Uncorrected Proofs:Copyright Cornell University 厂 Introduction Constructing the International Economy Rawi Abdelal,Mark Blyth,and Craig Parsons We have,as a rule,only the vaguest idea of any but the most direct consequences of our acts.Now the whole object of the accumulation of wealth is to produce results,or potential results,at a comparatively distant,and sometimes an infinitely distant date Thus the fact that our knowledge of the world is fluctuating,vague,and uncertain, renders wealth a peculiarly unsuitable topic for the methods of classical economic theory....About these matters there is no scientific basis on which to form a calcu- lable probability whatever.We simply do not know. John Maynard Keynes.1937 he world is.as they say.complicated.The world econom is especiall so.Unpredicted events often influence markets in improbable ways.Individu- als and organizations-firms,governments-surprise observers by behaving in ways that appear contrary to their presumed material interests as events defy the categories and concepts we construct to contain them.Crises recur with worri- some frequency.As the world internationalizes,these complications become more profound.Yet it would be difficult to find evidence of these complica- tions in much of the scholarly study of international political economy (IPE). Scholars of IPE have arrived at a comfortable certainty about how the world works.Most see the environment in which firms and governments operate as predominantly material.The incentives,they argue,that actors derive from the material structure of the economy determine what firms and governments do. These scholarly constructs are generally presumed by their constructors to cor- respond to reality rather than to represent stylizations of the heuristics that,we assume,inform decision making within organizations(Blyth 2007a).As their materialist theories and rationalist models become more sophisticated,these scholars have hoped that the world will become more knowable.And to some extent,progress has,in fact,been made.We grow more and more satisfied with our ability to explain the world to ever more detailed degrees. Tremendous gaps in our understanding still exist,however,because scholar- ship based on the connection between observer-deduced material incentives E L 1 3050-544_01Intro.indd 1 1125/20098:4609AM
1 Introduction Constructing the International Economy Rawi Abdelal, Mark Blyth, and Craig Parsons We have, as a rule, only the vaguest idea of any but the most direct consequences of our acts. Now the whole object of the accumulation of wealth is to produce results, or potential results, at a comparatively distant, and sometimes an infi nitely distant date. Thus the fact that our knowledge of the world is fl uctuating, vague, and uncertain, renders wealth a peculiarly unsuitable topic for the methods of classical economic theory. . . . About these matters there is no scientifi c basis on which to form a calculable probability whatever. We simply do not know. John Maynard Keynes, 1937 The world is, as they say, complicated. The world economy is especially so. Unpredicted events often infl uence markets in improbable ways. Individuals and organizations—fi rms, governments—surprise observers by behaving in ways that appear contrary to their presumed material interests as events defy the categories and concepts we construct to contain them. Crises recur with worrisome frequency. As the world internationalizes, these complications become more profound. Yet it would be diffi cult to fi nd evidence of these complications in much of the scholarly study of international political economy (IPE). Scholars of IPE have arrived at a comfortable certainty about how the world works. Most see the environment in which fi rms and governments operate as predominantly material. The incentives, they argue, that actors derive from the material structure of the economy determine what fi rms and governments do. These scholarly constructs are generally presumed by their constructors to correspond to reality rather than to represent stylizations of the heuristics that, we assume, inform decision making within organizations (Blyth 2007a). As their materialist theories and rationalist models become more sophisticated, these scholars have hoped that the world will become more knowable. And to some extent, progress has, in fact, been made. We grow more and more satisfi ed with our ability to explain the world to ever more detailed degrees. Tremendous gaps in our understanding still exist, however, because scholarship based on the connection between observer-deduced material incentives _S _E _L 3050-544_01Intro.indd 1 11/25/2009 8:46:09 AM Uncorrected Proofs: Copyright Cornell University
Uncorrected Proofs:Copyright Cornell University 2 Constructing the International Economy and observer-imputed rational responses leaves a great many empirical ques- tions unanswered.Some answers that seemed useful enough to many schol- ars,moreover,have over time been undermined by the wide diversity in the ways that firms and governments react to seemingly similar situations.The re- vealed rules of economic action seem to shift erratically across time and space. Increasingly,many scholars of international political economy have come to suspect that purely social phenomena intervene between actors and material structures.Structures do not come with instruction sheets for managers and policymakers(Blyth 2o03b).Instead,people can interpret their material en- vironments in very different ways.Indeed,so many "similar people"make so many "dissimilar choices"that our mainstream theories correspond,at best, only to time-and space-specific subsets of the world economy (Berman 2006; Vogel 1996). In this book we advance a strong version of the following claim:the assump- tion of a purely materialist view of theory is not-and never was-tenable.All political economy scholarship needs at least to consider,as a plausible hypoth- esis,that economies might vary substantially for nonmaterial reasons.In other words,the field needs to engage more systematically with constructivism,a the- oretical approach that emphasizes precisely those nonmaterial influences on both institutions and practices. The central insight of constructivism is that collectively held ideas shape the social,economic,and political world in which we live (Wendt 1999;Hacking 1999).That is,the world in which firms and governments act is neither trans- parent nor,in their own interpretations,similar across all cases.What objects mean to agents,how collectively held ideas differentially filter the environment for agents,how agents creatively respond to uncertainty,and how agents are coextensive and interdependent with the world in which they both act and ana- lyze all vary (and matter)too much for the assumptions of a simple material correspondence theory to hold.Building in diverse ways on this central insight, scholars across disciplines have increasingly embraced insights that either re- semble or are informed by constructivist theory. Beyond political science scholarship in international or comparative political economy,economic sociologists,for example,have for decades explored the influence of the social facts of the world on economic practices.'Increasingly, many economists believe that "cultural economics"and the influence of beliefs represent the cutting edge of research.2 Meanwhile,psychologists,anthropolo- gists,and (even some)political scientists continue to make progress in our un- derstanding of how identities and beliefs influence both individual preferences and societal practices.3 In political science,however,it is in the subfield of in- ternational relations where the revolution has cut most deeply.Today the study 1.See,for example,Swedberg 2003:Dobbin 2004a,2004b;Smelser and Swedberg 2005:and S Nee and Swedberg 2005. 2.See Di Tella and Dubra 2008:Greif 2006. E 3.See,for example,Geertz 1974;and Abdelal et al.2006. L 3050-544_01Intro.indd 2 1125/20098:4609AM
