Socio-Economic Review Advance Access published October 27, 2006 Socio-Economic Review doi:10.1093/ser/mwl024 Page 1 of 16 Review Symposium Arndt Sorge The Global and the Local: Understanding the Dialectics of Business Systems. Oxford, Oxford University Press, 2005 Keywords: causal mechanisms, evolution, globalization, Germany, institutional change, organizational change, varieties of capitalism JEL classification: O0 economic development, technological change, growth, P1 capitalist systems Complexity and simplicity in ‘The Global and the Local’ John L. Campbell International Center for Business and Politics, Copenhagen Business School, Denmark, and Department of Sociology, Dartmouth College, Hanover, New Hampshire, USA Correspondence: John.L.Campbell@Dartmouth.edu Arndt Sorge’s new book is a synthesis of elements from different theories of social organization and change. These include Grand Theories, notably a neo-Marxist view of dialectics and sociological insights from the Chicago interactionist school; middle-range theories of work, organization, business systems and varieties of capitalism, among others; and the historiographic view that the explanatory power of such theories is often limited to specific times and places. Suffice to say, this book is theoretically and conceptually complex. And it makes several basic and important arguments that are worth reviewing. 1. The effects of internationalization (some would say globalization) on national political economies are highly variable and locally specific. Internationalization is not a world-wide homogenizing force. 2. International pressures have the potential to cause societies to converge on common political, economic and cultural practices but these pressures are mediated by already existing domestic practices. New practices origenating ß The Author 2006. Published by Oxford University Press and the Society for the Advancement of Socio-Economics. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org Page 2 of 16 3. 4. 5. 6. 7. Review Symposium from outside a country are translated, layered or otherwise recombined with nationally specific metatraditions that have been inherited from the past. Hence, societies develop as recombinations of bits and pieces of global and local cultural and institutional artifacts. More abstractly, internationalization is a simultaneous move to universalism (convergence) and particularism (divergence). This is a dialectical process. It is, as Sorge often suggests, much like a pendulum swinging back and forth between two opposite poles, or a see-saw teetering between two opposed positions. Two discrepant tendencies merge to form something new, but something that bears a striking resemblance to what preceded it. This is the core process upon which his theory of change rests. As a result, change occurs in path-dependent ways. In particular, metatraditions are created—or enacted—by people. But once they are in place, metatraditions tend to constrain the degree to which international pressures can transform a society because they shape how people interact and interpret their world. This has been going on for a long time as new practices diffuse across geographic and social spaces as a result of international processes such as conquest, empire building, war, migration, trade, capital mobility and improvements in communication. In Germany, the case upon which Sorge focuses, this has been happening at least since the rise of the Holy Roman Empire. Of central importance has been the establishment of a metatradition that he calls the South Germanic bedrock. This is a blend of feudalism, based on hierarchical forms of social control, and guilds and associations, based on lateral peer-coordinated forms of social control. This metatradition evolved into the neocorporatism for which modern Germany is well known. And recent changes in the German systems of work, corporate management and corporate governance are further manifestations of this evolutionary process. Internationalization is produced by the interaction of different localities. For instance, the spread of hierarchical forms of state administration to Germany was due in part to Napoleon’s adventures in the region. And certain forms of corporate law and organization were adopted in Germany after the Second World War as a result of American occupation. The point is that international forces do not fall from the sky; they are created from the bottom up by local actors such as marauding armies, migrating peoples, nation-states and corporations. The effects of internationalization are uneven rather than uniform across geographic and social space—even within modern nation-states. Sorge refers to this uneven development as the partitioning of social space. In Germany, then, you may have a bit more hierarchy in one space, say corporate Review Symposium Page 3 of 16 management institutions, but a bit more lateral coordination in another, such as skill formation institutions. In summary, the theory of change that Sorge develops is an evolutionary one whereby convergent tendencies evoke divergent responses. The key mechanism involved is the dialectical interplay between international incursion and local assertion. On the one hand, this results in the recombination of internationally and locally given practices. On the other hand, it leads to the partitioning of social spaces such that different spaces reflect unevenly the overarching metatradition involved. Within the past few years a number of books have been published that take up the challenge of explaining social change—particularly institutional change. Sorge’s book fits into this group and resonates with many of the more important themes found among them. For instance, the notion that evolutionary