Transforming Economics: Perspectives on the Critical Realist Project (Economics As Social Theory)

Transforming Economics

Economics has become polarised. On the one hand, orthodox economists attempt to advance their discipline via an increasing use of mathematical modeling. On the other hand, heterodox economists argue that if economics is to progress and to be useful, then, rather than relying on the unthinking application of mathematical modeling, economists must tailor their analytical tools more closely to the nature of their subject-matter than has hitherto been the case. The contributors to this book fix their scholarly gaze on the heterodox section of economics, and in particular upon critical realist approaches to the subject. Experts from a variety of perspectives have come together in these pages to examine the impact and usefulness of critical realism in relation to the different spheres within economics. Notable for its contributions from such distinguished figures as the latest winner of the Nobel Prize for Economics, Clive Granger, Edward J. Nell and Peter J. Boettke, this book deserves to find a ready audience across the economics spectrum. Paul Lewis is a Newton Trust Lecturer in the Faculty of Economics and Politics and the Faculty of Social and Political Sciences, Cambridge University, and a Fellow of Selwyn College.

Economics as Social Theory Series edited by Tony Lawson University of Cambridge

Social theory is experiencing something of a revival within economics. Critical analyses of the particular nature of the subject matter of social studies and of the types of method, categories and modes of explanation that can legitimately be endorsed for the scientific study of social objects, are re-emerging. Economists are again addressing such issues as the relationship between agency and structure, between the economy and the rest of society, and between the enquirer and the object of enquiry. There is a renewed interest in elaborating basic categories such as causation, competition, culture, discrimination, evolution, money, need, order, organization, power, probability, process, rationality, technology, time, truth, uncertainty, value, etc. The objective for this series is to facilitate this revival further. In contemporary economics the label ‘theory’ has been appropriated by a group that confines itself to largely asocial, ahistorical, mathematical ‘modeling’. Economics as Social Theory thus reclaims the ‘Theory’ label, offering a platform for alternative rigorous, but broader and more critical conceptions of theorizing. Other titles in this series include: Economics and Language Edited by Willie Henderson Rationality, Institutions and Economic Methodology Edited by Uskali Mäki, Bo Gustafsson and Christian Knudsen New Directions in Economic Methodology Edited by Roger Backhouse Who Pays for the Kids? Nancy Folbre Rules and Choice in Economics Viktor Vanberg

Beyond Rhetoric and Realism in Economics Thomas A. Boylan and Paschal F. O’Gorman Feminism, Objectivity and Economics Julie A. Nelson Economic Evolution Jack J. Vromen Economics and Reality Tony Lawson The Market John O’Neill Economics and Utopia Geoff Hodgson Critical Realism in Economics Edited by Steve Fleetwood The New Economic Criticism Edited by Martha Woodmansee and Mark Osteeen What do Economists Know? Edited by Robert F. Garnett, Jr. Postmodernism, Economics and Knowledge Edited by Stephen Cullenberg, Jack Amariglio and David F. Ruccio The Values of Economics An Aristotelian perspective Irene van Staveren How Economics Forgot History The problem of historical specificity in social science Geoffrey M. Hodgson Intersubjectivity in Economics Agents and structures Edward Fullbrook The World of Consumption, 2nd edition The material and cultural revisited Ben Fine

Reorienting Economics Tony Lawson Toward a Feminist Philosophy of Economics Edited by Drucilla K. Barker and Edith Kuiper The Crisis in Economics Edited by Edward Fullbrook The Philosophy of Keynes’ Economics Probability, uncertainty and convention Edited by Jochen Runde and Sohei Mizuhara Postcolonialism Meets Economics Edited by Eiman O. Zein-Elabdin and S. Charusheela The Evolution of Institutional Economics Agency, structure and Darwinism in American institutionalism Geoffrey M. Hodgson Transforming Economics Perspectives on the critical realist project Edited by Paul Lewis

Transforming Economics Perspectives on the critical realist project Edited by Paul Lewis

First published 2004 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2004 Selection and editorial matter Paul Lewis; individual chapters the contributors All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested ISBN 0-203-41726-7 Master e-book ISBN

ISBN 0-203-68246-7 (Adobe eReader Format) ISBN 0–415–36966–5 (hbk) ISBN 0–415–36967–3 (pbk)

Contents

Contributors Acknowledgements 1

Transforming economics? On heterodox economics and the ontological turn in economic methodology

ix xiii

1

PAUL LEWIS

2

3

Transforming Post Keynsian economics: critical realism and the Post Keynesian project STEPHEN P. DUNN

33

Macroeconomic theory, (critical) realism and capitalism

55

JOHN SMITHIN

4

Critical Realism and Transformational Growth EDWARD J. NELL

5

Critical realism and econometrics: an econometrician’s viewpoint

76

96

CLIVE GRANGER

6

Critical realism and feminist economics: how well do they get along?

107

EDITH KUIPER

7

8

The agency–structure model and the embedded individual in heterodox economics JOHN B. DAVIS

132

Critical realism and the heterodox tradition in economics

152

SHAUN HARGREAVES HEAP

viii

Contents

9 Economics as Social Theory and the New Economic Sociology

167

PAUL LEWIS

10 The really real in economics SCOTT A. BEAULIER AND PETER J. BOETTKE

187

11 Addressing the critical and the real in critical realism

202

BEN FINE

12 Economics as symptom JEAN-PIERRE DUPUY 13 The Economics of Institutions and the Institutions of Economics

227

252

UGO PAGANO

14 A note on critical realism, scientific exegesis and Schumpeter

268

MÁRIO DA GRAÇA MOURA

15 Transforming methodology: critical realism and recent economic methodology D. WADE HANDS Index

286

302

Contributors

Scott A. Beaulier is a fourth year PhD candidate in George Mason University’s Economics Department. He is a research fellow at both the James Buchanan Center for Political Economy and the Mercatus Center’s Global Prosperity Initiative. His research is primarily concerned with the emergence of property rights systems and their relationship to economic development. Beaulier has done extensive field work in the Czech Republic, and he has published in several academic outlets. Before attending George Mason University, he completed his undergraduate work at Northern Michigan University, where he majored in Economics and History. Peter J. Boettke is the Deputy Director of the James M. Buchanan Center for Political Economy, a Senior Research Fellow at the Mercatus Center and a Professor of Economics at George Mason University. Before joining the faculty at George Mason University in 1998, he held faculty positions at Oakland University, Manhattan College and New York University. In addition, Boettke was a National Fellow at the Hoover Institution for War, Revolution and Peace at Stanford University during the academic year 1992-3. Boettke is the author of several books on the Soviet Union, including The Political Economy of Soviet Socialism: The Formative Years, 1918-1928 (Kluwer 1990); Why Perestroika Failed: The Economics and Politics of Socialism Transformation (Routledge 1993); and Calculation and Coordination: Essays on Socialism and Transitional Political Economy (Routledge 2001). Boettke is also the co-author (with David Prychitko) of a revised edition of Paul Heyne’s The Economic Way of Thinking (tenth edition, Prentice Hall 2002). John B. Davis holds a PhD in Philosophy from the University of Illinois (1993) and a PhD in Economics from Michigan State University (1995). He is Professor of Economics and Econometrics at the University of Amsterdam, and Professor of Economics at Marquette University, Milwaukee, USA. He is the author of Keynes’s Philosophical Development (Cambridge University Press 1994) and co-editor of The Handbook of

x Contributors Economic Methodology (Elgar 1988). His latest book is The Theory of the Individual in Economics: Identity and Value (Routledge 2003). Stephen P. Dunn, a 2003–4 Commonwealth Fund Harkness Fellow in Health Care Policy, is a senior policy advisor in the Department of Health in England. In 2003-4 he was a visiting scholar at the Graduate School of Business at Stanford University and a Health Policy Researcher at the University of California, San Francisco. He has received the K. William Kapp Prize from the European Association for Evolutionary Political Economy. He has published numerous articles on the methodology of economics, the economics of J.K. Galbraith, Keynesian economics and the theory of the firm. Jean-Pierre Dupuy is Professor of Social and Political Philosophy, École Polytechnique, Paris, and Director of Research at the CNRS (Philosophy). He was formerly the Director of CREA (Centre de Recherche en Épistémologie Appliquée), the philosophical research group of the École Polytechnique, which he founded in 1982. He is also a Professor in French and Political Science at Stanford University. His research focuses on paradoxes of rationality, or the classical philosophical problem of the antinomies of reason in the age of rational choice theory, analytic philosophy and cognitive science. Recent books include The Mechanization of the Mind – On the Origins of Cognitive Science (Princeton University Press 2000) and Self-Deception and Paradoxes of Rationality (CSLI Publications, Stanford University 1998). Ben Fine is Professor of Economics at the School of Oriental and African Studies, University of London. Recent books include Social Capital versus Social Theory: Political Economy and Social Science at the Turn of the Millennium (Routledge 2001); Development Policy in the Twenty-First Century: Beyond the Post-Washington Consensus (edited with C. Lapavitsas and J. Pincus, Routledge 2001); and The World of Consumption: The Material and Cultural Revisited (Routledge 2002). He is currently preparing a new edition of Marx’s Capital (with A. Saad-Filho) for Pluto Press and, as part of broader research on economics imperialism, a book on shifting relations between history and economics (with D. Milonakis). Mário da Graça Moura completed his undergraduate degree in Economics (Licenciatura) at Faculdade de Economia, Universidade do Porto, in 1989. He holds an MPhil in Economics (1993) and a PhD (1998) from the University of Cambridge. He is currently Lecturer in Economics at Faculdade de Economia, Universidade do Porto. He has published articles on Schumpeter and methodology in the Cambridge Journal of Economics and The European Journal of the History of Economic Thought.

Contributors xi Clive Granger is Emeritus Professor of Economics at the University of California, San Diego. His research centres on time series econometrics, where he has made fundamental contributions to the analysis of nonstationary data and of causal relationships. He was awarded, jointly with Robert Engle, the 2003 Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel. D. Wade Hands is Professor of Economics at the University of Puget Sound in Tacoma, Washington. He has written on a number of topics in the history and philosophy of economics. He is the author of Reflection without Rules: Economic Methodology and Contemporary Science Theory (Cambridge University Press 2001) and is one of the editors, along with John Davis and Uskali Mäki, of The Handbook of Economic Methodology (Edward Elgar 1998). His most recent book is Introductory Mathematical Economics (second edition, Oxford University Press 2004). Shaun Hargreaves Heap teaches economics at the University of East Anglia. His current research is on individual rationality in an historical and social setting and on the economics of film and television. His publications include Rationality in Economics (Basil Blackwell 1989), The Theory of Choice: A Critical Introduction (with others, Basil Blackwell 1992) and Game Theory: A Critical Text (with Y. Varoufakis, Routledge 2004). Edith Kuiper is a researcher at the Department of Economics and Econometrics at the Universiteit van Amsterdam. Her research interests are in the history and philosophy of economics. She is coeditor of Out of the Margin: Feminist Perspectives on Economics (Routledge 1995), author of The Most Valuable of All Capital: A Gender Reading of Economic Texts (2001), and co-editor (with Drucilla Barker) of Towards a Feminist Philosophy of Economics: An Anthology (Routledge 2003). Paul Lewis was educated at Peterhouse, Cambridge, and Christ Church, Oxford. He was a Research Fellow of Emmanuel College, Cambridge, before becoming a Newton Trust Lecturer in the Faculty of Economics and Politics and the Faculty of Social and Political Sciences, Cambridge University, and a Fellow of Selwyn College. His research interests include: the Austrian school of economics; economic sociology; the philosophy of the social sciences (especially economics, sociology and politics); the methodology of economics; postwar British politics; and applied microeconomics (especially the economics of quasi-markets and performance-related pay). Edward J. Nell is Malcolm B. Smith Professor of Economics in the Graduate Faculty of Political and Social Science of the New School University, New York. Educated at the universities of Princeton and

xii Contributors Oxford, his research concentrates on macroeconomic theory and policy, economic methodology, growth theory, business cycles, inflation and unemployment. He is the author of numerous articles and books, including Rational Economic Man (with Martin Hollis, 1974), Transformational Growth and Effective Demand (1992) and The General Theory of Transformational Growth (1998). Ugo Pagano is Professor of Economic Policy and Director of the PhD programme in Economics at the University of Siena and Recurrent Visiting Professor at CEU, Budapest. He received his PhD at the University of Cambridge in 1983, where he was a University Lecturer and Fellow of Pembroke College from 1985 to 1990. He has been the President of the Italian Association for the Study of Comparative Economic Systems (AISSEC) and a member of the council of the European Association for Evolutionary Political Economy (EAEPE), which in 1997 awarded him the Kapp Prize. He is the author of Work and Welfare in Economic Theory (Blackwell 1985) and the co-editor of Democracy and Efficiency in the Economic Enterprise (Routledge 1995), The Politics and the Economics of Power (Routledge 1999) and The Evolution of Economic Diversity. A member of the Santa Fe Institute Founding Group on ‘New Directions in Behavioural Sciences’, his research interests include the role of rationality and emotions in decision making and the analysis of evolutionary theories in social and in natural history. John Smithin is Professor of Economics in the Department of Economics and the Schulich School of Business, York University, Toronto, Canada. He holds a PhD from McMaster University, and has previously taught at the University of Calgary and Lanchester Polytechnic (now Coventry University) in England. His main research interests are in the fields of monetary economics, international finance and macroeconomics. He is the author or editor of Macroeconomic Policy and the Future of Capitalism (1996), Money, Financial Institutions and Macroeconomics (1997), What is Money? (2000) and Controversies in Monetary Economics (2003).

Acknowledgements

I am very grateful to Terry Clagne, Mary Donaldson, Clive Lawson, Tony Lawson, Nadine Lewis, Fintan Power and Jochen Runde for their assistance in the preparation of this volume, and to the contributors for their efforts and patience.

1

Transforming economics? On heterodox economics and the ontological turn in economic methodology Paul Lewis

Introduction1 One of the most significant recent developments in the methodology of economics has been the increasing attention devoted to the ontological commitments of economic theories, that is to what those theories presuppose about the nature of socio-economic reality. Whereas methodological debate in economics in the 1970s and 1980s focused on the epistemological issues associated with the construction and evaluation of theoretical claims, the 1990s witnessed an upsurge in interest in examining the (often implicit) assumptions that economists make about the constituents of the socio-economic world.2 In the vanguard of this ‘ontological turn’ is an approach known as critical realism. This approach, whose most prominent exponent in economics is Tony Lawson (1997, 2003), is prescriptive in orientation, being predicated on the claim that research into socio-economic life is most likely to bear fruit if it uses tools which are tailored to suit the nature of the socio-economic material under investigation. To this end, critical realists deploy philosophical arguments in order to develop an abstract account of the nature of the socio-economic world, which is used to inform subsequent attempts to characterise the methods appropriate for studying socio-economic life.3 One of the central claims advanced by critical realists, to anticipate the conclusion of a line of argument which will be set out in detail below, is that the methods employed by mainstream economists are poorly suited to the investigation of the socioeconomic world. And the reason why orthodox economists have persisted for so long in using methods that are largely inapplicable to socioeconomic reality, critical realists contend, is that they have engaged in far too little sustained, explicit (ontological) reflection about the nature of their subject-matter and the methods that are most likely to bear fruit in investigating it. However, while supporters of critical realism are sceptical about the merits of orthodox economics, they maintain that they share a good deal in common with various heterodox schools of thought. The origin of this

2 Paul Lewis common ground, critical realists argue, lies in the fact that heterodox economists have frequently invoked claims that carry ontological presuppositions in line with those explicitly defended in critical realism (Lawson 1997: xiii–xiv, xvii; Fleetwood 1999). A critical realist reading of heterodox economics suggests that many Austrian, Evolutionary, Feminist, Old Institutionalist, Intersubjectivist, Marxist, Post Keynesian and Social economists have emphasised the importance of adopting a methodology that is tailored to suit the nature of the socio-economic material under investigation, pitching their criticisms of orthodox economics and the case for their own preferred methods of economic analysis on grounds which presuppose (if only implicitly) something very like the socio-economic ontology delineated in critical realism. Proponents of critical realism argue that it provides philosophical resources that can help heterodox economists to make explicit, clarify and systematise their insights into the nature of socio-economic being and its implications for appropriate methodology for economics. The sustained, explicit reflection about ontological issues facilitated by critical realism, its advocates contend, is useful in a number of ways: it allows intuitions to be more fully articulated and developed; it helps to reveal internal inconsistencies in theoretical positions and arguments; it facilitates a greater sensitivity to the trade-offs involved in adopting one analytical framework in preference to another; and it enables researchers to identify more accurately the differences between competing approaches, thereby promoting a more informed, and therefore hopefully respectful and productive, dialogue between their proponents. This brings us to the current collection of essays, the purpose of which is first of all to solicit the views of heterodox economists concerning the claim that there are strong affinities between various heterodox schools of thought and critical realism, stemming from shared ontological presuppositions, and second to provide a forum for exploring and developing such commonality where it exists. More specifically, the authors were asked to assess from the vantage point provided by their own project how (if at all) the critical realist concern with ontological analysis coheres with their own concerns and to evaluate the contribution (if any) that critical realism can make to transforming the aims and practices of modern economics in order to facilitate a more relevant and fruitful discipline than exists at present. The objective of this opening chapter is to outline the essays contained in the current volume and also to provide an overview of the critical realist approach to economic methodology. The overview is intended to introduce the project of critical realism in economics to those readers who are unfamiliar with it, thereby ensuring that this collection is reasonably self-contained. It is to the task of outlining critical realism that we now turn our attention.

Transforming economics? 3

Critical realism and the ontology of the socio-economic world We have already noted the centrality to critical realism of the claim that successful social-scientific research requires that the analytical tools employed are tailored to the nature of the socio-economic material under scrutiny (Lawson 1997: xii–xiii, xviii, 16; Archer 1998: 72–3). Critical realists are methodological realists, advocating the use of methods that do not require researchers to employ convenient fictions in their theories. Of course, if critical realists contend that economists should tailor their analytical tools and modes of explanation to suit the characteristics of those features of socio-economic reality being studied, then it is incumbent upon them to provide an account of the nature of the latter. It is for this reason that supporters of critical realism have set such great store by the study of socio-economic being or ontology, using philosophical arguments to derive an account of the nature of socioeconomic reality which is then used to characterise the methods and mode of explanation appropriate for social research. A frequent point of departure for the critical realist analysis of the nature of the socio-economic world is the a posteriori observation of a paucity of stable event regularities in the socio-economic world. (An event regularity is a pattern of the form: ‘Whenever event or state of affairs x, then event or state of affairs y’.) Situations where event regularities obtain are said to be closed; situations in which such regularities are absent are termed open systems. The inability of econometricians to find many sharp and enduring event regularities indicates to critical realists that the socio-economic world is open (Lawson 1997: 70; 1998: 357–60). Critical realists contend that one reason for the observed openness of the socio-economic world is to be found in the perceived reality of human choice. For the possibility of genuine choice must surely entail that a person’s behaviour is not simply a determinate or single-exit response to her external environment. On this view, choice is an open-ended enterprise such that if in any given circumstances (x) a person chose to do y, then (s)he could have chosen to undertake a different course of action (not-y). But if that is indeed the case, then the possibility of genuine choice presupposes that the socio-economic world is open (Lawson 1997: 8–11). Critical realists build on this line of reasoning by arguing that if a person’s actions are indeed chosen, then it must be possible to give an account of them in terms of that person’s beliefs and desires; that is to say, human action must be intentional under some description. But such purposeful action requires that people have some idea about how to achieve their goals. For if people were so ignorant of the future that they were completely unable to discern a connection between their decisions and ensuing states of affairs, then intentional action would be impossible.

4 Paul Lewis But whence comes the knowledge that informs people’s decisions? The openness of the socio-economic world suggests that the objects of such knowledge are unlikely to be event regularities. Perforce, the relevant objects of knowledge must be non-empirical (Lawson 1997: 30–1). More specifically, advocates of critical realism suggest that people’s actions are informed and guided by their knowledge of (non-empirical) social structures (social rules and institutions). For example, long-term contracts are a social institution that enables economic actors to secure a degree of control over their future income and expenditure (in the case of households) and revenues and costs (in the case of firms), thereby providing them with a measure of assurance about the future consequences of their actions. While such contracts do not tie the future down completely, and so do not give rise to stable event regularities (there always remains the possibility, sometimes realised, that one or more parties will unexpectedly renege on their contractual commitments), long-term contracts (in conjunction with the broader network of institutions and rules that constitute the legal system) do provide people with enough confidence in the consequences of their actions to facilitate intentional agency. In a similar vein, money is a social institution that both serves to facilitate economic calculation and also enables people to deal with the vagaries of an uncertain future by providing them with the liquidity required to deal with unforeseen obligations and opportunities (Runde 1993: 388–93; Lawson 1998: 357–62; McKenna and Zannoni: 2000–1). This line of reasoning yields a metaphysics6 which suggests that the socio-economic world can be divided into three ontologically distinct domains: the actual (actual states of affairs and events, including people’s actions and practices); the empirical (our sensory experiences of the actual); and the realm of underlying social structures, which (as we shall elaborate in due course) both enable and constrain the activities of economic agents. Critical realists express the fact that the socio-economic world is composed of a number of ontologically distinct realms by saying that the socio-economic world is structured. Underpinning this account of the constituents of socio-economic reality is a theory of the relationship between social structure and human agency. In developing the latter, critical realists strive to find a middle way between the reductionist extremes of voluntarism (according to which social structure is in effect created ex nihilo by human agency) and determinism (which portrays people as cultural dupes whose actions are completely determined by their social environment). Instead of regarding social structure as being ontologically reducible to human agency, or vice versa, critical realists subscribe to a transformational model of social activity according to which structure and agency are recursively related; each is both a precondition for and a consequence of the other. It is possible to elaborate on this idea in the lexicon of heterodox economics by saying that

Transforming economics? 5 the interaction between structure and agency is an intrinsically historical process and as such is conceptualised as occurring in real historical time. On this view, the structure–agency relationship must be conceptualised as an inherently ‘tensed’ process whereby people continuously draw upon (pre-existing) social structures in order to act and in which their actions, taken together, subsequently lead either to the reproduction or the transformation of those structures (Archer 1995: 65–92, 154–8; Sztompka 1993: 199–201, 215–19, 224–9).7 We can elaborate on the significance of historical time for the structure–agency relationship by noting first that all human activity takes place within the context provided by a set of pre-existing social structures. At any particular moment people confront social structures that are ‘ready-made’ in the sense that they are the product not of their actions in the present, but of actions undertaken in the past. Perhaps most obviously, everyone is born into a world of pre-existing social structures. These structures are imposed involuntaristically upon the current generation of people, confronting them with a reality that is objective in the (moderate) sense that it is ontologically irreducible to their current subjective beliefs and actions (Layder 1997: 9–28, 108–16, 165–6, 246; Sayer 2000: 18, 35, 58–61). Hence, according to critical realism, pre-existing social structures, inherited ‘already determined’ from the past, are ontologically distinct from the practices of actors in the present. This is significant because the fact that antecedent social structures pre-exist and are therefore ontologically irreducible to the current exercise of human agency implies that social structure is not simply the voluntaristic creation of human agency. It is worth noting at this juncture that while critical realism suggests that historically given social structures are objective in the sense that they are ontologically irreducible to people’s current actions, it does not reify them by denying their ultimate dependence on human activity. On the contrary, critical realists take great pains to identify precisely whose activities are responsible for the social structures that exist at any particular moment in time (Archer 1995: 66, 72, 141–54). Two points are worth noting in this regard. First, critical realists readily admit that historically given social structures are the product of human action. However, the point is that their activity-dependence is past tense; the structures in question are the product of actions undertaken in the past, possibly by actors long since dead, and as a result are ontologically irreducible to current agency (in the present). Second, critical realists also accept that while the historically given social structures extant at a particular moment in time are the result of actions undertaken in the past, their continued existence depends upon current human agency. However, as has already been explained, current human agency is informed and so made possible by pre-existing social structure. Moreover, as we shall see below, at any particular instant in time both the incentives that people

6 Paul Lewis have to attempt either to preserve or to change the structures which initially confront them, and also the distribution of the resources required to act on those incentives, are themselves embodied in the antecedent social structures as an historically given product of actions undertaken in the past. What this shows, advocates of critical realism contend, is that the human agency essential for the (continued) existence of social structure is itself dependent in a variety of ways upon (antecedent) social structure. It is this mutual dependence that prompts critical realists to describe social structure and human agency as recursively related. More specifically, because current human agency takes place within the context provided by pre-existing social structures, and is itself dependent upon them, the relationship between social structure and human agency is best thought of as one in which current agency reproduces or transforms (preexisting) social structures rather than (voluntaristically) creates them out of nothing. And it is for this reason that the critical realist theory of the relationship between social structure and human agency is known as the transformational model of social activity (Archer 1995: 71–2, 137–41; Lawson 1997: 168–70). On this view, pre-existing social structures both facilitate and constrain current human agency. Historically given social structures are a prerequisite for human intentional agency. For instance, activities such as speaking and writing, setting a university examination paper, and drawing up and enforcing contracts, are made possible by preexisting social structures such as (respectively) rules of grammar, student–teacher relationships and the legal system. However, social structures not only facilitate human agency; they also constrain it. For the fact that social structure pre-exists and is therefore ontologically irreducible to current human agency implies that it enjoys a measure of autonomy from and influence over the latter (Archer 1995: 137–9). Someone who wants to communicate effectively in Britain has to speak English. A person who insists on speaking Spanish, say, is unlikely to be understood. Merely wishing that things were different, or acting as if they were so, will not make it so. This is not necessarily because of awkwardness or intransigence on the part of the person’s British interlocutors. Rather, it reflects the fact that while the rules of English grammar facilitate successful communication in Great Britain, they also compel one to speak English if one does indeed wish to communicate. Social structure can thus be seen to make a difference to the way in which people behave in the sense that the existence of a particular set of social structures (rules of grammar, for example) militates in favour of certain actions being undertaken (speaking English). For critical realists, this capacity to ‘make a difference’ (in this case to the behaviour of socioeconomic actors) is the hallmark of causal efficacy. Social structures, conceptualised as pre-existing and therefore as enjoying a degree of autonomy from current agency, can thus be seen to exert their own, sui generis causal influence on current social activity (Archer 1995: 139, 147–8,

Transforming economics? 7 176; Lawson 1997: 31–2, 57–8; Lewis 2000: 251–2). There are social structures (for instance, the language which people are taught) that cannot be altered immediately, even if all relevant parties wish to do so. For example, however much most of the indigenous population might wish to help a visitor to Britain by speaking Spanish, they are unlikely to be able to do so because, thanks to the activities of the parents and teachers who educated them in the past, they were taught not Spanish but English. Such social structures are the consequences of past actions and consequently are irreducible to the contemporary activity. The fact that ultimately, with sufficient time and effort, succeeding generations of people will be able to alter inherited social structures is undoubtedly true; but the point is that, until the (possibly quite lengthy) process of change has run its course, the inherited structures in question will continue to influence people’s choice of action and so exert an irreducible causal influence over the course of events in the socio-economic world (Archer 1995: 76–9, 85, 195–201; 1998: 77–8; 2000a: 465–9).8 Advocates of critical realism argue that it is possible to acknowledge the causal efficacy of social structure without suggesting that people’s actions are uniquely determined by those structures. For instance, while a person who is in Britain and who wishes to communicate with other people will have to use the pre-existing rules of English grammar in order to produce intelligible speech acts, those rules do not determine what (s)he says or writes. It may also be the case that the person in question is a taciturn character who chooses not to communicate at all. What this goes to show is that while critical realism regards social structure as being causally efficacious, it does not reduce people to the status of cultural dupes whose actions are dictated by pre-existing structures (Archer 1995: 195–6; Layder 1997: 22, 111–12, 200–2; Porpora 1998: 346–7). Additional insight into the relationship between human agency and social structure, and the causal efficacy of the latter, can be had from a closer examination of the way in which they come into contact with each other. Critical realists argue that two observations are particularly noteworthy in this regard. The first is the fact that society is typically structured along hierarchical lines in the sense that different people have different rights and responsibilities. For instance, landlords are obliged to ensure that the accommodation they provide meets certain safety standards, in return for which the landlord is entitled to insist that tenants pay the rent on time and keep the property clean and tidy. The second observation is that these rights and obligations usually exist independently of specific individuals. For example, new tenants usually have the same rights and responsibilities as their immediate predecessors. Advocates of critical realism argue that these two observations can be rendered intelligible once it is acknowledged that social structures are made up of sets of positions, each of which is associated with particular obligations and privileges that pre-exist any given occupant. And it is in

8 Paul Lewis the occupancy of these positions by specific individuals – and more specifically via the conditioning of the subsequent behaviour of the incumbents by the rights, obligations and (as we shall see) interests and resources which accompany the positions – that the interface of social structure and agency is to be found. Another significant observation is that the practices routinely followed by the incumbents of such positions tend to be directed towards the occupants of other positions. For example, the rights and obligations of landlords are oriented towards, and defined in terms of, their interactions with tenants. Supporters of critical realism infer from this that social structure is highly relational in nature and distinguish two varieties of relation. Internal relations obtain when the relata are constituted by the relations in which they stand to one another. The position of a landlord, say, is at least partly defined by the relations between landlords and tenants, suggesting that the positions of landlord and tenant are internally related (Lawson 1997: 64; Sayer 2000: 13). Two positions are externally related when neither is defined in terms of its relations to the other. The prevalence of cases such as landlords and tenants, teachers and students, and bosses and workers suggests ex posteriori that the socio-economic world is highly internally related. The vantage point provided by critical realism suggests, then, that social structure is most appropriately conceptualised as a nexus of often internally related positions, each of which is accompanied by a set of rights and obligations. These positions tend also to be associated with vested interests which, like the social structures themselves, constitute an objective reality that endows their occupants with reasons or incentives to pursue particular goals (Lawson 1997: 264; 1999a: 247; Porpora 1998: 352–3). Historically given social structures also influence current activity because at any particular instant in time the material and cultural resources required to prosecute particular courses of action (wealth, power, status, access to credit and the like) are distributed unequally between the various positions (along class, gender and racial lines, for example) (Layder 1997: 80–1, 122–3; Marsh 2002). Depending on their location in these social hierarchies, then, people are endowed with an historically given array of resources, which in turn constitutes an ontologically irreducible influence on their current ability to further their interests (Lawson 1997: 257; Porpora 1998: 349–52). Antecedent social structures, and the distribution of vested interests and resources they embody, comprise the social context in which people must act. Given that such structures are external to and autonomous from the current cohort of agents, they exert an irreducible causal influence on current activity. Of course, critical realism stops short of a reductive materialism in which people’s conduct is determined by their social context. The transformational model of social activity does not sustain the view that social structures act behind the backs of agents or that people are merely

Transforming economics? 9 the playthings of social structure. The fact that critical realism portrays people and social structure as ontologically distinct from one another is significant in this regard, for it implies that the two may possess different properties from one another. In the present context, it enables supporters of critical realism to claim that purposefulness, reflexivity, intentionality and the capacity to initiate activity characterise people without anthropomorphically attributing such properties to social structure. While social structures are causally efficacious, they are neither conscious decision-making entities nor ‘prime movers’ that initiate activity and subordinate actors to the dictates of some structural imperative (Bhaskar 1989: 35; Archer 1995: 195–6, 198; Layder 1997: 20, 100–1). On the contrary, as we have seen, the transformational model of social activity suggests that social structures influence socio-economic affairs only by conditioning people’s choice of action. Consequently, critical realists can speak of the causal impact that pre-existing circumstances exert on socioeconomic affairs without denying either the mediating role of people’s interpretations (and hence the necessity of a hermeneutic moment in social science)9 or the possibility that people may respond to their circumstances in an innovative way, and therefore without making any deterministic claims about the connection between people’s material circumstances and their actions. While interests always provide presumptive grounds for action, different individuals may not reach the same interpretation of given material circumstances, leading them to have different (and possibly even mistaken) views of where their interests actually lie. Furthermore, even if agents do have similar ideas of their interests, they may come to different conclusions about how best to pursue them. In particular, critical realists fully acknowledge that people may creatively devise new strategies for deploying (historically given) resources so as to achieve their desired goals. Moreover, it is even conceivable that, for a variety of reasons, people may act against their interests. For all of these reasons, the pre-existing pattern of interests and resources, while influencing people and motivating them to act in certain ways, does not uniquely determine how they behave. But, to repeat, what critical realists do claim is that pre-constituted social structures, together with the associated distribution of vested interests and resources, have a potentially important impact on people’s actions and therefore on the course of events in the socio-economic world, and that as a result they ought not to be excluded from attempts to explain socio-economic events of interest. Overall, then, critical realism implies that socio-economic life is best conceptualised as an intrinsically dynamic process of interaction between pre-existing social structures and current human agency, occurring in historical time. Social structures are a necessary condition for individual acts but it is only through (the totality of) the actions of individuals that they persist over time. Social structures should never be regarded as

10 Paul Lewis permanently fixed – they should never be reified – because, given their dependency on (potentially creative and so transformative) intentional agency, the scope for change is ever-present. Hence, both society in general and specific social institutions, such as the market, must be understood as inherently dynamic processes in which change is initiated not only by exogenous shocks but can also be endogenously generated as an integral part of social life (Lawson 1997: 34–5, 170–1, 187–8; Layder 1997: 81–2). For critical realists, then, ‘[T]he future … is forged in the present, hammered out of past inheritance by current innovation’ (Archer 1996: xxvi). Pre-existing social structures, the legacy of actions undertaken in the past, constitute the environment in which current activity occurs. As we have seen, these historically given structures condition (but do not determine) the behaviour of socio-economic actors in the present by laying down an initial distribution of resources and vested interests. The (reproduced or transformed) set of structures that emerges as the outcome of this process in turn forms the context for the next round of activity. Critical realism’s emphasis on ontological issues should not be taken to indicate that it neglects epistemology. Indeed, as has already been noted, the whole point of the critical realist analysis of socio-economic ontology is to facilitate a more informed account of the methods that are most appropriate for producing knowledge of socio-economic reality. So, what follows from the critical realist account of the nature of socio-economic being for the methods employed in economic analysis? The model of explanation sponsored by critical realism suggests that explaining some socio-economic phenomenon of interest consists in giving an abstract, usually discursive and always fallible account of its causes. On this view, socio-economic phenomena are to be explained as the outcome of the causal interplay over historical time between (antecedent) social structure and (subsequent) human agency. More specifically, the initial stage of an explanation involves the identification of the practices responsible for the phenomenon under investigation, after which it is necessary to uncover the social structures and tacit skills which facilitate those practices, together with any conscious and unconscious psychological factors which motivate them (Lawson 1997: 56–8, 191–271).10 The question which naturally arises is how can such information be obtained? In answering this question, supporters of critical realism readily acknowledge that while the socio-economic world pre-exists the activities of social researchers rather than being simply a construct of the latter, social scientists do not have immediate access to it from some God’s-eye view; there is no Archimedean point from which we can view the world As it Really Is.11 On the contrary, critical realists are epistemological relativists in the sense that they readily acknowledge that the world can be known only under particular, historically specific, theory-laden, and therefore fallible and corrigible, descriptions. All

Transforming economics? 11 knowledge is a theory-dependent social product, actively produced through the endeavours of social scientists as they attempt to apply, test, re-work and extend existing theories and data. To this end, critical realism postulates the existence of a transitive domain of knowledge – consisting of theories, observations, intuitions, (theory-laden) observations and the like – to complement the intransitive domain, or ontology, already discussed. Supporters of critical realism contend that by combining this epistemological relativism with their ontological realism, they can sustain a judgemental rationality according to which it is possible (fallibly) to discriminate between competing theories on the basis of their explanatory power, that is their success (or lack thereof) in illuminating the causal processes at work in the socio-economic world (Lawson 1997: 25–6, 58–9, 238–46).12 Of course, the claim that the most appropriate way to explain some socio-economic phenomenon of interest is to give a discursive account of its causes leaves critical realism somewhat at odds with orthodox economics, according to which explanation in economics proceeds by deducing explananda as the equilibrium outcomes of formal, rational choice models. Critical realists justify their preferred causal model of explanation on the grounds that it is better suited to the nature of the socio-economic material under investigation than the mathematical models so beloved of orthodox economists. According to proponents of critical realism, the various mathematical techniques and substantive assumptions utilised by orthodox economists reflect a long-standing commitment to deductivism. By ‘deductivism’ is meant a methodology which dictates that socio-economic phenomena should everywhere be analysed or ‘modelled’ in terms of theoretical results deduced from sets of axioms and assumptions of the form, ‘Whenever these conditions, then that outcome (or distribution of outcomes)’. Supporters of critical realism argue that the use of a deductivist methodology is analytic to orthodox economics (Lawson 1997: 86–107, 69–85; 1999a: 223–33). However, critical realists maintain that deductivism suffers from a number of limitations so far as the analysis of the socio-economic world is concerned. In the first place, while regularities between (sets of) actualities of the form ‘Whenever these conditions, then that (set or spread of) outcome(s)’ are central to deductivism, (as we have already seen) ex posteriori observation reveals the socio-economic world to be an open system in which such regularities are conspicuous by their absence. The paucity of regularities displaying the strictness presupposed by orthodox economics leads critical realists to be extremely sceptical about the usefulness of such models for socio-economic analysis (Lawson 1997: 70–1). This scepticism is heightened by the fact that the construction of closed-system models typically necessitates the distortion (within the theory) of aspects of the socio-economic world. The vantage point provided by critical realism suggests that it is necessary to make a number

12 Paul Lewis of ‘closure assumptions’ which are designed to exclude various sources of indeterminacy from the model and so facilitate the deducibility of theoretical results. First, people are modeled as if their conduct is determined by their preferences and the constraints they face. The standard model of rational choice exemplifies this modeling strategy: people are assumed to face a given means–ends framework that yields a single rational course of action. Questions of interpretation – of divining the meaning of prices and events and of judging their significance – are foreclosed, as too is the scope for people to display creative, entrepreneurial responses to their environment. On this view, the mainstream’s commitment to deductivism leaves it with an impoverished notion of choice because, as was argued above, behaviour that is merely a mechanical, automaton-like response to external circumstances can hardly be said to have been ‘chosen’. Second, deductivism also requires that the actions of individual agents yield a determinate outcome at the aggregate level. One way of attempting to construct models which possess this property is to assume that the socio-economic world is atomistic, so that the actions of different agents combine mechanically to produce an aggregate outcome that is no more than the additive sum of their individual behaviour. However, if the critical realist analysis reported above is correct in suggesting that the socio-economic world is highly internally related, then models which presuppose that it is atomistic ignore a potentially important aspect of socio-economic reality (Lawson 1997: 78–81, 98–102, 185–6). For supporters of critical realism, then, the mainstream’s neglect of the openness and holistic nature of the socio-economic world implies that orthodox economics provides a highly distorted account of socioeconomic reality. The dearth of stable closures in socio-economic life, taken in conjunction with the inability of models constructed along deductivist lines to do justice either to the reality of human choice or to internal relations, leads to the conclusion that the applicability of orthodox economic models to the socio-economic world is distinctly limited (Lawson 1997: 19–20, 86–133, 282; 2001b: 76–9). The fact that orthodox economics has persisted with a methodology that is so unsuited to its subject-matter contrasts sharply with the way in which critical realists set a premium on methods which have been tailored to suit the nature of the socio-economic world. The neglect of ontological issues displayed by mainstream economics leads advocates of critical realism to argue that orthodox economics is not realist enough and that the (epistemological) goal of generating knowledge about the socio-economic world would be better served by a shift of resources in the discipline of economics away from deductivist modeling towards those heterodox approaches that adopt something like a critical realist approach (Lawson 1999b: 270–4).

Transforming economics? 13 For, as was noted above, proponents of critical realism argue that the neglect of explicit ontological reflection displayed by orthodox economics is not representative of the discipline as a whole, being countered by the ontological orientation of many (though not all)13 heterodox schools of thought. Advocates of critical realism attempt to promote the development of heterodox economics in two main ways. The first lies in the use of the philosophical framework furnished by critical realism to clarify and refine various fundamental categories in socio-economic analysis – for instance, ‘social order’ (Fleetwood 1996), ‘tendency’ (Pratten 1998), ‘probability’ (Runde 1999), ‘competences’ (C. Lawson 1999, 2003), ‘technology’ (C. Lawson 2004), ‘markets’ (Fleetwood 2004a, 2004b) and ‘need’ (Lawson 2000) – and also to provide an account of the relationship between social structure and human agency (Archer 1995, 2000b; Lewis 2000, Forthcoming b; Lewis and Runde 2002). The elaboration of such basic concepts facilitated by critical realism, its proponents argue, can help to make explicit, clarify and so realise the full potential of the intuitions that heterodox economists have about the socio-economic world. In this way, so its advocates argue, critical realism can serve to clear the conceptual ground – or, in critical realist parlance, to underlabour – for economics (Lawson et al. 1996; Lawson 1997: 60–1; 1999c: 3, 14–15). We can elaborate on the notion of underlabouring by drawing a distinction between abstract philosophical analysis like that found in critical realism and substantive or concrete social scientific research. The analysis advanced by critical realism is pitched at a higher level of abstraction than concrete research, and is designed to provide guidance about the sort of methods that are likely to bear fruit in investigating the socio-economic world. More specifically, as we have seen, critical realism is intended to encourage researchers to develop (substantive) theories that do justice to the structured, intransitive, open, dynamic and at least in part holistic nature of socio-economic reality. However, critical realism does not attempt to answer substantive social scientific questions. It does not purport to provide mechanical or algorithmic procedures for answering substantive questions about the socio-economic world (Lawson 1997: 191). Indeed, there may be a number of concrete accounts that are consistent with critical realism, between which the latter does not distinguish. Just as there is no specific theory that can be designated as the deductivist theory of any particular substantive phenomenon, so too there is no such thing as the critical realist account. Rather, proponents of critical realism contend that the variety of substantive theories that are consistent with the general critical realist approach should be judged according to their explanatory power, that is their capacity to explain a wide range of substantive socio-economic phenomena of interest. And, to reiterate, such judgements are the prerogative of substantive social science, not of critical realism per se (Lawson 1999c: 14–15, 1999d: 3–6).

14

Paul Lewis It is worth emphasising that supporters of critical realism have always emphasised the fallibility of their claims concerning the ontology of the socio-economic world and the methods that are appropriate for investigating it. The vantage point provided by critical realism neither elevates its supporters to some Olympian vantage point from which they can pronounce Final Judgement on substantive research programmes nor offers an infallible algorithm whose mechanical application will yield absolute Truth (Lawson 1997: 43, 58–61, 191). Rather, advocates of critical realism hope to add another voice to the conversation in economics that, while eschewing any claim to absolute authority, highlights the importance of tailoring research methods to the nature of the socioeconomic material under investigation. And the significance of these arguments, advocates of critical realism contend, is that a reallocation of research effort away from the deductivist, closed-system approach of orthodox economics in favour of open-system methods of the sort often used by heterodox economists is warranted (Lawson 1999e: 50–1, 54 n.4). In addition, as we have seen, it is hoped that critical realism will offer conceptual resources that will assist heterodox economists to articulate more fully and so more persuasively their own views about the working of the socio-economic world. One highly significant (some would say insurmountable) impediment to the reallocation of resources just described lies in the fact that, however many insights they may yield, approaches which depart from a deductivist methodology are likely to be viewed with suspicion by orthodox economists on the grounds that they do not proceed in a ‘scientific’ manner. Mainstream economists believe that the use of formal, deductive models is the hallmark of the natural sciences and that economics has a legitimate claim to scientific status only if it proceeds in a similar fashion. For members of the economics academy, mathematical modeling offers economists a common, grammatical language that facilitates the rigorous derivation of results by deduction from precisely defined assumptions. This predilection for a mathematical language over more discursive styles of reasoning manifests itself in the condemnation of approaches which fail to conform to the rules laid down by the formal, deductivist grammar as ‘unscientific’, ‘lacking in rigour’ and a-theoretical (‘merely descriptive’ or, even more pejoratively, ‘nihilistic’) and in their exclusion from the domain of ‘proper’ economics. The feminist economist Diana Strassmann has expressed this point clearly as follows: To a mainstream economist, theory means model and model means ideas expressed in mathematical form … Mainstream economists believe proper models – good models – take a recognisable form: presentation in equations, with mathematically expressed definitions, assumptions, and theoretical developments clearly laid out … Theories that do not take this form may not be recognised as ‘theory’

Transforming economics? 15 so that the valorisation of mathematical language constrains what counts as theory in economics. (Strassmann 1994: 153–4, 156) Strassmann is not alone in airing such views. Heterodox economists of all schools often lament the fact that their work receives the sort of response described by Strassmann: Austrians (Rizzo 1992: 247–51; Boettke 1994: 605, 610; 1997: 21), Feminists (Strassmann 1993a: 54–7, 64–5; 1994: 153–6; Harding 1999: 129–30), Institutionalists (Hodgson 1988: 74; Samuels 1997: 137), Post Keynesians (Chick 1995: 20–3; Dunn 2000: 351) and development economists (Bauer 2000: 20) have all commented that their work has been denounced as unscientific and excluded from the domain of ‘proper’ economics simply because it is not expressed in terms of formal, mathematical models. The state of affairs just described constitutes the setting for the second way in which critical realism seeks to assist the development of heterodox economics, namely by providing philosophical arguments which contest the claim that schools of thought which eschew formal, deductivist methods must necessarily be unscientific. The perspective provided by critical realism suggests that the orthodox position on this issue is based on an erroneous account of the natural sciences. To see why, recall that deductivism implies that regularities of the form ‘Whenever this event or state of affairs, then that (set or spread of) outcome(s)’ are central to science. However, any account that places event regularities at the heart of science runs into grave difficulties when confronted with two widely accepted observations concerning experimental activity in natural science. First, with the notable exception of astronomy, most of the empirical generalisations or regularities that are held to be significant in the natural sciences occur only within the confines of scientific experiments. Second, notwithstanding the first observation, results obtained via controlled experiments are successfully applied outside the confines of the experimental set-up, in situations where sharp, stable empirical regularities are largely absent. These observations pose serious problems for the deductivist account of science. For the fact that most of the event regularities regarded as noteworthy by natural scientists occur only under the restricted conditions of experimental control bears the unfortunate implication that the existence of the regularities which (on the deductivist account of science) constitute the laws of nature depends on human intervention. In addition, if the regularities that are the sine qua non of deductivist explanations are largely confined to experimental conditions, then deductivism is able to explain neither what governs events outside the experimental environment nor the successful application beyond the laboratory of the results obtained therein. These two observations can, however, be rendered intelligible once it is acknowledged that the natural world is structured and intransitive. The

16 Paul Lewis metaphysics advanced by critical realism suggests that, like the socioeconomic world, nature consists not only of actual events and our experiences of them, as presupposed by deductivism, but also of ontologically irreducible structures and mechanisms which underlie and govern the flux of actual events. Furthermore, with one or two notable exceptions (such as the astronomical realm), the natural world is an open system in which events are influenced by a variety of different and often countervailing mechanisms. To take the familiar pedagogical example, the behaviour of a falling leaf typically does not conform to an empirical regularity simply because the leaf is affected in different, and conflicting, ways by gravitational, aerodynamic and thermal forces. Experimental activity, according to this perspective, is best thought of as an attempt to intervene in the natural world in order to insulate the activity of one particular mechanism from the impact of countervailing forces. The ultimate objective of such experiments is to engineer a closed system in which there is a one-to-one correspondence between the action of a mechanism and ensuing events, so that the impact of the mechanism in question can be identified empirically. And the results obtained in this way are applicable to non-experimental situations in virtue of the fact that, once triggered, the mechanism under investigation is operative and in play even in non-experimental situations (that is to say, the mechanisms identified in scientific experiments are atomistic or stable across different contexts) (Bhaskar [1975] 1997: 33–7; Lawson 1997: 27–30). What this line of reasoning suggests, then, is that it is only if the deductivist view that scientifically interesting generalisations about nature consist in event regularities is rejected and replaced by something like the critical realist perspective, according to which the objects of science are both structured and intransitive, is it possible to render intelligible the role of experimentation in the natural sciences (Bhaskar [1975] 1997: 33–6; Lawson 1997: 27–8). And the objective of science, on this critical realist account, is not to elaborate patterns at the level of observable events and states of affairs but rather to identify and illuminate the (non-empirical) causal mechanisms underlying empirical phenomena.14 The significance of this line of reasoning derives from the fact that as soon as the natural sciences are adequately conceptualised, it becomes apparent that the work of those heterodox economists who adopt something like a critical realist approach can be just as scientific as that of natural scientists, even though the heterodox theories in question are expressed discursively, in natural language, rather than in the mathematical language of formal models. For critical realism suggests that just as explanation in the natural sciences proceeds via the identification of the (non-empirical) causes of natural events and states of affairs, so (as we have seen) explaining some socio-economic phenomenon of interest involves the identification of the practices and

Transforming economics? 17 (non-empirical) social structures and unconscious motivations which gave rise to it. Viewed at a sufficiently high level of abstraction, then, explanation in the social sciences can be seen to proceed in the same way as in the natural sciences, involving a movement from a knowledge of some empirical phenomenon to knowledge of its underlying causes. And in suggesting that there is a logic of explanation that is common to the study both of nature and of society, critical realism supports the view that those heterodox approaches which subscribe to something like a critical realist perspective can be scientific in precisely the same (abstract) sense as natural sciences like physics and biology (Lawson 1997: 20, 56–7, 126).15 The significance of this point for heterodox economists lies in its implications for the way that heterodox economists are able to proselytise on behalf of their preferred style of economic analysis. Confronted by an orthodoxy that denies legitimacy to their work and excludes them from the conversation in economics on methodological grounds, heterodox economists may find in critical realism rhetorical resources – arguments – which can assist their attempts to rebut the charge that they do not employ a scientific methodology. Critical realism, by suggesting that heterodox economic theories can be expressed discursively and yet still be scientific (if science is properly understood), may help heterodox economists to establish the scientific credentials of their work and so (hopefully) gain access to the conversation in the discipline of economics (Lawson 1997: ii). The two aspects of the critical realist contribution outlined above – the elaboration of various concepts integral to socio-economic analysis and the defence of the discursive mode of theorising – go hand-in-hand, for it may well be possible to do justice to the insights found in heterodox economics analysis only if the formal, mathematical language preferred by orthodox economists is abandoned in favour of natural language. The point has been well made by (once again) Diana Strassmann (1993b: 152, 153; also see Lavoie 1994: 3–4): [A] critical barrier in economics to the incorporation and dissemination of feminist and interpretative [more generally, ‘heterodox’] thought has to do with restrictions on acceptable language, and the processes by which such restrictions are used to limit disciplinary membership and define acceptable economic practice. These issues extend to the very words that can be used in the discourse of an intellectual community … A consequence is that feminist and interpretative [heterodox] work in economics must often choose between redeveloping and explaining a panoply of concepts and presenting a more simplified and essentially inadequate exposition of arguments … In particular, by requiring ‘worthy’ research to conform to a particular conceptual framework and rhetoric, those with such power are able to constrain the nature of

18

Paul Lewis interpretations and texts [meaning] … [In this way t]he language of the powerful constrains and influences communication and insight … hinder[ing] the development of new conceptualisations.