2 Constructing the International Economy and observer-imputed rational responses leaves a great many empirical questions unanswered. Some answers that seemed useful enough to many scholars, moreover, have over time been undermined by the wide diversity in the ways that fi rms and governments react to seemingly similar situations. The revealed rules of economic action seem to shift erratically across time and space. Increasingly, many scholars of international political economy have come to suspect that purely social phenomena intervene between actors and material structures. Structures do not come with instruction sheets for managers and policymakers (Blyth 2003b). Instead, people can interpret their material environments in very different ways. Indeed, so many “similar people” make so many “dissimilar choices” that our mainstream theories correspond, at best, only to time- and space-specifi c subsets of the world economy (Berman 2006; Vogel 1996). In this book we advance a strong version of the following claim: the assumption of a purely materialist view of theory is not—and never was—tenable. All political economy scholarship needs at least to consider, as a plausible hypothesis, that economies might vary substantially for nonmaterial reasons. In other words, the fi eld needs to engage more systematically with constructivism, a theoretical approach that emphasizes precisely those nonmaterial infl uences on both institutions and practices. The central insight of constructivism is that collectively held ideas shape the social, economic, and political world in which we live (Wendt 1999; Hacking 1999). That is, the world in which fi rms and governments act is neither transparent nor, in their own interpretations, similar across all cases. What objects mean to agents, how collectively held ideas differentially fi lter the environment for agents, how agents creatively respond to uncertainty, and how agents are coextensive and interdependent with the world in which they both act and analyze all vary (and matter) too much for the assumptions of a simple material correspondence theory to hold. Building in diverse ways on this central insight, scholars across disciplines have increasingly embraced insights that either resemble or are informed by constructivist theory. Beyond political science scholarship in international or comparative political economy, economic sociologists, for example, have for decades explored the infl uence of the social facts of the world on economic practices.1 Increasingly, many economists believe that “cultural economics” and the infl uence of beliefs represent the cutting edge of research.2 Meanwhile, psychologists, anthropologists, and (even some) political scientists continue to make progress in our understanding of how identities and beliefs infl uence both individual preferences and societal practices.3 In political science, however, it is in the subfi eld of international relations where the revolution has cut most deeply. Today the study 1. See, for example, Swedberg 2003; Dobbin 2004a, 2004b; Smelser and Swedberg 2005; and Nee and Swedberg 2005. 2. See Di Tella and Dubra 2008; Greif 2006. 3. See, for example, Geertz 1974; and Abdelal et al. 2006. S_ E_ L_ 3050-544_01Intro.indd 2 11/25/2009 8:46:09 AM Uncorrected Proofs: Copyright Cornell University
Uncorrected Proofs:Copyright Cornell University Introduction 3 of conflict,war,and peace is well nigh dominated by constructivist arguments, which increasingly appear as commonsensical notions in popular media. Scholarly insistence on the importance of social constructions is so prevalent as to border on the cliche.Identities,norms,beliefs,and symbols seem to be everywhere. Everywhere,that is,except in the mainstream of international political econ- omy,which has remained resistant to this trend.As used to be the case elsewhere, the view of the world that still informs much political economy scholarship is materialist and rationalist.That is,the vast majority of IPE scholars assert that we can derive sufficient explanations of action from some function of ratio- nal responses to objective and largely knowable and transparent environments (Rogowski 1987;Keohane and Milner 1996). A recent survey by Jeffry Frieden and Lisa Martin goes so far as to argue that the field of IPE has approached a "consensus on theories,methods,analytical frameworks,and important questions,"at least for "North American scholar- ship"and less so for "much European scholarship"(200g:119).Though their review charts the many impressive accomplishments of a flourishing field,no- tably absent are not only "European"ideas,but also the norms and identities that characterize economic sociology and constructivist political economy.Para- doxically,as some IPE scholars have converged to an understanding of the field that mimics the analytical language of economics,economists themselves are embracing culture and beliefs more systematically as essential elements in a use- ful theoretical tool kit.Thus the basic approach has become standard:"Scholars identify the socioeconomic interests at stake,how groups are organized,and how organized and general interests express themselves in the context of do- mestic electoral,legislative,and bureaucratic institutions.On this basis,other work focuses on the interaction of states at the international level,emphasizing especially the impact of the informational environment and the strategic setting on relations among states"(Frieden and Martin 2003:145). According to this traditional thinking,political action varies not with different interpretations of the world,but with the resources people hold and the relative position they inhabit in arenas like markets(Frieden 1991b),security competi- tions (Powell 1996),and contests for domestic political power (Snyder 1991). In other words,people select certain actions as a rational response to their place in an environment implicitly characterized as an "obstacle course,"in which payoffs may be opaque,but they are knowable;and it is this "correspondence theory of the world(plus a behavioral rule or two)that drives action. The classic versions of such thinking in international relations-Marxism, Liberalism,and Realism-share this vision of an obstacle course composed of an exogenously given distribution of resources and technology.Each of these schools offers a certain theoretical description of the obstacle course,and 4.Katzenstein 1996a,1996b. 5.See Katzenstein,Keohane,and Krasner 1998;and Ruggie 1998. E L 3050-544_01Intro.indd 3 1125/20098:46-09AM
Introduction 3 of confl ict, war, and peace is well nigh dominated by constructivist arguments, which increasingly appear as commonsensical notions in popular media.4 Scholarly insistence on the importance of social constructions is so prevalent as to border on the cliché. Identities, norms, beliefs, and symbols seem to be everywhere. Everywhere, that is, except in the mainstream of international political economy, which has remained resistant to this trend. As used to be the case elsewhere, the view of the world that still informs much political economy scholarship is materialist and rationalist.5 That is, the vast majority of IPE scholars assert that we can derive suffi cient explanations of action from some function of rational responses to objective and largely knowable and transparent environments (Rogowski 1987; Keohane and Milner 1996). A recent survey by Jeffry Frieden and Lisa Martin goes so far as to argue that the fi eld of IPE has approached a “consensus on theories, methods, analytical frameworks, and important questions,” at least for “North American scholarship” and less so for “much European scholarship” (2003:119). Though their review charts the many impressive accomplishments of a fl ourishing fi eld, notably absent are not only “European” ideas, but also the norms and identities that characterize economic sociology and constructivist political economy. Paradoxically, as some IPE scholars have