge, MA: MIT Press. Grommé, F. and Scheel, S. (2020). Doing statistics, enacting the nation: The performative powers of categories. Nations and Nationalism 26 (3): 576–593.

Scaling Up, Down, and Across

307

Hogg, M. and Mullin, B. (1999). Joining groups to reduce uncertainty: Subjective uncertainty reduction and group identification. In D. Abrams and M. Hogg (eds) Social Identity and Social Cognition. Oxford, UK: Blackwell. Hooghe, L. and Marks, G. (2001). Multi-Level Governance and European Integration. Lanham, MD: Rowman & Littlefield. Homans, G.C. (1950). The Human Group. New York: Harcourt, Brace and Company. Hornsey, M.J. and Hogg, M.A. (2000). Subgroup relations: A comparison of mutual intergroup differentiation and common ingroup identity models of prejudice reduction. Personality and Social Psychology Bulletin 26 (2): 242–256. Hove, M. and Risen, J. (2009). It’s all in the timing: Interpersonal synchrony increases affiliation. Social Cognition 27 (6): 949–961. Insko, C.A., Hoyle, R.H, Pinkley, R.L., Hong, G-Y. and Slim, R.M. (1988). Individualgroup discontinuity: The role of a consensus rule. Journal of Experimental Social Psychology 24: 505–519. Jaspal, R., Carriere, K. and Moghaddam, F. (2016). Bridging micro, meso and macro processes in social psychology. In J. Valsiner, G. Marsico, N. Chaudhary, T. Sato and V. Dazzani (eds.), Psychology as a Science of Human Being: The Yokohama Manifesto. New York: Springer. Kavanagh, L. and Winkielman, P. (2016). The functionality of spontaneous mimicry and its inf luences on affiliation: An implicit socialization account. Frontiers in Psychology 7, Article 458. Legro, J. (2009). The plasticity of identity under anarchy. European Journal of International Relations 15 (1): 37–65. Levendusky, M. (2018). Americans, not partisans: Can priming American national identity reduce affective polarization? The Journal of Politics 80 (1): 59–70. Liebovitch, L., Coleman, P. and Fisher, J. (2020). Approaches to understanding sustainable peace: Qualitative causal loop diagrams and quantitative mathematical models. American Behavioral Science 64 (2): 123–144. Lynn, J.A. (1984). The Bayonets of the Republic. Urbana, IL: University of Illinois Press. Macy, M., Deri, S., Ruch, A. and Tong, N. (2019). Opinion cascades and the unpredictability of partisan polarization. Science Advances 5 (8). DOI: 10.1126/sciadv.aax0754. Malešević, S. (2013). Nation-States and Nationalisms: Organization, Ideology and Solidarity. Cambridge, UK: Polity Press. Marks, G. and Hooghe, L. (1996). European integration from the 1980s: State-centric vs. multi-level governance. Journal of Common Market Studies 34 (3): 341–378. Martinez, S., Valentim, V. and Dinas, E. (2022). Symbolic conf lict resolution and ingroup favouritism. Journal of Conflict Resolution, in press. McIntosh, D. N. (2006). Spontaneous facial mimicry, liking and emotional contagion. Polish Psychological Bulletin 37 (1): 31–42. Mead, G.H. (1934). Mind, Self, and Society. Chicago: University of Chicago Press. Messick, D.M. and Mackie, D.M. (1989). Intergroup relations. Annual Review of Psychology 40: 45–81. Meyer, R.D., Delal, R.S. and Hermida, R. (2010). A review and synthesis of situational strength in the organizational sciences. Journal of Management 36 (1): 121–140. Milward, A. (1992). The European Rescue of the Nation-State. Berkeley: University of California. Moussaïd, M., Kämmer, J.E., Analytis, P.P. and Neth, H. (2013) Social inf luence and the collective dynamics of opinion formation. PLoS ONE 8 (11): e78433. https://doi. org/10.1371/journal.pone.0078433 Nilsson, D. (2012) Anchoring the peace: Civil society actors in peace accords and durable peace. International Interactions 38 (2): 243–66. Osgood, C.E. (1962). An Alternative to War or Surrender. Urbana, IL: University Illinois Press.

308

National Identity

Over, H. and Carpenter, M. (2012) Putting the social into social learning: Explaining both selectivity and fidelity in children’s copying behaviour. Journal of Comparative Psychology 126: 182–192 Pettigrew, T.F. and Tropp, L.R. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology 90 (5): 751–783. Ramberg, B. (Ed.) (1993). Arms Control Without Negotiation: From the Cold War to the New World Order. Boulder, CO: Lynne Rienner. Rosenberg, S. and Beattie, P. (2019). The cognitive structuring of national identity: Individual differences in identifying as American. Nations and Nationalism 25 (1): 361–384. Rouhana, N.N. (2000). Interactive conf lict resolution: Issues in theory, methodology, and evaluation. In P.C. Stern and D. Druckman (eds) International Conflict Resolution After the Cold War. Washington, DC: The National Academies Press. Schakel, A.H. (2020). Multi-level governance in a ‘Europe with the regions.’ The British Journal of Politics and International Relations 22 (4): 767–775. Scheepers, D., Spears, R., Doosje, B. and Manstead, A.S. (2006). The social functions of ingroup bias: Creating, confirming, or changing social reality. European Journal of Social Psychology 17(1): 359–396. Sherif, M. (1966). In Common Predicament: Social Psychology of Intergroup Conflict and Cooperation. Boston: Houghton Mifflin. Singer, J.E., Radloff, L.S. and Wark, D.M. (1963). Renegades, heretics, and changes in sentiment. Sociometry 26: 178–189. Sztompka, P. (1994). The Sociology of Social Change. Oxford, UK: Blackwell. Tajfel, H. (1981). Human Groups and Social Categories: Studies in Social Psychology. Cambridge, UK: Cambridge University Press. Tarrant, M. (2002). Adolescent peer groups and social identity. Social Development 11 (1): 110–123. Taylor, P.J. (2014). The role of language in conflict and conflict resolution. In T.M. Holtgraves (ed), The Oxford Handbook of Language and Social Psychology. Oxford, England: Oxford University Press. Thibaut, J.W. and Kelley, H.H. (1959). The Social Psychology of Groups. Hoboken, NJ: John Wiley. Tilly, C. (1978). From Mobilization to Revolution. Reading, MA: Addison-Wesley. Todd, J. (2005). Social transformation, collective categories, and identity change. Theory and Society 34 (4): 429–463. Turner, J. (1987). Rediscovering the Social Group: A Self-Categorization Theory. Oxford UK: Blackwell. Valori, L, Picciolo, F., Allansdottir, A. and Garlaschelli, D. (2012). Reconciling long-term cultural diversity and short-term collective social behavior. Proceedings of the National Academy of Sciences 109: 1068–1073. Van Evera, S. (1995). Hypotheses on nationalism and the causes of war. In C. Kupchan (ed) Nationalism and Nationalities in the New Europe. Ithaca, NY: Cornell University Press. Van Wijk, J., Zietsma, C., Dorado, S., de Bakker, F. and Marti, I. (2019). Social innovation: Integrating micro, meso, and macro level insights from institutional theory. Business and Society 58 (5): 887–918. Whiten, A., McGuigan, N., Marshall-Pescini, S. and Hopper, L. (2009). Emulation, imitation, over-imitation and the scope of culture for child and chimpanzee. Philosophical Transactions of the Royal Society 364: 2417–2428. Wolfer, R., Faber, N.S. and Hewstone, M. (2015). Social network analysis in the science of groups: Cross-sectional and longitudinal applications for studying intra- and intergroup behaviour. Group Dynamics: Theory, Research, and Practice 19 (1): 45–61. Zentall, T. (2006). Imitation: definitions, evidence, and mechanisms. Animal Cognition 9: 335–353.

PART 6

Process and Outcome Justice

PART INTRODUCTION

This part consists of two chapters, an empirical analysis of 50 peace agreements and a synthesis of the vast literature on justice and negotiation. These chapters are preceded by an overview of the research on justice. As noted earlier, my interest in issues of justice and peace was sparked by a chance meeting with a colleague at a UK conference in the early 2000s. Generous support from several sponsors put these issues at the top of my research agenda. We learned about how justice principles operate during the process in several types of negotiations, how they are manifest in the outcomes, how they facilitate implementation of agreements, and their impact over the long term in post-war societies. While adding substantive knowledge to the justice literature, we also contributed innovative approaches for doing case studies with both a small and large number of cases. A good deal of the writing on justice pits it against peace. In this view, efforts to hold perpetrators of conf lict accountable would likely slow progress toward a peaceful resolution of differences between the protagonists. An expedient solution is to let perpetrators off the hook with mere slaps on the hand in order to get a cease fire or peace agreement. These agreements are unlikely to be of lasting value. An alternative is to focus attention primarily on the justice issues before turning to settlements. Although these agreements are likely to last longer, due in part to increased trust among the parties, the toll on lives taken in the meantime is a very high a price to pay. The dilemma is that peace must be achieved in a way that does not preclude justice (Zartman and Kremenyuk, 2005). We address this dilemma in our research. The approach taken focuses on justice during the negotiation processes and in outcomes. This includes recognizing both PJ and DJ principles. We then ask about the extent to which adhering to these principles facilitates implementation DOI: 10.4324/9781003293361-30

312 Process and Outcome Justice

of the agreements and their durability over time. We do not however focus on justice as accountability for crimes committed by persons or groups during the prosecution of a war, which are often issues taken up by truth and reconciliation commissions. We ask about contributions made by process and outcome justice to peace, but we also recognize the need to understand retributive and transformative processes. In addition to studying peace, we examined the role of justice in bilateral and multilateral negotiations over trade, arms control, and the environment.

17 DURABLE PEACE

We returned to research on peace agreements following the set of studies on trade, arms control, and the environmental cases. Support came from a Swedish Research Council grant. The funding was sufficient to perform a robust study on the justice and peace issues. Indeed, this was our most ambitious study on the topic. It consisted of a large representative sampling of cases, a very detailed measurement of durable peace, a two-stage modeling approach, and an evaluation of eight control variables. The results further our understanding of the justice and peace relationship as well as bridging micro- and macro- levels of analysis as this is discussed in the scaling article in Part 5. Above all perhaps is the clear support for the importance of a negotiation process that features PJ. This study strengthens the earlier findings in several issue areas. It also adds new findings about how PJ travels through time all the way through to a longer-term societal peace. Social conduct in the form of fair treatment and transparency stands out as a driver of both reconciliation and institutional change. They also set in motion transitional justice processes during the implementation of the agreements. The analyses call attention to community relations, an aspect of reconciliation, as a key to continued cooperation between former rivals. These consist of collaborative activities such as joint participation in re-building the collapsed infrastructures that resulted from the war. Examples of projects are new community centers and other shared facilities, highways and train lines, and health services. When in play, these activities increase the trust needed to perpetuate peace. But they need to be buttressed by improved security and governance institutions at the societal level. For a podcast interview about this article with the JCR editor Paul Huth see https://journals.sagepub.com/page/jcr/podcasts/index.

DOI: 10.4324/9781003293361-31

314 Process and Outcome Justice

Justice Matters: Peace Negotiations, Stable Agreements, and Durable Peace, Daniel Druckman and Lynn Wagner. Journal of Conflict Resolution 63 (2) (2019): 287–316. © Sage Publication Inc. Reprinted with permission.

JUSTICE MATTERS Peace Negotiations, Stable Agreements, and Durable Peace Daniel Druckman and Lynn M. Wagner

Attaining durable peace (DP) after a civil war has proven to be a significant challenge, as many negotiated agreements lapse into violence. Since the end of World War II, civil wars have occurred in all corners of the world, with a dramatic increase in frequency following the end of the Cold War particularly on the African continent. Many civil wars have been terminated with a peace agreement that ends the fighting. Fewer agreements have resulted in successful implementation or a lasting peace between the warring parties. Problems of implementing the terms of the agreements are evident in cases included in the Uppsala Conf lict Data Program’s (UCDP) Peace Agreements data set (1975–2011). Over 40 percent of the 216 intrastate peace agreements saw a resumption of violence within five years (Högbladh 2012). The question raised is: why do so many peace agreements fail to achieve their goal of bringing peace to war-torn nations? We address this question by building on the justice and negotiation, peace agreements, and civil war termination literatures. This study is motivated by a question about the relationship between justice and peace during the termination of civil wars: does justice in the negotiation process and in the outcome lead to peace in the postagreement period? This question suggests a path from negotiation, analyzed at a microlevel, to changes in the formerly war-torn society analyzed at a macro-level: do the ways in which negotiators interact, and the agreements they reach, increase the chances for obtaining peace in their societies? The literatures on justice and peace provide bases for the theoretical claims that support the hypotheses evaluated in this study. The justice literature, developed primarily in social and organizational psychology, has been divided between the importance of distributive justice (DJ) and procedural justice (PJ) concerns. The DJ literature, spearheaded by Adams (1965) and Deutsch (1985), emphasized the principles of equitable and equality outcomes. The PJ literature,

316 Process and Outcome Justice

initiated by Thibaut and Walker (1975) and developed further by Lind and Tyler (1988), called attention to the process principles of fair treatment, transparency, and voice. Both these literatures highlight the importance of fairness during the negotiation process and in its outcome. This is particularly evident in Deutsch’s (1985) equality principle, in Lind and Tyler’s (1988) group value model, and, more recently, in Hollander-Blumoff ’s (2017) findings about social conduct. Placed in the context of peace, these concepts are central to the development of relationships between competitors or adversaries. Building relationships between former combatants in civil wars is essential for building a culture of peace (BarTel 2008). Reconciliation is an important first step for building peace. It is ref lected most clearly in community or collaborative activities among the former combatants as shown by Balcells and Justino’s (2014) work on political violence in the context of civil wars. Less is known about the path from reconciliation to institutional reforms essential for sustaining peace: do institutions respond to improved relationships that follow peace agreements? This question arises in analyses of linkages between the microprocesses of negotiation and the macroprocesses of societal change (for an early treatment of this issue, see Druckman 1980). The earlier analyses informed our research question about micro- to macropaths. This study makes several new contributions. Continuing in the tradition of analyzing the justice–peace nexus in attempts to terminate civil wars (Albin and Druckman 2012; Wagner and Druckman 2017), we provide a novel data set of fifty cases selected with stratified random sampling procedures. The earlier studies analyzed an opportunistic (nonrandom) sample of sixteen cases. Another contribution consists of including a variety of contextual variables in the analysis. Impacts of factors in the broader negotiating environment are compared to those obtained from the negotiating process variables. Only one previous study compared the effects of context (e.g., alliance membership) with process (distributive vs. problem-solving approaches) on negotiations to end violent international conf licts (Irmer and Druckman 2009). Third, we distinguish between types of justice (PJ, DJ), context (eight control variables), and peace (stable agreements [SA], reconciliation, institutional change). A fourth contribution is a statistical evaluation of a path that travels from microlevel processes to macro-level societal changes. This double mediation analysis provides evidence for the micro- to macrolinkage concepts discussed in the peace and conf lict literatures (e.g., Balcells and Justino 2014). It also provides direction for building theories of peaceful societies in the aftermath of war. The sections to follow begin with a review of relevant literature that addresses the key questions of the study. Those questions are a basis for theory building discussed in the next section, followed by direct and indirect hypotheses. A description of eight control variables precedes a discussion of methods (case sampling, coding procedures, and analysis methods) and results, concluding with a section that develops implications for theory building on the relationship between justice and peace.

Durable Peace

317

Literature Review Justice preferences have been shown to guide the process and outcomes of negotiated exchanges. The preferences are both distributional and procedural. Findings from laboratory experiments show an overwhelming preference for equal allocations of resources (Deutsch 1985). The preference for equal outcomes was also found for negotiators attempting to resolve the difficult issues posed by terminating wars (Druckman and Albin 2011). Laboratory findings also showed that PJ principles produce more integrative agreements (Hollander-Blumoff and Tyler 2008), particularly those principles that ref lect social conduct (Hollander-Blumoff 2017). These principles have also been shown to produce more DP agreements by Albin and Druckman (2012). Correlations between the two types of justice have been found to vary by context with high correlations found for laboratory studies and lower correlations found for comparative case studies in the areas of trade, arms control, the environment, and peace agreements (see Druckman and Wagner 2016, for a summary of these results). Justice principles have been found to inf luence the stability of negotiated agreements in both field experiments and case studies. In their field studies of mediation, Pruitt et al. (1993) reported that PJ judgments were a stronger predictor of improved relations with the other party (SA) than goal achievement or satisfaction with the agreement. Adherence to PJ principles also increased perceptions of the legitimacy of agreements in felony cases (Casper, Tyler, and Fisher 1988) and with the cessation of violence in civil wars (Albin and Druckman 2012; Wagner and Druckman 2017). In their study of sixteen peace agreements, Wagner and Druckman (2017) extended the focus from stability following an agreement to long-term peace. They constructed a new DP index consisting of four parts: reconciliation, security institutions, governance institutions, and economic growth. The DP index correlated with the stability of the agreement (SA), defined as short-term absence of violence. A key finding was that SA mediated the effects of PJ on reconciliation: more adherence to PJ principles led to more reconciliation when the agreements were stable. This finding suggests that PJ may set in motion relational dynamics that continue after the agreement is reached and longer term reconciliation between the former combatants is achieved. Further analyses of this path are performed in this study with fifty cases. The results from these studies indicate that PJ and DJ inf luence negotiated outcomes, stability following those outcomes, and aspects of long-term peace. These are not however regarded as sovereign inf luences on these outcomes. With regard to negotiation outcomes, Hollander-Blumoff and Tyler’s (2008) laboratory simulation and Wagner and Druckman’s (2012) historical case analyses showed that problem-solving processes mediated the relationship between PJ and integrative outcomes. Druckman’s (1993) simulation of environmental negotiations showed that news coverage inf luenced negotiating f lexibility. With

318 Process and Outcome Justice

regard to stability, such contextual features as the conf lict environment (CE; Downs and Stedman 2002; Joshi and Quinn 2015; Walter 2002), the use of peacekeeping operations (PKOs) during implementation (Downs and Stedman 2002; Fortna 2008) and the type of agreement including power-sharing provisions (Mattes and Savun 2009; Walter 2002; Hartzell and Hoddie 2008) were found to inf luence the number of years that an agreement remains in force. Less is known about variables that inf luence DP. This study seeks to fill this gap. The impacts of justice on DP and reconciliation are evaluated in relation to many of the contextual variables reviewed above. (Online Appendix 1 provides a summary of the variables analyzed in the earlier studies organized by conf lict stages.) This study knits together these facets of the justice and peace literature by exploring the connections among process and outcome justice on the one hand and short- and long-term peace on the other. The results contribute to the development of a theory that emphasizes the importance of social relationships.

Theory Building: Model Specification and Evaluation Our aim is to contribute to the building of a theory about the relationship between justice and peace. Results obtained from the studies reviewed above render this a plausible goal. Questions are: what kind of theory is suggested? and what kind of evidence is needed to support the development of such a theory? The role played by justice in peace deals primarily with social relationships. Several converging bodies of work recognize the importance of relational dynamics at the interpersonal, intergroup, and community levels of analysis. A theme running through these analyses is social conduct. By the way they conduct themselves, negotiators can attain more or less integrative outcomes (Hollander-Blumoff 2017). By their joint activities, former belligerents can reduce the level of violence in their post-war-torn communities (Balcells and Justino 2014). Both types of interactions – formal negotiations and community relations – contribute to the building of peaceful societies, referred to also as cultures of peace (Bar-Tel 2008). Progress toward DP depends on transferring the gains obtained through reaching agreements to a sustained reconciliation and institutional change. Understanding these connections is the challenge for building a theory of justice and peace. The two-stage statistical models specified and evaluated in this study contribute to theory building. In the first stage, referred to as vertical modeling, we construct a multiple regression model for evaluating the relative strength of both justice and control variables on DP and reconciliation. Results show the possible importance of justice to peace in the context of a larger CE. The second stage, referred to as horizontal modeling, consists of developing a regressionbased double mediation model for evaluating a four-step path from negotiation process to DP. Results show the strength of the connection between the levels, going from microlevel interactions among peace negotiators to macro-level institutional change. This chain is set in motion by adherence to PJ principles during the negotiation. These modeling stages are complementary: the vertical

Durable Peace

319

modeling demonstrates the role of justice; the horizontal modeling shows how the justice–peace relationship may be bridged by fair outcomes and SA. The evidence provided by both analyses addresses the viability of a theory of peace based on relationship building.

Direct Effects Hypotheses This study focuses attention on the impacts of PJ and DJ on DP. Going further than previous studies, we evaluate a number of alternative explanations for the peace that follows the agreements.1 The portfolio of analyses is extended further by examining hypotheses on combined effects of the justice variables.

PJ Recent research by Hollander-Blumoff (2017) showed that courtesy, respect, and trust are the key components of PJ judgments. These findings suggest that PJ is a relational variable. Being treated well during the negotiation process is essential to fostering and sustaining positive relationships. It is also essential for sustaining peace over the long term. Thus, adhering to PJ principles during the process of negotiating should contribute to peace following the agreement. Indeed, strong correlations between PJ and DP were found in the sixteen cases analyzed by Wagner and Druckman (2017). The relationship was strongest on the reconciliation component of the DP index (r = .66). Since the reconciliation component of DP depends more on trusting relationships than the other components – security and governance institutions, economic growth – it would be expected to be most strongly inf luenced by PJ. The previous research on justice suggests two hypotheses for investigation in this study of fifty cases. Hypothesis 1: More adherence to PJ principles during the negotiation process will lead to stronger DP following the peace agreement. Hypothesis 1a: More adherence to PJ principles during the negotiation process will lead to stronger reconciliation following the peace agreement. The underlying constructs for each of the four PJ principles may differ. Fair treatment and transparency are about social conduct during the negotiation process. They have in common a focus on building or hindering the development of relationships between the negotiating teams (Lind and Tyler 1988; Heald 2006). By contrast, fair representation may be regarded as a norm settled on prior to the beginning of formal negotiations (Leventhal, Karuza, and Fry 1980). Party representation may be a sticking point prior to the initiation of the talks (e.g., Hume 1994). Voluntary (or involuntary) decisions refer to pressure exerted from the other parties (Barry 1996). Pressures may take the form of coercion or demands made for concessions. These distinctions suggest variation in the extent to which the four PJ principles are relational. Experiments by Lind, Tyler, and Huo (1997) and

320 Process and Outcome Justice

Hollander-Blumoff (2017) showed that PJ judgments were strongly inf luenced by perceived relationships. More specifically, courtesy and respect were the key elements of those judgments. These elements are likely to be inferred from social conduct during negotiation. Thus, it is suggested that fair treatment and transparency account for the hypothesized relationship between PJ and DP or reconciliation. This relationship is summarized by the following hypotheses. Hypothesis 1b: More adherence to principles of fair treatment and transparency during the negotiation process will lead to stronger DP following the peace agreement. Hypothesis 1c: More adherence to principles of fair treatment and transparency during the negotiation process will lead to stronger reconciliation following the peace agreement.

DJ PJ contains relational principles, DJ deals with matters of fair allocation. Research by Druckman and Albin (2011) on peace agreements showed that DJ was related to SA. Wagner and Druckman’s (2017) study extended the focus from SA to DP. They obtained moderate relationships between DJ and DP, weaker than the correlations between PJ and DP. Kabanoff (1991) distinguished between the DJ principles of equity and equality. The former relates more strongly to productivity, the latter to solidarity (see also Deutsch 1985). In both these studies, the DJ-DP relationship was accounted for by the equality principle. Equality also correlated with PJ and SA. The earlier research suggests the following hypotheses. Hypothesis 2: The more significant DJ principles are in the negotiated agreement the higher the assessments of DP following the peace agreement. Hypothesis 2a: The more significant the equality principle is in the negotiated agreement the higher assessments of DP following the peace agreement. PJ and DJ: Together, Hypotheses 1 and 2 suggest that PJ and DJ principles contribute to DP and reconciliation. Earlier case study research suggests the relevance of an additive model where PJ and DJ contribute relatively independent sources of variation to DP (Druckman and Wagner 2016). By combining PJ and DJ, we create a global justice measure. This combined measure of justice was used as a dependent variable (DV) in the regression analyses, based on the hypotheses stated above.

Indirect Effects Hypotheses The hypothesized justice effects are both direct, as discussed above, and indirect, as discussed in this section. Indirect effects posit one or more variables that mediate relationships between an independent variable (IV; e.g., PJ) and a

Durable Peace

321

DV (e.g., DP). The earlier studies on justice in peace agreements examined the effects of single mediators (DJ or SA). In this study, we extend the path to two mediating variables, referred to as serial mediation. Rationale for the extended path is time ordered, empirical, and conceptual. Peace negotiations precede implementation and societal changes. Justice issues that arise during negotiation and in the agreement have been shown in earlier analyses to inf luence postnegotiation peace processes. This hypothesized path illuminates a relationship between micro- and macro-level variables as discussed earlier (Balcells and Justino 2014). The hypothesized serial mediation path is built from the earlier results showing indirect effects. Those results indicate that both DJ and SA mediate relationships between PJ and DP as well as reconciliation (see Figure 17.1 in Wagner and Druckman 2017). These are, however, single mediators; DJ and SA were explored separately. The fifty cases assembled for this study provide an opportunity to examine an extended path from negotiation process and outcome to cessation of violence and societal change. Adherence to PJ principles may give impetus to signing agreements that incorporate such DJ principles as equality or power sharing (Hartzell and Hoddie 2003), are stable, and facilitate peacebuilding and institution change (Binningsbo et al. 2012). This path may also be self-sustaining as noted by Quinn, Mason, and Gurses (2007): “The longer the peace can be sustained, the less likely it is to break down” (p. 189). Thus, DP could be a path-dependent consequence of negotiation processes and outcomes. Two mediators, DJ and SA, are specified to explore the path from PJ to DP as well as the path from PJ to reconciliation. These progressions are evaluated by the serial multiple mediation model depicted in Figure 17.1. The serial mediation is developed from the six direct effects and three single mediation paths shown in the figure. The direct and indirect effects of PJ are estimated using the

DJ a1

PJ FIGURE 17.1

SA

a3 a2

b1 c’1

Serial Multiple Mediation. Adapted from Hayes (2013).

b2

DP

322 Process and Outcome Justice

coefficients from the following three equations, one for each of the mediators (DJ and SA) and one for the DV, DP (or reconciliation): DJ = iDJ + a1PJ + e DJ; SA = i SA + a 2PJ + a3DJ + e SA; DP = iDP + c ′1PJ + b1DJ + b2SA + e DP;

(1) (2) (3)

(equation [3] could also be: reconciliation (R) = i R + c′1PJ + b1DJ + b2SA + eR). The direct effect of PJ on DP is estimated by the coefficient c′1 in equation (3). Indirect effects of PJ on DP are estimated as the product of coefficients for variables connecting PJ and DP through the mediators. There are three mediators in this model, one through DJ (a1b1), one through SA (a 2b2), and one through both DJ and SA as serial effects (a1a 3b2). These indirect effects sum to produce a total indirect effect of PJ on DP (a1b1 + a 2b2 + a1a 3b2). When added to the direct effects, the result is the total effect of PJ on DP (see Hayes [2013] for references to applications of this model in other fields). The serial mediation model is summarized by the following hypotheses: Hypothesis 3: If negotiators adhere to PJ principles during the process, then DP will be enhanced when DJ principles are central to the agreement, and the agreements are stable. Hypothesis 3a: If negotiators adhere to PJ principles during the process, then reconciliation will be enhanced when DJ principles are central to the agreement, and the agreements are stable.

Control Variables An issue is that peace agreements do not occur randomly, and the factors that encourage parties to sign agreements are likely to affect the outcome of that agreement. Our focus on justice is one such factor. But, it may be only one type of variable that drives peace agreements and continues to shape implementation and long-term peace. Other variables, correlated with or independent of PJ and DJ, may also influence the agreements and the peace to follow. Correlated variables include trust perceptions, shared identities, and problem-solving processes. Although difficult to measure with archival materials, these variables are regarded as mechanisms that support PJ principles and, as such, are part of the explanation for the extent to which parties adhere to these principles (see Druckman and Wagner 2016). Relatively independent influences on the agreement and peace to follow include the CE, mediation, and media coverage. These variables are included in the analyses as controls and discussed as part of a set of contextual- and case-specific variables that may influence the progression from agreement to peace. Others include the goals of the parties to the conflict (Walter 2002; Joshi and Quinn 2015), whether a military stalemate exists (Walter 2002), and whether a democratic government exists (Joshi and Quinn 2015).

Durable Peace

323

Negotiating Context Three types of contextual variables are the CE, the extent of news coverage surrounding the negotiations, and PKOs.

CE Earlier research underscores the importance of the conf lict environment as an inf luence on implementing peace agreements. In this study, we create an index for the aspects of the CE that Downs and Stedman (2002) found to significantly inf luence the stability of the agreement in their sixteen cases: disposable resources, hostile neighboring states, and major power involvement.2 Following the earlier research, these external conf lict indicators would be expected to increase the difficulty of achieving stable or durable peace.3

News Coverage Media coverage has been shown to produce relatively inf lexible bargaining behavior in a number of experimental studies. Also referred to as audience effects, visibility of the talks to constituents made bargainers more reluctant to compromise than when they negotiated in private. These effects are heightened when the constituents were in an evaluative role that encouraged face saving by the bargainers (see Druckman, [1993] for a review of these studies). In this study, we coded the number of reports related to each case by three news agencies, BBC news, Reuters, and Agence France-Presse, during a three-month period prior to the signing of the peace agreement. Following the results obtained in the experiments, the impact of news coverage during the later stages would be expected to produce stronger effects than earlier coverage, which could affect both PJ (process) and DJ (outcome of agreement).

PKOs Civil war analyses suggest that the presence of peacekeepers can significantly inf luence whether violence resumes after the combatants sign a peace agreement. Downs and Stedman (2002), for example, assess the length of the PKO mission after the agreement is signed as an indicator of success. Walter (2002) identifies the importance of credible guarantees for ensuring the implementation of the terms of the agreement and finds that such guarantees are difficult to provide without outside assistance such as PKOs. Fortna (2008) finds that 75 to 85 percent of the civil wars she examined were less likely to slide back into civil war if PKOs were in place, based on the duration of the absence of war. Our fifty cases divide evenly between those that had PKOs and those that did not. This provides an opportunity to compare the two subsamples on the peace variables.

324 Process and Outcome Justice

Case Features The five case variables considered as control variables are type of agreement, mediator presence, length of negotiations, region, and decade.

Type of Agreement The relationship between the credibility of the commitments contained in a peace agreement and implementation of that agreement (Walter 2002; Mattes and Savun 2009) suggest that whether an agreement is partial or comprehensive will also affect its implementation. Disputants involved in drafting comprehensive agreements, for example, would have less uncertainty about the role of potential spoilers and fewer concerns about as yet unresolved issues, which would build confidence in the resulting peace. By contrast, the disputants in cases that produced partial agreements may have less confidence in the commitments contained in the agreement. Our fifty case sample includes thirty comprehensive and twenty partial agreements.

Mediator Presence Mediation was used in thirty-two of the fifty cases. Many of these cases involved some form of international mediation usually from several countries. A question asked is whether these interventions impacted on the postnegotiation peace processes. In this data set, the presence of mediators is a proxy for involvement of international actors in the peace process. Mediation was used considerably more often in partial (93 percent) than in comprehensive (63 percent) agreements. Thus, mediation was used in the more difficult cases, suggesting a negative relationship between mediation and postnegotiation peace processes.

Length of Negotiation The length of the negotiation process varied considerably across the cases. They ranged from 1 month (the cases of Niger 1995, Haiti 1993, Somalia 1997, Cote d’Ivoire 2003, and Angola 1989) to 49 months (the cases of Papua New Guinea 2001 and the Philippines 1996) with an average length of 10.5 months and a standard deviation of 12.8 months. Length may be an indicator of difficulty of the negotiation process, which, in turn, is a reflection of the CE: longer negotiations may occur for more intense conflicts. Indeed, the correlation between length and internal conflict is .37 (p < .02). Thus, DP may be less likely to follow long negotiations.

Region The fifty cases were divided among the regions as follows: Africa (twentythree), Asia (twelve), Europe (six), Central America and the Caribbean (six),

Durable Peace

325

the Middle East (three). The large representation of African cases ref lects the trend that Wallensteen, Melander, and Högbladh (2013) highlight: “some regions have been more prone than others to make peace agreements” (p. 129). Roughly 50 percent occur in African countries (see Elbadwi and Sambanis 2002). To the extent that these conf licts are more difficult to resolve, DP is less likely to follow the negotiations. The question turns on the implementation environments including regional power interest, spoilers, and valuable resources (Downs and Stedman 2002). Interestingly, the external CE index is significantly higher for the non-African than the African cases in this sample – bolstering the decision to consider whether regional differences play a role in the cases.