For critical realists, then, a philosophical critique of deductivism is an important part of the task of underlabouring for heterodox economics, for it is only by defending and so continuing to make use of discursive modes of theorising that heterodox economists will be able to exploit fully the potential of their preferred conceptual repertoire (Lawson 1999: 219–21).16 By demonstrating that heterodox theories can be expressed discursively and still provide causal explanations of the sort found in the natural sciences, critical realism’s account of the possibility of naturalism can help furnish heterodox economists with the rhetorical resources required to defend themselves against the charge that their work is unscientific and nihilistic. And, in turn, the reclamation of the mantle of scientific respectability on behalf of heterodox economics facilitated by critical realism can help to make a difference to the capacity of heterodox economists to retain their own preferred language and so do justice to their substantive insights about the socio-economic world.17 The above, then, constitutes an outline of the project of critical realism in economics. Ultimately, the success and longevity of a methodological project like critical realism depend centrally upon its capacity to appeal to practitioners by making good on its claim to provide insights which can help heterodox economists to develop their own projects in substantive economic analysis. One way of assessing the potential for the critical realist project to be successful in this endeavour is of course to determine how critical realism is viewed by a range of heterodox economists. This brings us to the contributions to the present volume. Stephen Dunn is optimistic that critical realism can serve as a catalyst for the development of Post Keynesian Economics. Dunn identifies two main areas in which critical realism can make a significant contribution. First, critical realism is said to offer a coherent methodological vision for Post Keynesian Economics, according to which the latter is unified by its commitment to a view of the socio-economic world as an open system, in whose working creative human choice, radical uncertainty and (organic) institutions figure prominently, and to the search for causal explanations of socio-economic phenomena of interest. Second, Dunn opines that critical realism can also assist Post Keynesians by deepening their understanding of radical uncertainty. Viewed from the vantage point provided by critical realism, radical uncertainty can be seen to be both a consequence of and a causal influence upon people’s actions: it is a consequence of people’s actions in the sense that it is an historically specific, emergent feature of the socio-economic world which is generated by the interplay between creative human agency and pre-existing social structures such as the accounting and monetary systems; and it is also a

Transforming economics? 19 causal influence on human conduct because the need to deal with their ignorance of the future induces people to behave in particular ways (remaining liquid, following conventions, entering into nominal contracts and so on) which would not be witnessed if people had actuarially certain knowledge of the consequences of their actions. The scope for critical realism to contribute to a more realistic, relevant and fruitful macroeconomics is the subject of John Smithin’s chapter. Smithin sets out his stall against orthodoxy economics’ insistence that macroeconomic theory possess rigorous microfoundations, with macroeconomic phenomena being explained solely in terms of the (inter)action of isolated, atomistic individuals. Smithin argues that critical realism’s affirmation of the ontological irreducibility and explanatory significance of macro-structural phenomena such as social institutions and classes provides the basis for a genuine macroeconomics that is quite in keeping with traditional, heterodox styles of thought. Smithin argues that, in stark contrast to the orthodoxy’s axiomatic-deductive microfoundations project, a macroeconomics informed by critical realism would investigate the ‘macrofoundations of microeconomics’, focusing on transcendental questions of the form ‘What must the overall macroeconomic structure of the economy be like in order for microeconomic phenomena of interest to be possible?’ and attempting to identify the social structures which underpin microeconomic phenomena of interest (as well as how people’s actions lead either to the reproduction or to the transformation of those structures). Smithin illustrates his argument by means of an account of how money, conceptualised as an ontologically irreducible (macro)social relation, facilitates a variety of microeconomic practices, and of how those practices in turn drive the evolution of the monetary system. Edward Nell uses an overview of the ontological issues raised by the analysis of the historical dynamics of capitalist economies as a vehicle for an evaluation of the relative merits of critical realism and his own theoretical framework. As befits someone whose preferred mode of economic analysis is entitled a transformational model of economic growth, Nell argues (consistent with critical realism’s transformational model of social activity) that capitalism is an open system, the development of which is generated by the interplay of social structure and creative human agency. However, while Nell is broadly sympathetic both to the philosophical ontology sponsored by critical realism and also to the transcendental style of argument used to justify it, he contends that there is more scope for the use of formal, closed-system modeling successfully to isolate the causal mechanisms at work in the socio-economic world than critical realists are willing to countenance. Clive Granger offers an econometrician’s view of critical realism. Granger argues that critical realism’s preoccupation with topics such as that of the ‘true’ model is unproductive, arguing that empirical analysis

20 Paul Lewis ought to concentrate on (what Granger sees as) the more reasonable objective of developing a series of improving approximations to the processes through which economic data are generated. Granger maintains, furthermore, that critical realists exaggerate the deficiencies of econometrics. In the first place, they are said to overstate the paucity of empirical event regularities in the socio-economic world. Secondly, according to Granger, critical realists fail to do justice to the panoply of new econometric techniques designed to enable econometricians to deal with the variations in parameters produced by changes in underlying generative mechanisms. The advent of new techniques devoted to dealing with time-varying parameters, regime shifts and structural breaks implies for Granger that contemporary econometrics is much more in line with critical realism than earlier forms of statistical analysis used in economics. Edith Kuiper investigates the compatibility of critical realism and feminist economics. More specifically, she explores the scope for critical realism usefully to inform a variety of aspects of feminist research in economics, most notably the definition of (feminist) economics; feminists’ understanding of (the nature of) gender; the desirability and possibility of a feminist epistemology; the impact of the masculine notion of human nature on the development of economics; and the conceptualisation of power utilised in economic research. Kuiper’s audit of critical realism credits its account of the layered ontology of the socio-economic world with the capacity to clarify and enrich both feminists’ accounts of gender and also their understanding of the nature and scope of feminist economics. She also contends that the transformational model of social activity provides a useful meta-theoretical framework for feminist analyses of the evolution of the discipline of economics. Kuiper illustrates her argument by reference to a historical case study, informed by critical realism, of the impact of gender on economic thought. However, moving to the debit side of the ledger, Kuiper is sceptical of critical realism’s account of human nature, arguing that critical realism retains an unreconstructed masculine notion of people in general and of (social) scientists in particular. Kuiper concludes, therefore, that while feminists can indeed profit by taking on board some aspects of critical realism’s ontological orientation, there remains a central place for feminist epistemologies which are more attuned than critical realism to the way in which the conduct and evaluation of social scientific research has been dominated and distorted by a masculine set of values. The starting point for John Davis’s contribution is the view, common among heterodox economists, that people are social beings who are embedded in networks of social and economic relationships. Davis goes on to observe, however, that while there is something of a consensus in heterodox economics about the shortcomings of the orthodox view of people as isolated individual atoms, there is rather less agreement about how the nature of embeddedness is to be understood, with some

Transforming economics? 21 approaches emphasising the impact of social institutions on people’s (inter)actions and others prioritising the influence of social values. Davis argues that critical realism’s account of the structure–agency relationship, supplemented by insights into the formation of people’s self-identities mined from the discipline of social psychology, can help to clarify and develop heterodox economists’ understanding of embeddedness in a fruitful way. Critical realism contributes to heterodox economics by drawing attention to the variety of channels through which social structure and human agency interact with one another, thereby helping heterodox economists to avoid the one-sidedness which sometimes characterises their accounts of the structure–agency relationship, while social psychology’s contribution lies in the way it facilitates a more elaborate account of human subjectivity which portrays people as active beings whose sense of themselves is continually constituted and reconstituted in a socially influenced process of self-evaluation. For Davis, then, the transformational model of social activity provides a way of incorporating key ideas drawn from social psychology into heterodox economists’ understanding of individuals as socially and economically embedded, thereby enabling heterodox economists to develop an account of the structure–agency relationship that both clearly differentiates their approach from that of orthodox economics and is explanatorily powerful. Sean Hargreaves Heap’s assessment of the scope for synergies between heterodox economics and critical realism also centres on the latter’s account of the relationship between social structure and human agency. Hargreaves Heap is sympathetic to the general thrust (if not the precise details) of the argument that critical realists use to justify their claim that the socio-economic world contains structures which are ontologically distinct from people’s actions. The putative irreducibility of social structure is particularly important for heterodox economics, he argues, because it underwrites a richer model of rational action than that sponsored by orthodox economics. For the idea that there are real social structures, possessing the sui generis causal power to shape people’s attributes, underwrites a picture of human nature according to which the scope for instrumentally rational behaviour is only one of a number of capacities which drive human action and one which, moreover, may be offset by the countervailing impact of higher-order aspects of human reason. The existence of particular social norms and rules may induce people to deviate from the pursuit of their own self-interest because conduct which conforms to those rules and norms expresses a person’s commitment to the latter and thus says something about the type of person they are. This notion of situated or, as Hargreaves Heap terms it, expressive rationality provides an important heterodox corrective to the orthodoxy’s emphasis on instrumentally rational behaviour and, according to Hargreaves Heap, has important practical consequences

22 Paul Lewis which critical realists (and heterodox economists more generally) would do well to emphasise. The goal of reviving the social-theoretic analysis of economic affairs is one that critical realism shares not only with much of heterodox economics but also with advocates of the New Economic Sociology. Rejecting the traditional division of labour between economics and sociology, according to which the investigation of the allocation of scarce resources between competing ends is the exclusive preserve of economics, with sociological analysis being confined solely to the investigation of the origins of those ends, New Economic Sociologists argue that the use of explanatory frameworks drawn from sociology can yield insights into the allocation of resources which are not captured by the conceptual tools used by (orthodox) economists. Like advocates of critical realism, supporters of the New Economic Sociology challenge mainstream economics on ontological grounds: they argue firstly that the orthodoxy’s atomistic model of man leaves it unable to do justice to the fact that people are social beings whose identities, values, beliefs and goals, and consequently whose conduct, are all profoundly shaped by the personal relations into which they enter; and secondly that the orthodox approach’s emphasis on equilibrium states renders it incapable of doing justice to the open-ended processes whereby economic institutions are socially constructed. Paul Lewis examines the relationship between critical realism and the New Economic Sociology via a case-study of the work of one of the latter’s founders, namely Mark Granovetter. Lewis argues that critical realism can usefully augment Granovetter’s approach by drawing attention to the fact, to which Granovetter’s preoccupation with personal relations arguably blinds him, that the social structures in which economic activity is embedded have an impersonal as well as a personal dimension, associated with the macro-social distribution of material, discursive and authoritative resources. Lewis illustrates how the ontological insights yielded by critical realism can benefit the New Economic Sociology by demonstrating that an awareness of the impersonal aspects of social structures facilitates a more sophisticated analysis of trust and cooperation than that provided by approaches which focus only on personal relations. Scott Beaulier and Peter Boettke engage with recent critical realist analyses of the Austrian school of economics, focusing in particular on the issue of the social-theoretic foundations of economic analysis. Beaulier and Boettke argue that there is more common ground, and consequently greater scope for gains from trade, between critical realists and Austrians than the former have hitherto realised. Critical realists have in the past argued that Austrians tend to subscribe to a rather impoverished notion of social structure, conceptualising it as being reducible to people’s (intersubjectively shared) meanings. This approach is inadequate, critical realists maintain, because it leaves Austrian economics bereft of the

Transforming economics? 23 conceptual resources required to do justice to the impact on socioeconomic affairs of non-hermeneutic aspects of social structure such as the distribution of access to positions of power and authority. Beaulier and Boettke contend that this criticism is unwarranted, arguing that Austrians do acknowledge that social structure is irreducible to people’s subjective beliefs and attitudes and so eschew the voluntaristic account of the structure–agency relationship which critical realists attribute to them. For Beaulier and Boettke, the source of this confusion stems from Austrians’ commitment to methodological individualism. They argue that, correctly understood, methodological individualism is not an ontological doctrine suggesting that social structures are exhausted by people’s conceptions, but is rather an epistemological position according to which the only way to gain access to and thereby understand socioeconomic phenomena of interest is via (an understanding of) people’s interpretations of the socio-economic world. Once this point is grasped, Beaulier and Boettke conclude, Austrian economics can be seen to be a thoroughly critical realist enterprise. Ben Fine assesses critical realism from the vantage point provided by Marxist political economy. Fine shares critical realism’s emphasis on the openness of the socio-economic world, on the limitations of methodologically individualist, deterministic and equilibrium-based approaches to economic theory, and on the merits of empirical work that goes beyond multivariate analysis. However, he goes on to argue that the scope for critical realism to promote the flourishing of heterodox economics is severely circumscribed, for two main reasons. In the first place, Fine contends that by focusing its critique of orthodox economics on the latter’s methodology, critical realism rests on an illegitimate and strategically ill-advised division between methodology and substantive theory. This dichotomy has strategic shortcomings because, given orthodox economists’ reluctance to engage in meta-theoretical debate, a methodological critique is unlikely to be effective in winning hearts and minds and therefore in shifting the balance of power in the discipline of economics. And it is illegitimate, Fine contends, both because much of what counts as orthodox economics is connected tenuously if at all with deductivism and also because, according to Fine, it prevents critical realists from addressing adequately the question of whether the economy should be regarded as a separate object of study, distinct from (the subjectmatter of) the other social sciences. This brings us to what Fine sees as the second major limitation of critical realism, namely its failure to address the question of how the (current) discipline of economics reflects the broader socio-historical context, and more specifically the (capitalist) mode of production in which it is embedded. A satisfactory analysis of this question, Fine maintains, would require critical realists to confront the socially and historically specific realities of contemporary capitalism, forcing them to deconstruct the substantive concepts of ‘capital’ and

24 Paul Lewis ‘labour’ sponsored by the latter and thereby undermining the division between (abstract) methodology and (concrete, substantive) analysis upon which the project of critical realism in economics currently rests. Jean-Pierre Dupuy compares and contrasts critical realism with the work of the French Intersubjectivist or ‘Économie des Conventions’ school of thought. The French School concurs with critical realism that a critical examination of the philosophical, and more specifically the ontological, underpinnings of orthodox economics has an indispensable role to play in the shift towards a more relevant and fruitful economics. Moreover, the French School and critical realism share similar ontologies, with the architects of the ‘Économie des Conventions’ approach claiming, consistent with critical realism, that the socio-economic world is characterised by the existence of internally related social structures which, possessing the emergent causal power to react back upon and condition the behaviour of the people from whose actions they arose, are both a cause and a consequence of human agency. And, again like critical realists, the French School portrays the principal task of economics as being the identification and illumination of the social structures and mechanisms which condition observable economic phenomena, rather than the elaboration of event regularities. However, Dupuy is sceptical of the way in which critical realists pitch their philosophical critique of orthodox economics at the rather abstract level of the latter’s deductivist methodology. He argues in the first place that it is necessary to descend from the methodological level in order to investigate how the dominance of orthodox economics is rooted in and so reflects the development of modern industrial economies. Dupuy also contends that a critique of the ontological presuppositions of orthodox economics requires closer attention to the substantive problems of economic theory, most notably the various paradoxical results that emerge from rational choice and game theory, than critical realism’s emphasis on method allows. Ugo Pagano introduces an element of reflexivity into the debate by exploring the scope for critical realism to inform an institutionalist analysis of the discipline of economics itself. More specifically, Pagano contends that critical realism offers a valuable conceptual framework for understanding how the institutional structure of academic economics affects the production of knowledge about the economy. Pagano focuses on two of critical realism’s central claims: firstly that there are ontologically irreducible social structures which condition people’s behaviour; and, second and relatedly, that the formal modeling techniques that characterise mainstream economics are inappropriate for studying the socio-economic world. Drawing on the transformational model of social activity, he argues that the institutions of the economics academy influence the behaviour of economists in a variety of ways: by moulding their values and habitual modes of thought and action; by structuring the incentives they face; and by influencing the flows of

Transforming economics? 25 information they receive. He goes on to explain how competition for positions of power and status within the institutions of academic economics, coupled with the inculcation of particular values and habits of thought in students and fledgling academics as they are selected for and socialised into the profession, generates a tendency for the discipline to favour formal modeling (based in particular on the assumption of utilitymaximisation) in preference to other, arguably more fruitful methods of research (which treat the capacity for instrumentally rational behaviour as only one of a variety of forces driving individual action). Mário da Graça Moura’s contribution sees critical realism being put to work in the service of the history of economic thought. Graça Moura examines the work of Joseph Schumpeter, focusing in particular on the oft-heard claim that Schumpeter’s writings contain a number of tensions and inconsistencies. Graça Moura eschews approaches which strive to downgrade the significance of such tensions and inconsistencies by portraying them as being of marginal importance. He also rejects attempts to dissolve the aforementioned tensions and inconsistencies by claiming that at some level of abstraction Schumpeter’s writings can legitimately be viewed as internally consistent. Graça Moura argues instead that critical realism’s emphasis on the ontological presuppositions of social and economic theories gives rise to another, more fruitful perspective on Schumpeter’s work. Rather than attempting to downgrade or to dissolve Schumpeter’s inconsistencies, an emphasis on ontological issues shows how their existence can be explained. Graça Moura attributes the manifest tensions and inconsistencies found in Schumpeter’s work to a clash between Schumpeter’s vision of the socio-economic world as an open system, in which entrepreneurship, innovation, novelty, uncertainty and structural change feature prominently, and his commitment to a deductivist analytical framework which is intrinsically incapable of doing justice to such phenomena. D. Wade Hands situates critical realism within the context of recent developments in the methodology of economics. He argues that critical realism forms part of what he terms ‘the new economic methodology’, a loose collection of approaches which have arisen over the past two decades in response to the widely perceived limitations of positivist and Popperian approaches to the methodology of economics. The latter sought to discover a set of rules for the conduct of economic analysis, conformity with which would guarantee the epistemic status of economic theories. And they did so by using accounts – rational reconstructions – of the history of the natural sciences as a basis for prescriptions about how social science ought to proceed, with little attempt being made to allow for any differences between the natural and socio-economic realms. Hands observes that critical realism departs from such approaches in two main ways. First, while there is clearly a normative dimension to critical realism, it does not purport to provide a set of rules or algorithm for the

26 Paul Lewis discovery of epistemically justified knowledge. Second, critical realists working in economics have systematically reconfigured their approach in order to ensure that it is well-suited to meet the demands of those researchers investigating the socio-economic world. A recognition of the fact that there is a good deal in common between critical realism and other contributions to the ‘new economic methodology’ is significant, Hands argues, both because it will help to alert critical realists to the availability of additional philosophical resources with which they can go about their task of underlabouring for heterodox economics, and also because it potentially opens the door for the wider dissemination of critical realist ideas among the community of economic methodologists and philosophically inclined and informed (heterodox) economists.

Notes 1 2 3

4

I am grateful to Tony Lawson for his helpful comments on an earlier version of this chapter. The term ‘ontology’ is used here to refer to the nature of (what exists in) the world, that is, the nature of being (Harré 1988: 100; Butchvarov 1995: 489). Of course, critical realism is far from being the only significant body of work that falls under the heading of the ontological turn. Another highly noteworthy contribution is to be found in the writings of Uskali Mäki, who in a series of case studies has sought to make explicit and to clarify the ontological commitments of different figures and methodological approaches in economics (Mäki 1990a, 1990b, 1992, 1997, 1998a, 2001a 2001b). Unlike advocates of critical realism, Mäki largely refrains from offering any direct philosophical critiques of the works in question on account of their (newly revealed) ontological presuppositions. Accordingly, and in contrast to critical realism’s prescriptive bent, Mäki’s approach falls under the heading of descriptive metaphysics (Strawson 1959; Mäki 1998b: 408–9). This common emphasis on ontological issues brings out a second important aspect of critical realism that differentiates it from the earlier epistemologically oriented literature in economic methodology, namely the fact that (according to its proponents) critical realism is strongly rooted in (heterodox) economics. Contributors to the earlier, epistemologically oriented tradition of economic methodology took their cue from developments in the philosophy of the natural sciences, most notably the work of growth of knowledge philosophers such as Popper and Lakatos, and consequently neglected the methodological work undertaken by heterodox economists (Dow 1997: 74, 87; Fleetwood 1999: 128–9). In contrast, while contributors to the project of critical-realism-in-economics have had recourse to the work of philosophers, notably Bhaskar (1989, [1975] 1997), in developing their ideas, they have also drawn critically upon the writings of a disparate group of largely heterodox economists (Commons, Hayek, Kaldor, Keynes, Marx, Menger and Shackle, to name but a few), thereby (it is claimed) ensuring that the project remains firmly grounded in the concerns of (heterodox) economists. Moreover, there is a two-way process of interaction between critical realism and heterodox economics in which ideas developed by heterodox economists have been employed to fill lacunae or to correct misconceptions within critical realism. For instance, critical realism lacked a notion of social order, which it acquired by borrowing a (somewhat

Transforming economics? 27

5

6 7 8 9 10 11

12 13

14

transformed) notion of spontaneous order from Hayek (Fleetwood 1995, 1996). It is also noteworthy in this regard that the interest in ontology displayed by the most prominent contributor to the critical-realist project, namely Tony Lawson, was evident in his own substantive research prior to his acquaintance with Bhaskar’s philosophical writings (Kilpatrick and Lawson 1980; Lawson 1981a, 1981b, 1998: 362 n.5). For detailed textual exegeses in support of this claim, see Fleetwood (1995, 1996, 2001), C. Lawson (1994, 1996), T. Lawson (1994, 1995, 2002), Lewis (2002, Forthcoming a, Forthcoming b, Forthcoming c), Lewis and Runde (1999, 2002, 2004), Mearman (2002), Montes (2003), Northover (1999), Pratten (1993) and Runde (2001). The term ‘metaphysics’ is used here in a non-pejorative sense to denote the set of (philosophical) categories in terms of which the ontology of the socioeconomic world is conceptualised (Harré 1988: 100; Butchvarov 1995: 489). For overviews of the transformational model of social activity, see Archer (1995) and Lawson (1997: 30–2, 56–8, 157–73). For applications to economics, see Pratten (1993) and Fleetwood (1995, 1996). For more on the relevant notions of causality, see Lewis (2000). On the necessity of a hermeneutic dimension in social science, see Lawson (1997: 34–5, 200–1, 223–5) and Sayer (2000: 17–18). More extensive discussions of the critical realist metaphysics and of the explanatory framework to which it gives rise are to be found in Archer (1995: 165–344) and Lawson (1997: 157–271). What critical realists claim in arguing that the socio-economic is intransitive is that at any given moment in time social structures pre-exist current agency (including the activity of social researchers), so that socio-economic reality is not simply an artefact of current activity (including social scientific research). For critical realists, then, socio-economic reality is ontologically irreducible to social scientists’ knowledge of it or discourse about it. But this is perfectly consistent with the possibility that social science may yield knowledge which, once made available to its subjects, may induce them to change their subsequent behaviour in a way that leads to the eventual transformation of the initial set of structures (Lawson 1997: 25). Fuller accounts of the critical realist epistemology can be found in Lawson (1997: 191–271; 2001a), Runde (1998) and Lewis (1999). A number of ostensibly heterodox approaches – for instance, neoRicardianism, analytical Marxism and French Intersubjectivism – adopt (often implicitly) a deductivist methodology and are therefore as vulnerable to critical realist criticisms as orthodox economics. See, for example, Pratten (1999), Fleetwood (2001) and Lewis and Runde (2002). In the lexicon of critical realism, a mechanism such as gravity is said to create a tendency for objects such as leaves to fall to the ground, but this tendency may not be fulfilled if the impact of gravity on the object in question is offset by other, countervailing forces such as the wind. The term ‘tendency’, on this view, denotes the fact that in virtue of their internal constitution mechanisms possess the power or capacity to act in particular ways and so to produce or facilitate particular effects. For instance, gravity has the capacity to cause objects to fall to earth and the rules of the highway code make safe driving possible. Such powers can exist whether or not they are triggered; petrol retains its capacity to burn even if it is not ignited, and the rules of the highway code do not lose their capacity to facilitate safe driving just because people choose not to use their cars. Moreover, a mechanism such as gravity is said to act transfactually, which term expresses in critical realist parlance the fact that the mechanism continues to act and influence outcomes even

28

Paul Lewis

though, thanks to the countervailing impact of other mechanisms, its impact may not be straightforwardly manifest in (regular sequences of) observable events. A transfactual statement is not a counterfactual; it does not express what would happen if conditions were different. Rather, it refers to something that is going on and having an effect, even if the actual outcome is also jointly co-determined by other influences. Tendency statements, then, must be interpreted as relating to the operation of real mechanisms which once triggered continue to act whatever the actual outcome. This returns us to the point, made in the main body of the text, that a mechanism such as gravity usually operates in the presence of other structures and mechanisms that can reinforce or negate its workings, and it is the combined effect of these multiple mechanisms that generates the flux of observed phenomena, the events and states of affairs of our ordinary, observable world. This critical realist perspective suggests moreover that causal laws refer not to empirical regularities, as deductivism implies, but rather to the capacity of mechanisms and structures to act in particular ways. On this view, the laws of science describe the activity of generative mechanisms and structures independently of any particular sequence or pattern of events (Lawson 1997: 20–5). 15 It is worth emphasising that the parallels which critical realism suggests exist between the study of nature and of society are to be found at a fairly high level of abstraction, in the mode of causal explanation which can be sustained by the two fields of study, and that the concrete research methods which (on a critical realist account) are appropriate for the study of the natural and socio-economic worlds may well differ. Laboratory experiments exemplify this point. The internal-relatedness of social structure and human agency implies that people’s actions are likely to differ according to the social context in which they are situated. The socio-economic world is not atomistic and results obtained in experiments involving human subjects may well prove to be little more than artefacts of the (artificial) social context provided by the laboratory. Hence, while laboratory experiments are of paramount importance in the natural sciences, they are likely to be less useful in investigating the socio-economic world (Greenwood 1991: 112–33; Siakantaris 2000). 16 For similar views, see Boettke (1994: 611) and Strassmann (1994: 156). 17 For a case study which examines how a commitment to the formal, mathematical dialect can prevent heterodox economists from exploiting the full potential of their insights into socio-economic life, see Lewis and Runde (1999).

References Archer, M.S. (1995) Realist Social Theory: The Morphogenetic Approach, Cambridge: Cambridge University Press. —— (1996) Culture and Agency: The Place of Culture in Social Theory, Revised edition, Cambridge: Cambridge University Press. —— (1998) ‘Social Theory and Analysis of Society’, in T. May and M. Williams (eds) Knowing the Social World, Buckingham: Open University Press. —— (2000a) ‘For Structure: Its Reality, Properties and Powers. A Reply to Anthony King’, The Sociological Review, 48: 464–72. —— (2000b) Being Human: The Problem of Agency, Cambridge: Cambridge University Press.

Transforming economics? 29 Bauer, P.T. (2000) ‘Disregard of Reality’, in P.T. Bauer, From Substance to Exchange and Other Essays, Princeton: Princeton University Press. Bhaskar, R. (1989) The Possibility of Naturalism: A Philosophical Critique of the Contemporary Human Sciences, Second edition, Hemel Hempstead: Harvester Wheatsheaf. —— ([1975] 1997) A Realist Theory of Science, Second edition, London and New York: Verso. Boettke, P.J. (1994) ‘Alternative Paths Forward for Austrian Economics’, in P.J. Boettke (ed.) The Elgar Companion to Austrian Economics, Cheltenham: Edward Elgar. —— (1997) ‘Where Did Economics Go Wrong? Modern Economics as a Flight from Reality’, Critical Review, 11: 11–64. Butchvarov, P. (1995) ‘Metaphysics’, in R. Audi (ed.) The Cambridge Dictionary of Philosophy, Cambridge: Cambridge University Press. Chick, V. (1995) ‘Is there a Case for Post Keynesian Economics?’, Scottish Journal of Political Economy, 42: 20–36. Dow, S.C. (1997) ‘Mainstream Economic Methodology’, Cambridge Journal of Economics, 21: 73–93. Dunn, S.P. (2000) ‘Wither Post Keynesianism?’, Journal of Post Keynesian Economics, 22: 343–64. Fleetwood, S. (1995) Hayek’s Political Economy: The Socio-Economics of Order, London and New York: Routledge. —— (1996) ‘Order without Equilibrium: A Critical Realist Interpretation of Hayek’s Notion of Spontaneous Order’, Cambridge Journal of Economics, 20: 729–47. —— (1999) ‘Situating Critical Realism in Economics’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. —— (2001) ‘What Kind of Theory is Marx’s Labour Theory of Value? A Critical Realist Inquiry’, Capital and Class, 73: 41–77. —— (2004a) ‘Preparing the Ground for a Viable Account of Labour Markets’, mimeo, Lancaster University. —— (2004b) ‘Where do Firms End and Labour Markets Begin?’, mimeo, Lancaster University. Greenwood, J.D. (1991) Relations and Representations, London and New York: Routledge. Harding, S. (1999) ‘The Case for Strategic Realism: A Response to Lawson’, Feminist Economics, 5: 127–33. Harré, H.R. (1988) The Philosophies of Science, Oxford: Oxford University Press. Hodgson, G.M. (1988) Economics and Institutions: A Manifesto for a Modern Institutional Economics, Cambridge: Polity Press. Kilpatrick, A. and T. Lawson (1980) ‘On the Nature of Industrial Decline in the UK’, Cambridge Journal of Economics, 4: 85–102. Lavoie, D. (1994) ‘Introduction: Expectations and the Meaning of Institutions’, in D. Lavoie (ed.) Expectations and the Meaning of Institutions: Essays in Economics by Ludwig Lachmann, London and New York: Routledge. Lawson, C. (1994) ‘The Transformational Model of Social Activity and Economic Analysis: A Reinterpretation of the Work of J.R. Commons’, Review of Political Economy, 6: 186–204.

30

Paul Lewis

—— (1996) ‘Holism and Collectivism in the Work of J.R. Commons’, Journal of Economic Issues, 30: 967–84. —— (1999) ‘Towards a Competence Theory of the Region’, Cambridge Journal of Economics, 23: 151–66. —— (2003) ‘Competence Theories’, in S. Fleetwood and S. Ackroyd (eds) Realism in Action in Organisation and Management Studies, London and New York: Routledge. —— (2004) ‘The Transformational Model of Technical Activity’, mimeo, Cambridge University. —— M. Peacock and S. Pratten (1996) ‘Realism, Underlabouring and Institutions’, Cambridge Journal of Economics, 20: 137–51. Lawson, T. (1981a) ‘Keynesian Model-Building and the Rational Expectations Critique’, Cambridge Journal of Economics, 5: 311–26. —— (1981b) ‘Paternalism and Labour Market Segmentation’, in F.S. Wilkinson (ed.) The Dynamics of Labour Market Segmentation, London: Academic Press. —— (1983) ‘Different Approaches to Economics Modelling’, Cambridge Journal of Economics, 7: 77–84. —— (1994) ‘The Nature of Post Keynesianism and its Links to Other Traditions’, Journal of Post Keynesian Economics, 16: 503–38. —— (1995) ‘Economics and Expectations’, in S.C. Dow and J. Hillard (eds) Keynes, Knowledge and Uncertainty, Aldershot: Edward Elgar. —— (1997) Economics and Reality, London and New York: Routledge. —— (1998) ‘Clarifying and Developing the Economics and Reality Project: Closed and Open Systems, Deductivism, Prediction, and Teaching’, Review of Social Economy, 56: 356–75. —— (1999a) ‘Critical Issues in Economics as Realist Social Theory’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. —— (1999b) ‘What Has Realism Got to Do with It?’, Economics and Philosophy, 15: 69–82. —— (1999c) ‘Developments in Economics as Realist Social Theory’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. —— (1999d) ‘Connections and Distinctions: Post Keynesianism and Critical Realism’, Journal of Post Keynesian Economics, 22: 3–14. —— (1999e) ‘Feminism, Realism and Universalism’, Feminist Economics, 5: 25–59. —— (2000) ‘Evaluating Trust, Competition and Cooperation’, in Y. Shionoya and K. Yagi (eds) Competition, Trust and Cooperation: A Comparative Study, New York, Berlin and Tokyo: Springer Verlag. —— (2001a) ‘Economics and Explanation’, Revue Internationale de Philosophie, 217: 371–93. —— (2001b) ‘Mathematical Formalism in Economics: What Really is the Problem?’, in P. Arestis, M. Desai and S. Dow (eds) Methodology, Microeconomics and Keynes: Essays in Honour of Vicky Chick, London and New York: Routledge. —— (2002) ‘Should Economics Be an Evolutionary Science? Veblen’s Concern and Philosophical Legacy’, Journal of Economic Issues, 26: 279–92. —— (2003) Reorienting Economics, London and New York: Routledge. Layder, D. (1997) Modern Social Theory: Key Debates and New Directions, London: UCL Press.

Transforming economics? 31 Lewis, P.A. (1999) ‘Metaphor and Critical Realism,’ in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. —— (2000) ‘Realism, Causality and the Problem of Social Structure’, Journal for the Theory of Social Behaviour, 30: 249–68. —— (2002) ‘Social Economics: A View from Cambridge’, paper presented at the 28th Annual Conference of the Eastern Economic Association, Boston, USA. —— (Forthcominga) ‘Boettke, the Austrian School, and the Reclamation of Reality in Modern Economics’, forthcoming in The Review of Austrian Economics. —— (Forthcomingb) ‘Structure and Agency in Economic Analysis: The Case of Austrian Economics and the Material Embeddedness of Socio-Economic Life’, forthcoming in J.B. Davis, A. Marciano and J.H. Runde (eds) The Elgar Companion to Economics and Philosophy, Cheltenham: Edward Elgar. —— (Forthcomingc) ‘Structure, Agency and Causality in Post-Revival Austrian Economics: Tensions and Resolutions’, forthcoming in the Review of Political Economy. —— and J.H. Runde (1999) ‘A Critical Realist Perspective on Paul Davidson’s Methodological Writings on – and Rhetorical Strategy for – Post Keynesian Economics’, Journal of Post Keynesian Economics, 22: 35–56. —— (2002) ‘Intersubjectivity in the Socio-Economic World: A Critical Realist Perspective’, in E. Fullbrook (ed.) Intersubjectivity in Economics: Agents and Structures, London and New York: Routledge. —— (2004) ‘Lachmann, Institutions and the Possibility of Socio-Economic Order: A Reconstruction’, mimeo, Cambridge University. McKenna, E.J. and D.C. Zannoni (2000–1) ‘Post Keynesian Economics and Nihilism’, Journal of Post Keynesian Economics, 23: 331–47. Mäki, U. (1990a) ‘Mengerian Economics in Realist Perspective’, in the Annual Supplement to History of Political Economy, 22: 289–310. —— (1990b) ‘Scientific Realism and Austrian Explanation’, Review of Political Economy, 2: 310–44. —— (1992) ‘The Market as an Isolated Causal Process: A Metaphysical Ground for Realism’, in B. Caldwell and S. Boehm (eds) Austrian Economics: Tensions and New Developments, Dordrecht: Kluwer Publishers. —— (1997) ‘Universals and the Methodenstreit: A Reexamination of Carl Menger’s Conception of Economics as an Exact Science’, Studies in the History and Philosophy of Science, 28: 475–95. —— (1998a) ‘Is Coase a Realist?’, Philosophy of the Social Sciences, 28: 5–31. —— (1998b) ‘Realism’, in J.B. Davis, D.W. Hands and U. Mäki (eds) The Handbook of Economic Methodology, Cheltenham: Edward Elgar. —— (2001a) ‘The Way the World Works (WWW): Towards an Ontology of Theory Choice’, in U. Mäki (ed.) The Economic World View: Studies in the Ontology of Economics, Cambridge: Cambridge University Press. —— (2001b) ‘Economic Ontology: What? Why? How?’, in U. Mäki (ed.) The Economic World View: Studies in the Ontology of Economics, Cambridge: Cambridge University Press. Marsh, D. (2002) ‘Pluralism and the Study of British Politics: It is Always the Happy Hour for Men with Money, Knowledge and Power’, in C. Hay (ed.) British Politics Today, Cambridge: Polity Press.

32

Paul Lewis

Mearman, A. (2002) ‘To What Extent is Veblen an Open-Systems Theorist?’, Journal of Economic Issues, 36: 573–80. Montes, L. (2003) ‘Smith and Newton: Some Methodological Issues Concerning General Equilibrium Theory’, Cambridge Journal of Economics, 27: 723–47. Northover, P. (1999) ‘Evolutionary Growth Theory and Forms of Realism’, Cambridge Journal of Economics, 23: 33–63. Porpora, D.V. (1998) ‘Four Concepts of Social Structure’, in M.S. Archer, R. Bhaskar, A. Collier, T. Lawson and A. Norrie (eds) Critical Realism: Essential Readings, London and New York: Routledge. Pratten, S. (1993) ‘Structure, Agency and Marx’s Analysis of the Labour Process’, Review of Political Economy, 5: 403–26. —— (1998) ‘Marshall on Tendencies, Equilibrium and the Statical Method’, History of Political Economy, 30: 121–63. —— (1999) ‘The “Closure” Assumption as a First Step: Neo-Ricardian Economics and Post-Keynesianism’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. Rizzo, M.J. (1992) ‘Afterword: Austrian Economics for the Twenty-First Century’, in B.J. Caldwell and S. Boehm (eds) Austrian Economics: Tensions and New Directions, Dordrecht: Kluwer Academic Publishers. Runde, J.H. (1993) ‘Paul Davidson and the Austrians: Reply to Davidson’, Critical Review, 7: 381–97. —— (1998) ‘Assessing Causal Economic Explanations’, Oxford Economic Papers, 50: 151–72. —— (1999) ‘On Popper, Probabilities and Propensities’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. —— (2001) ‘Bringing Social Structure back into Economics: On Critical Realism and Hayek’s Scientism Essay’, Review of Austrian Economics, 14: 5–24. Samuels, W.J. (1997) ‘The Crisis of Vision in Modern Economic Thought: A Review’, Advances in Austrian Economics, 4: 133–47. Sayer, A. (2000) Realism and Social Science, London: Sage. Siakantaris, N. (2000) ‘Experimental Economics under the Microscope’, Cambridge Journal of Economics, 24: 267–81. Strassmann, D.L. (1993a) ‘Not a Free Market: The Rhetoric of Disciplinary Authority in Economics’, in M.A. Ferber and J.A. Nelson (eds) Beyond Economic Man: Feminist Theory and Economics, Chicago: University of Chicago Press. —— (1993b) ‘The Stories of Economics and the Power of the Storyteller’, History of Political Economy, 25: 147–65. —— (1994) ‘Feminist Thought and Economics; Or, What do the Visigoths Know?’, American Economic Review Papers and Proceedings, 84: 153–8. Strawson, P. (1959) Individuals: An Essay in Descriptive Metaphysics, London: Methuen. Sztompka, P. (1993) The Sociology of Social Change, Oxford: Blackwell.

2

Transforming Post Keynesian economics Critical realism and the Post Keynesian project Stephen P. Dunn

Introduction1 Post Keynesian economics has many critics. To many it represents a loose association of ignorant and incoherent idiot savants who know what they are against but do not know what they are for. Unable to do the sophisticated maths employed by the mainstream, Post Keynesians carp from the sidelines, engaging in ritualistic debunking, without seeking to make any positive contribution to economic theory or to the study of economic phenomena. The heart of the problem is that many Post Keynesian economists are the slaves of a cadre of defunct economists; while earlier generations of Post Keynesian economists posed interesting questions, the interesting features of their work have now been internalised and superseded by recent advances in formal theory, reducing contemporary Post Keynesianism to the status of a degenerating research programme. The degeneracy of Post Keynesianism is exemplified, its critics argue, by its lack of coherence. Post Keynesian economics has been around for nearly thirty years but there is still confusion in the minds of many as to what it constitutes. To be sure progress has been slow and it is only over the past decade that Post Keynesians have begun to realise that they must seek to codify the key tenets of their research programme if they are to maintain their viability well into the next century. As Hodgson (2001: 230–1) has remarked: in the early years, Post Keynesianism lacked any developed methodological foundations … [However n]ot only was Post Keynesianism originally founded on weak and undeveloped methodological foundations, but also, by the close of the century, ‘Post Keynesian’ economics had still failed to provide itself with an agreed and sufficient set of common core principles around which dissidents could gather. This omission might well prove fatal.

34 Stephen P. Dunn At the same time the project systematised as critical realism is having an increasing impact within economics. The aim of this project has been to lay bare the methodological foundations of mainstream economics and argue for a new approach which recognises that the core problem faced by social scientists is that their subject matter rarely approximates the conditions of experimental control and that an alternative explanatory approach is required. An increasing number of people have contributed to this project and dialogues with other methodologists and various schools of substantive economic thought have been established (see Fleetwood 1999). Proponents of critical realism have been keen to stress the relevance of developing and engaging with other approaches to economics and economic methodology, arguing that critical realism possesses strong affinities with various heterodox traditions in particular (see Lawson 1994, 1997; Fleetwood 1999). Our aim here is to explore how critical realism can act as a catalyst to quicken the systematisation of the Post Keynesian research programme and enable it further to progress. The basis for this assessment is that, on the face of it, Post Keynesian economics exhibits many commonalities with critical realism. Like many critical realists, Post Keynesians oppose much of the empty formalism within the mainstream that masquerades as economic analysis and theory. Similarly many Post Keynesians are keen to emphasise the methodological distinctiveness of the Post Keynesian project and stress the importance of reflecting on such issues. A key aspect of such deliberations is the contention that the mainstream methodology conceals the impact of historical processes for theory and analysis and obscures the salience of uncertainty for economic outcomes. Moreover, there is the related point that while purporting to be the science of choice, orthodox economics effectively denies the reality and significance of real human choice. Such affinities and similarities can be seized upon to clear up some foundational issues in Post Keynesianism that currently lie unresolved. Informed by critical realism, it is possible to argue that the coherence of Post Keynesian economics lies principally at the methodological level. That is to say, Post Keynesians adopt a similar non-deterministic approach to economics to that of critical realists, one that can be understood in terms of the elaboration of the critical realist distinction between closed and open system theorising (Arestis, Dunn and Sawyer 1999; Dunn 2000). Post Keynesianism exhibits a coherent methodological vision, which can be summarised as a commitment to a nonergodic, opensystems mode of theorising which enables economists to take uncertainty seriously. However, noting these affinities and similarities gives rise to a series of important and urgent questions that need to be addressed if Post Keynesians are to continue to make progress in their quest to develop an alternative approach to economics to rival the dominant mainstream

Critical realism and the Post Keynesian project 35 approach (Lawson 1999). Firstly, is Post Keynesian economics anything more than a philosophical position? Is it just an inferior version of, or forerunner to, critical realism that has since been superseded by the systematic elaboration of the latter by Lawson (1997)? Are there some substantive assessments shared by all Post Keynesians, but not necessarily held by those who adopt a critical realist perspective? Similarly, if critical realism connects competing heterodox traditions, then what distinguishes them? The purpose of this chapter is to assess further how (if at all) critical realism can assist the continued development of the Post Keynesian project. We argue that it is only possible to do full justice to Post Keynesian insights about money, unemployment, effective demand, and political and economic institutions (e.g. nominal wage contracts) within an open-systems framework like that outlined by critical realism. Embedding Post Keynesianism in the vision of an open world promulgated by critical realism can help Post Keynesians to develop their ideas about money, contracts, conventions, effective demand and so on. Moreover we argue that the unifying recognition of the salience of money, radical uncertainty, institutions and the principle of effective demand is embedded in this ontological view. The structure of the chapter is as follows. In the subsequent section we discuss the nature of Post Keynesianism, examining its genesis and outlining its central themes. We then consider the nexus between critical realism and Post Keynesianism. We suggest that Post Keynesianism is coherent at the methodological level. More specifically, Post Keynesianism is committed to an open-systems mode of theorising which is characterised by an emphasis on agency, transformation, organic interdependence and causal explanation. We then highlight how the central themes of monetary accumulation under conditions of uncertainty and their relationship to the principle of effective demand are linked to the open-systems ontology that Post Keynesians espouse. But first we must start by addressing a key question: what is Post Keynesianism?