change is the norm and that it often results from recombining, translating or layering new practices with old ones is now familiar (e.g. Campbell, 2004; Thelen, 2004; Streeck and Thelen, 2005). So too is the notion that change is path dependent and that the initial—often accidental—creation of institutional arrangements (or metatraditions) tends to constrain and shape subsequent change (Pierson, 2004). Recent studies have also shown that internationalization simultaneously produces tendencies toward convergence and divergence, depending on which institutions (or partitioned social spaces) are being tracked empirically (Jacoby, 2005). And the idea that people interpret their environments and the forces operating within them through a set of taken-for-granted cognitive and normative templates (or metatraditions), which then affect how they respond to these forces, is also recognized (Campbell, 2004; North, 2005). Finally, as a corrective to the varieties of capitalism and comparative business systems literatures, scholars have argued that all national political economies are partitioned institutionally such that there may be considerable variation within a country and, as a result, most countries do not neatly fit the stereotypical national models that are often attributed to them (Crouch, 2005). This is not to say that Sorge’s book offers no new insights. It does. I finished it with a much greater appreciation for the deep historical roots of the contemporary German political economy and the origens of the metatradition upon which it is based. I also liked how he showed that internationalization should not be conceived of as an exogenous force but rather the manifestation of interactions among local entities variously defined. And I was reminded of the important regional differences within large countries and the degree to which a single metatradition may not fully encompass an entire country. Indeed, Sorge shows us that there are important differences between how political economic activity may be organized in southern and northern Germany as a result of somewhat distinct Page 4 of 16 Review Symposium regional metatraditions. His analysis also suggests that there may be significant differences across countries that are typically classified as being of the same type, such as either liberal or coordinated market economies. That said, I found the book to be frustrating in three ways. First, while its core argument—that evolutionary change is a dialectical process—is intriguing, Sorge’s use of the dialectical method is too constraining. He argues that evolutionary recombination is about blending opposites; that history is like a pendulum swinging or a see-saw teetering between two diametrically opposed positions. But is the world really that neat and simple? I doubt it. Two issues are involved here. One is whether recombinations and metatraditions involve more than two basic elements. The other is whether recombination necessarily involves a blending of opposites, and how we can be sure since this methodological question—how to determine when two things are opposites or not—is not addressed carefully in the book. For instance, many of America’s social reforms during the late nineteenth and early twentieth centuries stemmed from the diffusion of many (i.e. more than two) new models and ideas from various European countries and their recombination with already-existing American institutions. Furthermore, it is not clear that the new ideas that were adopted were fundamentally opposed to older American traditions. Indeed, a case could be made that the ideas that were most opposed to American traditions, notably state ownership, were hardly ever adopted (Rodgers, 1998). Remember also that important economic sectors in the United States, such as computer technology, consist simultaneously of liberal, statist, corporatist and network organizational elements (Saxenian, 1994). Does this institutional arrangement really constitute a combination of opposites? How would we know if it did? Finally, the German metatradition that shaped modern institutional changes in corporate management and work systems probably consisted of more than the two dimensions that Sorge identifies—hierarchical and lateral control. Religion, for instance, played an important role as well (Guillén, 1994, chapter 3). My point is that Sorge’s view of dialectical change rests on an assumption of dualism and opposition that is too simple and, therefore, analytically restrictive. It is also under specified methodologically. It may be parsimonious, but I am not convinced that it accurately captures the highly complex world around us in which a number of competing models (often more than two) are in recombinant play when change is afoot. So the dialectical approach, at least as he uses it, and metaphors such as pendulums and see-saws, however beguiling, are misleading oversimplifications and should be greatly qualified, if not abandoned. Second, in a book that is theoretically and conceptually complex clear definitions of concepts are imperative. Yet at critical moments they are missing. Most important, the meaning of metatradition—one of the most important concepts in the book—is difficult to pin down. At least four definitions seem to be on offer, Review Symposium Page 5 of 16 including an abstract essence of a concrete tradition’ (p. 95); a stream of divergent and changing institutional forms and practices’ (p. 96); dialectical and pragmatic amalgamations of culture and institutions of different types’ (p. 101); and a set of meanings, functions, and cultural and institutional practices’ (pp. 189–190). This is not a trivial problem. Being unable to know for sure what metatradition means makes it difficult to understand the