Decade The decades are distributed as post-2000 (fourteen cases), 1990s (twenty-four), and pre-1990s (twelve). About half of the civil wars occurred in the decade following the end of the Cold War. This sampling ref lects the higher incidence of peace agreements after 1989 than before. Wallensteen, Melander, and Högbladh (2013) note that this higher incidence coincides with the fact that, since 1989, there have been relatively few civil wars with clear-cut victories (see also Hironaka 2005). Of interest is the question whether the distribution of power within the international system relates to DP following the agreements: were the 1990s cases more difficult to resolve and less likely to result in DP than those that occurred before and after this decade?

Cases, Coding, and Analyses We code each case according to variables that are identified by our hypotheses and conduct statistical analyses to discover relations among them. Our coding incorporates information from data sets that peace agreement and civil war scholars often use – the UCDP and the Peace Accords Matrix (PAM). However, because our hypotheses require the coding of new independent (justice) and dependent (DP) variables, we chose fifty cases to represent the universe of available cases in these data sets. Because this is a novel data set, we describe the sampling and coding procedures in some detail followed by the statistical analyses performed.

Sampling Procedures While each peace agreement is unique to the conf lict it seeks to resolve and to the skills of the parties in combining, fractionating, or otherwise cobbling together their vision of a new direction for their country, similarities could stem from whether a comprehensive or partial agreement was reached. They could also stem from whether the agreements were signed during the same time period, when elements related to the international system would have

326 Process and Outcome Justice

similar inf luences. To ensure that the results from this study are not confined to a particular type of agreement or time period, a stratified random sampling procedure was used: comprehensive and partial agreements were represented as well as agreements struck across a variety of decades. These case differences are included in the analysis as control variables (see above). The cases range from the Bandaranaike–Chelvanayakam Pact in Sri Lanka, which was signed in 1957, to the Djibouti Agreement in Somalia, which was concluded in 2008. Given that cases in Africa are overrepresented in more recent decades, the selection by decade also ensured a more representative sampling of global regions. It is useful to address the question of a possible sampling bias. Our interest in negotiated agreements places constraints on the kinds of cases sampled. A willingness to negotiate an end to the war may dispose these parties toward cooperation. Although this possible disposition may skew our indicators of justice and peace, it does not preclude variation among them within the sample. As shown in Table 17.1, the standard deviations are relatively high for both the justice and peace variables. Thus, the sampled cases are suitable for exploring the hypothesized relationships with implications for a universe of negotiated peace agreements. Our question is what makes negotiated agreements engender enduring peace? Inferences from these results may be more difficult to draw for cases where negotiations failed or where the parties ended the conf lict without pursuing negotiations. Although the stratification criteria increased the representativeness of the sample, they also reduced the pool of eligible cases. One result was the inclusion of several cases from countries chosen as earlier cases (e.g., the cases of Angola 1989 and 2002). A question arises about the independence of these recurring cases. This question was addressed with a set of diagnostic steps. First, it is noted that only seven cases address similar conf licts in the same country; this consists of 14 percent of the total sample. Second, the average period between the same-country cases is eleven years; in all the recurring cases, the negotiators differed from the earlier to the later case. Third, the key indices (DJ, PJ, PJ + DJ, SA, DP) do not correlate significantly between the original and recurring cases. Fourth, the distinction between cases that were first or recurring peace agreements (based on the UCDP and United Nations [UN] Peacemaker data sets) produced nonsignificant p levels in multiple regressions in which the justice and control variables were regressed on DP and reconciliation. Fifth, p levels from multiple regressions run with a reduced sample of forty-three cases (excluding the recurring cases) are very similar to those obtained with the fifty case sample including those cases. Sixth, the critical values for the t ratios in the regression analyses change very little when the degrees of freedom are adjusted by subtracting the number of recurring cases in the sample. Thus, we conclude that lack of independence for these cases is not an issue. There is no need to adjust the sample. All analyses are performed on the full randomly chosen sample.4

Durable Peace

327

Data Sets Three data sets established the universe of possible peace agreement cases for this study. The PAM database consists of thirty-four civil war cases from 1989 to 2007, all of which achieved comprehensive agreements5 (all but two of these cases are also included in the UCDP data set). We randomly selected twentyfive of the PAM agreements for this study. We examined the UCDP data set to identify partial agreements and the UN Peacemaker data set to ensure variety among the decades. The total number of intrastate peace agreements examined in any one study depends on the criteria used for including cases. Joshi and Quinn (2015) analyzed 196 process or procedural, partial, and full agreements reached between 1975 and 2011. UCDP includes 216 cases signed between the warring parties active in an armed conf lict from 1975 to 2011 (see Högbladh 2012). The UN Peacemaker set includes 649 cases signed between 1947 and 2016. Thirty-two of the 34 PAM comprehensive cases overlap with the UCDP data set; the other 2 are included in the UN Peacemaker data set. Twenty-eight of the UCDP cases were coded as both partial agreements and agreements that were not signed by all warring parties.6 Online Appendix 2 lists the thirty comprehensive and twenty partial peace

TABLE 17.1 Descriptive Statistics.

Variables

Minimum

Maximum

Mean

Standard deviation

Durable peace Reconciliation Security institutions Governance institutions Economic stability Stable agreement PJ DJ Equality PJ + DJ CE Internal CE External CE Peacekeeping operation Mediation News Decade Region Length of negotiation Type of agreement

0 0 0 0 0 1 0 0.50 0 1 1.17 1 1 0 0 0.0 1 1 1 0.00

12 3 4 4 2 3 10 4.5 2 14 2 2 2 1 1 798 3 5 49 1

3.86 0.84 1.24 1.04 0.64 1.98 4.02 2.85 1.30 7.00 1.61 1.71 1.50 0.50 0.68 144.58 1.96 2.08 11.78 0.62

3.34 1.02 1.27 1.36 0.63 0.85 2.54 1.09 0.42 2.99 0.17 0.25 0.29 0.51 0.47 187.56 0.73 1.28 13.68 0.49

Note: CE = conf lict environment; PJ = procedural justice; DJ = distributive justice.

328 Process and Outcome Justice

agreements examined in this research. The UN Peacemaker subset selected for the current study includes five comprehensive and five partial agreements from 1957 to 1989. Overall, the UCDP data set is well represented in this study, with 4 of the 10 UN Peacemaker cases being included in the UCDP data set. Comprehensive agreements came from the PAM and the UN Peacemaker data sets. Partial agreements came from UCDP and UN Peacemaker. A question is whether our case selection confounds type of data set with type of agreement. We took steps to avoid this problem. Relevant here is the type of information available from the different sources. The PAM cases are a subset of UCDP: as noted above, all but two of the sampled PAM cases are also in UCDP. This overlap enabled our coders to use information from both (as well as other) sources in making their judgments7 (the coding sheets provide all the sources used; see note 8 for availability).

Coding the Justice and Peace Variables The following sections present the coding procedures used for each of the variables: PJ, DJ, SA, and DP. These procedures are summarized in Online Appendix 3.8

PJ Primary and secondary accounts of each negotiation process were evaluated to identify instances in which each of four PJ elements played a role in the negotiations: transparency, fair representation, fair treatment, and voluntary agreement. For those instances in which the justice component had a positive inf luence on the talks, either because relevant information was shared, stakeholders were included in the talks and had the opportunity to make their voice heard, or the outcome was accepted voluntarily, the PJ variable was assessed for whether it had a (3) highly significant, (2) important, or (1) marginal inf luence on the process. Coders were trained to distinguish among the three judgments in terms of the extent to which the instance was closely adhered during the process. For each of the four PJ components, we add the relevant scores and develop an average per component. We add the four averages to develop a “PJ satisfied” score. A graduate student research assistant coded each case according to guidelines established for the study. The principal investigators reviewed the data for nineteen randomly selected cases and developed independent coding assessments, which matched 87.8 percent of the primary coder’s assessments. The calibration procedure concluded with a discussion about those codes that differed, and adjustments to the coding were made. Using the 2000 case of Burundi, we provide examples in Online Appendix 3 of how each PJ component was coded.

DJ A second research assistant assessed the provisions in the signed agreements for each of the fifty cases according to four DJ principles: equality, proportionality, compensation, and need. DJ was judged in terms of its significance or centrality to the

Durable Peace

329

agreement. Each article in each peace agreement was assessed for whether it represented one of the four DJ principles. The role played by the principle was assessed on a scale that distinguished among highly significant (1.5), very important (1.25), important (1), less important (.75), or marginal (.5) role. Coders were trained to distinguish among these points on the scale in terms of the extent to which the textual passage referred to a central point in the conf lict. The intervals between the highly significant, important, and marginal codes are proportional to the intervals between the codes for the comparable PJ measures. However, our practice coding with the DJ measure suggested that we could make finer distinctions. Thus, two additional categories – very important and less important – were added to the coding scheme. The numerical values for each principle were averaged for the final DJ score. A quarter of the cases were randomly selected to calibrate the coding. Intercoder agreement on assigning codes to the DJ categories was 84.8 percent, and the process concluded with a discussion about differences and adjustments to the coding. Examples of coding decisions for each DJ principle are shown in Online Appendix 3 for the 1985 Uganda case. As noted in the section on hypotheses, we also add the scores for PJ and DJ to calculate a global justice index.

SA SA refers to the durability of the peace agreement during the first five years after it was signed. As noted above, earlier studies found this variable to play a mediating role between PJ and the reconciliation component of DP as well as between equality and DP. However, the coding was based on a two-year assessment of whether large-scale violence had been brought to an end while peacekeeping implementers are present and whether war was terminated on a self-enforcing basis, so that implementers can leave without war rekindling (Downs and Stedman 2002). The current study introduces a measure that is closer to that used in the civil war termination literature: whether conflict subsides for five years after the peace agreement is signed. Assessment of this variable draws from the UCDP data set9 regarding whether the peace agreement remained in force for five years after the agreement was reached and whether the conflict was active at the five-year mark. Stability coding was performed using a three-step scale: (3) cases that remained in force and for which conflict was not active at the five year mark, (2) cases for which conflict events were recorded during the first five years but that remained in force, and (1) cases for which sustained active conflict resumed and the agreement was no longer in place at the conclusion of five years. A second coder’s assessments differed from the first coder’s assessments for only two cases, for a 96 percent rate of agreement.

DP We assessed DP by evaluating four components during the eight years immediately following the signing of an agreement: extent of reconciliation, reform of security institutions, reform of governance institutions, and economic growth/

330 Process and Outcome Justice

stability. The theoretical and empirical bases for this index come primarily from Diehl and Druckman’s (2010) peace operation success framework and Paris’s (2004) examination of peacebuilding efforts after intrastate conf lict but also draw from the civil war termination research literature. Several reasons led to the decision to use an eight-year period for assessing DP. One is to allow sufficient time for institutional change, based in part on O’Reilly’s (2014) assessment that at least six years are required for economic institutions to stabilize following conf licts. We judge that eight years are the minimum needed to build institutions necessary for a lasting peace. We also wanted to ensure that the DP period would be several years longer than the five-year SA assessment of whether conf lict had subsided and the agreement remained in place. The reconciliation component encompasses activities that seek to redress past offenses. This includes efforts to address crimes and establish truth commissions, but Thoms, Ron, and Paris (2010) find that there is insufficient empirical evidence in the transitional justice literature to show that trials and truth commissions have state-level effects. We therefore also include reception and reintegration, community relations, and provisions for refugees in reconciliation (e.g., Balcells and Justino 2014; Hayner 2011). The growing field of transitional justice focuses on these and related strategies to address the legacy of human rights abuses with the objective of contributing to the foundation of societal peace.10 Institution building provides forward-looking components for the DP index. Paris (2004) highlights that democratic politics and capitalist economics “depend on public institutions to uphold rules, to maintain order, to resolve disputes impartially, and to regulate behavior incompatible with the preservation of market democracy itself ” (p.  205). We organize such institution-building efforts into three parts: security, governance, and economic institutions. Security institutions incorporate the extent to which military and police reform take place, demobilization and disarmament proceed, violent crime decreases, and ceasefire efforts hold within the country (Diehl and Druckman 2010; UN 2000). The governance institutions component encompasses power-sharing arrangements, rule of law and legal reform, electoral reform and holding elections, and human rights protection (Cousens, Kumar, and Wermester 2001). The economic component consists of conditions and actions that lead to economic stability, including measures of income equality, changing living standards, support programs from lending institutions, and the pursuit of conditions and policies for sustainable economic growth (O’Reilly 2014). To construct an index for the DP variable, a research assistant reviewed the information in the PAM and UCDP data bases, along with other sources, for accounts of each country’s experience after the peace agreement was signed regardless of the provisions incorporated into the peace agreement itself. The cases were coded for whether each of the sixteen DP elements was mostly or very successful, had limited success, or failure. The coder was trained to distinguish among these choices in terms of the extent to which it was realized in practice. These assessments were added to develop the final DP counts score,

Durable Peace

331

which ranged from 0 successful DP elements to 16 successful DP elements for each case. We also added these assessments for each of the four DP components to develop separate codes for reconciliation, security institutions, governance institutions, and economic stability. A calibration procedure was used to assess the replicability of the scores, with a second coder making independent assessments based on the case data. Calibration assessments consisted of comparing the assessments made by the first and second coders. They agreed on 86 percent of their judgments. The differences were reconciled by discussion. Examples of coding decisions are shown in Online Appendix 3 for the 2005 Indonesia case.

Statistical Analyses Direct and indirect effects were evaluated with regression analyses. The DP index and its components, which are the DVs for these analyses, were measured on an equally appearing interval scale.11 Bivariate correlations were also calculated for all pairs of variables (see Online Appendix 4). Direct effects – referred to as vertical modeling – were assessed with multiple regressions that included the justice variables (PJ, DJ, and combined PJ + DJ) and the eight control variables. The relatively large number of sampled cases makes it feasible to include these variables in our analyses: a favorable ratio of cases (fifty) to variables (ten) augurs well for including both IVs and control variables in the same multiple regression model (see Stevens [2012] on the ratio of cases to variables in multivariate statistical analysis). An attempt was made to address multicollinearity by running separate regressions with each justice variable. The results provide a vertical ordering of the variable effect sizes. The indirect effects or horizontal model was evaluated with regression-based boot-strapping procedures following Hayes (2013). These analyses examine the extent to which a causal variable (PJ) inf luences an outcome (DP) through one or more mediator variables (DJ, SA). The results provide a horizontal timeordered path of the variables.

Results Descriptive statistics for each variable coded in this study are shown in Table 17.1 above. The results are organized by the hypotheses beginning with the direct effects.

Direct Effects Hypothesis 1: More adherence to PJ principles during the negotiation process will lead to stronger DP following the peace agreement. When included in a multiple regression that includes the eight control variables and SA, PJ predicts DP ( β = .41, t = 2.79, p < .008). None of the control variables predicts DP (see Table 17.2).12 These findings provide strong support for the first

332 Process and Outcome Justice

hypothesis: PJ predicts DP following the agreement better than other contextual and case variables.13 Hypothesis 1a: More adherence to PJ principles during the negotiation process will lead to stronger reconciliation following the peace agreement. When included in a multiple regression equation that includes the eight control variables and SA, PJ predicts reconciliation ( β = .52, t = 3.34, p < .002). None of the control variables predicts reconciliation (see Table 17.2).14 Thus, PJ is a strong predictor of reconciliation following the agreement and is a considerably better predictor than other contextual and case variables.15 Hypothesis 1b: More adherence to principles of fair treatment and transparency during the negotiation process will lead to stronger DP following the peace agreement. When included in a multiple regression that includes the eight control variables and SA, fair treatment does not predict DP ( β = .223, t = 1.55, p < .13). None of the control variables predicts DP. When included in a multiple regression that includes the eight control variables, transparency strongly predicts DP ( β = .487, t = 2.87, p < .009). Only one of the control variables, comprehensive versus partial agreements, also predicts DP ( β = .421, t = 2.37, p < .03). The combined variable (fair treatment + transparency) also strongly predicts DP when included in a regression with the eight control variables ( β = .485, t = 2.42, p < .025). The only significant predictor among the control variables is comprehensive versus partial agreements (p < .04). Neither fair representation

TABLE 17.2 Procedural Justice and Control Variables Regressed on Reconciliation and DP.

 

Reconciliation

  Variable

β coefficients t

β Significance coefficients t

Significance

PJ Type of agreement Conf lict environment index Peacekeeping operation News coverage Mediation Length of negotiation Region Decade Stable agreement

.52 .19 -.12 -.10 -.08 -.19 -.07 -.08 .11 .40

.002 NS NS NS NS NS NS NS NS .01

.008 NS NS NS NS NS NS NS NS .002

3.34 1.34 -0.91 -0.79 -0.53 -1.30 0.45 -0.50 0.71 2.70

Note: PJ = procedural justice; DP = Durable Peace.

DP

.41 .18 -.02 -.16 -.004 .08 .05 .10 -.10 .46

2.79 1.37 -0.14 -1.37 -0.03 0.62 0.33 0.67 -0.67 2.40

Durable Peace

333

nor voluntary agreements predicted DP, in the single or multiple regressions including the control variables. Hypothesis 1c: More adherence to principles of fair treatment and transparency during the negotiation process will lead to stronger reconciliation following the peace agreement. When included in a multiple regression with the eight control variables and SA, fair treatment predicted reconciliation at the .07 level while transparency predicted reconciliation at the .03 level. The only significant control variable was comprehensive versus partial agreements in the transparency analysis (p < .03). The combined fair treatment + transparency analysis predicted reconciliation in multiple regression analyses (at .04). Neither fair representation nor voluntary agreements or their combination predicted reconciliation in the multiple regressions. Results obtained from a factor analysis of these variables showed that DP, reconciliation, fair treatment, and transparency loaded on a first factor. Only voluntary agreement loaded on a second factor. Fair representation did not load significantly on either factor. Hypothesis 2: The more significant DJ principles are in the negotiated agreement the higher the assessments of DP following the agreement. When included in a multiple regression equation that includes the eight control variables and SA, DJ is a weak predictor of DP ( β = .322, t = 1.61, p < .12). None of the eight control variables predicts DP. DJ does not predict reconciliation when included in a multiple regression with the eight control variables. Comprehensive versus partial agreements is the only control variable that predicts reconciliation (p < .03). DJ does predict the development of security institutions when included in a regression equation with the eight control variables ( β = .511, t = 2.63, p < .01). It does not predict government reform or economic growth. These results provide limited support for the hypothesis: DJ is a strong predictor of the development of security institutions and, although not significant, is a stronger predictor of DP than any of the control variables. Hypothesis 2a: The more significant the equality principle is in the negotiated agreement the higher assessments of DP following the agreement. The DJ principle of equality is not a significant predictor of DP, or any of the four components of this index in the multiple regressions that include the eight control variables. None of the control variables predicts DP or reconciliation. The multiple regression results do not support this hypothesis. When included in a multiple regression equation with the eight control variables, the PJ + DJ index predicts DP (β = .45, t = 3.14, p < .003; see Table 17.3). The

334 Process and Outcome Justice

combined index also predicts reconciliation when included in a multiple regression equation with the controls (β = .53, t = 3.41, p < .002). None of the controls predicts DP or reconciliation.16 The combined PJ + DJ index also predicts security reform at p < .01. These results provide strong support for this hypothesis. The combined index performs better than PJ with βs of .50 and .36, respectively. It performs better than DJ with βs of .50 and 32, respectively. Thus, PJ makes a somewhat stronger contribution to the relationship between the combined justice index and DP.17,18

Indirect Effects Hypothesis 3: If negotiators adhere to PJ principles during the process, then DP will be enhanced when DJ principles are central to the agreement and the agreements are stable. This hypothesis is an evaluation of the serial mediation model described above. The model posits two mediators: DJ and SA. A bootstrapping procedure was used to ascertain whether DJ and SA mediated the effects of PJ on DP. The procedure was run with 1,000 resamples and a 95 percent confidence interval (see Hayes 2013, model 6). First, we evaluated the direct effects of each mediating variable. The results showed that SA (p < .0001) and PJ (p < .001) were strong predictors of DP. These findings are consistent with the direct effects evaluated in the previous section. Second, we examined the indirect effects of these TABLE 17.3 PJ + DJ and Control Variables Regressed on Reconciliation and DP.

 

Reconciliation

  Variable

β Coefficients

PJ + DJ Type of agreement Conf lict environment index Peacekeeping operation News coverage Mediation Length of negotiation Region Decade Stable agreement

DP t

Significance

β Coefficients

t

Significance

.53 .17

3.41 1.16

.002 NS

.45 .15

3.14 1.16

.003 NS

-.10 -.08

-0.76 -0.66

NS NS

-.01 -.16

-0.07 -0.37

NS NS

-.07 .18 .07

-0.44 -1.22 -0.45

NS NS NS

0 .09 .04

-.01 .10 .36

-0.03 0.66 2.42

NS NS .02

.16 -.10 .43

Note: PJ = procedural justice; DJ = distributive justice; DP = durable peace.

0.002 0.66 0.27 1.12 -0.73 3.16

NS NS NS NS NS .003

Durable Peace

335

variables. The best model consisted of a path from PJ to DP through DJ and SA. Statistical significance is indicated by the confidence intervals for this path (lower, .0050, upper, .2168), where zero is not included in the interval. The serial path takes the following form: PJ → DJ → SA → DP This path shows a linear relationship between negotiation process (PJ), outcomes (DJ), and short-term (SA) and long-term (DP) peace. SA and DP develop from adherence to justice principles during the negotiations and in the negotiated outcome. Neither of two shorter paths (PJ → DJ → DP or PJ → SA → DP) were significant; zero was included in the range from the lower to upper confidence interval.19 These results provide strong support for the serial mediation model summarized by Hypothesis 3. Hypothesis 3a: If negotiators adhere to PJ principles during the process, then reconciliation will be enhanced when DJ principles are central to the agreement and the agreements are stable. Using the same bootstrapping procedure, we first evaluated the direct effects of each mediating variable. The results showed that PJ (p < .008) and SA (p < .01) were strong predictors of reconciliation. The DJ–reconciliation relationship reached border-line significance at p < .09. Second, we evaluated the indirect effects of these variables. Two significant models were found. One consisted of the following single-mediator path: PJ → DJ → reconciliation. Statistical significance is indicated by the confidence intervals for the path (lower, .0002, upper, .0846); zero is not included in the interval, indicating statistical significance. Another included SA in the path: PJ → DJ → SA → DP reconciliation. The confidence intervals for this indirect model did not include zero (lower, .0004, upper, .0440), thus indicating statistical significance. This path provides support for the serial mediation model summarized by this hypothesis.

Discussion The discussion is organized into three parts: justice and peace, context, and the bridge from micro- to macro-processes.

Justice and Peace The two-stage modeling results make evident that justice matters in the peace process following the termination of civil wars. The vertical (direct) analyses show that justice variables impact strongly on DP and reconciliation in the context of a larger CE. The horizontal (indirect) analyses show that PJ initiates

336 Process and Outcome Justice

a chain reaction that travels through outcomes, cessation of violence, and DP. These findings suggest directions for theory building. Justice is a meaningful psychological dynamic that could be critical if peace is to last. Both PJ in the process and DJ in the outcome contribute to reconciliation and DP in post–civil war societies. The strongest vertical effects occur for a combined justice variable that consists of the sum of PJ and DJ. These findings contrast with those obtained for a variety of contextual and case variables. These variables do not predict DP or reconciliation, rendering the justice–peace relationship robust. Further insights into the relationship are provided by the results of the serial mediation model shown in Figure 17.1. These results identify a horizontal path from process (PJ) to peace (DP) through outcome (DJ) and cessation of violence (SA). An important finding is the strength of PJ effects on DP and reconciliation. Interestingly, these effects are stronger than those found for DJ. It appears that conduct during the negotiation process is a key to lasting peace following a civil war. The analyses conducted on the components of PJ provide some clues about why the effect occurred. Stronger effects on DP and reconciliation were obtained for the fair treatment and transparency components of PJ than for fair representation and voluntary decision-making. Fair treatment refers to the way negotiators conduct the negotiation, particularly with regard to the opportunity to be heard and a show of consistency and impartiality during the interactions. Transparency refers to openness and accessibility of information about decisions, activities, and actions. When adhered to, these PJ components convey courtesy and respect from one negotiating delegation to the other. These findings are consistent with results obtained by Hollander-Blumoff (2017) in a laboratory simulation of a legal negotiation. She showed that PJ judgments depended on perceptions of courtesy and respect. Taken together, the findings from the two studies support a relational or group values model of PJ as described by Lind and Tyler (1988). Support for this model from two settings bolster the importance of relational concerns in negotiation. By their conduct, legal and peace negotiators inf luence the quality of agreements. In this study, more comprehensive agreements occurred when negotiators adhered to the two social conduct components of PJ: .47 for fair treatment and comprehensive agreements, .36 for transparency, and .39 for the combination of these components. The way that relationships are developed between adversaries during negotiation carry forward to how the parties interact in the years following the talks. They carry back to the motivations for developing those relationships prior to and during the process. Less is known about the etiology of social conduct: what are the underlying mechanisms? Clues come from the trust literature, particularly the work on ladders of trust (Irmer and Druckman 2009). Trust perceptions and conduct are intertwined: as trust moves up the ladder from calculus-based to shared group identity, social conduct improves, which in turn increases trust.

Durable Peace

337

Working together, the perceptual and behavioral processes improve and sustain the relationship. Social conduct and type of agreement come into play as well with regard to the reconciliation component of DP. Reconciliation is predicted from fair treatment + transparency and comprehensive versus partial agreements. These findings suggest that the inclusion of uncertainty-reducing measures may contribute to peace (see Mattes and Savun 2009; Joshi, Quinn, and Regan 2015). Confidence may be bolstered when the negotiating opponent’s behavior signals improved relationships, and there are no loose ends in the agreement. It is manifest in the reconciliation activities coded for our sample of peace agreements. The strongest impact of the social conduct PJ variables was on community reconciliation. Comprehensive agreements were particularly beneficial for the reparations and refugee provisions.

Context This study fills gaps in our understanding of the impacts of the negotiation process on peace. The importance of relational dynamics is evident in our regression analyses with controls: strong effects occur for the social conduct components of PJ when controlling for several contextual variables examined in earlier civil war termination studies (CE, PKOs, mediation) as well as other features of the cases such as news coverage and region. Thus, building relationships between former combatants is a key to establishing a culture of peace (Bar-Tel 2008). Relationships are particularly important for postagreement reconciliation but also inf luence security reform. Interestingly, they have no impact on economic development, supporting the earlier analyses of post–civil war change by O’Reilly (2014).20 By calling attention to certain features of the negotiation process, the results of this study contribute new knowledge to the literature on civil war termination and the consequences of peace agreement. They do not, however, suggest that we have identified a sovereign variable that is the key to peace. Rather, the findings broaden our perspective on the role of peace processes in civil war termination. The key is found in the sequence of stages that proceed from the CE through negotiations to ending wars and establishing peace. For example, more transparent negotiations may lead to increased confidence and less fear in implementing power-sharing provisions and engaging in reconciliation activities (Joshi and Quinn 2015). But the sequence also has a circular feature. DP may have a shelf life determined by the alternation of rises and dips in relationships between the former adversaries. This is well illustrated by the course of interactions between the Mozambican National Resistance (RENAMO) and the Mozambique Liberation Front (FRELIMO) following the 1992 peace agreement in Mozambique. In force for two decades, the agreement has been violated in recent years. RENAMO has become increasingly disenfranchised by the centralization of power held by FRELIMO. Moving in the direction of more dynamic models is a challenge for civil war and peace agreement researchers.

338 Process and Outcome Justice

Bridging Micro and Macro Processes Balcells and Justino (2014) raised the hope that more studies of civil wars would bridge micro- and macro-levels of analysis. This study is one such attempt. The serial mediation findings show a time-ordered or causal path from micro- to macro-level processes: PJ connects to DP and reconciliation through DJ and SA. In this section, we search for some possible reasons and ask questions about why this connection occurs. The mediating effect of SA is important. This variable stands out as a bridge from the budding relationships developed during the negotiation process to affirming those relationships during the reconciliation phase.21 SA preceded and predicted DP. However, the transition from SA to DP did not eventuate in eleven of the thirty-two cases that achieved SA. Justice played a role in this transition. The twenty-one transition cases were above the mean in PJ; the eleven nontransition cases were below the mean. The transition from SA to DP was due largely to effects on reconciliation and, specifically, on the community relations component of this index: this was the only aspect of reconciliation that correlated significantly (.51) with SA. Defined in this study as collaborative activities between the former combatants, this was also the key bridge in the micro- (e.g., recruitment) to macro-level (civil war duration) statistical analyses of political violence performed by many of the authors in the special issue edited by Balcells and Justino (2014). In our study, collaborative activities occur at the end of a path that travels from negotiation processes (at the micro-level) through to implementation of the agreements (at a meso-level). Another transition takes place during the negotiation. This goes from destructive to constructive conf lict processes. Of interest is the question of how and when this transition occurs. Framing the question in terms of the PJ indicators, we ask whether there are observable shifts in social conduct as the negotiation process unfolds. Answers to this question may come from timeseries analyses of changes in the fair treatment and transparency codes. The analyses would also reveal how the changes occur. They could be gradual, as when trust accumulates over time leading to comprehensive agreements to end violent wars (Irmer and Druckman 2009). Or they can be abrupt, as shown in analyses of critical moments in negotiation (Leary 2004). Whether gradual or abrupt, these relational changes may be critical for sustaining peace through the stages of the serial mediation path. Understanding the dynamics of both kinds of transitions – from SA to DP and from destructive to constructive conf lict – will increase the chances for agreements that bring peace to nations ravaged by civil wars.

Authors’ Note The complete data set is available at http://jcr.sagepub.com/.

Durable Peace

339

Notes 1 The large number of cases assembled for this study provides an opportunity to evaluate the generality of findings obtained in the three earlier studies. Further, this data set allows for an evaluation of a host of control variables derived from both the negotiation and the civil war termination literatures. 2 Downs and Stedman (2002) found that two of their four significant conflict environment (CE) variables are correlated, spoilers and hostile neighbors. As a result, our CE index includes only hostile neighbors. 3 We also create a new index for the internal CE. It consists of three parts, all assessed at the point that negotiations on the peace agreement began: months of violence, average number of deaths per month, and number of internally displaced persons. These indicators are salient features for assessing the situation of a war-torn country’s population prior to negotiation, and the degree to which the conflict affected daily life and provoked violent animosity among warring sides. More conflictual environments (cases in the highest quartile for the three indicators) would be expected to be associated with cases that are less likely to achieve stable and durable peace (DP). 4 One approach for addressing case interdependence is to adjust the standard errors in regression analyses. This decision is contingent on the outcomes of diagnostic tests on the cases in question. The outcomes of the diagnostics reported in this paragraph preclude the need for adjustments. 5 Peace Accords Matrix (PAM) defines “comprehensive” cases as pacts for which the major parties to the conflict were involved in the negotiations that produced the agreement, and the substantive issues underlying the dispute were included in the negotiation (Joshi, Quinn, and Regan 2015). 6 These twenty-eight cases were identified in the Uppsala Conflict Data Program (UCDP) data set (Högbladh 2012) based on two types of comments: about the signatories of the agreement and whether it was signed by all or some parties (c-sign); and about the type of agreement and various regulations (pa). 7 Our decision to select cases from PAM was based on our observation that PAM implementation data are better organized and more extensive than UCDP. However, information from both sources were available to the coders. 8 The coding decisions made for each variable and sources used for the coding of each of the cases are available from the authors. Coding guidelines are also available from the authors. 9 Forty-two of the cases are included in the UCDP data set. For the remaining eight cases, a research assistant reviewed case data and assessed this variable. 10 Transitional justice is conceived as part of the reconciliation component of our DP index. Our focus on the negotiation process and outcome leads us to concentrate on the impacts of procedural justice (PJ) and distributive justice (DJ), both considered as independent variables, on peace processes following the agreement, considered as dependent variables (DVs). 11 This assumption permits the computation of linear multiple regression and is essential for performing single or double mediation analyses. An alternative would have been to treat DP as a categorical variable and perform logit regressions. We did analyze the data with logits and found very similar results. Thus, our conclusions would have been virtually the same for the direct effects, but we would have forfeit the opportunity to examine indirect effects. 12 Results are similar when the two CE index components, internal and external CE, are included in the regression: PJ predicts DP at p < .03 when internal conflict is included in the model; PJ predicts DP at p < .007 when external conflict is included in the model; neither conflict index regresses significantly on DP. Similar results are obtained when reconciliation is the DV (see Hypothesis 1a). 13 Partial correlations were also computed controlling for each of the contextual and case variables in PJ (DJ) –DP and PJ-reconciliation relationships. These variables had minimal impact on the bivariate justice–peace correlations.