Post Keynesian economics The term Post Keynesianism is a relatively new one, having emerged in the early 1970s (Lee 2000, 2001). It has generally been viewed as synonymous with the vision and ideas of heterodox Cambridge economists such as, among others, Kaldor and Robinson, who formed the vanguard of the Keynesian revolution. These non-orthodox economists, while making several positive contributions to the development of alternative theories, were generally associated with a critique of orthodox economic theorising. The unintended consequence of this was that the sole unifying characteristic of Post Keynesianism was originally seen by

36 Stephen P. Dunn the economics establishment as its opposition to neoclassical economics (Hahn 1982, 1989; Solow in Klamer 1984). However, since its inception Post Keynesian economics has always been conceived, by its practitioners at least, as a positive programme with the aim of developing theoretical structures that incorporate categories which reflect such generalised features of capitalist experience as money, effective demand, cost-plus pricing, stagnation and unemployment. Indeed, by the mid-1970s Eichner and Kregel (1975: 1294, emphasis added) felt confident enough to claim that a diverse set of ideas, broadly subsumed under the label Post Keynesianism, contained the ‘potential for becoming a comprehensive, positive alternative to the prevailing neoclassical paradigm’. Since this modest start Post Keynesian economists have made farreaching contributions to economic theory, the history of economic thought and the methodology of economics. A number of review articles and volumes have been published on the subject of Post Keynesian economics (see for example Hamouda and Harcourt 1988; Chick 1995; Arestis 1990, 1996) and the Journal of Post Keynesian Economics has been established and continues to thrive. Recently Post Keynesians have felt confident enough to proclaim that Post Keynesianism represents a distinctive approach whose positive contribution has an impressive heritage reaching back to the classical economists (Arestis 1990, 1996; Chick 1995; Arestis, Dunn and Sawyer 1999; Holt and Pressman 2001). However, confusion in the mainstream over what constitutes a Post Keynesian approach to economics still persists.2 Even a leading Post Keynesian such as Dow (1990: 346) has acknowledged that there ‘is a sense of confusion, even among post-Keynesians, as to what postKeynesianism is’. So what exactly is Post Keynesianism? Most attempts at defining Post Keynesianism begin by examining the theories and approaches of those who explicitly accept the label, suggesting that Post Keynesianism is what Post Keynesians say it is. In this vein Lawson (1994) attempted to identify those features which are most prominent in Post Keynesian writings and recognised as being so. Lawson argued that, at a general level, five common features are discernible. Firstly, ‘orthodox economics’ is rejected on the grounds of various perceived deficiencies. Indeed, this was a key feature of the formative years of Post Keynesian economics and a reason why the latter’s claim to represent a progressive alternative approach has been called into question. Secondly, there is a significant emphasis on the methodological distinctiveness of the Post Keynesian approach (see for example Lawson 1994; Chick 1995; Dow 1996). Post Keynesian economics is explicitly realist in that there is an acknowledged desire to provide a clear understanding of actually existing real economies, an ‘economics of the real world’ to paraphrase Davidson’s (1972) seminal work on the

Critical realism and the Post Keynesian project 37 economics of Keynes. Post Keynesians frequently locate the inconsistencies and shortcomings of mainstream economics in their professed methodology (Robinson 1974, 1980; Davidson 1996; Dow 1996, 1997). Indeed, these methodological criticisms seem to reinforce the belief among the mainstream that Post Keynesian economics is concerned solely with negative critique, especially when leading orthodox economists view methodology as a distraction. Thirdly, there is an agreed research agenda that stems from Davidson’s (1982–3: 9–18) characterisation of the Post Keynesian approach as underpinned by six a posteriori propositions: 1 2 3 4 5 6

The economy is a historical process; In a world where uncertainty is undeniable, expectations have an unavoidable and significant effect on economic outcomes; Economic and political institutions play a significant role in shaping economic events; The relevance of the distribution of income and power in the study of economic processes; Real capital is non-malleable, embodies historical decisions and is conceptually distinct from financial capital; and Income effects are more dominant than substitution effects in creating and resolving economic problems.

This represents a unifying agenda. Indeed, a common point of departure is Joan Robinson’s (1956) intended aim of completing a ‘generalisation of The General Theory’. This entails a commitment to the principle of effective demand in the analysis of contemporary economies and links the economics of Keynes to the economics of Kalecki (who independently discovered this principle). This involves the rejection of ‘Say’s Law’ – that supply creates its own demand – in both the short and long run and suggests that the main economic problem is not the scarcity of resources but the scarcity of demand. Moreover the excess capacity observed in many advanced industrialised economies, stylised as high levels of unemployment, is taken as prima facie evidence for this point. Fourthly, Post Keynesians believe that in some sense the (diverse) array of theories collected under the label are potentially complementary and thus can ultimately be presented as a viable alternative to the conventional wisdom (Lavoie 1992; Arestis 1996; cf. Hamouda and Harcourt 1988). Indeed, Eichner and Kregel (1975) explicitly acknowledged this possibility. Eichner wished to present Post Keynesian theory as an integral whole, just as comprehensive and coherent as the neoclassical approach. Lavoie (1992: 1) reaffirms this commitment when he asserts his belief that ‘post-Keynesian economics can be presented within a framework that is just as coherent as the neoclassical framework, and that as a consequence it can offer a viable and coherent alternative to those that

38 Stephen P. Dunn are disenchanted with orthodox economics’. The fifth element of Post Keynesianism generally is the linkage of its approach to the traditions rooted in the economic writings of Keynes, Kalecki and some classical economists (Davidson 1996; Sawyer 1988; Arestis 1996).

Critical realism Critical realism is a philosophical approach to economics that has been systematised in the 1980s and 1990s, seeking to provide insights at a higher level of abstraction than substantive theory. It underlabours for social science and accordingly should not be conceived as a substitute for substantive scientific enquiry. One of the principal contributions that critical realism has sought to make is to identify the essence of mainstream economics. Many critics of mainstream economics, including some Post Keynesians, isolate and focus upon its substantive claims expressed through its theories of rationality and equilibrium. In sharp contrast critical realists such as Lawson (1997: 69–107) contend that mainstream economics is best understood and delineated in terms of its, generally implicit, deductivist, closed-system method. According to this perspective, mainstream economics seeks to elaborate (real or purely imagined) universal event regularities and to develop theoretical explanations of the (general) form ‘whenever event (type) X, then event (type) Y’. This approach applied to economics assumes that closed systems, and the event regularities that typify them, characterise the social realm. Critical realists note that, apart from astronomy, the majority of strict event regularities identified by science have been generated under the conditions of experimental control. Critical realists argue, in sharp contrast to mainstream economists, that the social world is not closed because the phenomena under consideration are transmutable, i.e. economic agents possess the capacity for creative, crucial, substantive choice. As Lawson (1997: 219) points out, ‘given the dependence of social mechanisms upon inherently transformative human agency, where human beings chose their courses of action (and so could have always acted otherwise), strict constancy seems a quite unlikely eventuality’. As a result, even while (event) regularities in the social realm are not ruled out, where they do arise they do so because of the unceasing interaction between the reflexive choices of human agents and the social structures and institutions that underpin them. This implies that such occurrences will be partial and multifaceted and neither predictable nor universal. Moreover, mainstream economics’ use of a closed-system ontology for the study of open processes results in inconsistencies at the level of method, social theory and methodology (Lawson 1997: 5–14, 36–42). In response to the widespread (explanatory) failures of deductivism, critical realism argues that the most philosophically tenable conception of

Critical realism and the Post Keynesian project 39 objective reality is one which is open, structured and governed by causal mechanisms (Lawson 1997: 43–65). Critical realism proposes a reorientation of social enquiry to questions of ontology. It advances a distinct, layered ontology that comprises three levels: the empirical domain of experience and impression, the actual domain of events, and the deep or real domain of structures, powers, tendencies and generative mechanisms. These domains are viewed as distinct and out of phase with each other and as being unlikely to give rise to strict event regularities. Critical realists argue that intransitivity occurs at the level of structure and not at the level of observed event regularities and thus that the aim of science is to identify and explain (retroduce) the structures and mechanisms that characterise the real domain (Lawson 1997: 24–6). This moves beyond the essentially flat ontology exhausted by events and experiences postulated by the deductivist approach. However, although critical realists maintain that, a posteriori, there appear to be clear limits to the possibility of bringing about closure in the social realm, the inescapable hermeneutic moment in the social sciences provides an alternative and compensatory route which facilitates the identification of underpinning causal mechanisms and fundamentally changes the nature of economic enquiry (Lawson 1997: 199–226). From this perspective the objective of social enquiry is to move beyond surface phenomena and develop abstractions that are appropriate and concerned with real social structures (and not artificial constructs). These social structures are not in any sense fixed or immutable but emerge and are reproduced or transformed through the intentions and actions of (economic) agents. As Lawson (1997: 35) notes, ‘the impossibility of engineering, and the absence of spontaneously occurring, closed social systems, necessitates a reliance upon non-predictive, purely explanatory, criteria of theory development and assessment in the social sciences’.

Open systems and transmutability As noted in the introduction, there are many connections between the methodology of Post Keynesian economics and critical realism. Post Keynesians incorporate many of the insights that can be gleaned from the realist perspective into their account of the processes of the social realm. Both critical realism and Post Keynesian economics view social positions, rules and relations as those that are created at any point in time through the shared and mimetically derived actions of agents. Social structure and human agency are recursively related, that is to say each is both a condition for and a consequence of the other (Lawson 1997: 160–73). Post Keynesians typically employ a methodology that explicitly, or implicitly, approximates a realist understanding of science, nature and society. Indeed the critical realist focus on methodology and the associated distinction between open and closed systems is broadly similar to the Post

40 Stephen P. Dunn Keynesian distinction between immutable (ergodic) systems and transmutable (nonergodic) systems. Post Keynesians argue that mainstream models are underpinned by the hypothesis of ergodicity. Ergodicity refers to the property by which the time and space averages that originate and are computed from any data-generating process either coincide for a series of infinite observations or converge as the number of observations increases (with a probability of one) for a finite number. That is to say, averages from past realisations collapse on the objective probability distribution that governs current and future outcomes. Under the ergodic hypothesis the passage of time does not affect the joint probability laws governing processes, so that history and time ultimately do not matter. Post Keynesians reject the axiom of ergodicity. This rejection is founded upon both hermeneutics and scientific realism, and ultimately divests the ‘axiom’ of the theoretical status typically accorded to it. It should be recalled that axioms, according to Hahn (1985: 5) at least, refer to incontrovertible statements of event regularities that are so widely agreed upon as to make additional justification unnecessary (see also Lawson 1997: 86–107). Davidson (1982–3, 1988, 1996) rejects the idea that the acceptance of the ergodicity axiom is a proposition that lies beyond dispute. Davidson’s criticism of the ergodicity assumption is twofold. First, Davidson argues that the fact that most macroeconomic time series are non-stationary – a sufficient but not a necessary condition for the existence of nonergodic processes – provides empirical evidence which contradicts the assumption of ergodicity. Second, and more theoretically, Davidson argues that ergodicity is undermined by people’s capacity for real, crucial choices. For the latter entails that in any given circumstances people could have acted differently than they in fact did, suggesting that stable, stochastic regularities of the sort required for ergodicity are unlikely to obtain. Hence, in discarding the ergodicity axiom, Post Keynesians reject the presupposition of the immutability of economic phenomena over time and thus acknowledge the importance and salience of real choice, fundamental uncertainty and history in economic analysis. The distinction made by Post Keynesians between ergodic and nonergodic processes corresponds closely to the critical realist distinction between closed and open systems. Assuming that observable events can be exhaustively described by some invariant conditional probability distribution is tantamount to saying that the process under consideration exhibits strict event regularities. The principal distinction is, however, that the regularities under consideration are expressed in probabilistic terms, i.e. they are stochastic event regularities. Stochastic process theory, in other words, is just one more (probabilistic) expression of orthodox economics’ commitment to deductivism (Lawson 1997: 17, 69, 76). In sharp contrast to mainstream economists, Post Keynesians oppose the characterisation of the social world as closed and the dogmatic

Critical realism and the Post Keynesian project 41 adherence to a formalistic and deductivist closed-system approach to modeling in the face of compelling evidence that social scientists have failed to find significant event regularities in the social realm. Following Keynes (CW XIV: 289), both Post Keynesians and critical realists recognise that ‘the economic environment is not homogenous over time’. Both Post Keynesians and critical realists accept that economics is a moral science, that economics deals with introspection, values, motives, expectations and psychological uncertainties: One has to be constantly on guard against treating the material as constant and homogenous. It is as though the fall of the apple to the ground depended on the apple’s motives, on whether it is worth falling to the ground, on whether the ground wanted the apple to fall, and on mistaken calculations on the part of the apple as to how far it was from the centre of the earth. (Keynes CW XIV: 300) Consistent with this, and also with critical realism, Davidson seeks to go beneath purely stochastic considerations, identifying the underlying causal mechanisms and emergent properties and structures that generate nonergodic time series. In a much-neglected aspect of his discussion of nonergodicity, Davidson (1982–3: 192) emphasises the link to Shackle’s concept of creative, crucial decision-making, arguing that the existence of crucial decision-making represents a sufficient condition for the existence of nonergodic processes. As we have already seen, where creative human agency is important, simply processing information from the past is likely to provide insufficient information about the course of future events to guide current activity. In linking nonergodicity to Shackle’s concept of crucial decisionmaking, Davidson advocates a broader, more creative view of agency than that contained within mainstream models of human behaviour. Accordingly he has expanded the concept of ergodicity to incorporate non-stochastic processes (Davidson 1991). Deterministic models of decision-making which are elaborated in logical time require Savage’s ordering axiom – the presumption, at least in principle, that agents can make a transitive ordering over all possible outcomes. This involves a preprogrammed future and invokes a substantive rationality that is inconsistent with the creativity which Post Keynesians impute a posteriori to agents. Post Keynesians recognise that it is impossible to form a transitive ordering over a yet-to-be-created future in which circumstances inconceivable at the point of origination emerge. Subsequently, and in response to the numerous meanings that could be imputed to nonergodicity, as well as to encompass developments in complexity and chaos theory, Davidson (1996) has reformulated his discussion in terms of a distinction between immutable and transmutable

42 Stephen P. Dunn economic processes. Immutability encompasses the ergodic and ordering axioms and embodies ‘the presumption of a programmed stable, conservative system where the past, present and future reality are predetermined whether the system is stochastic or not’ (Davidson 1996: 480–1). In models which assume that socio-economic reality is immutable, history is predetermined and choice is neither genuine nor matters. Under such a reformulation immutability refers to attempts to elaborate (real or imagined) universal event regularities and to develop theoretical structures of the general form ‘whenever event (type) X, then event (type) Y’. Thus the Post Keynesian conceptualisation of time closely parallels Lawson’s discussion of closed systems. In contrast the broader notion of transmutability advanced by Post Keynesians encompasses the stochastic discussion of nonergodicity within a creative and emergent conceptualisation of history in which choice is genuine, matters and can make a difference in the long run – not least in affecting liquidity considerations and influencing the employment path of an economy over time. As Arestis (1996: 117) observes, ‘[t]he essence of uncertainty in Post Keynesian economic theory is grounded in a nonergodic, nondeterministic world understood as an open-system’. All Post Keynesians share this vision, and this common ground can be used by them to underpin their claim that Post Keynesian economics is coherent, especially at the methodological level. In summary, the commitment to transmutable open-system theorising is the methodological glue that binds Post Keynesians and critical realists together. Post Keynesians argue that mainstream economics is characterised by a methodology of ergodic, closed-system reasoning which ultimately prohibits a substantive role for money and its associated institutions, historical time and the nature of choice under uncertainty.3 Post Keynesian methodology contains a number of (organically linked) elements which are shared with critical realism (Arestis, Dunn and Sawyer 1999). They are: (i) a transformational conception of human agency which presupposes openness where choice is genuine and matters; (ii) an organicist conception of social reality with its concomitant rejection of atomism and methodological individualism; which (iii) prevents (full) closure and the elaboration of a deterministic (ergodic) account of (maximising) economic phenomena; and hence (iv) the aim of economic science is then explanation and not prediction. Like critical realists, Post Keynesians aim to provide causal explanations of concrete economic phenomena which are couched in terms of discursive, abstract, but nevertheless descriptively adequate statements of the phenomena concerned, as opposed to other approaches that by their own admission represent fictitious idealisations (Hicks 1979; Shackle 1979; Davidson 1980; cf. Lawson 1997). While some further clarification and debate may still be needed, a fundamental methodological vision that characterises Post Keynesian economists can be discerned – a commitment to open-system

Critical realism and the Post Keynesian project 43 theorising, an approach which exhibits, at its core, an emphasis on agency, transformation, organic interdependence and explanation. Thus while, as Lawson (1999: 13) points out, such ‘discussion and debate over connections and limitations are still at an early stage’, it is suggested that Post Keynesian coherence and synergies with critical realism can be found at the methodological level. However, although we can accept that Post Keynesianism and critical realism may share similar ontological perspectives and that this can underpin claims of methodological coherence, the question arises of what (if anything) actually distinguishes Post Keynesianism and critical realism? As Lawson (1999: 9) has asked, is Post Keynesianism ‘ultimately anything more than a philosophical position?’

Post Keynesian economics: an open-systems approach While critical realism is ontologically bold, it is epistemologically cautious. The ‘primary aim of the project of critical realism in economics in particular is to bring ontological considerations (back) into the economics picture and to indicate real possibilities in the social realm, [although] it cannot determine a priori which possibilities are to be actualised in any local context’ (Lawson 1999: 7). The project systematised as critical realism does not in itself seek to uncover or investigate the specific structures and emergent processes that warrant scientific investigation. Proponents of critical realism, however, argue that it is the task of the individual sciences to assess and investigate the specific social structures, totalities and processes that emerge and warrant further examination. Nevertheless it is unlikely that a definitive critical realist account would emerge in these individual sciences. Given that the world is dynamic and open, it is not unsurprising that social scientists utilising a critical realist framework compose different explanatory accounts of certain phenomena. Moreover even if a dominant explanation does emerge, on account of its ability to render intelligible the relevant phenomenon, it does not follow that such a rationalisation is conclusive. All knowledge is provisional, fallible and possibly transient. Science is itself a social enquiry and the conclusions of science are not in any sense fixed or immutable but emerge and are reproduced, or transformed, through the actions of scientists. According to the general critical realist perspective, the objective of economics is to understand and explain the institutions, conventions and social processes by which human agents reproduce and transform society. A Post Keynesian economics informed by critical realism ‘attempts to identify those structures, mechanisms, and so on, that allow us to think about questions of employment, output, growth, inflation, and distribution, and from which we can retroduce scientific statements about observable outcomes to explanations in terms of the causal mechanisms

44 Stephen P. Dunn that facilitated or produced these outcomes’ (Rotheim 1999: 80). Inasmuch as Post Keynesianism can represent a distinct realist approach, it seeks to investigate the specific institutional structures that generate different levels of employment across different epochs, cultures and continents.4 From this perspective Post Keynesian economics can be viewed as a specific mode of enquiry into developed capitalist economies which argues that they are best understood through their financial institutions and the money–credit nexus. At the centre of this methodological core is the principle of effective demand. This principle is embedded in the rejection of the ergodic axiom and a closed-system approach to economic modeling. It embodies the recognition that certain structures and institutions, such as the wage bargain and the civil law of contracts, may have emerged as a sensible response to mitigate the impact of an uncertain future. As Arrow and Hahn (1971: 356–7, emphasis added) have argued: ‘the terms in which contracts are made matter. In particular, if money is the good in terms of which contracts are made, then the prices of goods in terms of money are of special significance. This is not the case if we consider an economy without a past or a future [i.e. in a closed theoretical system] … If a serious monetary theory comes to be written, the fact that contracts are made in terms of money will be of considerable importance.’ Closed theoretical models such as general equilibrium constructions, however, entail an economy where money is merely a neutral numeraire which does not impinge on real production and consumption decisions. In contrast, an open-systems approach to economic theorising provides a foundation for ‘serious monetary theory’ by emphasising the need for contracts made in terms of non-neutral money. In the absence of conditions of ergodicity or closure, i.e. the absence of strict event regularities, stochastic or otherwise, agents are truly uncertain as to the future. In such an uncertain, open world, if society organises its production and trading relationships via the use of nominal contracts, then we have what Keynes referred to as an entrepreneurial system, where money is used to settle all obligations that arise in production and exchange agreements (Davidson 1972, 1994; Rotheim 1999). Such a perspective recognises the fact that production takes place through historical time and that in contemporary capitalism money-denominated contracts facilitate the efficient organisation of time-consuming production (and exchange) activities by providing some assurances to the contracting parties as to the delivery of future commitments (Davidson 1988: 333). Similarly such contracts provide information as to the likely relationship between the current and future costs and commitments and allow agents to plan in the context of an uncertain future. Moreover liquid assets such as money help people deal with uncertainty by providing them with the ability to respond to unforeseen contingencies and discharge contractual commitments when they arise.

Critical realism and the Post Keynesian project 45 Post Keynesians recognise the importance of specific entrepreneurial institutions – particularly the civil law of contracts – and the use of money in mitigating the impact of uncertainty and allowing agents and organisations to plan for the future (Davidson 1972; Rotheim 1999). This implicitly suggests an overlap with the (older) institutionalists (see Dunn 2000), but draws attention to the Post Keynesian interest in the specific institutions that generate different levels of employment within different periods in history. Classical theory, which requires neutral money and recontracting without penalty in order to maintain a closed theoretical schema, is clearly incompatible with Post Keynesian economics. It is the recognition of the salience of uncertainty and of historical time as they bear upon the decision to act which, in conjunction with the essential properties of money, namely a zero elasticity of both production and substitution, represents a necessary and sufficient condition for the emergence of unemployment. As the future becomes more uncertain, individual agents deciding to hold non-reproducible, non-substitutable, liquid assets, i.e. money, are then able to defer entering into contractual commitments with other agents and are saved to make a decision another day. However while such decisions to hold money are sensible from an individual perspective, they do not necessarily make sense from the point of view of the whole economy. Unemployment develops as the demand for liquidity expands in response to an increase in dispersed ‘fundamental’ uncertainty. As Hahn (1977: 31, 39) stated, an unemployment equilibrium can occur as long as there are ‘resting places for saving other than reproducible money … if we study an economy which is not a barter economy … then any non-reproducible asset allows for a choice between employment-inducing and non-employment inducing demand.’ It is this recognition of the substantive role for money in a credit production economy in which goods buy money and money buys goods, but goods do not buy goods, that undermines the a priori demonstration of the effectiveness of the market in co-ordinating economic activity. Davidson (1996: 20) notes correctly that Arrow and Hahn (1971: 361) have demonstrated that all general (full employment) equilibrium existence proofs are jeopardized in a world with fixed money contracts over time; no general equilibrium may exist. A nonergodic setting provides the analytical basis for the use of fixed money contracts and therefore provides for the possibility of the existence of long-period unemployment equilibrium – and the possibility of the non-existence of a general (i.e., full employment) equilibrium in the absence of deliberate government policy to assure there is never a lack of aggregate effective demand.

46 Stephen P. Dunn It is for these reasons that Post Keynesians develop theoretical structures that highlight how political and economic institutions can act as a channel for effective demand and influence the amount of activity in an economy and why they explicitly consider the distributional and normative implications of their analysis (Arestis and Sawyer 1998, 1999). It is the openness or transmutability of the socio-economic world, set in historical time and displaying radical uncertainty, which, in conjunction with the essential properties of money avarice, underpins Post Keynesian explanations of unemployment. Post Keynesian economists have developed theoretical frameworks that highlight how political and economic institutions act as a conduit for effective demand determining the level of activity in an economy, yielding the possibility of involuntary unemployment (Robinson 1962; Davidson 1994). The fact that the solution to such problems, as Keynes noted, lies outside the operations of individuals, relates to identification of the importance of the fallacy of composition (which is committed when a conclusion is drawn about a whole based on the features of its constituents when, in fact, no justification is provided for the inference) and its nexus to a monetary economy existing in historical time.5 As Davidson notes: In a nonergodic [open] system plagued by persistent long-period underemployment equilibrium, society can intelligently control and improve the performance of the economy by developing governmental policies to influence, directly or indirectly, entrepreneurial animal spirits sufficiently to offset agents’ excessive aggregate demand for liquidity. Keynes’ suggestion (CW VII, pp. 377–9) that policies be designed to assure that there is never a lack of effective demand does not require a prescient government. All that is necessary is the development of institutions that act as a balancewheel by providing a guiding influence (a) in stimulating effective demand whenever the private sector displays a propensity to produce a lack of effective demand and (b) in reducing demand whenever over full employment conditions prevail. (Davidson 1996: 21) A Post Keynesian economics informed by critical realism recognises the associated problems with fine-tuning and hydraulic Keynesianism and the importance and influence (sometimes positive, sometimes negative) of government activism and regulation upon entrepreneurial confidence and expectations. While it is clear that neither governments nor individuals are omniscient or omnipotent, it is possible to draw attention to the causal mechanisms and social structures that underpin the ability of governments to pump-prime effective demand and repay and sustain loans and finance deficits above and beyond the level that we might think prudent from an individual point of view.6 That is to say, Post Keynesian

Critical realism and the Post Keynesian project 47 economists have highlighted the fact that governments differ from individual agents in many important dimensions, not least in terms of their democratic accountability, their relative longevity, the fact that they can borrow at rates substantially below those at which individuals can; and their legal authority to levy taxes, print money and enforce contractual (debt) obligations. It is these mechanisms and structures, which are embedded in social conventions and practices, that underpin the identification of the significance of the principle of effective demand. Moreover such caution in policy making recognises that knowledge of the economy is provisional, fallible and historically contingent and is grounded in a retroductive methodology. Nevertheless it does not undermine the a posteriori conclusion that a monetary production economy passing through real time is unlikely to generate a high and stable level of employment and thus there is prima facie evidence that as mass unemployment develops, there also develops a role for what Arestis and Sawyer (1998: 187) refer to as ‘coarse’ tuning, whereby ‘over the medium term the government seeks to ensure an adequate level of aggregate demand through the usual channels of the balance between taxation and government expenditure and … through the balance between savings and investment’.

Uncertainty as an emergent property Nevertheless while the ontological distinction between closed and open systems is broadly similar to the distinction between immutable and transmutable processes, it is clear that the Post Keynesian discussion of uncertainty can be further extended with reference to the critical realist method. As discussed above, Post Keynesians have been keen to follow Keynes and integrate the fact that individuals cannot know the future into their economic analysis.7 However Keynes’s and much Post Keynesian discussion, while suggestive of a transformative conception of the economic process, does not seek to identify the generative mechanisms and interactions that give rise to uncertainty and reverberate through the state of long-term expectations. Occasionally Post Keynesians appear to employ and treat uncertainty as an a-historical deus ex machina in order to drive a stake into institutional stability. Such an approach, however, leaves the origins and emergence of the salience of uncertainty unaddressed. Why is the future uncertain? Is it just that we simply cannot know the future? Or are there reasons why our knowledge about the future is limited? And how can uncertainty increase over time? By employing the critical realist method, we can conceptualise uncertainty both as a product of human agency and also as having a causal and irreducible impact upon human agency. The critical realist method recognises the salience of the concept of emergence for scientific enquiry. Emergence may be defined ‘as a

48 Stephen P. Dunn relationship between two features or aspects such that one arises out of another and yet, while perhaps being capable of reacting back on it, remains causally and taxonomically irreducible to it’ (Lawson 1997: 63). Uncertainty can be viewed as a dynamic emergent process of interaction between pre-existing social structures, such as money and accounting practices, and human agency, which is reproduced and transformed in significance over time.8 The reality of uncertainty, as it bears down upon economic action, is embedded in the inherited practices of the present and reproduced and transformed in its contemporaneous significance over time. Conceptualising uncertainty in this way allows us to understand the fact that it is only comparatively recently that the notions and reality of risk and uncertainty have emerged into everyday discourse and become themselves objects of scientific (real) enquiry. The widespread recognition and usage of forward-looking epistemological concepts such as risk and uncertainty are comparatively recent phenomena. Except for some marginal instances in the Middle Ages there was no concept of risk or uncertainty. Giddens dates the emergence of the notion of risk to the sixteenth and seventeenth centuries. He argues that Western explorers first used it as they journeyed off into the unknown in search of new continents, cultures and experiences in the pursuit of riches and wealth. The term ‘risk’ seems to have entered the English language through Spanish or Portuguese, where it was used to refer to sailing into uncharted waters. Originally, in other words, it had an orientation to space. Later, it became transferred to time, as used in banking and investment – to mean calculation of the probable consequences of investment decisions for borrowers and lenders. It subsequently came to refer to a wide range of other situations of uncertainty. (Giddens 1999) It is the forward-looking nature of monetary accumulation that characterises the emergence of capitalism, giving rise to the increasingly widespread usage of future orientated concepts such as risk and uncertainty, simultaneously underscoring their reality. As Weber (1930: 76) pointed out, the chief characteristic of the modern capitalist economy is that it is rationalised on the basis of rigorous calculation, directed with foresight and caution towards the economic success which is sought in sharp contrast to the hand-to-mouth existence of the peasant, and to the privileged traditionalism of the guild craftsman and of the adventurers’ capitalism, orientated to the exploitation of political opportunities and irrational speculation.

Critical realism and the Post Keynesian project 49 What is more, the emergence of such epistemological concepts reflects the embryonic reality of uncertainty. The emergence of risk and uncertainty as salient scientific categories within contemporary society is embedded in the emergent novelty that arises via social structures and conventions, such as the invention of double entry bookkeeping. In modern industrial societies, the term ‘uncertainty’ is typically employed to refer to situations which we have very little historical experience of confronting. Uncertainty finds its salience and genesis in the creative, emergent processes associated with competition and capital accumulation. As Schumpeter (1942: 83) observed, ‘[t]he fundamental impulse that sets and keeps the capitalist engine in motion comes from the new consumers’ goods, the new methods of production or transportation, the new markets, the new forms of industrial organisation that capitalist enterprise creates’. It is such nascent novelty, which occurs as a result of the creative actions of some agents, that generates radical uncertainty for others, especially around the decision to invest (Shackle 1972). As Cliffe Leslie – one of the leading figures in the English historical school – recognised, variety, disruption and uncertainty were themselves embedded in the corrupting pursuit of avarice: The desire for wealth, or of its representative – money – instead of enabling the economist to foretell values and prices, destroys the power of prediction that formerly existed, because it is the mainspring of industrial and commercial activity and progress, of infinite variety and incessant alteration in the structure and operations of the economic world. (Leslie 1888: 223) The evolving reality of risk and uncertainty is closely intertwined with the social structures and conventions that underscore the calculation of profit and loss and concomitant accounting processes and practices which fuel the pursuit of avarice and the creatively destructive competitive process. And whereas some structures and conventions pertain to particular macroeconomic contexts, others are more localised, such as the particular positions and associated practices, rules and relations generated and found within particular microeconomic institutions and processes. Uncertainty emerges as the outcome of the very impact of our transformative actions upon the world. Of course the individual elements generating uncertainty are multi-layered and cross-cutting but it is clear that we can identify specific social structures that fuel the incessant drive to introduce new novel product innovations and technological revolutions which generate uncertainty.

50

Stephen P. Dunn

Conclusion In summary, critical realism can be viewed as part of a more general philosophical project aimed at understanding the most general statements that can be made about science and social theory. Post Keynesianism can be understood as a critical realist elaboration of the study of macroeconomics under uncertainty, of how society copes with scarcity and uncertainty. Nevertheless while representing different traditions and approaches with different aims and objectives, both Post Keynesians and critical realists are united in the belief that a posteriori the search for sharp empirical regularities in economics is misplaced and both identify the reason for this in their understanding of the distinctiveness of human choice. Both wish to refute the repeated claims of the mainstream that whatever the defects of their programme there is no alternative to their project. Post Keynesian economics is still quite young and while many Post Keynesians have given penetrating accounts of the inconsistencies, fallacies and limitations generated by the mainstream project, they have not always been clear about the foundational source of these ills. This is where the underlabouring role of critical realism can help inform and clarify the development of a retroductive macroeconomics. The codification of the realist approach embodied in Tony Lawson’s Economics and Reality represents a potential catalyst for the Post Keynesian project to achieve a similar foundational breakthrough. There exists a fundamental methodological vision shared by all Post Keynesians who argue that history, time and money matter. Post Keynesians are committed to opensystem theorising, which is characterised by its stress on agency, transformation, organic interdependence and explanation. Critical realism offers the prospect of developing an alternative methodological foundation for such a project. Post Keynesian economics offers the prospect of developing a macroeconomic framework that is consistent with this vision. What is more, the critical realist critique of the generally implicit, deductivist, closed-system method of the mainstream is particularly instructive in explaining the obfuscation of the reality of uncertainty within orthodox economics. The orthodox method of closed-system reasoning misrepresents uncertainty by conceptualising it as a subjective probability embedded within a closed-system conceptualisation of the economic process. That is to say, the method of mainstream modelers precludes the recognition and incorporation of the salience of radical uncertainty into their theoretical schemas, despite the widespread evidence of the absence of strict event regularities within the social realm. Post Keynesians reject the closed, ergodic approach of mainstream modelers as it ultimately renders the concept of choice vacuous and hinders the development of ‘serious monetary theory’. In contrast Post

Critical realism and the Post Keynesian project 51 Keynesian monetary theory is founded on the recognition that the world is in a fundamental sense ‘open’, and that this underscores the recognition of the salience of money-denominated contracts and their nexus to the principle of effective demand. Moreover this rejection of the ergodic axiom distinguishes Post Keynesians not only from orthodox neoclassical theory but also from other schools of thought, which, implicitly or explicitly, invoke such closed-system methods and argue for a predetermined (ergodic) path denying the reality of uncertainty. Nevertheless Post Keynesian economics should not be thought of as the critical realist account of the macroeconomics of credit production monetary economies. It is, however, the strongest and most substantive articulation of the theoretical implications of assuming that the economic environment is not homogenous over time. In the absence of sharp empirical regularities characterising the social realm, decisions about the future will be filled with uncertainty. Post Keynesian economics recognises this fact in elaborating the principle of effective demand.

Notes 1

2 3 4

5 6 7

I would like to thank Tony Lawson, Andrew Mearman, Paul Lewis and the participants of the Workshop of Realism and Economics, King’s College, Cambridge University, for their comments on earlier versions of this work. The usual disclaimer applies. See for example the puzzlement expressed by proponents of neoclassical economics such as Hahn (1982, 1984, 1989), Solow, Blinder and Lucas in Klamer (1984) and Backhouse (1988). We shall discuss this assertion in greater detail below. To quote Tony Lawson (1994: 526) on this: ‘Critical realism per se does not license any particular substantive claim. Of course where critical realism is accepted, the aim will be to discriminate between competing accounts on the basis of explanatory power and the like. However, the extent to which this is possible will be severely dependent upon, amongst other things, the context of analysis. But more generally, all theory is fallible; it is historically specific and potentially transformable. It is, then, not surprising, and it is perfectly desirable, that competing accounts are sought – even if the aim must be continually to seek to determine, and then provisionally at least to maintain, those accounts that provide the adequate (explanatorily powerful) expressions of the relevant aspect(s) of reality.’ See also Lawson (1997: 277–81), who highlights the relationship of policy to emancipatory goals. See Arestis and Sawyer (1997, 1998) for a more detailed elaboration of a set of ‘Keynesian policies for the new millennium’. Indeed the General Theory can be viewed as the exemplification of this recognition. As Coddington (1983: 53; emphasis added) has pointed out, ‘it is not the fact of uncertainty that is important for Keynes’s argument, but rather how individuals are supposed to respond to the fact of uncertainty’. The essence of the General Theory is that ‘our desire to hold money as a store of wealth is a barometer of the degree of our distrust of our own calculations and conventions concerning the future’ (Keynes CW XIV: 116).

52 8

9

Stephen P. Dunn As critical realists point out, human activity appears to depend on networks of social structures, which are pre-existing in the sense that they are the product of the creative actions of agents, made in the past. Everyday activities in Western societies such as speaking, writing, reading and withdrawing monies from automated teller machines are dependent on pre-existing social structures, which constrain and enable action. And as they pre-exist any individual act, they can be viewed as autonomous and potential objects of enquiry. Such structures impinge on current actions as an objective reality that is ontologically distinct from and irreducible to individual subjective beliefs and actions. Uncertainty, as it pertains to action, can thus be conceived as emerging from social structures such as money and a range of profit and loss accounting procedures. As Giddens (1999) has observed, ‘[m]odern capitalism embeds itself into the future by calculating future profit and loss, and therefore risk [and uncertainty], as a continuous process. This wasn’t possible until the invention of double entry bookkeeping in the 15th Century in Europe, which made it possible to track in a precise way how money can be invested to make more money.’

References Arestis, P. (1990) ‘Post Keynesianism: A New Approach to Economics’, Review of Social Economy, 48: 222–46. —— (1996) ‘Post-Keynesian Economics: Towards Coherence, Critical Survey’, Cambridge Journal of Economics, 20: 111–35. —— and M.C. Sawyer (1997) The Relevance of Keynesian Economic Policies Today, London: Macmillan. —— (1998) ‘Keynesian Policies for the New Millennium’, Economic Journal, 108: 181–95. —— (1999) ‘The Macroeconomics of New Labour’, Economic Issues, 4: 39–58. ——, S.P. Dunn and M. Sawyer (1999) ‘Post Keynesian Economics and Its Critics’, Journal of Post Keynesian Economics, 21: 527–50. Arrow, K.J. and Hahn, F.H. (1971) General Competitive Analysis, Edinburgh: Oliver and Boyd. Backhouse, R.E. (1988) ‘The Value of Post Keynesian Economics: A Neoclassical Response to Harcourt and Hamouda’, Bulletin of Economic Research, 40: 35–41. Bhaskar, R. (1978) A Realist Theory of Science, Second edition, London: Harvester Wheatsheaf. Chick, V. (1995) ‘Is There a Case for Post Keynesian Economics?’, Scottish Journal of Political Economy, 42: 20–36. Coddington, A. (1983) Keynesian Economics: The Search for First Principles, London: George Allen and Unwin. Davidson, P. (1972) Money and the Real World, London: Macmillan. —— (1980) ‘Causality in Economics’, Journal of Post Keynesian Economics, 2: 576–84. —— (1982–3) ‘Rational Expectations: A Fallacious Foundation for Studying Crucial Decision-making Processes’, Journal of Post Keynesian Economics, 5: 182–97. —— (1988) ‘A Technical Definition of Uncertainty and the Long-run Nonneutrality of Money’, Cambridge Journal of Economics, 12: 329–37. —— (1991) ‘Is Probability Theory Relevant for Uncertainty? A Post Keynesian Perspective’, Journal of Economic Perspectives, 5: 129–43. —— (1994) Post Keynesian Macroeconomics Theory, Aldershot: Edward Elgar.

Critical realism and the Post Keynesian project 53 —— (1996) ‘Economics and Reality’, Journal of Post Keynesian Economics, 18: 479–508. Dow, S.C. (1990) ‘Post Keynesianism as Political Economy: A Methodological Discussion’, Review of Political Economy, 2: 345–58. —— (1991) ‘The Post Keynesian School’, in D. Mair and A.G. Miller (eds) A Modern Guide to Economic Thought, Aldershot: Edward Elgar. —— (1996) The Methodology of Macroeconomic Thought: A Conceptual Analysis of Schools of Thought in Economics, Aldershot: Edward Elgar. —— (1997) ‘Methodological Pluralism and Pluralism of Method’, in A. Salanti and E. Screpanti (eds) Pluralism in Economics: Theory, History and Methodology, Aldershot: Edward Elgar. Dunn, S.P. (2000) ‘Wither Post Keynesianism?’, Journal of Post Keynesian Economics, 22: 343–64. Eichner, A. and J. Kregel (1975) ‘An Essay on Post-Keynesian Theory: A New Paradigm in Economics’, Journal of Economic Literature, 13: 1293–1314. Fleetwood, S. (ed.) (1999) Critical Realism in Economics: Development and Debate, London and New York: Routledge. Giddens, A. (1999) Reith Lectures – ‘Runaway World’: Lecture 2 – Risk, London: BBC, http://news.bbc.co.uk/hi/english/static/events/reith_99/week2/week2.htm. Hahn, F. (1977) ‘Keynesian Economics and General Equilibrium Theory’, in G.C. Harcourt (ed.) The Microfoundations of Macroeconomics, London: Macmillan. —— (1982) ‘The Neo-Ricardians’, Cambridge Journal of Economics, 6: 353–74. —— (1984) Equilibrium and Macroeconomics, Oxford: Basil Blackwell. —— (1985) ‘In Praise of Economic Theory’, The 1984 Jevons Memorial Fund Lecture, London: University College. —— (1989) ‘Robinson–Hahn Love–Hate Relationship: An Interview’, in G.R. Feiwel (ed.) Joan Robinson and Modern Economic Theory, London: Macmillan. Hamouda, O. and G. Harcourt (1988) ‘Post-Keynesianism: From Criticism to Coherence?’, Bulletin of Economic Research, January. Reprinted in C. Sardoni (ed.) On Political Economists and Modern Political Economy: Selected Essays of Geoff Harcourt, Routledge: London. Hicks, J. (1979) Causality in Economics, Oxford: Basil Blackwell. Hodgson, G.M. (2001) How Economics Forgot History: The Problem of Historical Specificity in Social Science, London: Routledge. Holt, R. and S. Pressman (eds) (2001) The New Guide to Post Keynesian Economics, London: Routledge. Keynes, J.M. (1971) The Collected Writings of John Maynard Keynes, London: Macmillan. Klamer, A. (1984) Conversations with Economists: New Classical Macroeconomists and Opponents Speak Out on the Current Controversy in Macroeconomics, Totawa, NJ: Rowman and Allanheld. Lavoie, M. (1992) Foundations of Post Keynesian Economic Analysis, Aldershot: Edward Elgar. Lawson, T. (1994) ‘The Nature of Post Keynesianism and Its Links to Other Traditions: A Realist Perspective’, Journal of Post Keynesian Economics, 16: 503–38. —— (1997) Economics and Reality, London: Routledge. —— (1999) ‘Connections and Distinctions: Post Keynesianism and Critical Realism’, Journal of Post Keynesian Economics, 22: 3–14. Lee, F.S. (ed.) (2000) ‘On the Genesis of Post Keynesian Economics: Alfred S. Eichner, Joan Robinson and the Founding of Post Keynesian Economics’, Research in the History of Economic Thought and Methodology, 18–C: 9–40. —— (2001) ‘The Social History of Post Keynesian Economics in America, 1971–1995’, Journal of Post Keynesian Economics, 23: 141–62.

54

Stephen P. Dunn

Leslie, T.E.C. (1888) Essays in Political Economy, Second edition, London: Longmans, Green. Robinson, J.V. (1956) The Accumulation of Capital, London: Macmillan. —— (1962) Economic Philosophy, Harmondsworth: Penguin. —— (1974) ‘History versus Equilibrium’, in Contributions to Modern Economics, Oxford: Blackwell, 1978. —— (1980) ‘Time in Economic Theory’, Kyklos, 33: 219–29. Rotheim, R. (1999) ‘Post Keynesian Economics and Realist Philosophy’, Journal of Post Keynesian Economics, 22: 71–103. Sawyer, M. (ed.) (1988) Post Keynesian Economics, Aldershot: Edward Elgar. Schumpeter, J.A. (1942) Capitalism, Socialism and Democracy, London: Allen and Unwin. Shackle, G.L.S. (1972) Epistemics and Economics: A Critique of Economic Doctrines, Cambridge: Cambridge University Press. —— (1979) ‘On Hick’s Causality in Economics: A Review Article’, Greek Economic Review, 1: 43–55. Weber, M. (1930) The Protestant Ethic and the Spirit of Capitalism, London: Allen and Unwin.

3

Macroeconomic theory, (critical) realism and capitalism John Smithin

Introduction Does the methodological project of critical realism (Lawson 1997, 1999a; Fleetwood 1999; Lewis and Runde 2002) have relevance to the discipline of macroeconomics? This chapter argues that it does. This is not to suggest that it has much in common with the modern mainstream approach to the subject. The claim is rather that it connects up with those projects which seek to transform current macroeconomics into a more realistic and relevant discipline.1 One of the major difficulties in pursuing theoretical research on macroeconomics and monetary economics in the contemporary academic environment is the pressure to conform to a narrow and well-defined set of methodological criteria in the presentation of ideas and research results. This pressure is both perceived and real. It is certainly ‘perceived’ by graduate students, for example, as they undergo the educational/socialization process preparing them for careers in academia (Colander and Klamer 1990). It is real in the sense that it is genuinely difficult, most would say impossible, for work which does not conform to the accepted set of criteria to be published in the most prestigious outlets (Tobin 1986: 350), and hence receive attention from the relevant audience. This has consequences for the academic careers of those who are willing to undertake non-conforming research, and therefore on the incentives for scholars to pursue questions which cannot easily be fitted into the accepted framework. Presumably most observers (except perhaps for those insiders committed to the status quo for either careerist or ideological reasons) would accept that this situation is unlikely to further the growth of knowledge in economics or any other discipline.2 Hence, the present chapter will not belabor this point. Rather, it seeks instead to evaluate the methodological criteria mentioned above, and to discuss the impact these have had on the way in which research is actually conducted. It then suggests a substantive approach to macroeconomic processes in a capitalist economy which does not necessarily conform to the ‘approved’

56

John Smithin

criteria, but which nonetheless does have roots in more traditional theory and claims explanatory power, particularly in terms of the political economy of the system. In pursuing these objectives, the chapter examines the extent to which the precepts of critical realism may help either to sustain this specific approach or, more generally, to provide the underpinnings for a revitalization of macroeconomics. The critical realists have been at pains to point out that the role of methodology is to ‘underlabour’ for research activity as such (Lawson 1999b: 3; Lewis and Runde 2002: 211–12), and hence does not sponsor any particular substantive theory. Nonetheless, critical realism does make the claim that there are certain general features of socio-economic reality that any theory must take into account, and this seems best illustrated with reference to a substantive example, whether or not the specific claims put forward are ultimately widely accepted. A final section will additionally discuss the controversial role of mathematical methods in macroeconomic theory (Harcourt 1995; Dutt and Amadeo 1990: 164), as this is likely to be a point of tension between critical realists and research practitioners (Lawson 1994; Rotheim 1999). One of the main conclusions reached is essentially positive. The question raised, and many of the answers given, by the critical realist school on methodology do point towards the development of a more useful macroeconomics than currently exists in the academy. Acceptance of something like the critical realist account of the nature of social reality would seem to make possible a macroeconomics which is not as completely hamstrung, as is contemporary orthodoxy, by obeisance to the axiomatic-deductive microfoundations project.