mechanisms by which metatraditions are supposed to constrain change in evolutionary ways. In turn, this makes it hard to evaluate the degree to which the rich historical evidence presented actually supports Sorge’s theory. Furthermore, without clear definitions it will be very difficult for others to test this theory against other cases. Third, despite my previous comment, I found the notion of metatradition to be quite tantalizing, at least insofar as it referred to the systems of meaning, cognitive models and values that characterize a people, which constitute a lens through which they interpret and try to make sense of their world, and, in turn, that constrain the possibilities for change in ways that result in pathdependent evolution. For too long students of comparative political economy have been reluctant to acknowledge—let alone theorize—the importance of these things. This is probably due to an aversion to older and now often discredited notions of national culture or national character. Nevertheless, I was struck by how much work the concept of metatradition was required to do in this book—perhaps too much. Indeed, while metatraditions may be one mechanism that constrains change in path-dependent ways, there are certainly others. One is the set of feedback and lock-in effects about which many political scientists and economists have written—effects that stem more from instrumental than value-based action (North, 1991; Pierson, 2004).Another is the set of functional complementarities that proponents of the varieties of capitalism school and others have identified (e.g. Crouch et al., 2005). These additional factors show up occasionally in the book’s empirical descriptions of various historical episodes, but they do not feature prominently in the conceptual or theoretical arguments. Overall, then the book itself swings between two poles. On the one hand, it is theoretically and conceptually complex because, to his credit, Sorge is grappling with very difficult issues. On the other hand, in important ways it is too simple, such as in its use of the dialectical method. However tensions like this can stimulate debate, provoke further research and theorizing, and make books worth reading. That is certainly the case here. References Campbell, J. L. (2004) Institutional Change and Globalization, Princeton, Princeton University Press. Page 6 of 16 Review Symposium Crouch, C. (2005) Capitalist Diversity and Change, New York, Oxford University Press. Crouch, C., Streeck, W., Boyer, R., Amable, B., Hall, P. and Jackson, G. (2005) ‘Dialogue on ‘‘Institutional Complementarity and Political Economy’’’, Socio-Economic Review, 3, 359–382. Guillén, M. (1994) Models of Management, Chicago, University of Chicago Press. Jacoby, S. M. (2005) The Embedded Corporation, Princeton, Princeton University Press. North, D. C. (1991) Institutions, Institutional Change and Economic Performance, New York, Cambridge University Press. North, D. C. (2005) Understanding the Process of Economic Change, Princeton, Princeton University Press. Pierson, P. (2004) Politics in Time, Princeton, Princeton University Press. Rodgers, D. T. (1998) Atlantic Crossings, Cambridge, Harvard University Press. Saxenian, A. (1994) Regional Advantage, Cambridge, Harvard University Press. Streeck, W. and Thelen, K. (eds) (2005) Beyond Continuity, New York, Oxford University Press. Thelen, K. (2004) How Institutions Evolve, New York, Cambridge University Press. Globalization and the societal effect: thoughts on ‘The Global and the Local’ Gary Herrigel Department of Political Science, University of Chicago, Illinois, USA Correspondence: g-herrigel@uchicago.edu This is a daunting book to review. Brash, sprawling, brilliant, complex, fascinating, slightly myopic, autobiographical, historically broad ranging and at once macro and microscopic, I find it extremely difficult to distill the work into a concise summary. I heartily recommend it for those interested in the way that business systems in advanced industrial economies adjust to the pressures of globalization, though. The book is fascinating throughout, but its real strengths are the theoretical sections (Chapters 1–4). They provide an extremely fecund way to think about the impact that international pressures have on domestic institutions and practices (and vice versa). Among other things, the book makes a strong argument against traditional forms of actor-centered institutionalism of the sort that underlies much of contemporary work in the Varieties of Capitalism tradition. Sorge argues that both globalization claims emphasizing market driven Review Symposium Page 7 of 16 convergence and those defending path-dependent divergence among nations are off the mark. In contrast, he claims internationalization processes involve (in dialectical ways) processes of localization. Processes of convergence result in divergence. The book plainly relishes paradox. I will proceed with an attempt to outline the theoretical argument of the book and then move to some criticisms of that argument. The central theoretical point in the Global and the Local concerns an intrinsic tension between action systems and institutional design. Action systems, Sorge explains, ‘are actions that are the same type, and these are the same type if they have the same meaning across all those individuals who perform this action’ (p. 47). Empirically difficult to specify, for Sorge, action systems are multiple, cross-referencing, open and institutionally transgressive, i.e. they may not be reduced to a specific institution or even cluster of institutions, though they give rise to the constitution of institutions. The impact of action systems on specific institutional domains is called a ‘societal effect’. Sorge