Progress in International Relations Theory The BCSIAStudies in International Security book series is edited at the Belfer Center for Science and International Affairs at Harvard University’s John F. Kennedy School of Government and published by The MIT Press. The series publishes books on contemporary issues in international security policy, as well as their conceptual and historical foundations. Topics of particular interest to the series include the spread of weapons of mass destruction, internal con- ºict, the international effects of democracy and democratization, and U.S. de- fense policy. Acomplete list of BCSIAStudies appears at the back of this volume. Progress in International Relations Theory Appraising the Field Colin Elman and Miriam Fendius Elman, editors BCSIA Studies in International Security MIT Press Cambridge, Massachusetts London, England ©2003bytheBelferCenterforScienceandInternationalAffairs JohnF.KennedySchoolofGovernment,HarvardUniversity Cambridge,Massachusetts02138 (617)495-1400 Allrightsreserved.Nopartofthisbookmaybereproduced,storedinaretrieval system,ortransmittedinanyformorbyanymeans—electronic,electrostatic,magnetic tape,mechanical,photocopying,recording,orotherwise—withoutpermissioninwriting fromTheMITPress,5CambridgeCenter,Cambridge,MA02142. ThisbookwastypesetinPalatinobyTeresaLawsonandwas printedandboundintheUnitedStatesofAmerica. LibraryofCongressCataloging-in-PublicationData Progressininternationalrelationstheory:appraisingthefield/ ColinElmanandMiriamFendiusElman,eds. p. c.m—(BCSIAstudiesininternationalsecurity) Includesbibliographicalreferencesandindex. ISBN0-262-05068-4(hc.:alk.paper)—ISBN0-262-55041-5(pbk.:alk.paper) 1.Internationalrelations—Methodology.I.Elman,Colin.II.Elman,MiriamFendius. III.Series. JZ1242.P762003 327.1’01—dc21 2002032170 PrintedintheUnitedStatesofAmerica 10 9 8 7 6 5 4 3 2 1 Wegreatlyappreciatepermissiontoincludepreviouspublicationsinthefollowingjournals: Chapters 1 and 2 in Colin Elman and Miriam Fendius Elman, “How Not to Be Lakatos Intolerant:AppraisingProgressinIRResearch,”InternationalStudiesQuarterly,Vol.46,No.2 (June2002),pp.231–262;Chapter4inJonathanM.DiCiccoandJackS.Levy,“PowerShifts and Problem Shifts: The Evolution of the Power Transition Program,” Journal of Conºict Resolution,Vol.43,No.6(December1999),pp.675–704;Chapter5inAndrewMoravcsik, “Taking Preferences Seriously: A Liberal Theory of International Politics,” International Organization, Vol. 51, No. 4 (Autumn 1997), pp. 513–553; Jeffrey Legro and Andrew Moravcsik,“IsAnybodyStillaRealist?”InternationalSecurity,Vol.24,No.2(Fall1999),pp. 5–55; and Chapter 8 in Robert L. Jervis, “Realism, Neoliberalism, and Cooperation: UnderstandingtheDabate,”InternationalSecurity,Vol.24,No.1(Summer1999),pp.42–63. On the cover: The Copernican solar system. We gratefully acknowledge permission to reproducethisillustrationfromtheUniversityofChicagoPress.RedrawnfromWilliamD. Stahlman from Galileo: Dialogue on the Great World Systems, revised, edited, and annotated by Giorzio de Santillano, University of Chicago Press, © 1953. We also thank DonnaWhippleforresearchanddesignassistance. Contents Foreword Thoughts about Assaying Theories vii Kenneth N. Waltz Acknowledgments xiii Chapter 1 Introduction: Appraising Progress in 1 International Relations Theory Colin Elman and Miriam Fendius Elman Chapter 2 Lessons from Lakatos 21 Colin Elman and Miriam Fendius Elman Part I Applying Lakatos: Judging Theoretical and Empirical Progress in IR Theory Chapter 3 Institutional Theory as a 71 Research Program Robert O. Keohane and Lisa L. Martin Chapter 4 The Power Transition Research Program: 109 A Lakatosian Analysis Jonathan M. DiCicco and Jack S. Levy Chapter 5 Liberal International Relations Theory: A 159 Scientific Assessment Andrew Moravcsik Chapter 6 A Lakatosian View of the 205 Democratic Peace Research Program James Lee Ray Chapter 7 Operational Code Analysis as a Scientific 245 Research Program: A Cautionary Tale Stephen G. Walker Chapter 8 Realism, Neoliberalism, and 277 Cooperation: Understanding the Debate Robert Jervis Chapter 9 The Progressiveness of Neoclassical 311 Realism Randall L. Schweller Chapter 10 “Is” and “Ought”: Evaluating Empirical 349 Aspects of Normative Research Jack Snyder Part II Commentaries on Lakatos, and Beyond Chapter 11 Explanation and Scientific Progress 381 David Dessler Chapter 12 Measuring Intra-programmatic Progress 405 Roslyn Simowitz Chapter 13 Kuhn vs. Lakatos? The Case for Multiple 419 Frames in Appraising IR Theory John A. Vasquez Chapter 14 A Lakatosian Reading of Lakatos: What 455 Can We Salvage from the Hard Core? Andrew Bennett About the Contributors 495 Index 499 About the Belfer Center for Science and International Affairs Foreword Thoughts about Assaying Theories Kenneth N. Waltz S tudents sometimes ask, with a hint of exasperation, why I assign Lakatos in seminars on international-political theory. One easily thinks of a number of reasons to omit him: Philosophy of science is a subject that demands clarity and precision; Lakatos’s prose is opaque and vague; reading his well-known essay on “Falsification and Research Programmes” provides no clear guide to the evaluation of theories.1 One certainly is not told just what to do. Yet the answer to why we should take Lakatos seriously is simple: He demolishes the simplistic notions about testing that have been and remain part of the intellectual stock of most students of political science. How can we assay theories? Karl Popper gave a pleasingly simple answer. First, make a conjecture, preferably a bold one such