340 Process and Outcome Justice

14 Calculation of an odds ratio representing the chances that a particular outcome will occur (high reconciliation) given a particular exposure (high PJ) compared to the odds of the outcome occurring in the absence of that exposure (low PJ) shows that the odds are just under 3:1 that reconciliation will occur when PJ principles are adhered to during the negotiation process. 15 PJ predicted security reform at p < .05. It did not predict governance reform or economic growth. Decade predicted economic growth. None of the other control variables predicted any of the DP components. 16 Odds ratio calculations show that the odds are 4:1 that DP will occur when negotiators adhere to both these justice principles. Similar odds are obtained for reconciliation. 17 Interactive effects between PJ and DJ were also evaluated. Referred to as compensatory effects, this hypothesis posits that PJ has a stronger impact on DP when parties receive unfavorable outcomes; it also posits that DJ effects are stronger when negotiators violate PJ principles. This hypothesis was not supported. 18 Economic growth following the peace agreements is flat with an average score across the fifty cases of .64 on a scale that ranges from 0 to 3. The most change occurs on support from lending institutions (change occurs on twenty-five of the fifty cases) with either no change (improved living standards) or marginal change (policies for sustaining economic growth) on the other components. Factor analysis results show that the economic growth variable is isolated from the other variables used in the analysis: it loads strongly only on a factor that accounts for about 10 percent of the total variance explained in the correlation matrix. The other DP components have strong loadings on a factor that explains 50 percent of the total variance. Thus, economic growth did not occur during the recovery period and was not correlated with the other components of DP. Nor was it correlated with the justice variables. Correlations were also computed with a three-component DP index that excludes economic growth. The pattern of correlations between this index, stable agreement (SA), and the justice variables is only marginally higher than those reported with the four-component index. The two indexes are very highly correlated (r = .98). 19 The shorter paths (single mediating variables) are stronger when evaluated with Sobel’s z statistic. DJ is shown to mediate the relationship between PJ and DP at a borderline level of significance (z = 1.54, p < .06, one-tailed) as well as between PJ and reconciliation (z = 1.43, p < .08, one-tailed). SA also mediates the relationship between DJ and DP (z = 1.65, p < .05, one-tailed). 20 This may be due to the time needed for the development of institutions that would facilitate the investments needed for economic recovery following civil wars. The eightyear time period used in this study may be insufficient. If so, a longer period for observing economic development is warranted for future studies. 21 This bridge is reinforced by successful peacekeeping missions, which are also shown to be influenced by the justice variables. Adherence to principles of PJ for the sample of peacekeeping operation cases led to shorter duration missions followed by DP. Conceivably, the cooperation encouraged by PJ led negotiators to provide consent for the missions, which, in turn, increased the chances for both efficient and effective missions.

References Adams, J. Stacy. 1965. “Inequity in Social Exchange.” In Advances in Experimental Social Psychology, edited by L. Berkowitz, 267–99. New York: Academic Press. Albin, Cecilia, and Daniel Druckman. 2012. “Equality Matters: Negotiating an End to Civil Wars.” Journal of Conflict Resolution 56 (1): 155–82. Balcells, Laia, and Patricia Justino. 2014. “Bridging Micro and Macro Approaches on Civil Wars and Political Violence: Issues, Challenges, and the Way Forward.” Journal of Conflict Resolution 58 (8): 1343–59.

Durable Peace

341

Barry, Brian. 1996. Justice as Impartiality. Oxford, UK: Oxford University Press. Bar-Tel, Daniel. 2008. “Reconciliation as a Foundation of Culture of Peace.” In Handbook on Building Cultures of Peace, edited by Joseph de Rivera, 363–77. Berlin-Heidelberg, Germany: Springer. Binningsbo, Helga Malmin, Cyanne E. Loyle, Scott Gates, and Jon Elster. 2012. “Armed Conf lict and Post-civil War Justice, 1946–2006: A Dataset.” Journal of Peace Research 49 (5): 731–40. Casper, Jonathan D., Tom R. Tyler, and Bonnie Fisher. 1988. “Procedural Justice in Felony Cases.” Law and Society Review 22 (3): 483–507. Cousens, Elizabeth M., Chetin Kumar, and Karin Wermester, eds. 2001. Peacebuilding as Politics: Cultivating Peace in Fragile Societies. Boulder, CO: Lynne Rienner. Deutsch, Morton. 1985. Distributive Justice: A Social-psychological Perspective. New Haven, CT: Yale University Press. Diehl, Paul F., and Daniel Druckman. 2010. Evaluating Peace Operations. Boulder, CO: Lynne Rienner. Downs, George, and Stephen J. Stedman. 2002. “Evaluation Issues in Peace Implementation.” In Ending Civil Wars: The Implementation of Peace Agreements, edited by Steven J. Stedman, Donald Rothchild, and Elizabeth M. Cousens, 43–70. Boulder, CO: Lynne Rienner. Druckman, Daniel. 1980. “Social-psychological Factors in Regional Politics.” In Comparative Regional Systems, edited by Werner J. Feld and Gavin Boyd, 18–55. New York: Pergamon. Druckman, Daniel. 1993. “The Situational Levers of Negotiating Flexibility.” Journal of Conflict Resolution 37 (2): 236–76. Druckman, Daniel, and Cecilia Albin. 2011. “Distributive Justice and the Durability of Peace Agreements.” Review of International Studies 37 (3): 1137–68. Druckman, Daniel, and Lynn Wagner. 2016. “Justice and Negotiation.” Annual Review of Psychology 67:387–413. Elbadwi, Ibrahim, and Nicholas Sambanis. 2002. “How Much War Will We See? Explaining the Prevalence of Civil War.” Journal of Conflict Resolution 46 (3): 307–34. Fortna, Virginia Page. 2008. Does Peacekeeping Work? Princeton, NJ: Princeton University Press. Hartzell, Caroline A., and Matthew Hoddie. 2003. “Institutionalizing Peace: Power Sharing and Post-civil War Conf lict Management.” American Journal of Political Science 47 (2): 318–32. Hartzell, Caroline A., and Matthew Hoddie. 2008. Crafting Peace: Power-sharing Institutions and the Negotiated Settlement of Civil Wars. University Park: Penn State University Press. Hayes, Andrew F. 2013. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-based Approach. New York: The Guilford Press. Hayner, Priscilla B. 2011. Unspeakable Truths: Transitional Justice and the Challenge of Truth Commissions, 2nd ed. New York: Routledge. Heald, David. 2006. “Varieties of Transparency.” Proceedings of the British Academy 135: 25–43. Hironaka, Ann. 2005. Neverending Wars. Cambridge, MA: Harvard University Press. Högbladh, Stina. 2012. “Peace Agreements 1975–2011: Updating the UCDP Peace Agreement Dataset.” In States in Armed Conflict 2011, edited by Therése Pettersson and Lotta Themnér, 39–56. Uppsala, Sweden: Department of Peace and Conf lict Research Report 99, Uppsala University.

342 Process and Outcome Justice

Hollander-Blumoff, Rebecca. 2017. “Formation of Procedural Justice Judgments in Legal Negotiation.” Group Decision and Negotiation 26 (1): 19–43. Hollander-Blumoff, Rebecca, and Tom R. Tyler. 2008. “Procedural Justice in Negotiation: Procedural Fairness, Outcome Acceptance, and Integrative Potential.” Law & Social Inquiry 33 (2): 473–500. Hume, Cameron. 1994. Ending Mozambique’s War: The Role of Mediation and Good Offices. Washington, DC: United States Institute of Peace Press. Irmer, Cynthia, and Daniel Druckman. 2009. “Explaining Negotiation Outcomes: Context or Process?” Negotiation and Conflict Management Research 2 (3): 209–35. Joshi, Madhav, and J. Michael Quinn. 2015. “Is the Sum Greater than the Parts? The Terms of Civil War Peace Agreements and the Commitment Problem Revisited.” Negotiation Journal 31 (1): 7–30. Joshi, Madhav, J. Michael Quinn, and Patrick M. Regan. 2015. “Annualized Implementation Data on Comprehensive Intrastate Peace Accords, 1989–2012.” Journal of Peace Research 52 (4): 551–62. Kabanoff, Boris. 1991. “Equity, Equality, Power, and Conf lict.” Academy of Management Review 16 (2): 416–41. Leary, Kimberlyn. 2004. “Critical Moments in Negotiation.” Negotiation Journal 20 (2): 143–45. Leventhal, Gerald S., Jurgis Karuza, and William R. Fry. 1980. “Beyond Fairness: A Theory of Allocation Preferences.” In Justice and Social Interaction, edited by Gerold Mikula, 167–228. New York: Springer. Lind, E. Allan, and Tom R. Tyler. 1988. The Social Psychology of Procedural Justice. New York: Plenum. Lind, E. Alan, Tom R. Tyler, and Yuen J. Huo. 1997. “Procedural Context and Culture: Variation in Antecedents of Procedural Justice Judgments.” Journal of Personality and Social Psychology 73 (4): 767–80. Mattes, Michaela, and Burcu Savun. 2009. “Fostering Peace after Civil War: Commitment Problems and Agreement Design.” International Studies Quarterly 53 (3): 737–59. O’Reilly, Colin W. 2014. “Investment and Institutions in Post-civil War Recovery.” Comparative Economic Studies 56(1): 1–24 Paris, Roland. 2004. At War’s End: Building Peace after Civil Conflict. New York: Cambridge University Press. Pruitt, Dean G., Robert S. Pierce, Neil B. McGillicuddy, Gary L. Welton, and Lynn M. Castrianno. 1993. “Long-term Success in Mediation.” Law and Human Behavior 17 (3): 313–30. Quinn, J. Michael, T. David Mason, and Mehmet Gurses. 2007. “Sustaining the Peace: Determinants of Civil War Recurrence.” International Interactions 33 (2): 167–93. Stevens, James P. 2012. Applied Multivariate Statistics for the Social Sciences, 5th ed. Boston, MA: Allyn and Bacon. Thibaut, John W., and Laurens Walker. 1975. Procedural Justice: A Psychological Analysis. Hoboken, NJ: John Wiley. Thoms, Oskar N. T., James Ron, and Roland Paris. 2010. “State-level Effects of Transitional Justice: What Do We Know?” The International Journal of Transitional Justice 4 (3): 329–54. UN (United Nations). 2000. Report of the Panel on United Nations Peace Operations. New York: United Nations. Wagner, Lynn, and Daniel Druckman. 2012. “The Role of Justice in Historical Negotiations.” Negotiation and Conflict Management Research 5:49–71.

Durable Peace

343

Wagner, Lynn, and Daniel Druckman. 2017. “Drivers of Peace: The Role of Justice in Negotiating Civil War Termination.” Group Decision and Negotiation 26 (1): 45–67. Wallensteen, Peter, Erik Melander, and Stina Högbladh. 2013. “Peace Agreements, Justice and Durable Peace.” In Rethinking Peacebuilding. The Quest for Just Peace in the Middle East and the Western Balkans, edited by Karin Aggestam and Annika Björkdahl, 125–139. London, UK: Routledge. Walter, Barbara F. 2002. Committing to Peace: The Successful Settlement of Civil Wars. Princeton, NJ: Princeton University Press.

18 JUSTICE AND NEGOTIATION

The review article in this chapter was an attempt to connect two major themes of the book, justice and negotiation. We do this with the help of a framework adapted from Figure 3.1. The six boxes of the framework, shown in Figure 18.1 of this article, encompass the various topics discussed. Decisions about justice made prior to negotiation deal with framing and preferences for the DJ principles that presage the outcome. The idea of anticipatory justice surfaces in the literature as an expectation for achieving fair outcomes. But we also show that less powerful groups may be fooled into believing that benefits will be forthcoming because they are given opportunities to voice their concerns. Even when power is symmetrical, issues of trust and shared identities loom large. There is circularity in the relationship between trust and justice. Some degree of trust may be needed to take a chance on transparency and fair treatment; when invoked these principles are likely to increase trust. The article carries these processes through to outcomes and the way they are implemented following the agreement. The ideas of directionality and circularity suggest alternative perspectives on the role of justice in negotiation. One perspective views justice through the lens of paths that move through linear time from one stage to another. An example is the modeling done in the preceding article on peace agreements. Another perspective provides a systemic lens where variables are moving together through time. The justice-trust relationship is an example. Here the emphasis is on an interplay among these factors similar to the work on conf licting interests and values discussed in Part 3. We conclude that these perspectives can be used together. Paths are helpful for identifying the variables that set the cycles into motion. The article concludes with a large set of new hypotheses and raises questions for further research. By knitting together two streams of my career, this article closes the journey on an optimistic note about what this field has to offer the generations of social scientists to follow. DOI: 10.4324/9781003293361-32

Justice and Negotiation

Justice and Negotiation, Daniel Druckman and Lynn Wagner. Annual Review of Psychology 67 (2016): 387–413. Annual Reviews. Reprinted with Permission.

345

JUSTICE AND NEGOTIATION Daniel Druckman and Lynn M. Wagner

Introduction This review examines the role played by principles of justice in negotiation. Our focus is on how justice preferences guide the process and outcome of negotiated exchanges. We are less concerned about the use of justice principles to regulate allocation decisions made by authorities or other actors charged with unilateral decision making. Focusing primarily on the two types of principles that have received the most attention, distributive justice (outcomes of negotiation) and procedural justice (process of negotiation), we introduce the topic by reviewing the most relevant experimental and field or archival research on the roles played by these justice principles in negotiation. Our coverage of the literature is broad and deep. While emphasizing more recent studies, we draw on earlier research for providing insights into the connection between justice and negotiation. The recent literature is embedded in the context of a larger body of work. A discussion of the methods used in these studies precedes a review organized in terms of a framework that highlights the concept of negotiating stages. Following the review of relevant literature for each negotiating stage, we identify hypotheses that emerge from existing research to point the way for further studies of the relationship between justice and negotiation. Justice principles can affect negotiation processes and outcomes in a number of ways. In negotiation simulations, subjects negotiating a contract on behalf of a homeowner and contractor, respectively, were found to be more willing to accept an agreement if they assessed the process as fair, and they reached better joint outcomes if they reported a more collaborative negotiation process (HollanderBlumoff & Tyler 2008). In a related finding, opportunities for expression (voice) or the perception of control during a negotiation led to more positive evaluations

Justice and Negotiation

347

of the decision-making process and to judgments of fairness in the outcome (Lind et al. 1990, Tyler 1987). Similar findings are reported for cases with life and death stakes. The negotiations leading to the Rome General Peace Accords, which brought an end to a 15-year deadly conf lict in Mozambique, have been held up as an example for the resolution of civil wars. The Community of Sant’Egidio, a Catholic nongovernmental organization (NGO), led the mediation efforts in these talks and fostered “a genuine political process based on dialogue between the warring factions” (Bartoli 1999, p. 265). The relationship established during the negotiations carried over into the implementation of the agreement. By contrast, many of the messages exchanged between rebels and government representatives in the talks seeking to conclude the civil war in Sierra Leone through the Abidjan Peace Agreement of 1996 were conveyed through intermediaries, public statements, and on the battlefield. Justice principles of fair play and transparency were not adhered to in the negotiation process, and the agreement failed to establish a lasting peace (Wagner & Druckman 2015). Other examples of the role of justice come from the arenas of trade and environmental negotiations. Weaving through these negotiating processes are attempts to resolve various procedural issues. The usual result is a mix of adhering to and violating such principles as fair representation, transparency, and voluntary decision making. When these principles are satisfied, the result is often an effective agreement: Examples include the World Trade Organization talks on agricultural issues in 2004 and the Convention on Early Notification of a Nuclear Accident negotiated in 1986. When the principles are violated, the talks are less effective in producing a durable agreement: Examples include the 1990 General Agreement on Tariffs and Trade (GATT) negotiations on textiles and the 2009 Copenhagen Climate Change Conference. When some principles are satisfied while others are violated, the result is often agreements that favor only a few of the negotiating parties, as was the case with the 1992 North American Free Trade Agreement (NAFTA) and the Basel Convention on the Control of Transboundary Movements of Hazardous Wastes and Their Disposal negotiated in 1989. Fairness. Transparency. Equality. Each of these terms is an element of a broader concept of justice. A close examination of the experiments and high-stakes negotiations discussed above reveals that justice is a complex concept, both in relation to attaining just outcomes and to establishing just processes. An understanding of the outcomes in each of these examples necessitates an appreciation of the various types of justice that could be incorporated into an outcome as well as the types of justice involved in the decision-making process. This review of the existing literature seeks to provide such an appreciation, and in bringing together the justice literature as it relates specifically to negotiations, this article examines a topic not yet explored in the Annual Review of Psychology. Related reviews focus primarily on negotiation, with scant reference to justice concepts. Although the Bazerman et al. (2000) review of research on social factors

348 Process and Outcome Justice

that affect negotiation presents some complementary literature, its review of ethics and sacredness evaluates questions of deception in the negotiation process and the implications of negotiations for issues perceived as being taboo. Our review defines justice as a multifaceted process and outcome concept. Kelman (2006) identifies approaches to peacemaking on a larger scale, through conflict settlement, conflict resolution, and reconciliation. The Thompson et al. (2010) review examines negotiation processes and outcomes, focusing on five levels of analysis: intrapersonal, interpersonal, group, organizational, and virtual. Their study evaluates the implications of each level for integrative and distributive bargaining, which involves what the negotiators are doing at the table rather than the fundamental justice principles establishing how they have organized the talks. On the concept of justice, this article offers the first comprehensive literature review of how justice contributes to the negotiation process and acceptance of negotiated outcomes. Miller (2001) examines links among disrespect, anger, and injustice and offers a brief summary of the concepts of procedural and distributive justice, but not as they relate specifically to the negotiation context.

Distributive Justice and Negotiation Distributive justice (DJ) refers to principles for allocating benefits or burdens among the members of a group or community. Four DJ principles are emphasized in the literature: equality, proportionality or equity, compensation, and need. One or more of these principles surface in the outcome of a negotiation. Perhaps the most ambitious research program on DJ in negotiation outcomes was conducted by Deutsch (1985), who investigated preferences for DJ principles in laboratory negotiation tasks. Clear findings emerged from his experiments with college students. The results, along with results from other experiments, indicated an overwhelming preference for the distributive principle of equality and, to a lesser extent, need. These findings are consistent with those obtained by a number of other investigators. Deutsch does note that the findings can be a function of the particular context of the experiments: namely, money as the resource to be distributed, a workplace as the social context, and American college students as the participants. However, he also claims that these conditions worked against a preference for egalitarianism and would seem to encourage a preference for proportionality or equity. Yet, despite the claims for robustness of the equality preferences, Deutsch emphasizes the importance of the type of distributive system. He contrasts an egalitarian, solidarity-oriented system, characterized by positive social relations and a sense of similarity, with a meritocratic, economic-oriented system, characterized by impersonal social relations and a sense of differences. Equality is the preferred principle in the former, whereas proportionality is sought by negotiators in the latter type of social system. By taking the normative system into account, Deutsch runs into a problem. His own results, showing preferences for equality in a meritorious system, run counter to this hypothesis. Further,

Justice and Negotiation

349

equality distribution preferences have been shown to correlate with higher productivity than preferences for distributions based on an equity principle (Cook & Hegtvedt 1983). The empirical results contradict expectations based on norms and raise an issue about the difference between normative and empirically based theory, which is an age-old question in this literature (cf. Rawls 1958 with Deutsch 1985 or Lind & Tyler 1988). Deutsch’s hypothesis does, however, garner support in two recent studies. Harmon & Kim’s (2013) laboratory research showed that equity, rather than equality or need, is regarded as a fair DJ principle in performance-based situations in which negotiators compete for economic rewards. The Wagner & Druckman (2015) archival study of peace agreements showed that DJ principles can lead to the emergence of normative systems. They found a strong correlation between proportionality or equity and the economic component of an index of durable peace: The more central was the proportionality principle in the agreements, the more stable was the economic environment following the negotiation. Neither proportionality nor the economic component correlated with any of the parts of the durable peace index (reconciliation, security institutions, governing institutions). On the other hand, equality correlated with the other parts of the index, indicating that the noneconomic aspects of durable peace are inf luenced by agreements that embody this principle. Earlier research showed that positive relationships or liking (Mikula & Schwinger 1973), long-term relationships (Mikula 1980), and an emphasis on the team aspect of a relationship (Lerner 1974) led to preferences for an equality rule. Recent results in the realm of peace agreements call attention to the instrumental value of distributive principles. Druckman & Albin (2011) found that the equality principle mediated the relationship between the intensity of the conf lict environment and the durability of the peace agreements: When equality was central in the agreements, the intensity of the conf lict had less inf luence on durability. Another study by these authors showed that equality also mediated the relationship between procedural justice (PJ) and durability: PJ increased durability when equality was central in the agreements (Albin & Druckman 2012). Despite normative systems that would seem to encourage proportionality, the peace negotiators looked forward by paving the way for a political system in which the former rebels would be treated as equal citizens of their country. Those negotiating delegations that insisted on equality provisions carved out durable agreements. Thus, distributive principles may be used in more practical or strategic and less normative ways. These DJ findings suggest interesting issues. One is the distinction between normative and empirical approaches to the study of justice. This distinction is also a source of tension in some issue areas. For example, nonproliferation norms in arms control negotiations are viewed by developing countries as being unjust. Agreements to halt the development of nuclear technologies are seen to widen the inequality gap (Müller 2013). Another issue is the relative preferences for different DJ principles in varied contexts and situations. For example, the equality

350 Process and Outcome Justice

principle is essential for durable peace agreements but relatively unimportant for negotiating effective trade agreements (Albin & Druckman 2014a). These issues are discussed in the sections to follow. Similar issues arise with regard to PJ in negotiation, to which we now turn.

Procedural Justice and Negotiation Procedural justice refers to principles for guiding the negotiation process toward agreements. These principles include fair treatment and fair play, fair representation, transparency, and voluntary decisions. One or more of these principles surface during the negotiation process either positively, as, for example, more fair play or transparency, or negatively, as, for example, a lack of fair play or transparency. Positive adherence to one or more of these principles usually moves the process in the direction of agreement, whereas negative adherence often sustains impasses. An example of research that explored relationships between PJ and the negotiation process and its outcomes was conducted in two experiments by Hollander-Blumoff & Tyler (2008). These investigators developed a simulated dispute between two lawyers representing a home-owner and contractor, respectively. The dispute was a conf lict of interest over a contract for the construction of an in-ground, custom-designed swimming pool: The homeowner (contractor) desired as much (little) work as possible for as little (much) money as possible. A set of distributive issues – including various aspects of the construction and monetary value – were at stake in the negotiation. The first study focused on the distributive issues. The results showed that (a) the more procedurally fair the process was rated, the more willing negotiators were to accept the agreement; (b) the more procedurally fair the process was rated, the more that negotiators indicated they felt better and collaborated more during the process; and (c) negotiating dyads that indicated they were more collaborative and had good feelings during the negotiation reached better joint outcomes. The second study introduced an integrative element to the negotiation problem. Negotiators were offered an opportunity to create extra value that had not existed in the previous version of the problem. Procedural justice inf luenced each of the three measures of integrative outcomes: more disclosures, higher joint outcomes, and more Pareto-optimal (most mutually beneficial) outcomes occurred for procedurally just negotiating dyads. Taken together, the two studies show that PJ inf luences the quality of negotiated outcomes: Greater PJ encourages agreements as well as more integrative outcomes when they are available. Wagner & Druckman (2012) obtained similar findings in a study of 11 historical cases of agreements negotiated between governments. In addition to measures of PJ and outcomes, this study added indicators of DJ, problem-solving processes, and the durability of the agreements. These variables were coded from archival documentation about the cases. Thus, rather than the subjective questions asked in the Hollander-Blumoff & Tyler (2008) study, these investigators performed

Justice and Negotiation

351

content analysis of the documents: DJ and PJ were coded in terms of principles that surface in the outcome (for DJ) or during the discussion texts (for PJ). Problem solving was measured with indicators of acknowledgment, disclosure, willingness to cooperate, and brainstorming or reframing. One of three outcome types were coded for each part of the agreement: integrative, compromise, or asymmetrical. Durability consisted of the number of years that the agreement was in force. The key finding from this study was that problem-solving processes mediated the relationship between PJ and negotiation outcomes: When problem-solving processes were used, the outcome was more strongly inf luenced by the PJ principles. Specifically, more integrative outcomes emerged when problem-solving processes were set in motion by adherence to PJ principles. However, without problem solving, PJ would not lead reliably to integrative outcomes. This finding extends the Hollander-Blumoff & Tyler (2008) results in two ways. The PJ-integrative outcome relationship is shown to depend on problem solving, which is similar to their measure of disclosure. The international domain provides another context for demonstrating the role played by justice in negotiation. Similar results from the different settings bolster the case for external validity. Another finding of interest was that DJ principles were more central in durable agreements. The PJ/problem-solving/integrative outcome cluster did not relate to durability. (See Albin & Druckman 2012 for findings on the mediating effects of the DJ principle of equality on the relationship between PJ and durability.) The DJ/PJ distinction is particularly relevant to negotiation, the one referring primarily to outcomes or allocation decisions, the other primarily to the way the process is conducted. Yet, these should not be considered as monistic formulations of justice. Indeed, combinations of principles may be relevant to particular situations. This is especially likely in complex negotiations in which different principles guide distributions on different issues. The idea of compound justice is illustrated by Zartman et al. (1996): “For example, if the two parties claim different interpretations of inequality (equity versus compensation) for different aspects of an issue (or different issues), an equality principle can be produced by offsetting one with the other” (p. 88). Similarly, violations of the PJ principles of transparency and fair representation by secret discussion among members of a small coalition may be offset by presenting a proposal for ratification by all the parties, thus satisfying the PJ principle of voluntary decisions (e.g., Quimpo 2001). The secret discussions often contribute to efficiency at the cost of representation or voice. Attempts made to reconcile these two objectives have been effective in such areas as negotiations over arms export control policy (Müller 2013). Thus, various principles may be applied in combination or sequentially to produce a fair distribution (see also Cook & Hegtvedt 1983). These examples illustrate a bridging function for justice principles, an area that has received limited attention in the research literature. Justice principles may also serve as heuristics that are easy to implement. The DJ principles of equality and equity are generally understood and can be used to justify an allocation decision. The problem, however, is mutual acceptance

352 Process and Outcome Justice

of the “correct” principle: Negotiators may differ on whether the goal of negotiation is to maximize gains (preferring equity) or enhance social harmony (preferring equality). These differences can be mollified by PJ. Perceptions of fair procedures have been shown to offset perceptions of unfair outcomes and facilitate compromise (Tyler & Blader 2003). This is another example of the combined use of DJ and PJ principles. These authors also suggest a strong relationship between the types of principles. Indeed, Hauenstein et al. (2001) report an average correlation between PJ and DJ, across the 63 studies used in their meta-analysis, of 0.64. The findings from these studies shed light on the role played by these principles in negotiation. Negotiators who view the talks as being fair or who adhere to PJ principles during the process usually engage in problem solving and attain better outcomes. It is also the case, however, that these findings were obtained from only a few studies. Further insights into the role of justice in negotiation are sought by canvassing a large empirical literature on justice in related settings. We organize the findings in terms of negotiation stages to provide analytical specificity. We generate hypotheses to be explored in laboratory and field negotiations, and we provide a basis for organizing future reviews of findings from the studies to be conducted during the next decade.

Methodology The empirical literature on justice has consisted of both experiments and case or field studies. Both approaches have contributed to the development of theory. The complementary strengths and weaknesses of these methods (see Druckman 2005) are understood in terms of issues of causation and levels of analysis. These issues are discussed in this section. The logic of causation guides the design of experiments. Necessary conditions include both control over the administration of independent variables (e.g., performance or relational-based task) and assessment of dependent variables (e.g., preference for a type of justice principle) as well as over the inf luence of possible confounding variables that threaten the validity of causal inferences. Many of the studies reviewed in this article satisfy these conditions (e.g., Deutsch 1985, Hollander-Blumoff & Tyler 2008, Harmon & Kim 2013); some studies go further by isolating mediating or moderating variables (e.g., perceptions of trust) that help to explain the independent variable–dependent variable relationship (e.g., Colquitt 2001, Hauenstein et al. 2001). Further advantages include replication and direct observation of negotiating processes. However, these internal validity strengths are gained at the cost of relevance to the complex, nonlaboratory negotiation settings in which justice plays a role before, during, and after the conclusion of negotiation. This broadened perspective is captured by case studies. The difficulties involved in deriving causal inferences from case data are offset by wider and deeper probes of the role of justice in negotiation. Wider probes

Justice and Negotiation

353

refer to an expanded portfolio of variables to analyze. Deeper probes refer to longer time periods to observe the inf luence of justice variables during several phases of negotiation, including postnegotiation processes. Further, historical case studies provide opportunities for comparative research. Although these strengths are gained at the cost of causal validity, some progress has been made on this front as well. One advance has been to specify the time-ordering of variables; for example, adherence to PJ principles precedes outcomes that embody DJ principles, which leads to more or less durable implementation. Confidence in the time-ordering of the variables and a large number of cases encourages a search for mediating variables (see Albin & Druckman 2012, Chebat & Slusarczyk 2005). Another advance consists of the precautions taken to reduce the possibility of reverse causation; for example, from effective outcomes to process justice. These include developing conceptually distinct definitions of the variables, using different material for coding the justice variables, and interspersing the order of coding for the key variables such as justice and effectiveness (Albin & Druckman 2014a). A third advance is the care taken to avoid selection biases when choosing a sample of cases for analysis. This is done by drawing random or representative samples from a defined universe of cases within issue areas (e.g., trade, security, environmental negotiations). These procedures bring us closer to bridging the internalexternal validity gap. The gap is reduced further when both experiments and case studies are included in a research project. Other differences between laboratory and cases include the size of effects and levels of analysis. With regard to the former, Lind & Tyler (1988) note: “The laboratory and scenario methods used in early research may have caused us to underestimate the magnitude of PJ effects. Field research results have not only confirmed the findings of laboratory and scenario studies on PJ, but in fact have usually shown stronger PJ effects” (p. 206). This occurs, at least in part, because many of the field studies are conducted in contexts that support PJ, such as democratic institutions, and benefit from the combined effects of several independent variables operating in the same direction. Nonetheless, the correlation between PJ and DJ is consistently high across studies with varied methodologies, although there is a difference in the average size of correlations between dispute resolution and reward allocation settings (Hauenstein et al. 2001). This indicates a problem for assessing independent effects of the two types of justice, particularly when global assessments are made. It encourages the use of experimental or statistical controls in justice studies. Laboratory research on negotiation has been conducted at a micro level, where inf luences are largely situational. Field and archival research has, on the other hand, construed negotiation as a process embedded in a larger domestic or international context. These macro-level variables are considered in our framework to be inf luences on justice decisions in negotiation. But the interplay between levels can also be considered as mutual inf luences, from negotiation to collective processes and vice versa. For example, justice principles that guide negotiated

354 Process and Outcome Justice

distributions may affect mobilization for collection action, which inf luences future negotiations (Cook & Hegtvedt 1983). But it is also the case that collective action decisions are negotiated often in large conferences involving diverse constituencies. An interesting question concerns the use of justice principles to reduce complexity and facilitate decisions in these settings. Other questions involve the path from negotiation to implementation, and some of these questions are discussed in the section on the implementation of negotiated agreements.

Justice and Negotiation Stages The set of studies reviewed above makes a case for the importance of justice principles in negotiation. In this article, we explore further the role played by and the importance of justice principles in negotiation, drawing on literature from several social science disciplines using a variety of methodologies, including experiments and field or archival studies. But, given the limited amount of research on justice in negotiation (see Conlon 2012), we also draw on related literatures that have implications for negotiation. The relevant literature is organized in terms of a framework that includes various parts of negotiation. These parts are depicted in Figure 18.1 as boxes connected by arrows. The variables within the boxes are highlighted in the sections to follow. The stages consist of prenegotiation, processes, outcomes, and implementation. Background factors and conditions are considered to be inf luences from the negotiating parties and the situations within which they negotiate. The factors in these boxes are discussed in conjunction with the stages. Each section below concludes with a summary of key points made about the role of justice in that negotiation stage and includes hypotheses to be explored in further research.