The microfoundations literature As mentioned, an obvious constraint on the pursuit of macroeconomic research in mainstream academia is the repeated insistence that macroeconomic theory should be based on what are usually called ‘microfoundations’ or ‘micro-underpinnings’ (Harcourt [1977] 1986; Scarth 1996; Turnovsky 1999). At the textbook level the difference between microeconomics and macroeconomics is said to be that between the investigation of the workings of individual markets and that of the economy as a whole. However, this is not the point of the microfoundations literature. A better definition of microeconomics for these purposes would be a pure theory of individual decision-making, essentially divorced from the institutional or social context in which the choice is supposed to take place.3 The insistence on microfoundations, from this point of view, is therefore the insistence that an explanation of macroeconomic phenomena should be based on the logic of the interaction of individual decisions by atomistic agents behaving in a uniform way, without reference to higher-level social structures or groupings. A

Macroeconomic theory, (critical) realism and capitalism 57 commitment to methodological individualism is shared by several schools of thought in economics, including the Austrian School, associated with such writers as Menger ([1871] 1950), Hayek (1988, 1994) and Mises (1978), and also the contemporary mainstream or neoclassical school. There are, however, different interpretations as to what is acceptable economic theory in practice. In particular, the former group rejects mathematics, while the latter embraces it. For the mainstream neoclassical economists, therefore, another methodological obiter dicta, which seems closely connected with the above, is the demand that theory should be ‘rigorous’ (Sargent 1979: xiii; Tobin 1980: x). If this term was used in its most general sense, presumably this would be unexceptionable. However, in contemporary academic economics the application of the term rigor seems to be restricted only to certain types of mathematical optimization problem. In macroeconomics, this is frequently the problem solved by the ‘representative agent’ or ‘representative individual’ (Kirman 1992), a mythical character whose choices supposedly reflect those of the society as a whole and are taken as illustrative of the aggregate of such choices in the macroeconomy. At the operational level, the first-order conditions of the agent’s dynamic maximization problem are taken to be valid mathematical expressions also of aggregative behavior. Money demand or labor supply equations derived from the individual optimization exercise are treated as the corresponding functions at the aggregative level. And, furthermore, no postulated behavior at the aggregate level is regarded as a legitimate hypothesis unless it can be derived from this or some similar exercise. This was not, however, the position taken by the original ‘macro thinkers’ (Nell 1998), such as Keynes (1930, 1936), Kalecki (1971), Kaldor (1956, 1982), Robinson (1956), Lerner (1943), Weintraub (1959), Davidson and Smolensky (1964) and others,4 during the years when macroeconomics was first developing as an academic discipline. Many of these writers were willing to use mathematical expressions of macroeconomic magnitudes, along with other methods. However, their aggregative functions were not usually based on putative microeconomic assumptions, but plausible conjectures or hypotheses as to the behavior of the system as a whole. Examples would be Keynes’s (1936: 246–7) ‘psychological propensities’, presumed differences in the consumption/ saving behavior of different social classes, or institutional data such as the existence of banks or labor unions. But this is anathema to contemporary methodological purists, and whatever the underlying motives of such strictures it is reasonable to suggest that at least one practical consequence of these attitudes is to preclude certain types of questions being asked. It should be made clear, however, that the implied censorship of macroeconomic discussion is actually rather more subtle than either the defenders or opponents of the academic status quo usually allow. It is not usually acknowledged, for example, that given any initial hypothesis

58

John Smithin

regarding aggregative behavior it is always possible (as a purely formal exercise) to work backwards, as it were, and specify particular functional forms for utility functions, maximands and constraints, which in combination will yield the originally hypothesized results.5 The issue is therefore better framed, as I now see it, by saying rather that there is really no point in doing this from the perspective of actually acquiring knowledge about how the system operates. The objective of such exercises can only be such things as (a) communication with colleagues whose training is in the standard methodology,6 (b) the careerist objective of proving to one’s colleagues that one ‘can do the math’, (c) being able to publish in the approved journals, (d) the obligations of graduate student instruction, and so on. The important point which should be made, however, is that this laborious procedure does not actually add anything in terms of discoveries about how the economy works.7

The methodological debate There are diverse reasons behind the emphasis on methodological individualism in conventional economic theory. In terms of macroeconomics the basic idea seems to be that aggregates (such as total investment or total consumption), or such concepts as social classes, are not legitimate subjects for investigation because they can have no independent existence apart from their individual components or members. There is no such thing as a working class, for example, just a number of individual workers with their own goals, ambitions and activities, which may or may not resemble those of their peers. The phrase attributed to the former British Prime Minister Margaret Thatcher that ‘there is no such thing as society’ expresses this point of view in an extreme form. This may be going too far for the majority of academics, but, even if proponents of similar views would not deny the actual existence of social institutions and groupings, they do insist that the evolution and development of these entities can (and should) be explained as the outcome of spontaneous and voluntary choices by atomistic actors pursuing their individual self-interest. One argument made within critical realism is that the microfoundations project is driven ultimately by a commitment to deductive methods of analysis (Lawson 1999b; Lewis and Runde 2002). The derivation of results via deduction requires that certain closure conditions must hold (Lawson 1997: 98–9), which implies inter alia that aggregates must always be reducible to their constituent parts. This is a problem in the social sciences, in which organic or internal relationships between the actors are prevalent. Hence, the methodology chosen is at odds with the material under investigation.8 The deeper motivations causing scholars and others to adopt the worldview described above are presumably also many and varied. The most important would presumably be concerns that any untoward

Macroeconomic theory, (critical) realism and capitalism 59 emphasis on classes or groups is to deny the importance of individual agency, personal responsibility and individual liberty, and hence may ultimately lead to collectivist or totalitarian tendencies in politics. In other words, to take any other view is the Road to Serfdom, as Hayek (1944) put it in the title of his most famous book. All of the above might be described as the ‘legitimate’ motivations for the individualist methodological stance. However, at the sharp end of academic trench warfare, involving editorial decisions at academic journals, hiring, tenure, and promotions procedures, and the allocation of research funding, it is sometimes difficult to believe that any such philosophical reflection is going on. At that level, the outcome of the methodological debate seems to hinge much more on academic politics, peer-group pressure, careerism, cronyism, and other manifestations of the social environment of the economics profession. From the point of view of macroeconomic theory, the consequences for the development of the discipline are more or less obvious. The point is that macroeconomics inevitably falls under suspicion when looked at through the lenses of the orthodox approach to economic methodology. As argued by Ingham (1996), macroeconomics, the underlying objective of which is to discover the ‘law(s) of motion’ (Marx [1867] 1976: 92) of the economic system as a whole, of necessity takes on a somewhat sociological character. This, however, is something akin to sacrilege to true believers in axiomatic neoclassical theory.9 In our own era, there have therefore been conscious attempts to eradicate the topic of macroeconomics, as this would have been understood by such writers as Keynes (1936), Kalecki (1971), or even Samuelson (1948), from the corpus of economics altogether. According to Lucas (1987: 107–8): The most interesting recent developments in macroeconomic theory seem to me describable as the re-incorporation of aggregative problems such as inflation and the business cycle within the general framework of ‘microeconomic’ theory. If these developments succeed the term ‘macroeconomic’ will simply disappear from use and the modifier ‘micro’ will become superfluous. We will simply speak, as did Smith, Ricardo, Marshall and Walras of economic theory. As against this, it can be argued that critical realism does provide guidance as to a more appropriate methodological stance for macroeconomic theory, given its inevitable overtones of social investigation. According to Lawson (1997: 157), who has played a leading role in introducing these notions to economists, the term critical realism encompasses both a general philosophy of science (transcendental realism) and a specific theory of social ontology.10 It is the latter, in particular, which seems relevant to breaking the impasse reached by the microfoundations of macroeconomics literature. This ontology affirms the

60

John Smithin

real existence and the importance of social structures constituted by such things as rules, positions, and relations, and relatively enduring institutions. The basic definition of social structure, however, is a set of ‘continually reproduced inter-dependencies’ (Lawson 1997: 159), meaning that the focus is on the process whereby the structure reproduces itself over time (or is transformed over time) as a result of the mutual interaction, decisions, and activities of its members. Moreover, the positional situations within the structure can be (in fact, are likely to be) internally (organically) related, as well as externally (atomistically) related. The advantage of this approach, from the point of view of thinking about macroeconomics, is that it is able to steer a middle course between the two extremes characterizing much of the discussion of the structure–agency relationship in the social sciences. These are, on the one hand, the reductionism, implicit in the orthodox view discussed above, that social structures result simply from the actions of atomistic individuals, and, on the other, the deterministic view, objectionable on political and other grounds to many, that individual behavior is completely controlled (rather than simply shaped or constrained) by an external social structure such as membership of a social class. The underlying vision is that social structures do exist and are the precondition for human action, but, at the same time, the structures also depend on action. They are not fixed and can be transformed over time through both individual initiative and collective action. The usual illustration given is that of a specific language. The English language, for example, with its rules of grammar and vocabulary, certainly existed before any of the current speakers of English were born. It is not of their creation, and yet to a large extent their actions and thoughts are shaped or conditioned by this pre-existing social institution. Just as obviously, though, the language continues to evolve as time progresses, as a result of both the intentional and unintentional human agency of current and future English speakers. The English language is a real social structure, it is reproduced and has considerable continuity over time. Yet, to recognize this, and to accept that the structure has consequences for the course of events (such as the writing of this chapter), is not to deny the efficacy of the individual embedded in the structure. It is possible for (a) an individual to ‘coin a phrase’ which passes into the language, and (b) for usage to change over time as a result of unconscious collective practice. From the point of view of macroeconomic thought, the argument must be that the development of economic structure has much the same character, and can be thought of in an analogous manner. If accepted, this then appears to completely undercut the insistence on atomistic microfoundations held to have been so deleterious to the development of useful macroeconomic theory in the modern era. Or, putting the point in a more positive way, it makes possible an alternative approach to

Macroeconomic theory, (critical) realism and capitalism 61 macroeconomics which has a more realistic view of the complexities of social relationships. The social world is conceived of as an open system, but it is also structured. Social research would consist of the study of social structure and the ways in which this both constrains and enables human action, and the associated ‘powers, tendencies and mechanisms’ (Lawson 1997: 21) which may be discovered. One of the main objectives would be to uncover those structures, tendencies, and mechanisms which are relatively enduring, which have been reproduced in a recognizable way over some period of time, and can therefore provide some type of theoretical explanation for the actual course of events during that period. Whether or not the critical realists would themselves accept this view, it seems to me that this approach does provide something of a magna carta for macroeconomic theory as this was traditionally conceived.11 Specifically, it establishes the crucial point that the treatment of social classes, groupings, and aggregates does not imply an anthropomorphic stance where human propensities such as will, intention, or purpose are attributed to some collectivity. But, in addition to this defensive stance, it can also be argued that it becomes possible to go even further, and to insist that what is required is not so much the ‘microfoundations of macroeconomics’ but, on the contrary, the ‘macrofoundations of microeconomics’ (Crotty 1980: 23).12 That is to say, the relevant question to ask is what the social structure must be like, or how the social structure has evolved, such that so-called microeconomic rationality, economizing, optimization, rent-seeking, strategic behavior, etc., seem to be a guide to the conduct of some or many of the individuals who find themselves embedded in that structure. If the social structure was different (e.g., an isolated monastic community), presumably the types of conduct required or imposed would also be different. Is homo economicus, in short, a cause or a consequence of the development of the social system of capitalism? This is a basic question for both defenders and opponents of this particular method of economic organization. The idea of searching for the macrofoundations of microeconomics, rather than the other way around, would seem to resonate in particular with the transcendental arguments used by critical realists in their accounts of social ontology. In effect, the general question that is asked is ‘what must the social structure be like for human choice to be possible?’, and the method employed to answer it is that of abduction or retroduction (Lawson 1997: 24), rather than simply induction or deduction. The similar question suggested here is ‘what must the overall macroeconomic structure be like in order for the microeconomic phenomena of interest to the neoclassical economist to be possible?’.13 It is not, after all, a very convincing argument to simply appeal to an inbuilt human disposition to ‘truck, barter, and exchange’ (Smith [1776] 1981: 30).

62

John Smithin

Towards a monetary theory of production By way of illustration, it seems clear that if we are considering the historically specific ‘mode of production’ (Marx [1859] 1970: 20) of capitalism, the social institution or structure most in need of investigation is that of money. By this is meant not only the primary concept of a ‘money of account’ (Keynes 1930: 3) which makes possible price lists, monetary calculation, and accounting, but also the entire apparatus of banks, central banks, and other financial institutions involved in the production of money, its social control, and the granting of credit. From a common-sense point of view, economic activity under capitalism is all about money, making money, spending money, saving money and so forth (Smithin 2000: 1). However, as explained by Ingham (1996, 1999, 2000), the serious study of money from this perspective has apparently been neglected by both mainstream economics and mainstream sociology for reasons that are not entirely coincidental. Ingham (1999: 105) has insisted, on the contrary, that ‘money itself is a social relation’. Like laws, language, customs, and other such things, it is a relation between agents, as opposed to the usual concerns of economics with the relationships between agents and goods (in consumption), or goods and goods (in production). This perspective is congruent with such insights as Schumpeter’s distinction ([1954] 1994: 276–8) between ‘monetary’ and ‘real’ analysis (Rogers 1989: 3–17), and Keynes’s concept of ‘monetary production’ (Rotheim 1999: 90–95). The basic idea of the latter is that the economic system under which we live and that has evolved over the past several centuries is, in fact, pre-eminently a monetary system (Smithin 1994). Those responsible for setting production in train, whether entrepreneurs or corporations, must first acquire monetary resources to do so. The ultimate proceeds of productive activity from the sale of goods and services are also sums of money. Output and employment outcomes depend upon expectations of money receipts relative to cost. Moreover, the ultimate reward structure of the society, and the distribution of power and prestige, also depend on the accumulation of wealth denominated in financial terms. The explanation of this can only be some tacit social arrangement (social structure) entailing that possessors of money, as validated by custom and convention, should have unique claims to the social product. Some writers, such as Winslow (1999) and Dostaler and Maris (2000), have pointed to the psychological motives which give rise to the drive for accumulation and the ‘love of money’, and which were recognized by both Keynes and Freud. These surely do help to explain some features of the genesis of the money system, and also certain pathological behaviors once it has developed, such as ‘manias, panics and crashes’ (Kindleberger [1979] 1989; emphasis added). At the same time, however, once a social structure of this type exists, the behavior of the agents embedded therein,

Macroeconomic theory, (critical) realism and capitalism 63 rational or otherwise, is both constrained and facilitated precisely by that structure. Therefore, in principle, the role of money in capitalism seems no more difficult to understand than (say) the role of physical force in acquiring privileges and prerequisites in a different type of society. However, this perspective is quite different from that of the standard neoclassical approach in which ‘money does not matter’, or ‘money is a veil’ over a system the fundamental basis of which is assumed to be the rational barter transactions of atomistic agents. In such a world money supposedly emerges from the market as an optimizing response to the technical inefficiency of barter (Menger 1892). The concept of money as a social relation, however, implies that, if anything, the reverse is true (Smithin 2000: 4). The market emerges as a consequence of money and monetary practice. It is true that monetary systems, practices, and structures evolve and are transformed over time, as are all other human institutions (Chick 1986; Hicks 1989). However, it is also reasonable to argue that there has been a substantial degree of reproduction and continuity of this particular social institution from the early modern period. This is what justifies discussion of a coherent (albeit evolving) capitalist system in the first place. To put the point concretely, there is little doubt that Lorenzo de Medici, Henry Thornton or Walter Bagehot, if they were brought back to life today, would easily be able to understand the role and functions of contemporary central banks, commercial banks, and financial markets. Moreover, they would surely also understand (as many modern pundits and commentators apparently do not) that the computerization of the payments system, an obvious surface feature of monetary practice today, is primarily a change of form rather than substance. Another relatively enduring feature of capitalist monetary production is the concept of the firm or enterprise. This is also closely connected with the social phenomenon of money. According to Dillard (1988: 300): [A] monetary theory of production requires a theory of business enterprise, which is the dominant economic institution of modern civilization. Money has very special meaning for business. It is both the means and the end of business activity. Dillard (ibid.) also quotes Keynes from an early draft of the General Theory: The firm is dealing throughout in terms of money. It has no object in the world except to end up with more money than it started with. That is the essential characteristic of an entrepreneur economy. In terms of the evolution of social systems, there is an implication that monetary concepts must be antecedent to the development of business

64

John Smithin

firms. However, once given notions of money, there is an evident ‘pecuniary logic’ (Dillard 1988: 300) to the development of specialized social institutions devoted to the practical realization of Marx’s M-C-C1-M1. The business firm itself also evolves, from single proprietors, to partnerships, to limited liability concerns characterized by the separation of ownership from control (Robinson 1956), to the multinational corporation.14 Schumpeter ([1942] 1992: 132) originally made the individual entrepreneur the hero of his particular social drama. Moreover, this view now enjoys a considerable revival, as in the proliferation of business school courses on ‘entrepreneurship’. However, from another perspective, both the ‘innovative entrepreneur’ and the hidebound ‘old economy’ corporation trying to preserve market share have the same object in view. They are both trying to ‘make money’. It is just a question of how best to do this in any given set of circumstances.15 In any concrete economic theory, these nuances about the evolution of the system and its relative dynamism would need to be taken into account and evaluated. The point here, however, is that the element of relative continuity is also present, in that an enduring feature of capitalism is some kind of institution organized for the purpose of literally making money by business activity. The third important institution of capitalism, then, is wage labor. If money is important in the social system in terms of both basic survival and power and prestige, and there is no other avenue for acquiring it, working for money wages in the employ of the business firm is a basic modus vivendi for many (empirically for most) of the actors in the system. Again, it is obvious that labor as an institution also changes over time. For example, labor unions may be weak at some stages of social evolution, then powerful, then weak again. There may be social regulation as to hours of work or minimum wages, which prevails in some periods and not in others. But the basic idea of (having to) work for money wages, whether this is as a laborer or computer programmer, has clearly been an enduring feature of capitalist economics. The discussion thus far has not dealt explicitly with the role of the state, and space constraints prevent an extended discussion of the problems of public finance. However, in adopting (in some sense) a more realistic and organic view of the development of social and economic institutions, it seems clear that the concept of a state which is independent of, and antipathetic to, the ‘natural economy’ is a misleading way of thinking about the relationship between the public and private sectors. The state is inevitably bound up with, and has an internal relationship to, the other social and economic institutions. Therefore the very concept of ‘government interference’ in an otherwise pristine world of laissez-faire is by definition nonsense. The government and the economy must impinge on one another in various ways, as they are part and parcel of the same fabric of social relationships.

Macroeconomic theory, (critical) realism and capitalism 65 As a practical matter the state has the decisive role in the maintenance and operation of the monetary system, identified above as one of the key social institutions of capitalism, and in money creation. Modern ‘neochartalists’, such as Wray (1998), have revived the theory of Knapp ([1924] 1973) suggesting that money is a ‘creature of the state’ (Wray 1998: 12). Keynes (1930: 4) put this somewhat more cautiously when he prefaced a discussion of chartalism with the phrase ‘the state or community’. What seems indisputable is that in the modern world control over the monetary system is also a prerequisite for control of the other levers of policy such as fiscal policy, social policy, and labor legislation (Smithin 1999). For example, consider the contemporary situation of the formerly independent European countries which have adopted the single currency or Euro. They have clearly given up control over money, but have inevitably found that this is bundled up with a loss of control over the other policy instruments also (Marterbauer and Smithin 2000). This shows that the state itself evolves over time as other social institutions do. Forces exist within contemporary capitalism pushing for a weakening of the traditional nation state, and its replacement by regional or global entities, which are larger but perhaps less powerful in a sense relevant to the business/financial interests. The process can be connected with the underlying political economy of the system and the issue of which groups are to have a decisive say in the making of macroeconomic policy.

The political economy of the macroeconomic system Pursuing the theme of the relative continuity of some of the underlying social structures, one of the most informative sketches of the basic political economy of the system remains that by Keynes in his Tract on Monetary Reform (1923: 5–40). Keynes examined the impact of unanticipated inflation on three broad strata of society. These were an ‘investing class’ of rentiers or financial capitalists, a ‘business class’ of entrepreneurs, manufacturers, and merchants, and the ‘earner’ or labor. These obviously correspond fairly closely to the underlying social institutions of capitalism identified above. And this is the point, also, at which the hard-line methodological individualism of contemporary neoclassical economics would be supposed to preclude any useful modern equivalent to an approach along these lines. Keynes, however, anticipated the neoclassical objection that the threefold class division of income does not necessarily correspond to the individual distribution. He conceded (1923: 5) that ‘the same individual may earn, deal and invest’, but nonetheless thought it important to investigate the effects of inflation on each of the different income sources, particularly in terms of the incentives to engage in each activity. It may also be true that the particular class divisions suggested by Keynes are now obsolete, the role of the investing class, for example,

66

John Smithin

being played by pension funds and other institutional investors, and that of the business class by large corporations. If there is a basic continuity of capitalistic economic methods, however, meaning that Keynes’s environment eighty years ago sufficiently resembles our own, we can still discuss the effects of inflation on the incentives for the provision of financial capital, the organization of productive activity, and the supply of work effort respectively. Keynes’s argument was that inflation need not be of much concern to either non-financial business or labor. In times of rising prices labor has little difficulty in securing wage increases to match, while non-financial business may actually gain as the money value of stocks of finished goods and work-in-process appreciates on their hands. The financial interests, however, whose wealth is denominated in terms of money, do have a strong aversion to inflation. Strangely enough, given Keynes’s later reputation in monetarist circles, he actually came out in favor of price stability as the best means to mediate these social conflicts. However, a crucial point was glossed over in the original discussion and continues to be in the contemporary debate. Although the usual focus is on the impact of inflation as such, in other words the surface phenomenon, on a closer look it becomes clear that the underlying cause of conflict is actually the real interest rate changes that must occur as part and parcel of any attempt to change inflation by monetary means (the actual monetary policy instrument being the nominal rate set by the central bank). In the case of the business class, for example, for there to be any genuine profiteering from inflation, the real rate of interest must fall. Otherwise, in a continuing production process, refinancing next period’s output at the new higher prices would eliminate the nominal profits obtained by selling today’s goods at similarly higher prices. In the case of the earner, Keynes’s argument was that labor can cope reasonably well with inflation itself. Nonetheless, labor does have a definite interest in the real interest rate changes which may be necessary to change inflation. In the case of high real interest rates (designed to reduce inflation), there may be a tendency both for unemployment to rise and for real wages to fall. Even for financiers, it must be true that in any relatively sophisticated financial environment the real problem is not so much inflation as such, but the extent to which financial capital is protected from inflation, its real return. Very few individuals or institutions, after all, do keep substantial quantities of the proverbial banknotes under the mattress. In a high inflation environment, a genuinely troubling circumstance from the rentier point of view would be if real interest rates are negative, so that the required inflation protection is not present. Significantly, this did in fact occur in recent history, in several jurisdictions during the 1970s. It may well be the case, therefore, that although policy objectives tend to be articulated in terms of targets for the inflation rate, the underlying point of contention is the real rate of return on financial instruments itself. The

Macroeconomic theory, (critical) realism and capitalism 67 late-twentieth-century ‘conservative revolution’ in economic theory and policy-making might then be linked to the capture of the central banking policy agenda by the financial interests around 1979–82, the motivation being not only the high rates of inflation in the 1970s, but also the more urgent problem that financial capital was not being adequately compensated for this (Smithin 1990, 1996). The course of events in recent decades could then be explained as the playing out of this revolution, and its effects on the actual economy. The interpretation of monetary policy issues and events along these lines obviously involves identifying, in what seems to be the manner advocated by critical realists, the relatively enduring social structures under capitalism, and investigating the way they operate and the conflicts they generate. The chain of reasoning requires, first, recognition of the role which the need to acquire, use, and accumulate monetary resources plays in the functioning of the social system. Second, there is the realization that the ‘object of desire’ (Keynes 1936: 235), money, is not immutably fixed in supply but is endogenously created and destroyed by the process of bank lending itself. This is a progressive element in that it is possible for producers to acquire financing without previous saving to set in motion new activity. Inevitably, however, the same circumstances threaten previously entrenched interests. Third, there is an understanding that interest rates are not ‘natural’ or market-determined phenomena, but are themselves one of the main means for mediating political conflict between the social groups involved in the process of production. Whatever the accuracy of the substantive claims, it can be seen that this style of argument is more informed by, and consistent with, the prescriptions of critical realism than is the neoclassical microfoundations literature criticized above.

Additional comments on methodology One issue which remains controversial is the role of mathematics in macroeconomic theorizing. It is necessary to comment briefly on this here, as the advocates of critical realism, cited here as providing material for a revival of the original macroeconomics project, tend to deprecate the use of mathematical methods in economics. The use of mathematics is associated with axiomatic-deductive reasoning, Humean law, closed systems, and so on, and hence is not thought consistent with the precepts of critical realism (Fleetwood 1997; Lewis and Runde 1999). There is ample reason, of course, to sympathize with this perspective when contemplating current offerings in ‘leading’ academic journals. By now, the neoclassical school has not only ‘crept off to hide in thickets of algebra’ (Robinson 1962: 112), but seems well and truly lost. Hence, there may well be general agreement amongst currently heterodox economists that the

68

John Smithin

balance is not right in economic research and instruction, and that mathematics should be reduced to its ‘proper place’ (Harcourt 1995: 206). On the other hand, a difficulty for the present argument is that many of the scholars cited above as being important macro thinkers also used mathematical methods of various kinds in their work, albeit of a more accessible variety than those of contemporary mainstream economics. Contemporary non-neoclassical scholars also have recourse to certain types of mathematical reasoning. A good case in point for present purposes is the text by Lavoie (1992: 7–10), who explicitly adopts a realist methodological stance,16 and goes on to provide an exemplar of the use of simple mathematical methods, combined with other types of reasoning, in an authoritative exposition of Post Keynesian macroeconomics.17 Moreover, although the discussion (above) of a plausible macroeconomic environment did not resort to a mathematical idiom, there are clearly aspects of it which might well have been expressed in such a way. Given an ontology which conceives of the social realm as an open rather than a closed system, it is easy to explain why a certain suspicion of mathematical methods in the social sciences is justified. The use of mathematical reasoning and, in particular, equilibrium constructs would seem to imply closed systems, and results of the form ‘if x then y’ (Lawson 1994: 507–9; 1997: 16–19; Fleetwood 1997: 128). How then might this particular circle be squared in order to preserve a useful macroeconomic theory with a plurality of research methods? Potentially, one of the original starting points of the critical realist approach, as explained by Fleetwood (1997: 128–30) with reference to Bhaskar (1978), may itself provide a basis for some progress along these lines. The relevant issue is actually the status of experimental results even in the natural sciences. The natural world is also conceived of as an open and evolutionary system (Lawson 1998), and Humean law as such does not apply in the natural environment any more than in the social. Therefore, what the experimental situation does is to provide a local and artificial closure for the purpose of identifying certain mechanisms which are believed to be present in the actual system, but can be observed more accurately in isolation. An implication is that even the constant event conjunctions important to natural science do not occur ‘spontaneously’ (Fleetwood 1997: 128), but only in the artificially closed environment demarcated by experimental design. On the other hand, the results are often successfully applied in the real-world open system outside the experimental situation. The point is that the mechanisms identified in the experimental environment are still present in the larger system, and do in some circumstances dominate the actual course of events. However, if the course of events is otherwise, this does not imply that the mechanisms which have been discovered no longer exist, but simply that they have been interfered with, or superceded, by other tendencies that are also present. In the often-cited example of the ‘falling leaf’ (Fleetwood 1997:

Macroeconomic theory, (critical) realism and capitalism 69 128–9), if the leaf is blown away by the wind or caught on a roof, this does not mean that the gravitational mechanism has ceased to operate. Hence, the ‘layered ontology’ (Fleetwood 1997: 128–30) suggested by critical realism, involving the overlapping domains of the empirical, the actual, and the ‘deep’, becomes a useful framework for explaining both the actual course of events, and their apparent lack of synchronicity with ultimate causes. The way in which accounts of social science differ from the above is the lack of any genuine analogue to the experimental stage. There is now a sub-discipline of neoclassical economics called ‘experimental economics’, but, again, this is primarily concerned with the reactions of atomistically related groups of individuals to choices of various kinds. It does not attempt to identify any potential social mechanisms which might carry over to the real setting. Therefore, in the realm of macroeconomics, in the absence of genuine experimentation, the simple mathematical model can be thought of as in some sense the equivalent of the artificial closure of the experimental environment. The working out of the macro model is then part of the attempt to identify some of the relevant ‘tendencies and mechanisms’, believed to exist in the current social structure, in a quasiexperimental way. They may be useful, therefore, in that regard (Harcourt 1996). However, when we write down a macro model, we do not (or should not) thereby assert that an open social system is now closed, that the reality of human choice is denied, that the results of the model will directly translate into constant event conjunctions in the real system, or that the model could be applied to the actual course of events without doing extensive factual research. ‘Modeling’ on this view represents one method of identifying some of the potential tendencies and mechanisms which may exist and be operative/dominant in the actual system. It is important to recognize also that the model can never be anything more than this, a point seemingly lost sight of in conventional usage. Possibly, critical realists would continue to be skeptical of this limited case for some mathematical exposition of economic theory on the grounds that formal methods also presuppose atomism and a deductive method. In the ‘organic’ social world, it is a question not only of countervailing tendencies, but also of the stability of the mechanisms identified by formal analysis. However, it can be argued that macro methods involving aggregative functions, ‘propensities’, accounting relationships, and so on, which may already comprehend some of the internal/organic relationships between the social partners, need not necessarily entail reductionism. Historically, Keynes was the economist most sensitive to the question of organicism versus atomism in the social realm.18 Yet, his central contribution to economic theory, the ‘principle of effective demand’ (Keynes 1936: 25–30), was assuredly a mathematical model, specifically D = D1 + D2, D1 = (N), Z = I(N), D = Z, with D2 exogenous. Presumably, there was no assumption of social atomism in these

70

John Smithin

equations on Keynes’s part, and it this type of non-reductionist procedure that I am advocating here rather than neoclassical microfoundations. Stability, however, does remain a serious issue in the social world. Explanatory success will presumably depend upon the presence of such things as Lawson’s ‘demi-regs’ (Lawson 1997: 204–9),19 and the point stressed earlier that some social structures at least will be relatively longlived and enduring. How does the position developed here differ from the views of Lucas in his defense of mathematical modeling of the ‘new classical’ variety? According to Lucas ([1980] 1981: 271): ‘One of the functions of theoretical economics is to provide fully-articulated artificial economic systems that can serve as laboratories in which policies … can be tested out at much lower cost.’ The difference is that (on the same page) Lucas explicitly disavows any attempt at ‘realism’ in the construction of such models. He goes on to say ([1980] 1981: 272) that theory is ‘not a collection of assertions about the behavior of actual economy but … an explicit set of instructions for building … a mechanical imitation economy’. There is a requirement that the model be able to ‘mimic’ the behavior of an actual economy, and it is presumed that good mimicry will also lead to good prediction or forecasting. But that is all. The approach is purely instrumental. On the other hand, the position defended here would take for granted realistic basic assumptions, and also that the object of the exercise is to identify mechanisms believed to exist in the real world. The useful theory will not, in fact, be ‘fully articulated’. It will remain clear that what is being abstracted from are other tendencies which may be present in practice, and the openness of the real system. Moreover, the theoretical effort need not necessarily be regarded as a failure if prediction fails in any particular instance, the goal being explanation rather than prediction. Emphasis on the realism of the assumptions to be made in macroeconomic models seems analogous to Lee’s (2002: 792–801) concept of ‘grounded theory’, although there would clearly be scope for dispute about what constitutes adequate grounding in any particular exercise.

Conclusions This chapter has criticized contemporary mainstream macroeconomic research which is grounded in methodological individualism and the search for putative microfoundations based on the assumption of representative agents. It is argued that the methodological perspective of critical realism might provide the foundations for a more fruitful and sustainable macroeconomics project. In particular, the view of social institutions as ‘continually reproduced interdependencies’ is able to steer a middle course between the extremes of reductionism and determinism in the social sciences. In my own view, the style of research which would emerge would be a more traditional macroeconomics in the manner of

Macroeconomic theory, (critical) realism and capitalism 71 Keynes or the contemporary Post Keynesian school (Lawson 1994, 1999). The use of mathematics per se would not be rejected. However, the role of these methods in providing only partial local closures of what is in principle an open social system would be better understood. Moreover, mathematics would not occupy any exclusive or privileged position among the different modes of explanation (Harcourt 1987). The key to factual research would be to identify those social institutions and mechanisms which are relatively enduring in the course of a particular historical epoch. It has been suggested here that under capitalism some of the most crucial such institutions are money, the business firm and entrepreneurial activity, and wage labor. All of these have themselves undergone change and evolution, but their relative endurance over time explains why contemporary economists can still find relevance (if not necessarily agreement) in the works of such earlier writers as Thornton, Marx, Wicksell, Weber, Schumpeter, Keynes, and Kalecki, to cite a number of diverse examples.

Notes 1

2 3

4 5 6

7

I would like to thank Geoff Harcourt, Tony Lawson, Paul Lewis, Bill Scarth, Hana Smithin and Neil Wolff for many helpful comments and suggestions which have improved this paper. However, the usual disclaimer should certainly apply, particularly given the nature of the subject matter. Hodgson (1988: xiv) goes so far as to refer to ‘methodological fascism’ when describing this state of affairs. ‘Decision-making’ is a better term than ‘choice’, because, as a number of writers have pointed out, and in spite of claims to the contrary, the neoclassical theory does not actually allow for much choice as such. See Wray (2000: 63), who quotes Veblen on ‘self-contained globules of desire’, Hodgson (1996: 104–5), Lawson (1997: 8–11) and Lewis and Runde (1999: 38–9). This list of references is obviously not intended to be exhaustive, but rather to illustrate what is meant by such terms as ‘traditional macroeconomics’ in the present context. However, see the remarks of Lewis and Runde (1999: 49–50) and Rotheim (1999: 100) on the usefulness, or otherwise, of such exercises. See, for example, Paraskevopoulos, Paschakis and Smithin (1996), and Kam, Mansoorian and Smithin (1998). From the point of view of the present writer these contributions were precisely exercises in obtaining ‘non-neoclassical’ results from more or less orthodox premises. Similar remarks could be made on the more general topic of marginalism in economics. For example, Lavoie (1992: 136–7), discussing the Post Keynesian theory of pricing, remarks that the same results can always be obtained in a marginalist framework by (e.g.) assuming imperfect competition and constant costs over the relevant range. See also Tarshis (1980: 10–12). So, the main point of debate is not so much technique as the substantive implications of the theory of prices employed. In particular, the key issue is whether or not the setting of prices allows for the reality of the generation of profits or a surplus, which the neoclassical concept of perfectly competitive equilibrium does not.

72 8

9 10 11 12 13

14 15 16

17 18

19 20

John Smithin One implication of this, presumably, is that the chance of explanatory success using the standard methodology would be slim even when attention is turned by orthodox theorists to topics previously identified as being important by work in other traditions. There are several examples of this (e.g., endogenous growth, the supply-side effects of money and finance, the average opinion problem, etc.), which might be cited. However, as pointed out by Dutt and Amadeo (1990: 163), when such issues do finally make the agenda there is usually no reference at all to the earlier heterodox literature. See the remarks by R.E. Hall reported by Hodgson (1988: xiii). Also see Samuelson (1988: 321). Lawson (1997: 157–8) provides references to the earlier development of this literature, with particular emphasis on the work of Bhaskar (1978). This phrase echoes an expression used by Schumpeter about Walras. See Tobin (1980: 33). I am most grateful to Geoff Harcourt for suggesting this reference. An example of an approach which seems to ask such a question would be the types of argument used by Post Keynesian writers who would ground the existence of important features of the social structure (such as money and nominal contracts) in the nature of the macro environment faced by economic actors, particularly its transmutability or non-ergodicity. See (e.g.) the discussion by Lewis and Runde (1999: 44–53) of the work of Davidson. See also the discussion of the ‘social relation’ of money below. For Tarshis (1980: 10) the key institutions in contemporary capitalism are ‘multi-national oligopolistic mega-corps’. I am indebted to Neil Wolff for an illuminating discussion on the subject of entrepreneurship. Dow (1999: 15) points out that Lavoie does not actually use the modifier ‘critical’ in his discussion of realism (as is true of a number of other Post Keynesian writers), but nonetheless concludes that this is the sense of the argument. See also the discussion by Harcourt (1996: 94–8), including reflections on some of his own earlier contributions (e.g., Harcourt 1965). According to Keynes (1933: 286): ‘The atomic hypothesis which has worked so splendidly in Physics breaks down in Psychics’. Harcourt (1995: 206) mentions that according to contemporary opinion it does not work particularly well in physics either. That is to say, the partial regularities of the kind that do seem to obtain in the social world, particularly of a ‘contrastive’ nature (Lawson 1997: 204–8). Lawson explicitly argues that ‘in accepting (the) ex posteriori assessment that the world is open and structured, it does not follow (as some have supposed) that Post Keynesians [and presumably others] ought thereby not to engage at all in formalistic methods’ and ‘critical realism … is quite consistent with a limited and informed methodological pluralism’ (1999a: 7, 8; emphasis added).

References Bhaskar, R. (1978) A Realist Theory of Science, Leeds: Harvester Press. Chick, V. (1986) ‘The Evolution of the Banking System and the Theory of Saving, Investment and Interest’, Economies et Sociétés, 20: 237–50. Colander, D. and A. Klamer (1990) The Making of an Economist, Boulder, Colorado: Westview Press. Crotty, J. (1980) ‘Post Keynesian Economic Theory: An Overview and Evaluation’, American Economic Review Papers and Proceedings, 70: 20–25.

Macroeconomic theory, (critical) realism and capitalism 73 Davidson, P. and E. Smolensky (1964) Aggregate Supply and Demand Analysis, New York: Harper & Row. Dillard, D. (1988) ‘The Barter Illusion in Classical and Neoclassical Economics’, Eastern Economic Journal, 14: 299–318. Dostaler, G. and G. Maris (2000) ‘Dr. Freud and Mr. Keynes on Money and Capitalism’, in J. Smithin (ed.) What Is Money?, London: Routledge. Dow, S.C. (1999) ‘Post Keynesianism and Critical Realism: What is the Connection?’, Journal of Post Keynesian Economics, 22: 15–33. Dutt, A.K. and E.J. Amadeo (1990) Keynes’s Third Alternative: The Neo-Ricardians, Keynesians and the Post Keynesians, Aldershot: Edward Elgar. Fleetwood, S. (1997) ‘Critical Realism: Marx and Hayek’, in W. Keizer, B. Tieben, and R. van Zijp (eds) Austrian Economics in Debate, London: Routledge. —— (ed.) (1999) Critical Realism in Economics: Development and Debate, London and New York, Routledge. Harcourt, G.C. (1965) ‘A Two-Sector Model of the Distribution of Income and the Level of Employment in the Short-run’, Economic Record, 40: 103–17. —— ([1977] 1986) ‘Introduction’, in G.C. Harcourt (ed.) The Microeconomic Foundations of Macroeconomics, New York: Stockton Press. —— (1987) ‘Theoretical Methods and Unfinished Business’, in D.A. Reese (ed.) The Legacy of Keynes, San Francisco: Harper & Row. —— (1995) ‘On Mathematics and Economics’, in Capitalism, Socialism and Post Keynesianism, Aldershot: Edward Elgar. —— (1996) ‘How I Do Economics’, in S.G. Medema and W.J. Samuels (eds) How Do Economists Do Economics?, Cheltenham: Edward Elgar. Hayek, F.A. (1944) The Road to Serfdom, London: Routledge and Sons. —— (1988) The Fatal Conceit, Chicago: University of Chicago Press. —— (1994) Hayek on Hayek, Chicago: University of Chicago Press. Hicks, J.R. (1989) A Market Theory of Money, Oxford: Oxford University Press. Hodgson, G. (1988) Economics and Institutions, Cambridge: Polity Press. —— (1996) ‘Towards a Worthwhile Economics’, in S.G. Medema and W.J. Samuels (eds) How Do Economists Do Economics?, Cheltenham: Edward Elgar. Ingham, G. (1996) ‘Some Recent Changes in the Relationship between Economics and Sociology’, Cambridge Journal of Economics, 20: 243–75. —— (1999) ‘Money Is a Social Relation’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. —— (2000) ‘Babylonian Madness: On the Historical and Sociological Origins of Money’, in J. Smithin (ed.) What is Money?, London: Routledge.. Kaldor, N. (1955/56) ‘Alternative Theories of Income Distribution’, Review of Economic Studies, 23: 94–100. —— (1982) The Scourge of Monetarism, Oxford: Oxford University Press. Kalecki, M. (1971) Selected Essays on the Dynamics of a Capitalist Economy 1933–1970, Cambridge: Cambridge University Press. Kam, A.E., A. Mansoorian and J. Smithin (1998) ‘A Note on Endogenous Time Preference and Monetary Non-superneutrality’, Working Paper no. 98–10, Department of Economics, York University. Keynes, J.M. (1923) A Tract on Monetary Reform, London: Macmillan. —— (1930) A Treatise on Money (2 volumes), London: Macmillan. —— (1933) Essays in Biography, London: Macmillan.

74

John Smithin

—— (1936) The General Theory of Employment, Interest and Money, London: Macmillan. Kindleberger, C.P. ([1979] 1989) Manias, Panics and Crashes, New York: Basic Books. Kirman, A.P. (1992) ‘Whom or What Does the Representative Individual Represent?’, Journal of Economic Perspectives, 6: 117–36. Knapp, G.F. ([1924] 1973) The State Theory of Money, New York: Augustus M. Kelley. Lavoie, M. (1992) Foundations of Post-Keynesian Economic Analysis, Aldershot: Edward Elgar. Lawson, T. (1994) ‘The Nature of Post Keynesianism and Its Links to Other Traditions’, Journal of Post Keynesian Economics, 16: 503–38. —— (1997) Economics and Reality, London: Routledge. —— (1998) ‘Clarifying and Developing the Economics and Reality Project: Closed and Open Systems, Deductivism, Prediction and Teaching’, Review of Social Economy, 56: 356–75. —— (1999a) ‘Connections and Distinctions: Post Keynesianism and Critical Realism’, Journal of Post Keynesian Economics, 22: 3–14. —— (1999b) ‘Developments in Economics as Realist Social Theory’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. Lee, F.S. (2002) ‘Theory Creation and the Methodological Foundations of Post Keynesian Economics’, Cambridge Journal of Economics, 26: 789–804. Lerner, A. (1943) ‘Functional Finance and the Federal Debt’, Social Research, 10: 38–51. Lewis, P.A. and J. Runde (1999) ‘A Critical Realist Perspective on Paul Davidson’s Methodological Writings on – and Rhetorical Strategy for – Post Keynesian Economics’, Journal of Post Keynesian Economics, 22: 35–56. —— (2002) ‘Intersubjectivity in the Socio-economic World: A Critical Realist Perspective’, in E. Fullbrook (ed.) Intersubjectivity in Economics: Agents and Structures, London: Routledge. Lucas, R.E. ([1980] 1981) ‘Methods and Problems in Business Cycle Theory’, in Studies in Business-Cycle Theory, Cambridge, MA: MIT Press. —— (1987) Models of Business Cycles, Oxford: Basil Blackwell. Marterbauer, M. and J. Smithin (2000) ‘Fiscal Policy in the Small Open Economy within the Framework of Monetary Union’, WIFO Working Paper 137/2000, Vienna. Marx, K. ([1859] 1970) A Contribution to the Critique of Political Economy, Moscow: Progress Publishers. —— ([1867] 1976) Capital, Volume 1, London: Penguin Classics. Menger, C. ([1871] 1950) The Principles of Economics, Glencoe, Illinois: Free Press. —— (1892) ‘On the Origin of Money’, Economic Journal, 2: 239–55. Mises, L. von (1978) Notes and Recollections, South Holland, Illinois: Libertarian Press. Nell, E.J. (1998) ‘Functional Finance in Retrospect and Prospect’, mimeo, New School for Social Research. Paraskevopoulos, C.C., J. Paschakis and J. Smithin (1996) ‘Is Monetary Sovereignty an Option for the Small Open Economy?’, North American Journal of Economics and Finance, 7: 5– 18. Robinson, J. (1956) The Accumulation of Capital, London: Macmillan.

Macroeconomic theory, (critical) realism and capitalism 75 —— (1962) Economic Philosophy, London: C.A. Watts. Rogers, C. (1989) Money, Interest and Capital, Cambridge: Cambridge University Press. Rotheim, R.J. (1999) ‘Post Keynesian Economics and Realist Philosophy’, Journal of Post Keynesian Economics, 22: 71–103. Samuelson, P.A. (1948) Economics: An Introductory Analysis, New York: McGrawHill. —— (1988) ‘The Passing of the Guard in Economics’, Eastern Economic Journal, 14: 319–29. Sargent, T.J. (1979) Macroeconomic Theory, New York: Academic Press. Scarth, W.M. (1996) Macroeconomics: An Introduction to Advanced Methods, Toronto: Dryden. Schumpeter, J.A. ([1942] 1992) Capitalism, Socialism and Democracy, London: Routledge. —— ([1954] 1994) History of Economic Analysis, London: Routledge. Smith, A. ([1776] 1981) An Inquiry into the Nature and Causes of the Wealth of Nations (2 volumes), Indianapolis: Liberty Fund. Smithin, J. (1990) Macroeconomics after Thatcher and Reagan, Aldershot: Edward Elgar. —— (1994) Controversies in Monetary Economics, Aldershot: Edward Elgar. —— (1996) Macroeconomic Policy and the Future of Capitalism, Aldershot: Edward Elgar. —— (1999) ‘Money and National Sovereignty in the Global Economy’, Eastern Economic Journal, 25: 49–61. —— (ed.) (2000) What Is Money?, London, Routledge. Tarshis, L. (1980) ‘Post-Keynesian Economics: A Promise That Bounced?’, American Economic Review Papers and Proceedings, 70: 10–14. Tobin, J. (1980) Asset Accumulation and Economic Activity, Chicago: University of Chicago Press. —— (1986) ‘The Future of Keynesian Macroeconomics’, Eastern Economic Journal, 12: 347–56. Turnovsky, S.J. (1999) Methods of Macroeconomic Dynamics, Cambridge, MA: MIT Press. Weintraub, S. (1959) A General Theory of the Price Level, Output, Income Distribution and Economic Growth, Philadelphia: Chilton. Winslow, T. (1999) ‘Kleinian Psycho-analysis and Keynes’s Treatment of Capitalist Purposiveness’, mimeo, York University. Wray, L.R. (1998) Understanding Modern Money, Cheltenham: Edward Elgar. —— (2000) ‘Modern Money’, in J. Smithin (ed.) What is Money?, London: Routledge.

4

Critical Realism and Transformational Growth Edward J. Nell

Introduction1 Critical Realism (CR) approaches the explanation of economic events in terms of an ontology that sees ‘structured objects’, possessing various powers, lying ‘beneath the surface’ of the visible sensory world. These objects and their powers exist independently of our knowledge or perception and form the basis of the causal relationships that science describes. From the Transformational Growth (TG) perspective this is surely acceptable, details aside. CR goes on to address some of the special problems that arise in connection with such ‘structures’ in the social world, and here it overlaps with the TG perspective. In particular both stress that social structures are based on obligations and promises, and both note that structures maintain their existence through interdependent production and reproduction. But they treat these matters somewhat differently. Once these differences are properly understood, however, they lead directly to the main province of TG, namely providing an account of the way these structures evolve and develop over time.