as all swans are white. Then, search for falsifying instances. Thousands or millions of white swans do not prove that all swans are white, but just one black swan proves the conjecture false. Simply multiplying observations that appear to offer confirmation will not do, because one cannot know of the lurking instance that would defeat the “theory.” Popper’s idea of the “critical” test rests on a distinction between trying to prove truth and being able to demonstrate falsity. Popper believed that the latter is possible; the former, not. 1. Imre Lakatos, “Falsification and the Methodology of Scientific Research Programmes,” in Imre Lakatos and Alan Musgrave, eds., Criticism and the Growth of Knowledge: Proceedings of the International Colloquium in the Philosophy of Science, London, 1965, vol. 4 (Cambridge: Cambridge University Press, 1970), pp. 91–196. viii | progress in international relations theory In 1970, Errol Harris published an insightful but little-known book, Hypothesis and Perception.2 His title suggested that the results of tests require interpretation, and his text developed the argument that seemingly critical tests are at best problematic. Tests attempting to falsify a theory are conducted against background information that in its day is taken for granted. How can we know that the background information is valid? Perhaps the black bird one thought a swan was really a turkey. That nothing is both empirical and certain is a proposition established long ago by David Hume and Immanuel Kant. If the bold conjecture seems to flunk the critical test, the scientist- observer still has decisions to make about the implications that are to be drawn from the outcome. Lakatos takes the problem up at this point. His dictum is that “we cannot prove theories and we cannot disprove them either.”3 He was right for this reason among others: Facts are no more independent of theories than theories are independent of facts. The validity of theories does not depend on facts that are simply given. Theory and fact are interdependent. As the English astronomer Sir Arthur Stanley Eddington put it, “We should not put overmuch confidence in the observational results that are put forward until they have been confirmed by theory.”4 A moment’s thought reveals the wisdom of his advice. The earth is the center of the universe, and the sun and other heavenly bodies swirl around it: These beliefs were among the “facts” accepted in antiquity and through the Middle Ages. They were easily “verified” by looking around; they conformed to everyday experience. From Copernicus onward, however, new theories changed old facts. Are these thoughts relevant for today’s political scientists? We have to believe so when we read the following statement in a widely consulted manual on the design of social inquiry: “A theory must be consistent with prior evidence about a research question.” To drive the 2. Errol Harris, Hypothesis and Perception (London: Allen and Unwin, 1970). 3. Lakatos, “Falsification and the Methodology of Scientific Research Programmes,” p. 100 (emphasis in original). 4. Sir Arthur Stanley Eddington, New Pathways in Science (New York: Macmillan, 1935), p. 211. foreword | ix point home the manual’s authors quote Stanley Lieberson’s pronouncement that a “theory that ignores existing evidence is an oxymoron.”5 Ironically, these thoughts are recorded in a chapter titled “The Science in Social Science.”6 The authors’ science, however, is of the medieval variety. Theories, they say, may “emerge from detailed observation, but they should be evaluated with new observations.” In this positivist perspective, facts are a source of theories and their arbiters as well. A theory is tested by confronting it with “the hard facts of empirical reality.”7 Yet seeming facts exist in infinite number. Which facts are to be taken as providing evidence for or against a theory? Because of the interdependence of theory and fact, one cannot give a simple answer. As Goethe put it, “The highest wisdom is to realize that every fact is already a theory.”8 According to the manual, a theory must indicate what evidence would show the theory wrong. According to Lakatos, a theory cannot specify the observations that would over- throw it.9 According to the manual, there is an asymmetry between proving something true and proving something false. The latter is possible; the former, not. Theories are to be tested by trying to falsify them. Yet according to Lakatos, “some of the most interesting experiments result ... in confirmation rather than falsification.”10 Steven 5. Gary King, Robert O. Keohane, and Sidney Verba, Designing Social Inquiry: Scientific Inference in Qualitative Research (Princeton, N.J.: Princeton University Press, 1994), p. 19. 6. Ibid., chap. 1. 7. Gary King, Robert O. Keohane, and Sidney Verba, “The Importance of Research Design in Political Science,” American Political Science Review, Vol. 89, No. 2 (June 1995), pp. 475–476. 8. Johann Wolfgang von Goethe, Wisdom and Experience, selections by Ludwig Curtius, trans. and ed. with an introduction by Hermann J. Weigand (New York: Pantheon, 1949), p. 94. 9. King, Keohane, and Verba, Designing Social Inquiry, p. 19. Lakatos puts his point this way: “exactly the most admired scientific theories simply fail to forbid any observable state of affairs.” Lakatos and Musgrave, Criticism and the Growth of Knowledge, p. 100 (emphasis in original). 10. Lakatos, “Falsification and the Methodology of Research Programmes,” p. 115. Harris argues that among natural scientists, falsification is a seldom used