Prenegotiation Several themes are discussed in this section in terms of their relevance to the prenegotiation stage. These include types of framing, tasks, preference for distributive principles, and anticipatory justice.

Framing Mikula & Wenzel (2000) call attention to the importance of a shared moral community for the acceptance of justice principles. When this occurs, the principles become guides for structuring a fair process (PJ) and for agreeing on fair outcomes (DJ). Divergent perceptions of injustice can elicit social conf licts. When this occurs, justice principles become issues in negotiation. They may also be used tactically to defend positions, further exacerbating the conf lict. In these situations, conf licts are resolved only when the proposed resolution leads to better outcomes for both/all parties.1

Justice and Negotiation

Prenegotiation Framing of norms Framing of group boundaries Framing in terms of selfor collective interests Preference for distributive principles

Background factors Past experience, expectations/anticipations Power (a)symmetries Conflict intensity Institutional setting

FIGURE 18.1

Processes PJ and motivational orientations Bargaining or problemsolving processes Shared identities and trust False justice

Conditions Competitive or collaborative task Divisions within teams Mediation SociaI networks Spoilers

355

Outcomes Type of agreement: joint benefit, compromise, asymmetrical, impasse DJ principles: equality, equity, need-based outcomes

Implementation Adherence to PJ principles Type of distributive outcome Postnegotiation spoiler dynamics

Justice and Negotiation: A Framework. Abbreviations: DJ, distributive justice; PJ, procedural justice.

The concept of a shared moral community captures both shared identities and shared cognitions. These variables have been shown to inf luence negotiation outcomes. (See, for example, Swaab et al. 2002 on shared cognitions and Gelfand et al. 2006 on shared identities.) In an attempt to integrate these literatures, Swaab et al. (2007) showed how they interact during the process leading to integrative agreements; shared identities were both the product of, and precursor to, the development of shared cognitions. Working together, these variables increased the chances of obtaining integrative outcomes. Regarded also as components of justice, identities and cognitions may be regarded as mechanisms that explain the relationship between adherence to shared principles of justice (such as equality) and durable outcomes (see Albin & Druckman 2012). The shared moral community argument is also similar to Tyler’s (2000) discussion of the way people frame group boundaries. The wider the definition of inclusiveness, the more likely that adherence to PJ principles will contribute to favorable joint negotiation outcomes. The way that people frame their identities inf luences the impact of PJ on outcomes. Interestingly, perceived identities can be inf luenced through experimental priming. Druckman & Olekalns (2013) showed that primed shared identities inf luenced decisions following a negotiating crisis. Deutsch’s (1985) egalitarian/solidarity versus meritorious/economics-based distinction is similar to the distinction made by Lind & Tyler (1988) between a group value and self-interest model. An emphasis on one or the other orientation can be framed by the negotiators or third parties prior to the negotiation. Attempts to invoke shared identities place an emphasis on group values, which

356 Process and Outcome Justice

increase the salience of PJ principles and problem solving during the process. The cooperative context engendered by this framing may also be instrumental in creating civic values that serve to restore damaged justice and reconcile deteriorating relationships, as noted by Johnson & Johnson (2012) in their discussion of constructive controversy. A focus on self-interests orients negotiators toward allocation decisions that inf luence the outcome. This focus may decrease the salience of PJ in the process in favor of a competitive bargaining process. Competitive negotiators seek advantageous agreements but may also strive for fair allocations. The key point is that they approach negotiation as tacticians with an eye on the outcome rather than the process. Thus, prenegotiation framing can inf luence the process by orienting negotiators toward fair processes or toward advantageous (or fair) outcomes. Barrett-Howard & Tyler (1986) found that PJ is more important in situations in which maintaining social harmony and maximizing welfare of all group members are the goals. These are negotiating situations that may give rise to cognitive (differences over means) rather than interest (differences over ends or outcomes) conf licts. Bremer & Hammond (1977) showed that these are difficult conf licts to resolve. They did not, however, investigate the role of PJ in facilitating resolutions.

Tasks and Setting The study by Harmon & Kim (2013) on trust repair is also interesting for the prenegotiation phase. They found that trust is restored when explanations for a breech of trust were based on equity rather than equality or need. Needbased justifications were least effective, especially when the benefits accrue to the person violating the trust. This effect is mediated by perceptions of fairness. It appears that equity is regarded as a fair DJ principle in performance-based situations. Equality, on the other hand, may be the guiding principle in group solidarity situations. A study by Cohn and colleagues (2000) showed that the dispute-resolution setting made a difference with regard to the PJ principle used to guide decisions: impartiality is more important in court-like settings, whereas voice (opportunities to be heard) is more important in settings that focus on rights and morality. Can negotiating issues be distinguished along these lines, perhaps the distinction between interests and values? This study also has implications for process: PJ considerations trump DJ in the adjudication of a rights claim (values); it was considered unjust to deny PJ to the undeserving. Further, the correlation between PJ and DJ may be somewhat stronger for dispute resolution than for reward allocation contexts: The average correlation computed across 39 dispute resolution studies was 0.72, whereas the correlation computed across 55 reward allocation studies was 0.59; the difference between these correlations is not, however, statistically significant (Hauenstein et al. 2001). Thus, the relationship between these variables is, to some extent, context specific.

Justice and Negotiation

357

Preference for Distributive Principles The research reviewed by Kabanoff (1991) suggests that power differences orient parties toward the distributive principle of equity, with distributive decisions favoring the stronger party. These decisions may not be the result of a negotiation, serving to frustrate the weaker parties and leading them to withdraw from the relationship. The power differences may also preclude negotiations. One way to overcome these differences is to initiate protective contracts. Thibaut’s (1968) research shows that bargainers were inclined to enter into protective contracts that insured equity (fairness) in the distribution of resources (DJ) when the conf lict of interest (CI) was high. These contracts reduced the impacts of intense conf licts, defined in terms of asymmetrical power [one party could distribute resources that either favored him/herself considerably (high CI) or only slightly (low CI)]. A challenge, however, is posed to both weaker and third parties. They must persuade the more powerful parties that it is also in their interest to sign such contracts. This is more likely to occur when longer-term relational issues are at stake. Whether achieved by contractual mechanisms or as a by-product of organizational structure, perceived power equality orients parties toward the distributive principle of equality, which usually leads to negotiation based on reciprocity. However, a lack of power differentiation may also produce overt conf lict that cannot be resolved by resorting to differences in power (Kabanoff 1991). These implications are discussed further in the section on the negotiation process. More generally, Kabanoff (1991) suggests that organizational structure inf luences preferences for the equality and equity principles. A preference for equality is likely in less hierarchical, more horizontal organizations; a preference for equity occurs in more hierarchical, power differentiated organizations. (See Kabanoff ’s 1991 Figure 1 for the various combinations of structure and process.) Structure is more inf luential for negotiations conducted in tightly structured organizations such as the military but less inf luential in loosely structured organizations such as the United Nations or other international organizations. Druckman’s (2006) distinction between strong and weak cultures is relevant. Accepted institutional routines guide reward systems in strong organizational cultures; more f lexible routines encourage alternative routes to advancement in weaker cultures. These structures shape distributive norms in the direction of equity for strong cultures or equality for more loosely structured organizations.

Anticipatory Justice The study by Goldman et al. (2013) on anticipatory justice is relevant in thinking about justice and prenegotiation. Anticipatory justice refers to thinking about an event that has not yet been experienced. With regard to mediation, the findings show that it is more likely to be accepted when both DJ and PJ are expected.

358 Process and Outcome Justice

More interesting, however, is the finding that the PJ effect depends on anticipatory DJ. Another interesting finding is that fear of the opponent (anticipated negative effects of the opponent’s voice) leads to reluctance to choose mediation, and this effect depends on the anticipatory distributive injustice or costs incurred to an organization from a harmful opponent.

Summary The review in this section highlights the inf luence of decisions made during the prenegotiation stage and the emergence of justice principles. With regard to PJ, the framing of group boundaries can either expand or limit the perception of inclusiveness. A broad definition of community is likely to enhance shared identities, which encourages adherence to PJ principles. The research is less clear about the mechanisms of framing or reframing. Perceptual change is a longterm process, as recognized by practitioners who design interactive problemsolving workshops (e.g., Rouhana 2000). Although progress has been made on forging collaborative relationships among adversaries in several conf lict arenas, the interactive technologies have not addressed connections between cooperative motives (or shared identities) and PJ principles. Nor have justice researchers explored the impact of prenegotiation agreement on the choice of PJ principles for guiding the negotiation process. Two hypotheses are suggested for further research: H1: Agreement on PJ principles is more likely when parties perceive shared identities or common membership in a moral community. H2: To the extent that parties agree on PJ principles during prenegotiation, these principles will guide the negotiation process. With regard to DJ, preferences for equality or equity principles have been shown to be inf luenced by the framing of the negotiation (performance or solidarity), the relative power of the parties (asymmetrical, symmetrical), and the structure of the organization in which the talks are embedded (hierarchical, horizontal): Performance (solidarity), unequal (equal) power, and hierarchical (horizontal) structures lead to equity (equality) preferences. The prenegotiation conditions that promote equality are more likely to also encourage agreement on using PJ principles to guide the process. The PJ principles then lead to equality outcomes and durable agreements, as shown by Albin & Druckman (2012) in their research on civil wars. Two hypotheses are suggested: H3: Preferences for distributions based on the principle of equality are more likely to occur when the negotiation is framed in terms of solidarity rather than competitive performance. H4: To the extent that parties agree on DJ principles during prenegotiation, these principles will inf luence preferences for certain types of outcomes.

Justice and Negotiation

359

Negotiation Processes The discussion of processes in this section emphasizes the themes of motivational orientations and PJ, shared identification and trust, false justice, and power.

PJ and Motivational Orientations The type of justice most relevant to the process of negotiation is PJ. There is an interesting parallel between the Lind and Tyler models of PJ and negotiation models. Lind & Tyler’s (1988) distinction between self-interest and group value explanations for adhering to PJ principles is similar to bargaining and problemsolving approaches to negotiation. Both self-interest and bargaining focus primarily on the goal of maximizing returns for self. Both group value and problem solving focus on the goal of maintaining or improving relationships. Perhaps the key distinguishing feature for both the PJ and negotiation models is orientation toward self or other (Pruitt & Carnevale 1993). PJ judgments and negotiating behavior derive from similar sources of motivation referred to by Donohue & Hobbler (2002) as affiliation and power orientations. These parallels provide a link between justice and negotiation process. Viewed across the spectrum of negotiation stages, motivational orientations may be developed prior to negotiation, are manifest in the process as competitive or cooperative behavior, emerge as compromise or integrative agreements, and inf luence the durability of the agreement. This sequence is evident in the negotiation findings obtained by Hollander-Blumoff & Tyler (2008) and Wagner & Druckman (2012) discussed above. A broad survey of the PJ literature (e.g., Tyler 2005) suggests that PJ effects on perceptions and behavior are robust across cultures and are generally noninstrumental or valued for their own sake (e.g., Machura 2003, Sugawara & Huo 1994, Wemmers et al. 1995). However, the particular PJ principle used is sensitive to the situation (Tyler 1988). Situational dimensions include type of authority encountered, formal and informal encounters, and characteristics of the experience, but also prior views and expectations, which may lead to biases (Tyler 2000). One particular PJ principle, relevant to negotiation, is the opportunity for voice, referred to also as fair treatment. Studies by Tyler (1987) and by Lind et al. (1990) show that opportunities for expression (voice) or the perception of control over the process leads to more positive evaluations of the decision-making process and to higher fairness judgments. Interestingly, predecision voice led to higher fairness judgments than postdecision voice. More generally, opportunities for voice can ameliorate the prospect of undesirable distributive outcomes. This is often seen in electoral politics, where groups remain loyal to their candidate despite policy differences: A heretical leader is more desirable than a consistently abject alternative (see Hirschman 1970). And, as Peterson (1999) showed, opportunities for voice may have positive effects on less – but not more – intense conf licts. Voice can have the effect of either reducing or intensifying conf lict depending on the sensitivity

360 Process and Outcome Justice

of the issues and the context in which the opinions are expressed. These robust findings on PJ effects beg the question of what accounts for them. Tyler & Blader (2003) suggest that procedures are important because they shape peoples’ social identity within groups, and these identities in turn shape values, attitudes, and behavior: Social identity may be a mediator between PJ and satisfaction or fairness judgments. The PJ feature that carries the most socialidentity-relevant information is the perceived quality of interpersonal treatment, which is relevant to one’s perceived status and value in the group. Tyler & Blader (2003) connect PJ to the identity literature. Implications for negotiation process may be in the realm of within-team or delegation perceptions concerning confidence in and legitimacy of the negotiating representative. These perceptions are likely to change through the course of negotiation, which is monitored by delegation members and policy makers to whom negotiators are accountable. With regard to complex, multi-tiered negotiations between bitter adversaries, Lilja (2012) shows that there is value in developing social networks within the rebel parties. By increasing the interconnectedness of factions and thus cohesion, they become a more unified negotiating team with increased confidence in their representatives. The networks provide strong ties between individuals who control resources. Lilja’s (2012) analyses suggest that the unity enhanced the team’s f lexibility as well as their adherence to PJ principles. This case study raises the issue of how to structure a group or delegation in order to enhance internal trust, with implications for f lexibility in negotiation.

Shared Identities and Trust Perhaps more important for negotiation is the extent to which the parties share an identity. Shared identities have been shown to produce improved processes and outcomes (Druckman & Olekalns 2013, Kramer et al. 1993). Thus, to the extent that PJ increases the perception of a shared identity, it should also improve the negotiation. More broadly, shared identities are part of a cluster of covarying relationships that include PJ, trust, and problem-solving processes. Positive effects on outcomes and relationships from adhering to PJ principles may be explained by increased trust and problem-solving behavior. The Irmer & Druckman (2009) study showed that movement from calculus to identity-trust led to more comprehensive peace agreements. The Wagner & Druckman (2012) study showed that problem-solving processes mediated the relationship between PJ and integrative outcomes in historical international negotiations. The Druckman & Olekalns (2013) study showed that trust (as confidence that the other will honor commitments) propelled negotiators to continue talking in the face of a crisis even when transaction (process) costs were high. Holtz (2013) developed a model of reciprocal relations between trust and perceived justice. This model posits that the development of trust, which may form rapidly, precedes and inf luences employee perceptions of justice.

Justice and Negotiation

361

These studies raise questions about causal sequences among these variables: Is trust a precondition for PJ, shared identities, and problem solving? Or, does trust emerge from agreement on PJ principles and/or problem solving? If trust is a precondition, as Holtz (2013) suggests, then an early focus on creating conditions for increasing perceptions of trust would be beneficial. If, however, trust is an emergent process, as suggested by the Irmer & Druckman (2009) finding, then focusing first on establishing PJ rules would be advised. On the other hand, if the trigger is problem solving, as suggested by the Wagner & Druckman (2012) finding, then encouraging these behaviors would increase trust. It may be that the variables in this cluster are intertwined or cyclical rather than sequential. Bolstering any one of them would have ramifying effects on the others. These are issues that remain to be explored. A related issue is the role of justice principles in trust violation and repair. Explanations for violations may be framed in terms of justice principles. Equity principles work best to repair trust when the perceived violation occurs in task performance situations. Equality is better in situations that emphasize the socioemotional aspects of relationships (Harmon & Kim 2013). Less is known about the role of PJ principles in trust repair. One issue is the extent to which PJ principles repair trust when the specific principles do not fit the situation or provide sufficient explanation for the violation. Trust is important as well with regard to mediation. Acceptance of a mediator’s suggestions has been shown to depend on the extent to which he or she is judged as being fair. Conlon et al. (1994) showed that negotiators accepted a mediator’s suggestions when they perceived him or her to be fair. A suggestion to reject a compromise decision in favor of a problem-solving process was accepted after the mediator demonstrated an understanding of the compromise or equal losses outcome. That understanding enhanced perceptions of the mediator’s fairness or trust that, in turn, bolstered the negotiators’ willingness to take risks. The risks consisted of revealing sensitive information that aided problem solving toward integrative solutions. These mediation process variables may, however, be moderated by the parties’ expectation of receiving a better outcome (Schuller & Hastings 1996). The studies reviewed above support the idea that positive PJ perceptions are beneficial to both/all parties in negotiation: They promote shared identities and increase trust perceptions. However, it would be interesting to identify the conditions under which PJ is used instrumentally to achieve better outcomes for self, following the Lind & Tyler (1988) self-interest model of procedural justice. Relevant themes are false justice, compensatory justice, and asymmetrical power.

False Justice Despite the benefits of PJ, there is the possibility that these principles can be abused. Referring to the idea of “false consciousness,” Lind & Tyler (1988) call attention to the dangers of PJ manipulation. Low-power groups can be fooled

362 Process and Outcome Justice

into believing that there will be distributive gains when they are given voice. This has been shown to occur at the macro, meso, and micro levels of analysis. At the macro level, politically disadvantaged groups are discouraged from mobilizing for actions that would improve their outcomes by fostering societal beliefs in the fairness of political institutions (Tyler & McGraw 1986). At the meso level, “an organization might introduce task-assignment procedures that appeared to allow workers voice prior to the allocation of task assignments when in fact the voiced preferences and values are never really considered” (Lind & Tyler 1988, p.  201). At the micro level, giving people voice in small groups may exacerbate conf lict when these people fundamentally disagree (Peterson 1999). This effect was also shown to occur in attempts to negotiate peace agreements. Opportunities for voice given to rebel groups in negotiation often backfire when the agreement that emerges favors the government. Interestingly, PJ principles were adhered to even in some of the most difficult conf lict environments: The correlation between PJ and conf lict intensity across 16 peace negotiations was moderate (−0.49) (Druckman & Albin 2011). But, many of these cases – for example, the 1994 Rwanda negotiations – did not produce durable agreements.

Power In asymmetrical power situations, the more powerful party prefers equitable distributions (favoring him- or herself ) without alienating (disloyalty, withdrawal) less powerful parties. They may do this by espousing procedural justice, which can be used manipulatively as false justice. An exception, documented by Kapstein (2008), occurred in the Uruguay round of the GATT negotiations. In those talks, the more powerful developed countries made larger concessions than the less powerful developing countries. This was interpreted as a form of compensatory justice, which contributes to trade balance in the international system. Of interest is the question of conditions under which different types of justice principles are used in negotiation. A key tension experienced by negotiators is between striving for equitable or equal distributive outcomes. This tension is acute when the negotiating parties are near-equals in power (Mulder 1977). The jockeying for advantage that occurs between near-equals prolongs the process and risks impasses. Both parties challenge the equity of rewards received by the “stronger” party. Both are motivated by self-interest: When PJ is motivated by self-interest, it is likely to lead to equitable distributions (to reinforce power relations). A challenge in this situation is to orient the parties toward group values: When PJ is motivated by group values, it is more likely to lead to equal distributions (to reinforce group cohesiveness). The connection between the Lind & Tyler (1988) self-interest and group values models and task or solidarity motives is made by Kabanoff (1991). He discusses the organizational designs that would balance efficiency (rewards based on performance) with cohesion or solidarity (rewards based on

Justice and Negotiation

363

relationships). This balance is often achieved through negotiation processes. We know less about the possible causal sequences from motivation to PJ adherence to distributive outcomes.

Summary In this section, we discussed the relationship between PJ and negotiation processes. The review suggests that this relationship can be understood in terms of alternative motivational orientations, trust, and power. The distinction between affiliation and power sets in motion particular negotiation processes as summarized by two hypotheses: H1: Motivational orientations that focus on affiliation increase the likelihood of problem-solving processes. H2: Motivational orientations that focus on power increase the likelihood of competitive bargaining processes. These contrasting orientations can also be construed in terms of the Lind & Tyler (1988) models of PJ, referred to as group values and self-interest: An emphasis on group values is more likely to lead to problem solving, whereas a self-interested negotiator is more likely to encourage competitive bargaining. A group values emphasis strengthens group identity, with implications for both intrateam and interteam negotiations as summarized by the following hypotheses: H3: Adherence to PJ principles with an emphasis on group values strengthens group identity, which increases trust in the team’s negotiating representative. H4: Adherence to PJ principles with an emphasis on group values increases the shared identity between the opposing representatives, leading to problemsolving processes. A self-interest emphasis reduces group identity with implications also for intra- and interteam negotiations, as summarized by the following hypotheses: H5: Adherence to PJ principles with an emphasis on self-interest weakens group identity, which decreases trust in the negotiating representative. H6: Adherence to PJ principles with an emphasis on self-interest decreases the shared identity between opposing representatives, leading to competitive bargaining processes. Another interesting relationship is between PJ and power. The review suggests that an emphasis on group values or self-interests may moderate the impact of unequal power on negotiation processes. This relationship is summarized by the following set of hypotheses:

364 Process and Outcome Justice

H7: Asymmetrical power between opposing negotiating representatives enhances cooperation or competition, depending on whether negotiators emphasize group values or self-interests. H7a: Group-value negotiators are more likely to use their power advantage to encourage adherence to noninstrumental PJ principles, which leads to problem-solving processes. H7b: Self-interested negotiators are more likely to use their power advantage to encourage adherence to instrumental PJ principles, which leads to competitive bargaining processes. Viewed through a PJ lens, negotiation processes are driven by the alternative orientations of group values or self-interest. The former emphasizes joint gains, whereas the latter is concerned primarily with maximizing own gains. This distinction is made also by the well-known dual-concern model of negotiation (Pruitt & Carnevale 1993) and reinforced by the relatively strong effect sizes for orientation obtained in meta-analyses of bargaining behavior (Druckman 1994). Extending the path from orientations to processes further, we can derive implications for negotiation outcomes, as discussed in the next section.

Negotiation Outcomes Negotiation outcomes are usually considered in terms of the type of agreement, as joint benefits, compromises, or impasses (see Figure 18.1). A considerable amount of research on negotiation has focused on relationships between these outcomes and processes, such as bargaining or problem solving. Results have shown that processes inf luence outcomes (e.g., Hopmann 1995, Irmer & Druckman 2009, Wagner 2008). Justice provides another lens for viewing outcomes, which are considered in terms of distributional benefits (see Figure 18.1). These benefits, or DJ principles, also emerge from processes, which are construed in terms of PJ principles.

Relationships Between PJ and DJ The substantial correlations between PJ and DJ across the 63 studies analyzed by Hauenstein et al. (2001) are consistent with results on process-outcome relationships from the negotiation studies. However, the correlations are inconsistent across negotiation issue areas. Correlations among various indexes of PJ and DJ for samples of peace agreements range from 0.06 to 0.39 (see Wagner & Druckman 2015). The correlations for a sample of bilateral and multilateral trade negotiation cases range from 0.38 to 0.65 (Albin & Druckman 2014a). For a sample of arms control negotiation cases, the correlations range from 0.06 to 0.25 (Albin & Druckman 2014b),

Justice and Negotiation

365

and for a sample of environmental cases, they range from 0.01 to 0.32. Thus, the strength of correlations between these variables is contingent on issue area. A more promising line of investigation may be to consider mediating variables. As we discussed previously, PJ effects on outcomes are mediated by problemsolving processes (Hollander-Blumoff & Tyler 2008, Wagner & Druckman 2012): Adherence to PJ principles leads to integrative outcomes when negotiators engage in problem solving (versus bargaining). PJ effects on the durability of the negotiation outcome have been shown to depend on the centrality of the DJ principle of equality in the agreement (Albin & Druckman 2012): Adherence to PJ principles leads to durable outcomes when equality is emphasized in the agreement. These findings suggest that PJ/DJ relationships depend on other features of the process (bargaining and problem solving), the outcome (integrative or compromise, equality), and context (issue area). Thus, the justice variables are understood as part of a sequence of processes, outcomes, and implementation activities. These relationships have also been shown to depend on perceived time frame. Joy & Witt (1992) found that PJ and DJ are more strongly related for employees with a long-term perspective on their organization. Another way of thinking about interactions between PJ and DJ is in terms of compensatory effects of the two types of justice. Brockner and his colleagues showed that negative effects of unfavorable outcomes, such as a small pay raise or no promotion, can be mitigated by adherence to PJ principles during the process. Indeed, the more severe the outcome, the more salient the effects of PJ (Brockner et al. 1992). Similarly, receiving favorable outcomes, such as an unexpected pay raise or promotion, can mitigate the negative effects of violations of PJ principles (Brockner & Wiesenfeld 1996). Negative events (unfair procedures, bad outcomes) seem to heighten sensitivity to the outcome received or to the procedures used to determine the outcome. More broadly, democratic political systems are based on fair procedures, particularly with regard to legal institutions. This fundamental tenet provides citizens with access to legal procedures and due process. It also cushions the disappointment of distributive losses suffered from negotiated outcomes and may avoid costly civil conflicts between groups (Tyler 1994). Outcomes are evaluated as well in terms of perceived fairness. Interestingly, these perceptions may not be related strongly with satisfaction. Outcome fairness is based on social comparisons, whereas satisfaction is based more on the extent to which one’s own outcomes correspond with expectations or preferences (Brickman 1975). There may also be a temporal effect for the relative contribution of PJ and DJ to perceptions of fairness: PJ is the stronger inf luence when procedural information is available before information about outcomes (van den Bos et al. 1997). Further, the fair process/fair outcome distinction depends on perceptions of authorities or institutions for process (PJ) and specific outcomes such as pay raises (DJ) (Folger & Konovsky 1989).

366 Process and Outcome Justice

Impacts of DJ Principles Achieving outcomes that are “just” may fulfill a universal human desire, but as Zartman et al. (1996) indicate, the negotiating participants’ interpretations of what is just are likely to vary. The different types of justice that can be represented in outcomes introduce the possibility for the creation of trade-offs, but they also present a challenge for negotiators to understand the types of just outcomes and their effects. Although equality, equity, need, and compensation are all called distributive “justice,” the decision mechanisms used to reach each outcome type, the distribution accorded under each principle, and the expected outcomes from each type are different. Proportionality (or equity) and equality have received the most attention in the DJ literature. Equitable outcomes indicate that negotiators have agreed to a distribution that is proportional to negotiators’ inputs (Adams 1965, Homans 1961). By contrast, equality refers to outcomes in which there is an equal distribution of resources or burdens among the negotiators (Deutsch 1985). Thus, actors are assumed to have the same stake in the outcome, an assumption that may lead to different implications than those derived when equity is the guiding principle. A compensatory distribution of resources would involve a division that indemnifies one or more parties for undue costs or burdens (Piaget 1948). The distribution of resources based on parties’ needs would signify that negotiators agreed to an outcome that assigns value in proportion to the strength of one or more party’s needs (Burton 1986). Proportional, or equitable, outcomes rely on assessments of what the relevant input (or, in Homan’s 1961 terms, investments and contributions) is for a particular decision and an agreement regarding the appropriate division of rewards relative to that input. Since these assessments are made by individuals (Adams 1965), they could make room for the tradeoffs identified by Zartman et al. (1996), particularly if each negotiator values the inputs differently. Adams (1963) and others have looked into whether inputs vary based on perceptions of inequality and found that student subjects who believed they were being paid more than they deserved increased their relative inputs as a way to reduce inequities. Experiments have also revealed that the reverse may occur: Those who feel underpaid are likely to reduce their inputs. Deutsch (1985), however, critiques this theory for overlooking the interactional element of negotiations and for assuming that individuals typically try to maximize their material outcomes. As discussed previously, the objectives of outcomes that emphasize equality and proportionality differ. Similarly, the objectives of outcomes that emphasize need also differ. Kabanoff ’s (1991, p. 420) observation that “equity [proportionality] was the preferred principle when productivity was emphasized, and equality was chosen when solidarity was the goal” highlights that outcome and implementation are on a continuum, where decisions at one stage can anticipate the requirements of the next. Deutsch (1985) notes that need is the prominent principle of distributive justice if fostering personal

Justice and Negotiation

367

development and welfare is the primary goal of negotiation. He suggests that “caring” and “solidarity” (equality) orientations are similar and differ in similar ways from the “economic” (proportionality/equity) principle, although need is “characterized by a more direct and explicit responsibility for the fostering of the personal development and personal welfare of others in the group” (Deutsch 1985, p. 45). These outcome-related concepts may take shape during the prenegotiation phase and influence the postnegotiation process. As negotiators develop their preference for distributive principles and assess their relevant inputs before they enter into a negotiation, implicit or explicit assumptions regarding the value of distributive outcome types are made. These concepts also indicate outcome preferences that could ensure implementation, such as convincing the other negotiator that they are receiving more than they deserved. The concept of distributive justice also introduces a mechanism through which negotiations may become recursive in a search for justice. If parties believe they did not receive equitable distributions, they will be distressed, which in turn motivates them to try to restore equity (Walster et al. 1978). Postagreement negotiations are often stimulated by distress and by motivation for the restoration of equity.

Summary The research on negotiated outcomes reveals that justice is a dynamic concept. It may be construed on a continuum that is set in motion by justice issues raised during the process, continuing through the outcome and to implementation activities. The relationship between justice as an outcome and the extent to which justice is involved in the procedures to reach that outcome on the one hand, and the expected type of implementation on the other, suggests a pattern of overlapping influences across negotiation stages. Some of these influences are summarized in the form of the following hypotheses: H1: The strength of the relationship between procedural and distributive justice will vary by issue area. H2: Equality outcomes enhance the durability of an agreement, particularly when negotiators adhere to PJ principles during the process. H3: Adherence to PJ (DJ) principles can compensate for unfavorable distributive (procedural) outcomes. H4: Perceived fairness is not strongly correlated with outcome satisfaction: The former is based on social comparisons; the latter is based on expectations or preferences. H5: Equitable/proportional outcomes enhance productivity, equal outcomes enhance solidarity, and need-based outcomes enhance personal development and welfare.

368 Process and Outcome Justice

Implementation of Agreements (Postnegotiation) The discussion in this section on implementation includes considerations of both procedural and distributive justice issues.

Influence of PJ The durability of agreements has been shown in field experiments on mediation to be inf luenced by PJ judgments. Pruitt and his colleagues (1990, 1993) found that long-term compliance with agreements is more strongly predicted by PJ perceptions than by joint problem solving processes, goal achievement, or satisfaction with the agreement. Interestingly, reaching agreement or the quality of those agreements did not predict durability. Adherence to PJ principles has also been shown to increase compliance with agreements in difficult felony cases (Casper et al. 1988). These findings may be due, in part, to the positive impact of PJ on the relationship between the parties (Tyler & Blader 2003). Postagreement negotiation processes include carrying out the terms of agreement as well as continuing negotiations over those terms, new issues that arise, or the relationship. In their discussion of intergovernmental negotiation, Spector & Zartman (2003) conceive the postagreement phase as a series of related encounters. Procedures and outcomes from one negotiation are likely to influence the next negotiation. This focus on relationships places an emphasis on a long process of reconciliation or trust building (see also Rosoux 2013). Thus, research on trust and PJ reviewed above in the section on the process is relevant as well to the postnegotiation stage. Particularly important is the cyclical idea of PJ-trust-problem solving (Holtz 2013, Wagner & Druckman 2012) and the way that PJ influences trust repair (Harmon & Kim 2013). But the path from PJ in the process to equality outcomes to durability also contributes to the success of implementing agreements (Albin & Druckman 2012). That success may, however, be challenged by groups with a political or economic interest in sustaining or escalating the conflict.