Outline of Critical Realism CR has been offered as a foundation for a critique of neoclassical economics and, of equal importance, as the basis for a critique of the alleged underpinnings of econometrics. Neoclassical economic theory is said to function at the level of appearances and events; it fails to develop an analysis of ‘deep structures’. It simply relates observables to one another – choices and actions are put together into ‘equilibrium’ patterns. It is claimed that neoclassical economics makes no effort to unearth a deeper level. Econometrics is the practical manifestation of this approach to theory; correlations are precisely Humean ‘constant conjunctions’, and the search for them is the practice of relating observables to one another. Taking this a step further, Critical Realists argue that when Econometrics seeks to go beyond good correlations and impute causality, the notion it employs – Granger causality – is defective, being strictly

Critical Realism and Transformational Growth 77 Humean, depending as it does, mechanically, on temporal priority. Yet it can be argued that temporal priority is neither necessary nor sufficient for causality, using the concepts in their normal sense. Theory is indeed necessary to the practice of traditional econometrics; outcomes certainly do depend heavily on the assumed ‘priors’. But the conventional approach has no way to justify either the priors themselves or the need for them. For the most part, neoclassical ideas are simply taken for granted, and used to establish the required ‘priors’. The procedures then find the best fits, subject to those assumptions. CR looks for stable existing structures of objects, objects underlying the observable phenomena. These give rise to ‘mechanisms’ which are the operating systems of the world; describing these mechanisms in laws or law-like statements will tell us how the world works. Theories are fullscale descriptions of these mechanisms, showing their inner logic. These mechanisms are stable; they have to be, for intentional human action to be possible, for structures defining the ‘world’ to persist. The relationships here are hierarchical, and hold between the level of observable phenomena and underlying ‘intransitive’ objects possessing powers (objects, that is, which exist independently of observers and of knowledge about them). The existence of these objects – and the possibilities of their powers – are established by transcendental arguments. Such arguments take the form: this is what has to be the case for the world to be as it is. These arguments establish the outlines of what there is in the world – what CR calls the philosophical ontology. Objects are defined by their essences, and this enables us to understand them at ‘their own level of being’ (Lawson, 1997). To fill in the details, to say what these structures are like, and what powers they have, then requires practical science, real investigations. Once the structures and their powers are determined and delimited, it can be seen just how they give rise to the practical mechanisms which govern the world. These, in turn, are described by law-like statements, and when such statements are gathered together and fully articulated, they form theories, telling us how the world works. As a first approximation this surely passes muster, but it has to be admitted that it is not clear what a ‘structure’ is. Examples suggest that many different kinds of objects and mechanisms are bunched together. TG takes a similar approach, starting from the idea of an agent. Then the outlines of ‘what there is’ – or ‘what there has to be’ – are established through ‘conceptual analysis’, and these are next filled by means of ‘field work’, and finally the story is pulled together in the form of theories that are developed interactively with practical and statistical research. The initial development of CR applied to the natural world, but the basic relationships are considered to hold equally in the natural and social worlds. The differences do not lie in the way theories are formed and justified. Not that the two are the same in all respects; far from it – the

78

Edward J. Nell

social domain appears to be open and so must be described by theories that reflect and acknowledge that openness. (In a sense, theories are always ‘indeterminate’ – underdetermined by the evidence – but that is a different matter.) Social theories, however, are based on underlying mechanisms just the same as the theories of natural science, ‘the social sciences are concerned to study intransitive objects in exactly the way of natural ones’. Moreover, ‘being, or existence, means the same thing in both contexts even if the forms of being are different’ (Lawson 1997: 223). (Here TG tends to differ, taking the view that the role of ‘intention’ in social action creates a dimension in social scientific analysis that does not exist in natural science.) Realism requires establishing what exists. Existence is established by the practice of referring, which has two aspects, conceptual and practical. Conceptual reference means finding descriptions which will draw the attention of others, enabling the object to be located or isolated, picked out from the crowd. Practical reference means using material methods (machinery, instruments, scientific or practical equipment) to determine the exact details and properties of the object in question. Practical reference and conceptual reference will normally interact, that is, each will correct, inform and guide the other. Each will enable the other to become more exact. TG takes a somewhat different path, heading, however, in the same direction. It seeks to establish ‘conceptual truths’ about the agents, starting with the idea in general.2 Agents are rational animals, mortal, have free will, have active minds, and are material bodies, subject to the laws governing material bodies. These features of ‘agency’ are then used, as we shall see shortly, to explain the continued existence of a social system by means of ‘reproduction.’

Realism and determinism Models have to refer to what exists; that is what is meant by ‘realism’. But models also have to exhibit relationships that are like those in reality. In particular, the social order is held to be ‘open’, that is, it cannot be circumscribed and summed up in a deterministic model. Nor can it be described in terms of stochastic regularities of the sort presupposed by econometricians (Lawson 1997: 76–7; also Lewis and Runde 1999: 38–9). The reasons for this openness, however, could be made clearer.3 Is it simply a fact, or is it necessarily the case? An important implication, however, is drawn from this alleged fact, namely that social and economic theories cannot be tested by prediction. Being open, outcomes are, in some sense, ‘indeterminate’. Instead, economists and social scientists must look for ‘tendencies’, generated by structures. (TG would consider that a model of such ‘tendencies’ could have a perfectly determinate

Critical Realism and Transformational Growth 79 outcome – and would argue, as we shall see, that such closures are useful even if they are tidier than reality.) Neoclassical theory is rejected as empiricist and deductivist. This has been argued before, so while the charge is not at all new,4 it is still as controversial as ever. And while it is clearly true of a great deal of what passes for econometric research, it is not so clear that mainstream economic theory can be both empiricist and deductivist at the same time. Followers of the Rational Expectations approach, in particular, take the view that there is a single, correct account of the world. Agents know this, and plan their behavior in accord with it. They optimize, that is make the best choices they can subject to their constraints so that their actions can be deduced from their preferences and constraints. These choices will be determinate, and will be set in the context of the correct account of the way the system works. In short, the resulting theory will be deterministic, testable on the basis of its predictions. It describes a closed system – and therefore will be regarded by CR as profoundly misleading because it ignores the very essence of real human choice, namely the capacity of an agent to do otherwise than he or she in fact did.5 As Martin Hollis (1994: 185–6) has put it: ‘[H]omo economicus is a very mechanical individual, a mere throughput between given preferences and an automatically computed rational choice’. This is deductivist, certainly, but hardly empiricist. How could the maximizing agent be wrong if s/he follows the algorithm correctly and has reasonable knowledge of the constraints? How could an empirical test falsify such a model? On the other hand, if we were to take the line of the early American empirical economists, e.g. H.L. Moore, E. Working and H. Working, and H. Schulz, and simply try to identify statistically based supply and demand curves, there would then be no deductivist component. (Of course to identify the curves there have to be ‘priors’, but these need not come from theory.) Still, consider the mainstream approach in a general form, as presented, say, by Samuelson in his Foundations of Economic Analysis (1947). Rational agents face market variables; for given values of these they maximize. As the market variables change, repeated application of maximizing traces out a set of response variables as a function of the market ones. These can then be aggregated over the set of agents. Put together the functions for the two sides of the market, and solve for the values that satisfy both sides. That will be the market equilibrium. (To take it a step further, we take all markets together, to define a general equilibrium. This, however, requires special and hard-to-justify restrictions on the form of the functions.) Let’s ask now whether this is fully coherent, if taken as a general model of behavior. It is presumably, in some sense or other, a ‘true’ or accurately predicting account. We will come to that later. What is supposed to be included in this model? Is it supposed to be ‘complete’? Does it include,

80

Edward J. Nell

for example, the effects of the understanding by participants of the true account or correct predictions? We build a model of the world, or of the part of the world about which we are theorizing. Does our theory of how the world works include the effects of this theory in guiding our behavior and informing our insight? The way these effects are brought about and how they play out must surely be the subject of a further development of the theory. But then, what of the knowledge represented by this further development – will it, too, not have an impact? Which in turn would have to be formulated in a still further development – and so on, ad infinitum. Consider a variation on this argument – this may shed some light on why social theories must be ‘open’ in at least some ways. Supposing there is a model that will predict an agent’s action from physical data. What happens when the agent studies the model? Having studied it, will the agent still act as the model predicts? Suppose that the agent is arguing the case for free will, and the model predicts that he will so argue in scholarly debate. He could then in his argument deny free will – and thereby affirm it, by proving the model wrong! His other alternative is to do as the model says he will, and argue the case for free will. Either way, he advances the case for free will. If that is what the model predicts, however (that he will seek to advance the case for free will), the model cannot be disconfirmed, since its prediction is satisfied whichever way he acts. So it fails the falsifiability test. Nothing that the agent does can disconfirm it. On the other hand, if the model is supposed to predict what he will say, as opposed to what he will do, he can certainly advance his cause by opting to argue against free will on this occasion! Could, say, E. O. Wilson or Daniel Dennett – both supporters of closed deterministic systems for explaining the whole of human behavior – argue that he would not or could not do this? Suppose they were to argue that we have a higher-order model that will predict what the agent will do when he reads or studies the first-order model. The first-order model does not apply to cases that involve the model itself. The second-order model tells us how the agent will react to the first-order model. But then suppose he studies the second-order model? There will have to be a third-order model to tell us how the agent will react to the secondorder one. Indeed, at whatever level a model is constructed, when the model itself predicts the behavior of the agent who now comes to study the model, there will have to be a model of a still higher order. Only by developing such a higher-order model will it be possible to encompass the case where the agent learns what the given model (one level below) says about him. This leads to an infinite regress – which is to say, the system is necessarily open-ended. Once we take into account the model’s impact on the agent whom the model models, it seems no longer possible to obtain determinate results. This suggests that some kinds of openness may be

Critical Realism and Transformational Growth 81 necessary in social theory, a position taken by the Transformational Growth approach. Clearly the question should be pursued further.6

Uniqueness and change A CR account provides a description of a ‘structured ontology’. This is certainly opposed to Methodological Individualism. It can be described perhaps as a kind of Methodological Structuralism – identifying the underlying structures is the basis of the approach. Once they have been clearly described and their features determined, it can then be shown how they generate the surface phenomena. Or rather, tend to generate them. Clearly getting the account of the structures right is crucial, and this is where ‘realism’ comes into play. The structures cannot be misdescribed, or be ‘assumed’ to have attributes or powers which they in fact do not. Realism means providing a true account of the underlying structures and their powers and relationships. Given a ‘realistic’ account of the underlying structures raises the question of whether there is just one such true account. Is the identification of objects and structures unique? Are there equally acceptable alternative accounts, some useful in one way, others in other ways, perhaps? Of course there may be many incomplete accounts, each quite different from the rest – the old story springs to mind of the blind men each describing the elephant according to the part he could touch. But can there be more than one significantly different complete account? (Complete, that is, in regard to identifying essentials.) Speculating, it seems likely that CR would readily argue that there may well be a variety of true, but incomplete, accounts of an object. All theoretical investigations are context-dependent in the sense that the various aspects of its subject-matter that a theory focuses upon will clearly depend upon the question under investigation. Hence, CR would argue that there may well be a variety of true, but incomplete, accounts of some object. Such accounts tell the truth – they do not ascribe properties (such as perfect foresight) to an object under investigation which turn out to be false when strictly predicated of it. But they do not tell the whole truth in the sense of providing a complete account of that object. CR, like TG, would argue that there could be only one complete and true account of an object, but CR would hold that it is practically impossible to achieve this, the reason being that the world is complex and messy so that it is impossible to comprehend it in its entirety in one go (Lawson 1997: 230–1; Runde 1996). By contrast, TG argues that completeness in regard to social objects and social structures cannot be achieved in principle, for reasons outlined earlier, namely that part of a complete account will be selfunderstanding, leading to an infinite regress. On the other hand, TG would hold that the nearer one gets to a complete account, the more the different lines of inquiry will tend to converge. In a sense there is only one

82

Edward J. Nell

true, complete account, but it can only be approached, never reached. (Moreover, the account may have to change as it approaches the truth more closely, for the increase in knowledge – especially selfunderstanding – may alter the basic structures.) An important distinction in this respect, not original with CR but emphasized by it, is that between abstraction and idealization. CR points out that, just because it is impossible to comprehend the entirety of (complex aspects of) the socio-economic world in one go, it does not follow that theories and models are justified in employing descriptively false assumptions. The tendency to do so ignores the distinction between abstraction and idealization (Runde 1996; O’Neill 1996: 41–4; Lawson 1997: 108–33, 227–37). Abstraction is a process of focusing on particular aspects of some (concrete) phenomenon, with the aim of individuating or picking out particular features while ignoring others. Notably, it is not the case that the existence of the neglected features is denied; rather they are (momentarily) left out of focus and relegated to the periphery of our attention. Abstraction, then, is a matter of bracketing features of the phenomenon under investigation rather than of denying their existence. Abstract reasoning makes claims that (hopefully) do not hinge on the neglected features of objects to which the reasoning is applied. Idealization, in contrast, involves the ascription of features to an object that it does not in fact possess, that is, features that are false when predicated of it. Thus, theorists who use idealization invoke fictions, objects that exist only in the realm of ideas. Idealization generally comes in one of two forms. The first of these involves the use of ideal or limit types, that is ‘enhanced’ versions of entities or situations which possess features that are (in some sense) perfect, complete or pure.7 Examples would be perfect information, or perfect foresight. The second form is to invent a fictitious being or activity, study its properties, and then contend that the facts of the world are ‘as if’ they had been generated by the invented agent or process. An example would be a ‘representative agent’. We might also ask whether a complete account includes an account of the propensity of an object or structure to change. Can we clearly and definitively describe a structure if it is in the process of becoming something else? To what extent can we explain or even adequately describe what is happening, if the source that is generating the surface phenomena is itself undergoing a basic change? In fact while change is discussed at some length in CR, it is not treated as inherent in the understanding of social structure. CR appears to be static in many respects – social objects and structures are identified and described. They may or may not change; this will be a matter to investigate. There is no necessary presumption one way or the other. It’s worth quoting Lawson at some length:

Critical Realism and Transformational Growth 83 by recognising the centrality of action and the processes of social reproduction/transformation to all aspects of human life including society and economy, an intrinsically dynamic perspective for social science is seen to be unavoidable … [C]hange is integral to … social or economic objects, structures or systems. We have seen that although social structures are a condition for individual acts it is only through the actions … of individuals in total that any social structure endures when it does, that it is reproduced or transformed. Social structures, in consequence, can never be regarded as somehow fixed … There is always fluidity and movement. Even when, over a restricted region of time and space, a structure might be held to have been reproduced intact as it were … this is always on the basis of intrinsically dynamic and always potentially transformative human practice … Change, then, (as with continuity) is endemic to social life … It follows from this transformational conception of social activity that if change, or at least the potential for change, is always present, the analysis of change per se is of no greater (or lesser) significance to social explanation than the understanding of continuity and reproduction. Both (relative) change and (relative) continuity need to be addressed and accounted for in science. Ex post, to be sure, certain structures are found to be relatively more enduring than others. But this does not license any view to the effect that the relatively more enduring structures are somehow more natural, normal or given. Put differently, there is no ontological asymmetry between relative continuity and relative change. It is a mistake to regard any social structure or system as something that would somehow reproduce itself unless obstructed from doing so. In consequence, there is equally no epistemic or logical asymmetry between the explanation of relative continuity and relative change. Tradition or fashion, routine or novelty, habit or deliberation, imitation or invention, preservation or innovation … are in principle open to, and require, social explanation. In the social world change and continuity are on a par as real possibilities. (Lawson 1997: 170–1) For CR stasis and change are on a par; both need to be studied. TG certainly accepts this, but takes a further step: the potential for change is inherent in the conditions of social stasis, and will eventually manifest itself. Social systems will either develop or collapse; they will not endure unchanged forever. Of course, there are great differences in the rapidity and magnitude of change. In CR, evolution – as a process inherent in the system – is not discussed at all. In particular, there is no discussion of in what ways these ‘structured objects’ and their powers might change, or develop. Why and under what circumstances might they change? What systematic pressures

84

Edward J. Nell

for change have emerged, and why? (Under what circumstances, if any, could there be no such pressures?) Nor is there discussion of how such objects might come into existence or go out of existence. Where did they come from? How long will they last, and where will they go? Could they last forever and, if not, will their demise be accidental or the result of a determinate process? Three kinds of processes of change can usefully be distinguished, even though they may overlap or blend into one another at times: •

•

•

the normal life cycle: birth, growth, maturity, senescence, death and decay – ‘dust into dust’. This applies not only to people, but to animals of all sorts. It also applies to inanimate objects, ordinary manufactured things like tables, chairs and automobiles, and natural objects, like stones, mountain ranges and planets. But it doesn’t seem to apply to elements or to molecules and atoms. In economics we have the product life cycle, and Marshall discussed the life cycle of business firms. the transformational life cycle: eggs become chickens, caterpillars butterflies, and human eggs turn into embryos, then foetuses and finally become babies. Feudalism develops into capitalism. The initial stage is fundamentally different – it works differently – from what it turns into. evolutionary processes: humans have developed from homo habilis to homo sapiens sapiens, competitive markets evolve into oligopolies, a population of family firms is transformed into a group of modern corporations, and then into a set of interlocking networks. Animals are given special characteristics through selective breeding, and now through genetic engineering. Or animals may develop certain characteristics through ‘arms races’, as when antelopes become faster and more graceful, because only those can escape the leopards and live to breed; and, of course, the leopards likewise become faster and more skillfully stealthy, since the slow and blundering starve.

Institutions and what exists Let’s try to deal with these issues. First, just what is under discussion when CR refers to ‘structures’ in economics? They are supposed to ‘underlie’ economic phenomena. How, exactly? What does ‘underlie’ mean? A possible example runs as follows. Prices and incomes are paid in money in market transactions but, following the neo-Ricardian tradition, it could be argued that underlying such market phenomena are the barter exchange ratios and real income shares that reflect the conditions of reproduction and the relative potency of the social classes, as expressed in business and civil institutions. To justify such a claim we have to dig deep, to determine what exists in society, and how such existence is maintained,

Critical Realism and Transformational Growth 85 that is, what continued existence depends on, leaving aside all transient, ephemeral matters. To study change, we have to know what is changing; something has to provide continuity in the process. So we have to establish what there is. Consider the example of the ‘nation’, the bearer of the economic variables ‘national income’ and ‘GNP’. An important first step – which can only be sketched here – is to distinguish the nation and its characteristics and relationships from its current citizens and their characteristics and relationships. This applies to institutions at all levels. The club and the interests of the club should not be identified with its present membership and their interests – though of course there may be a large overlap, but, just the same, the country is distinct from its current population. How can the club exist apart from its members, the office from its holder, the job from the worker? Institutions must be identified in relation to persons,8 but they are defined in terms of functions to be fulfilled, which imply duties to be carried out. These are allocated to various offices or positions, which are granted powers and responsibilities. The whole is made up of these parts, fitted together to perform complementary functions. The relation of part to whole here is that of component to composite, not item to list or unit to aggregate. The component is defined by its role in the composite, as piston to engine, captain to army, sales manager to company. (CR would describe these as ‘internal relations’, which is to say that the connection is, in part, necessary or conceptual.) To hold such an office or position is to step into a role, to which one must be appointed or somehow confirmed, and which carries duties, rights and powers. To be appointed, one must be qualified, one must have the appropriate knowledge, skills and abilities, that is, one must be a certain type, the product of socialization. There are stable, established ways of supporting a population and reproducing it; these are carried out by people trained for that purpose. The well-defined activities and the training that prepares people to do them can be said to be institutionalized. Social structures depend on rules and on agents following the rules, and knowing how to apply the rules correctly in new circumstances, innovating if necessary to make the rules fit. In short, institutions are what exist in society, what make up society. To say that an institution exists is to say that there is a set of practices that enable it to be maintained and replaced, both in regard to equipment and personnel. These practices are themselves organized and grouped in such a way that they are the responsibilities of appropriate roles. The roles, in turn, are well defined in terms of powers and responsibilities, calling for appropriate skills. Training and socialization practices prepare agents; selection procedures assign trained or socialized agents to the roles for which they are (best) suited, as older agents retire. This is a process of

86

Edward J. Nell

reproduction, as agents are trained and prepared, then selected to perform the activities of society, where these activities interlock, so that what each uses up in its normal procedures is replaced by the output of other activities in their normal operations. (Again we see that these interlocking activities are ‘internally related’.) Goods and services are used to produce goods and services, which in turn support agents acting in roles that manage and carry out the work of producing and using the goods and services, in the process creating and bringing into actuality a certain set of ‘lifestyles’. If this is a reasonable account of the ‘underlying structures’ of a social order, then it would not seem unreasonable to accept, at least provisionally, the neo-Ricardian account above as a first step in a CR analysis of economic phenomena. But, to date, while CR surely points in the right direction, it has not systematically addressed the question of how what exists in society – the basic social order – is maintained and reproduced. Nor is it even clear that what exists has been reliably identified (Parsons 1999; Eriksson 1998). Until this is done, we can’t move on to examine how these things emerge and disappear, or change and develop – or decay.

Transformational Growth and institutional development By contrast these questions are at the center of the TG approach, which rests on what might be called ‘Methodological Institutionalism’. This is, indeed, itself a kind of structuralism, in that the institutions, which are defined by formal relationships, do in a sense generate the observed phenomena.9 Institutions may be related to one another in formal and legal ways, and these relationships – internal relationships – may also be considered structures, as in the account sketched above. The important point is that institutions rest on obligations – promises that agents, appointed to various positions, will do their duties.10 The institutions, as a whole, function because the agents in the various roles do as they are supposed to, they carry out their duties, on time and to a reasonable standard. Expectations – including market expectations – typically rest on obligations; bondholders expect to be paid – the bond is a promise to pay. We know that the police are sworn to uphold the law, and we expect them to do so. We know that a certain firm has promised to license one of its patents to another firm. We expect it to honor its promise.11 The powers attributed to or conferred on certain offices also rest ultimately on obligations. The traffic police have the power to pull drivers over – because it is the duty of citizens to obey the law. The general can command his troops to march or to open fire, because it is the duty of soldiers to obey; indeed they have taken an oath to do so. The manager of the company can hire and fire employees. It is a power conferred upon

Critical Realism and Transformational Growth 87 him by the board; it is part of their job for employees to obey their superiors. To perform a duty, do a job properly, or fulfill a function requires following rules. But rules cannot be followed mechanically, because circumstances change.12 The rule has to be adapted to changed circumstances, which means understanding its intent, and then seeing how that intent can best be carried out in the new or changed conditions. This requires analysis and creativity. It requires an active mind (Nell 1998: chs 3, 4), the most fundamental reason for the ‘openness’ of socioeconomic reality. The active mind is implied by the fact that agents, acting under obligations, must be able to apply the rules to new situations. They must be able to redefine their obligations, and what it means to meet them, to adapt to new circumstances. This implies that agents can always reconceptualize their obligations, that they can always invent or create new powers, develop new forms of control over nature, and define new relationships with each other.13 This is an important foundation for the Transformational Growth approach; it is always possible to change the way we relate to the world. But agents cannot do this just as they please; the ability of agents to bring about changes and innovations will depend on the framework of institutions within which the activity of innovation is carried out. The central institution here is the market, for it provides both incentives and support to the active mind in rethinking the way we relate to the material world. In particular, it supports developing control over the world. To follow rules, to perform a job, requires acting, which means using up energy and resources. It also means getting exhausted, becoming tired, worn out, aging. (CR agrees and, indeed, examines the ‘material causes’ – in Aristotle’s sense – of social activity.) Resources – energy and materials – have to be replaced. The replacements will have to be made on a regular basis, and the act of making them will wear out tools and equipment. With developed technologies, more is likely to be produced than will be used up in the aggregate in the production processes. This sets the stage for an ‘input–output’ approach, as in the Classical models of production and distribution. If the jobs in production and management are to be performed on a regular basis, then personnel replacement and renewal must also be regular. That is to say, if the institutions are permanent, then training of new staff will have to take place regularly as well. Suitably prepared new agents will have to replace the old, as the latter retire or die. Institutions are staffed by appointment; an agent must be chosen or selected in some way, by public criteria, and then, in some public ceremony, appointed, sworn in, given the office. We can understand this by distinguishing ceremonies which mark entering into office, and assuming the corresponding powers and responsibilities, from those

88

Edward J. Nell

which mark graduating from an institution of training or socialization, and receiving the appropriate qualification. Appointments are normally made from among those who have earned the appropriate qualifications. The obligations of agents in certain positions in given circumstances may be debatable, but they are nevertheless facts and there are socially defined procedures to determine them. They are presented in statements of the form ‘in this society it is the duty of agents in position x to perform actions y in circumstances z’, or ‘it is a norm of this society that in circumstances z, agents in position x will do actions y’. Although they are general, these statements are not ‘law-like’ generalizations, nor are they ‘analytic’. Nor, finally, are they normative statements, though they report normative rules. They report facts, namely, that in certain circumstances, in certain (or perhaps all) human societies, certain norms hold sway.14 Determining their truth requires what TG calls ‘field work’, in the tradition of anthropology, guided by conceptual analysis (Nell 1998: chs 3, 4). Institutions must be reproduced; they only exist, continue to exist, if they are supported and reproduced. The conditions for successful reproduction define the limits or boundaries of the society, economy. This is not the place, nor is there space, to offer more than a quick outline, one that might suggest the directions in which a Transformational Growth inquiry takes us. As we indicated above, two broad kinds of institution can usefully be distinguished, service or producing institutions, which carry out the various functions of society – producing food, clothing, shelter, arranging transport and entertainment, providing defense, promoting physical, mental and spiritual health, effecting cures, and so on. Positions in these are filled by appointment of qualified persons. On the other hand, there are the socialization and education or training institutions. People pass through these, acquiring qualifications which enable them to be appointed to positions in the institutions that produce, defend and maintain the fabric of society.15 Service or producing institutions, then, carry out the activities that enable the social system to maintain and reproduce itself, while the socializing and training institutions prepare the next generation. Such a system can be modeled in greater or less detail, showing how it works. That is, the internal relations between the roles and the institutions can be simplified and presented all together, abstracting from inessentials, and organized to allow us to focus on whatever questions are uppermost. Here TG distinguishes between ‘bearers’ and the variables they bear (Nell 1966). Bearers are institutions (as defined by conceptual analysis and field work), variables are (measurable) characteristics of the bearers and their activities. Households are the bearers of the variables Consumption and Disposable Income. Firms are the bearers of Output and Investment. Models are made up of relationships between variables, but these

Critical Realism and Transformational Growth 89 relationships are constrained by the fact that they must be consistent, collectively, with the conditions for the continued existence of the set of bearers. For example, we might divide output into capital goods and consumer goods (and services), then show how capital goods and labor are used in producing capital goods and consumer goods; how output is sold for investment and consumption; and sales revenue divided into wages and profits. Then we might fit all this together, first into Social Accounts, then, by making some simple hypotheses about what the various kinds of expenditure depend on, into a model that determines the level of output and employment. Wait a minute! ‘Determines’? Earlier we agreed with CR that the economic system, and the social order generally, must be considered essentially open. How can we have a determinate model? Nothing easier – and it is done all the time. The model shows how the system works, how it is supposed to work, and it may well also be used to show how it might break down, and what sorts of things might cause it to do so. No one can predict exactly how a given car is going to run, but the engineer’s design tells us how it is supposed to run. No one expected the World Trade Center to implode, but the architect’s plans can be used to reconstruct how it happened. Determinate models show us how the system, whether designed or emergent, is supposed to work, how it all fits together and produces the outcomes that keep it going. But that doesn’t mean that such models tell us what will happen. Other outcomes are possible, parameters may change, relationships may shift, innovations may take place. But the model, if it is a good one, shows us how the structure hangs together. Let’s develop this. If our conceptual analysis shows that the underlying mechanisms have quantitative aspects – and our field work gives us some idea of what the relationships look like – then we need to set up quantitative models. We want to understand how the mechanisms work, so we need to study them ‘in isolation’. The system is open, but to see how a given mechanism works it is necessary to set it apart, to examine it on its own, by artificially holding other factors constant. This is Marshall’s method of ‘ceteris paribus’. We know the other factors are not constant, and we know that our mathematical assumptions are only approximations. But we want to see how the system works. So we apply the model, using it to interpret the world. We compare what the model tells us to what happens, as far as we can measure that statistically. In this way the model can be assessed, its explanatory power determined. Its shortcomings will become apparent, and ways of improving the model may be suggested. But it is also likely that the model itself will call for new ways of looking at the world, pointing to some new dimensions to measure. This will bring more information, with which to develop the model. The improved model may then suggest even more new aspects to measure, bringing still further improvements in the model.

90

Edward J. Nell

In other words, we enter on an interactive interrogation of the world, using the model to understand what is happening, and using that understanding to improve the model (Block and Majewski 1998). We approach the truth; we can get a picture of it, but we never capture it. Let’s consider an example of such a form of ‘explanation’. TG contrasts two general ‘models of adjustment’ of capitalist societies. Each is abstract and quite general, but nothing has been ‘idealized’. Each is presented ‘mathematically’, although there is no pretense that the functions could be fitted to data. But it is claimed that the functions correctly represent directions of variation, and rough relative orders of magnitude. The first claims that the early capitalist societies, running up to World War I, were characterized by a stabilizing ‘price mechanism’, accompanied by a stabilizing but unreliable monetary and financial system, which periodically broke down. That is, the price mechanism worked ‘countercyclically’ to enable these economies to adjust to various kinds of external shocks; the banking system complemented this, as the interest rate also moved countercyclically. By contrast, after World War II, there is little trace of such a mechanism, but in its place there is an unstable or volatile mechanism, the ‘multiplier-accelerator’, which actually amplifies external shocks. Stabilization is provided by the government, partly through its sheer size (a new feature), partly through the fact that its budget is weakly countercyclical, and partly through discretionary policy. Contrasting these models creates a framework in which many different features of early and late capitalism (underdeveloped and developed economies) can be classified and analyzed (Nell [ed.] 1998). Moreover, the countercyclical adjustments in the markets of early capitalism created important pressures on business. Profits were squeezed, bringing risks of bankruptcy; to counteract that, businesses innovated in ways that made costs more flexible. The combined effect of these innovations was to change the system, moving it towards the second pattern of adjustment. Next, given that reproduction will never be perfect, there will always be variations, ‘mutations’, if you like, and this leads to a process by which institutions evolve. That is, changes will take place, and more successful forms will tend to be preserved and passed on. In a limited sense, this has to be the case. The system has to be reproduced; the environment of the society/economy will be changing, and there will always be slight variations, ‘mutation’, in the design and production of goods, services, skills, training and so on. Some of these will be better suited to the changing environment, some worse. They will compete, in short, and over time the system will gradually change. In addition, however, the active mind will be at work. There may be large, deliberate changes. For institutions can be and often will be modified by intentional, organized action – although the results will very often be quite different from what was intended. But in these cases also, selection procedures will pick the variations that survive and are passed along. So institutions develop, evolve.

Critical Realism and Transformational Growth 91 Selection may be based on biological reproduction – those institutions survive which enable the group to reproduce more effectively than groups organized by other institutional forms. Or selection may be based on something else. The market is a selection process; it evaluates how successful technologies and business practices are. Those which are profitable are supported and flourish; those which are not, however aesthetically pleasing or morally sound, will perish. Of course a society which is profitable will also be successful in biological reproduction. The system evolves, grows, and increasingly comes to manage, even dominate, the material world. Productive power and control over nature increase. The ability to shape the material environment, including human health and physical well-being, develops in many diverse directions. When this vast increase in our abilities is understood in conjunction with the fact that the underlying social structure rests on obligations, such an expansion in our powers has a significant moral implication. For as Kant pointed out, ‘ought’ implies ‘can’ or, alternatively, ‘cannot’ implies ‘not ought’. There is no duty to do what we cannot do. But as we come to be able to do more and more, our duties correspondingly increase. Indeed, they have to be redefined, re-examined and rethought. Whole new areas may open up, as when we come both to understand and to control social processes, making us – arguably – responsible for the impact on the population. (A social policy benefits one group, but is statistically certain to cause a small number of deaths. We know this; how are we to make the call?).16 Institutions are related to one another in a complex that permits the social order to be reproduced and to expand and change. But how this complex works may be anything but transparent. Those who fill the principal roles, and make the crucial decisions, may not themselves fully understand how the system works. For the working of the system as a whole may be quite different from the way the various parts work (Nell 1996). In economics this may give rise to various ‘paradoxes’, of which the famous ‘paradox of thrift’ in macroeconomics is perhaps the best known (Nell 1996: ch. 5).17

Conclusions CR takes some important steps toward developing a foundation for social and economic inquiry. TG can deepen its philosophical foundations by studying CR. TG’s ‘conceptual analysis’ can be deepened by considering CR’s transcendental arguments, and its exploration of ontology. The emphasis on ontology, and starting from the conditions necessary for the social world to exist, is surely right on track. But CR needs to develop its idea of structures and intransitive objects further. The methods for identifying these objects need to be specified and explained. Different kinds of objects and mechanisms need to be identified and classified.

92

Edward J. Nell

Causality in these mechanisms can only be superficially similar to that in natural science, so the working of the mechanisms needs spelling out. Moreover, CR could go further in the study of change, and might be encouraged to explore some of the issues opened up by TG. Several major opportunities seem to open up. First, CR could do more to examine the basic nature of the underlying structures in social systems, in particular the extent to which they rest on promises and obligations. Second, for these structures to continue to exist they must be reproduced, which requires interdependence (hence obligations). Models of reproduction could be examined; the solutions of such models will not be any conventional kind of ‘equilibrium’. Besides reproduction, however, social systems have characteristic ways of working, and these do not usually depend on ‘maximizing’ or ‘economizing’, though such behavior may be important at times. Studying how obligations and commitments influence behavior in the context of a social system that regularly ‘reproduces/expands/etc.’ may open up many new directions and provide important new approaches to social explanation. Finally, this leads directly to the study of the problems inherent in the way a system ‘works’ at any time; given the active mind, such problems can be expected to generate innovations, and this sets up a process leading to evolutionary change.

Notes 1

2

3

4

Thanks to M. Artabili, Will Milberg and Paul Lewis for helpful comments. Paul Lewis also provided editorial suggestions and drew my attention to a number of overlapping concerns between CR and TG that I had overlooked. TG employs transcendental arguments, though it doesn’t so label them. That conceptual truths exist is established by considering the denial; if true it would be a conceptual truth (Nell 1998: Ch. 3, esp. 88–9). The characteristics of agents are established in similar fashion – but must be filled out by field work. Lawson himself at one point claims merely that ‘the evidence suggests that it is an open system’, pointing to ‘fifty years of econometric failure’ to find ‘event regularities’ Lawson (1999a: 221). However, in Lawson (1997: 174–88), and in Lewis and Runde (2002), it is argued that social science must rest on human agency, and that human agency implies the possibility of genuine choice, including innovation. This is close to the idea of the ‘active mind’ in Transformational Growth. Hollis and Nell (1975) charged that neoclassical economics rested its defense of its unrealistic assumptions on positivism – if it predicts well unrealistic assumptions don’t matter. But in taking this line, the mainstream created a dilemma for itself in regard to rationality. For if a rational analysis is correct, then if agents do not behave in conformity with the model, it is their behavior, not the model, which is wrong. The model does not predict, it prescribes. But if the assumptions are unrealistic, the model will not apply, so its strictures on what should be done are irrelevant.

Critical Realism and Transformational Growth 93 5

6

7

8

9

10

11

12 13

Interestingly, this form of neoclassical thinking is not so easy to dismiss as positivist. Indeed, it may be said to look for the underlying structures that generate tendencies. The predictions that it comes up with rest on rational calculations and the processing of information, clearly fallible activities, as befits ‘tendencies’ (cf. Hands 1999). However, see Lawson’s reply to Hands (Lawson 1999b: 223–33). Lawson (1997: 71–85) touches on some of these points (in reference to Lucas). Lewis and Runde (2002) also address questions of agency and intersubjectivity. Critical realists argue that theorizing can proceed by abstraction alone; it is not necessary to employ idealizations/fictions. Why, then, do orthodox economists do so? The answer, critical realists argue, is that the idealizations employed by orthodox economists are the result of the need to construct models in accordance with the dictates of deductivism. Identified and re-identified in Strawson’s sense (Strawson 1959). Institutions are particulars in relation to the social and economic variables of which they are the bearers (e.g. households, firms, banks in relation to consumption, investment and lending), but they cannot be identified and re-identified ‘on their own’, so to speak. They must be identified in relation to persons, and their acts. We can distinguish at least a large and a small sense of ‘institution’. Banks, corporations, labor unions, government agencies, schools, universities and hospitals are ‘institutions’ in the larger sense. Not walking on the grass, talking in a whisper in libraries, shaking hands, and saluting the flag are institutions in the small sense, which will not be considered here. Making a promise is an act that changes social relationships. This adds a dimension to the earlier discussion of the necessity of ‘openness’. Can making promises, and keeping them, be fully determined? Here is a potential counter-example: consider the act of making the only promise you will ever make, namely, ‘never to keep a promise’. Could this be the determinate outcome of a model? If so, it must of course be clear just what you do: do you keep your only promise, or not? (Do you, the Liar, for once tell the truth, after all, as you acknowledge your lies?) And you can’t wriggle free by claiming you do both, since then you would do nothing. They cancel. But you acted, you made a promise, or an assertion. You had an intention to act. Promises are the basis of contract, of credit, of obligations and duties. If a deductivist model cannot predict promises and keeping of promises, it cannot successfully model social behavior. It would be possible to appeal here to ‘transcendental’ arguments to show that a social system must rest on valid expectations, and that these, in turn, must rest on promises (Nell n.d.). As for economics, one need merely think of the importance of money, interest and credit, of ‘advances’, of ‘discounting’, and so forth, all of which rest on promises. Wittgenstein paid special attention to this set of problems (cf. Kripke 1982). TG finds this a point at which social inquiry differs importantly from the examination of natural systems. In social analysis it must always be possible to distinguish what an agent does from what happens to the agent. No such distinction has to exist in Physics, nor in most Chemistry. In Biology, however, it begins to be important, and it is fundamental in the social sciences.

94 14

15

16

17

Edward J. Nell Positivism and most allied forms of empiricism will have trouble classifying statements of this type. They are not empirical generalizations about behavior, nor are they analytic truths, nor are they normative directives. Some roles are taken on almost automatically at certain stages of life by virtually everyone who is ‘normal’. Thus one is born into a family and arrives to maturity at the appropriate age – one becomes a man or a woman, an adult, where these can be thought of as generic roles, enabling the person to become a candidate to take on a specific role, say, warrior or priest or wife and mother, for example. Consider food inspection, or meat-packing regulation. Greater regulation increases costs, and the higher cost leads to poorer diets in lower income groups, with a consequent rise in illness and malnutrition. Less regulation permits more contaminated food to get through to the supermarkets, resulting in a predictable number of outbreaks of food poisoning and a small but predictable number of deaths. How are we to balance these? These issues pose practical problems, with important and new moral dimensions; modern economies have to work them out and come up with answers. For example, the Theory of Transformational Growth argues that the value of (commodity) money was expressed in terms of an ‘invariant measure’, by way of the Golden Rule, even though no one knew what this was or how it worked. It also argues that the Craft Economy was (partially) stabilized by an automatic ‘price mechanism’, which worked whether or not anyone was aware of it (Nell 1998, 2004).

References Block, T. and R. Majewski (1998) ‘Methodology and Historical Macroeconomics’, in E.J. Nell (ed.) Transformational Growth and the Business Cycle, London and New York: Routledge. Eriksson, R. (1998) ‘Was Keynes a Realist?’, Review of Political Economy, 10: 175–98. Fleetwood, S. (2001) ‘Causal Laws, Functional Relations and Tendencies’, Review of Political Economy, 13: 201–20. Hands, D.W. (1999) ‘Empirical Realism as Meta-method: Tony Lawson on Neoclassical Economics’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. Hollis, M. (1994) The Philosophy of Social Science, Cambridge: Cambridge University Press. Hollis, M. and E.J. Nell (1975) Rational Economic Man, Cambridge: Cambridge University Press. Kripke, S. (1982) Wittgenstein on Rules and Private Language, Cambridge, MA: Harvard University Press. Lawson, Tony (1997) Economics and Reality, London and New York: Routledge. —— (1999a) ‘Critical Issues in Economics as Realist Social Theory’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. —— (1999b) ‘Developments in Economics as Realist Social Theory’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge.

Critical Realism and Transformational Growth 95 Lewis, P. and J. Runde (1999) ‘A Critical Realist Perspective on Paul Davidson’s Methodological Writings on – and Rhetorical Strategy for – Post Keynesian Economics’, Journal of Post Keynesian Economics, 22: 35–56. —— (2002) ‘Intersubjectivity in Economics: A Critical Realist Perspective’, in E. Fullbrook (ed.) Intersubjectivity in Economics: Agents and Structures, London and New York: Routledge. Nell, Edward J. (n.d.) ‘Consciousness and Contradiction’, unpublished MS. —— (1966) ‘Variables, Laws and Induction’, unpublished MS. —— (1996) Making Sense of a Changing Economy, London and New York: Routledge. —— (1998) The General Theory of Transformational Growth, Cambridge: Cambridge University Press. —— (ed.) (1998) Transformational Growth and the Business Cycle, London and New York: Routledge. —— (2004) ‘Monetizing the Classical Equations: A Theory of Circulation’, Cambridge Journal of Economics, 28: 173–203. O’Neill, O. (1996) Towards Justice and Virtue, Cambridge: Cambridge University Press. Parsons, S. (1999) ‘Economics and Reality: A Philosophical Critique of Transcendental Realism’, Review of Political Economy, 11: 455–66. Runde, J. (1996) ‘Abstraction, Idealisation and Economic Theory’, in P. Arestis, G. Palma and M. Sawyer (eds) Markets, Unemployment and Economic Theory: Essays in Honour of Geoff Harcourt, Volume II, London and New York: Routledge. Samuelson, P.A. (1947) Foundations of Economic Analysis, Cambridge, MA: Harvard University Press. Strawson, P. (1959) Individuals, London: Methuen.

5

Critical realism and econometrics: an econometrician’s viewpoint Clive Granger

Introduction I will present my personal reactions to the ideas concerning critical realism as presented in the paper ‘A Realist Theory of Economics’ by Tony Lawson (1994), the book Economics and Reality (Lawson 1997), and the collection of essays edited by Steve Fleetwood (1999) in the book Critical Realism in Economics: Development and Debate. These will be called ‘the paper,’ ‘the book,’ and ‘the collection’ respectively. I will concentrate on the paper, which I find to be more precise and quotable and usually easier for me to understand, given my background. I do not view myself as an economist but would accept the classification of being an econometrician, although some of my colleagues may consider me to be more of a statistician. In any case I do have an affinity for economic data and strongly believe in the value of its use when analyzed properly. I am more familiar with the use of data than with economic theory. As with anyone, I can only draw on the accumulation of my lifetime world experiences, including what I have been taught, read, or discovered in my research which has been both theoretical and empirical. In the paper Lawson introduces a new viewpoint, critical realism, which is defined in the book, pages 157/8. The usual procedure is followed by first attempting to show that the presently used methods are deficient and in comparison the new ideas are clearly superb. As usual, both arguments are overemphasized; the old methods are probably not as bad as stated and the new not as good, as will be argued below.

What is reality? To an econometrician the question of reality can be equated with asking about the data generating mechanism. As we observe new data every month, or over some other period, there must be some mechanism that generates it. To ask what forms this data, it is convenient to start with the belief that economics is basically a decision science. Therefore, the fundamental building blocks are the decisions made by large numbers of

An econometrician’s viewpoint 97 decision makers in the economy: consumers, investors, employers, and policy makers. In a modern developed economy there will be several million such agents; in the United States there are roughly 200 million of them. The decisions will be made largely independently; that is, without interactions between the agents. They are based on the available information at the time the decision is made and these information sets may well contain common components. The components can include interest rates, tax rates, publicly announced policies, and expectations. It follows that the decisions are not statistically independent. The process by which a decision is made may be adequately discussed by microeconomic theory, or it may not. It should be noted that the information set that is an input to these decisions is dynamic, with evolving contents that will contain both previous decisions and their outcomes. When I choose what to have for lunch one day, I will remember what I ordered on previous days and how much I enjoyed those choices. An important difference between natural sciences, such as Physics, Astronomy, Meteorology, or Chemistry, and any social science, such as economics, is that the individual elements in the social sciences (the agents) are independently acting decision makers with different objectives, experiences, needs, information, and desires. In my opinion, it is thus clear that there is a distinct difference between the two types of sciences; that there is no reason to believe that the rules or laws of the natural sciences or the methodologies developed for them are necessarily relevant for social sciences. It is remarkable how many times discussions of economic methodology use examples from the physical sciences, such as falling leaves and gravity. Such examples are much less convincing than would be examples from a social science. The old argument amongst economic philosophers is to ask if economics is a science? This is surely one of the least interesting questions one can ask. Will the answer impact on how we do economics? The data generating mechanism (DGM) is only an abstract concept used by econometric theorists as the ultimate aim of an asymptotic analysis. It is not considered an achievable structure or objective. Because it is so complex, there is even some debate about its existence. Finding the truth is not considered to be a plausible objective for an empirical analysis in most cases. The considerable attention paid to the topic of the true model in discussions of economic philosophy and methodology has little relationship with the reality of such analysis. The DGM is only an approximation to the generator of the economy which has to describe the process by which each decision is made, and so is inherently unobservable. Even if one could observe the consumption of every good, for example, the amount of information would be overwhelming and so it is usual and sensible to aggregate. The data available to econometricians for analysis, particularly in the area of macroeconomics, has usually undergone a number of reconfigurations, such as temporal and cross-sectional aggregations,

98

Clive Granger

seasonal adjustments, and various filters and redefinitions. It follows that any model built can only hope to be an adequate approximation to the DGM. The aim of the modeling exercise is to form a sequence of improving approximations as more data becomes available and better model specifications are discovered. In correspondence, Tony Lawson points out that there is ‘a difference between an object of analysis and an analysis of an object.’ I certainly agree, but in my opinion there is an extra complication for the DGM, which an econometrician would like to be an object of analysis but which is virtually never attainable as such. One can only obtain, by analysis of outputs of the object, the DGM models that are hoped will approximate to it. We cannot realistically expect that the model of the process, that is the DGM, will be the process that is being modeled. Nevertheless, it is hoped that economically useful approximations can be achieved, in terms of better decisions, by our modeling techniques based on data analysis. Decisions are made by many different people at many different places and at different times. The ‘macro economy’ does not strictly exist; there is no place and time one can go to and watch it in operation. What is called the macro economy is an approximation of the core elements of the aggregations of the components of this economy. A stock price generated on the New York Stock Exchange, for example, comes from a market at a known location. The time at which it is generated is known and recorded with little uncertainty. Prices generated on an electronic exchange do not have a specific location but may have a known time of generation and this is going to be the usual marketplace of the future. A further difference between the natural sciences and the decision sciences, such as economics, is whether the variables exist in continuous time or not. Time itself is obviously continuous, or should be taken to be so, and most physical activities and chemical reactions occur in continuous time. However, decisions cannot be made continuously; by their basic nature they are discrete in that information has to accumulate for a new decision to be made. In some circumstances, decisions can be made frequently but this is still not strictly speaking continuous time. In some parts of economic theory, particularly finance, it is convenient to use the mathematics of continuous time but the resulting models will only approximate the actual discrete data. In many ways the approximation can be a useful one, but not all features go through from continuous to discrete time.