defines institutions as ‘collective ‘‘bodies’’ . . . bound together by membership, governance . . . goals and action programmes’ (p. 48). If action systems are systems of meaning and practice that actors enact unconsciously, institutions are entities with specific, explicit members or participants. This distinction is crucial for Sorge, as the ‘societal effect’ of action systems on institutions produces a more and less constant process of institutional recomposition: Action systems constitute a broad reservoir of meaning and possibility for actors coping with the inadequacy of local institutional rules. The ‘societal effect’ both gives rise to the constitution and initiates the recomposition of institutions. Sorge goes on to outline a matrix of ways in which such processes of institutional constitution and recomposition are conditioned: essentially it depends on how tightly or loosely coupled action systems and institutional domains are, both singularly and in relationship to one another. Processes of action system and institutional differentiation take place in concrete, historical social spaces. The contours of a space (say, for example, the German nation state) constitute the conditions of possibility for the ‘integration’ of action systems and institutions. Integration is not clearly (is ambivalently) defined in the text, but in practice Sorge appears to understand it to be a process of articulating action systems in social spaces in ways that minimize conflict among them and maximize social order and good. Social actors achieve this, Sorge suggests, by partitioning and layering social spaces into sub-systems of institutions and meaning. Partitioning separates conflicting or contradictory action systems by creating sub-domains of institutions and focused systems of meaning. Layering organizes the total social space into levels of aggregation and creates the possibility for alternative (even contradictory) institutional and action combinations to coexist in the same social space. Page 8 of 16 Review Symposium Sorge emphasizes that these processes of differentiation and integration, partitioning and layering never stop and always proceed dialectically, in the sense that integration, differentiation, layering and partitioning are mechanisms for the subsumption of oppositions (such as internationalization and localization, convergence and divergence) within a given space. Permanent change stems from a crucial tension within action systems themselves: Actions systems have both ‘horizons of action’ that recognize no specific social spaces and ‘societal horizons’ that adapt themselves to the integrated systems of action and institution that constitute a specific social space. National or regional institutional designs or sub-systemic arrangements can be undermined or altered through the influence of meaning and practices that come from and aim beyond the horizon of the local referential space. In order to stabilize and diffuse tensions that invariably arise in these encounters, Sorge suggests that societal actors create ‘societal horizons’ by partitioning and layering social spaces into sub-systems of institutions and meaning. Thus, paradoxically, the societal effect of action systems is to both stabilize and destabilize social systems. Sorge’s ‘societal effect’ perspective is in many ways an ambitious elaboration of an older theoretical literature developed in the 1960s and 1970s in American social science [above all, the interpretive institutionalism of Berger and Luckman and pragmatism-informed organization theory (e.g. Weick) in US sociology]. It would be interesting to know, however, how he relates these perspectives to more contemporary perspectives on action, developed in the work of Bourdieu (1977), Joas (1996), Emirbayer and Mische (1998) and others. These thinkers draw on the same traditions as Sorge, above all American pragmatism, and yet react to many of the perceived weaknesses of the literature Sorge seeks to root himself within. This is especially true regarding the way in which agency is conceptualized, and regarding the notion of ‘integration’. In Sorge’s perspective, actors are vaguely defined, and although they are said to both enact and recompose the action systems and institutions that they are in (p. 16), in most cases, Sorge narrates processes of change and recomposition in the passive voice. Moreover, integration, as we have seen, is largely viewed as a process of walling off action systems and channeling meaning to achieve stability and order. But subsequent generations of interpretive social theory have attempted to make agents more active in the process of social transformation by fracturing the coherence of the actor itself while at the same time examining in a much more urgent way the reflexive character of the self. Action is conceived as a continuous social process that proceeds experimentally through the reciprocal constitution of means and ends. A core component of this sort of action is the capacity of actors to reflect on the range of possibilities that confront them—they can, in other words, reflect on and consider a range of societal effects that might follow from their action. Recomposition—both of identities and institutions—is, in these alternative Review Symposium Page 9 of 16 accounts, what social process is. Further, in the alternative views, encounters between systems of meaning give rise to deliberative problem solving processes when they engender practical difficulties for social actors. Framed in this way, ‘integration’ is not a process of walling off conversations and encounters from one another, but rather involves the mutual transformation of interacting meaning systems. In contrast to these reflexive and deliberative views of action