Distributive Outcomes The archival research by Druckman & Albin (2011) showed that equality outcomes (fairness) decreased the impact of intense conf licts on the durability of the agreements: For intense conf licts, an emphasis on the equality principle reduced the inverse correlation between intensity and durability. The problem, however, is to create an environment that encourages parties to desire equality outcomes. This may be more problematic in demanding social environments where conf licts are intense and relationships are deteriorating. Mittone & Ploner (2012) showed that equity rather than equality is the guiding distributive principle in

Justice and Negotiation

369

these settings. But it is also the case that mediation is more likely to be used in deteriorating situations (Bercovitch & Diehl 1997). Mediators can be instrumental in moving parties to seek PJ in the process and anticipate equality as an outcome (see Goldman et al. 2013), particularly if they have a reputation for being fair (Conlon et al. 1994) or derive their authority from engaging in fair practices (Pruitt et al. 1993). The literature on reactions to perceived inequity/inequality and on ways to restore equity/ equality is relevant. Experimental subjects generally react to restore an equilibrium by compensating the inequitably underpaid subject (Leventhal et al. 1969, Törnblom 1977). Thus, the distributive principle of compensation may replace equity in the postnegotiation period. However, when the inequity is created by chance rather than intention, compensation is not provided to restore equity (Garrett & Libby 1973, Leventhal et al. 1969). Ascription of responsibility for the inequity may be an important influence on reactions to inequity (see also Cook 1975). Another reaction to inequitable distributions is to leave the situation, which has important implications for implementing the agreement and future negotiations (Cook & Hegtvedt 1983). The issue for negotiation is whether negotiators are motivated to restore equity through compensation or to justify inequitable outcomes by believing that they are consistent with a belief in a just world (Lerner 1980; see also Kapstein 2008 on compensatory concessions from developed to developing countries in the GATT Uruguay round of trade negotiations). This issue has not benefited from research to date. The idea of compensation to restore equity was shown to be related to age. Different DJ principles were used by children of different ages following negotiation in the study by Solomon & Druckman (1972): equity for the youngest group (7–9 years old), equality for the middle-aged group (10–12 years old), and compensation for the oldest group (13–15 years old). These differences are consistent with Piaget’s (1948) stage theory of development of attitudes about resource distribution. An implication is that a compensation rule takes longer to be understood but kicks in before adulthood, suggesting that diplomats would resort to it as a mode for restoring inequity and preserving relationships.

Role of Spoilers A major problem for implementing agreements is the presence of spoilers. These are “leaders and parties who believe the emerging peace threatens their power, world view, and interests and who use violence to undermine attempts to achieve it” (Stedman 2000, p. 178). Spoilers may be inside or outside the negotiation and may have different types of goals, referred to by Stedman (2000) as limited, greedy, and total. Justice considerations come into play with regard to strategies for managing spoilers. These strategies include inducement, socialization, and coercion. Inducement may work best for spoilers with limited goals. These include a larger

370 Process and Outcome Justice

share of the distribution or side payments. Greedy spoilers may also be satisfied with material rewards, but care must be taken to avoid whetting their appetite for more resources. A larger problem occurs with total spoilers, whose demands are often nonnegotiable. The only option may be to resort to coercion if the spoilers’ power base prevents the use of legal channels for prosecution. For each of these spoilers, a socialization strategy is appropriate but difficult to implement given the time frame needed for change to occur. A challenge is to persuade the spoilers to adhere to a common set of norms that include PJ principles. This strategy may work better for internal spoilers who are reluctant to destroy the negotiation process. The prospects of increased voice and legitimacy may satisfy their needs. With regard to peace processes, Neu (2012) raises the dilemma of pursuing retributive justice for war criminals. Difficulties in prosecuting them increase the chances that war criminals will disrupt the process. This prospect is reduced when they are taken out of action by legal prosecution. However, it may also be the case that the groups to which the war criminals belong may rally around them and increase the intensity of protest and violence. On the positive side, such escalation of spoiler demands may serve to unite the opposing negotiating parties. They now have a shared goal of neutralizing the spoilers and a shared task of developing conflict management strategies. These cyclical action-reaction dynamics pose interesting questions for research on implementation of agreements.

Summary The relevant research on implementation of negotiated agreements was reviewed in this section. Compliance with agreements has been shown to be inf luenced by PJ perceptions as well as by the distributive concepts of equality and compensation. These relationships can be understood in terms of a connected sequence of hypotheses as follows: H1: Adherence to PJ principles leads to an emphasis on equal outcomes for the parties. H2: PJ adherence and equal outcomes enhance the relationship between negotiating parties. H3: Improved relationships increase the chances that the negotiating parties will comply with the terms of the agreement. H4: Improved relationships enhance the willingness of more powerful (or advantaged) negotiating parties to offset unequal agreements through compensation during the implementation period. Taken together, this set of hypotheses suggests the following path: Adherence to PJ principles during the process → centrality of the equality principle in outcomes → improved relationships → compliance with the

Justice and Negotiation

371

terms of agreement → use of compensatory principles to adjust unequal outcomes. This path may also have a recursive feature by looping back to negotiation processes. Future talks between the parties may be facilitated by the strengthened relationships between them. These relationships develop in concert with adherence to PJ (process) and DJ (outcomes) principles, both of which contribute to durable agreements. However, the relationships may be threatened or strengthened by the presence of spoilers during the implementation process.

Conclusion: Perspectives on Justice and Negotiation In this review, we situate justice in a broad framework of influences on and processes of negotiation (see Figure 18.1). In this concluding section, we attempt to develop implications of the review for perspectives on the topic. Two ways of construing justice variables emerge from the discussion. One consists of considering justice as part of a system of interacting variables. Another approach views justice through a chronological lens of negotiating stages. A discussion of both is followed by an attempt to provide an integration of the approaches. A systemic approach emphasizes mutual influences of variables relevant to negotiation. This includes bidirectional effects, where justice simultaneously influences and is influenced by other variables through the course of a negotiation. An example is the way that justice covaries with motivational orientation, trust, and shared identification. A willingness to adhere to DJ or PJ principles is encouraged by a social climate in which negotiators perceive trusting relationships, share a larger professional or personal identity, and view the negotiation as a problem to be solved rather than as a contest to be won. The cyclical feature of these covarying factors is that change in any one – for example, information that violates an expectation of affective trust – reverberates through the system. This cycle links the way negotiation is conducted (PJ principles) with the principles in the outcome (DJ), relationships among the parties (trust), motivation (orientation), and social connections (shared identities). The feature of mutual influence also allows for reverse causation, for example, from justice (motivation) to motivation (justice). The systemic idea does, however, raise questions about the interplay among these factors, for example, the difference between reinforcing effects (high trust, cooperation, and PJ lead to integrative outcomes) and offsetting effects (adherence to PJ compensates for low initial trust) or the likelihood of trumping effects (trust is a prerequisite for adherence to PJ principles). A practical question is how to set the cycle into motion. Implications for this question come from the research on prenegotiation framing and anticipatory justice, which is discussed in conjunction with the second perspective.

372 Process and Outcome Justice

A different perspective is provided by situating justice concerns in a sequence of negotiation stages. Paths and cumulated effects on negotiating behavior are emphasized by this chronological perspective. Framing concerns loom large during the prenegotiation period. Of particular importance is the way that justice norms and preferences are developed in the context of structural and experiential factors. These concerns shape various aspects of the negotiation process, including decisions about PJ principles and exchange routines that emphasize bargaining or problem solving. DJ principles such as equality also come into play during the process but surface as formulae that underwrite the agreements. Adhering to these principles during the implementation phase bolsters the chances for compliance with the agreements as well as dealing with those actors intent on spoiling them. But a stage-like process also provides negotiators with experiences that shape expectations for future negotiations. The expectations in turn influence the frames that guide new rounds of negotiation. Nonlinear (systemic) and linear (stages) approaches to theory are often considered to be alternative, even competing, conceptual vantage points. However, they may also be viewed as complementary. Both a dynamic interplay among negotiation variables and a chronological path of influences through the course of negotiation are relevant to an understanding of the role played by justice. For example, cooperative orientations that emerge from shared frames encourage adherence to PJ principles during the early days of negotiating. New information about the other’s strategy or inferences about the other’s intentions may provoke a reaction that reduces a willingness to abide by those principles. This development in turn reduces perceived trust, moving the talks away from agreements that embody DJ principles or an integration of needs and preferences. The systemic perspective highlights the interplay among these variables. The chronological perspective calls attention to particular variables – e.g., anticipatory justice, PJ principles, DJ principles, presence of spoilers – that set the cycles into motion. The combined perspectives may be a useful framework to guide further research about the role of justice in negotiation.

Further Questions Table 18.1 brings together the hypotheses developed in this literature review and can be used to develop further questions that connect the stages. For example, H1 for outcomes suggests that the relationship between procedural and distributive justice will vary by issue area: Do agreements on PJ and DJ principles have stronger influences on outcomes in certain issue areas (prenegotiation H1, H3, and H4)? Are power asymmetries (processes H2, H7, H7a, and H7b) related to the preferences for distribution (prenegotiation H3 and outcomes H5)? And if parties perceive shared identities or common membership in a moral community and therefore agree on PJ principles (prenegotiation H1), do they use problem-solving processes (processes H1, H4)?

TABLE 18.1 Hypotheses for Future Studies on Justice and Negotiation

Processes

Outcomes

Implementation

H1: Agreement on procedural justice (PJ) principles is more likely when parties perceive shared identities or common membership in a moral community. H2: To the extent that parties agree on PJ principles during prenegotiation, these principles will guide the negotiation process. H3: Preferences for distributions based on the principle of equality are more likely to occur when the negotiation is framed in terms of solidarity rather than competitive performance. H4: To the extent that parties agree on distributive justice (DJ) principles during prenegotiation, these principles will inf luence preferences for certain types of outcomes.

H1: Motivational orientations that focus on affiliation increase the likelihood of problem-solving processes. H2: Motivational orientations that focus on power increase the likelihood of competitive bargaining processes. H3: Adherence to PJ principles with an emphasis on group values strengthens group identity, which in turn increases trust in the team’s negotiating representative. H4: Adherence to PJ principles with an emphasis on group values increases the shared identity between the opposing representatives, leading in turn to problem-solving processes. H5: Adherence to PJ principles with an emphasis on self-interest weakens group identity, which in turn decreases trust in the negotiating representative.

H1: The strength of the relationship between PJ and DJ will vary by issue area. H2: Equality outcomes enhance the durability of an agreement, particularly when negotiators adhere to PJ principles during the process. H3: Adherence to PJ (DJ) principles can compensate for unfavorable distributive (procedural) outcomes. H4: Perceived fairness is not strongly correlated with outcome satisfaction: The former is based on social comparisons; the latter is based on expectations or preferences.

H1: Adherence to PJ principles leads to an emphasis on equal outcomes for the parties. H2: PJ adherence and equal outcomes enhance the relationship between negotiating parties. H3: Improved relationships increase the chances that the negotiating parties will comply with the terms of the agreement. H4: Improved relationships enhance the willingness of more powerful (or advantaged) negotiating parties to offset unequal agreements through compensation during the implementation period. (Continued)

Justice and Negotiation

Prenegotiation

373

Prenegotiation

Processes

Outcomes

H6: Adherence to PJ principles with an emphasis on self-interest decreases the shared identity between opposing representatives, leading in turn to competitive bargaining processes. H7: Asymmetrical power between opposing negotiating representatives enhances cooperation or competition depending on whether negotiators emphasize group values or self-interests. H7a: Group-value negotiators are more likely to use their power advantage to encourage adherence to noninstrumental PJ principles, which in turn leads to problem-solving processes. H7b: Self-interested negotiators are more likely to use their power advantage to encourage adherence to instrumental PJ principles, which in turn leads to competitive bargaining processes.

H5: Equitable/proportional outcomes enhance productivity, equal outcomes enhance solidarity, and need based outcomes enhance personal development and welfare.

Implementation

374 Process and Outcome Justice

TABLE 18.1 (Continued)

Justice and Negotiation

375

Summary Points The key points made in this review are organized below by section. 1

2

3

4

5

6

Distributive Justice (DJ) in Negotiation. Different DJ principles are emphasized in different issue domains. For example, equality is important in securing durable peace agreements. Equity or proportionality is a preferred principle when issues of economic stability are at stake. Procedural Justice (PJ) in Negotiation. Adherence to PJ principles encourages integrative negotiating agreements, particularly when the negotiators engage in problem solving or disclosure during the negotiating process. Methodology. Laboratory and field research on justice are complementary research strategies. Conducted primarily at a micro level of analysis, laboratory studies provide insights into the role of justice in negotiating processes and outcomes. A focus primarily on macro-level variables in field or archival research provides insights into the influence of the institutional contexts in which negotiation occurs. A challenge for analysts is to develop methodological strategies that facilitate exploration of linkages between these levels. Prenegotiation. Decisions made prior to negotiation influence the way negotiators discuss the issues. The framing of group boundaries influences perceptions of inclusiveness and shared identities, which in turn increase (decrease) adherence to PJ principles. Preferences for the DJ principles of equality or equity are influenced by whether the negotiation is framed as focused on building or reinforcing relationships (solidarity) or on encouraging competitive performance. Negotiation Processes. Negotiation processes are influenced by alternative models of PJ. A group values emphasis gears the process in the direction of problem solving. PJ principles are used in noninstrumental ways. The mechanisms for this effect are strengthened intra- and interteam identity and increased trust. A focus on self-interest orients the process toward competitive bargaining. The mechanisms for this effect are weakened team identity and decreased trust. PJ principles are viewed as instrumental for achieving desired outcomes. Negotiation Outcomes. The distinction between equal and equitable outcomes has received considerable attention in the research literature on justice. Equality is the preferred principle when the negotiation is framed in terms of solidarity. Equity is preferred when the negotiation is framed in terms of economic productivity. The former promotes group values, whereas the latter is geared toward self-interest. A third type of distribution is based on need. Emphasizing the welfare of others, this distribution compensates for losses experienced by the other negotiator.

376 Process and Outcome Justice

7

8

This outcome may also compensate for a lack of fairness in the negotiation process. Implementation of Agreements. A key to the implementation of agreements is the relationship among the negotiating parties. The relationship is strengthened by adherence to PJ principles during the process and by outcomes that emphasize the distributive principle of equality. Yet despite improved relationships, an agreement can unravel when some members of negotiating delegations or third parties have incentives to spoil or undermine the implementation. Strategies for dealing with spoilers depend on the distinction between their goals as limited, greedy, or total. Conclusion. Negotiation can be understood as a system of interacting variables. Justice perceptions influence and are influenced by relational, motivational, and identity variables. The sequential perspective used in this review provides further insights into these cyclical processes. Following a negotiation through its stages reveals how early framing encourages (or discourages) adherence to PJ principles, which in turn leads to certain types of distributive outcomes that remain in place over time or go off course through actions taken by spoilers. Systemic and sequential perspectives are shown to be complementary.

Note 1

Rawls (1958) distinguished between utilitarian and justice principles. He argued that unequal outcomes can be justified when benefits accrue to the entire community. The justification is bolstered when procedures used for attaining these outcomes are considered to be fair by members who share the same moral community. These circumstances, present in most democratic societies, do not vitiate the central organizing principle of equality.

References Adams JS. 1963. Toward an understanding of inequity. J. Abnorm. Soc. Psychol. 67:422–36 Adams JS. 1965. Inequity in social exchange. Adv. Exp. Soc. Psychol. 62:335–43 Albin C, Druckman D. 2012. Equality matters: negotiating an end to civil wars. J. Confl. Resolut. 56:155–82 Albin C, Druckman D. 2014a. Procedures matter: justice and effectiveness in international trade negotiations. Eur. J. Int. Relat. 20:1014–42 Albin C, Druckman D. 2014b. Bargaining over weapons: justice and effectiveness in arms control negotiations. Int. Negot. 19:426–58 Barrett-Howard E, Tyler TR. 1986. Procedural justice as a criterion in allocation decisions. J. Personal. Soc. Psychol. 50:296–304 Bartoli A. 1999. Mediating peace in Mozambique. In Herding Cats: Multiparty Mediation in a Complex World, ed. CA Crocker, FO Hampson, P Aall, pp. 245–74. Washington, DC: US Inst. Peace Press

Justice and Negotiation

377

Bazerman M, Curhan JR, Moore DA, Valley KL. 2000. Negotiation. Annu. Rev. Psychol. 51:279–314 Bercovitch J, Diehl PF. 1997. Conf lict management of enduring rivalries: frequency, timing, and short-term impact of mediation. Int. Interact. 22:299–320 Bremer B, Hammond KR. 1977. Cognitive factors in interpersonal conf lict. In Negotiations: Social-Psychological Perspectives, ed. D Druckman, pp. 70–103. Beverly Hills, CA: Sage Brickman P. 1975. Adaptation level determinants of satisfaction with equal and unequal outcomes in skill and chance situations. J. Personal. Soc. Psychol. 32:191–98 Brockner J, Tyler TR, Cooper-Schneider R. 1992. The inf luence of prior commitment to an institution on reactions to perceived unfairness: The higher they are, the harder they fall. Adm. Sci. Q. 37:241–61 Brockner J, Wiesenfeld BM. 1996. An integrative framework for explaining reactions to decisions: interactive effects of outcomes and procedures. Psychol. Bull. 120:189–208 Burton JW. 1986. The history of international conf lict resolution. In International Conflict Resolution: Theory and Practice, ed. EE Azar, JW Burton, pp.  40–55. Boulder, CO: Lynne Reinner Publ. Casper JD, Tyler TR, Fisher B. 1988. Procedural justice in felony cases. Law Soc. Rev. 22:483–507 Chebat J-C, Slusarczyk W. 2005. How emotions mediate the effects of perceived justice on loyalty in service recovery situations: an empirical study. J. Bus. Res. 58:664–73 Cohn ES, White SO, Sanders J. 2000. Distributive and procedural justice in seven nations. Law Hum. Behav. 24:553–79 Colquitt JA. 2001. On the dimensionality of organizational justice: a construct validation of a measure. J. Appl. Psychol. 86:386–400 Conlon DE. 2012. Introduction to the special issue on justice, conf lict, and negotiation. Negot. Confl. Manag. Res. 5:1–3 Conlon DE, Carnevale P, Ross WH. 1994. The influence of third-party power and suggestions on negotiation: the surface value of a compromise. J. Appl. Soc. Psychol. 24:1084–113 Cook KS. 1975. Expectations, evaluations, and equity. Am. Sociol. Rev. 40:372–88 Cook KS, Hegtvedt KA. 1983. Distributive justice, equity, and equality. Annu. Rev. Sociol. 9:217–41 Deutsch M. 1985. Distributive Justice: A Social-Psychological Perspective. New Haven, CT: Yale Univ. Press Donohue WA, Hobbler GD. 2002. Relational frames and their ethical implications in international negotiation: an analysis based on the Oslo II negotiations. Int. Negot. 7:143–67 Druckman D. 1994. Determinants of compromising behavior in negotiation: a metaanalysis. J. Confl. Resolut. 38:507–56 Druckman D. 2005. Doing Research. Methods of Inquiry for Conflict Analysis. Thousand Oaks, CA: Sage Druckman D. 2006. Uses of a marathon exercise. In The Negotiator’s Fieldbook: The Desk Reference for the Experienced Negotiator, ed. AK Schneider, C Honeyman, pp. 645–56. Washington, DC: Am. Bar Assoc. Druckman D, Albin C. 2011. Distributive justice and the durability of peace agreements. Rev. Int. Stud. 37:1137–68 Druckman D, Olekalns M. 2013. Motivational primes, trust, and negotiator’s reaction to a crisis. J. Confl. Resolut. 57:966–90 Folger R, Konovsky MA. 1989. Effects of procedural and distributive justice on reactions to pay raise decisions.Acad. Manag. J. 32:115–30

378 Process and Outcome Justice

Garrett J, Libby WL. 1973. Intention and chance reward allocations in the dyad as determinants of equity.J. Personal. Soc. Psychol. 28:21–27 Gelfand MJ, Smith V, Raver J, Nishii L, O’Brien K. 2006. Negotiating relationally: the dynamics of the relational self in negotiations. Acad. Manag. Rev. 31:427–45 Goldman B, Pearsall M, Shapiro D. 2013. When is mediating employee grievances chosen versus rejected as a dispute-resolution procedure: an anticipatory justice perspective. Presented at Int. Assoc. Conf l. Manag., 26th, Takoma, WA Harmon D, Kim P. 2013. Trust repair via distributive justice rationales: the contingent implications of equity, equality, and need. Presented at Int. Assoc. Conf l. Manag., 26th, Takoma, WA Hauenstein NMA, Mcgonigle T, Flinder SW. 2001. A meta-analysis of the relationship between procedural justice and distributive justice: implications for justice research. Empl. Responsib. Rights J. 13:39–56 Hirschman AO. 1970. Exit, Voice, and Loyalty. Cambridge, MA: Harvard Univ. Press Hollander-Blumoff R, Tyler TR. 2008. Procedural justice in negotiation: procedural fairness, outcome acceptance, and integrative potential. Law Soc. Inq. 33:473–500 Holtz BC. 2013. Trust primacy: a model of the reciprocal relations between trust and perceived justice.J. Manag. 39:1891–923 Homans GC. 1961. Social Behavior: Its Elementary Forms. New York: Harcourt, Brace & World Hopmann PT. 1995. Two paradigms of negotiation: bargaining and problem solving. Ann. Am. Acad. Polit. Soc. Sci. 542:24–47 Irmer C, Druckman D. 2009. Explaining negotiation outcomes: process or context? Negot. Confl. Manag. Res. 2:209–35 Johnson DW, Johnson RT. 2012. Restorative justice in the classroom: necessary roles of cooperative context, constructive conf lict, and civic values. Negot. Confl. Manag. Res. 5:4–28 Joy VL, Witt LA. 1992. Delay of gratification as a moderator of the procedural justicedistributive justice relationship. Group Org. Manag. 17:297–308 Kabanoff B. 1991. Equity, equality, power, and conf lict. Acad. Manag. Rev. 16:416–41 Kapstein EB. 2008. Fairness considerations in world politics: lessons from international trade negotiations. Polit. Sci. Q. 123:229–45 Kelman HC. 2006. Interests, relationships, identities: three central issues for individuals and groups in negotiating their social environment. Annu. Rev. Psychol. 57:1–26 Kramer RM, Newton E, Pomeranke PL. 1993. Self-enhancement biases and negotiator judgment: effects of self-esteem and mood. Org. Behav. Hum. Decis. Proc. 56:110–33 Lerner MJ. 1974. The justice motive: “equity” and “parity” among children. J. Personal. Soc. Psychol. 29:539–50 Lerner MJ. 1980. The Belief in a Just World: A Fundamental Delusion. New York: Plenum Leventhal GS, Weiss T, Long G. 1969. Equity, reciprocity and reallocating rewards in the dyad. J. Personal. Soc. Psychol. 13(4):300–5 Lilja J. 2012. Trust and treason: social network structure as a source of f lexibility in peace negotiations. Negot. Confl. Manag. Res. 5:96–125 Lind EA, Kanfer R, Earley PC. 1990. Voice, control, and procedural justice: instrumental and noninstrumental concerns in fairness judgments. J. Personal. Soc. Psychol. 59:952–59 Lind EA, Tyler TR. 1988. The Social Psychology of Procedural Justice. New York: Plenum Machura S. 2003. Fairness, justice, and legitimacy: experiences of people’s judges in South Russia. Law Policy 25:123–50

Justice and Negotiation

379

Mikula G, ed. 1980. Justice and Social Interaction. New York: Springer Mikula G, Schwinger T. 1973. Liking towards one’s partner and need for social approval as determinants of allocations of group rewards. Psychol. Beiträge 15:396–407 Mikula G, Wenzel M. 2000. Justice and social conf lict. Int. J. Psychol. 35(2):126–35 Miller DT. 2001. Disrespect and the experience of injustice. Annu. Rev. Psychol. 52:527–53 Mittone L, Ploner M. 2012. Asset legitimacy and distributive justice in the dictator game: an experimental analysis. J. Behav. Decis. Mak. 25(2):135–42 Mulder M. 1977. The Daily Power Game. Leiden: Martinus Nijhoff Müller H. 2013. Conclusion: agency is central. In Norm Dynamics in Multilateral Arms Control: Interests, Conflicts, and Justice, ed. H Müller, C Wunderlich, pp.  337–65. Athens: Univ. GA Press Neu J. 2012. Pursuing justice in the midst of war: the International Criminal Tribunal for the Former Yugoslavia. Negot. Confl. Manag. Res. 5:72–95 Peterson RS. 1999. Can you have too much of a good thing? The limits of voice for improving satisfaction with leaders. Personal. Soc. Psychol. Bull. 25(3):313–24 Piaget J. 1948. The Moral Development of the Child. Glencoe, IL: Free Press Pruitt DG, Carnevale PJ. 1993. Negotiation in Social Conflict. Pacific Grove, CA: Brooks/ Cole Pruitt DG, Peirce RS, McGillicuddy NB, Welton GL, Castrianno LM. 1993. Long-term success in mediation. Law Hum. Behav. 17:313–30 Pruitt DG, Peirce RS, Zubek JM, Welton GL, Nochajski TH. 1990. Goal achievement, procedural justice, and the success of mediation. Int. J. Conf l. Manag. 1:33–45 Quimpo NG. 2001. Options in the pursuit of a just, comprehensive, and stable peace in the Southern Philippines. Asian Surv. 41(2):271–89 Rawls J. 1958. Justice as fairness. Philos. Rev. 67(2):164–94 Rosoux V. 2013. Is reconciliation negotiable? Int. Negot. 18:471–93 Rouhana NN. 2000. Interactive conf lict resolution: issues in theory, methodology, and evaluation. In International Conflict Resolution After the Cold War, ed. PC Stern, D Druckman, pp. 294–337. Washington, DC: Natl. Acad. Press Schuller RA, Hastings PA. 1996. What do disputants want? Preferences for third party resolution procedures. Can. J. Behav. Sci. 28:130–40 Solomon D, Druckman D. 1972. Age, representatives’ prior performance, and the distribution of winnings with teammates. Hum. Dev. 15(2):244–52 Spector BI, Zartman IW, eds. 2003. Getting It Done: Postagreement Negotiation and International Regimes. Washington, DC: US Inst. Peace Stedman SJ. 2000. Spoiler problems in peace processes. In International Conflict Resolution After the Cold War, ed. PC Stern, D Druckman, pp. 178–224. Washington, DC: Natl. Acad. Press Sugawara I, Huo YJ. 1994. Disputes in Japan: a cross-cultural test of the procedural justice model. Soc. Just. Res. 7:129–44 Swaab R, Postmes T, Neijens P, Keirs MH, Dumay ACM. 2002. Multi-party negotiation support: the role of visualization’s inf luence on the development of shared mental models. J. Manag. Inform. Syst. 19(1):129–50 Swaab R, Postmes T, van Beest I, Spears R. 2007. Shared cognition as a product of, and precursor to, shared identity in negotiations. Personal. Soc. Psychol. Bull. 33:187–99

380 Process and Outcome Justice

Thibaut J. 1968. The development of contractual norms in bargaining: replication and variation. J. Confl. Resolut. 12:102–12 Thompson LL, Wang J, Gunia BC. 2010. Negotiation. Annu. Rev. Psychol. 61:491–515 Törnblom KY. 1977. Magnitude and source of compensation in two situations of distributive injustice. Acta Sociol. 20:75–95 Tyler TR. 1987. Conditions leading to value-expressive effects in judgments of procedural justice: a test of four models. J Personal. Soc. Psychol. 52:333–44 Tyler TR. 1988. What is procedural justice? Criteria used by citizens to assess the fairness of legal procedures. Law Soc. Rev. 22:301–55 Tyler TR. 1994. Governing amid diversity: the effect of fair decision-making procedures on the legitimacy of government. Law Soc. Rev. 28:809–31 Tyler TR. 2000. Social justice: outcome and procedure. Int. J. Psychol. 35(2):117–25 Tyler TR, ed. 2005. Procedural Justice, vols. I and II. Burlington, VT: Ashgate Publ. Tyler TR, Blader SL. 2003. The group engagement model: procedural justice, social identity, and cooperative behavior. Personal. Soc. Psychol. Rev. 7:349–61 Tyler TR, McGraw KM. 1986. Ideology and the interpretation of personal experience: procedural justice and political quiescence. J. Soc. Issues 42:115–28 van den Bos K, Vermunt R, Wilke HAM. 1997. Procedural and distributive justice: What is fair depends more on what comes first than on what comes next. J. Personal. Soc. Psychol. 72(1):95–104 Wagner L, Druckman D. 2012. The role of justice in historical negotiations. Negot. Confl. Manag. Res. 5:49–71 Wagner L, Druckman D, .2015. Drivers of peace: the role of justice in negotiating civil war termination. Presented at Intl. Stud. Assoc. Conf., 56th, New Orleans Wagner LM. 2008. Problem-Solving and Bargaining in International Negotiations. Leiden: Martinus Nijhoff Publ. Walster E, Walster GW, Berscheid E. 1978. Equity: Theory and Research. Boston: Allyn & Bacon Wemmers JA, van der Leeden R, Steensma H. 1995. What is procedural justice: criteria used by Dutch victims to assess the fairness of criminal justice procedures. Soc. Justice Res. 8:329–50 Zartman IW, Druckman D, Jensen L, Pruitt DG, Young HP. 1996. Negotiation as a search for justice. Int. Negot. 1:79–98

PART 7

Conclusions

PART INTRODUCTION

This Part is divided into two chapters. First, I look back on the inf luences that shaped the career. Then I look forward to some recent research that increases our understanding of many of these ideas, showing how the past is prelude to a thoroughly modern social science informed by advances in theory and methods. While crafting this concluding Part, I came upon the realization that time is not a linear concept. Rather it seems to be circular in the sense of a weaving together of past memories, current moments, and future aspirations: Where we have been shapes our decisions in the present, which limits the options we have for the future. The shadows of the past and of the future are always in the present. It is this circularity that is referred to as lived experience.

DOI: 10.4324/9781003293361-34

19 LOOKING BACK

It seems like just yesterday when I was sitting in the library at Duke University pouring over the writing by early 20th century social philosophers and scientists. That was 1961–1962. Now, in the third decade of the 21st century, sitting at the same desk where I began writing this book, I am taken aback at the realization that many of those ideas shaped my research, consulting, and teaching career over the course of 56 years. In this chapter of the concluding Part, I discuss how these ideas emerged as cornerstones for the research and various forms of practice. The chapter is divided into two sections: “Relating to Classical Ideas” and “An Evidence-Based Approach.”

Relating to Classical Ideas Two broad ideas provide a kind of epistemological framework for the research on display in this book. One is the behavioral perspective introduced by Watson (1925). The other is Lewin’s (1951) principle of contemporaneity. The earlier experimental research and the more recent comparative case studies share the view that recording what we observe provides a more solid basis for accumulating knowledge than interpreting what we hear. This distinction is subject to debates between our colleagues who lean toward positivism – which prioritizes reliability of observations – and those who prefer phenomenological or constructivist philosophies, which prioritizes subjective data. But the distinction is not hard and fast. My colleagues and I have railed against turning these perspectives into ideologies (see Druckman, 2005, chapter 1). A good deal of our research has analyzed the role played by perceptions in explaining links between situations and behavior. This type of analysis has been aided by the advent of statistical mediation procedures (Hayes, 2022). Front and center in much of the research is a focus on situations. The preference for analyzing proximal inf luences on behavior stems less from conviction DOI: 10.4324/9781003293361-35

Looking Back 385

than from the quest for attributing cause to observed behavior. The research on turning points (TPs) is a case in point. Recall the framework developed to analyze TPs in negotiation: precipitants lead to departures, which produce consequences. The issue of causation loomed large in our coding and analyses. This was addressed by adhering to short periods of time between each part of the process trace: for example, a summit meeting is followed quickly by a new proposal, which produces a framework agreement. This line of thought was inspired by Lewin’s proposition and is supported by a large body of research reviewed in my earlier chapter on “Situations” (Druckman, 2008b). The research can be traced to Newcomb’s (1929) study of trait inconsistency and Hartshorne et al.’s (1930) study of “character” through to Frederickson’s (1971) typology of situations and Mischel and Shoda’s (1995) reconceptualization of personality and situations to my situational levers article (Chapter 9) in Part 2 of this collection (see also de Fornel and Quéré, 1999). Searching for a causal nexus between situations and behavior has deep roots in the social sciences, which continue in the current work on negotiation, identity, and justice. The foundation for many of the projects discussed in this book comes from other classical writing in the 1950s and 1960s. Simmel’s (1955) proposition on the intensifying inf luence of contrasting ideologies set the stage for the work on interests and values. Sherif ’s ideas about individual and group levels of analysis focus on products of an interaction process that includes “developing reciprocities among individuals, organizational structures, and group products, like social norms” (Sherif and Sherif, 1956: 342). A few years later Thibaut and Kelley (1959) developed a model of reciprocal roles that spurred our research on models of responsiveness in international negotiation. A more complex form of responsiveness is Goffman’s (1969) idea of expression games, where intentions are judged and attributions are made from nonverbal signals. This work was instrumental in our analyses of laboratory data on nonverbal communication. A type of nonverbal signaling is Schelling’s (1980) concept of focal points. This idea is incorporated in our turning points framework as a precipitant to departures in the negotiation process. (See also Druckman and Rosoux, 2016, on a comparison of focal and turning points.) In their 1965 book on labor negotiations, Walton and McKersie elucidated the dynamics of boundary roles, which I extended to the domain of international negotiation. Our research on robot mediation extends earlier work on the value of external interventions, such as the broken water tank, which served as a superordinate goal that increased cooperation between competing groups in the Sherif and colleagues field experiment (Sherif and Sherif, 1956, chapter 9).