Out with the old … The paper on page 258 states: ‘Even accepting that contemporary economics has proven to be notoriously unsuccessful over the last fifty years or so, is not the proper solution just to try that little bit harder?’ Leaving aside the ‘proper solution’ comment for the moment, is it really

An econometrician’s viewpoint 99 true that conventional economics has been such an obvious failure? It obviously depends on the criteria being applied. If one expects to forecast every event perfectly, to know before every crisis what policy to apply to prevent it and to ensure that all economies grow steadily with both low inflation and low unemployment, for example, then economists can be judged failures. The conclusion is less clear if more realistic criteria are used. An alternative perspective comes from the observation that every major university in the world has a department of economics and for many of these universities the quality and ranking of their departments are important. Economics is an area of study frequently chosen by both undergraduate and graduate students, this last because there is a continuously strong demand from industry and government for their knowledge and talents. There is a Nobel Memorial Prize in economics. There are more than 220 academic journals covering various parts of the subject as well as many popular journals and newspaper sections. If economics were such a failure or a fraud, would it really get such attention? I also find most of my colleagues excited by the success of their profession; for example, forecasts of policies about deregulation, uses with complicated public auctions, helping economies recover relatively quickly these days from financial crisis have been attributed to useful economic knowledge. I am sure that real economists can provide many other examples. Is economics just a pretend (or pseudo) science whose secret will eventually come out? I doubt it. When did you last see a university department in Alchemy, Phrenology, or Astrology? These real world facts do not mean that there is nothing wrong with the present form of economic theory or empirical modeling. Constructive criticism can only be helpful, provided researchers are open to suggestions.

… and in with the new It is useful to replace something that is claimed to be unsatisfactory with something that is very much better. If critical realism is adopted as the approach to economics, according to Lawson (1994: 282), ‘human emancipation emerges as a real possibility’ and ‘policies and strategies can be formulated … with the aim … of replacing structures that are unwanted with others that are needed and empowering; to facilitate, in other words, a greater or more desirable or equally distributed range of human opportunities.’ Few people would not approve an approach that could achieve such aims, but it would not be considered a failure if they were not all completely reached, which is more realistic. As economic theories and models are human constructs, the question arises whether it is best to discard all that has been constructed in the past and then to start again or to build on the best of the old material and to continue with new ideas. To discuss this question, it is worth considering

100

Clive Granger

the particular criticism of the current methods, which I will do in the next section.

Economic regularities Lawson criticizes orthodox economics for relying too heavily on rules such as ‘whenever event X then event Y,’ where such regularities can be either deterministic or stochastic. He uses the term ‘deductivism’ to describe the logical sequence that leads to such rules. I will call them just one aspect of a regularity. The point is made that such event regularities are not pervasive (1994: 261), and that they ‘are hardly ubiquitous’ (1994: 262) but ‘arise only in rather special circumstances.’ There are a number of other similar comments in both the book and the paper suggesting that economists cannot rely on the apparent regularities that they believe they have found. Of course it would be as ridiculous to imply that there are no regularities in economics as it is to claim that any regularity discovered is stable and reliable (see Lawson [1997: 204] on ‘demi-regularities’). The economy is effectively stochastic and so will be any relationship and thus it may not hold at all times or hold only to a varying extent. It is worthwhile discussing a few simple examples of regularities. From its definition an unemployment rate will always be positive, and so a forecast of this variable will be positive. This is an example of a regularity, although not a very interesting one. If we consider inflation rates, most observed values are positive, although not quite all. Again forecasts will be positive and this is a useful fact from an empirical regularity. Of course inflation is very difficult to forecast, so that other useful regularities are not claimed. An area where there appears to be considerable useful regularity involves the impact of seasonality. Unemployment rates, for example, change seasonally as the school year ends and so high school and university graduates enter the job market in line with the varying demand for workers in seasonal industries. Many other major economic series contain important seasonals which provide examples of a useful regularity. ‘Seasonably’ is not mentioned in the book’s index but then neither is ‘microeconomics’ nor ‘macroeconomics’ nor ‘industrial’ or ‘international economics.’ Lawson in both the paper and the book also argues against causality and forecastability, as though they never occur. Again an idea with some merit is oversold. People who have an inoculation have a lower probability of getting a particular disease; those who smoke have a greater probability of getting cancer; if a company increases the price of its product when other prices are stable, there is likely to be a decrease in sales. There is ample empirical evidence available to support these and other regularities. In each case a stochastic form of causality, and similarly of forecast, is used, as is now common amongst philosophers. A rather trivial form of regularity occurs in the collection. If a paper has the phrase ‘critical realism’ in the title, it will

An econometrician’s viewpoint 101 usually contain the term ‘transcendental’ in the text as well as the names ‘Lawson’ and ‘Bhaskar.’ This is an empirical regularity, not always reliable, not based on a theory, and not particularly useful, but frequently occurring. Other examples are given later. If one agrees that steady regularities do exist, it still allows the possibility that they can be improved. Other apparent regularities are quite unstable and require a different viewpoint for further investigation. Thus, if critical realism can provide a new perspective, it will prove to be valuable.

The econometrician’s viewpoint Consider the following two descriptions of a modeling format: (a) ‘whenever event of type X then event of type Y’ or more generally ‘whenever event X then event Y always follows, as long as condition e holds’ (a stochastic form can be used); or (b) ‘explanation entails drawing upon antecedently established knowledge of relatively enduring structures and mechanisms (rather than revealing them), and investigating the matter of their joint articulation in the production of the novel event in question.’ The first statement is associated with the ‘classical’ economic approach; the second with the proposed new methodology. Statement (a) might remind an econometrician of a simple bivariate regression of the form X = a + bY + e, which is a relationship discussed and estimated in early chapters of fairly elementary textbooks. In my opinion statement (b) is much closer to the methodology used or at least attempted by modern dynamic econometricians. For that reason the new methods should be attractive to empirical modelers.

Essential aspects of the new viewpoint Lawson lays out the essential differences between the original ‘empirical realism’ and the new ‘transcendental realism’ in a table on page 263 of the paper. No similar table appears in the book, I believe. In the ‘empirical realism’ part of the table there are just two sections entitled ‘actual’ and ‘empirical.’ In the ‘transcendental realism’ part these two sections are repeated identically but there is a new first section called ‘deep’ which contains as objects ‘structures, mechanisms, powers, and tendencies.’ Clearly, it is these components that are viewed as the core elements of the new viewpoint. The argument is made, particularly in the paper, that the regularities that classical economic models attempt to utilize are disrupted, and are thus destabilized, by leaving out the effects of these deep components. From the econometrician’s viewpoint, what properties should we expect these deep variables to have? Either they should be always ponderous, slowly changing, or they could have occasional

102

Clive Granger

substantial changes if there is a sudden generally agreed swing to a new viewpoint by a society, for instance. Examples of deep variables might be global warming, population levels, political swings from left to right and back, or level of capital investment stock. An example of variables with regularities but slow responses is educating young women in a developing country, thus eventually lowering birth rates and increasing women’s productivity; or a company spending money on research and development which will eventually produce superior profits. It is clear that these deep components cannot fluctuate rapidly, as they would then be confused with other unobserved, quickly changing, almost white noise, components such as various exogenous shocks and measurement errors. Such fast moving variables hinder estimation of an existing regular relationship but do not produce unreliable results if one has enough data. Naturally, if the regularity is small and hidden in a large amount of noise, it will be difficult to find, but this is an old, well-known point and not the one that Lawson is making. Let us take population size as an example of a deep variable. It is possible that some economic relationships may change as population changes due to scale effects; either parameters could change or functional forms. As population is slowly moving and observable, an econometrician would consider it to be plausible to model these changes either based on some theory or using a time-varying parameter technique. If the deep parameter was suspected but not observed, there are many techniques that can be used to explore its existence and approximate its effects. Recent journal articles on time series analysis are frequently concerned with structural breaks, breaking processes, regime shifts, timevarying parameter models, non-parametric curve fitting, and long memory components both in mean and variance. It could be argued that these techniques, which often appear to be successful compared to simpler models, would at least partly capture or approximate an unobserved deep, slow-moving component with occasional large value changes. I am not implying that no difficulties exist; it is not easy to model the impact of a slowly changing component such as aggregate consumption or GNP measured quarterly on a rapidly changing, frequently observed series, such as daily or hourly stockmarket rates. Such a relation is expected, in both directions, and should be found by a careful analysis but this is not an easy task when building a highfrequency model for stock prices, say. Nevertheless, I would conclude that if there are important deep variables, econometric tools exist to analyze them and their impact on possible relationships and thus many of them will already have been found and approximated. In that sense, the new economics of critical realism is already here. It is true that if the new viewpoint is accepted and if specific deep variables can be identified and measured to some extent, the quality of modeling will greatly improve. However, it will still be based on regularities, but now with a better

An econometrician’s viewpoint 103 specification. I am not sure if that is what Tony Lawson wants to occur as a result of his criticism of the current economics.

What economy? The paper, the book, and the collection all have titles that emphasize the interaction of critical realism with economics, but their contents pay little attention to this relationship. In this they may be continuing a tradition of economic philosophy of placing greater emphasis on philosophy than on economics. The indexes of the book and the collection do not contain terms such as ‘microeconomics’ or ‘macroeconomics.’ Several of the essays in the collection make no, or virtually no, mention of economics. However, Boylan and O’Gorman (1999: 137) say ‘the development of the “crisis” in contemporary economics follows from the breakdown of the post-war Keynesian paradigm during the course of the 1970s.’ Here, and elsewhere in these contributions, the economy seems to be equated with the macroeconomy, whereas most economists work elsewhere, for example on industrial, labor, investment, financial, and international trade questions. This lack of vision, together with the few specific applications, is a continual disappointment. In the collection Hands (1999: 175–8) quotes Arrow and Hahn (1971) as an example of economic practice; a twenty-eight-year-old book, even though a classic, is hardly likely to be considered to be mainstream economics in 1999. The only clear exception to this criticism is Ingham (1999), who discusses money, successfully criticizing the classical approach but being much less clear in his attempt to use the new approach to suggest a better theory.

About econometrics Even if economic philosophers have usually been indefinite about the economics that is being criticized, they seem to be in general agreement that econometrics is often at fault and there is a widespread dislike of economic forecasting. These views can be found in the book, the collection, and in places such as Deborah Redman’s Economics and the Philosophy of Science (1991) and Mark Blaug’s The Methodology of Economics: Or How Economists Explain (1992). A not untypical citation in Blaug (1992: 246) is as follows: ‘Victor Zarnowitz (1968) sums up present day achievements in GNP forecasting in the United States in these words: “the record of economic forecasters in general leaves a great deal to be desired”.’ The quotation illustrates two points: a renowned forecaster, such as Zarnowitz, is prepared to criticize his area of study; and the use by the philosopher of a twenty-four-year-old quote described as ‘present day’ whereas, in fact, it may be totally out of date. If one wants to be critical of an area of research, an obvious strategy is to find respected members of the field who are being constructively critical themselves,

104

Clive Granger

such as Sims, Learner, and Hendry (in econometrics), and emphasize the critical comments from them without mentioning the corrective or improved techniques they propose. I find this unbalanced presentation of an active debate is sometimes reported by economic philosophers. In the early years, the attitude of econometricians was supposed to be to take a fully specified model from a theorist and just estimate any unknown parameters as well as possible using relevant data. This strategy was usually not satisfactory as the resulting model did not fit the data in many important ways. The econometrician has to search for models with a wider specification and which include the model based on the theory as a special case, and then estimate parameters. It is certainly true that stochastic regularities are required to discover these specifications that consider relationships between economic variables, rather than events. The criticisms of econometrics found in books by philosophers concern a subject rarely recognized by modern econometricians and thus have little relevance or impact. This is not the place to defend the methodology of econometrics, nor would I want to attempt to do so. Not all of the methodology used is satisfactory by any means, and one would be trying to defend a moving target. Econometrics remains a dynamic, vital, and hopefully improving area of research. Some of the improvement occurs because leading practitioners criticize the methodology and suggest improvements. The area has provided a number of empirically successful studies, although not many can be explained in simple terms. One early success in econometrics, which leads to a clear regularity, deserves a mention. This is the random walk model for prices from speculative markets. This is found to be correct, to a close approximation, so that returns are unforecastable, at least linearly, and again to a close approximation. This was found to be true for both current and old price data from various stock and commodity markets, for example. Both high and low frequency data, such as daily and monthly, were used. Originally proposed by Bachelier in 1900 and confirmed by empirical studies in the early 1960s, the theorists eventually used the idea as the basis of the ‘efficient market hypothesis.’ The lack of forecastability of returns is a regularity; it continues to be observed, almost without exception, for many studies using a variety of techniques, different data sets, and different time spans. The regularity is not unstable or irregular. It was discovered by empirical studies without the use of an initial economic theory. Many other apparent regularities have been found, such as the positive autocorrelations for volatility measures associated with speculative prices, but this relationship is less stable, or more stochastic. It is sufficiently reliable to be the basis of many risk control techniques used by investment companies. This is not to say that in the future improved techniques may not be found.

An econometrician’s viewpoint 105 A specific criticism of econometrics occurs in the book on page 71 and is an interesting one. It is called the ‘Lucas Critique.’ A quotation from an article by Lucas in 1976 starts with ‘the most striking [sign] is the indifference of econometric forecasters to data series prior to 1947.’ The quotation goes on to say that theory suggests that ‘more observations always sharpen parameter estimates and forecasts’ and yet data ‘from 1929–1946 are rarely used as a check on the post-war fits.’ To put this in context, if you had a theory with associated estimated models about a specific actual economic question for country X, would you expect the same model to be useful for the same question in another country? Except in financial areas, the answer is usually no; at least different parameter values are needed and quite possibly a different specification. The reason is that the countries differ in institutional structure, in tastes, in culture, in human capital, and so forth. The same reasoning holds true when comparing a country at two distant time intervals. Is a country in the period 1926–1946 really very similar in technology, and in other basic properties, to itself in the period 1979–1999? It is because laws, institutions, and most other aspects of an economy change over thirty years that econometricians no longer concentrate on fixed parameter models. As stated before, a great deal more attention is now paid to flexible specifications, including time-varying parameters, regime shifts, and structural breaks, which could be thought of as being more in phase with the critical realist approach.

Evaluation Once a theory has been determined or a model specified and estimated, that is not the end of the process. It is still necessary to evaluate the theory or model, to decide if it is any good or if it is better than some specific alternatives. The comparison is actually easier to do; one just needs an appropriate and acceptable criterion. It is virtually impossible to evaluate a model without establishing its purpose, and the purpose has to be specified exactly. An inexact purpose, such as ‘explaining’ or ‘understanding’, is not useful as one cannot know if the purpose has been achieved. The purpose for which an evaluation is best developed is that of forecasting, which is discussed in my book (Granger 1999). The idea is proposed that a decision is based on a forecast and an economic measure of the value arising from the decision can be used to evaluate the forecast. In particular, it can be used to compare alternative decisions based on different sets of forecasts. Here, a forecast is not taken to be just a mean or expectation but rather a predictive distribution. It is common practice in economics to welcome promising new techniques and methodologies, but they are not accepted until they have been evaluated by practical implementation. Using this criterion, critical realism should be welcomed as an interesting development with

106

Clive Granger

potential. However, it cannot be fully accepted until it is shown that it can produce solutions to some realistic, actually occurring economic questions. These solutions should be shown to be, in some pragmatic sense, superior to the solutions obtained by the older fashioned economics that we are being advised to reject. This will be the critical development in this area, to my pragmatic mind.

References Arrow, K.J. and F.H. Hahn (1971) General Competitive Analysis, San Francisco: Holden-Day. Blaug, M. (1992) The Methodology of Economics: Or How Economists Explain, Second edition, Cambridge: Cambridge University Press. Boylan, T. and P. O’Gorman (1999) ‘Critical Realism and Economics: A Causal Holist Critique’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. Fleetwood, S. (ed.) (1999) Critical Realism in Economics: Development and Debate, London and New York: Routledge. Granger, C.W.J. (1999) Empirical Modeling in Economics: Specification and Evaluation, Cambridge: Cambridge University Press. Hands, D.W. (1999) ‘Empirical Realism as Meta-Method: Tony Lawson on Neoclassical Economics’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. Ingham, G. (1999) ‘Money Is a Social Relation’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. Lawson, T. (1994) ‘A Realist Theory of Economics’, in R.E. Backhouse (ed.) New Directions in Economic Methodology, London and New York: Routledge. –—— (1997) Economics and Reality, London and New York: Routledge. Redman, D. (1991) Economics and the Philosophy of Science, Oxford: Oxford University Press.

6

Critical realism and feminist economics How well do they get along? Edith Kuiper

Introduction1 In the journal Feminist Economics Lawson (1999) and Harding (1999) initiated a debate over the contribution of critical realism as a way out of some major problems in feminist (economic) theorising. Lawson focuses on what he perceives as the main feminist response to the universalisation of values, namely the rejection of all generalisation. Feminist theorists could profit from critical realism, states Lawson, but do not get beyond naive realism (Lawson 1999: 27). Feminist economists’ focus on the epistemological side of matters should be replaced by a stronger focus on ontology. Harding is sceptical, especially on this last point, as she suggests that the feminist focus on epistemology has a strong strategic basis. More generally, she states that assumptions about reality and ontological arguments have strategic value, and she stresses the importance of gaining insight into how a particular set of ontological beliefs fit with other parts of networks of beliefs, moral and political stands, etc. (Harding 1999: 132). Besides the issues of universalisation and generalisation there are several more concerns that feminist theorists and economists and critical realists share. In feminist economic research in general the conceptualisation of fundamental concepts such as the economic agent, gender, and the relation between gender and knowledge must be addressed. In research on economic science in particular, issues like the relation between economic science and society and the role of power in feminist economic research require theoretisation. Some solutions and approaches of feminist economists to these issues share critical realist features such as the acknowledgement of distinct ontological levels in society and in science, as identified by Harding, Scott and others. On the other hand, feminist economists bring in perspectives and issues that may cause concerns for critical realists. It is the aim of this chapter to make these issues more explicit and elaborate them further. This chapter focuses on some central issues in feminist economics and poses the question of how (if at all) feminist economics may benefit from

108

Edith Kuiper

critical realism in dealing with these issues. What has critical realism to offer when we want to understand the approaches taken by feminist economists in their research and the direction in which feminist economics is going? Should feminist economists perhaps turn to the critical realist perception of ontology to come to a shared perception of reality and to direct the further articulation of a feminist economic research programme? What are the limits of the usefulness of critical realism for feminist economic research? What can critical realists learn from feminist economists in their assessments of the economy and economic science in particular? I discuss these issues in relation to a research project on a relevant topic in feminist economics: the relation between gender and the history of economic science. More specifically, the research project I dwell on deals with the impact of notions of femininity and masculinity on economics, in particular on the structure and the content of economic theorising through the history of economic thought.

Some central issues in feminist economic research Feminist economics took off as a more or less distinct field of research in the early 1990s. By then researchers in a range of countries all over the world who were dealing with ‘women’s or gender issues’ came to the understanding that orthodox economic theory and research had a tradition of misrepresenting women and gender relations and that this had fundamental implications for most if not all economic theories (see e.g. Ferber and Nelson 1993). At the first international conference on feminist economics (1993 in Amsterdam) most contributions addressed the dominant, neoclassical economic paradigm. Critiques were directed at methodological research, conceptual and theoretical research, empirical research, historical research and policy research, and new empirical and theoretical research was presented. It took until the late 1990s before feminists had built a research tradition and a body of literature that made it possible to reflect on feminist economic research in more general terms and to come to epistemological statements on the way feminist economists conduct research and to ontological statements on the way gender structures economic science (see Nelson 1995a; Longino 1998; Hewitson 1999; Jacobsen 2003). In their research feminist economists apply various conceptualisations of sex and gender and most of them define gender on one specific level of economic reality and/or economic research.2 Julie Nelson, for instance, sees gender as a basic metaphor in economic research (Nelson 1995b); Folbre (1994) regards gender as part of the structure of constraints; and Elson (1991) sees gender differences as the outcome of social processes. Gender is thus generally defined at the level at which the research is directed and where power is located. Do these differences stem from

Critical realism and feminist economics 109 differences in contextual perspectives or strategy and have they to be perceived as (fundamentally) incompatible, or can they be compared and related to one another in one more general framework? An important issue here is the under-theorisation of the gender concept. What would an analytical concept of gender in economics look like? Should it perhaps, as critical realists suggest, be based on a shared concept of ontology? While feminist economists work at getting new and revolutionary topics on the economist’s agenda, economic science is increasingly perceived as coinciding with the rationalist approach to individual behaviour that typifies neoclassical economics. As the neoclassical orthodoxy has claimed all of economics, so too it has claimed relevance for all aspects of life. The feminist response to this tendency has been twofold: either to work within the neoclassical programme and to press towards the extension of this theoretical approach to an even greater extent; or to make use of other methods, concepts and theories that do not fit the neoclassical programme at the risk of being considered as noneconomic and excluded from the economic discussion. Thus feminist economists are confronted with the issue of what economics is and should be about,3 and how to achieve and attain a position in the field. Similar, and for feminist economists as important, is the definition of the field of study of feminist economics.4 In order to deal with such a question it is necessary to have an idea about how to assess when and to what extent gender is relevant, also if not especially in relation to other aspects like class, race and nationality. Can critical realism perhaps provide feminist economists with the tools required to unravel the difficulties surrounding this issue? A third issue I want to mention here is related to what Harding (1986) calls ‘the science question in feminism’ – the fundamental relatedness of science and masculine values and/or the Western male subject – which makes the use of (economic) science as a tool for women’s empowerment problematic because of the risk of reproducing these very masculine values. Lawson’s suggestion that a critical realist ontology can provide feminist economics with both a scientific and shared set of notions about economic reality puts the issue of the possibility, desirability and necessity of a feminist epistemology and ontology of economics firmly on the table, together with the question of the extent to which these would have critical realist features. Along with these more general issues, there are three more issues I would like to take up, which are more specifically linked to the conceptualisation of the relation between gender and (economic) science. Describing the main approaches of feminist researchers, Harding (1987) distinguishes ‘feminist empiricism’ that accepts the positivist and objectivist perceptions of knowledge and merely adds women and women’s issues to their analyses; ‘feminist standpoint’ that conceives women’s social, cultural and biological experience as fundamentally

110

Edith Kuiper

different from that of men, and which has major implications for the way science has to be conducted; and ‘feminist postmodernism’ that challenges the assumptions of the epistemological project upon which modernist approaches to science, including feminist empiricism and the feminist standpoint, are based. In her valuable contribution to feminist economics, Harding (1995) also theorises the importance of marginal perspectives to science for obtaining a ‘strong objectivity’ (rather than ‘weak objectivity’ that is posited as coinciding with value-neutrality). How can we build on this and come to an understanding of the relation between gender and economics? And can critical realism promote such understanding? Most feminist economic methodological research focuses on the content and realism of assumptions and the value-ladenness of economic concepts.5 Until recently postmodernist feminist economic analyses addressed mainly the content and dominance of neoclassical economics.6 Feminist research on the history of economic science shows three main lines of research, the first being the research on the writings of economists such as Adam Smith, John Stuart Mill and Alfred Marshall on women and ‘women’s issues’.7 The second line of thought addresses the work of women economists and their economic research.8 The third line of research investigates the gendered character of general economic concepts such as capital, investment, productivity and efficiency.9 To theorise and investigate the relation between gender and the emergence of economic concepts and theories we need to attain insight into the way economic concepts have been constructed in relation to gender within their specific social and academic historical context, and into the role of gender in the development of economic science over time. Can critical realism provide us with a framework for the analysis of the relation between gender and economics in which feminist and orthodox economic science are perceived as interrelated and mutually constitutive approaches to science and as being both time and context dependent? A specific issue that deserves attention in this respect is the impact that the conceptualisation of the Western Modernist Subject has had on economic science. The rationalist perception of economic research still dominates both feminist and critical realist discussions of epistemology and ontology: economics is predominantly regarded as a rational endeavour, conducted by rational and autonomous economists. This leads to suggestions about alternative definitions of the economic agent without, however, acknowledging the social and academic embeddedness of the current concept of the rational agent, which hampers the development of effective strategies for change. Since critical realism does not challenge Western Modernist thought in this respect, the question here will rather be what this limitation of critical realism means for feminist economics, and what feminist economists might contribute in this respect.

Critical realism and feminist economics 111 The last issue I want to mention concerns the problems centring on the notion of power in (feminist) economic research. In most feminist theory the notion of gender is perceived as inherently and explicitly linked to that of power. Though gender processes and power processes can be distinguished, they are often intertwined. In the ongoing negotiation processes in which gender is given content and meaning, power plays an important if not decisive role. This also accounts for the construction and use of (gender) concepts in economic research. The theoretisation and understanding of the role of power in economic research, and in feminist economic research in particular, is not yet very well developed in feminist economics. Critical realists claim that their approach acknowledges and theorises power in the economy. Can it also illuminate the workings of power in (feminist) economics?

What can critical realism mean to feminist economics? We are thus left with three general issues: A) the theoretisation of gender, B) the definition of (feminist) economics, C) the possibility and desirability of a feminist epistemology and ontology of economics. On the relation between gender and economic science there are three more specific issues: D) the theoretisation of gender in a changing science and society, E) the impact of the Western Subject, F) the conceptualisation of power in (feminist) economic research. Before turning to a discussion of the contribution of critical realism on these issues, let us first recall the main propositions of this approach to economics. Critical realism as articulated by Lawson (1997) builds on Bhaskar’s theory of transcendental realism about the natural world, according to which the objects of science research are ‘structured’ (irreducible to events) and ‘intransitive’ (existing and acting independently of their being identified) (Lawson 1997: 28). It articulates an ontology that assesses the existence of three ontologically distinct but related realms: ‘the empirical’ (experiences and impressions); ‘the actual’ (actual events and states of affairs) and ‘the real’ (structures, powers, mechanisms and tendencies). Critical realism extends this approach to the social world, aiming at theorising society scientifically. It conceptualises social reality as a hierarchically structured and open system, in which social structure and human agency are seen as both distinct and highly interdependent. Though to a certain limited extent capable of free choice, human agents are located and situated in social structures (social rules and relations). The position agents take in these structures determines their responsibilities and range of action (see Lawson 1997: 164, see also Lawson 1999: 33). Social structures exist independently of the individual’s perception and actions, but are continually reproduced and modified by human agency in general (Lawson 1997: 169).

112

Edith Kuiper

To recognise that the social sciences require an intrinsically dynamic perspective, Lawson stresses the importance of getting away from the old notions of creation and determinism to work instead with the concepts of reproduction and transformation. Social change is seen as an ongoing process, which consists of both accidental changes and necessary changes ‘as an integral part of what the system or object in question is’ (Lawson 1997: 171). Reproduction and transformation are perceived as the (partly unintended) consequence of human action, as conditioned by underlying structures, rules, mechanisms and tendencies. ‘The primary aim of science is the production of knowledge of mechanisms that, singularly or in combination, produce the phenomena that are actually manifest’ (Lawson 1997: 26). The knowledge of these mechanisms will enable social scientists and policy makers to work towards more desirable outcomes. According to Lawson, social science perceived as such ‘provides a perspective on science, nature, society and economy that is not only explanatorily powerful but also able to preserve the intuition that human social history is explicable and yet actively made’ (Lawson 1997: 289). Let us now, after this brief and general summary, turn to a more specific discussion on how critical realism bears upon the issues in feminist economics identified above. A The theorisation of gender In most mainstream research the attention paid to gender does not go much further than the indication of whether being male or female is statistically significant in relation to the subject matter. Most (feminist) economists address ‘women’ and ‘men’ as unproblematic statistical categories, although they are based on biology. The introduction of ‘gender’ as a social construction, in opposition to ‘sex’ as fixed and based in biology, in the 1970s made it possible for feminist academics to study gender differences as the (changeable) result of social processes. In her discussion of feminist economics as a field of research, Hewitson (1999) distinguishes between a) empirical feminist economists who strive for equality between women and men, taking science and economic theory as given, b) feminist economists who perceive science and economic theories as gendered and as linked to a masculine perspective on the economy, and c) those feminist economists who see sex/gender differences as produced by a range of discourses. In Women’s Studies’ discourse ‘gender’ refers in a general manner to the content and meaning given to masculinity and femininity. Sandra Harding defines gender more specifically as ‘an analytic category within which humans think about and organize their social activity rather than as a natural consequence of sex difference’ (Harding 1986: 17). In order to outline and discuss the possible extensions of the economists’ conceptual toolbox concerning gender, I would like first of all

Critical realism and feminist economics 113 to distinguish various levels in feminist economic research. As shown in Figure 6.1, level 0 is what could be referred to as ‘reality’; it is here that we see the revealed differences between women and men (it may perhaps be best compared to what critical realists indicate as the realm of the actual). Level 1 is how economists think about these differences, how they make explicit use of perceptions of gender or apply these as tacit knowledge while constructing their economic arguments (comparable to the realm of the empirical in critical realism). The last level (level 2) is the level of analytical concepts used to assess and address how people think or how to think about differences between women and men (since critical realists make no distinction between levels 1 and 2, this level would also be considered as part of the realm of the empirical; that of experience and impression). Feminist economists use the term ‘gender’ to address the state of affairs concerning gender differences (level 0) and also to address the workings of gender in economic theorising, as for instance is done by Nelson (1995b) (level 1). ‘Gender’ is also used as an analytical concept by theorists like Scott and Harding (level 2). To distinguish between these uses of the term ‘gender’, I use gender in its general meaning of ‘notions of masculinity and femininity’. Notions of masculinity and femininity can of course be applied at all these three levels. At level 1 the use of gender is more precisely referred to as ‘perceptions of gender’ and ‘the use of sex/gender distinctions’, at level 2 we speak of ‘analytical gender concepts’. When we continue with the conceptualisation of the analytical gender concept, the works of feminist theorists are the logical candidate to turn to. Joan Scott, Sandra Harding and others introduced analytical gender concepts that distinguish various levels, and these concepts enable one to investigate the way notions of masculinity and femininity enter scientific reasoning. Generally three levels are distinguished: gender symbolism, gender structure and individual gender (see Harding 1986). Sometimes four levels are indicated and discussed: gender symbols, normative gender notions, gender structure and gender identity (Scott 1986). There Level 2:

Analytical concepts of gender

Analysing the way people, or groups of people, perceive gender Level 1:

Perceiving gender

Applying sex/gender distinctions and notions of masculinity and femininity Level 0:

Gender differences

Figure 6.1 Levels of abstraction in conceptualising gender

114

Edith Kuiper

is no hierarchy in the importance of these levels, nor is it indicated how these various levels precede or impact on each other. Scott stresses that the notions of gender applied and constructed at these various levels are interrelated in the sense that they confirm or counteract each other. Gender, thus defined, is a descriptive rather than an explanatory concept (Davis, Leijenaar and Oldersma 1991: 5). Echoing Lawson’s account of critical realism, Scott states that the process of negotiation about gender notions at these various levels is an open process. The actual outcomes of the mutual impact of the working of gender at these levels can therefore only be stated a posteriori. In feminist economics, as elsewhere in the social sciences, operationalising this analytical gender concept is complicated because of the distinct ontological character of these levels. Instead of including all these levels in the analyses, many researchers reduce the other levels to that of gender structure and/or only pay attention to those levels that seem directly relevant depending on the issue under scrutiny (see also Tonkens 1998). Critical realism’s emphasis on ontology might prompt feminist economists to discuss more explicitly the status, use and theoretisation of these levels. A more explicit perception and use of these levels could enrich feminist economic research, as it would stimulate researchers to take all these levels more systematically into account. B The definition of (feminist) economics Lawson (1999) refers to the generalisation and universalisation that has been paramount in economic science from a masculinist and Western perspective. Lawson claims that feminist economists have withdrawn into contextualism. While this may be the case in other social sciences, I have my doubts about whether this is indeed the dominant response of feminist economists; both empirical and theoretical postmodernist research is still rather scarce in feminist economics, as is feminist research on the history and sociology of economics (see Lawson 1999: 45). Feminist economists apply a range of theories and methods in their research, which is not the same as turning away from generalisation. Moreover, there is a tendency for feminist economists – like feminist theorists and neoclassical economists – to understand (almost) everything from one specific perspective. The tendency to address almost everything from a gender perspective is a logical one as the alternative in economics seems to be to understand nothing from that perspective. This, however, tends to overstate the gender aspect, especially in relation to other aspects like race, class and nation. Thus the relevance, content and meaning of gender in social processes should be investigated rather than assumed or constructed according to specific traditional lines, as has been done in orthodox economics. Feminist researchers are aware of the contextuality of the definition of

Critical realism and feminist economics 115 gender they apply; what is defined as femininity and as masculinity is different for economists and, say, football players. This, however, is not yet reflected in the theoretical approaches applied. Sandra Harding, for instance, states that the hierarchical and asymmetrical definition of gender has a universal character (Harding 1986: 18). Julie Nelson (1992, 1995b) does a similar thing when she defines the hierarchical and asymmetrical perception of gender as a hierarchical dualism basic to economic science. To investigate the relevance of the way gender is constructed and given meaning in science and in society at large, both an analytical gender category and a theoretical framework that enables investigation of these matters are required. Critical realism may provide a framework in which the content of these relations can be investigated; the relative positions of women and men can be assessed and related to race, class and (other) power relations. In her discussion of the definition of economics as a field of research, Nelson (1993a) suggests that economics should be defined not in terms of its methodology, but rather as the study of the process of provisioning. The discussion of what is to be perceived as feminist economics has only recently got off the ground, with suggestions ranging from a field of research directed to the improvement of the position of women (Strober 1994), to theories that stress its political character in general (Power, Mutari and Figart 2003), to definitions that perceive the improvement of economic science as an inherent part of feminist economics (Hewitson 1999). Seen in the light of this discussion of the usefulness of critical realism for the feminist issues on the table here, I can subscribe to Lawson’s suggestion that there are ‘possible advantages to feminist explanatory and emancipatory projects from engaging (or engaging more fully) in the sort of explicit ontological analysis associated with modern versions (at least) of scientific realism’ (Lawson 1999: 26). One would have to be cautious, however, about accepting a prescriptive ontology for feminist economics in this stage of the discussion, at which point opening up the discourse and the articulation of the central issues is still in progress. C The possibility and desirability of a feminist epistemology and ontology If you tell critical researchers that after criticism comes construction, and that new theory is not possible without criticism of what was, you are not telling them anything new. That feminist economists will shift from a focus on epistemology towards ontology is important, necessary and to some extent unavoidable if feminist economics is to become a mature field of study. However, a one-sided focus on ontology is neither required nor desirable. The focus on epistemology as we now know it in feminist economics has a particular aim: the need for a thorough understanding of

116

Edith Kuiper

how we have come to know through a masculine set of values and how these have had an impact on scientific research; in what way these values and theories have prevented us from understanding social and economic processes instead of illuminating them (see also Harding 1999). Working and stating an ontology without fully acknowledging these insights will throw us back into old traps that feminist economists are keen to avoid. Feminist economists are just finding out how masculine values have been of fundamental influence in the emergence of economics as a science. If the influence has been as fundamental as it currently seems, and if masculine values have had a major impact on the definition of the subject matter, the choices of methods and the construction of the main theoretical concepts, then the development of economics as a science cannot be well understood without paying attention to gender. What is clear, also from a critical realist point of view, is that epistemologies and ontologies that do not recognise gender as fundamentally structuring economic knowledge production contribute to the reproduction of traditional and hierarchical perceptions and relations of gender. Whether feminist researchers should strive for a feminist epistemology and/or ontology is an issue not yet decided in women’s studies. These discussions are now coming up in economics. In my view both a feminist epistemology and ontology are possible and required. This means that feminist economists and other economists should work towards an epistemology that acknowledges the importance of the various ways in which gender structures the production of economic knowledge and prescribes a practice in which these and other differences are taken account of as a requirement for ‘good economic knowledge’ (see also Peter 2001). The development of a feminist ontology and epistemology should go together; neither should be perceived as prior. Economic science would be perceived as part of society and the researcher as specifically positioned in that society. The variety of ontological perceptions that go with the variety of these positions would be acknowledged and accounted for. Such a methodology would not state certain gender relations or notions of gender as natural or as given but would acknowledge the variety of notions of gender and facilitate their investigation. Critical realism can offer feminist economists a way to theorise the economy and the development of economic science in real time, acknowledging the embeddedness of economics in social and economic structures through highly complex processes, which are to some extent irreversible. Especially useful for feminist economists is the notion that we are born into a situation we have no complete knowledge about nor a (complete) say in; that (some) structures are of a larger grade of complexity than agents grasp and/or were already in place before we or the agents under investigation came about. This is highly relevant for

Critical realism and feminist economics 117 feminist economists, who have to deal with a fully articulated academic tradition, and might help us to come to effective strategies for change. As it is now stated, however, the critical realist claim that the researcher has no influence on his or her topic of research is too strong, and needs adjustment and further theoretisation. Like other agents in society, an economic researcher’s perspective and impact on his/her topic of research are related to that researcher’s position and relations, and thus highly context and topic dependent. There is a difference between the scope of the influence of a Nobel Prize winner, a textbook writer and an applied researcher who works within the accepted analytical framework on an uncontested topic. These differences should be acknowledged and theorised. D The theorisation of gender in a changing science and society Turning to the theoretisation and investigation of the gendered character of economic science, we can say that both feminist theorists (see e.g. Harding 1986, Scott 1986, Haraway 1991) and critical realists (see e.g. Lawson 1997, 1999) perceive science as a social institution. Economic science can be investigated for the use made of social sex/gender distinctions and the construction of specific perceptions of gender, and its contribution to the reproduction and transformation of gender and to gender differences in society at large can be indicated. Though this endeavour can be pursued by taking a critical realist approach, the critical realist view on scientific knowledge does not require it, as does feminist economics, but rather discourages it. When we decide to conduct such research anyhow, we see that, although critical realists like Tony Lawson claim that competing theories and empirical statements will eventually be judged by reference to the empirical reality, critical realism at the same time makes it possible to scrutinise the practice of economic science and assess the way in which relations, positions and social structure influence scientific development. Critical realism gives us tools to investigate how the development of scientific thought is subject to social structures, rules, mechanisms and tendencies, which are difficult if not impossible for an individual or a (small) group to escape. It gives us tools to investigate how economic and gender concepts are used and become ingrained in social concepts and economic and political arrangements, which in turn set the limits for scientific development. Critical realism suggests that social, economic and political changes will induce changes in science and, in turn, scientific output also has its impact on society. In real time these changes will mostly occur in sequence. Scientific changes, for instance, follow after large investments in specific fields of research; technological changes will follow the constructions of new concepts and technology developed by science. By

118

Edith Kuiper

taking a historical perspective and addressing (economic) science as a social institution in real time, these interrelations between science and society at large can be indicated and investigated. The critical realist approach to time and change may also contribute to a more sophisticated view of change in relation to notions of gender and gender relations. Although the assigning of a fixed ontological status to sexual differences is rightly called essentialism,10 feminist economists and others are confronted with gender differences that cannot be changed overnight and sometimes not even over a lifetime. We are then looking at structures that have to be acknowledged as fixed when we consider the short term (at level 0 in Figure 6.1). Instead of regarding these differences as either biological or cultural, gender differences can also be recognised, as critical realism suggests, as fixed or flexible depending on the length of the time period taken into consideration. This implies that in the short run notions such as weakness, dependence, unrelatedness to others or, for instance, the tendency to take responsibility for one’s children may be considered as fixed, while in the longer run increasing areas of gender differences have to be considered as flexible and as such subject to change. Taking a long-term historical approach can be fruitful and even required where it concerns deeply ingrained structures. Thus major changes in gender identity, gendered concepts, gender structure and gender symbolisms can also be investigated and shown in their interrelation. E The impact of the Western subject The specific character of the Western masculine subject concerns not only Rational Economic Man – representing the object of research – but also and not least the scientist or scholar him/herself. The features of rationality, isolation and autonomy are not only assumed to apply for the economic agent, but are also ascribed to the economist him/herself.11 The impact of this last statement has not yet been fully thought through in economics, unlike in other sciences. It is especially here that critical realism leaves us, and joins the other Modernists (see also Barker 2003, Peter 2003b). The masculine character of rational economic man has been discussed and indicated by many. Nelson (1995b) suggests a person in relation as an alternative to the rational economic agent. Lawson (1997) similarly suggests a person in relation and relative to his/her position. Both thus restate the homogeneous individual – now complemented with the feature of having relationships – that stands in the tradition of the Western Subject without the acknowledgement of its fundamental masculine character. Lawson even goes so far as to state a common human nature (if perceived at a high enough level of abstraction) based in biological features (‘genetic constitution’) (Lawson 1999: 46). It is this perception of human agents that Lawson proposes to take as a starting point for social improvements and

Critical realism and feminist economics 119 social and economic policies. Thus the traditional Western perception of human agency (free choice, human rights, dichotomies between subject and object, rationality and emotionality), both as object and as subject in economic research, is kept intact and reproduced. The perception of the role of knowledge and academics in society that is linked to that of the Western Subject is one that posits science as able to decide on matters outside the political realm based on empirical arguments; as value and context free and as helping academics and politicians to instruments and arguments that are technical rather than political to improve society. Lawson may attempt to rebut this criticism by claiming that critical realism regards science as a social institution like any other. His epistemology, however, is overtly optimistic about the scope for comparing and evaluating theories and ontologies on the basis of empirical research (Lawson 1999: 44; see also Peter 2003b). As others have pointed out in greater detail, critical realism does not theorise the social practices in science through which scientific knowledge comes about (Peter 2003a), nor does it acknowledge the economic, social, political and personal interests involved in the reproduction of bodies of thought, as is the case in economics (Barker 2003). As it now stands, critical realism does not recognise and theorise the structuring power of language in science nor the positions, power and (personal) interests of researchers in established bodies of thought that importantly determine who will be heard and who will not be heard in the academic discussions (Barker 2003, Peter 2003a). This, however, requires more, not less, epistemological research. F The conceptualisation of power As for economics in general, power has been a problematic concept for feminist economics. The link between ‘gender’ and ‘power’ has been more implicit than theorised. Sandra Harding describes gender as ‘an asymmetrical category of human thought, social organization, and individual identity and behavior’. She also indicates that ‘[m]asculinity requires the conception of woman as “other” … Femininity is constructed to absorb everything defined as not masculine, and always to acquiesce in domination by the masculine’ (Harding 1986: 55, emphasis as in the original). In this perception, the use of gender as a distinction between masculinity and femininity is considered to be always of a hierarchical and asymmetrical character. According to Harding, it is the use of this asymmetrical category of human thought at various levels in society – together constituting a sex/gender system – that brings about the social inequality between women and men. Harding’s analytical gender concept implies a power aspect and at the same time does not acknowledge changes in perceptions of masculinity and femininity, as it considers a specific definition of masculinity and

120

Edith Kuiper

femininity as applicable to all cultures.12 Joan Scott stresses the importance of the historical gender context, the use of language, the various ways in which gender is given meaning and the role of power in these processes. In ‘Gender: A Useful Category of Historical Analysis’ (1986), Scott distinguishes two aspects of gender: the layered character of gender in the sense that gender is defined and confirmed at various points in social life, and its power aspect.13 Scott perceives the definition and use of notions of gender as means of (re)producing power relations. She does not, however, directly link power to the definition of masculinity (and femininity). Instead, she advocates that the outcomes of these processes be established empirically and be assessed only ex post. Critical realists distinguish between various kinds of power processes (e.g. the accumulation of capital versus labour), some of which bring about gender differences. It contributes to the clarity of the analysis to distinguish these from gender processes, through which power is reproduced. Thus the power of the agent can be perceived as diffuse, as an aspect of relations and as significantly constituted by the positions people occupy (see also Lawson 1997). However, the way critical realism conceptualises structures, power, mechanisms and tendencies – as a deeper level of reality – seems to be based in or at least highly related to the specific interests of researchers: as these structures and mechanisms contain the deeper level of reality that triggers specific human behaviour, uncovering these structures and mechanisms provides the researcher with power to change this behaviour and, ultimately, events. This is a matter of the interests and power of the researcher, an aspect of critical realist ontology that remains implicit and cannot be assessed and reflected upon as the epistemological aspects of the critical realist approach have been shifted to the background. Though critical realism may yet be silent on this specific issue, and even go so far as to exclude these issues from the analysis of scientific research, it does provide us with a means to analyse economic science as a social institution. Below I report on such an endeavour: the project contains a historical and dynamic approach to economic science from a feminist/gender perspective, making use of critical realist notions of ontology to theorise, investigate and describe the relation between gender and economic science.

A research project on gender and economic science 1 The aim and scope of the project The research project I describe here addresses the impact of gender on economic theorising. It proposes an institutional and historical approach that acknowledges both the social character of economic science, scientific and social change, and the interaction between structures and agents in

Critical realism and feminist economics 121 economic science. To investigate the impact of gender on economics, economic science is perceived as a real time and irreversible phenomenon. Economic texts from three periods in the history of economic thought (early Political Economy, early neoclassical economics and the recent period) were scrutinised as empirical material. In this research I started from feminist theory and elaborated an analytical framework making use of critical realist notions. 2 Theorising gender, economic science and society To investigate the construction and use of gender in economic science it is necessary first to assess the use made by economists of sex/gender distinctions (level 1 in Figure 6.1). To address the way economists think about and deal with the differences between women and men and give meaning to gender, I distinguish two extremes: a biological concept of gender differences (regarding social, economic and physical differences between women and men as biological – as ‘natural’, unchangeable and as explainable by the sex difference between women and men) and a cultural concept of gender differences (regarding social, economic and physical differences between women and men as cultural – as the result of social and therefore changeable processes).14 These sex/gender distinctions should be distinguished from ‘perceptions of gender’ that not so much address the importance of the biological vis-à-vis the cultural differences, but instead refer to the associations and images ascribed to masculinity and femininity.15 The perception of gender can thus be investigated both where a biological concept of gender differences is used and where a cultural concept of gender differences is applied. A hierarchical and asymmetrical gender concept (see e.g. Harding 1986) thus contains a specific kind of perception of gender, namely one that perceives gender (femininity/masculinity) as ‘hierarchical’ – the masculine pole is more highly valued than the feminine pole – and ‘asymmetrical’ – the masculine pole is positively defined, whereas the feminine pole is defined as merely a negative opposite of the masculine pole (Harding 1986; Nelson 1992, 1995b). The hierarchical and asymmetrical perception of gender is, however, a specific and extreme perception of gender, whereas other forms and definitions in which ‘femininity’ is more positively articulated and valued are also possible. In the research discussed here, gender is regarded as being constantly (re)negotiated and (re)defined on multiple levels and in multiple locations in society in reference to women’s and men’s ideal, desired or achieved social position, characteristics and perceived role in biological reproduction. Gender and power are seen as related but also as separable. Social power processes, including gender processes, are perceived as together giving rise to gender differences (see also Brouns 1993; Schaapman 1995; Keuzenkamp 1998).