converged to an understanding of the fi eld that mimics the analytical language of economics, economists themselves are embracing culture and beliefs more systematically as essential elements in a useful theoretical tool kit. Thus the basic approach has become standard: “Scholars identify the socioeconomic interests at stake, how groups are organized, and how organized and general interests express themselves in the context of domestic electoral, legislative, and bureaucratic institutions. On this basis, other work focuses on the interaction of states at the international level, emphasizing especially the impact of the informational environment and the strategic setting on relations among states” (Frieden and Martin 2003:145). According to this traditional thinking, political action varies not with different interpretations of the world, but with the resources people hold and the relative position they inhabit in arenas like markets (Frieden 1991b), security competitions (Powell 1996), and contests for domestic political power (Snyder 1991). In other words, people select certain actions as a rational response to their place in an environment implicitly characterized as an “obstacle course,” in which payoffs may be opaque, but they are knowable; and it is this “correspondence theory of the world” (plus a behavioral rule or two) that drives action. The classic versions of such thinking in international relations—Marxism, Liberalism, and Realism—share this vision of an obstacle course composed of an exogenously given distribution of resources and technology. Each of these schools offers a certain theoretical description of the obstacle course, and 4. Katzenstein 1996a, 1996b. 5. See Katzenstein, Keohane, and Krasner 1998; and Ruggie 1998. _S _E _L 3050-544_01Intro.indd 3 11/25/2009 8:46:09 AM Uncorrected Proofs: Copyright Cornell University
Uncorrected Proofs:Copyright Cornell University 4 Constructing the International Economy argues that generally smart and resourceful actors figure out what the world "really"looks like and strategize accordingly.In comparative politics (and in some parts of international relations),more recent institutionalist scholarship emphasizes the man-made organizational aspects of the obstacle course (Pier- son 2004;Thelen and Streeck 2005).This focus on the man-made aspects of the environment leads to a stress on mechanisms of feedback and path depen- dence since it implies that actions at one point change the man-made con- straints for subsequent action,but otherwise both views are equally rationalist in downplaying interpretation.5 Both posit rational people responding to an obstacle course that any human being would perceive fairly similarly.From neo-Realists to rationalist institutionalists and back again,material or organi- zational features dictate the interests that explain actions.Ideas,norms,and their attendant uncertainties are conspicuous by their absence. This is so primarily because the context for choice is characterized as essen- tially asocial.In security studies,that context was defined by the fact of anarchy, that is,the absence of a world government,and the standard starting point was to deduce meaningful hypotheses about state behavior from that simple, asocial fact.In much of IPE,the basic context was seen as the market,albeit an unusually austere,utterly material market.Yet the "facts"of anarchy or of the market,scholars have repeatedly shown,have taken us not very far toward a richer understanding of how the world works. Change,however,is afoot,even within scholarship on the international po- litical economy.An increasing number of scholars have highlighted empirical puzzles and posed research questions during the past fifteen years that they see as answerable only through attention to the influence of social construc- tions.These have not been marginal issues or irrelevant findings.Such scholars have argued that the relationship between the welfare state and the interna- tional economy (Blyth 2002),the trade and financial practices of developed and developing countries(Duina 2006;MacKenzie 2006),the organization of the international monetary and financial system(Best 2005;Abdelal 2006 and 2007),the character and logic of European integration (Parsons 2003;Jabko 2006),or the policies underlying Eurasian disintegration (Abdelal 2001;Her- rera 2005)have all varied in major ways due to the influence of norms,identi- ties,and social interpretations. As highly regarded as many of these contributions are,though,their collec- tive impact on the field as a whole has not yet been transformative.Engagement with alternate constructivist hypotheses is still not an obligation for scholars publishing in leading journals or with university presses.International politi- cal economy is still regularly practiced and taught without serious analysis of the norms and identities that could provide the foundation for a more useful, convincing account of the empirical puzzles that drive research.Scholars whose S 6.Other"historical institutionalist"and "sociological institutionalist"scholars take the "institu- E tionalist"label in partly or fully constructivist directions. L 3050-544_01Intro.indd 4 1125/20098:4609AM
4 Constructing the International Economy argues that generally smart and resourceful actors fi gure out what the world “really” looks like and strategize accordingly. In comparative politics (and in some parts of international relations), more recent institutionalist scholarship emphasizes the man-made organizational aspects of the obstacle course (Pierson 2004; Thelen and Streeck 2005). This focus on the man-made aspects of the environment leads to a stress on mechanisms of feedback and path dependence since it implies that actions at one point change the man-made constraints for subsequent action, but otherwise both views are equally rationalist in downplaying interpretation.6 Both posit rational people responding to an obstacle course that any human being would perceive fairly similarly. From neo-Realists to rationalist institutionalists and back again, material or organizational features dictate the interests that explain actions. Ideas, norms, and their attendant uncertainties are conspicuous by their absence. This is so primarily because the context for choice is characterized as essentially asocial. In security studies, that context was defi ned by the fact of anarchy, that is, the absence of a world government, and the standard starting point was to deduce meaningful hypotheses about state behavior from that simple, asocial fact. In much of IPE, the basic context was seen as the market, albeit an unusually austere, utterly material market. Yet the “facts” of anarchy or of the market, scholars have repeatedly shown, have taken us not very far toward a richer understanding of how the world works. Change, however, is afoot, even within scholarship on the international political economy. An increasing number of scholars have highlighted empirical puzzles and posed research questions during the past fi fteen years that they see as answerable only through attention to the infl uence of social constructions. These have not been marginal issues or irrelevant fi ndings. Such scholars have argued that the relationship between the welfare state and the international economy (Blyth 2002), the trade and fi nancial practices of developed and developing countries (Duina 2006; MacKenzie 2006), the organization of the international monetary and fi nancial system (Best 2005; Abdelal 2006 and 2007), the character and logic of European integration (Parsons 2003; Jabko 2006), or the policies underlying Eurasian disintegration (Abdelal 2001; Herrera 2005) have all varied in major ways due to the infl uence of norms, identities, and social interpretations. As highly regarded as many of these contributions are, though, their collective impact on the fi eld as a whole has not yet been transformative. Engagement with alternate constructivist hypotheses is still not an obligation for scholars publishing in leading journals or with university presses. International political economy is still regularly practiced and taught without serious analysis of the norms and identities that could provide the foundation for a more useful, convincing account of the empirical puzzles that drive research. Scholars whose 6. Other “historical institutionalist” and “sociological institutionalist” scholars take the “institutionalist” label in partly or fully constructivist directions. S_ E_ L_ 3050-544_01Intro.indd 4 11/25/2009 8:46:09 AM Uncorrected Proofs: Copyright Cornell University