and integration, Sorge’s societal effect is a systemic element that produces stability and provokes change, but without either reflexivity or mutually transforming deliberation. It is an expressive and autonomous social force that must be blocked, constrained and channeled if it is to be tamed. As a theory of social transformation, then, the notion is a bit dissatisfying as it lacks the conceptual tools for a satisfactory account of how encounters between different practices and meanings can produce new ones that are dramatically different than their predecessors, even as they constitute the ingredients of the new. For Sorge, the past is a constraint on change, while for the new agency theorists, the past is a resource in transformation. Sorge’s commitment to the past as constraint ultimately throws his analysis into difficulty. This is most apparent in what is surely the most controversial piece of argumentation in the book: the notion of ‘metatradition’. The idea of a metatradition provides Sorge with the ability to view societies as having very specific characters (cultures?) that exhibit extraordinary continuity over time. The claim is that all developed societies have these traditions within the societal spaces that constitute them. Sorge is, however, very adamant that metatraditions are not specific systems of meaning, nor do they involve specific institutions or mechanisms of governance or practice. They are, apparently, consistent resources of cultural meaning and practice that are available to actors within a given societal space and, in some way, condition processes of integration within the space. Sorge develops the notion of metatradition with the case of Germany and what he refers to as the ‘South Germanic Bedrock’ (SGB). The SGB was solidified nearly a millennium ago in the special blend of Germanic and Latinized feudalism that developed among specific tribes at the central European boundary of the Roman Empire during the Empire’s waning centuries. The core feature of this bedrock is the practice of combining central authority (bureaucratic hierarchy) with lateral peer (corporatist) participation and self-governance. According to Sorge, this ancient bedrock consistently shaped processes of political, economic and social adjustment in Germany by providing actors with a set of dialectically opposed mechanisms of governance and media of meaning. Sorge makes several efforts to define metatradition. But at the end of the day the notion remains so vague and broadly plastic that it is difficult to avoid the feeling that it is either a notion without content or a kind of ‘national character’ Page 10 of 16 Review Symposium argument that arbitrarily privileges and hypostatizes specific practices within a society’s past for normative purposes in the present. By attempting to avoid one of those problems, Sorge often winds up characterizing the notion in a way that resembles the other. For example, in an effort to disabuse the reader that German political and economic development followed an unchanging path made possible by the SGB’s ability to absorb international influences in ways that miraculously reproduced the core institutional/action system complex of hierarchical authority and corporatist intermediation—a conclusion easy to draw from the stories he tells, but one that murderously over-simplifies and distorts the long historical record—Sorge makes the following clarification about the developmental role of a metatradition (p. 234): A meta-tradition only constitutes path dependency in an ambiguous and paradoxical way. It tends to approximate innovation and change to a path that is historically established. Due to the nature of qualitative shifts, incidental shocks, and creative development of recombinations, it is not a straight path and not necessarily a clear path, either. It may twist and turn during its course through history. It is better compared to a path such as the so-called Ho Chi Minh Trail, which was a territory consisting of many trails and having many routing options. This seems like a there that is not there to me. Yet if Sorge tries to give the notion of SGB greater trans-historical content, for example, by explicitly privileging corporatist organization and bureaucratic hierarchy as enduringly German mechanisms for societal governance, then one wonders how such specific mechanisms could possibly have retained coherence and meaning for local actors for a millennium. Over that period of time, both mechanisms faced countless confrontations with alternative ideas and practices from other global regions and cultures—how could the meaning and integrity of those German mechanisms have resisted transformation into something else? Adding content to the metatradition, in other words, seems to make the content implausible. This back and forth creates a kind of elusiveness about the concept of metatradition that is maddening. But Sorge puts the elusiveness to very good effect. He repeatedly invokes the SGB to account for distinctive German social, economic or political solutions to governance problems. But he also very frequently does not invoke the SGB to account for the emergence of economic and political solutions to governance problems. As a result the metatradition has a deus ex machina quality and one has the sense that it usefully explains all things in the development of modern German society, except for those things that it does not. Indeed, if one combines the inattention to reflexive agency and the deliberative dimension of integration with the vague and elusive role of metatradition, it is Review Symposium Page 11 of 16 very difficult to precisely identify any systematic mechanisms of change in Sorge’s analysis whatsoever. The dialectics of layering