122

Edith Kuiper

Perceptions of gender and economic concepts are thus also conceived as being subject to an ongoing process of redefinition in (economic) science. This process of redefinition, however, is not seen as fully flexible but as to a large extent limited by accepted symbols, institutions and regulations, accepted knowledge and language, and unconscious behaviour and habits. These tend to (re)produce established gendered concepts and limit the extent to which a change in the perception of gender is possible. Economic science is conceptualised as a social institution, and as a distinct social institution. The aim of science – and at the same time its legitimisation – is here understood as its claim to produce objective or, rather, value-free knowledge, and scientific value-neutrality is taken to be based on the perceived division between society and/or politics and science.16 The internal development is addressed by acknowledging and indicating the differences between the perceptions of gender in science and those in society at large. Scientific concepts are analysed as being defined and given meaning both within the context of academic debate and traditions, and within the social and historical context. Besides rational considerations and scientific conventions, making use of constants outside (economic) science is seen as one of the strategies used in the negotiations over theories and gender.17 ‘Nature outside culture’ or ‘science outside economics’ (physics and other exact sciences) can be referred to and used as a foundation for theories. The biological concept of gender that posits gender differences as fixed is also used in this way. Table 6.1 shows the analytical framework and indicates that gender is perceived here as given content and meaning at various levels in society and in science, though not necessarily in the same manner. Scientific texts make use of both academic and social perceptions of gender. Though very interesting and relevant, the (role of gender in the) process determining which concept and/or theory becomes successful has not been investigated in this research project (for an example of such research, see Scott 1988). Instead, successful texts have been selected as outcomes of these processes. In line with critical realist notions about the Table 6.1 Gender in science and society Gender

Science

Society

Gender symbolism

Metaphors, basic concepts

Ideals, metaphors

Gendered concepts

Dichotomies, concepts

Social dichotomies

Gender structure

Academic and social rules, laws

Social rules and laws

Gender identity

Academic and social role models

Social role models

Critical realism and feminist economics 123 duality of structure, I suggest that once observations, concepts and theories have become accepted in science and, in a broader context, within society, these observations, concepts and theories in some cases attain the status of ‘facts’ and become naturalised. When the ‘fact’, ‘concept’ or ‘law’ has attained its status as value-free, its gender value-laden character is lost from sight. Standardised facts, concepts, and theories are subsequently used to build on, either by elaborating them, or by using them as support, or by arguing against them, thus setting the boundaries of what can be spoken of scientifically. Social and academic acceptance of concepts may prevent change where change is due. Some data, concepts or theories, however, are not only accepted within academia, they also become socially implemented and/or accepted in society on such a scale that, although based on understandings and explanations that are scientifically no longer perceived as valid, they are very hard, if not impossible, to change. It is here that social and academic power play comes in. One of the options that can be pursued by academics involved in these power processes is to supplement rather than change these concepts and/or theories. Other options are to replace whole sets of theories and/or introduce new techniques. The outcomes of these processes depend strongly on social and academic power positions (for instance, the command over financial resources). Since the use of sex/gender distinctions and perceptions of gender are expected to differ in various contexts and thus also between science and society at large,18 the social or contextual19 sex/gender distinction and perceptions of gender may coincide and/or reaffirm the internal scientific perceptions of gender. These perceptions, however, may also contradict one another. Though social perceptions of gender may change over generations, this takes time and changes are expected to show a lag before they are adopted in science. In the short run economic texts are expected to make use of some current but mostly lagging social symbols, structure, etc. In the long run, it is expected that (some) scientific concepts and products are used in society and, in turn, references to social changes and new perceptions of gender are made in science. The analytical distinction between society and science, and the analysis of economic texts from various periods in the history of economic science, make it possible to account for the reflexive effects of perceptions of gender between economic science and society. 3 Economic texts as empirical material A text emerges and is positioned by the author both in an academic tradition and in a social context. Economic texts are of a dual nature: they are a means in the negotiations over gender and power and (successful)

124

Edith Kuiper

texts are also the results of these negotiations. As many influences and considerations play a role in these negotiations, the outcomes can only be established a posteriori (see also Scott 1986; Lawson 1997). Concrete seminal texts are also historical representations that are constantly reread and reinterpreted by scholars. This is done within a different historical context from the one in which the text was written, where perceptions of economic and gender relations were different. Texts acquire content and meaning in a context of literary, cultural and social conventions, and are read by people who are themselves part of and socialised in these networks of conventions (see e.g. Meijer 1996). By focusing on texts, it is possible to investigate what was selected, used and stressed by the author, which is, albeit in various ways and to a varying extent, related to the (social and academic) position and relations of the author. By selecting texts by authors with varying positions in (economic) science, the impact of their different positions may be compared. The focus on texts provides the reader with concrete and limited empirical material, and makes the research conducted repeatable. Similar research can be conducted by (many) others, possibly also on other texts, which will provide us with a clearer and more detailed picture of the relation between gender and (the history of) economics. It can easily be extended and applied to address the issues of race and class and for the investigation of the interrelations between the (mostly implicit) use of these categories. 4 Empirical and theoretical results of the research project The analytical framework proposed in this research project provides a means for analysing and explaining the way gender enters economic reasoning, including economics texts that are generally regarded as gender neutral ‘since these do not address women’. The analysis of Petty ([1662] 1986), ([1672] 1986), ([1676] 1986), Smith ([1759] 1976), ([1776] 1979), Jevons ([1871] 1965), Marshall ([1890] 1930), Becker (1981, 1991) and Pencavel (1986) shows the use of a masculine perspective in all these texts. The use of gender notions could be indicated on various levels. Gender symbols, for instance, played a fundamental role in Petty’s work, gender identity in Smith ([1759] 1976), gender structure in Smith ([1776] 1979), the focus on the behaviour of men and the supposed role of women in Jevons ([1871] 1965) and Marshall ([1890] 1930). Assumptions about sex and gender differences and the use of standardised gendered economic concepts could be shown to provide the basis of reasoning in Becker (1981, 1991) and Pencavel (1986). The analysis makes it possible to trace the use in these texts of gender meanings as applied at various levels in science and in society at large. It

Critical realism and feminist economics 125 was thus possible to account for the use of gender symbols (Land and Labour, Nature and Man), next to references to gender structure (definitions of work, workers, productivity, etc.), established gender concepts and language (e.g. public–private dichotomy) and notions of gender identity, and show how these were inserted and applied to construct the economic argument in the text. To get a grip on the hidden gender meanings, knowledge of the historical and academic contexts is required and this brings in neglected aspects of economic history such as the economic position and contribution of women as producers inside and outside the family. It also sheds light on the economic use of notions of race and class in these texts. The analytical distinction between the academic and the social context made it possible to indicate the uses of different notions of gender and to assess the use and impact of various definitions of femininity and masculinity in economic science. In Petty’s texts, for instance, the social symbol of ‘Land as the Mother and Labour as the Father of Wealth’ structures his economic argument on value, while at the same time he uses the academic notion of Man as an isolated being as the starting point for his accounts (see Petty 1672: 183; Petty 1676: 286; see also Kuiper 2001). By addressing more periods over a long period of time, the change in perceptions of gender applied, and in the definition and use of economic concepts, could be investigated. Where Adam Smith, for instance, was involved in discussions around the public–private distinction and on women’s and men’s role in these realms, by the end of the nineteenth century these concepts were socially accepted to the extent that Marshall could base his definition of economic science on them.20 Similarly, established gender concepts, which were dealt with as naturalised notions, were shown to become basic concepts that economists later on used as building blocks for their arguments. Where, for instance, Jevons and Marshall referred to eugenic, sexist and racist notions about the motives of the labourer, Pencavel (1986) could lean on these established notions, expressed in mathematical formulae (see Kuiper 2001). In addition to these insights, a mechanism came to the fore that is here suggested as responsible for the reproduction of a masculine perspective on the economy in economic science. This mechanism contains the recurring and many times restated link in the history of economics between male economists, the historical notion of the Western Male Subject and the value-neutrality of economic scientific research and theorising. This mechanism works under the condition of the absence of women from the academic realm, i.e. the absence of objections to the claims concerning the identity of the masculine perspective with scientific value-neutrality.

126

Edith Kuiper

Conclusions What has this brought us on the identified issues in feminist economics? Can critical realism help feminist economists to make a shift to ontological issues and statements? Can it provide us with the means for the conceptualisation of gender and deal with the variety of meanings of femininity and masculinity that exist over the globe? Can critical realism indicate how to assess the relevance of gender to economics? And finally, can critical realism provide a way to conceptualise economic science to investigate the impact of gender, including the impact of the Western subject and power? Critical realism can indeed help feminist economists to become more aware of the ontology they are applying and to develop more articulate positions in this respect. It may also provide feminist economists with an analytical framework with which to assess and theorise about what it is that feminist economists are doing as a necessary and integral part of economic science, conceived as a social institution. It may provide feminist economists with a way to conceptualise gender in a more open manner and to analyse the use of gender in (economic) science and base their claims on empirical epistemological research, instead of on normative statements only. It offers a framework that acknowledges the multiplicity of positions and relations, and theorises social, economic and academic change. Like many feminist economists, however, critical realists are overly optimistic about the autonomy of the academic mind and the rationalist features of academic research. Most critical realists, like many feminist theorists and other economists, overlook the social and psychological interests of the academic person. Acknowledgement is required of the fact that even critical realists are subjects of gender and other power processes that take place in the process of generating knowledge. And the assessment and integration of such issues in a critical realist approach to science requires research on, for instance, the relation between the social position of the researchers and the knowledge produced. The fundamental influence of the academic conception of the Western subject on economic research, including critical realism and feminist economics, has not yet been thought through in economics. It may now be well established that the rational economic agent is a man in all his features, but we have yet to acknowledge fully the effects of four hundred years of academic investigation on the basis of the perceptions of male, upper-class white British and American economists. Nor do we know yet how to conduct economic research that does justice to considerations of gender, race, class, nation and power. As long as the features of ‘Western Academic Man’ remain fundamental to both feminist economics and critical realism, issues of epistemology will – hopefully – remain very high on the feminist economists’ agenda, as should also be the case for critical

Critical realism and feminist economics 127 realists. Until that issue is solved, understood and dealt with satisfactorily, feminist economists have their own path to follow and a substantial and fundamental contribution to make. In the meantime, feminist economists will use and apply the available and, for their context of action, relevant approaches. Critical realism is certainly one of them and as such worth exploring for its potential to elaborate feminist economic theorising on economic science and the economy.

Notes 

 



       





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µ)HPLQLVW3ROLWLFDO(FRQRP\¶VHH3RZHU0XWDUL DQG)LJDUW  6HHHJ)HUEHUDQG1HOVRQ DQG6HL]  6HHHJ)HLQHU 3RODQ\LDQG6WUDVVPDQQ DQG+HZLWVRQ  6HH HJ 3XMRO   5HQGDOO   )ROEUH   *URHQHZHJHQ   DQG )RUJHW  6HHHJ&RRSHU 'LPDQG'LPDQGDQG)RUJHW DQG*URHQHZHJHQ   6HHHJ)ROEUHDQG+DUWPDQQ )ROEUH 3XMRO 3LFFKLR  6HL] DQG%DUNHU  6HHHJ)HUEHUDQG1HOVRQ  /DZVRQ¶VVWDWHPHQWRQDERXWZKDWIHPLQLVWHFRQRPLVWVSHUFHLYHDVµWKHIXQGDPHQWDO GLVSRVLWLRQ RI PDOH HFRQRPLVWV¶ QDPHO\ µWR SRUWUD\ KXPDQ DJHQWV DV UHODWLYHO\ LVRODWHGVHOIFRQWDLQHGLQGLYLGXDOV¶ /DZVRQ LVWKHUHIRUHLQFRPSOHWH µ,Q YLUWXDOO\ HYHU\ FXOWXUH JHQGHU GLIIHUHQFH LV D SLYRWDO ZD\ LQ ZKLFK KXPDQV LGHQWLI\WKHPVHOYHVDVSHUVRQVRUJDQL]HVRFLDOUHODWLRQVDQGV\PEROL]HPHDQLQJIXO QDWXUDO DQG VRFLDO HYHQWV DQG SURFHVVHV$QG LQ YLUWXDOO\ DOO FXOWXUHV ZKDWHYHU LV WKRXJKW RI DV PDQO\ LV PRUH KLJKO\ YDOXHG WKDQ ZKDW LV WKRXJKW RI DV ZRPDQO\¶ +DUGLQJ  µ7KHFRUHRIWKHGHILQLWLRQUHVWVRQDQLQWHJUDOFRQQHFWLRQEHWZHHQWZRSURSRVLWLRQV JHQGHULVDFRQVWLWXWLYHHOHPHQWRIVRFLDOUHODWLRQVKLSVEDVHGRQSHUFHLYHGGLIIHUHQFHV EHWZHHQWKHVH[HVDQGJHQGHULVDSULPDU\ZD\RIVLJQLI\LQJUHODWLRQVKLSVRISRZHU¶ 6FRWW  7KHVHWZRFRQFHSWVFDQEHSHUFHLYHGDVWZRH[WUHPHVDVWZRHQGVRQDOLQHDOODW OHYHO  LQ )LJXUH   7KH FXOWXUDO FRQFHSW RI JHQGHU FRQFHLYHV RI GLIIHUHQFHV EHWZHHQ ZRPHQ DQG PHQ DV SXUHO\ VRFLDO DQG FXOWXUDO FRQVWUXFWV DQG VR PDNHV LW SRVVLEOH WR DFNQRZOHGJH WKH FXOWXUDO DQG KLVWRULFDO FKDUDFWHU RI GHILQLWLRQV RI µPDVFXOLQLW\¶ DQGµIHPLQLQLW\¶

128  





 

Edith Kuiper 1HOVRQ GHILQHV JHQGHU DV µWKH DVVRFLDWLRQV VWHUHRW\SHV DQG VRFLDO SDWWHUQV WKDW D FXOWXUHFRQVWUXFWVRQWKHEDVLVRIDFWXDORUSHUFHLYHGGLIIHUHQFHVEHWZHHQPHQDQG ZRPHQ¶ 1HOVRQE  +DUGLQJ PDNHV D GLVWLQFWLRQ EHWZHHQ WKH REMHFWLYLW\ RI VFLHQWLILF NQRZOHGJH DQG YDOXHQHXWUDOLW\ UHJDUGLQJ ZKLFK VKH VWDWHV WKDW µ>Z@KDW WKH GHIHQGHUV RI WKH IXQGDPHQWDO YDOXHQHXWUDOLW\ WKH SXULW\ RI VFLHQFH UHDOO\ PHDQ WKH\ VD\ LV WKDW VFLHQFH¶V ORJLF DQG PHWKRGRORJ\ DQG WKH HPSLULFDO FRUH RI VFLHQWLILF IDFWV WKHVH SURGXFHDUHWRWDOO\LPPXQHIURPVRFLDOLQIOXHQFHVWKDWORJLFDQGVFLHQWLILFPHWKRG ZLOOLQWKHORQJUXQZLQQRZRXWWKHIDFWXDOIURPWKHVRFLDOLQWKHUHVXOWVRIVFLHQWLILF UHVHDUFK¶ +DUGLQJ  $FFRUGLQJWR+DUGLQJ WKLVYLHZRIVFLHQWLILFODZVµQDWXUDOODZV¶ FODLPVWREH EDVHG LQ D UHDOLW\ RXWVLGH FXOWXUH ZKLFK LV SUHVHQWHG DV UHTXLULQJ LQWHUSUHWDWLRQ E\ PHDQVRIVFLHQFHLQRUGHUWREHDSSOLHGVRFLDOO\DQGSROLWLFDOO\2EMHFWLYLW\DVFODLPHG E\PRVWSRVLWLYLVWVDFFRUGLQJWR+DUGLQJGHPDQGVVFUXWLQ\RI DQ\WKLQJH[FHSWRI VFLHQFHLWVHOI 7KLV LV VLPLODU WR ZKDW +HOHQ /RQJLQR GLVWLQJXLVKHV DV FRQVWLWXWLYH DQG FRQWH[WXDO YDOXHV &RQVWLWXWLYH YDOXHV DUH LQWHUQDO WR WKH VFLHQWLILF FRPPXQLW\ DQG FRQWH[WXDO YDOXHVDUHSROLWLFDOVRFLDODQGFXOWXUDOMXGJHPHQWVDQGSUHIHUHQFHVVHH6HL] ±/RQJLQR>@  1%&RQWH[WXDOGRHVQRWPHDQ H[WHUQDO WRVFLHQFHDVVFLHQFHLVFRQVLGHUHGSDUWRI VRFLHW\ µ(FRQRPLFVLVDVWXG\RIPHQDVWKH\OLYHDQGPRYHDQGWKLQNLQWKHRUGLQDU\EXVLQHVV RIOLIH%XWLWFRQFHUQVLWVHOIFKLHIO\ZLWKWKRVHPRWLYHVZKLFKDIIHFWPRVWSRZHUIXOO\ DQG PRVW VWHDGLO\ PDQ¶V FRQGXFW LQ WKH EXVLQHVV SDUW RI KLV OLIH¶ 0DUVKDOO >@  

References Barker, D.K. (1995) ‘Economists, Social Reformers, and Prophets: A Feminist Critique of Economic Efficiency’, Feminist Economics 1: 26–39. —— (1999) ‘Coherence, Realism and Feminist Economics’, paper presented at the IAFFE Conference in Ottawa, Canada, and at the HES Conference in Greensboro, North Carolina. —— (2003) ‘Emancipatory for Whom? Comment on Critical Realism’, Feminist Economics 9: 103–8. Becker, G.S. (1981) A Treatise on the Family, Cambridge, MA: Harvard University Press. —— (1991) A Treatise on the Family: An Enlarged Edition, Cambridge, MA: Harvard University Press. Brouns, M. (1993) De homo economicus als winkeldochter: theorieën over arbeid, macht en sekse (The Homo Economicus as an Old Maid: Theories, Power and Gender), Amsterdam: SUA. Cooper, B. (1993) ‘Harriet Martineau’s “Embodies Principles”: Whose Body, What Principles?’, Paper presented at the conference ‘Out of the Margin: Feminist Perspectives on Economic Theory’, Amsterdam. Davis, K., M. Leijenaar and J. Oldersma (eds) (1991) The Gender of Power, London: Sage Publishers. Dimand, M.A., R.W. Dimand and E.L. Forget (1995) Women of Value: Feminist Essays on the History of Women in Economics, Cheltenham: Edward Elgar. —— (eds) (2000) A Biographical Dictionary of Women Economists, Cheltenham: Edward Elgar.

Critical realism and feminist economics 129 Elson, D. (ed.) (1991) Male Bias in the Development Process, Manchester: Manchester University Press. Feiner, S.F. (1995) ‘Reading Neoclassical Economics. Toward An Erotic Economy of Sharing’, in E. Kuiper and J. Sap, with S.F. Feiner, N. Ott and Z. Tzannatos (eds) Out of the Margin: Feminist Perspectives on Economics, London and New York: Routledge. Ferber, M.A. and J.A. Nelson (eds) (1993) Beyond Economic Man, Chicago: University of Chicago. Folbre, N. (1991) ‘The Unproductive Housewife: Her Evolution in the NineteenthCentury Economic Thought’, Signs 16: 245–55. —— (1992) ‘ “The Improper Arts”: Sex in Classical Political Economy’, Population and Development Review 18: 105–21. —— (1993) ‘How Does She Know? Feminist Theories of Gender Bias in Economics’, History of Political Economy 25: 167–84. —— (1994) Who Pays for the Kids?, London and New York: Routledge. —— and H. Hartmann (1988) ‘The Rhetoric of Self-Interest: Ideology of Gender in Economic Theory’, in A. Klamer, D.N. McCloskey and R. Solow (eds) The Consequences of Economic Rhetoric, Cambridge: Cambridge University Press. Forget, E.L. (1999) The Social Economics of Jean-Baptiste Say: Markets and Virtue, London and New York: Routledge. Gremmen, C.C.M. and J.A. Westerbeek van Eerten (1988) De kracht van macht: Theorieën over macht en hun gebruik in vrouwenstudies (The Force of Power: Theories on Power and Their Use in Women’s Studies), The Hague: STEO. Groenewegen, P. (ed.) (1994) Feminism and Political Economy in Victorian England, Aldershot: Edward Elgar. —— (1995) A Soaring Eagle: Alfred Marshall 1842–1924, Cheltenham: Edward Elgar. Gustafsson, S.S. (1997) ‘Feminist Neoclassical Economics: Some Examples’, in G. Dijkstra and J. Plantenga (eds) Gender and Economics: A European Perspective, London: Routledge. Haraway, D. (1991) Simians, Cyborgs, and Women: The Reinvention of Nature, London: Routledge. Harding, S. (1986) The Science Question in Feminism, Ithaca: Cornell University Press. —— (ed.) (1987) Feminism and Methodology, Bloomington: Indiana University Press. —— (1995) ‘Can Feminist Thought Make Economics More Objective?’, Feminist Economics 1: 7–32. —— (1999) ‘The Case for Strategic Realism: A Response to Lawson’, Feminist Economics 5: 127–33. Hewitson, G. (1999) Feminist Economics: Interrogating the Masculinity of Rational Economic Man, Cheltenham: Edward Elgar. Jacobsen, J. (2003) ‘Some Implications of the Feminist Project in Economics for Empirical Methodology’, in D. Barker and E. Kuiper (eds) Towards a Feminist Philosophy of Economics: An Anthology, London and New York: Routledge. Jevons, W.S. ([1871] 1965) The Theory of Political Economy, New York: August Kelley. Keuzenkamp, S. (1998) ‘Gender, macht en het emancipatiebeleid’ (Gender, Power and Emancipation Policy), paper presented at the congress Vrouwenstudies ev Vrouwenkwesties, Amsterdam.

130

Edith Kuiper

Kuiper, E. (2001) ‘The Most Valuable of all Capital.’ Gender Readings of Economic Texts, Amsterdam: Thela Thesis. Lawson, T. (1997) Economics and Reality, London and New York: Routledge. —— (1999) ‘Feminism, Realism, and Universalism’, Feminist Economics 5: 25–59. Longino, H. ([1987] 1996) ‘Can There Be a Feminist Science?’, reprinted in A. Garry and M. Pearsall (eds) Women, Knowledge and Reality: Explorations in Feminist Philosophy, New York and London: Routledge. —— (1998) Opening lecture at ‘Out of the Margin II/IAFFE Summer Conference’, Amsterdam. Marshall, A. ([1890] 1930) Principles of Economics, 8th edition, London: Macmillan. Meijer, M. (1996) In tekst gevat. Inleiding tot een kritiek van representatie (Captured in Text. An Introduction to a Critique of Representation), Amsterdam: Amsterdam University Press. Nelson, J.A. (1992) ‘Gender, Metaphor, and the Definition of Economics’, Economics and Philosophy 8: 103–25. —— (1993a) ‘The Study of Choice or the Study of Provisioning? Gender and the Definition of Economics’, in M.A. Ferber and J.A. Nelson (eds) Beyond Economic Man, Chicago: University of Chicago Press. —— (1993b) ‘Value Free or Valueless? Notes on the Pursuit of Detachment in Economics’, History of Political Economy 25: 121–45. —— (1995a) ‘Feminism and Economics’, Journal of Economic Perspectives 9: 131–48. —— (1995b) Feminism, Objectivity and Economics, London and New York: Routledge. Pencavel, J. (1986) ‘The Labor Supply of Men. A Survey’, in O. Ashenfelter and R. Layard (eds) Handbook of Labor Economics, Volume I, Amsterdam and New York: North Holland. Peter, F. (2001) ‘Rhetoric vs Realism in Economic Methodology: A Critical Assessment of Recent Contributions’, Cambridge Journal of Economics 25: 571–89. —— (2003a) ‘The Reality of Economic Rhetoric: Economic Knowledge and Epistemic Practices’, in D. Barker and E. Kuiper (eds) Towards a Feminist Philosophy of Economics: An Anthology, London and New York: Routledge. —— (2003b) ‘Critical Realism, Feminist Epistemology, and the Emancipatory Potential of Science: A Comment on Lawson and Harding’, Feminist Economics 9: 93–101. Petty, W. ([1662] 1986) ‘A Treatise of Taxes and Contributions’, in C. Hull (ed.) The Economic Writings of Sir William Petty, New York: August Kelley. —— ([1672] 1986) ‘The Political Anatomy of Ireland’, in C. Hull (ed.) The Economic Writings of Sir William Petty, New York: August Kelley. —— ([1676] 1986) ‘Political Arithmetick’, in C. Hull (ed.) The Economic Writings of Sir William Petty, New York: August Kelley. Picchio, A. (1992) Social Reproduction: The Political Economy of the Labour Market, Cambridge: Cambridge University Press. Polanyi, L. and D. Strassmann (1995) ‘The Economist as Storyteller: What the Texts Reveal’, in E. Kuiper and J. Sap, with S.F. Feiner, N. Tot and Z. Tzannatos (eds) Out of the Margin: Feminist Perspectives on Economics, London and New York: Routledge. Power, M., E. Mutari and D.M. Figart (2003) ‘Beyond Markets: Wage Setting and the Methodology of Feminist Political Economy’, in D. Barker and E. Kuiper

Critical realism and feminist economics 131 (eds) Towards a Feminist Philosophy of Economics: An Anthology, London and New York: Routledge. Pujol, M. (1992) Feminism and Anti-feminism in Early Economic Thought, Aldershot: Edward Elgar. Rendall, J. (1987) ‘Virtue and Commerce: Women in the Making of Adam Smith’s Political Economy’, in E. Kennedy and S. Mendus (eds) Women in Western Political Philosophy, Kant to Nietzsche, Brighton: Wheatsheaf Books. Schaapman, M. (1995) Ongezien onderscheid: n analyse van de verborgen achtswerking van sekse (Unseen difference. An Analysis of the Hidden Power Effects of Gender), The Hague: Ministry of Social Affairs. Scott, J. (1986) ‘Gender: A Useful Category of Historical Analysis’, The American Historical Review 91: 1053–75. —— (1988) Gender and the Politics of History, New York: Columbia University Press. Seiz, J. (1993) ‘Feminism and the History of Economic Thought’, History of Political Economy 25: 185–201. —— (1995) ‘Epistemology and the Tasks of Feminist Economics’, Feminist Economics 1: 110–18. Smith, A. ([1759] 1976) The Theory of Moral Sentiments, D.D. Raphael and A.L. Macfie (eds), Oxford: Clarendon Press. —— ([1776] 1979) An Inquiry into the Nature and Causes of the Wealth of Nations, R.H. Campbell and A.S. Skinner (eds), Oxford: Clarendon Press. Strassmann, D.L. (1993a) ‘Not a Free Market: The Rhetoric of Disciplinary Authority in Economics’, in M.A. Ferber and J.A. Nelson (eds) Beyond Economic Man, Chicago: University of Chicago Press. —— (1993b) ‘The Stories of Economics and the Power of the Storyteller’, History of Political Economy 25: 147–65. Strober, M. (1994) ‘Rethinking Economics through a Feminist Lens’, American Economic Review 84: 143–7. Tonkens, E. (1998) ‘Gender in welke lagen? Kanttekeningen bij de gelaagde concepten van Scott, Harding en Hagemann-White en hun toepassing in Empirisch onderzoek’ (Gender in What Layers? Remarks on the Layered Gender Concepts of Scott, Harding and Hagemann-White and Their Use in Empirical Research), Tijdschrift voor Gender Studies 1: 42–9.

7

The agency–structure model and the embedded individual in heterodox economics John B. Davis

I Introduction Two main differences distinguish mainstream and heterodox economics’ ways of understanding the individual. First, whereas mainstream economics treats individuals as atomistic beings, heterodox economics generally characterizes individuals as being embedded in networks of social and economic relationships. Second, whereas there is a fair amount of agreement across mainstream economists over what the atomistic conception of the individual generally involves, different schools of heterodox economists tend to disagree over how to understand the nature of the individual and individual embeddedness. Indeed, heterodox economists within different schools also often disagree over how individual embeddedness is to be understood. This suggests the need for developing a concrete strategy for investigating the topic of the embedded individual in heterodox economics. My strategy in this chapter is first to distinguish two broad approaches within heterodox economics to understanding the individual as a socially and economically embedded being, and then to attempt to show how a critical realist understanding of individual and society helps us understand these two approaches. Preliminary to doing this, however, I set out a general understanding of what it means to talk about individual embeddedness, within which these two broad approaches can be distinguished and explained. This threepart discussion – general understanding, two broad approaches, and particular heterodox schools – describes the overall order and structure of the chapter. The next two sections of the chapter set forth a general understanding of what it means to talk about individuals as being embedded in networks of social and economic relationships. Section II begins by surveying a collection of concepts used in describing embedded individuals that have been developed in the social psychology literature on the subject of the individual’s ‘self-concept.’ Section III then advances a general conception of the embedded individual based on a critical realist understanding of the relationship between individual and society. Here I draw on Lawson’s

The agency–structure model

133

Economics and Reality Part III agency–structure analysis to describe a complex dynamic between individuals and social-economic structures which transforms each as well as the relationship between them over time. Lawson’s analysis is general in that the form or character of the means by which individuals and structures each operate upon one another is left intentionally uninterpreted in the interest of characterizing the relationship between them in the broadest possible terms. ‘Structures’ for Lawson include not just more highly institutionalized social arrangements and social positions, but also uncodified, more informal social norms, routines, and conventions. Indeed, it is differences over the relative importance of these respective types of means by which individuals and structures influence one another – and therefore differences over interpretation of the concept of ‘social structure’ – that provide us with a basic distinction between the two broad heterodox approaches to understanding individual embeddedness. Section IV distinguishes two broadly different approaches to understanding the embeddedness of individuals in networks of social and economic relations in terms of two different types of means by which individuals and society influence one another: institutions and social values. The institutions approach works by examining how institutions structure interactions between individuals, and how individuals in turn influence the structure of society by creating and transforming institutions. The social values approach works by examining how social values structure interactions between individuals, and how individuals in turn influence social values. Though I dichotomize two approaches to understanding the agency–structure relationship, there are clearly important similarities between the two approaches. This reflects the fact that they both draw on a single understanding of individuals as socially and economically embedded rather than atomistic. Section V then briefly summarizes thinking about agency and structure in reference to six heterodox economic schools – institutionalist economics, social economics, radical/Marxist economics, Post Keynesian economics, feminist economics, and ecological economics – according to their respective reliance upon one or both of the two approaches set out here. The section concludes with summary remarks regarding what critical realism offers to the overall framework.

II Social psychology’s concept of the individual: the ‘self-concept’ The concept of the individual, or more specifically the ‘self-concept’ as it is generally termed in the social psychology literature (Rosenberg 1979; Gecas 1982; Demo 1992), dates from the late 1960s, and reflects the nature of social psychology as an interstitial social science shared by sociologists and psychologists. The concept has been said to be ‘rapidly becoming the

134

John B. Davis

dominant concern within social psychology … as part of the general intellectual shift from behavioral to cognitive and phenomenological orientations’ in both sociology and psychology (Gecas 1996). Those with a more sociological orientation within social psychology usually examine the social antecedents of individual behavior, whereas those with a more psychological orientation within social psychology rather tend to investigate the social consequences of individual motivation and behavior. Social psychological research on the concept of the self as a whole combines these two orientations by investigating how individual identity is reflexively constituted and reconstituted out of a socially influenced process of self-evaluation, self-imagining, and self-awareness engaged in by the individual. The self-concept is thus a concept of individual identity understood as ‘situationally variable,’ yet also relatively stable, as suggested by the idea of a ‘moving baseline’ subject to fluctuation and change (Demo 1985). The concept reflects social psychology’s comparatively more dialectical mode of investigation in comparison with the more uni-directionally oriented dynamics of sociology and psychology, which examine either the impact of social structure on the individual or the impact of the individual on social structure. Here I provide only a very brief review of a number of themes in the social psychology literature that bear on the individual self-concept. In my view, the social psychology literature on the self-concept is valuable for thinking about individuals as socially and economically embedded, because of its emphasis on the individual as a being continually constituted and reconstituted in a socially influenced process of selfevaluation. The self-concept itself in the most general sense, then, is that understanding which individuals acquire of themselves as a result of the influence of social structures upon their processes of self-appraisal (Rosenberg 1979). Within this general understanding, the self-concept is conceived both (i) in terms of self-appraisal processes in which the individual thinks of herself as a subject, an ‘I,’ and a knower, and (ii) in terms of self-appraisal processes in which the individual sees herself as an object for others, a ‘me,’ and a self known by others. Social psychologists with the former, more psychological orientation explain the self-concept in terms of individual desire to establish a sense of self, using the term ‘self-esteem’ to emphasize ‘the evaluative and emotional dimensions of the self-concept.’ Social psychologists with the latter, more sociological orientation use the concept of individual identity to ‘focus on the meanings comprising the self as an object’ within a social context (cf. Gecas 1982: 4). The former emphasis on self-esteem makes the individual activity or agency that produces a self-concept central, whereas the latter emphasis on identity gives greater weight to society’s influence on individual self-evaluation.

The agency–structure model

135

Both orientations, however, assume that how others judge an individual influences the formation of the individual’s self-concept (whether in the self-esteem sense or in the identity sense), and use this to argue that the self-concept involves a sort of ‘reflected’ self-appraisal. At the same time, it is also recognized that the self-concept as a kind of reflected self-appraisal still lacks good empirical foundations, since while it is plausible to say that people’s self-concepts reflect the way they believe others see them, there nonetheless seems to be little evidence supporting close correspondence between people’s self-concepts and the way people are actually seen by others. The problem is that it is not obvious just how people’s ‘self-evaluations are influenced by the feedback received by others’ (Shrauger and Schoeneman 1979). Possible reasons for this have been suggested for both the self-esteem and identity understandings of the concept. The way in which individuals establish self-esteem may be selective with respect to good and bad opinions of others as well as with respect to whose opinions are observed. Alternatively, the way in which individuals establish an identity concept may reflect norms of social interaction in a culture, whereby customary practices condition the form of appraisal from others, and create differences between how others perceive an individual and express their views of that individual (Goffman 1959). Thus in addition to the basic psychological and sociological types of approaches to the self-concept itself, there are also distinct psychology-based and sociology-based strategies for explaining the ways in which others’ opinions feed back upon and affect individuals’ self-appraisal.1 But for our purposes, the most important aspect of the social psychology literature on the self-concept is its shift toward greater emphasis on the dynamic effects of interaction between the individual and society. Thus while the earlier social psychological literature on the self-concept tended to emphasize constancy and stability in individuals’ self-concepts, in order to develop a coherent conception of a ‘maintained’ self, there now appears to be increasing agreement that emphasis should also be placed upon a ‘plasticity of characteristics previously assumed to be stable throughout the life course’, implying a more ‘“open” or “unfinished” character of the human organism in relation to its environment’ (Dannefer 1984: 107; cf. Demos 1992). One such view, widely influential in the social sciences and health fields, is of the selfconcept as an organized succession of different types of ages, for example developmental age as a set of meanings associated with personal passage from youth through old age, social age as the social meanings of age at various transitions and turning points in social life, and historical age as meanings associated with the experience of different cohorts and generations in history (Elder 1991). But there are many other ways in which the plasticity of the individual could be understood when taking the idea of the individual as a sort of ‘moving baseline.’ Thus in principle

136

John B. Davis

the social psychological self-concept should be amenable to employment in economics frameworks that emphasize dynamic, mutually constituting relationships between individual and society. What, then, does the social psychology thinking on the self-concept have to offer heterodox economics on the subject of the relationship between individual and society? In the first place, it offers new resources in the idea of the individual as specifically a reflexive being, or as a being that forms a self-concept in a social context. Heterodox economists generally stress the influence of society on the individual per se, but less often investigate how this influence produces changes in individuals’ selfconcepts. Indeed the issue of individual or agent identity itself infrequently arises in heterodox economics. Second, the social psychology of the individual is valuable in that it emphasizes the idea of the individual as an active being. Individuals are actively engaged in reconstituting their self-concepts in light of social influences upon them. Generally methodologically holist, heterodox economics emphasizes the active influence of society on individuals, but less often explains how individuals in turn actively influence social structures. Third, social psychology’s self-concept offers opportunities to heterodox economics in that its combination of the different causal strategies of sociology and psychology makes reciprocal cause and effect relationships central to explaining both individual and society as both cause and consequence of one another. Heterodox economics, by understanding individuals as socially and economically embedded, already employs these themes, but would be better able to develop them were it to incorporate a more developed concept of the agent as an active being.2 In this latter regard, the agency–structure model of individual and society, as developed in Lawson’s Economics and Reality, provides a way of incorporating the key ideas involved in social psychology’s self-concept in heterodox economics’ general understanding of individuals as socially and economically embedded, since Lawson’s analysis of the agency–structure model is both rooted in a heterodox economics vision, and employs a thinking about individuals and agency quite close to social psychology’s self-concept. I thus now turn to Lawson’s analysis to further develop a heterodox account of the relationship between individual and society that explains human agency as reflexive and active and the relationship between individual and society as one of reciprocal cause and effect.

III The agency–structure model of individual and society The inspiration behind the agency–structure model lies in the idea that the opposition between methodological individualism and methodological collectivism can be overcome by regarding agency and structure as dynamically interactive. An early formulation is Giddens’

The agency–structure model

137

influential ‘structuration theory,’ which treats individuals and social structures as interdependent, or as a duality, such that each may be said to help constitute the other, especially through recurrent social practices (1976, 1984). Bhaskar (1979 [1989]) and Archer (1995) revised and extended Giddens’ thinking, principally by seeing reality as stratified and multi-layered with emergent properties differentiating one layer or level from another (cf. Hodgson 2000: 5–13; also Collier 1994). Lawson developed this latter, critical realist conception of the agency–structure relationship specifically for economics in Part III of his Economics and Reality (1997), and used it primarily to critique the methodological posture of mainstream economics. I use it instead to understand the embedded individual in heterodox economics, and carry out an ontological analysis of agents in economic life. Central to this latter project is how the model accounts for the reciprocal effects of human agency and social structure upon one another, and how the model treats human agency as intentional activity. I argue that this framework is in need of a fuller understanding of the individual, and attempt to provide one along the lines developed in social psychology. In my view, then, an agency–structure model expanded to include an account of individuals can account for processes of social reproduction and transformation that both bear upon and help us understand the identity of individual economic agents. This fuller account with its alternative conception of the individual economic agent as socially and economically embedded is what I believe is most needed for heterodox economics to distinguish itself from orthodox economics. The basic idea of the agency–structure model, then, is that social structure depends on human agency, and human agency depends on social structure. First, consider how social structure depends upon human agency. The agency–structure model of individual and society is specifically a critical realist analysis in that, as a realist model for a social science, the model makes ‘social structure dependent upon human agency … open to transformation through changing human practices which in turn can be affected by criticising the conceptions and understandings on which people act’ (Lawson 1997: 158, original emphasis). That is, social structure changes, because human practices change, because people critically evaluate the ideas which those practices reflect. This does not imply, of course, that human agency is an unmediated force acting on social structure free of all social influence. The process of replacing one set of ‘conceptions and understandings’ with another clearly involves the idea that individuals operate upon an inherited set of materials. So indirectly social structure still plays a role in criticism. But this raises a long-debated question. Might this not imply that human agency is ultimately only a product of social structure, and that it is a mistake to say that social structure depends on agency?

138

John B. Davis

One way to resist this conclusion is to say that when we take individuals as agents of change, their self-conceptions and selfunderstandings, which individuals continually organize and reorganize to create their self-concepts, play a central role in organizing the great mass of ‘conceptions and understandings’ individuals regularly have about the world. The process of forming a self-concept, involving the replacement and abandonment of one set of self-conceptions with another, is understood in social psychology as a reflexive and active process in which individuals absorb a range of conceptual materials from society (say, a grasp of customs, understandings of markets, knowledge of legal principles, etc.), and then actively integrate these materials in terms of their self-conceptions (say, that I ought to behave in some way or another, that I can participate in such-and-such a market in some fashion, that the law applies to me in a particular manner). That they may be said to do this implies that they are able to distinguish themselves as selves (as I’s or me’s) about which they organize these materials. As a property of individuals as agents, that is, reflexivity implies that individuals possess a measure of detachment or relative autonomy from the social influences operating upon them though they draw upon them. The critical realist understanding of the agency–structure relationship implicitly invests individuals with this active capacity when it describes them as regularly engaged in ‘criticising the conceptions and understanding on which people act.’ What social psychology adds to this is the idea that this criticism also extends to self-conceptions and self-understandings, and the capacity to engage in this further criticism enables us to see individuals as relatively autonomous agents. From this perspective, it is fair to say that in the agency–structure model, social structure influences individuals, yet is also itself dependent upon human agency. Now consider the other side of the agency–structure model, or the reciprocal dependence of human agency on social structure. Given what has been said above, how exactly are we to understand this dependence and the way that society influences individual agency? Lawson makes two points in this regard. First, in rejecting the reductionist project of methodological individualism, he argues that ‘social structures (rules, relations, positions)’ are real in the sense of being irreducible to the actions of individuals (1997: 161). Second, social structure is said to be a precondition for individual action in that, as said above, individuals draw upon it when acting. More specifically, in rejecting the orthodox understanding of rationality, Lawson substitutes a theory of situated rationality whereby: ‘At any given point in time any individual is situated in a range of positions’ associated with which are ‘rules to draw upon, obligations to fulfill, structures of power to utilise and be influenced by’ (1997: 187). These positions, and the rules, obligations, and structures of power upon which they depend, constitute the social preconditions for

The agency–structure model

139

individual action which help explain how human agency depends upon social structure. What we may add to this in drawing on social psychology’s selfconcept is further understanding of just how these social preconditions impose their influence on individuals. When we add a self-concept to our understanding of individuals, we say that society’s rules, obligations, and structures of power help determine the form and shape of individuals’ self-conceptions. For example, a particular social obligation, such as that an individual return a favor, involves a set of expectations of that individual which create a set of self-understandings for that individual. Society’s rules, obligations, and structures of power, that is, really only become preconditions for individual action when they are translated into conceptions that apply specifically to particular individuals in terms of the self-conceptions they generate. The theory of embedded rationality broadly emphasizes the relationship between social structure and individual positions within a social structure. The social psychology selfconcept specifically links social structure and positions through the particular individual’s apprehension of this relationship, as formulated in terms of that individual’s self-conceptions and ultimately self-concept. Social structure, then, is a precondition for human agency in the sense of situating rationality in social positions, but, because individuals engaged in forming self-concepts actively grasp their social positions in applying rules, obligations, and expectations to themselves, human agency is dependent upon but not determined by social structure. Thus human agency depends upon social structure, just as social structure depends upon agency. The agency–structure model of individual and society consequently aims to explain the reproduction and transformation of both human agency and social structures. A consequence of this is that it may be said to avoid the (twin but opposite) errors of voluntarism and reification/determinism, where voluntarism explains social structures as essentially the creation of individuals (as in some methodological individualist accounts), and reification/ determinism sees individuals as essentially the product of social structures (as in some methodological collectivist accounts). The way in which the model achieves this is by emphasizing that human agency is intentional activity. From this two things follow. First, ‘most structural reproduction and/or transformation arises as the unintended product [of human intentional activity], whether or not desired or even recognised’ (Lawson 1997: 169). Though human agency creates social structures, intentional activity cannot fully explain the evolution of social structures (the error of voluntarism), because its results are in significant degree unintended. We might add to this that if individuals operate in some fundamental sense in terms of their self-concepts, then the scope of their activities is necessarily restricted, and the consequences of their actions are likely often unintended. Second, ‘If the reproduction/transformation

140

John B. Davis

of social structure is rarely an intended project, it is equally the case that the individual agents are not always aware, certainly not discursively or self-consciously so, of the structures (such as language rules) upon which they are drawing’ (1997: 169). We may add here that if individuals’ selfconcepts are by nature always open and unfinished, they cannot be expected to smoothly accommodate all social influences and feedback from others, so that the individual’s awareness of the import of social structures is always incomplete. This reinforces the idea that though social structures influence human agency, they do not determine intentional human activity (the error of reification/determinism). Yet if this account depends upon treating human agency as intentional activity, what are we to say about the pervasiveness of highly routinized forms of activity in human society? Lawson allows that much of human behavior does not take on the aspect of being intentional. If we distinguish between discursive and tacit consciousness, where the former involves conscious reflection on a particular subject, and the latter involves individuals monitoring their conduct at a subconscious level, then it seems obvious that individuals often rely upon a body of tacit knowledge and skills in order to do such things as follow rules, observe norms, and act in accordance with conventions without reflecting directly upon doing so. But such activity is still purposeful, and thus also fairly treated as a form of intentional activity. Were the individual interrupted in the performance of an episode of rule-following, typically a reason for following the rule could be given. This verdict is important not just because routine is such an important part of human behavior, but because understanding routinized behavior as a form of intentional activity is important for developing a more complete understanding of the individual and human subjectivity in the agency–structure model. Generally, Lawson’s discussion of the processes of reproduction and transformation in the agency–structure model tends to associate these processes with the evolution of social structures. But to be complete, the processes of reproduction and transformation of agency and social structure should also be associated with the evolution of human subjectivity, the agency side of the model. Here his discussion of routines as a form of intentional activity opens the door to one way to go about this. Lawson asks why so much human activity is routinized. Part of the answer lies in the simple fact that we seem unable to apply discursive decision-making to all occasions in life where it might be possible to do so. But if this is necessary to explaining routines, it is not sufficient, since other responses to this inability, such as inactivity, are also available. A sufficient explanation that Lawson contemplates is that this human inability causes individuals anxiety and stress, and that individuals accordingly rely on routinzed behavior to reduce this anxiety and stress. Then the fuller account of the agency–structure model that includes an explanation of the evolution of human subjectivity would involve

The agency–structure model

141

describing how routines evolve together with individuals’ evolving strategies for combating anxiety and stress. Yet it is not far from saying this to the social psychology idea of the individual self-concept. Indeed Lawson adds that individuals may be thought to have a need for ‘inner security’ which in turn may reflect a need for ‘a significant degree of continuity, stability and sameness in daily affairs’ (1997: 180). A way of understanding continuity and stability in life is in terms of how they promote continuity and stability in an individual’s self-concept. Recalling the ‘moving baseline’ expression, individuals might be said to rely on routinized activity, because doing so promotes continuity and stability in their self-concepts. Lawson in fact suggests just such an argument when (following R.D. Laing and Giddens) he treats a need for inner security as a need for an ‘ontological security,’ where the ‘psychological origins of ontological security are to be found in basic anxiety-controlling mechanisms … hierarchically ordered as components of personality’ (Giddens 1984: 50; quoted in Lawson 1997: 183). Ontological security, that is, applies to a subject, or the individual. Thus, the processes of reproduction and transformation that are explicitly associated with the evolution of social structures in Lawson’s agency–structure model seem equally applicable to the evolution of human subjectivity. Here I don’t recommend this or any other particular account of how to go about understanding the evolution of human subjectivity. It is enough for the purposes of this chapter to emphasize that a general agency–structure model is a model of interactive processes of reproduction and transformation affecting both social structures and human subjectivity. There is one minor point about Lawson’s discussion to be made before passing on to the two main approaches or ways of understanding individual embeddedness in heterodox economics. The agency–structure model, by using the term ‘structure’, seems to favor one of the two general approaches identified at the beginning of this chapter regarding the means by which individuals and society influence one another, namely, that approach which emphasizes the role of socio-economic institutions rather than that which emphasizes social values. However, it is clearly not Lawson’s intent to favor one of the two general approaches, since his discussions of ‘structure’ include a variety of social apparatuses which are better represented in terms of social values, or perhaps value structures, than in terms of institutionalized social arrangements. More careful discussion of this distinction must be postponed to the next section. But it seems fair to say at this point that Lawson’s discussions of rules and conventions allow a role for social values, and need not only be understood in terms of more institutionalized social arrangements. ‘Structure’ of course also refers to institutionalized social arrangements. The agency–structure model, then, is general in that it allows a role for

142

John B. Davis

both types of mechanisms of influence upon individuals, while ‘structure’ is to be understood both in terms of institutions and in terms of values.