Uncorrected Proofs:Copyright Cornell University Introduction 5 work is informed by constructivism have uncovered a wealth of evidence for novel explanations of important aspects of the world economy,but given their diverse starting points and theoretical orientations,the relationships among them remain unclear.The collective force of a constructivist research agenda can be hard to see,and not only for the mainstream IPE theorists who have paid little heed so far.Even constructivists themselves have difficulty specifying the common ground and different emphases within this growing body of work. For constructivism to become a more systematic part of central debates in all of political economy,this diversity must be organized into a more coherent set of complementary positions. This book aims to effect that organization,but not by proposing a synthetic constructivist"theory of everything."Though one of constructivism's challenges is terminological confusion,there is also real,substantive diversity among con- structivists.Scholars can agree on the importance of socially constructed inter- pretation but still make a variety of distinct arguments about the world.From this mix of positions we propose to lay out what constructivism in IPE looks like, in three ways.First,we outline and relate several different arguments for why scholars might see reasons for attention to social construction,inviting the wid- est possible array of scholars to engage with constructivism.Second,we examine points of terminological or theoretical confusion that create unnecessary barri- ers to engagement between constructivist and nonconstructivist work,and also among styles of constructivism.Third,we articulate several substantively differ- ent varieties of constructivism,organizing the tool kit that both constructivists and their critics can use to debate how much and when social construction matters. Our inspiration for this book is the most important "first generation"con- structivist project,Peter Katzenstein's 1996 book The Culture of National Security. Some departures from that model underscore,however,that the challenges for "second generation"constructivists are somewhat different.Katzenstein and his collaborators chose the terrain of long-dominant rationalist-materialist realism in security studies to elaborate the core themes of a constructivist research pro- gram.A decade later,with constructivist arguments spreading across issue areas and diversifying in theoretical logics,we survey and organize constructivist work to suggest its implications for the major subfield it has penetrated least,IPE. In so doing we mean both to broaden and to sharpen debates in IPE,and to advance constructivist theorizing more generally. Speaking to Two Audiences without Being Janus-Faced In this book we seek to advance the debate over the place of constructivism in IPE scholarship in two directions.In this introduction we try not to sing to the choir.That is,we try to lay out the broadest possible case for taking construc- tivism seriously for skeptics of constructivism.Prior to the logical,epistemologi- cal,and empirical force of any given constructivist claim,we need to spell out E L 3050-544_01Intro.indd 5 1125/20098:4609AM
Introduction 5 work is informed by constructivism have uncovered a wealth of evidence for novel explanations of important aspects of the world economy, but given their diverse starting points and theoretical orientations, the relationships among them remain unclear. The collective force of a constructivist research agenda can be hard to see, and not only for the mainstream IPE theorists who have paid little heed so far. Even constructivists themselves have diffi culty specifying the common ground and different emphases within this growing body of work. For constructivism to become a more systematic part of central debates in all of political economy, this diversity must be organized into a more coherent set of complementary positions. This book aims to effect that organization, but not by proposing a synthetic constructivist “theory of everything.” Though one of constructivism’s challenges is terminological confusion, there is also real, substantive diversity among constructivists. Scholars can agree on the importance of socially constructed interpretation but still make a variety of distinct arguments about the world. From this mix of positions we propose to lay out what constructivism in IPE looks like, in three ways. First, we outline and relate several different arguments for why scholars might see reasons for attention to social construction, inviting the widest possible array of scholars to engage with constructivism. Second, we examine points of terminological or theoretical confusion that create unnecessary barriers to engagement between constructivist and nonconstructivist work, and also among styles of constructivism. Third, we articulate several substantively different varieties of constructivism, organizing the tool kit that both constructivists and their critics can use to debate how much and when social construction matters. Our inspiration for this book is the most important “fi rst generation” constructivist project, Peter Katzenstein’s 1996 book The Culture of National Security. Some departures from that model underscore, however, that the challenges for “second generation” constructivists are somewhat different. Katzenstein and his collaborators chose the terrain of long-dominant rationalist- materialist realism in security studies to elaborate the core themes of a constructivist research program. A decade later, with constructivist arguments spreading across issue areas and diversifying in theoretical logics, we survey and organize constructivist work to suggest its implications for the major subfi eld it has penetrated least, IPE. In so doing we mean both to broaden and to sharpen debates in IPE, and to advance constructivist theorizing more generally. Speaking to Two Audiences without Being Janus-Faced In this book we seek to advance the debate over the place of constructivism in IPE scholarship in two directions. In this introduction we try not to sing to the choir. That is, we try to lay out the broadest possible case for taking constructivism seriously for skeptics of constructivism. Prior to the logical, epistemological, and empirical force of any given constructivist claim, we need to spell out _S _E _L 3050-544_01Intro.indd 5 11/25/2009 8:46:09 AM Uncorrected Proofs: Copyright Cornell University