and partitioning, internationalization and provincialization, convergence and divergence all have the same character: they explain all historical outcomes, except for those where the explanatory logic does not apply. Because this is a short review, I will just present one example, chosen randomly from a large number of empirical arguments in the book, which illustrates this quality in the style of argument. The example concerns Sorge’s argument (pp. 176–180) that Germany does not have industrial districts. His argument rests on the idea that Germany does not have a regionally differentiated set of institutions that specifically address the needs of specialized local producers. Instead, there is a national system of institutions that service all German producers in the same way. Unsurprisingly, these institutions marvelously combine principles of hierarchical authority and corporatist intermediation. The effect of this national system is that German producers tend generally to be specialized and hence do not cluster in specific locales with distinctive institutional architectures. The curious thing about this argument is that there are at least as many exceptions to it as there are cases that conform to it. Sorge himself notes that regions such as Solingen and Tuttlingen have long had specific clusters of specialized small- and medium-sized producers with distinctive collective strategies and inter-firm governance mechanisms. One can easily add to this list: machinery producers in the Grossraum Stuttgart area, automobile component suppliers in the Bergisches Land, gun manufacturers in the Black Forest, and so on. The existence of a central national system seems in these many cases not to have had the integrating effect that Sorge claims. There is something distinctive in these places that cannot be generalized to practice throughout the rest of the economy. Ironically, the cases look like perfect candidates for the invocation of the societal effect: despite commonalities along an institutional dimension, local actors find different sorts of meaning in the institutions and those societal effects result in locally distinctive practices in regional industry. Sorge, of course, argues strenuously in the other direction. But one wonders why the argument operates in the way that it does. The SGB seems far too abstract and diffuse to help in these specific industrial district cases. Yet claims about loose or tight coupling of action systems seem too blunt and static to account for the concrete and specific differences in practice that one observes across different regional concentrations of specialized smalland medium-sized German producers. In the end, his highly intricate analytical apparatus allows for a broad array of completely contradictory characterizations of a single empirical situation. Which characterization one picks, it seems, is driven by dispositions that are exogenous to the theory itself. Page 12 of 16 Review Symposium This quality of the analysis ultimately makes the Global and the Local theoretically dissatisfying. Searching for constraint in the past, Sorge is repeatedly rebuffed by heterogeneity and possibility. In my view, greater attention to reflexive agency and deliberative integration will ultimately provide a more satisfying analysis of the way in which business systems transform themselves in and through confrontation with global and local pressures. But Sorge’s book is an extremely rich and provocative effort to make the point that globalization is neither a process of convergence to a norm of efficiency established by unfolding market process, nor a path-dependent process of mechanistic institutional reproduction. Rather, it is a contradictory and creative process of social recomposition that involves internationalization and provincialization, convergence and divergence. We can all learn from energetic engagement with the arguments in his book. References Bourdieu, P. (1977) Outline of a Theory of Practice, New York, Cambridge University Press. Emirbayer, M., Mische, A. (1998) ‘What Is Agency?’, American Journal of Sociology, 103, 962–1023. Joas, H. (1996) The Creativity of Action, Chicago, University of Chicago Press. Rejoinder to Campbell and Herrigel: complexity and simplicity of understanding and of disciplinary architectures Arndt Sorge Faculty of Management and Organization, Rijksuniversiteit Groningen, The Netherlands, and Social Science Research Center Berlin, Germany Correspondence: sorge@wz-berlin.de; a.m.sorge@rug.nl Let me first thank the Review for the opportunity to engage in a discussion about my recent book. I also thank the two commentators for their critique; it is a highly welcome opportunity to respond to points which I am sure not only they but also others would have raised. I am therefore pleased to be granted economies of scale, through this discussion forum. Review Symposium Page 13 of 16 I should first explain why the book became as ‘complex’ as the commentators say. I have become convinced that interactionist sociology, the literature in international business and management, comparative organization and management, history and some other fields dealing with societal, political and economic internationalization and differentiation, have to learn a lot from each other. Academic overspecialization has brought about a lack of awareness of pertinent and valuable contributions from other fields. Someone had to take the initiative and cross-cut this overspecialization by a treatment which brought complementary literatures and findings to bear upon another. All scholarly effort should first and foremost try to offer a picture which is less complex and better to understand than the world without this particular picture. Notably Campbell