An Evidence-based Approach An important contribution made by the research discussed in this book has been exploring many of these ideas systematically, benefiting from advances made in methodologies as discussed in Druckman (2005) and Druckman and Donohue

386

Conclusions

(2020). A comparison between the approaches taken in two well-known books on negotiation is instructive. These books were written by authors from the same generation and institutional contexts. The hugely popular prescriptive book by Fisher and Ury (1981) provides an example of compelling ideas that are not connected to bodies of empirical research. Indeed, there are no citations to research articles, not even to the forerunner of their “best alternative to a negotiated agreement” idea put forth a couple of decades earlier by Thibaut and Kelley (1959). Nonetheless it has sold over a million copies and has been translated into 18 languages. The excellent book – in my opinion a classic among classics – by Walton and McKersie (1965) provides a contrasting treatment of negotiation. This book is rooted deeply in research traditions and performs systematic analyses of a set of cases. It did not make it to a second edition. Written for a large and diverse audience, Fisher and Ury accomplished their purpose. Written for a scholarly audience, Walton and McKersie also accomplished their goal, garnering around 5,000 citations and providing the basis for a sequel by Walton et al. (1994). The collection of articles chosen for inclusion in this book have in common the kind of evidence-based approach to knowledge used by Walton and McKersie (1965). This approach is evident in both the empirical studies and reviews. Some advantages include public documentation, cumulation, and replication. A disadvantage is the frustration of the time it takes to go through the steps of design, data collection and analysis, writing, peer review, re-writing, and publication. We often ask during the course of a career – or a particular study – whether the results are worth the effort. There is little doubt about my answer to this question. For me the key is the idea of cumulation. At any point in time a body of knowledge has developed even while waiting for new studies to see the light of print. The review articles on turning points, nationalism, and justice are examples of what we know about these topics at the time of publication. Metaanalyses of bargaining studies provide a more precise approach to cumulating findings (Druckman, 1994; De Dreu et al., 2000). Graduate school matters in ways not appreciated until later in a career. I took for granted the design and methodological skills that were developed in the early years. These skills were honed on-the-job where I participated in various projects discussed in this book. More important than skills, however, was an attitude about methods f lexibility gained in my time with mentors from the psychology and political science faculties. This attitude encouraged me to move between quantitative (meta-analyses, statistical mediation) and qualitative (frameworks, focused comparisons) approaches. It also led to discovering new ways for applying “old” techniques without letting methods drive the substantive contributions made by projects. One of these methods was the pair comparison analysis used in the study on ethnocentrism (see Chapter 14). The technique was useful in the research on situational levers (see Chapter 9), in work on democratic peace theory (Cuhadar and Druckman, 2015), in consulting on developing tools for negotiating delegations, and in more recent work on motivations for identifying with political parties.

Looking Back 387

For our practice colleagues the question of interest is how to bridge the gap between research and practice. Our earlier discussion about training is one way to do this. The thematic narratives are intended to provide evidence-based advice for dealing with sticky problems often faced by diplomats. The advice takes the form of brief prescriptions: for example, discourage quick agreements by avoiding rapid concession exchanges. The primary take-away is about the bridge between research and practice. Evidence is a useful source of insights for practice. Practice provides useful experiences that stimulate new research. Evidence is also useful for policy-making. In the MBFR example discussed earlier, our team suggested solutions from analyzing patterns of movement made by members of the two alliances. We learned that a Warsaw Pact compromise was actually a negotiating tactic designed to extract NATO concessions (Druckman and Hopmann, 2002). Our recent research on the role of civil society in peace processes suggest that international commissions are the most propitious form of involvement for sustaining peace. On candidate Trump’s art of the deal negotiating prescriptions, we used research findings to critique his advice: for example, Trump’s suggestion to use your leverage as a more powerful party is challenged by Schelling’s (1980) “paradox of weakness”; his idea of fighting back is more likely to place the negotiators on a collision course. Our recent proposal to negotiate a way out of the tragedy in Ukraine is based on a willingness to embrace f lexibility by making concessions that would effectively end the destruction of Ukraine and its population. A sticking point in our proposal is to ensure guarantors to protect against further incursions (Druckman and Meerts, 2022). During the course of a research and practice career, I have becomes increasingly impressed with the value of social science as a research enterprise and as a collection of applied fields. Knowledge accumulation benefits from a process where research feeds on itself in the sense of cumulating findings. Ample illustrations of applications to skills training and policy-making drive the point home that we trade the security of our secluded ivory towers for attempting to inf luence how non-social scientists conduct their careers and live their lives.

20 LOOKING FORWARD

In this final chapter I ask about the shadow of future research on negotiation, identity, and justice. Do recent articles on these topics suggest ways forward? This question is addressed for each of five themes.

Ingroup (Partisan) Bias Three articles provide promising ideas for a next generation of research on ingroup biases in intergroup conf lict. An older article by Scheepers et al. (2006) raises two issues not taken up in recent studies. One issue deals with functions served by ingroup bias. On the one hand the bias may be functional in serving group goals. This is likely in competitive situations where group cohesion enhances group performance. On the other hand, the bias may be dysfunctional when expressed in an unstable intergroup context. An example of the latter is the war in Ukraine where ingroup-favoring biases have escalated the conf lict leading to increasing deaths of Ukrainian citizens. Another issue deals with the role played by structural factors such as legitimacy, power, and access to resources. In an intergroup context, these factors are understood in terms of asymmetries. The research and practice challenges are to design negotiating situations that encourage more powerful or legitimate parties to use their advantage constructively by carving out fair solutions. The more recent article by Bohm et al. (2020) reviews what the ingroup-bias literature has shown and where it is heading. Their suggestions for three types of interventions that can de-bias intergroup relations are particularly interesting. One consists of changing incentives by emphasizing the rewards that come from doing things together. Another is to change social categorizations through social contact or by increasing the salience of other shared group memberships. A third is to change perceptions of opposing interests by reducing the illusion of morality DOI: 10.4324/9781003293361-36

Looking Forward 389

as self-interest. One way to do this is to encourage a calculation of self-interest benefits along with the benefits to others. A third article comes from American politics. Abramowitz and Webster (2018) noted a rise in negative partisanship, which they claim has been responsible for polarization in the American electorate. Their analyses of electoral data show that the divisions are based more on dislike for the opposing political party than liking for one’s own party. With regard to ingroup bias, this refers to the difference between liking and disliking as the primary source for the strength of the bias. Yet, despite the large volume of research on bias, there are virtually no experiments that contrast the two sources. With colleagues, I have begun to probe the impacts of these sources. As the research develops, we will learn about the conditions where disliking for the other group is stronger than liking for one’s own group. Together, the three articles chart paths for expanding our knowledge about ingroup bias, which started with the ethnocentrism article presented in Chapter 14.

Values and Interests The key challenge in this line of research is how to deal with differences in values, including ideologies and worldviews. The approaches have been divided between confronting contrasting values directly or working around them by focusing on “smaller” issues. A host of interventions have been evaluated as discussed in Part 3. At the heart of the problem is the extent to which the value differences are perceived as threats to identities (Druckman, 2020). The stronger the threat, the less likely that conf licts will be resolved or even settled. Strong identity threats are best dealt with by working around the values. Disputants retain their value-based identities while cooperating on their shared interests. (See Druckman and Meerts, 2022, for a proposal to negotiate an end to the Ukraine invasion by working around the strong identities.) Superordinate goals, delinking strategies, and other-affirmation exercises are examples. Less is known about how these effects occur. Alternative explanations are increased perspective taking, willingness to engage in log-rolling, increased pro-social motivation, or enhanced commitment to problem-solving (Harinck and Druckman, 2017). These are the ideas that open new research frontiers for studying value conf licts.

Justice Justice is a fertile area for research. Our Annual Review article in Part 6 (Chapter 18) offers 22 hypotheses organized by negotiation stages. They suggest a considerable number of gaps to be filled. Standing out from this set are three questions. One is the need for clarification about causal chains involving trust, identities, and justice. The confusion is whether trust precedes or follows just procedures and outcomes. More likely the dynamics take the form of circular rather than directional relationships, similar to the interplay between interests and values.

390

Conclusions

Another gap is on the under-studied issues of compensatory justice. We know little about how compensation works to offset power asymmetries and unfair outcomes. A third gap concerns transitional justice. Attempts to implement agreements are fraught with the dilemma of balancing restorative justice with forward-looking changes in governing and security institutions. The question is this: Must issues of forgiveness be resolved before progress is made on creating a more just society? Welch (2022) has recently raised the provocative issue of a universal definition of justice. For him this consists of a discrepancy between entitlements (what people expect to receive) and benefits (what they actually receive). While recognizing the value of shared definitions for a general theory of justice, we also recognize the burden of contested meanings of these concepts in practice. Negotiators can be stuck in bitter quarrels about their entitlements before they come to terms with their respective benefits. We have here an interesting question for research: What are the processes for resolving entitlement issues? How do these issues block attempts to reach agreements?

Boundary Roles A rich research literature has developed on boundary roles since 1967, when I published my first article on this topic. The following 25 years of research is reviewed in a meta-analysis, where I showed that group representation is not a strong constraint on reaching agreements ( Druckman, 1994). Since then the experimental literature on this topic has turned towards impacts of different kinds of constituencies. In summary, constituents’ attitude – whether hawkish or dovish, symmetric or asymmetric or with regard to ethical conduct – matters (e.g., Aaldering and Kopelman, 2022). Yet questions remain. One concerns the strength of the representative’s connection to the group: strongly connected representatives are likely to be more inf luenced by these orientations. Another concerns the communication process between constituents and their representatives: The further the constituents are from the negotiation process, the easier it is for the representative to justify the agreement in terms compatible with constituents’ views. A third issue concerns the long-term or relational consequences of the agreement: Hawkish constituents may jeopardize relationships. Thus, future research questions about representative-constituency dynamics deal with connection, communication, and relationships. The divided constituencies in the laboratory can be considered as an analogue to the divisions within a nation. Similar identity issues exist at both levels as discussed in the scaling paper (see Chapter 16). However, constituencies are much larger and more diverse at the national level. Issues of durability, spread, and the type of governing regime come into play. Although some research has been done on impacts of these factors on decisions to mobilize for war (Cuhadar and Druckman, 2015), we know little about how representatives weigh these factors when facing decisions about international violence or humanitarian crises. We also

Looking Forward 391

know little about the role of representatives in sustaining the changes wrought by agreements. This is a topic that connects representation with peace-building.

Levels of Analysis A theme that runs through the research discussed in this book is levels of analysis. From the early laboratory analyses of individual and dyadic decisions to the idea of negotiation as a microcosm of national policy, we have been keen on understanding how connections occur between micro and macro levels. In the turning points research, the levels concept is used as separate lenses for analysis (Druckman, 2004). A different set of questions are posed at each level: for example, questions about interpreting the other’s moves, questions about synchronizing interactions, and questions about institutional norms and rules. In the work on peace agreements, questions about sequences from small groups to societal change are posed. Less is known about circular causation where changed societal institutions feed back on negotiation processes. The circularity issue is confronted directly in the paper with Steve Wood on national identity (see Chapter 16). Many examples are given about top down (from collective to individual decisions), bottom up (from negotiations to national policy), and simultaneous (workshops with national policy makers) inf luences. As noted at the end of that paper, new lines of inquiry should focus on mechanisms, for example, the way that group or community dynamics connect citizens’ attitudes to societal norms. As the most recent formulation of levels of analysis, the scaling paper (Chapter 16) points to the discoveries that await the next generation of research on negotiation, identity, and justice. Looking forward, I hope that readers will take away some of the lessons learned from this career. The lessons f low from taking a long-term view of a career. This allows for f lexibility, perseverance, and resilience. By f lexibility, I refer to both adaptation and agency. Adaptation is needed when the changes are largely out of the researcher’s control. Agency is needed for creating opportunities to pursue research goals in the context of change. By perseverance, I refer to the grinding feature of research with a view toward continuous improvement. Research is a process, not a project. By resilience, I refer to treating disappointments and perceived crises as opportunities. Recall that negotiating crises precipitated turning points in the international cases that we analyzed. The spiral career recounted in this attempt at prosopography has been full of turning points. Looking forward to a few more in my current projects and many in your blossoming careers. Bringing these career experiences to the present, I ask about lessons learned for current PhD graduates. An earlier attempt to answer this question was made in an article about consulting careers written 22 years ago (Druckman, 2000). Here I update that advice by considering new challenges wrought by technological progress and changes in academic job markets. Current graduate-student cohorts and their mentors are aware of these challenges. Less clear is whether

392

Conclusions

they are prepared to provide useful advice. The career depicted in this book provides lessons that can be valuable. Moving between research and consulting requires adapting to change. This means keeping up with, and sometimes mastering, innovations in ways for doing analyses. In the consulting phases, I was keen on pleasing the sponsors and contributing to the academic literature: Questions asked had both problem-solving and theoretical implications. Pulling these dual challenges off required a f lexible mindset. I rarely knew where my paths would lead, but the mindset prepared me for adapting quickly to change and playing a role in fostering those changes. The past two decades have seen dramatic changes in the pace of new research technologies: for example, the big-data revolution, advances in causal modeling, dynamical systems analyses, and machine learning (see Druckman and Donohue, 2020). As a result, current social science graduates have an advantage. They can move more nimbly among the various research and consulting opportunities that come their way. Unlike my generation, these graduates are not limited to making a choice between being a professor or carving out an applied career. They can do both. To make this happen, however, requires developing the mindset and skills needed in modern graduate programs (see Druckman and Donohue, 2022). In wrapping up the journey, I return to the first-year graduate school term paper mentioned early in the introductory chapter. That paper develops the idea that people’s internal lives ref lect the external world in which they live. The career described in this book is an example of this connection. Each of the research streams discussed was situated in the working spaces that I occupied. The bargaining experiments were performed in a professional atmosphere that encouraged basic research. The work on international negotiation cases was done in the context of State and Defense department consulting. The justice research was conducted in academic climates that encouraged continuing grant support. On the one hand, I was responding to the structural imperatives of the institutions. This is compatible with a behavioral philosophy. On the other hand, I was developing an internal cognitive template that ref lected the institutional environments. This is compatible with the theme of the term paper. Indeed, this two-pronged philosophy has sustained a 56-year social science career.

REFERENCES1

Aaldering, H. and De Dreu, C.W. (2012). Why hawks f ly higher than doves: Intragroup conf lict in representative negotiations. Group Processes and Intergroup Relations 15: 713–724. Aaldering, H. and Kopelman, S. (2022). Dovish and hawkish inf luence in distributive and integrative negotiations: The role of (A)symmetry in constituencies. Group Decision and Negotiation 31 (1): 111–136. Abramowitz, A.I. and Webster, S.W. (2018). Negative partisanship: Why Americans dislike parties but behave like rabid partisans. Advances in Political Psychology 39 (Supplement 1). Adorno, T.W., Frenkel-Brunswik, E., Levinson, D.J. and Sanford, R.W. (1950). The Authoritarian Personality. New York: WW Norton. Albin, C. and Druckman, D. (2012). Equality matters: Negotiating an end to civil wars. Journal of Conflict Resolution 56 (2): 155–182. Albin, C. and Druckman, D. (2014a). Bargaining over weapons: Justice and effectiveness in arms control negotiations. International Negotiation 19 (3): 426–458. Albin, C. and Druckman, D. (2014b). Procedures matter: Justice and effectiveness in international trade negotiations. European Journal of International Relations 20 (4): 1014–1042. Albin, C. and Druckman, D. (2017). Negotiating effectively: Justice in international environmental negotiations. Group Decision and Negotiation 26 (1): 93–113. Aubert, V. (1963). Competition and dissensus: Two types of conf lict and conf lict resolution. Journal of Conflict Resolution 7 (1): 26–42. Axelrod, R. (1980). Effective choice in the prisoner’s dilemma. Journal of Conflict Resolution 24 (1): 3–25. Bass, B. (1966). Effects on the subsequent performance of negotiators of studying issues or planning strategies alone or in groups. Psychological Monographs. Whole Number 614. Bendahmane, D.B. and McDonald, J.W. (1986). Perspectives on Negotiations: Four Case Studies and Interpretations. Washington, DC: Foreign Service Institute. Blake, R.R. and Mouton, J.S. (1961). Loyalty of representatives to ingroup positions during intergroup competition. Sociometry 24 (2): 177–183.

394

References

Bohm, R., Rusch, H. and Baron, J. (2020). The psychology of intergroup conf lict: A review of theories and measures. Journal of Economic Behavior & Organization 178: 947–962. Brehmer, B. and Hammond, K.R. (1977). Cognitive factors in interpersonal conf lict. In D. Druckman (ed) Negotiations: Social-Psychological Perspectives. Beverly Hills, CA: Sage. Clarke, R.A. (2004). Against All Enemies: Inside America’s War on Terror. New York: The Free Press. Coddington, A. (1968). Theories of the Bargaining Process. Chicago: Aldine. Coser, L. (1956). The Functions of Social Conflict. New York: The Free Press. Crump, L. and Druckman, D. (2016). Turning points and international environments: Multilateral negotiations in the GATT and the WTO. International Negotiation 21 (1): 1–42. Cuhadar, E. and Druckman, D. (2015). Representative decision-making: Challenges to democratic peace theory. In M. Galluccio (ed) Handbook of International Negotiation. Dordrecht, NL: Springer. De Dreu, C.K., Weingart, L.R. and Kwon, S. (2000). Inf luence of social motives on integrative negotiation: A meta-analytic review and test of two theories. Journal of Personality and Social Psychology 78 (5): 889–905. de Fornel, M. and Quéré, L. (eds) (1999). La Logique des Situations. Nouveaux Regards sur L’écologie des Activités Sociales. Paris: Editions de l’EHESS. Deutsch, M. (1985). Distributive Justice. New Haven, CT: Yale. Downs, G. and Stedman, S.J. (2002). Evaluation issues in peace implementation. In S.J. Stedman, D. Rothchild, and E. Cousins (eds) Ending Civil Wars: The Implementation of Peace Agreements. Boulder, CO: Lynne Rienner. Driver, M.J. (1982). Career concepts: A new approach to career research. In R. Katz (ed) Career Issues in Human Resource Management. Englewood Cliffs, NJ: Prentice-Hall. Druckman, D. (1967). Dogmatism, pre-negotiation experience, and simulated group representation as determinants of dyadic behavior in a bargaining situation. Journal of Personality and Social Psychology 6 (3): 279–290. Druckman, D. (1970). Position change in cognitive conf lict as a function of the cuecriterion relationship and the initial conf lict. Psychonomic Science 19: 91–99. Druckman, D. (1971a). On the effects of group representation. Journal of Personality and Social Psychology 18 (2): 273–274. Druckman, D. (1971b). The inf luence of the situation in inter-party conf lict. Journal of Conflict Resolution 15 (4): 523–554. Druckman, D. (1973). Human Factors in International Negotiation: Social-Psychological Aspects of International Conflict. Sage Professional Paper in International Studies 2, 02–020. Beverly Hills and London: Sage. Druckman, D. (1977). Boundary role conf lict: Negotiation as dual responsiveness. Journal of Conflict Resolution 21 (4): 639–662. Druckman, D. (1978). The monitoring function in negotiation: Two models of responsiveness. In H. Sauermann (ed) Bargaining Behaviour. Tubingen, Germany: J.C.B. Mohr. Druckman, D. (1980). Social-psychological factors in regional politics. In W.J. Feld and G. Boyd (eds) Comparative Regional Systems. New York: Pergamon. Druckman, D. (1990). The social psychology of arms control and reciprocation. Political Psychology 11 (3): 553–581. Druckman, D. (1993). The situational levers of negotiating f lexibility. Journal of Conflict Resolution 37 (2): 236–276.

References

395

Druckman, D. (1994). Determinants of compromising behavior in negotiation: A metaanalysis. Journal of Conflict Resolution 38 (3): 507–556. Druckman, D. (1995). Social-psychological aspects of nationalism. In J.L. Cameroff and P.C. Stern (eds) Perspectives on Nationalism and War. Luxembourg: Gordon and Breach. Druckman, D. (2000). The social scientist as consultant. American Behavioral Scientist 43: 1565–1577. Druckman, D. (2001a). Negotiation and identity: Implications for negotiation theory. International Negotiation 6 (2): 281–291. Druckman, D. (2001b). Nationalism and war: A social-psychological perspective. In D.J. Christie, R.V. Wagner and D.D. Winter (eds) Peace, Conflict, and Violence. Upper Saddle River, NJ: Prentice Hall. Druckman, D. (2002). Settlements and resolutions: Consequences of negotiation processes in the laboratory and in the field. International Negotiation 7 (3): 313–338. Druckman, D. (2003a). Linking micro and macro-level processes: Interaction analysis in context. The International Journal of Conflict Management 14 (3/4): 177–190. Druckman, D. (2003b). Puzzles in search of researchers: Processes, identities, and situations. The International Journal of Conflict Management 14 (1): 3–22. Druckman, D. (2004). Departures in negotiation: Extensions and new directions. Negotiation Journal 20 (2): 185–204. Druckman, D. (2005). Doing Research: Methods of Inquiry for Conflict Analysis. Thousand Oaks, CA: Sage. Druckman, D. (2006a). Group attachments in negotiation and collective action. International Negotiation 11 (2): 229–252. Druckman, D. (2006b). Uses of a marathon exercise. In A.K. Schneider and C. Honeyman (eds) The Negotiator’s Fieldbook. Washington, DC: American Bar Association. Druckman, D. (2007). Negotiating in the international context. In I.W. Zartman (ed) Peacemaking in International Conflict: Methods & Techniques. Washington, DC: United States Institute of Peace. Druckman, D. (2008a). Toward integrated knowledge. In S. Cheldelin, D. Druckman and L. Fast (eds) Conflict. New York: Continuum. Druckman, D. (2008b). Situations. In S. Cheldelin, D. Druckman and L. Fast (eds) Conflict. New York: Continuum. Druckman, D. (2010). Frameworks, cases, and experiments: Bridging theory with practice. International Negotiation 15 (2): 163–186. Druckman, D. (2015). Negotiating as a group representative: Constraints and opportunities. International Negotiation 20 (1): 25–40. Druckman, D. (2020). Looking back and looking forward in negotiation and conf lict research. International Negotiation 25 (1): 130–150. Druckman, D., Adrian, L., Damholdt, M.F., Filzmolser, M., Koszegi, S.T., Seibt, J. and Vestergaard, C. (2021). Who is best at mediating a social conf lict? Comparing robots, screens, and humans. Group Decision and Negotiation 30 (2): 395–426. Druckman, D. and Albin, C. (2011). Distributive justice and the durability of peace agreements. Review of International Studies 37 (3): 1137–1168. Druckman, D., Ali, F. and Bagur, S. (1974). Determinants of stereotyping in three cultures. International Journal of Psychology 9 (4): 293-302. Druckman, D., Broome, B. and Korper, S. (1988). Value differences and conf lict resolution: Facilitation or de-linking? Journal of Conflict Resolution 32 (3): 489–510. Druckman, D. and Donohue, W. (2020). Innovations in social science methodologies: An overview. American Behavioural Scientist 64 (1): 3–18.

396

References

Druckman, D. and Donohue, W. A. (2022). Working together: Bridging the researcherpractitioner gap. Negotiation and Conflict Management Research. 15(3). doi: https://doi. org/10.34891/20220406-435 Druckman, D., Druckman, J. and Arai, T. (2004). e-Mediation: Evaluating the impacts of an electronic mediator on negotiating behavior. Group Decision and Negotiation 13 (6): 481–511. Druckman, D. and Ebner, N. (2018). Discovery learning in management education: Design and case analysis. Journal of Management Education 42 (3): 347–374. Druckman, D. and Green, J. (1986). Political Stability in the Philippines: Framework and Analysis. Denver, CO: Denver Monograph Series in World Affairs 22, No. 3. Druckman, D. and Green, J. (1995). Playing two games: Internal negotiations in the Philippines. In I.W. Zartman (ed) Elusive Peace: Negotiating an End to Civil Wars. Washington, DC: Brookings. Druckman, D. and Harinck, F. (2022). Trust matters in negotiation. Group Decision and Negotiation 31 (6): 1179–1202. Druckman, D. and Harris, R. (1990). Alternative models of responsiveness in international negotiation. Journal of Conflict Resolution 34 (2): 234–251. Druckman, D., Harris, R. and Ramberg, B. (2002). Computer-assisted international negotiation: A tool for research and practice. Group Decision and Negotiation 11 (3): 231–256. Druckman, D. and Hopmann, P.T. (2002). Content analysis. In V.A. Kremenyuk (ed) International Negotiation. San Francisco: Jossey-Bass. Druckman, D., Husbands, J.L. and Johnston, K. (1991). Turning points in the INF negotiations. Negotiation Journal 7 (1): 55–67. Druckman, D., Karis, D. and Donchin, E. (1983). Information-processing in bargaining: Reactions to an opponent’s shift in concession strategy. In R. Tietz (ed) Aspiration Levels in Bargaining and Economic Decision Making. Berlin: Springer. Druckman, D. and Ludwig, L.D. (1970). Consensus on evaluative descriptions of one’s own nation, its allies, and its enemies. Journal of Social Psychology 81 (2): 223–234. Druckman, D., Martin, J., Allen Nan, S. and Yagcioglu, D. (1999). Dimensions of international negotiation: A test of Ikle’s typology. Group Decision and Negotiation 8 (2): 89–108. Druckman, D. and Meerts, P. (2022). War or words: How to end the Russian invasion in Ukraine. Young Diplomat Review: Building the Bridge Between Theory and Practice. April, 2022. https://www.youngdiplomat.org/post/war-or-words-how-to-end-theinvasion-of-russia-in-ukraine Druckman, D. and Mitchell, C. (1995). Flexibility in negotiation and mediation. The Annals of the American Academy of Political and Social Science 542: 10–23. Druckman, D., Mitterhofer, R., Filzmoser, M. and Koeszegi, S.T. (2014). Resolving impasses in e-negotiation: Does e-mediation work? Group Decision and Negotiation 23 (2): 193–210. Druckman, D. and Olekalns, M. (2013). Punctuated negotiations: Transitions, interruptions, and turning points. In M. Olekalns and W.L. Adair (eds) Handbook of Research on Negotiation. Cheltenham: Edward Elgar. Druckman, D., Parlamis, J. and Burns, Z. (2022). Can loyal party members be f lexible negotiators? Impacts of constituent support, term limits, and bipartisan roles. The International Journal of Conflict Management 33 (5): 741–761. Druckman, D. and Robinson, V. (1998). From research to application: Utilizing research findings in negotiation training programs. International Negotiation 3 (1): 7–38. Druckman, D. and Rosoux, V. (2016). Focal points and turning points in negotiation: A comparative analysis. Negotiation Journal 32 (2): 127–150.

References

397

Druckman, D., Rozelle, R. and Baxter, J. (1982). Nonverbal Communication: Survey, Theory, and Research. Beverly Hills, CA: Sage. Druckman, D., Rozelle, R., Krause, R. and Mahoney, R. (1974b). Power and utilities in a simulated interreligious council: A situational approach to interparty decision making. In J. Tedeschi (ed) Perspectives on Social Power. Chicago: Aldine. Druckman, D., Rozelle, R. and Zechmeister, K. (1977) “Conf lict of Interest and Value Dissensus: Two Perspectives.” In D. Druckman (ed) Negotiations: Social-Psychological Perspectives. Beverly Hills: Sage. Druckman, D. and Vaurio, E. (1983). Regimes and selection of political and military leaders: Brazilian cabinet ministers and generals. Journal of Political and Military Sociology 11 (2): 301–324. Druckman, D. and Wagner, L. (2016). Justice and negotiation. Annual Review of Psychology 67: 387–413. Druckman, D. and Wagner, L. (2019). Justice matters: Peace negotiations, stable agreements, and durable peace. Journal of Conflict Resolution 63 (2): 287–316. Druckman, D. and Wallensteen, P. (2016). Summit meetings: Good or bad for peace? Global Summitry: Politics, Economics, and Law in International Governance 2 (2): 71–92. Druckman, D. and Zechmeister, K. (1973). Conf lict of interest and value dissensus: Propositions in the sociology of conf lict. Human Relations 26 (4): 449–466. Durkhein, E. (1895). Les Regles de la Methode Sociologique. Paris: Felix Alcan. Ericsson, A. and Pool, R. (2016). Peak: Secrets from the New Science of Expertise. Boston: Houghton Miff lin Harcourt. Ferster, C.B. and Perrott, M.C. (1968). Behavior Principles. New York: New Century. Fisher, R. and Ury, W. (1981). Getting to Yes: Negotiating Agreement Without Giving In. Boston: Houghton Miff lin. Follett, M.P. (1942). Dynamic Administration: The Collected Papers of Mary Parker Follett. Eds. H.C. Metcalf and L. Urwick. New York: Harper & Row. Frederickson, N. (1971). Toward a taxonomy of situations. Presented at the Annual Meeting of the American Psychological Association, Washington, DC. Gamson, W.A. (1968). Power and Discontent. Homewood, IL: The Dorsey Press. Gellner, E. (1983). Nations and Nationalism. Ithaca, NY: Cornell. Goffman, E. (1959). The Presentation of Self in Every Day Life. New York: Doubleday & Company. Goffman, E. (1969). Strategic Interaction. Ph iladelphia: University of Pennsylvania Press. Guilford, J.P. (1954). Psychometric Methods. New York: McGraw-Hill. Guner, S. and Druckman, D. (2000). Identification of a princess under incomplete information: An Amarna story. Theory and Decision 48 (4): 383–407. Haas, E.B. (1997). Nationalism, Liberalism, and Progress: The Rise and Decline of Nationalism. Ithaca, NY: Cornell. Harinck, F. and Druckman, D. (2017). Do negotiation interventions matter? Resolving conf licting interests and values. Journal of Conflict Resolution 61 (1): 29–55. Harinck, F. and Druckman, D. (2019). Values and interests: Impacts of affirming the other and mediation on settlements. Group Decision and Negotiation 28 (3): 453–474. Hartshorne, H., May, M.A. and Shuttleworth, F.K. (1930). Studies in the Nature of Character. Volume 2: Studies in Organizational Character. New York: Macmillan. Hayes, A.F. (2022). Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach (3rd ed.). New York: The Guilford Press. Hogg, M.A. and Mullin, B.A. (1999). Joining groups to reduce uncertainty: Subjective uncertainty reduction and group identification. In D. Abrams and M.A. Hogg (eds) Social Identity and Social Cognition. Oxford: Blackwell. Homans, G.C. (1950). The Human Group. New York: Harcourt, Brace and Company.