Uncorrected Proofs:Copyright Cornell University 6 Constructing the International Economy why differently inclined scholars should accept that social construction could matter for what they care about explaining,and why it consequently deserves consideration as a plausible alternative in their research designs.To this end we lay out four different "paths to constructivism":rationales that start from different theoretical positions but all suggest that social construction matters. We then address the much-discussed divide between nonconstructivist "causal- explanatory"scholarship and the constructivist focus on meaning and "consti- tutiveness."We argue that many constructivist claims are more open to direct engagement with traditional causal-explanatory schemas than skeptics on either side may realize.One does not have to "give up on causation"or "devolve into mere description"to take constructivism seriously.Some constructivist scholar- ship does resist causal claims,of course,but we suggest that even across an ex- plicit version of this epistemological divide,scholars can usefully engage much more than they do. In sum,we aim to bring constructivism into conversation with the mainstream of rationalist-materialist theorizing in IPE.In important ways this effort at pro- moting engagement also seeks to reverse a growing geographical segregation in political economy.In European journals and faculties that are otherwise in- creasingly integrated with American social-science debates,arguments about social construction have become very influential in studies of international or- ganizations,trade,finance,and markets.In American IPE,meanwhile,success- ful careers are still routinely made without any real engagement of any variant of constructivism (whatever its provenance)(Cohen 2007;Higgot and Watson 2008;Ravenhill 2008).From our point of view,this means that large numbers of smart,well-meaning scholars on both sides of the Atlantic are forgoing de- bates that could improve all their arguments.They should and can fruitfully debate one another. Why American IPE Is Different,and Why This Matters Although scholars outside the United States do not need reminding of this fact, it is worth recalling how "American"the origins of IPE are (see Cohen 2007, 2008).The origins of IPE date to a period when concern over the United States' putatively terminal economic and military decline was at its height-or depth, as it were-during the 197os and 198os.More comparatively oriented schol- ars picked up this declinist thesis too,through the analysis of how some states handled the oil and inflation shocks of the 197os better than others,and sought to explain this variation accordingly(Katzenstein 1976b).For others scholars, though,most notably Robert Gilpin,what we were seeing,once again,was an ex- ample of hegemonic decline (Gilpin 1981).Indeed,whether the United States was seen as a“malignant'”or“benign”hegemon(Russett 1985)ran second- ary to the consensus that the United States was in fact in decline,and the main research problem was then how to understand this and,if possible,arrest it. E_ As a consequence,the entire research program of hegemonic stability theory L 3050-544_01Intro.indd 6 1125/20098:4609AM
6 Constructing the International Economy why differently inclined scholars should accept that social construction could matter for what they care about explaining, and why it consequently deserves consideration as a plausible alternative in their research designs. To this end we lay out four different “paths to constructivism”: rationales that start from different theoretical positions but all suggest that social construction matters. We then address the much-discussed divide between nonconstructivist “causalexplanatory” scholarship and the constructivist focus on meaning and “constitutiveness.” We argue that many constructivist claims are more open to direct engagement with traditional causal-explanatory schemas than skeptics on either side may realize. One does not have to “give up on causation” or “devolve into mere description” to take constructivism seriously. Some constructivist scholarship does resist causal claims, of course, but we suggest that even across an explicit version of this epistemological divide, scholars can usefully engage much more than they do. In sum, we aim to bring constructivism into conversation with the mainstream of rationalist-materialist theorizing in IPE. In important ways this effort at promoting engagement also seeks to reverse a growing geographical segregation in political economy. In European journals and faculties that are otherwise increasingly integrated with American social-science debates, arguments about social construction have become very infl uential in studies of international organizations, trade, fi nance, and markets. In American IPE, meanwhile, successful careers are still routinely made without any real engagement of any variant of constructivism (whatever its provenance) (Cohen 2007; Higgot and Watson 2008; Ravenhill 2008). From our point of view, this means that large numbers of smart, well-meaning scholars on both sides of the Atlantic are forgoing debates that could improve all their arguments. They should and can fruitfully debate one another. Why American IPE Is Different, and Why This Matters Although scholars outside the United States do not need reminding of this fact, it is worth recalling how “American” the origins of IPE are (see Cohen 2007, 2008). The origins of IPE date to a period when concern over the United States’ putatively terminal economic and military decline was at its height—or depth, as it were—during the 1970s and 1980s. More comparatively oriented scholars picked up this declinist thesis too, through the analysis of how some states handled the oil and infl ation shocks of the 1970s better than others, and sought to explain this variation accordingly (Katzenstein 1976b). For others scholars, though, most notably Robert Gilpin, what we were seeing, once again, was an example of hegemonic decline (Gilpin 1981). Indeed, whether the United States was seen as a “malignant” or “benign” hegemon (Russett 1985) ran secondary to the consensus that the United States was in fact in decline, and the main research problem was then how to understand this and, if possible, arrest it. As a consequence, the entire research program of hegemonic stability theory S_ E_ L_ 3050-544_01Intro.indd 6 11/25/2009 8:46:09 AM Uncorrected Proofs: Copyright Cornell University
Uncorrected Proofs:Copyright Cornell University Introduction 7 (Gilpin 1981;McKeown 198g)about how international regimes might "stave off"the inevitable undersupply of international public goods that the hegemon provides (Krasner 1982),and how international institutions could be second- best solutions to decline (Keohane 1984),came to the fore and dominated the field for a decade.? What is of interest here,then,is how,despite the certainty pertaining to this scholarship and its thoroughly materialist underpinnings,practically all its key claims ran aground on the unexpected politics and historical events of the late 198os and early 1ggos.The end of the Soviet Union,the unification of Ger- many,the renaissance of the European Union,and the reemergence of global finance as a major force in international politics combined to send the suppos- edly rising hegemon,Japan,into a recession now in its second decade,while the declining hegemon was suddenly,and quite unexpectedly,upgraded to "hy- perpower"status.That such a large swath of the field missed what was going on by such a wide margin might have suggested that focusing only on material factors,such as relative shares of world trade,was perhaps less illuminating than one might hope.In reply to this failure,mainstream IPE picked itself up and set about looking for the materialist determinants of political action under new conditions.They quickly settled on internationalization and globalization(Frie- den 1991b;Keohane and Milner 1996). While attentive to new shifts in the global economy that had blindsided the older school,such as the sudden importance of finance as well as trade,this next- generation school committed the same error of assuming too much stability and predictability in the world based on seeing the environment as a materialist obstacle course.For example,some of the most influential materialist analy- ses of how financial globalization would affect agents'interests provide clear predictions as to which sectors and which actors,with which assets,would win and lose under globalization(Frieden 1991b;Keohane and Milner 1996).Thus predictions about preferences and likely coalitions could be made.Unfortu- nately,although these models provide clear predictions,the reality they purport to explain has not,once again,been very compliant.As empirical work on the preferences of real financial actors shows clearly,some sectors do not organize despite their assets,and some actors have exactly the opposite or even no prefer- ences at all over alternative policies (McNamara 1998;Mosley 2003;Jabko 2006; Abdelal 2006;see also Crystal 2o03b).Similarly,consider materialist scholar- ship on the future of the welfare state which saw open capital flows leading to a race to the bottom in taxes,the death of expensive models of welfare,and a convergence on a neoliberal form of capitalism(Kurzer 1993).Instead,what we see today are the high-spending and high-taxing social democratic welfare states ranking among the most economically successful in the global economy. 7.Indeed,the two positions that dominated the field,realism and neoliberal institutionalism, differed on little more than the imputed behavioral rule governing state behavior(Baldwin 1993; Powell 1994). E L 3050-544_01Intro.indd 7 1125/20098:4609AM