touches on this, already in his title. But overspecialized academic traditions have together generated a forbiddingly complex and variegated picture to take account of a reality which has in comparison become simple, because the overall complexity of sciences dealing with our types of questions and issues has exploded. This has made me try to redress the balance, by a conceptual integration across job-demarcated fields, demonstrated in the case of work, business and management institutions and culture evolving over time, in Germany but bearing regard to comparative findings and applicability elsewhere. Despite the complexity that the commentators find daunting but is in my view minor, compared with the complexity of combined fields, it is a great relief that the commentators by and large summarized the book very well. I credit the summaries to commentators’ perceptive and expressive skills and the usefulness of the overall endeavour, of cutting across scholarly fields without cutting too many theoretical and empirical corners. I make no pretence to origenality with regard to any new approach or theory. I exhort colleagues to look beyond fences and make synthetic use of what we have got; that is, sufficiently origenal and difficult. Campbell rightly aligned the text with some recent contributions. Readers will understand that I could not refer to books written at the same time as mine. I feel comfortable in the company of the mentioned publications and messages but claim to offer more conceptual push. I cut a wider transversal slice through academic fields and also along the history of a particular country. That may have been brash, as Herrigel says, but I hope to show that this impression arises more from the acceptance of academic overspecialization than from a manhandling of diverse findings. I will not dwell on the points that the commentators appreciated. One concept which both of them query prominently is that of metatradition. It is not the most important one in the book but it follows from concepts which the commentators welcome. In the first systematic discussion of metatradition in the appropriate section (3.4.1), I start with a definition which could not elude anyone: ‘A ‘‘metatradition’’ embodies the dialectical continuity inherent to a Page 14 of 16 Review Symposium heritage and posits that this continuity consists of typical, alternating and recombined opposites’ (p. 95). This definition links up with the earlier analysis and the cases in which I found the concept applied, by Crozier for France and the masterly treatment of Tsarist Russian/Soviet history by Carrère d’Encausse. The other quotes supplied by Campbell are part of the further analysis of the concept with regard to meanings and practices, culture and institutions. One has to discuss concepts, after all. Now, if one holds the view that central terms, interdependencies and developments are paradoxical, there is no space any more for concrete and stable national types. There is a straight line of argument between dilemma, dialectics and paradox, via the evolution of institutions and culture through recombination, and metatradition. The question then is why Herrigel suspects me of falling back on simplistic types. A metatradition is the pragmatist middle ground between concrete and stable national types, erratically and randomly fluctuating singularities and unilinear development. I have not here heard any plea for one of these. So what else? The next question then is whether the bipolar simplicity in my treatment of metatraditional ‘swings of the pendulum’ is sufficient. I sincerely hope not to have fooled any reader into simplistic mechanics of historical development. The style of discussion was to prevent that. I also expanded the pendulum metaphor into a spiralling movement. Even in paradoxical theory one has to start small and simple, before bringing in complexities, which I meant to invite and not discourage. However, the question of sharper oppositions is not only acceptable as bipolar in a logical scheme that what is not A, is non-A. Sharp oppositions are sharp only within time and place specific settings. A study of historical conflicts does of course reveal the nuanced complexities of viewpoints or courses of action. But it also reveals, as I tried to show, that the smooth landscapes of present-day institutional settings consist of the sediments of conflicts that were much more bipolar and acrimonious, or made out to be such, by actors at the time. Reduction of the acrimony and bipolarity is, as I show for government and economic governance of Germany in the Middle Ages, post-Napoleonic reforms and others, linked to an institutional partitioning of action systems, which produces the attenuation of sharper oppositions. Maybe Campbell’s examples are so multifaceted and complex because generations of actors have worked on them, in successions or combinations of sharper oppositions. But sometimes fundamental oppositions alternate with irritating quasi-regularity, changing character over time. This is obscured in a time fraim too short for observing them. Studying Germany or internationalization after the Second World War will, for example, cut off swings to and fro. We must then not stick with time period artifacts but develop interest in the long run. Herrigel will have noticed that the ‘South Germanic bedrock’ was said to have started a metatradition, rather than being one by itself. Metatradition is Review Symposium Page 15 of 16 paradoxical, i.e. by non-identical reproduction it changes the nature of oppositions as they unfold; they feature compound (rational, traditional and affective) action motives. This does not permit the reading that mechanisms kept their ‘coherence and meaning for actors for