398

References

Hopmann, P.T. (1977). Bargaining within and between alliances on MBFR: Perceptions and interactions. Presented at the Annual Meeting of the International Studies Association, St. Louis MO. Hopmann, P.T. (1995). Two paradigms of negotiation: Bargaining and problem solving. The Annals of the Academy of Political and Social Science 542: 24–47. Hopmann, P.T. and Druckman, D. (1981). Henry Kissinger as strategist and tactician in the Middle East. In J.Z. Rubin (ed) Dynamics of Third Party Intervention: Kissinger in the Middle East. New York: Praeger. Iklé, F. (1964). How Nations Negotiate. New York: Harper & Row. Irmer, C. and Druckman, D. (2009). Explaining negotiation outcomes: Process or context? Negotiation and Conflict Management Research 2 (3): 209–235. Kedourie, E. (1993). Nationalism (4th ed.). Oxford: Blackwell. Kelman, H.C. (1972). The problem-solving workshop in conf lict resolution. In R.L. Merritt (ed) Communication in International Politics. Urbana: University of Illinois Press. Kimble, G.A. (1961). Hilgard and Marquis’ Conditioning and Learning. New York: AppletonCentury-Crofts. Krause, R., Druckman, D., Rozelle, R. and Mahoney, R. (1975). Components of value and representation in coalition formation. Journal of Peace Science 1 (2): 141–158. Lall, A. (1966). Modern International Negotiation: Principles and Practice. New York: Columbia University Press. Leary, K. (2004). Critical moments in negotiation. Negotiation Journal 24 (2): 143–145. LeVine, R.A. and Campbell, D.T. (1972). Ethnocentrism: Theories of Conflict, Ethnic Attitudes and Group Behavior. New York: John Wiley & Sons. Lewin, K. (1951). Field Theory in Social Science: Selected Theoretical Papers. ( Edited by Dorwin Cartwright) New York: Harper. Lind, E.A. and Tyler, T.R. (1988). The Social Psychology of Procedural Justice. New York: Plenum Press. McClelland, C.A. (1972). The beginning, duration, and abatement of international crises: Comparisons in two conf lict arenas. In C.F. Hermann (ed) International Crises: Insights from Behavioral Research. New York: Free Press. McDonald, J.W. and Bendahmane, D.B. (1990). U.S. Bases Overseas: Negotiations with Spain, Greece, and the Philippines. Boulder, CO: Westview Press. Merritt, R.L. (1972). Effects of international student exchange. In R.L. Merritt (ed) Communication in International Politics. Urbana: University of Illinois Press. Mischel, W. and Shoda, V. (1995). A cognitive-affective theory of personality: Reconceptualizing situations, dispositions, dynamics, and invariance in personality structure. Psychological Review 102 (2): 246–268. Mittelmark, M.B., Rozelle, R.M. and Druckman, D. (1977). Accountability and role behavior of bargainers. Presented at the Annual Meeting of the American Psychological Association, San Francisco, CA. Monheim, K. (2015). How Effective Negotiation Management Promotes Multilateral Cooperation: The Power of Process in Climate, Trade, and Biosafety Negotiations. London: Routledge. Newcomb, T.M. (1929). The consistency of certain extrovert-introvert behavior patterns in 51 problem boys. Teachers College Contributions to Education 382: 123. Newhouse, J. (1973). Cold Dawn: The Story of SALT. New York: Holt. Osgood, C.E. (1962). An Alternative to War or Surrender. Champaign, IL: University of Illinois Press. Popper, K.R. (1959). The Logic of Scientific Discovery. Abington-on-Thames: Routledge. Pruitt, D.G. and Lewis, S.A. (1975). Development of integrative solutions in bilateral negotiations. Journal of Personality and Social Psychology 31 (4): 621–633.

References

399

Rokeach, M. (1960). The Open and Closed Mind. New York: Basic Books. Sawyer, J. and Guetzkow, H. (1965). Bargaining and negotiation in international relations. In H.C. Kelman (ed) International Behavior: A Social-Psychological Analysis. New York: Holt, Rinehart & Winston. Scheepers, D., Spears, R., Doosje, B. and Manstead, A.S. (2006). The social functions of ingroup bias: Creating, confirming, or changing social reality. European Review of Social Psychology 17 (1): 359–396. Schelling. T. (1980). The Strategy of Conflict. Cambridge, MA: Harvard University Press. Sherif, M. (1936). The Psychology of Social Norms. New York: Harper & Row. Sherif, M. and Sherif, C.W. (1956). An Outline of Social Psychology. New York: Harper & Row. Siegel, S. and Fouraker, L. E. (1960). Bargaining and Group Decision Making. New York: McGraw-Hill. Simmel, G. (1955). Conflict & The Web of Group-Affiliations. New York: The Free Press. Stoll, R.J. and McAndrew, W. (1986). Negotiating strategic arms control, 1969–1979: Modeling the bargaining process. Journal of Conflict Resolution 30 (2): 315–326. Strauss, A. (1956). The Social Psychology of George Herbert Mead. Chicago: University of Chicago Press. Tajfel, H. (1981). Human Groups and Social Categories: Studies in Social Psychology. Cambridge: Cambridge University Press. Thibaut, J.W. and Kelley, H.H. (1959). The Social Psychology of Groups. New York: John Wiley & Sons. Thibaut, J.W. and Walker, L. (1975). Procedural Justice: A Psychological Analysis. Hillsdale, NJ: Erlbaum. Turner, J.C. (1987). Rediscovering the Social Group: A Self-Categorization Theory. New York: Basil Blackwell. Walton, R.E., Cutcher-Gershenfeld, J.E. and McKersie, R.B. (1994). Strategic Negotiations: A Theory of Change in Labor – Management Relations. Boston, MA: Harvard Business School Press. Walton, R.E. and McKersie, R.B. (1965). A Behavioral Theory of Labor Negotiations: An Analysis of a Social Interaction System. New York: McGraw-Hill. Watson, J.B. (1925). Behaviorism. Chicago: University of Chicago Press. Welch, D. A. (2022). Understanding justice. International Negotiation (published online ahead of print 2022). doi: https://doi.org/10.1163/15718069-bja10074 Zartman, I.W. and Kremenyuk, V. (eds). (2005). Peace Versus Justice: Negotiating Forwardand Backward-Looking Outcomes. Lanham MD: Rowman & Littlefield. Zechmeister, K. and Druckman, D. (1973). Determinants of resolving a conf lict of interest: A simulation of political; decision making. Journal of Conflict Resolution 17 (1): 63–88.

Note 1 This is the complete set of references cited in the chapters with the exception of the reprinted articles. They contain their own references.

INDEX

Page numbers in italics indicate figures and page numbers in bold indicate tables. Abidjan Peace Agreement 347 Abramowitz, A. I. 291, 389 Abzug, B. 23 Academy for Educational Development (AED) 20–21, 44 accountability 77, 93–94, 257, 268, 293, 312 Adams, D. 261 Adams, J. S. 315, 366 Adelman, K. 24, 46, 288 Adjustment-Evaluation-Decision Cycle 84 Adorno, T. 55, 259 affirming the other 32, 123 agreement: distributive 19; distributive justice and 40– 43; f lexibility and 17, 100, 103, 108, 111–115; frameworks for 156–157; inf luences and 151–153; integrative 19; justice research and 4 –5, 10; non-standard 19; prenegotiation and 148–149; responsiveness and 95; robot mediators and 18–19; role-player theories and 108–109; situational levers and 111–112; turning points and 32, 149–150, 159, 177–178; unilateral position-formation and 70–71; value differences and 17 Agreement of Friendship and Cooperation (AFC) 155 Albin, C. 5, 40, 317, 320, 349, 358, 368 Ali, F. 35, 263 alliance negotiations 22–24, 77 Allport, G. A. 241

altercentrism 276 American Psychological Association 39n2 anticipatory justice 344, 357–358 Aquino, C. 10, 27, 31, 113, 123 arms control: nonproliferation norms 349; patterns of responsiveness and 155; procedural justice and 42; statistical analysis of 144–145; turning points and 28; US-Soviet 3, 23–24, 28; see also Intermediate-Range Nuclear Forces (INF ) Arms Control and Disarmament Agency (ACDA) 23–24, 33n2, 44– 45, 49n2 Aronson, E. 236, 247 Arrow, K. 47 Ashforth, B. 299 Aubert, V. 121, 125, 131–132 Australia 14, 48– 49 Australian National University (ANU) 14, 48 Australian Research Council 48 authoritarianism 37–38, 55, 259, 270, 293 Axelrod, R. 133, 139n6 Back, K. 290 Bagur, S. 35, 263 balance of power 27–28, 179 balancing perspective 123 Balcells, L. 316, 338 bargaining: concession-making and 107–108; conf lict resolution and 16–17;

Index

distributive 13, 17; f lexibility and 17; gaming experiments and 16; ideologies and 128; integrative 13, 17; internal dynamics in 81; media coverage and 99, 103, 106, 108–109, 114–115, 322; mediation and 18; negotiation process and 13, 152; principle of dissent 130; skills training and 33; turning points and 16; see also collective bargaining bargaining experiment 103, 105–107, 107 bargaining process analysis system 152, 205 Barrett-Howard, E. 356 Bartos, O. J. 80– 81 base rights: FSI research and 45; negotiation and 45; parallel negotiations and 30; Philippines negotiation 148; tactics and 30; threshold adjustment and 76; turning points and 3; use of language and 30; US-Spain talks 22–23, 29–30, 112, 114, 143–144, 155–157, 158, 159–168, 177–178, 205 Bass, B. 58, 72, 73n1, 74n8, 235, 304n3 Bazerman, M. 347 Beagle Channel, 1977–79 183, 191 behavioral approaches: f lexibility and 100; negotiating strategies and 136; negotiation process and 165, 169; observation and 384–385; research and 384; turning points and 166; variety notion and 171n7 behaviorism 12, 28, 99 beliefs 2 , 27, 236, 296, 298, 303, 361–362 benchmarks 148, 205 Bennett, A. 181 Bercovitch, J. 199n2 Berkowitz, L. 241 Berman, M. 84, 89, 96 bias: ethnocentrism and 35, 223, 227, 230, 238–239, 247, 278; halo error and 223; ingroup-outgroup 4, 35, 230–231, 238, 243, 260–262; selective 230, 238, 247; shifting alliances and 247 Biden, J. 303 bilateral monopoly experiment 80 bilateral negotiation 146, 171n1, 291; see also international negotiation bilateral-study condition 59– 60, 62, 66, 66, 67– 69, 71–72 Blader, S. L. 360 Blake, R. R. 17, 35, 60, 73, 73n1, 240, 267, 304n3 Bohm, R. 388 Bonoma, T. V. 80, 84– 85, 87 Booz Allen Hamilton 14, 46

401

boundary role conf licts (BRC): bidirectional responsiveness and 78; building a package 89–92; degree of agency 92–94; expectation-evaluationadjustment models 17; group representation and 16; interacting pressures and 95; monitoring function in 96; negotiator-as-bargainer model 79– 88, 96–97; negotiatoras-representative model 88–94, 96–97; reconceptualization of 94–96; responsiveness and 95 boundary roles: alliance negotiations and 77; concept of 76; constituent attitudes and 390; f lexibility and 2; international negotiation and 2 , 76–78, 385; representation and 77 Brazil 46 Brecher, M. 269 Bremer, B. 356 Brewer, G. 47 Brewer, M. B. 260, 274 Brexit 293, 301–302 Brody, R. A. 226, 228 Brown, B. R. 107 Brown, R. 261 Brubaker, R. 290 Bulska, D. 210 Bunker, E. 29 bureaucratic (governmental) politics 24, 74n4, 88, 96, 111 Burnight, R. G. 276 Cameroff, J. L. 37 Cameron, M. A. 178, 186 Campbell, D. T. 13, 34, 131–132, 223, 225–227, 233, 246, 248n1, 290 career development: academic roles and 14, 45, 48; AED and 20–21, 44; Australian higher education and 14, 48– 49; conf lict coaching and 15; consulting and 20–28, 33n1, 45– 48, 50, 391–392; f lexibility and 14; IJR and 44; international negotiation and 9 –10, 17, 20; LeVine-Campbell ethnocentrism project 13, 34; national identity and 10; negotiation and 13; NRC and 28, 37–38, 47– 48, 49n6, 50; research and 10–11, 50–51, 392; social sciences and 9, 12–13; study of human behavior and 9, 12; transitions and 14–15, 392 Carlsburg Foundation 18 case studies: durable peace (DP) 324–338; justice in negotiation 352–353; justice

402

Index

principles and 317, 320, 353; national identity and 293; responsiveness and 205; turning points in negotiation 177–199; U.S.-Spain base rights talks 146–171 Castro, F. 206 Catton, W. R., Jr. 240 Chamberlain, N. 206 Chasek, P., 199n2, 200n4, 210 Cherryholmes, C. H. 228 China 111, 191, 265, 269 Christie, D.J. 37 Chun, N. 230, 238 circularity 344, 391 civil society 295–296, 386 civil wars: data sets 327–328; institutionbuilding and 340n20; justice and 315; levels of analysis and 338; mediation efforts and 347; peace agreements and 315, 329, 336; peacekeeping operations (PKOs) and 323; political violence and 316; procedural justice and 317; reconciliation and 336; resumption of violence and 315, 323 Clarke, R. 45 Coddington, A. 16, 26, 31, 77, 79, 83, 116n2, 169, 206, 211 coercion 148, 293, 319, 370 cohesion 270, 291, 298, 360, 362 Cohn, E. S. 356 collaboration 44, 49n1, 346 collective action: attitudinal climate and 271; group loyalty and 267, 269–272; groups and 35–37; justice principles and 354; mobilization for 38, 252, 293, 298; negotiation for 354; policymaking and 271 collective bargaining: boundary roles and 76; conf lict resolution and 72, 74n8; dogmatism concept and 55; emotional attachment and 35; prenegotiation and 59, 61– 63; research and 2 , 57 collective behavior: expert effect 293–294; group loyalty and 254, 267–277, 292; groupthink and 294; holistic dynamical systems analysis and 295; individual loyalties and 271; majority effect 293–294; opinion and 293–294, 305n6; regional politics and 4; social network analysis (SNA) and 292; societal variation and 293; sociopolitical change and 295; stereotypes and 267 collective consciousness 222 common identity 297, 305n9; see also group identity

comparative reciprocity model 26, 144–145 compensation 348, 369, 390 compromising 16, 35, 101, 109, 160, 267 concession-making: arms control and 144; bargainer expectations and 83– 87; bargaining behavior and 107–108; f lexibility and 386; negotiator-asbargainer model 79– 80; responsiveness and 76, 79– 83, 87; security negotiation and 195–196; turning points and 150 Conference on the Committee on Disarmament (CCD) 25 conf lict: cognitive 17; context and 124, 197; ideologies and 27; impact on daily life 339n3; of understanding 125; intergroup 255, 254–258; internal/external 339n12; values and 2 , 17, 30–31; see also boundary role conf licts (BRC); conf lict resolution conf lict coaching 15 conf lict environment (CE) 323, 339n2, 339n3 conf lict of interest: communication opportunity and 124; competition and 126–127, 138n2; cross-cutting interests and 132–133, 135, 139n6; decision-latitude of representatives and 125; defining 124–125; ideologies and 124–137, 138n3; polarizing effects of 132, 135, 137, 139n5; political decision-making and 127, 139n3; prenegotiation experience and 125; principle of dissent 130; propositional format and 127–135; resistance to compromise and 124; social systems and 129; within-team differences 130–133, 139n4; values and 125–127 conf lict resolution: affirming the other 123; bargaining and 16–17; base-rights cases and 30; bilateral focus and 60, 62, 71–72; face-to-face interactions 37; game theory and 125; group loyalty and 59; nationalism and 37; negotiation and 30–32; opposition of interests and 16; prenegotiation and 59– 66, 66, 67– 69, 71–72; problem-solving 116n2, 317; realpolitik theory of war and 266; settlement-oriented 116n2; simulation approach and 123; team positionplanning and 59; unilateral focus and 72–73; see also group conf licts Conlon, D. E. 361 consequences: escalation/de-escalation 191–192, 192, 193–195, 198–199;

Index

negotiation process and 178, 181, 184, 208–209; positive/negative 185; precipitants and 189–190, 191, 192–195; process departures and 190, 191; sequence of 191 contemporaneity 384 content analysis 23–24, 45 context: collective action and 252; conf lict of interest and 124, 197; identity and 4; inf luences in negotiation 151–153, 195, 215, 323; national identity and 297, 302; negotiation process and 29; political 29–30, 139n3, 266, 278; simulations and 123; social behavior and 12–13 continual three-fold choice 26, 28, 31 Conventional Armed Forces in Europe (CAFE) 26 Convention on Early Notification of a Nuclear Accident 347 Cooper, F. 290 Cooperative Cross-Cultural Study of Ethnocentrism 13, 34, 223, 248n1 Copenhagen Climate Change Conference 347 Cortina, P. 156–157 Coser, L. 12, 121, 127–128, 139n5, 236 Cottam, R. W. 264–265, 267 creativity 19, 19n2 cross-cutting interests 132–134, 136, 138, 139n6 Crow, W. J. 226, 228–229, 248n1 Crowne, D.P. 74n4 Crump, L. 209, 214 Cuhadar, E. 15 cultures of peace 5, 9, 318 Cyprus Dispute 29 Dahrendorf, R. 133 Dean, J. 45 Defense Advanced Research Projects Agency (DARPA) 46 DeLamater, J. 256 Delanty, G. 297, 300 Deutsch, K. 300 Deutsch, M. 5, 40, 266, 316, 348–349, 366 Diehl, P.F. 14, 330 Diesing, P. 101 diplomacy 28–30, 84; see also shuttle diplomacy direct effects hypotheses 319–320, 331–334 directionality 344 directional model 144 disposition theory 243, 248

403

dissent 130, 273 dissonance-theory formulation 61 distributive bargaining 13, 17 distributive justice (DJ): compensation and 348; context and 349; durable peace and 319–322, 333, 334, 335–336, 338; economic growth and 317; effectiveness and 42; equality and 320, 348–350, 351, 357–358; equal outcomes and 5; fair allocation and 320; fairness and 316, 356; governance institutions and 317; justice principles and 354; need and 348; negotiation outcomes and 364–367; negotiation process and 317, 346–350, 375; outcomes and 351; peace agreements and 40– 41, 312, 317, 328–329, 349; principles of 41, 348–349; procedural justice and 352–354; proportionality (equity) 348–349, 351–352, 356–357, 366; reconciliation and 317; security institutions and 317; sentimental attachments and 230, 240; social conduct and 317; stability of agreements and 320; trade negotiation and 350; transparency and 316 Dobrynin, A. 24 Doha Declaration 209 Donchin, E. 49n3 Donohue, WA. 15, 214, 359, 385 double statistical mediation 58, 316, 318, 339n11 Downs, G. 40, 323–324, 339n2 Driver, M.J. 9 Druckman, D. 2 –3, 5, 16–17, 19n2, 27, 31, 33, 35, 47, 57, 59, 63– 64, 73n2, 80, 84– 85, 87, 96, 101, 126, 128–129, 146, 152, 171n4, 196, 199n2, 206–210, 221, 262–263, 268, 272, 286, 289–291, 302, 317–320, 330, 349, 350, 355, 357–359, 360–361, 368–369, 385 Druckman, J. 18 Druckman, M. 248n1 Duckitt, J. 259 Dunteman, G. 73n1, 235, 304n3 durable identity 293–294, 305n5 durable peace (DP): balance of power and 27; challenges of 315; community relations and 318; concept of 41; conf lict environment (CE) and 323, 339n3; direct effects hypotheses 319–320, 331–334; distributive justice and 319–322, 333, 334, 335–336, 338, 339n10; economic growth and

404

Index

330; equality principle and 320, 333, 349; fairness and 332–333; formal negotiations and 319; governance reform and 330; implementation environments and 325; indirect effects hypotheses 320–322, 334–335; institution-building and 330, 340n20; justice and 5, 28, 41; micro-level interactions and 302; negotiating context and 323; peace agreements and 41, 315; peacekeeping operations (PKOs) and 323, 340n21; procedural justice and 5, 287, 319–322, 332, 334, 336, 338, 339n10, 349; reconciliation and 319, 330–331, 332, 333, 334, 335–338; security institution reform and 330; serial multiple mediation path 321–322, 335–336, 338; stability of agreements and 322, 335–336, 338; statistical analysis of 331; theory development and 319; transitional justice and 339n10; transparency and 332–333; see also peace agreements Durkheim, E. 222, 289, 292 dynamic equilibrium models 137 Ebner, N. 33 economic growth: distributive justice and 333; durable peace and 41, 317, 330; peace agreements and 340n18 effective outcomes 42– 43, 353 Egypt 27 Eighteen-Nation Committee on Disarmament (ENCD) 25 e-mediation: development of 18; inference training and 46; negotiation and 18; robotic mediation and 18–19, 19n2 Emerson, R. M. 241 emotional attachment 35–36, 287 empirical approaches: conf lict of interest and 127–128, 137; group dynamics and 290; justice and 349, 352; national identity and 38; negotiating crises and 208; negotiating objectives and 25; negotiation process and 91, 154, 386; social sciences and 5; turning points and 177, 208 enemies: ethnocentric bias and 232–233, 238–239, 247, 269, 271; good vs. evil folk theory and 265–267; images of 265–267; stereotypes and 264–266; strong vs. weak 233, 239–240, 244, 247; theories of war and 265–267;

universal bias hypothesis and 223, 234–236 environmental negotiations: context and 317; precipitants and 210; procedural justice and 42– 43; process tracing and 200n4; progress in 179; role of justice in 347; shared expectations and 180; turning points in 177–178, 195 equality: distributive justice and 41, 315, 320, 348–350, 351, 356; durability and 349–350, 368; equal measures 41; equal shares 41; equal treatment 41; principle 333; mediation and 368–369; negotiation outcomes and 366–367, 375; negotiation process and 348; organizational structures and 357; procedural justice and 349, 358 equal outcomes: distributive justice and 5, 41; implementation success and 41; negotiation process and 317; procedural justice and 370–371; solidarity and 367; trust and 5 equity: compensation and 369; distributive justice and 348–349, 351–352, 356–357, 366; intense conf lict and 368; negotiation outcomes and 366–367, 375; organizational structures and 357; power differences and 357, 362–363; trust and 356 Ericsson, K. A. 213 ESP research 48 ethnic cleansing 269 ethnocentrism: alliance structure and 236, 247; bias and 35, 223, 227, 230–231, 238–239, 247; egocentricity and 231; enemy nations and 232–233; ethnophaulisms and 225, 248n2; as function of rater’s role 241–242, 242, 242, 243; group loyalty and 269, 273, 278; intergroup relations and 34–35, 226, 235, 243, 248n3; nationalism and 10, 233, 269; negative 276; overassimilation and 236; patriotism and 259; simulations and 34–35, 223, 230–236, 238–248, 248n3; as sociopsychological syndrome 225; status mobility and 241; stereotypes and 35; theory 225, 226, 243, 248, 248n3; theory of State and 225; universal bias hypothesis 223, 233–236, 243; universality and 226–227 Etzioni, A. 257 Europe: arms control and 10, 23, 26; cooperative systems and 37; nationalist

Index

ideology and 38; regional policy and 36, 301; social formation in 300 European Community 272, 276 European Union (EU): Brexit and 293, 301–302; development of 299; multilevel interactions and 287; national identity and 300–301, 303; Negotiation Club 43n1; regional policy and 36, 301; supra-national institutions and 299–301, 305n10; three-level concept of government in 300–301, 305n11 Evan, W. M. 130, 139n4 expectation-evaluation-adjustment models 17, 26, 31, 206 experiments: bargaining and 16, 103, 105–107, 107; boundary role conf licts (BRC) 78–97; demand characteristics and 245; frameworks and 98; logic of causation and 352; prenegotiation experience and 59–73; representation and 35; simulations and 245; situational levers and 100–116 expression games 385 fairness: anticipatory justice and 344; conf lict reduction and 368; distributive justice and 316, 356, 357; durable peace and 332–333; justice and 347; negotiation outcomes and 365, 367, 376; negotiation process and 206, 347; peace agreements and 316; procedural justice and 320, 328, 336, 350, 351, 360, 365; reconciliation and 320; social conduct and 320, 336; transparency and 5, 313; trust and 344 fair representation 350, 351 Falklands/Malvinas Islands Crisis 29, 183, 191 false justice 361–362 Federal Trade Commission (FTC) 48 Ferguson, C. K. 60 Feshbach, S. 258–260 Fishbein, M. 273 Fisher, R. 386 Fiske, S. T. 289 f lexibility: adaptation and 391; agency and 391; bargaining and 17, 105–107; compromising and 101; concessionmaking and 386; defining 100–101; diagnosing 110–115; dogmatism concept and 55; meta-analysis of 109–110; negotiation privacy and 98; overcoming constraints on 32;

405

simulation approach and 103–104, 108–109, 115; situational levers and 2 , 98–100, 103–104; studying the issues and 57; timing variable and 104, 115; turning points and 2 –3; values and 2; see also negotiating f lexibility focal points 17, 385 folk theory of war 265–267 Follett, M.P. 13 Ford, G. 159 Foreign Service Institute (FSI) 28–30, 45 Fortna, V. P. 323 Fouraker, L.E. 16 frameworks: continual three-fold choice 26, 28, 31; expectation-evaluationadjustment models 17, 26, 31; experiments and 98; Helsinki process 26; ICR interventions and 296–297; Iklé typology on negotiating objectives 25–26; international negotiation and 21, 22; Justice and Negotiation 354–355, 355, 356–358; negotiation process and 146–149; SawyerGuetzkow Model of Negotiation 19–21, 21, 37, 344; turning points (TPs) and 144, 175, 184, 185, 208–209 Franco, F. 23, 112, 144, 160, 207 Frederickson, N. 385 Fry, F. 47 game theory 32, 77, 125 Gellner, E. 38 General Agreement on Tariffs and Trade (GATT) 209, 347, 362 George, A. L. 181 George Mason University (GMU) 14, 48 Gladwell, M. 206 Glenn, E. S. 125 Goffman, E. 12, 385 Goldman, B. 357 Goldstone, J. 47 Gorbachev, M. 3, 24, 28, 179, 196, 208 governance: distributive justice and 317; durable peace and 41, 318, 330; institution-building and 330; justice and 9; multilevel interactions and 300–301 Green, J. 2 , 47 Gregor, A. J. 225 Grommé, F. 297 group attachments: conf lict resolution and 35; national identity and 36–37, 285; nationalism and 255–257;

406

Index

regional politics and 36–37; social sciences and 36 group conf licts 121–123; see also conf lict resolution group identity: individual identity and 299; mimicry and 287–288; nationalism and 261; political versus experimental 304n4; regional politics and 36–37, 222; self-worth and 262; SIT and 221–222; social network analysis (SNA) and 291; stereotypes and 225, 262; xenocentrism and 276 group loyalty: cohesion and 270; collective action and 267, 269–272; collective behavior and 254, 267–277, 292; collective images and 262; conditionality of 297; conf lict resolution and 59; decision-making and 271; ethnocentric bias and 269, 273, 278; group norms and 270; groupthink and 270–271; hostility toward other groups and 254–255; ingroupoutgroup 259, 272–275, 278–279; intergroup conf lict and 254–258; larger entities and 257; modifying 271–272; nationalism and 268–269, 277–278; negotiation and 17; political actions and 298; reference group theory and 275–277; renegades and heretics 273; representative role and 267–268; scaling hypothesis and 272, 277; social psychology of 254–258, 278–279; stereotypes and 262, 278; transitory 291 group norms 38, 257, 267, 270 group representation 10, 16, 59, 64, 390 groups: categorization and 261; competition and 261; concept of 290; constituent inf luence in 36; cooperative systems and 257; dynamics of 289–290; emotional attachment and 35–36; face-to-face interactions 37; human needs and 255, 278; ingroupoutgroup orientations 259; loyalties and 35–36; militarily and politically strong enemies 239–240; minimal group (MGP) studies and 290–291; perceptions of other groups 290; social identity and 360; social network analysis (SNA) and 291 groupthink 270–271, 294 Guetzkow, H. 13, 19–20, 37, 116n4, 139n8, 226, 228, 248n1 Guilford, J. P. 230, 232 Gulliver, P. H. 149

Gumplowicz, L. 225 Gurses, M. 321 Haas, E. B. 38 Hall, W. E. 210 halo error 223, 230, 232, 234, 248n5, 248n6, 249n10 Hamblin, R. L. 241 Hammond, K. R. 60, 71, 125, 356 Harinck, F. 10, 27, 58 Harmon, D. 349, 356 Harris, R. 144, 199n2, 206 Harrison, H. A. 71 Hartshorne, H. 385 Harvard Program on Negotiation 3 Haskel, B. G. 146 Hauenstein, N. M. A. 364 Helsinki process 26 Henderson, W. D. 270 Hermann, C. F. 226 Hermann, M. G. 226 Higgins, E. T. 208 Hobbler, G. D. 359 Hoffmann, S. 266 Högbladh, S. 325 Hogg, M. 221 holistic dynamical systems analysis 295 Hollander-Blumoff, R. 316, 318–320, 336, 350–351, 359 Holst, J. J. 106, 117n11 Holsti, O. R. 266 Holt, R. R. 265–267 Holtz, B. C. 361 Homans, G. C. 12, 230, 240, 290 Hong, S. G. 240 Hong Kong 183 Hopmann, P. T. 17, 26–27, 47, 77, 82, 84, 96, 153, 169 horizontal modeling 319, 331 Hove, M. 288 human agency 212, 391 Huo, Y. J. 319 Husbands, J. 28 Hussein, S. 265 Huth, P. 313 IAM–TWA negotiations 198 identity: coerced 293; competing 296; conditionality of 297; context and 4; cooperative responding and 288, 304n1; durability of experimental 291, 304n4; durable 37–38, 293–294; ICR interventions and 296–297; levels of analysis and 4; overlapping 296; procedural justice and 360; shared

Index

360–361; spread and 293; theory development and 3 – 4; values and 389; see also group identity; national identity ideologies: conf lict of interest and 124–137, 138n3; cross-cutting interests and 132–133, 139n6; group conf licts and 121–123; heightened conf lict and 27; intensification of commitment to 128, 385; interests and 5, 27, 31, 122–123; nationalism and 38; negotiation and 27; political decision-making and 136; within-team differences 131; values as 2 Iklé, F. 21, 25–27, 31, 33n2, 84, 139n3 images: collective 262; content vs. evaluative emphases of 262–264; diversity in descriptions of 262–264; embedded in schemas 265–267; of the enemy 265; theories of war and 265–267 indirect effects hypotheses 320–322, 334–335 inference training 46 inf luences: facilitating effects and 152; group boundary framing and 375; interferences on negotiation 152; negotiation process and 151–153, 159–161; responsiveness and 95 ingroup bias: competition and 261–262, 278; conf lict and 259–260; ethnocentrism and 35, 230–231; functions served by 388; group loyalty and 260–262, 278–279; hostility and 261–262; interventions for 388–389; loyalty–f lexibility relationship 290–291; perceptual accentuation and 261; positive regard and 261, 389; regional politics and 4; representation and 290–291; self-categorization theory and 261; simulations and 230–231; social identity theory (SIT) and 260–261; stereotypes and 35 Insko, C. A. 261, 291 Institute for Conf lict Analysis and Resolution (GMU) 14, 48 Institute for Juvenile Research (IJR) 14, 44 Institute of World Affairs 33 integrative bargaining 13, 17 interactive conf lict resolution (ICR) 296–298, 302 interests: accountability pressures and 77, 93–94; cross-cutting 130, 132–134, 135, 136, 138, 139n6; ideologies and 5, 27, 31, 122–123; opposition of 16, 388;

407

overlapping 139n6; patriotism and 266; values and 2 , 10, 17, 31–32, 37, 385, 389–390; see also conf lict of interest; self-interest intergroup relations: conf lict and 257–258; ethnocentrism and 34–35, 226, 235, 243; group loyalty and 255; militarily and politically strong/weak enemies 239–240, 244; sentimental attachments and 37 Intermediate-Range Nuclear Forces (INF): negotiating crises and 208; negotiation progress in 178; turning points in 3, 28, 175, 177, 183, 196, 208 international decision-making 102, 116n4 international exchange programs 36, 303 International Institute for Applied Systems Analysis (IIASA) 98 international law 24, 184 international negotiation: bargaining process and 171n3; benchmarks and 148; bidirectional responsiveness and 78; boundary roles and 76, 78; career development and 9 –10, 20; comparative reciprocity and 26; conf lict resolution and 30–32; descriptive studies and 171n2; diplomacy and 28–29; extension agreements and 170; government interests and 26–27; inf luences in 151–153; models of 21, 22; monitoring trends in 26; negotiator-as-bargainer model 79– 88; negotiator-as-representative model 88–94; overcoming constraints on f lexibility 32; patterned exchanges in 177–178; position formation and 61; process of 147–153; relative power and 30–31; responsiveness and 79– 83, 87; rhythms and patterns in 146, 169; side effects and 146; simulation approach 171n2; stages of 147–149; tactics and 32; taxonomies and schematic representations 171n2; turning points in 32, 177; types of cases 182, 183; types of issues in 32; values and 31 international relations: bargaining f lexibility and 17; critical junctures in 215n5; group loyalty and 268; negotiating process and 20; processcontext relationships and 304; simulation approach and 226, 236, 243–245; turning points and 212 Inter-Nation Simulation (INS): analyses of variance 237, 241–242; analysis and 231–233; as dynamic equilibrium