Introduction 7 (Gilpin 1981; McKeown 1983) about how international regimes might “stave off ” the inevitable undersupply of international public goods that the hegemon provides (Krasner 1982), and how international institutions could be secondbest solutions to decline (Keohane 1984), came to the fore and dominated the fi eld for a decade.7 What is of interest here, then, is how, despite the certainty pertaining to this scholarship and its thoroughly materialist underpinnings, practically all its key claims ran aground on the unexpected politics and historical events of the late 1980s and early 1990s. The end of the Soviet Union, the unifi cation of Germany, the renaissance of the European Union, and the reemergence of global fi nance as a major force in international politics combined to send the supposedly rising hegemon, Japan, into a recession now in its second decade, while the declining hegemon was suddenly, and quite unexpectedly, upgraded to “hyperpower” status. That such a large swath of the fi eld missed what was going on by such a wide margin might have suggested that focusing only on material factors, such as relative shares of world trade, was perhaps less illuminating than one might hope. In reply to this failure, mainstream IPE picked itself up and set about looking for the materialist determinants of political action under new conditions. They quickly settled on internationalization and globalization (Frieden 1991b; Keohane and Milner 1996). While attentive to new shifts in the global economy that had blindsided the older school, such as the sudden importance of fi nance as well as trade, this nextgeneration school committed the same error of assuming too much stability and predictability in the world based on seeing the environment as a materialist obstacle course. For example, some of the most infl uential materialist analyses of how fi nancial globalization would affect agents’ interests provide clear predictions as to which sectors and which actors, with which assets, would win and lose under globalization (Frieden 1991b; Keohane and Milner 1996). Thus predictions about preferences and likely coalitions could be made. Unfortunately, although these models provide clear predictions, the reality they purport to explain has not, once again, been very compliant. As empirical work on the preferences of real fi nancial actors shows clearly, some sectors do not organize despite their assets, and some actors have exactly the opposite or even no preferences at all over alternative policies (McNamara 1998; Mosley 2003; Jabko 2006; Abdelal 2006; see also Crystal 2003b). Similarly, consider materialist scholarship on the future of the welfare state which saw open capital fl ows leading to a race to the bottom in taxes, the death of expensive models of welfare, and a convergence on a neoliberal form of capitalism (Kurzer 1993). Instead, what we see today are the high-spending and high-taxing social democratic welfare states ranking among the most economically successful in the global economy. 7. Indeed, the two positions that dominated the fi eld, realism and neoliberal institutionalism, differed on little more than the imputed behavioral rule governing state behavior (Baldwin 1993; Powell 1994). _S _E _L 3050-544_01Intro.indd 7 11/25/2009 8:46:09 AM Uncorrected Proofs: Copyright Cornell University
Uncorrected Proofs:Copyright Cornell University 8 Constructing the International Economy The Multiple Paths to Constructivism Our point is not to pillory any and all materialist claims about politics.We wish to encourage engagement and debate,not to insist on the obvious superiority of a different selective take on the world.We think that a focus on material factors alone,one that does not even engage constructivist alternatives,limits what IPE scholars of any persuasion can hope to achieve.The same is true in reverse. But arguing generally for wide debate across theoretical alternatives is only a small step toward taking constructivism seriously.Mainstream IPE scholars tend to neglect constructivism not because they think their own approaches so airtight that clashes with alternatives are unimportant,but because they think constructivism in particular cannot offer legitimate or coherent alternatives. Many see constructivism as“relativistic,”“lacking a research program,”“being against science,"and "being against explanation"(Keohane 1993).This is not simply demonization,of course:though constructivists sometimes use somewhat different terms,many describe themselves as having distinct views of science and explanation.Yet as suggested above,we think that some misunderstandings lurk beneath this divide,and that even the real differences it does comprise are not obstacles to serious engagement.A first step to make the case for such engagement is to consider why scholars from a variety of theoretical points of departure might come to suspect that social construction deserves their atten- tion.What are the different paths to constructivism? Most self-consciously constructivist work to date justifies itself by what we call the path of meaning.Others arrive at similar analyses,however,through quite distinct logics.We see paths to constructivism through cognitive science,en- vironmental uncertainty,and the emphasis on subjectivity in postmodern or post-structuralist theory.All these logics provide rationales for doing empiri- cal research that asks how much social construction matters.Each also carries some distinct implications about how social constructs tend to arise,and about the mechanisms for their effects.At this step,however,our key point is that dif ferent scholars could agree to engage constructivist analyses while perceiving different justifications for doing so. We begin our survey with the path of meaning,since it is the classic route to this literature.Then we address the paths of cognitive science and environmen- tal uncertainty,which we think are most likely to be persuasive to our audience of IPE skeptics.Finally,we discuss how postmodern theorists can and should play a positive role in these discussions and debates. The Path of Meaning Human dependence on meaning and interpretation is the foundation of most explicitly constructivist scholarship.In a line of thinking that extends from Max Weber through a variety of intermediaries in sociology to contemporary political- S science constructivists such as Nicolas Onuf and Alexander Wendt,the open- E ing for socially constructed variation in action lies not in the unpredictability or L 3050-544_01ntro.indd8 1125/20098:4609AM