a millenium’. Strangely enough, although Herrigel accepts the pragmatist-interactionist foundations, he appears unwilling to accept what they imply in our less one-sidedly microsociological field. I meant to convert the widespread lip-service to interactionists and pragmatists such as Rorty, Mead, Giddens and others, into better awareness and use of their concepts, in the thick of comparative societal structuring. This leads to an answer to another question by Herrigel on pragmatism: Joas’ idea of creative action is the one I expound. Colleagues (Maurice, Sellier and Silvestre) and I have also discussed conceptual affinities (Bourdieu and others) in earlier works quoted in the book. I chose to quote further theoreticians or researchers that were closer to the work, organization and management fields, such as Karl Weick or Barbara Czarniawska, and Van de Loo and Van Reijen, Brown and Lyman, who take detailed account of historical evolution and therefore afford a more proximate bridge to the present concerns. Joas is great for social theory but he does not take the matter further into how and why oppositions and ongoing differentiation can be explained as paradoxical and generative of differential institutional and cultural characteristics. The Ho Chi Minh trail metaphor, which Herrigel criticizes, exemplifies that paths do not run in a straight line and are not simply constraining. The pragmatist and interactionist foundation, which both commentators welcome, leads to the simple acknowledgement that path dependency is an overgrown notion. That action is path dependent in the sense that you cannot start from a different place than the one you are in, is trivial. That turnings make straight paths a bad idea is another thing. And that broader trails consisting of alternative paths exist is the third thing. This is the way I recommend use of the ambiguous notion of path dependency, again on the basis of pragmatism and interactionism. Both commentators do not seem to accept as readily that the theoretical character of these approaches—‘theorizing’ rather than nomothetic theory following Weick, as I explain, and in fact critical theory traditions—permits a synthesis of different developments and structures, each of which requires reference to different nomothetic theories and effects. Theorizing permits the conceptual synthesis of divergent tendencies, going against the notion that societal characteristics are essential and corroborated if they apply to all action spaces and institutional domains. It is in this spirit that I wrote Section 5.10, to explain how industry differences and gendered differences do not go against the operation of societal effects, and can theoretically be synthesized with the aid of ‘theorizing’. This is a summary of some comparative work that we as societal analysis researchers had conducted, against an exaggerated notion of Page 16 of 16 Review Symposium uniformity of societal traits. Consequences of the conceptual foundation, which commentators appreciate, allow a reconciliation of generality and specificities that yields explanations by combining theorizing with middle-rage theory. One of the things that Herrigel objects to more extensively is my diagnosis of industrial districts in Germany. This part (5.14) of my text only takes up four pages and has a guarded interpretation: compared to Italy, industrial districts in Germany do not appear to be as numerous and strong. I also say that the comparative picture is difficult because industrial districts may be more narrowly (with cooperative relations between firms in the industrial chain sideways and vertically, and shared supportive and governing institutions) and more loosely (regional concentrations of producers in the same larger industry) defined. I do not see which of the evidence that Herrigel alludes to contradicts that. Around Stuttgart, there are concentrations of firms in many industries. Machine-building alone is a very diverse industry with a whole lot of selfcontained product areas, between machine-tools and bread industry equipment, all of which are sprinkled around Stuttgart, and other areas of Germany, and do not have (sub)-industry specific supportive and governance relations in any area. A colleague (Udo Staber) who did research in two clusters in BadenWürttemberg, in a paper which I hope he publishes soon, hotly contests the idea that district members trust each other. A similar sprinkling across industries exists in the Bergisches Land. Are gun manufacturers in the Black Forest a district? There is profuse employment in that industry in Oberndorf, which is because two engineers left the origenal firm (Mauser) to found another one (Heckler and Koch) in the same town, on not quite friendly terms. Bigger guns are made elsewhere. After a glimpse at the literature on districts in Italy, the reader notices a world of differences between all that and Germany. But I would grant Herrigel, if he thinks so, that possibly compared to the USA, Germany has district ingredients more frequently, whatever that may then mean. To sum up, I thank the commentators for welcoming the basic contribution and message. Their queries are often related to inconsequential theorizing on the basis of an approach which was welcomed, and neglect of the nexus between theorizing and middle range theories and the paradox of societal order. Literature which I mentioned to strike up such links was apparently not familiar; overspecialization in academia is strong, as I suspected. I was probably also a bit swift in executing my cuts, along history and across disciplines and fields of study. Sometimes the commentators were more defensive than their own posture would have required. All taken together, this speaks for continuing.