408

Index

model 137; ethnocentrism and 34–35, 223, 230–234, 234, 235–236, 238–248, 248n3; general social processes and 244–246; halo error and 232, 234, 248n5, 248n6; international negotiation and 20, 226; roles of individuals in 227–229; universal bias hypothesis and 227, 233–236, 243; WBSI simulation 228–229 interpersonal synchrony 288–289 intra-regional relationships 36–37 Irmer, C. 361 Israel 27, 36, 269, 296, 302 Jackson, H. 24 Janis, I. 270 January 6, 2021 insurrection 122 Japan 111–112, 113, 183 Jaspal, R. 286 Jennings, H. H. 230 Jensen, L. 169 Jervis, R. 179 Jochemczyk, L. 210 Johnston, K. 28 Joshi, M. 327 Joy, V. L. 365 justice: anticipatory 344, 357–358; compensatory 363, 390; compound 351; cultures of peace and 5, 9; defining 390; durable peace and 5, 28, 41; effectiveness and 42; empirical approaches to 349, 352; equality and 347; fairness and 347; field research and 353, 375; governance and 9; implementation of peace agreements and 10; laboratory studies and 375; normative approaches to 349; peace agreements and 311, 313, 315–317, 321, 336; preferences in 317; restorative 390; retributive 370; social conduct and 318; transitional 339n10; transparency and 347; trust and 344; truth and reconciliation commissions 312; see also distributive justice (DJ); justice principles; procedural justice (PJ) Justice and Negotiation framework: framing in 354–356; negotiation outcomes 364–367; negotiation process 359–364, 371; postnegotiation 368–371; prenegotiation 354–358; stages of negotiation 354, 355 justice principles: collective action and 354; distributive justice and 41, 354; durable peace and 335, 340n16;

hypotheses for future studies on negotiation and 373–374; negotiation and 4 –5, 10, 40, 311, 344–354, 359–364, 371–372, 376, 389–390; outcomes and 364–367; peace agreements and 286; postnegotiation and 368–371; prenegotiation and 354–358, 375; procedural justice and 286, 354; shared moral community and 354–355; stability of agreements and 317, 335; trust violation and repair 361 Justino, P. 316, 338 Kabanoff, B. 320, 357, 362, 366 Kapstein, E. B. 362 Kavanagh, L. 287 Kayatani, M. 230, 238 Kedourie, E. 38 Kelley, H. 139n3, 241, 292, 385–386 Kelley, H. H. 12, 47, 60 Kelman, H. C. 256, 348 Kennan, G. 45 Kent, D. P. 276 Kim, P. 349, 356 King, T. D. 169 Kissinger, H. 24, 27, 32, 156–157, 160 Klineberg, O. 262 Kramer, R. M. 260 Kressel, K. 101 Lall, A. 24–27, 33n2 Lambert, W. E. 262 Langley, J. 199n2 Larsen, E. 46 Latin America 276 leadership: alliance negotiations and 23; collective dynamics and 292; group loyalty and 271, 274; negotiation progress and 208; stereotypes and 265; succession crises 178, 184; turning points and 178 Lebanon 183, 191 Lebow, R. N. 269 Leites, N. 139n3 Lemyre, L. 261 levels of analysis: circular causation and 391; civil society and 295–296; civil wars and 338; collective 285; false justice and 361–362; identity and 4; interactions in 295–296; intergroup 285; interpersonal 285; national identity and 286–295, 298–299; negotiation and 4; simultaneous interactions and 303; structures and processes in 4

Index

Levendusky, M. 297 LeVine, R. A. 34, 223, 225–227, 233, 248n1 Levine, T. R. 206 Lewin, K. 5, 12, 384–385 Lewis, M. 61 Lewis, S. A. 55 Liebovitch, L. 295 Lilja, J. 360 Lind, E. A. 5, 40, 316, 319–320, 336, 353, 355, 359, 361, 362–363 Linder, D. 236, 238, 247 loyalty–f lexibility relationship 290–291 Ludwig, L. D. 35, 248n1 Lynn, J. A. 270 MacDougall, J. A. 130, 139n4 Mackie, D. M. 260 Macquarie University 14, 49 macro-levels: collective dynamics and 297–298; ICR interventions and 296–297; micro-levels and 303; national identity and 286–287, 289, 302; supra-national institutions and 299 Macy, M. 294, 301 Madoff, B. 206 Malešević, S. 269, 286, 298 Marcos, I. 49n4 Marshall, A. 49n2 Marwell, G. 125 Marx, K. 128 Mason, T. D. 321 matching strategies 288 Mathtech 14, 45– 46 McClelland, C. A. 143, 153, 163, 171n5, 205 McClintock, C. G. 71 McCloskey, R. J. 23, 45, 155 McDonald, J. 28–29, 45 McKersie, R. B. 13, 16–17, 56, 76, 106, 385–386 Mead, G. H. 12 media coverage: bargaining decisions and 99, 103, 106, 108–109, 115, 322; f lexibility and 17, 114–115; nationalism and 292; peace agreements and 322, 337; precipitants and 184; stereotypes and 264 mediation: advice and 18; analysis and 18; bargaining and 18; diagnosis and 18; equality and 368–369; novelty and 19; peace agreements and 323; procedural justice and 369; robotic 18–19, 19n2,

409

56; stability of agreements and 368; trust and 361; see also e-mediation Melander, E. 325 Merton, R. K. 275 meso-levels: collective dynamics and 298; multilevel governance and 300–301; national identity and 286, 289; negotiation process and 302–303; pathways theory and 303; social formation of networks and 300; social media narratives and 298 Messick, D. M. 260 meta-analysis 109–110, 117n14 micro-levels: collective dynamics and 298; ICR interventions and 296–297; interpersonal synchrony and 289; macro-levels and 303; matching strategies and 289; mimicry and 289; national identity and 286–287, 289, 302; party loyalty and 297; social innovation and 303 Middle East 10, 27–28, 32, 36 Mikula, G. 354 Milburn, T. 44, 266 Miller, D. T. 348 mimicry: emotional attachment and 287; family attachments and 287–289; formation of 287–288; interpersonal relations and 287–288; matching strategies and 288; national identity and 287; synchrony and 288–289 minimal group (MGP) studies 290–291 Mischel, W. 385 Mitchell, C. 2 , 240 Mittone, L. 368 monitoring activities: boundary roles and 76, 78, 95–96; movement and 96; negotiation process and 17, 49n3, 78, 85– 86, 150–151, 152, 213; preferences and 96; turning points and 205–207, 213 Montreal Protocol, 1987 183, 200n4 Moskos, C. C. 270 Moussaid, M. 293 Mouton, J. S. 17, 35, 60, 73, 73n1, 240, 267, 304n3 Mozambican National Resistance (RENAMO) 337 Mozambique 337, 347 Mozambique Liberation Front (FRELIMO) 337 Mueller, J. E. 266, 290 multilateral negotiation: diplomacy and 33n2; impact of 24–25; institutional

410

Index

contexts for 214; international law and 24; patterns in trade 3; procedural justice and 42– 43; procedures and 24 Mutual Balanced Force Reductions (MBFR) 22–24, 26, 45, 148 National Academies (NAS) 48, 49n7 National Democratic Front (NDF) 27, 31, 113, 123 national identity: attitude changes and 297, 305n8; authoritarian–democratic distinctions in 38; citizen perceptions and 36–37; coercion and 293; collective dynamics and 4, 292; competing identities and 296, 298; durable 37–38, 293, 305n5; ethnocentrism and 10; EU and 299–300; group attachments and 36–37, 285; heightened in foreign countries 297, 302, 305n9; interaction dynamics and 297, 305n9; intergroup 4; interpersonal dynamics and 4, 287–289; levels of analysis and 286–295, 298–299, 302; loyalties and 36; mechanisms for 304; media coverage and 292; micro/macro level interactions 286–287, 289; mimicry and 287; negative 276–277; networks and 300; pathways to theory of 286–287, 299, 301–303; political polarization and 294; political rhetoric and 298; simultaneous interactions and 287; SIT and 221–222; social communication and 300; social psychology of 254–255; spread and 293, 305n5; stereotypes and 252, 263–265; supra-national institutions and 299–300; theory development and 3 – 4, 37; universal bias hypothesis and 252 nationalism: aggressiveness and 258; conf lict resolution and 37; as doctrine 38; durable identities and 37–38; ethnic cleansing and 269; ethnocentrism and 233, 269; EU development and 299–300; group attachments and 255–257; group dynamics and 296; group identity and 261; group loyalty and 268–269, 272, 277–279; integrative theory of 286; interdisciplinary analysis of 286, 298, 304; militaristic approach and 258–259; patriotism versus 4, 258–260, 279; reference group theory and 275–277; responsiveness to myths 37; self-esteem and 261; self-images and 269; sentiment and 38, 256; simulation

approach and 272; social psychology of 254, 256, 277, 279, 291 National Mediation Board 198 National Research Council (NRC) 14, 28, 37–38, 47– 48 NATO 23, 26, 30, 45, 183 negative ethnocentrism 276 negotiating crises: INF talks and 208; issue framing and reframing in 208; turning points and 150, 159–160, 161–162, 166–168, 207–208, 214; US-Spain base rights talks 150, 160, 161, 166–168, 206–207 negotiating f lexibility: characteristics of 55–56; diagnosing 110–115; negotiating stages and 17; prenegotiation and 57–58; situational levers and 100–101; timing variable and 104, 115; values and 2 negotiation: alliance 22–23; backchannel communications and 24; boundary roles and 2 , 77; continual threefold choice 26; diplomatic 28–30; holistic framework for 19; hypotheses for future studies on justice and 373–374; international politics and 25; intervening variables and 58; justice and 4 –5, 10, 40, 311, 344–354, 371–372, 376, 389–390; levels of analysis and 4; monitoring function in 17, 49n3; multilateral 24; objectives and 25–26; pre-negotiation experience 2; private settings and 106, 117n11; problem-solving processes and 350–351; reciprocity in 26; side effects and 26; simulation approach 101–102, 346; situational levers and 2 negotiation outcomes: distributive justice and 364–367; equality and 366–367, 375; equity and 366–367, 375; inf luence of process on 364; justice principles and 364–367; perceived fairness and 365, 367; problem-solving processes and 364–365; relationships between procedural and distributive justice 364–365 negotiation process: bargaining in 13, 148–149, 152; benchmarks and 148; collaboration and 346; comparative analysis of 182, 184–191, 195–198, 199n2; conf licts and 197; consequences and 178, 181, 184, 185, 208–209; context and 317–318, 323, 337; distributive justice and 346–350, 375;

Index

environmental negotiations and 179; equality and 347; fairness and 206, 347; false justice 361–362; framework for 146–149; impasses and 214; inf luences in 151–153; large-sample analysis 177, 199n2; media coverage and 322; monitoring activities in 151–151, 152, 214; motivational orientations 359–360; peace agreements and 337; persuasive debate in 152; power differences and 362–363; precipitants and 178, 181, 184, 186, 209, 385; procedural justice and 346, 350–352, 359–360, 363–364, 375; procedures for advancing 149, 171n4; process departures and 178, 181, 184, 186, 196, 208–209; region of validity procedures 171n4; security issues and 179–180; shared identities and 360–361; smallsample analysis of 199n2; social conduct and 318–319; social factors and 347–348; stages of 148–149, 205, 346, 354, 355, 389; strategizing activities in 150–151; threshold adjustment and 206; trust and 360–361; turning points in 149–150, 177–180, 184, 186–191, 195–197, 199n3; verbal behavior in 153–154, 154, 155 Negotiator Assistant (NA) 18 Neu, J. 370 Newcomb, T. M. 385 Newhouse, J. 27 Nitze, P. 24 Nixon, R. 265 non-standard agreements (NSAs) 19 nonverbal communication 46, 49n3, 385 North American Free Trade (NAFTA) negotiations 178–179, 183, 196, 347 Northwestern University 13, 34 Norwegian Foreign Ministry 40 nuclear conf lict 28 Oakes, P. 261 Obama, B. 292 objectives: changes in strategy and 82– 83; conf licting expectations and 76; conf lict over 157, 159–160; negotiation and 21, 25–26, 55; of outcomes 366 Olekalns, M. 3, 208, 355, 360 O’Mahony, P. 297 O’Reilly, C. W. 330, 337 organizational structures 209–210, 214, 357 Orne, M. T. 245

411

Osgood, C. E. 17, 212, 266, 288 outcome justice see distributive justice (DJ) outgroups: ethnophaulisms and 225, 248n2; group loyalty and 272–273, 278–279; negative/positive feelings 274; stereotypes and 35, 225; see also ingroup bias out-of-phase pattern of activity 161 overassimilation 236 pair-comparison procedure 103, 106, 386 Parlamis, J. 58, 289, 291 Palestine 36, 274, 296, 302 Palmore, E. B. 248n2 Panama Canal Treaties 29, 183 Paris, R. 330 Partial Nuclear Test Ban Treaty (1963) 82 patriotism: cooperative approach and 258; emotional attachment and 258, 279; ethnocentrism and 259; nationalism versus 4, 258–260, 279; regional politics and 301 patterned exchanges 177–178 Peace Accords Matrix (PAM) 325, 327–328, 330, 339n5, 339n7 peace agreements: case study in 324–338; conf lict environment (CE) and 323, 339n2; distributive justice and 40– 41, 312, 317, 328–329, 349; durable peace and 41, 302, 315; economic growth and 340n18; implementation success and 323; justice and 10, 295, 311, 313, 315–317, 321, 336; long-term peace and 317; media coverage and 322, 337; mediation and 323; microlevel interactions in 302; negotiation length and 324; negotiation process and 337; power-sharing provisions and 318; procedural justice and 40– 41, 43, 286–287, 312–313, 317–318, 322, 328; reconciliation and 316; relationship building and 316, 318–319, 337; resumption of violence and 315; social conduct and 318; stability of agreements and 317–318, 329; see also durable peace (DP) Peace Agreements data set (UCDP) 315 peacekeeping operations (PKOs) 318, 323, 340n21 Pell, C. 47– 48 Peoples’ Revolution 10, 30, 49n4 perceptions: accentuation of 261; fairness and 352, 354, 356, 361, 365; group

412

Index

loyalty and 268, 271, 278, 297; national identity and 269; of other nations 264–266; procedural justice and 359–360, 368, 370; regional politics and 36–37; simulation approach and 223–224, 228; structures and 36–37; trust 322, 336, 352, 361 Peterson, R. S. 359 Pettigrew, T. F. 303 Pew Case Studies in International Affairs 25, 175, 181, 184 Pezzulo, L. 45 Philippines: Aquino government and 27, 31, 113, 123; base rights negotiations and 148; debt (1983–86) 183; Marcos regime in 46– 47, 49n4; National Democratic Front (NDF) in 27, 31, 113, 123; Peoples’ Revolution in 10, 30, 49n4 Piaget, J. 256 Ploner, M. 368 political decision-making: characteristics of 136; collective dynamics and 292; competition for power 136; dyad vs system 136; ideologies and 27, 126–127, 136, 139n3; loyalty–f lexibility relationships in 291; operating model for 135–136, 139n8; paths to outcomes 192–193; propositional format and 135; simulation approach and 129, 135–138; utility theory model of 139n3 political negotiations: cases in 183; external events and 188; paths to outcomes 193, 194–195; precipitants and 188; types of 184 political parties: collective dynamics and 292; conf lict of interest and 126; conservative bias and 206; crosscutting interests and 133; ideologies and 123; loyalty–f lexibility relationship and 291; nationalist rhetoric and 298; negative partisanship and 389; negotiating f lexibility and 289; party loyalty and 294, 297; voter identity and 297 political polarization 294, 389 position formation: bilateral 59– 60, 62, 65–73; international negotiation and 61; prenegotiation and 69–71, 73; team solidarity and 69; unilateral 59– 63, 66–70, 71–72 postnegotiation: distributive outcomes and 368–369; implementation of agreements and 368–371, 376; justice

principles and 368–371; procedural justice and 368, 370–371; role of spoilers in 369–370, 376 power differences: compensation and 390; distributive justice and 357; equitable distributions and 357, 362–363; group loyalty and 277; policymaking and 271, 301; procedural justice and 43, 362–363, 364 precipitants: consequences and 189–190, 191, 192–195; escalation/de-escalation 198–199; external factors 184, 188, 189, 189–190; inside/outside 189; negotiation process and 178, 181, 184, 208–209, 385; procedural factors 184, 189, 198–199; process departures and 190, 198–199; process tracing and 188; security negotiation cases 188, 209; substantive factors 184, 189; thirdparty intervention and 190–191, 210; types of negotiation 189 prenegotiation: anticipatory justice and 357–358; bilateral focus and 59– 60, 62, 66, 66, 67– 69, 71–72; collective bargaining and 59, 61– 63; conf lict resolution and 59– 66, 66, 67– 69, 71–73; distributive principles in 357; f lexibility and 2; framing in 354–356, 372, 375; intervening variables and 58; NAFTA talks and 178–179, 196; perceived commitment and 68– 69; position formation and 69–71, 73; tasks and setting 356; team positionplanning and 59; unilateral focus and 59– 63, 66, 66, 67– 69, 72, 73n2; unresolved issues 68, 68 Prespa agreement (2018) 296, 302 Press, F. 49n7 Price, D. K. 128–129 principle of dissent 130 Prisoner’s Dilemma game 230, 238 private settings 106, 117n11 problem-solving conf lict resolution 116n2, 317 procedural justice (PJ): distributive justice and 352–354; durable peace and 287, 319–322, 332, 334, 336, 338, 350; effectiveness and 42– 43; equality and 349, 358; fairness and 320, 328, 336, 350, 351–352, 359, 365; fair representation and 350, 352; false justice and 361–362; group values and 363–364; identity and 360; improved relations and 317; integrative outcomes

Index

and 317; justice principles and 354; mediation and 368; negotiation outcomes and 364–365; negotiation process and 317, 346, 350–352, 359–360, 363–364, 375; peace agreements and 40– 41, 43, 286–287, 312–313, 317–318, 328; peacekeeping operations (PKOs) and 340n21; perceived identities and 355; perceived relationships and 320; postnegotiation and 368, 370–371; power differences and 43, 362–363, 364; principles of 350; problem-solving processes and 350–351; self-interest and 362–364; shared identities and 360–361; social conduct and 5, 317, 338; social harmony and 356; social psychology of 40; stability of agreements and 368; transparency and 320, 328, 336, 350, 351; voluntary decisions and 350, 351 procedures: consensus-building 179; environmental negotiations and 210; multilateral negotiation and 24; negotiation process and 171n4; perceptions of fair 352, 356, 365; see also procedural justice (PJ) process departures: consequences and 189–190, 191; escalation/de-escalation 196; negotiation process and 178, 181, 184, 208–209; precipitants and 190; security negotiation cases 189; turning points and 184 process tracing: causation and 385; precipitants and 188; turning points and 175, 180–181, 187–188 proportionality 348–349, 351, 356–357, 366; see also equity propositional format 127–135 prosopography 9, 11n1, 50, 413 Pruitt, D. G. 55, 149, 317, 368 Putin, V. 37–38, 303 Putnam, R. D. 77 Quinn, J. M. 321, 327 Radloff, L. S. 236, 273 Ramberg, B. 146, 288 Rapoport, A. 60, 71, 125–126, 136, 171n4 Raser, J. R. 226, 228–229 Reagan, R. 3, 24, 196, 271 realpolitik theory of war 265–267 reconciliation: distributive justice and 317; durable peace and 319, 330–331, 332, 333, 334, 336–338; fairness and 320;

413

peace agreements and 316; transparency and 320, 337 reference group theory 275–277 Regan, P. M. 327 regime stability 49n4 regional politics 4, 36–37 representation: accountability and 93–94; effects of 35; group 10, 16, 59, 64, 390; group loyalty and 267–268; negotiation and 79, 88, 92; negotiator agency and 92–94; prenegotiation and 77; procedural justice and 42; selfrepresentation 64 resistance to compromise 63, 94, 124 responsiveness: boundary role conf licts and 95; concession-making and 76, 79– 83, 87; indirect 84; interdependence and 155; international negotiation and 82– 83, 88; models of 144–145; mutual 81, 83; patterns of 155 Richardson arms-race model 80, 82 Rieken, H. W. 230 Risen, J. 288 Robinson, C. 238 robotic mediation 18–19, 19n2, 56 Rokeach, M. 55, 236 Romans, G. C. 241 Rome General Peace Accords 347 Ron, J. 330 Rosenblatt, P. C. 225 Ross, M. H. 133, 257 Rothman, G. 236 Rouhana, N. N. 296–297 Round Table talks (1989 Warsaw) 210 Rovira, J. 23, 155 Rozelle, R. M. 46 Rubin, J. Z. 27, 107 Rusk, D. 60 Russa-Baltic States, 1990–94 183 Russia 25, 222, 252 Sage International Studies Series 20 Sandusky, J. 206 Sawyer, J. 19–20, 37 Sawyer-Guetzkow Model of Negotiation 19–21, 21, 37, 344 scaling hypothesis 272, 277 Schar School of Policy and Government (GMU) 14 Scheel, S. 297 Scheepers, D. 388 Schelling. T. 5, 13, 17, 47, 127–128, 385–386 schemas 265–267

414

Index

Scowcroft, B. 45 Seabed arms control negotiation 148 Sears, D. O. 259 security institutions: distributive justice and 333; durable peace and 41, 317, 327, 330, 349; restorative justice and 390 security negotiation: antagonistic 179; cases in 183; concession-making and 195; external events and 180, 188; framing-reframing issues 209; joint fact-finding tasks 209; paths to outcomes 192, 193, 192–194; precipitants and 188, 209; process departures and 189; third-party intervention and 191; turning points in 195; types of 184; verification procedures and 209 Seibt, J. 18 self-categorization 221, 261 self-esteem 221, 261 self-interest 266, 355, 356, 362–364, 389 sentiment 38, 230, 240, 256 settlement-oriented conf lict resolution 116n2 shared identities 360–361 Sherif, C. W. 5, 239, 241 Sherif, M. 5, 12, 73n1, 212, 230, 239–241, 290, 385 Shoda, V. 385 shuttle diplomacy 27–30, 32 Siegel, S. 16 Silverstein, B. 265–267 Simmel, G. 5, 12, 121, 127–128, 385 Simon, H. A. 47, 213 Simulated International Processes project 248n1 simulation approach: balancing perspective and 123; bias and 223–224; dynamic equilibrium models 137; experimental 101–102, 116n5; fairness and 346; f lexibility and 103–104, 108–109, 115; halo error and 223; international decision making 102, 116n4; mediator effects in 105; negotiation and 101–102; paircomparison procedure 103; political decision-making and 129, 135–138; role-player theories and 102, 108–109, 115; scaling hypothesis and 272; simplification of real-world setting and 101; situational levers and 103–104; universality and 222; validity studies and 226; variables and 246–247; see also Inter-Nation Simulation (INS)

Singer, J. E. 236, 247, 273, 304n3 situational levers 2 , 98–100, 103–104 skill development 6, 33 Smith, P. M. 261 Smith, T. C. 82, 84, 96 Snow, C. P. 126, 136, 139n5 Snyder, G. H. 101 social conduct: distributive justice and 317; fairness and 320, 336; justice and 318; negotiation process and 318–319; peace agreements and 316, 318; procedural justice and 5, 317, 338; transparency and 320, 337; trust and 5, 336; type of agreement and 337 social identity theory (SIT) 221, 260–261, 360–361 social media 291, 298 social network analysis (SNA) 291–292 social psychology: beliefs and attitudes in 2; collective behavior and 4, 292; group loyalty and 254–258, 278–279; groups and 37, 290; international conf lict and 20; nationalism and 254–256, 277, 279, 291, 304; negotiation process and 213; procedural justice and 40; sentimental attachments and 38; situational effects and 77 social sciences: behavioral approaches in 5; central concepts in 32; competing hypotheses and 34–35, 39n1; empirical approaches in 5; ethnocentrism and 34–35; group attachments and 35–36; group identities and 36–37, 289–290; methodologies in 27; representation effects and 35; sense-making in 12; study of human behavior and 9, 12–13 social systems 128–129, 133–134, 136 Society for the Psychological Study of Social Issues 27, 39n2 Solomon, D. 369 Soviet Union: arms control and 24; fall of 28; good vs. evil folk theory and 266; INF talks and 28, 112–113, 208; multilateral negotiation and 25; relations with the U.S. 269; START negotiations and 23–24, 45– 46; stereotypes and 264, 266, 271 Spain: Agreement of Friendship and Cooperation (AFC) 155; NATO membership and 23, 30; US base rights talks 22–23, 29–30, 112, 114, 143–144, 155–157, 158, 159–168, 177–178 Spector, B. I. 82, 368 spoilers 369–370, 376

Index

spread 293, 305n5 stability of the agreement: distributive justice and 320; durable peace and 322, 335–336, 338; mediation and 368; peace agreements and 317–318, 329; procedural justice and 368 Stagner, R. 128 Starr, H. 266 status-giving 241 Stedman, S. J. 40, 323, 339n2, 369 stereotypes: cognitive aspects of 263–264; collective behavior and 267; competing hypotheses and 35; of disliked nations 263–264; enemy nations and 264–266; evaluation adjectives and 35; group identity and 262; group loyalty and 278; intergroup 231; motivational aspects of 263–264; national identity and 252; outgroups and 35, 225; theories of war and 265–267 Stern, P.C. 37, 47 Strategic Arms Limitation Talks (SALT) 24, 148, 183 Strategic Arms Reductions Talks (START) 23–24, 28, 45– 46 strategizing activities 150 Strauss, A. 12, 179 structures: f lexibility and 17; negotiations and 4, 9; organizational 209–210, 214, 357; perceptions and 36–37; regional politics and 36–37 Stuart, D. 266 sub-national identity 297–298 Sumner, W. G. 223, 225, 235, 240 supra-national identity 297, 299–301, 305n9 Swaab, R. 355 Swartz, M. J. 233, 240, 276 Swedish Research Council 41, 313 synchrony: concession-making and 77; interpersonal relations and 288–289; mimicry and 288–289; synchronicity and 212–214; negotiation process and 214, 288; patterns of responsiveness and 288–289; turning points and 207 systems (scientific) theory of war 265–267 Sztompka, P. 297 Taiwan Straits Confrontations 143, 171n5, 205 Tajfel, H. 38, 221–222, 260 Taylor, P. J. 288 team solidarity 69 Terhune, K. W. 256

415

theories of war 265–267 Thibaut, J. W. 12, 40, 241, 292, 316, 357, 385–386 Thompson, L. L. 348 Thoms, O. N. T. 330 threshold adjustment: base rights negotiations and 76, 144; monitoring function and 213; offers and counteroffers in 213; responsiveness and 169; turning points and 206, 213 Thunberg, G. 13 Tilly, C. 290 time-series analysis 181, 205 tit-for-tat strategy 144 Todd, J. 295 Tomlin, B. W. 178, 181, 186 Tracy, B. H. 94, 146 trade negotiation: cases in 183; distributive justice and 350; external events and 188; paths to outcomes 194, 194, 195; patterns in 3; precipitants and 188, 200n9; role of justice in 347; turning points in 195; types of 184 transparency: distributive justice and 316; durable peace and 332–333; fairness and 5, 313; justice and 347; negotiation process and 347; procedural justice and 320, 328, 336, 350, 351; reconciliation and 320, 337; social conduct and 320, 337; trust and 344 Treaty of Friendship and Cooperation 23, 156 Treaty on the Non-Proliferation of Nuclear Weapons 25 trend model 144 Tropp, L. R. 303 Trump, D. 36, 206, 292, 386 trust: equal outcomes and 5; equity and 356; fairness and 344; justice and 344, 360–361; mediation and 361; negotiation process and 360–361; perceptions of 322, 336, 352, 360–361; social conduct and 5, 336; transparency and 344; violation and repair 361 truth and reconciliation commissions 312 truth-default theory 206 Turner, J. C. 221, 261 turning points (TPs): abrupt changes and 208; bargaining and 16; case chronology and 180–181; comparative analysis of 185–191, 195–197; concept of 171n5, 204, 215n1; concessionmaking and 150; crises and 150, 160, 161, 166–168, 207–208, 214; defining

416

Index

177, 205; environmental negotiations and 177–180, 195; external events and 178–180; f lexibility and 2 –3; historical overview of 204–205; impasses and 150, 160; inf luence of conference structures on 210; INF talks and 3, 28, 175, 177, 196; internal processes and 179; international negotiation and 32, 177; levels of analysis and 211–213; monitoring and 205–207, 214–215; NAFTA talks and 178–179; negotiation process and 28, 149–150, 178, 184, 186–191, 195–197, 199n3; organizational structures and 209–210, 214; process tracing and 180–181, 187–188; security negotiation 195; social climate and 208; system changes and 212; theory development and 211–215; third-party intervention and 191, 210; three-part framework of 144, 175, 182, 185–186, 208–211, 385; threshold adjustment and 206, 213; trade negotiation 195; US-Spain base rights talks 143–144, 159–162, 166–168, 177–178, 196, 204–205 Tversky, A. 47 Tyler, T. R. 5, 40, 316, 317, 319, 336, 350–351, 353, 356, 359–361, 362–363 Ukraine 37–38, 222, 252, 386, 388 unilateral initiatives 17 unilateral position-formation 61, 66, 66, 67– 69, 71–72, 73n2 unitary-rational actor models 88 United Nations 33, 33n2, 84, 197, 357 United States: Agreement of Friendship and Cooperation (AFC) 155; arms control and 3, 23–24, 28; foreign policy towards China 265, 269; INF talks and 28, 112–113, 208; Japan and 111–112, 113; multilateral negotiation and 25; Soviet relations and 266, 269, 271; see also U.S.–Spain base rights talks United States Institute of Peace (USIP) 18, 33 universal bias hypothesis: enemy nations and 223, 232, 234–236; ethnocentrism and 223, 233–236, 243; intergroup competition and 235; Inter-Nation Simulation (INS) and 233–236, 243; national identity and 252 University of Melbourne 14 University of Queensland (UQ) 14, 40, 48– 49

University of Southern Queensland 14, 49 UN Peacemaker data set 327–328 Uppsala Conf lict Data Program (UCDP) 315, 325, 327–328, 330, 339n6, 339n9 Uppsala University 40 Uruguay Round (GATT) 209, 362 Ury, W. 386 U.S.–Japanese administrative agreement 111–112, 113 U.S.–Republic of China mutual security talks 111–112 U.S.–Spain base rights talks: background to 155; content analysis of statements 162–168; diagnosis and outcomes in 112, 114; difference in objectives in 156; f lexibility and 114; impasses and 150, 160, 168; inf luences in 159–160; leadership succession crises and 178; NATO membership demands and 23, 30, 157, 159; negotiating behavior and 166–168; negotiating crises and 150, 160, 162, 166–168, 171n8, 207–208; negotiating rhetoric in 162–168; negotiation and 22, 29; negotiation stages 157, 159; negotiator activities 161–162, 178; out-of-phase pattern of activity 161–162; overview of negotiations 155–157; persuasive appeals in 161; processes and inf luences 158; threshold adjustment and 76, 144; time-series analysis and 143; trends in soft/hard behavior 164, 164, 165–166, 166, 167, 167, 169, 170; turning points and 143–144, 159–162, 166–168, 171n8, 177–178, 183, 196, 204–207 U.S.–Turkey trade agreement 111–112 Valadez, J. J. 116n4 value dissensus: academic-intellectual community and 126; bilateral 130–131; competition and 125; conf lict of interest and 127, 130–131; ideological conf lict and 125, 131, 134; intensity of conf lict and 137–138 values: as beliefs 2; conf lict and 2 , 17, 30–31; conf lict of interest and 125–127; differences in 389; f lexibility and 2; identity and 389; as ideologies 2; interests and 2 , 10, 17, 31–32, 37, 385, 389–390 Van Doorn, J. A. A. 138n3 Van Wijk, J. 303 Verba, S. 246 verbal behavior 153–154, 154, 155 vertical modeling 318–319, 331

Index

Vidmar, N. 35 VienNA system 18 Vietnam peace negotiations 147–148, 183 Wagner, L. 43, 286, 302 , 317, 319–320, 349, 350, 359, 360, 361 Walcott, C. 153 Walker, L. 40, 316 Wallensteen, P. 325 Walter, B. F. 324 Walton, R. 13 Walton, R. E. 16 –17, 56, 76, 106, 385–386 war criminals 370 war games 49n2 Wark, D. M. 236, 273 Warsaw Pact 26, 45, 386 Watson, J. 12 , 384 Webster, S. 291 Webster, S. W. 389 Welch, D. 390

417

Wells, R. 199n2 Wenzel, M. 354 Western Behavioral Sciences Institute (WBSI) 228 –229 Wilkenfeld, J. 18 Wilson, W. 230, 238 Winkielman, P. 287 Witt, L. A. 365 Wolf, F. M. 117n14 Wood, S. 4, 13, 15, 221, 285, 391 World Events Interaction Survey 153 World Trade Organization (WTO) 209, 347 xenocentrism 276 Zander, A. 270 Zartman, I. W. 84, 89, 94, 96, 97n1, 123, 147, 351, 366, 368 Zechmeister, K. 31, 126, 128 –129 Zimbabwe Independence 29