8 Constructing the International Economy The Multiple Paths to Constructivism Our point is not to pillory any and all materialist claims about politics. We wish to encourage engagement and debate, not to insist on the obvious superiority of a different selective take on the world. We think that a focus on material factors alone, one that does not even engage constructivist alternatives, limits what IPE scholars of any persuasion can hope to achieve. The same is true in reverse. But arguing generally for wide debate across theoretical alternatives is only a small step toward taking constructivism seriously. Mainstream IPE scholars tend to neglect constructivism not because they think their own approaches so airtight that clashes with alternatives are unimportant, but because they think constructivism in particular cannot offer legitimate or coherent alternatives. Many see constructivism as “relativistic,” “lacking a research program,” “being against science,” and “being against explanation” (Keohane 1993). This is not simply demonization, of course: though constructivists sometimes use somewhat different terms, many describe themselves as having distinct views of science and explanation. Yet as suggested above, we think that some misunderstandings lurk beneath this divide, and that even the real differences it does comprise are not obstacles to serious engagement. A fi rst step to make the case for such engagement is to consider why scholars from a variety of theoretical points of departure might come to suspect that social construction deserves their attention. What are the different paths to constructivism? Most self-consciously constructivist work to date justifi es itself by what we call the path of meaning. Others arrive at similar analyses, however, through quite distinct logics. We see paths to constructivism through cognitive science, environmental uncertainty, and the emphasis on subjectivity in postmodern or post-structuralist theory. All these logics provide rationales for doing empirical research that asks how much social construction matters. Each also carries some distinct implications about how social constructs tend to arise, and about the mechanisms for their effects. At this step, however, our key point is that different scholars could agree to engage constructivist analyses while perceiving different justifi cations for doing so. We begin our survey with the path of meaning, since it is the classic route to this literature. Then we address the paths of cognitive science and environmental uncertainty, which we think are most likely to be persuasive to our audience of IPE skeptics. Finally, we discuss how postmodern theorists can and should play a positive role in these discussions and debates. The Path of Meaning Human dependence on meaning and interpretation is the foundation of most explicitly constructivist scholarship. In a line of thinking that extends from Max Weber through a variety of intermediaries in sociology to contemporary politicalscience constructivists such as Nicolas Onuf and Alexander Wendt, the opening for socially constructed variation in action lies not in the unpredictability or S_ E_ L_ 3050-544_01Intro.indd 8 11/25/2009 8:46:09 AM Uncorrected Proofs: Copyright Cornell University
Uncorrected Proofs:Copyright Cornell University Introduction 9 complexity of the material world but in its inert,almost meaningless relationship to human existence and choice (Onuf 1989;Wendt 199g).While people are fairly rational and their views and actions indeed vary with their material sur- roundings,views and actions also vary a great deal in terms of the myths,iden- tities,symbols,norms,and conventions that people construct to motivate and prioritize their actions.From this point of view,not even the least uncertain material environment is free of potential variation in meaning. Thus the identity of“teacher,,”“president,.”or“central banker”is constituted through the authority relationships that make it possible to recognize an agent as a teacher,a president,or a central banker (Hall zo0g).The emergence of such social constructs and the role identities they both presume and make pos- sible therefore narrows choices in action from a potentially wider range of op- tions that were conceivable in terms of nonconstructivist causes.Theories that neglect how actors assign meanings and how meanings vary may not just miss out on "thick description"of what actors think.If the presence of certain mean- ings constitutes certain actors as actors of type X rather than Y,and thereby causes certain actions to obtain over other possible (from a materialist stand- point)outcomes,then nonconstructivist theories may be systematically and demonstrably wrong. Indeed,a strong version of meaning-oriented constructivism holds that socie- ties and policymakers rarely,if ever,interpret the world around them in purely material terms.Rather,they endow the economies in which they are embedded with social purposes.These purposes are embedded in a variety of collective identities,including national identities.s Economic activity and international interdependence tend to be treated not as ends in themselves or as brute mate- rial constraints,but rather as modes of acting within the world according to dif- ferent constituted identities.How societies interpret the material processes of production and distribution,or how policymakers recognize patterns of inter- dependence as natural or worrisome,reflects purposes shared among members of an identity group.Such purposes may be in a process of either constant con- testation or broad,taken-for-granted acceptance.Such societal identities tend to be sui generis rather than categorical. An alternate,more institutional approach to meaning in IPE finds that norms of appropriate behavior are implied by state identities of various types(Buko- vansky 2002).Thus the boundaries of legitimate policymaking vary for "devel- oped,”“European,.”and“civilized”states..In this sense,.international norms define the boundaries of choice and thereby affect how societies,policymak- ers,and market participants discern the meaning of various policy stances.In macroeconomic policymaking,this effect is particularly profound because of the endogeneity of policy outcomes.As Jonathan Kirshner and Ilene Grabel have pointed out,policies that are deemed illegitimate by the international financial community,composed also of market participants,simply cannot S 8.See Abdelal 2001 and Abdelal et al.2006 on the purposive content of identities. E L 3050-544_01Intro.indd 9 1125/20098:4609AM
Introduction 9 complexity of the material world but in its inert, almost meaningless relationship to human existence and choice (Onuf 1989; Wendt 1999). While people are fairly rational and their views and actions indeed vary with their material surroundings, views and actions also vary a great deal in terms of the myths, identities, symbols, norms, and conventions that people construct to motivate and prioritize their actions. From this point of view, not even the least uncertain material environment is free of potential variation in meaning. Thus the identity of “teacher,” “president,” or “central banker” is constituted through the authority relationships that make it possible to recognize an agent as a teacher, a president, or a central banker (Hall 2009). The emergence of such social constructs and the role identities they both presume and make possible therefore narrows choices in action from a potentially wider range of options that were conceivable in terms of nonconstructivist causes. Theories that neglect how actors assign meanings and how meanings vary may not just miss out on “thick description” of what actors think. If the presence of certain meanings constitutes certain actors as actors of type X rather than Y, and thereby causes certain actions to obtain over other possible (from a materialist standpoint) outcomes, then nonconstructivist theories may be systematically and demonstrably wrong. Indeed, a strong version of meaning-oriented constructivism holds that societies and policymakers rarely, if ever, interpret the world around them in purely material terms. Rather, they endow the economies in which they are embedded with social purposes. These purposes are embedded in a variety of collective identities, including national identities.8 Economic activity and international interdependence tend to be treated not as ends in themselves or as brute material constraints, but rather as modes of acting within the world according to different constituted identities. How societies interpret the material processes of production and distribution, or how policymakers recognize patterns of interdependence as natural or worrisome, refl ects purposes shared among members of an identity group. Such purposes may be in a process of either constant contestation or broad, taken-for-granted acceptance. Such societal identities tend to be sui generis rather than categorical. An alternate, more institutional approach to meaning in IPE fi nds that norms of appropriate behavior are implied by state identities of various types (Bukovansky 2002). Thus the boundaries of legitimate policymaking vary for “developed,” “European,” and “civilized” states. In this sense, international norms defi ne the boundaries of choice and thereby affect how societies, policymakers, and market participants discern the meaning of various policy stances. In macroeconomic policymaking, this effect is particularly profound because of the endogeneity of policy outcomes. As Jonathan Kirshner and Ilene Grabel have pointed out, policies that are deemed illegitimate by the international fi nancial community, composed also of market participants, simply cannot 8. See Abdelal 2001 and Abdelal et al. 2006 on the purposive content of identities. _S _E _L 3050-544_01Intro.indd 9 11/25/2009 8:46:09 AM Uncorrected Proofs: Copyright Cornell University
Uncorrected Proofs:Copyright Cornell University 10 Constructing the International Economy succeed:capital outflows sparked by an out-of-bounds policy can undermine a choice that,at another historical moment,may have been a perfectly plausible response to a policy challenge (Kirshner 2o0gb;Grabel 200g;Abdelal 2007). Michael Barnett and Martha Finnemore (2004),for example,have examined the role of international organizations in fixing meanings,thereby constituting the legitimate boundaries of policymaking.Although such processes are rarely observed as evidence of the attempt to fix meanings,political and economic organizations such as intergovernmental agencies and firms in fact regularly contest and thereby attempt to define the boundaries of acceptable practices and legitimate policy through the socialization (and persuasion)of other actors in the system.Thus the decline and eventual fall of otherwise regular economic transactions-buying and selling persons,invading indebted countries,mon- etizing government debt,and Asian "crony capitalism,"for example-were pre- cipitated by the purposeful efforts of ideational entrepreneurs to brand them illegitimate (Keck and Sikkink 1998;Finnemore 2003;Hall 2003). The Path of Cognition Another prominent line of theorizing that leads to constructivism takes a path- way through cognition.Many experimental psychologists and cultural sociolo- gists have long argued that human beings depend on heuristics and shortcuts to organize action and choice (Kahneman,Slovik,and Tversky 1982;Swidler 1986).Their core observation is not that agents search for significance and purpose,but that human beings filter information from the environment via heuristics and biases and consider it in highly selective ways that vary across social settings.While such dependence on heuristics might be rational in a very broad sense since strictly rational information gathering and calculation could leave us paralyzed in overanalysis(Rabin zooo;Mercer 2005),the point is that rationality per se tells us little about which heuristics or shortcuts to employ.3 Instead,we tend to pick up certain scripts,schemas,cognitive maps,frames,or analogies from locally available social constructions (or we hit on new ones in action). This basic psychological view of mental processes has become increasingly mixed with a logic of meaning since the 196os,as most cultural sociologists have replaced Weber's notion of meanings as socially constructed "values"with a more cognitive view of meanings as "taken-for-granted"social constructions (with people able to conceive of only certain legitimate meanings).Still,this literature clearly offers a distinctive logic of selective cognition to suggest that social construction matters,with particular mechanisms of both learning and filtering information coming to the fore.Social constructs shape action not so much because agents need meaning to inform action,but because they depend S 9.Psychologists have documented many heuristics or biases that seem to be common to all human beings to some degree,but the directions and degrees in which these universal biases are E activated is generally seen as dependent on social constructions. L 3050-544_01ntro.indd10 1125/20098:4609AM
10 Constructing the International Economy succeed: capital outfl ows sparked by an out-of-bounds policy can undermine a choice that, at another historical moment, may have been a perfectly plausible response to a policy challenge (Kirshner 2003b; Grabel 2003; Abdelal 2007). Michael Barnett and Martha Finnemore (2004), for example, have examined the role of international organizations in fi xing meanings, thereby constituting the legitimate boundaries of policymaking. Although such processes are rarely observed as evidence of the attempt to fi x meanings, political and economic organizations such as intergovernmental agencies and fi rms in fact regularly contest and thereby attempt to defi ne the boundaries of acceptable practices and legitimate policy through the socialization (and persuasion) of other actors in the system. Thus the decline and eventual fall of otherwise regular economic transactions—buying and selling persons, invading indebted countries, monetizing government debt, and Asian “crony capitalism,” for example—were precipitated by the purposeful efforts of ideational entrepreneurs to brand them illegitimate (Keck and Sikkink 1998; Finnemore 2003; Hall 2003). The Path of Cognition Another prominent line of theorizing that leads to constructivism takes a pathway through cognition. Many experimental psychologists and cultural sociologists have long argued that human beings depend on heuristics and shortcuts to organize action and choice (Kahneman, Slovik, and Tversky 1982; Swidler 1986). Their core observation is not that agents search for signifi cance and purpose, but that human beings fi lter information from the environment via heuristics and biases and consider it in highly selective ways that vary across social settings. While such dependence on heuristics might be rational in a very broad sense since strictly rational information gathering and calculation could leave us paralyzed in overanalysis (Rabin 2000; Mercer 2005), the point is that rationality per se tells us little about which heuristics or shortcuts to employ.9 Instead, we tend to pick up certain scripts, schemas, cognitive maps, frames, or analogies from locally available social constructions (or we hit on new ones in action). This basic psychological view of mental processes has become increasingly mixed with a logic of meaning since the 1960s, as most cultural sociologists have replaced Weber’s notion of meanings as socially constructed “values” with a more cognitive view of meanings as “taken-for-granted” social constructions (with people able to conceive of only certain legitimate meanings). Still, this literature clearly offers a distinctive logic of selective cognition to suggest that social construction matters, with particular mechanisms of both learning and fi ltering information coming to the fore. Social constructs shape action not so much because agents need meaning to inform action, but because they depend 9. Psychologists have documented many heuristics or biases that seem to be common to all human beings to some degree, but the directions and degrees in which these universal biases are activated is generally seen as dependent on social constructions. S_ E_ L_ 3050-544_01Intro.indd 10 11/25/2009 8:46:09 AM Uncorrected Proofs: Copyright Cornell University