IV Heterodox economics’ two approaches to the agency–structure relation In this section I contrast two broadly different types of ways of understanding the reproduction and transformation of both social structures and human subjectivity according to two different kinds of means by which individuals and society each influence one another. At the beginning of this chapter, I characterized the difference between these two approaches as a difference between how more highly institutionalized social arrangements and social positions influence and are influenced by individuals, and how less formal social norms, routines, and conventions influence and are influenced by individuals. I thus begin here by explaining (i) how individuals influence the transformation and evolution of institutions and social values (Lawson’s topic of how social structure depends upon agency). Then I turn to the reciprocal matter of (ii) how social-economic institutions and social values differ in the ways in which they operate upon individuals (Lawson’s topic of how agency depends upon social structure). Finally, I outline (iii) the overall agency–structure cause-and-effect system, in order to provide an account of its dynamics in terms of tensions between both kinds of factors operating on both individuals and social-economic structure in the form of institutions and social values. (i) Individuals’ different kinds of influence on institutions and social values respectively (social structure’s dependence on agency) Individuals may be said to influence institutions when they engage in interaction which departs from what institutions permit or require, and which subsequently brings about changes in existing institutions or leads to new ones. It seems fair to say that much individual action and interaction between individuals is not structured by institutions. Take, for example, antitrust laws as social-economic institutions determining the scope of individual action for individual business firms. In the past such laws have generally been national in scope, leaving the efforts on the part of firms to restrict competition among themselves across national boundaries largely unregulated. In response, in many countries existing antitrust laws have been reinterpreted to accommodate extra-territorial economic behavior, while new supranational antitrust agreements between countries have also begun to emerge. Thus institutions have been influenced by interaction between individuals (or firms taken as individuals) occurring outside of the framework of existing institutions.

The agency–structure model

143

Understanding these firms in self-concept terms, we might say that this overall development begins when firms determine how exceptions in existing laws apply to themselves. That is, they entertain new selfconceptions by, in Lawson’s language, ‘criticizing’ the practices and institutions under which they operate. This then creates the opportunity for their interaction with one another in restraint of competition, which in turn may lead to change in institutions. Thus individual action and interaction outside of institutional frameworks tend to lead to change in those frameworks. Alternatively, individuals may be said to influence social values when they share personal values different from widely held social values, and their doing so subsequently brings about a change in social values. Personal values may both coincide with and also differ from widely held social values. When they differ, the ‘personal’ character of an individual’s values is reinforced by their apparent lack of widespread acceptance. But appearances may be deceiving, and it is possible that so-called ‘personal’ values are shared by many individuals, unbeknownst to them. Should individuals, then, discover that they in fact share what they believed was largely personal, their personal values may begin to acquire a social character as well. For example, consider an electoral process. Prospective voters may share views about candidates and outcomes before public discussion and debate define electoral issues. As the latter proceed, however, individuals may discover they share others’ opinions and values. In self-concept terms, individuals find that their ‘personal’ values are shared when they formulate new self-conceptions regarding how their ‘personal’ values compare with those of others. Social values may thus emerge from this recognition of shared personal values, and consequently social values are influenced by personal values. Note that both of these two means by which individual action and decision influence social structure – interactions between individuals affecting institutions and sharing personal values affecting social values – are supra-individual in character in that they explain action and decision in terms of relationships between individuals. This implies that in a complete agency–structure model we ought to be able to explain the human subjectivity side of the model not just in terms of the behavior of collections of relatively independent individuals but in a stronger sense of individuals somehow acting together. That is, we need some coherent means of discussing how individuals use the language of ‘we’ as well as the language of ‘I’ and ‘me’ if we are to fully explain human subjectivity and individuality. This discussion goes beyond the immediate framework of this chapter, but I do treat it in an investigation of the concept of collective intentionality in connection to the theory of the embedded individual (Davis 2003).

144

John B. Davis

(ii) Influences of institutions and social values on individuals (agency’s dependence upon social structure) First, I suggest that when social-economic institutions influence individual action and decision-making, they do so chiefly by altering the scope of action and individual decision-making. An institution, say a national labor law or an environmental regulation, limits the space of activity in which individuals can act, and thus leads them to focus their decisionmaking on an allowable space of activity. Individuals’ subjectivity and self-concepts are consequently developed and transformed primarily in terms of their extent of application, rather than in terms of changes in their meaning or content. Some might wish to argue that institutions also influence the content of action and decision-making. Rather than argue this point, here I attempt to follow others more knowledgeable than I about the nature and history of thinking about institutions, who emphasize that institutions are best defined as networks of social relations and positions generally accepted by those to whom they apply (cf. Rutherford 1994: 182n). Thus institutions appear to have their primary impact on individuals by determining the space of allowable activity. This is also consistent with their characterization as more highly settled, more formal social arrangements and associated sets of social positions. Second, then, social values may be said to influence decision-making and individual subjectivity by altering their content and basis. That is, should individuals rely on some given set of reasons or rationales when acting in certain circumstances, being influenced by social values normally means that they adopt new reasons or rationales for how they will act in those same or similar circumstances in the future. Moreover, since these new reasons or rationales constitute a new basis for individual action, they should also be thought to involve changes in individuals’ selfconceptions and self-concepts. For example, adopting new religious views typically changes an individual’s actions and interactions with others. At the same time, a change in religious views generally means a change in the individual’s self-concept as it applies to the individual’s perceived relation to the supernatural. At issue here is less the scope of existing views than the character of those views itself. Thus it seems fair to say that the impact that social values have on individuals is primarily in terms of the meaning and content of their actions. To be sure, because institutions and social values are similar in important respects, it is somewhat artificial to parcel out their effects as I have done here. Thus, as will emerge in what follows, a good part of the justification for the distinction above lies in what it permits us to say about complex processes of reproduction and transformation in an expanded agency–structure model. I now turn to an account of the model that combines the discussion of the last two sections.

The agency–structure model

145

(iii) The overall agency–structure model Institutions and social values influence the scope and content of individuals’ action and decision-making respectively, while individuals’ interactive activity and shared values influence institutions and social values respectively. These processes clearly occur simultaneously. Though institutions create a framework within which individuals act, as for example exists with any set of laws, individuals’ interactions with one another may test that framework, perhaps causing laws to be rewritten. And though social values determine a basis for individual action, as when a conception of human rights guides individual decision-making, individuals may find that they share values regarding those rights at odds with that conception, perhaps leading to revision in a community’s conception of human rights. At the same time, it seems reasonable to say that these two pairs of reciprocally occurring processes (institutions/ individual interaction and social values/shared personal values) also somehow influence one another, and that we ought in principle to be able to describe how this occurs. It pays, however, to be cautious about attempting to over-describe the agency–structure model as if the social world were to be captured as a complete system of interconnections, since the social world is arguably stratified, such that some levels of activity are emergent upon others, and thus strictly speaking not interconnected.3 Thus to avoid portraying the world as a system of interconnections, and yet still provide an account of the overall agency–structure model, in what follows I conceptualize the relationship between these two pairs of reciprocal processes in terms of possible tensions between them in an overall process that reproduces and transforms both agency and structure. Further, since change originates on the agency side of the agency–structure model, I explain these tensions as arising from the actions of individuals, first in terms of how they generate a change in social values, which are then in tension with society’s institutions, and second in terms of how they generate change in society’s institutions, which are then in tension with its social values. Consider first how change in a society’s institutions may be brought about by change in its social values, for example in terms of changes in the relationship between a society’s legal framework and its conception of human rights. In the late eighteenth and early nineteenth century there was little tension between the basic legal framework of the United States, the Constitution, and reigning social values that guaranteed human rights to the European-derived population while denying them to the Africanderived population. Yet by the middle of the nineteenth century social values regarding human rights were in a process of change (especially in the North), and were increasingly perceived to be at odds with existing legal institutions. More specifically, the social values/shared personal values reciprocal process underwent change as the shared personal values of

146

John B. Davis

religiously inspired slavery Abolitionists gained weight in the determination of Northern social values. This change in social values helped generate change in US legal institutions by the middle of the nineteenth century, including amendments to the Constitution itself. It is true that this institutional change was also brought about by interaction between individuals outside of the framework of existing legal institutions, the war between the North and the South being an obvious example. But most historians agree that the new values in the North regarding the unacceptability of slavery played a key role in bringing about the war and subsequent institutional change, and thus a change in social values may be seen as bringing about the change in institutions.4 Both agency and social structure were reproduced and transformed in this process, but it is worth attending a bit more closely to change and transformation on the agency side of the model in light of the emphasis placed on understanding the embedded individual and subjectivity in this chapter. Thus, a change in shared personal values comes about when individuals inherit conceptual materials (the influence of structure on agency) and then, in ‘criticizing’ these materials and the practices they entail, come to entertain new self-conceptions. Northern Abolitionists, that is, grasped the moralities of a society accepting of human rights for some but not for others, found the self-conceptions these moralities implied unacceptable, and formed new self-conceptions regarding their personal stances toward slavery. As these new self-conceptions and the personal values they involved began to be shared by many individuals, social values regarding human rights and slavery began to change. Here, the question naturally arises as to what brought about the abandonment of old self-conceptions and the adoption of new ones in this process? But in an important sense this question is inappropriate. The agency– structure model presupposes that a genuine dynamic exists with influence on change arising in both agency and structure. The expanded agency–structure model with a more developed account of individuals and subjectivity explains the dependence of structure on agency by the treatment of individuals in social psychological terms as reflexive and active. In short, individuals simply have a genuine capacity to criticize their conceptions and self-conceptions. Consider now how change in a society’s social values can be brought about by change in its institutions. The feudal system in medieval Europe involved a set of economic institutions, such as manorial production and inherited class relationships, and a set of associated social values, such as that individuals have a customary, inherited place and vocation in life and that ruling classes had paternalistic responsibility for the well-being of the laboring classes. Market behavior, as was to emerge with capitalism, involves an alternative set of economic institutions, such as contracts and the wage system, and its own set of social values, such as the rights of private property and the idea of the individual as a locus of freedoms.

The agency–structure model

147

When in the seventeenth and eighteenth centuries individuals began to place greater reliance on market relationships with one another, they also began to put in place new economic institutions not consistent with the social values of feudalism. In this case, changes in individual interaction brought about change in economic institutions, which in turn generated tensions between institutions and social values. Of course it might alternatively be argued that change in shared personal values was also a source of change in the social values of the time. I don’t argue that this was not the case. But it seems many historians would agree that a change in economic practices played a particularly important role in Europe’s abandonment of feudalist social values, and thus a change in institutions may be seen as bringing about a change in social values. Again, in light of the emphasis placed in this chapter on the theory of the embedded individual and the agency side of the agency–structure model, it is worth attending a bit more closely to how changes in individual interaction might reflect individuals’ changing selfconceptions. Thus, when individuals restricted themselves to customary, feudal economic relationships, their self-conceptions were traditional, inherited ones. But when they began to engage in market transactions, they had to see themselves in new, unfamiliar capacities. Taking individual interaction outside of established institutions to be primarily a matter of changes in the scope of activity, this change in individual selfconceptions and self-concepts was a matter of not just a change in the scope of individual activity, but also a change in scope of individuals’ selfconceptions, which were enlarged to include the capacity to produce for exchange. The emergence of market institutions solidified this change. Again, I do not attempt to explain why individuals were able to seek out new forms of economic interaction after centuries of traditional economic relationships. The agency–structure model makes agency dependent upon structure, but also structure dependent upon agency, and the appearance of new forms of individual interactions in markets counts as evidence of the latter. The emphasis here lies rather on developing a clearer conception of changes in agency, or individuals’ subjectivity. Thus whereas changes in shared personal values alter the basis on which individuals form self-conceptions, changes in individual interaction enlarge the scope of those self-conceptions. Of course, my two examples – the nineteenth-century rise in opposition to African-American slavery in the United States and the emergence of market behavior in medieval Europe – are only highly schematic characterizations of complex historical processes that have been treated at much greater length by historians. My purpose in setting them forth is to illustrate two things: first, that the agency–structure model can advantageously be broken down into two types of reciprocal processes (institutions/individual interaction and social values/shared personal values), and, second, that enlarging the model to incorporate a

148

John B. Davis

more developed understanding of embedded individuals and subjectivity adds to the explanatory power of the model. Since this chapter seeks to use the agency–structure model to help develop a conception of the embedded individual for heterodox economics, I do not attempt here to further elaborate or defend this interpretation of the model, but rather now turn to a classification of heterodox economics schools both according to their apparent reliance or emphasis on institutions and/or social values as means by which individuals and society influence one another, and according to their associated conceptions of the embedded individual.

V Six heterodox schools: concluding remarks The six schools of heterodox economics I address, roughly according to their order of emergence in contesting post-war orthodox economics, are: institutionalist economics, social economics, radical/Marxist economics, Post Keynesian economics, feminist economics, and ecological economics. None are explicitly understood by their proponents in terms of the agency–structure model, but all arguably rely on the principal tenet of the model, namely, that agency and structure, however identified, are dependent upon one another. Thus all generally avoid the (twin but opposite) errors that face social science and social theory, voluntarism and reification/determinism, and accordingly all are neither methodological individualist nor methodological collectivist in orientation. First, then, how may these schools be classified according to their emphasis on institutions and/or social values as the means by which agency and structure influence one another? Institutionalist economics and social economics are relatively easy to characterize in that traditionally the former has emphasized institutions and the latter social values as the chief means by which agency and structure influence one another. Institutionalism does not ignore social values, but largely treats them as reflecting institutional relationships. Social economics does not ignore society’s institutional structures, but generally explains them as reflective of social value attachments. In contrast, radical/Marxist and Post Keynesian economics seem to involve conflicting views regarding the roles of institutions and social values. Radical/Marxist economics emphasizes the institutionalized nature of class relationships underlying labor exploitation and capital accumulation, but also emphasizes the role of class consciousness in revolutionary and liberation struggles, which involves social values. Post Keynesian economics emphasizes the institutionalized nature of liquidity and uncertainty in a capitalist market economy, but on the social value side also emphasizes investor animal spirits and the question of public commitment to demand management. Finally, feminist and ecological economics seem to have evolved in the weight they assign to institutions

The agency–structure model

149

and social values, though in different directions, with feminist economics now placing greater weight on social values and ecological economics now placing greater weight on institutions. Feminist economics originally gave greater weight to laws and institutionalized systems of discrimination in the economy in explaining gender inequality, but has increasingly emphasized gendered social attitudes, as are for example manifested in household relationships and child-rearing, as important. Ecological economics originally emphasized public consciousness regarding the environment, but has given increasing weight to changing legal frameworks limiting corporate activities. What, then, may we conclude from this regarding different conceptions of the embedded individual in these heterodox schools? In institutionalism, individual embeddedness is understood chiefly in terms of the institutions/individual interaction reciprocal processes relating agency and social structure. That is, individual activity is influenced by institutions (agency depends on structure), and individual activity changes institutions (structure depends on agency). In social economics, individual embeddedness is understood chiefly in terms of the social values/shared personal values reciprocal processes relating agency and social structure. That is, social values and shared person values influence and depend upon one another. In radical/Marxist and Post Keynesian economics, individual embeddedness can be explained on both bases. On the one hand, this produces richness in explanation across these two schools when each is taken as a whole. On the other hand, since the two schools tend to be divided between those giving greater emphasis to either the institutions/individual interaction or the social values/shared personal values processes, general conclusions within the two schools about individual embeddedness are difficult to achieve. In feminist economics, individual embeddedness is increasingly explained in terms of the social values/shared personal values processes, though with the institutions/individual interaction processes as a framework within which this occurs. Similarly, in ecological economics, the institutions/individual interaction processes are explained within the social values/shared personal values processes. These two latter schools perhaps offer the most explanatorily powerful accounts of individual embeddedness in their implicit integration of both sets of processes in overall agency–structure analyses. What these differing emphases suggest is that a general account of the individual as socially and economically embedded for heterodox economics could be based on a critical realist agency–structure model enlarged with social psychology thinking about the self-concept. Different heterodox schools give different kinds of emphasis to institutions and social values, but the agency–structure model can make use of both types of means in terms of two reciprocally occurring processes, institutions/individual interaction and social values/shared personal

150

John B. Davis

values, which each have effects upon one another. This overall framework, I suggest, is reflective of heterodox economics as a research programme alternative to mainstream economics. From this perspective, the different schools within heterodox economics might be classified according to their implicit implementation of this framework. At the same time, in so doing a general account of the individual as socially embedded begins to emerge for heterodox economics. The explicit reflection about the agency–structure relationship prompted by an encounter with critical realism might also help heterodox economists to avoid a tendency which some of them display towards the provision of one-sided accounts of the means by which agency and structure influence one another. The risk that heterodox schools such as institutionalism and social economics run is that by concentrating so heavily on institutions and values respectively, they pre-judge the importance of those factors and thereby run the risk of offering a distorted analysis of the processes through which agency and structure interact with one another. Critical realism provides a timely reminder that the question of whether the interplay between structure and agency centers primarily on institutions or on social values cannot be answered a priori but only ex posteriori through concrete research. Critical realism, then, leaves conceptual room for a variety of processes through which agency and structure interact and as a result is more able to do justice to the complexity of the structure–agency relationship than schools which prejudge its nature. And by alerting heterodox economists of these schools to potentially explanatorily significant avenues of influence that might otherwise be ignored, critical realism helps to further the development of explanatorily powerful economic analysis.

Notes 1

2 3

4

These different strategies also arise in connection with other standard (and related) topics of investigation in the social psychological literature such as ‘role playing’ and ‘social comparisons,’ which similarly concern how others’ opinions influence individuals’ self-appraisal. See Lewis (2000) for a discussion of causal strategies in connection with the debate between Bhaskar and Harré. Archer (1995) advanced this criticism of Giddens’ structuration theory. Emergence may be defined as ‘a relationship between two features or aspects such that one arises out of the other and yet, while perhaps being capable of reacting back on it, remains causally and taxonomically irreducible to it’ (Lawson 1997: 63). Another example along these lines is the struggle over the political rights of women. A social value transformation regarding whether women were entitled to basic political rights was caused by the increasing influence of shared personal values of those in the Suffragette movement in the early twentieth century. This led to institutional legal change beginning with voting rights to women, and continuing with more recent legislation such as the 1962 Equal Pay Act.

The agency–structure model

151

References Archer, M. (1995) Realist Social Theory: The Morphogenetic Approach, Cambridge: Cambridge University Press. Bhaskar, R. (1979 [1989]) The Possibility of Naturalism: A Philosophic Critique of the Contemporary Human Sciences, Second edition, Brighton: Harvester. Collier, A. (1994) Critical Realism: An Introduction to Roy Bhaskar’s Philosophy, London: Verso. Dannefer, D. (1984) ‘Adult Development and Social Theory: A Paradigmatic Reappraisal’, American Sociological Review, 49: 100–16. Davis, J. (2003) The Theory of the Individual in Economics, London and New York: Routledge. Demos, D. (1985) ‘The Measurement of Self-Esteem: Refining our Methods’, Journal of Personal Psychology, 48: 1490–1502. —— (1992) ‘The Self-Concept over Time: Research Issues and Directions’, American Review of Sociology, 18: 303–26. Elder, G. (1991) ‘The Life Course’, in E. Borgotta and M. Borgotta (eds) The Encyclopedia of Sociology, New York: Macmillan. Gecas, V. (1982) ‘The Self-Concept’, American Review of Sociology, 8: 1–33. —— (1996) ‘Self-Concept’, in A. Kuper and J. Kuper (eds) The Social Science Encyclopedia, Second edition, London: Routledge. Giddens, A. (1976) Central Problems in Social Theory, Berkeley: University of California Press. –—— (1984) The Constitution of Society, Cambridge: Polity Press. Goffman, I. (1959) The Presentation of Self in Everyday Life, New York: Doubleday. –—— (1967) Interaction Ritual, New York: Doubleday. Hodgson, G. (2000) ‘Structures and Institutions: Reflections on Institutionalism, Structuration Theory and Critical Realism’, unpublished manuscript. Lawson, T. (1997) Economics and Reality, London: Routledge. Lewis, P. (2000) ‘Realism, Causality and the Problem of Social Structure’, Journal for the Theory of Social Behaviour, 30: 249–68. Mead, G. (1934) Mind, Self, and Society, Chicago: University of Chicago Press. Rosenberg, M. (1979) Conceiving the Self, New York: Basic. Rutherford, M. (1994) Institutions in Economics: The Old and New Institutionalism, Cambridge: Cambridge University Press. Shrauger, J. and T. Schoeneman (1979) ‘Symbolic Interactionist View of the SelfConcept: Through the Looking Glass Darkly’, Psychological Bulletin, 86: 549–73. Stryker, S. (1968) ‘Identity Salience and Role Performance’, Journal of Marriage and Family, 30: 558–64.

8

Critical realism and the heterodox tradition in economics Shaun Hargreaves Heap

I Introduction This chapter offers a perspective on the contribution that critical realism has made to heterodox economics. The value of such an exercise depends in part on correctly identifying aspects of the critical realist project; so I begin with the usual disclaimers in this regard. The same applies, of course, for the heterodox tradition. I will begin, however, by characterising what seem to me to be two important aspects of that heterodox tradition which distinguish it from the orthodoxy, so that I can locate the contribution of critical realism and avoid any early misunderstandings. First, heterodox economists typically take philosophical argument more seriously than their orthodox counterparts. That is, the heterodox tradition believes that sorting out the philosophical foundations of the discipline is important, whereas the orthodoxy is largely insouciant to such concerns. This has led heterodox economists both to question the philosophical foundations of the orthodoxy and to engage with the recent postmodern turn in philosophy (unlike orthodox economics which has conspicuously ignored postmodern currents). In this context, critical realists have, to my mind, supplied useful buttresses against both the positivism that underpins some orthodox economics and also some contemporary relativist tendencies in economics (and social science more generally). Second, heterodox economists typically work with a richer, if vaguer, model of human agency than that found in neoclassical economics. There is no single alternative model but heterodox economists tend to emphasise the way that people are socially and historically located. So people are less individualistic, rely quite explicitly on norms and are less able to calculate what is to be done for the best than their orthodox counterparts. Again, I think critical realists have made significant contributions to this aspect of the heterodox tradition. I consider these two contributions in more detail in the next two sections. Section IV focuses on another contribution: the particular way

The heterodox tradition in economics 153 that critical realism characterises the dispute with orthodox economics. I am less inclined to find critical realism helpful to heterodox economics in this regard than the others, but this may be more a matter of appearance than substance. I hope it will be clear from the summary that I have given above that I find much to applaud in the critical realist relation with heterodox economics. However, I am not a ‘card carrying’ critical realist and I also try to explain why this is the case in these sections. This involves drawing out differences and since differences tend to attract the eye more than agreements, it is worth keeping in mind the general remarks that I have made here about the contribution of critical realism. In the final section, I draw some conclusions.

II Contributions to the philosophy of science: between positivism and idealism1 An important debate in the philosophy of science is between those who think that reality is discovered in science through some version of empirical investigation and those who believe in the social construction of ‘reality’. This is a version of the ‘objectivity versus relativism’ debate that has engulfed most social sciences at one time or another and which has been rekindled recently by what is known as the postmodern turn. Like critical realists, I believe that one needs to avoid both extremes. As a result, I think that the heterodox tradition is well served by arguments, like those of critical realism, which position it in this way. I shall examine the critical realists’ way of doing this in this section, beginning with their attack on ‘positivism’ and following through to their resistance to relativism. They have focused much of their attention on the ‘positivist’ version of the objectivity side of this debate. This is for the very sensible reason that ‘positivism’ supplies one of the philosophical defences of the method followed by many mainstream economists. In particular they have taken issue with two related aspects of positivism. The first concerns the depiction of causation, which goes back to Hume, as the constant conjunction of events. The second is the behaviourism which often follows from this: that is, a disregard for any deep kind of ontological investigation of the world. The two are connected because once science deals with event regularities, there is no need to enquire any deeper into what has produced these events. So, for instance, the orthodox axiomatic approach to choice illustrates both aspects. It focuses on the axioms of choice rather than the psychology of choosing and investigates choice empirically at the level of actual choices. The point of attack on positivism is set out by Lawson as follows:

154

Shaun Hargreaves Heap outside of astronomy most of the constant event conjunctions occur only under strict conditions of experimental control. In other words they are not spontaneous in nature but a product of human intervention … results or insights obtained through controlled experimental activity … are nevertheless successfully applied outside of the experimental situation. Put differently, constant conjunctions of events are in fact extremely rare, spatio-temporally restricted and usually artificially produced, while law-like knowledge appears to be generally available and useful, and some (but only some) of it experimentally corroborated … It is easy enough to see that these observations raise severe problems for any conception which ties laws … to constant conjunctions of events … For, if such regularities occur only in such restricted experimental set-ups, this bears the rather constricting implication that science and its results, far from being universal, are effectively fenced-off from most of the goings-on in the world. (Lawson 1997: 27)

The identification of constant conjunctions of events is even more restricted in a social science like economics because it is an open system. The source of this openness is the fact of human choice. ‘Now if choice is real any agent could have done otherwise; each agent could always have acted differently than he or she in fact did … A necessary condition for this [choice] is that the world, social as well as natural, is open in the sense that events could have been different’ (Lawson 1997: 30). The alternative to the positivist concern with event regularities that is proposed by critical realism flows directly from the view that the world has a structured reality, that is the view that ‘the world consists of more than the actual course of events and our experiences. Rather three domains of reality are distinguished, namely the empirical (experience and impression), the actual (actual events and states of affairs) and the real (structures, powers, mechanisms and tendencies)’ (Lawson 1997: 21). Science then is no longer confined to ‘the seeking out of constant event conjunctions, but aims at identifying and illuminating the structures and mechanisms and powers and tendencies that govern or facilitate the course of events’ (Lawson 1997: 23). The knowledge that we have of these ‘structures, mechanisms and powers’ is, however, always locally and historically mediated. This is inescapable: ‘knowledge … exists in a historically specific, symbolically mediated and expressed, practice dependent form’ (Lawson 1997: 59). Nevertheless, ‘the recognition of a social and pragmatic dimension to all social being, including scientific practice, does not necessitate an “anything goes” position … Rather, the attachment of epistemological relativism to ontological realism facilitates a judgemental rationality’ (Lawson 1997: 59).

The heterodox tradition in economics 155 Actually, what allows the fact of a reality to foreclose on an ‘anything goes’ approach is the ability to infer something about that reality. If choice is real then human actions must be intentional under some description. And intentionality is bound up with knowledgeability. For human beings must have some knowledge at least of the conditions that render their actions feasible. In turn again, knowledge presupposes a degree of endurability in the objects of knowledge sufficient to facilitate their coming to be known. Now if, as widely reported, scientifically significant event regularities do not often occur in the social realm, the enduring objects of knowledge must lie at a different level, at that of the structures which govern but are irreducible to events, including human activities. (Lawson 1997: 31) What are these social structures? It is easy enough to see that identifiable social structures exist. Items such as (societal) rules, relations and positions clearly depend on human agency as well as condition our every day (physical) activities. The human intentional activities of speaking, writing, driving on the public roadways, cashing cheques, playing games, giving lectures and so forth would be impossible without such social material conditions as rules of grammar, the highway code, banking systems, rules of play, teacher–student relations, etc. (Lawson 1997: 31) This is a kind of transcendental argument which is being deployed to prevent a retreat into relativism: one takes as self evident the fact that we are intentional agents and asks what is presupposed in the world by this attribute. This type of argument is supplemented by another. There is a further method for gaining knowledge about structure: retroduction. Retroduction, in contrast to both induction and deduction, is to move from the level of phenomenon identified to a different ‘deeper’ level in order to explain the phenomenon, to identify a causal mechanism … But can anything be said about this process of retroduction? Those working on critical realism, including myself, have generally rested content with making the observation that it ‘relies upon a logic of analogy and metaphor’. (Lawson 1999: 10) I have three comments to make on this positioning of critical realism in the philosophy of science.

156

Shaun Hargreaves Heap

First, it self consciously (see Lawson 1997: xvi) takes a different route to what seem to me to be broadly the same conclusions about the nature of science found in the lineage in the philosophy of science that can be traced from Quine’s (1953) famous essay on the dogmas of empiricism to Kuhn’s (1962) and Lakatos’s (1970) work on the history of science. That is, science does not consist of some simple form of empirical investigation of the world: there are always unexamined foundational assumptions in science which are upheld by what is, in effect, a collective judgement among a community of scientists.2 Does taking a different route to this conclusion matter? It seems to in the sense that the critical realist route produces a clear view on the structured nature of reality which sets an agenda for how science should proceed. In comparison the Quine and Kuhn type arguments have reinvigorated a form of pragmatism which is less clear about the nature of reality and much less prescriptive for science. Pragmatism holds that theories should be distinguished by whether they enable us to engage more or less successfully with the world and this leaves plenty open for debate: not least who are ‘they’ and what is ‘success’? Local interpretations of ‘they’ and ‘success’ seem unavoidable and so pragmatism is liable to license a variety of views; and it does, even within a single locale like today’s western academy. For example, in the hands of someone like Quine, the pragmatic judgement that a natural scientific method (which privileges logic and formal rigour) has proved amazingly successful in this respect has provided the foundation for an analytic philosophy with those same ingredients. Equally, a similar pragmatic judgement by Rorty (1989) with respect to, say, the influence of Freud in the social world encourages an hermeneutic philosophy. This may look like an important difference, with pragmatists seeing social science as a kind of critical engagement between an ineliminable plurality of views whereas critical realism offers an apparently more settled view on such matters as the structured nature of reality and how to go about science. However, given what critical realists also say about the local mediation of knowledge claims by the historically specific practices of science, such differences might be more apparent than real. Nevertheless, one point of dispute is likely to survive any sympathetic reading here. The privileging of ‘retroduction’ is unlikely to find universal favour among pragmatists. Indeed, pragmatists are much more likely to cast the role of ‘analogy’, ‘metaphor’ and the like in a postmodern way as devices which are used to persuade a scientific community of knowledge claims rather than as some device which can be used to gain insight into reality. Second, there is an important argument here about how to prevent the retreat into relativism (which is shared by pragmatists). It is to play the ‘realist’ card. There are simply some constraints on the ideas that I can maintain about the world which are supplied by the nature of that world.

The heterodox tradition in economics 157 For instance, it is evident that having the idea that ‘humans can fly by flapping their hands’ will not do the trick given our physiognomy. Likewise, the idea of ‘driving on the left’ will be successful only when others hold the same idea. These are trivial examples, but they are enough to show that there are natural and social constraints on the ideas that enable any individual to engage successfully with the world. This shared move is transparently important for the heterodox tradition’s positioning in the wider ‘objectivity versus relativism’ debate in the philosophy of social science. What is perhaps less obvious is that it is also important for heterodox engagement with orthodox economics. I shall say more on this below. For now I want simply to suggest that some philosophical defences of the orthodoxy in economics appear to owe little to positivism and more to the kind of relativism which makes economics a kind of self-referential game of the Wittgensteinian sort.3 Thus by avoiding both extremes, critical realism actually helpfully undermines two important methodological supports to the practice of orthodox economics (and not just one as is sometimes supposed). Third, I am in complete sympathy with the quasi transcendental argument that gives us some knowledge of the social world. This, then, is another important aspect of the contribution of critical realism which surfaces in the course of their resistance to the extremes of positivism and relativism. However, I am not persuaded that the actual critical realist argument here shows that social structures are irreducible. Since this relates directly to a second aspect of the heterodox tradition, I shall leave further discussion of this for the next section.

III Ontology matters: situated rationality Critical realists make ontology matter during their attack on positivism and their quasi-transcendental argument delivers a very specific conclusion about the ontology of the social world. There are social structures which are ontologically irreducible. I argued … that because social structure is a precondition of intentional human action and interaction … then the properties of social structure cannot be reduced to, or explained completely in terms of, the capabilities of the human agent. (Lawson 1997: 176) The key to this claim is the thought that people could not act intentionally without a certain degree of knowledgeability and this presupposes some social rules (see the earlier quotation). The difficulty with this claim is that it seems to overlook arguments which attempt to show that social rules could be the deposits of earlier interactions between intentional agents who do not have any social rules (see Schotter 1981 and Sugden 1986). In

158

Shaun Hargreaves Heap

other words, there are arguments which make such social rules reducible. So the critical realist claim regarding irreducibility cannot be taken as some self-evident piece of (logical) inference from the fact of intentionality. Instead, it needs some further justification in the form of an explanation of what is wrong with these alternative accounts of the matter. This has occupied much of my recent work and so I am parti pris. This said, I feel that the critical realist contribution to this aspect of the heterodoxy has been less than helpful, as it has encouraged jumping to conclusions on irreducibility rather too quickly. This is not to say that the claim is wrong. Indeed quite the reverse is the case, as I agree with critical realists on this matter. The point is that it takes a longer argument to arrive at this conclusion. The key starting point for such an argument is to be sure what model of individual action is being used to support the claim on reducibility/irreducibility; otherwise it is impossible to judge whether reduction is or is not possible. In the case of the particular arguments from Schotter (1981) and Sugden (1986), they have a bounded version of the standard economic model of instrumental rationality and show how in repeated interactions, say, to take Lawson’s example of the highway code, of driving on the road, a convention will quickly emerge where either everyone drives on the right or everyone drives on the left. This proof seems impeccable in its own terms. Nevertheless, there are two points of criticism. One is that there are chance elements and the peculiarities of the belief updating processes that determine which of the conventions emerge. So the explanation does not run solely in terms of instrumentally rational individuals. These extra elements may not seem to count for much if they are simply determining whether a population drives on the right or the left, but in more complicated social interactions the selection of one convention over another carries much more interesting implications. For example, property rights are, in this tradition, often explained as the result of a convention emerging in a repeated interaction around some disputed resource that has the form of a Hawk–Dove game. In cases like this, a piece of extraneous information can be used to distinguish between the parties to the Hawk–Dove interaction and can form the basis of a convention that assigns one of the parties to the Hawk and the other to the Dove role. So the selection of convention effectively determines the allocation of de facto property rights. The second weakness in the evolutionary account of social rules is the failure to explain how social rules of this sort seem to come with a set of validating beliefs. Thus a full account of why people drive on the left, say, ought to acknowledge that people seem to believe that conformity with the convention has a quasi-moral character. In other words, someone who drove on the right in these circumstances would not just be foolish, they would also be in some degree morally culpable. The point is even clearer

The heterodox tradition in economics

159

where the observation of a convention of property rights is not just sensible, it is usually supported by a set of beliefs which makes acceptance of the rights the ‘right’ or ‘just’ thing to do. This last observation is what informs, in my own work, a different sense of rational agency: expressive rationality (see Hargreaves Heap 1989, 2000). The idea here is that people do not just act so as to secure their objectives (as instrumental rationality holds), they also want to make sense of the world in which they act and this often leads them to act in ways which express beliefs about how to make sense of the world. When these beliefs are shared and encoded in norms, expressive acts of this kind are a form of norm-guided action. This is not the place to develop these ideas, I mention them in outline simply because they are not dissimilar to what Lawson, for example, says on the topic of ‘situated rationality’. Situated rationality proposes three levels of agency. There is the purposive level, there is a reflexive monitoring level and there is an unconscious level which supplies an orientation towards the avoidance of anxiety (see Lawson 1997: chapter 13). It would only take some loose translations like the equation of anxiety avoidance with ‘making sense of the world’, reflexive monitoring with expressive acts, and purposive with instrumental to see that on matters of substance we have much in common. The difference in a sense is whether one has to develop such a model of human agency to see what is wrong with reducibility claims which use a much narrower version of human agency (instrumental reason). I want to put this difference in some perspective with two observations. The first is that the Sugden and Schotter evolutionary arguments are just two of many which attempt to give an account of action exclusively in instrumentally rational terms. There are further important arguments in this vein in the rational expectations literature and in standard game theory. They too need to be considered if the irreducibility claim is to be advanced. Second, the irreducibility claim is important for heterodox economics in three respects: the claim forms the basis for heterodox arguments against (a) methodological individualism (of a particular kind) and (b) the imperial expansion of the instrumentally rational (rational choice or economic) model of rational action to the other social sciences (e.g. see Coleman 1990 and Becker 1996); (c) the richer model of rational agency associated with developing this claim is in part what distinguishes heterodoxy from orthodoxy. In other words, I think the irreducibility claim is important for heterodox economics and critical realism is helpful when it asserts the claim, but the claim needs more careful defence than it currently receives from critical realism. I shall develop some of these observations in the next section from a slightly different perspective.

160

Shaun Hargreaves Heap

IV The attack on mainstream economics The final part of my assessment of critical realists’ contribution to heterodox economics concerns their part in the attack on mainstream economics. Critical realists make a distinctive contribution here by making the grounds for attack strictly methodological, in a narrow sense of epistemological (although see my later comments on this below). One primary object of critical realism in economics has been to identify the basic nature or character of mainstream economics. Most critics of the mainstream project have, in attempting to characterise it, isolated particular substantive claims, such as theories of rationality, or of equilibrium. It has been the contention of critical realism, in contrast, that it is at the level of method that its essence lies. (Lawson 1999: 4) Some care is required in assessing this advice. Critical realists characterise orthodox method as ‘deductivist’ and note that positivism supplies one possible underpinning for this method. So they do not want to claim that the orthodoxy is governed by positivism. On the other hand, in practical terms, most of their criticism of deductivism is based on the criticism of positivism, so it is difficult not to see the methodological target as positivism (e.g. see Lawson 1992). With this qualification in mind, I have reservations about making method in this sense the point of an heterodox attack. 1

2

3

As a matter of fact, many mainstream economists are not ‘positivist’ in this strict sense. There are for example varieties of empiricist practitioners in economics who nevertheless believe in ontology. Equally there are theorists who seem totally unconcerned with any kind of empirical investigation (see Blaug 1980 and McCloskey 1983, who provide examples of this, albeit while drawing different conclusions). Of course, to the extent that deductivism can be rendered distinct from positivism, then this particular reservation may disappear. Most economists do not think about their method. This, after all, is the Kuhnian insight about normal science. You have to suspend disbelief at the foundations in order to get on with science. Critical realists acknowledge this but do not seem to draw the inference that I do, which is that a methodological argument of this sort (whether it is pitched at deductivism which is distinct from positivism or not) simply won’t engage most people’s attention in the orthodoxy. In part because of 2 but also because of my underlying pragmatism, I believe that the way to persuade the mainstream that there is

The heterodox tradition in economics 161 something wrong is to engage with them ontologically as it affects actual economic analysis. This last observation appears in marked contrast to, for instance, Tony Lawson when he says: Probably the ‘request’ I come across most frequently in the context of discussing critical realism is for examples of ‘critical realist substantive theory’ meaning the critical realist accounts of some highly concrete phenomena. In my view there can be no such thing. (Lawson 1999: 14) It may be that I am reading too much into this and the earlier quotation because it is plain from my discussion in the previous sections that critical realist argument against positivism precisely has the effect of making ontology matter. And once ontology matters, it is difficult to see why it should not become centrally part of the terrain on which the orthodoxy is challenged. Indeed there are obvious ways in which this is exactly what critical realists do (e.g. when ‘situated rationality’ is contrasted with the rational choice model). I shall conclude this section with a brief sketch of my ontological engagement with the orthodoxy because it reveals several further points of congruence with critical realism. My ontological challenge has focused, as I suggested earlier, on the inadequacy of an instrumental understanding (or the rational choice model) of individual agency when understood as the exclusive account of rational action. I have focused my attack here in part because the exclusive reliance seems to threaten to impoverish social life. There is more to social life than instrumental exchange but, in so far as we come to think of ourselves as exclusively rational choice type of people, social life will become no more than a set of instrumental exchanges. This goes back to the potential for theories in social science to become self-fulfilling and my reason for being concerned with getting the ontology right. What makes this really worrying is twofold. First, there is some evidence of this effect with the rational choice model (see Frank et al. 1993, where those subjects with a training in economics were more likely to defect in a prisoners’ dilemma experiment than those without). So the potential for a theory to become self-fulfilling in the social sciences is real in this case. Second, there is the nature of the impoverishment that could follow. The early Jon Elster (1983) was especially eloquent on what might be lost in moving to a world where social life consists only of instrumental exchange. Likewise, there is a history of concern here, for example see Adorno and Horkheimer (1972), whose critique of capitalism turns in part on the analysis of the elevation of this form of reason. In outline, though, it is easy to see what might be at stake here. There are a range of experiences from love and friendship to aesthetic and moral feelings

162

Shaun Hargreaves Heap

which are often distinguished from other kinds of experience precisely because they do not involve exchange or instrumental return. Of course some redescription of the actions associated with these experiences can always turn them into the product of an exchange. This is always possible even if it seems laboured but, once perceived in this way, these experiences can only begin to lose what once made them distinctive. This is the sense of impoverishment that worries me and it turns on exactly the point that is recognised by critical realism: the potential transformative influence of theory (e.g. Lawson 1997: 277). One tactic I have used in my engagement with mainstream economics has been a kind of cataloguing of how some important pieces of economic analysis depend on other notions of (rational) agency. My favourite examples here are those settings which are captured by a game where there are multiple Nash equilibria and those where people make an inspired or entrepreneurial decision. The one helps introduce a sense of norm-governed behaviour and the other a more creative or existential type (Hargreaves Heap 1989 and Hargreaves Heap and Varoufakis 1995). Another tactic is similar to that of critical realists: it involves the use of a kind of transcendental argument. In particular, in recent times I have started from the self-evident fact that we are reflective agents and that these reflections have a motivational force. That is, we reflect on what we do and derive a sense of self respect from these reflections or the reverse, as when we experience feelings of embarrassment, shame or guilt; and the anticipation of these feelings affects our actions. The question then is what kind of model of rational action can accommodate these feelings and their attendant actions. In particular, can the rational choice model account for these feelings and their motivational force? This is not the place to give a detailed answer to this question: suffice to say that I claim that either the instrumental model must be complicated with a two-tier structure of preferences or some fresh notion of expressive agency is required (Hargreaves Heap 2000). The overlap with the idea of ‘situated rationality’ will be plain. So, setting aside the issue of whether the attack can lead on ontology, there is no dispute either that ontology matters or over the broad understanding of what makes an individual act. I would put in a plea for my way of describing this on two grounds. One is that it maintains points of contact with the orthodox view. This is helpful both because it would be foolish to deny that in some aspects of life the instrumental/rational choice account does explain action and because it is possible to show that the model is inadequate in other settings. In this way, the language of rational choice can be used to help define the respects in which it is inadequate. The other ground is that my language enables one to draw on a body of psychological evidence with useful effects. In particular, I argue that, in the presence of additional or more complicated forms of agency, the prescriptions of the rational choice model are liable to be deeply

The heterodox tradition in economics 163 misleading because of the possible interaction between the different kinds of motivation (Hargreaves Heap 1989). The point here is essentially the same as that made by Frey (1992), who draws on the psychological literature which uses the terms intrinsic and extrinsic reason. Extrinsic reason is akin to instrumental reason and the problem is that an adjustment of the material incentives to do something which increases the extrinsic/instrumental reason for doing something can reduce the intrinsic/expressive reason for doing it. With such an interaction between the two types of reason, the result is much weaker (and possibly even perverse) as compared with what had been expected with reference to the operation of instrumental reason alone. The existence of such interaction effects is receiving increasing empirical support from laboratory experiments in economics and it helps explain some well known, and surprising from the perspective of instrumental reason, stylised facts like the failure of payments-by-results systems. I mention these points in outline because it seems to me that it is precisely this kind of pragmatic argument that is liable to persuade the mainstream and this is the kind of argument which the heterodox tradition needs to take to the orthodoxy.

V Conclusion Here is a checklist. ‘Yes’ to realist anchors, ontology, transcendental arguments, metaphor and analogy, richer models of rational agency and critiquing the mainstream. Critical realism has helped heterodox economists substantively in these respects. ‘No’, however, to making all these things matter because they are denied by positivism, making the critique of orthodox economics turn on the critique of positivism and to arguments which make social structures self evidently irreducible. In these respects, I don’t feel that critical realism has been helpful on matters of substance to the heterodox tradition (although recall my earlier caveat with respect to whether deductivism can be separated from positivism). Nevertheless, even in these respects, by being a part of the debate (indeed often provoking it), critical realism has helped heterodoxy to see things more clearly. I struggled in my first draft to make sense of this checklist. How can one best characterise this mixture? My early attempt cast it in terms of my pragmatism overlapping in some respects with that of critical realism: that is, I was coming to similar conclusions but for different reasons because, while pragmatism shares with critical realism a critique of positivism (and relativism), it has drawn different conclusions for the conduct of science. In rewriting the chapter, I am now inclined to summarise matters slightly differently. There is a kind of ambitious claim which critical realists might want to make in regard to heterodox economics. They might want to say that

164

Shaun Hargreaves Heap

critical realism offers a way of cohering the disparate character of heterodox economic thought. ‘Rally round the banner of critical realism, so to speak, and the heterodoxy will enjoy the kind of coherence that can truly challenge the orthodoxy.’ In other words, the heterodoxy might in this way become the new orthodoxy. I deliberately phrase this in terms of ‘might’ throughout because I do not know whether such claims are made. Nevertheless, it is easy to see how such a claim might be made and how, given what I have said earlier in the chapter about the contribution of critical realism to strengthening heterodox argument, it might have some substance. The claim, however, has another angle to it. It provokes, in effect, a question regarding what heterodoxy is aiming for. Now this, it seems to me, is a great service because I am not aware of this as a question that is addressed much by heterodox economists. In particular, to put the choice starkly, is heterodoxy aiming to become a new orthodoxy in the same vein as the old orthodoxy (that is, with shared principles of one kind or another that define a programme of research which is then conducted with indifference to those outside the programme) or does it wish to supplant this notion of orthodoxy (for example, by encouraging a kind of continuous debate between a plurality of views)? There is a disclaimer that I should finish with. I have found it very useful to write this chapter for my own work. It has made my own work seem more coherent than it really is, in the sense that it might suggest I have been executing some carefully laid plan. Indeed, it was this sense of an emerging individual coherence which seeded the larger idea of critical realism helping to cohere the heterodox tradition. In my own case, though, my behaviour has for the most part, as for most economists, I imagine, been more akin to blundering about than some measured execution of a coherent plan. Naturally it is gratifying to find some order to this roaming and this is why the chapter has been useful to me. However, it would be disingenuous not to recognise that the order is in part an ex post invention. This, though, makes my disclaimer slightly disingenuous because this is, of course, the point. It is the scope for invention in human life that I hope we will not lose from economics. The ability, after all, to create experiences through our own description and understanding does seem rather extraordinary and it would be a pity to constrain this talent artificially by working with a single rather simple and ultimately not especially flattering model of human agency; and in this I am surely joined and helped by critical realism.

Notes 1

I take this way of characterising the extremes from Lewis (1999). Positivism is, of course, also frequently discussed by Lawson among others because it supplies a basis for what he refers to as a ‘deductivist’ form of explanation found in mainstream economics.

The heterodox tradition in economics 165 2

3

The totality of our so-called knowledge or beliefs, from the most casual matters of geography and history to the profoundest laws of atomic physics or even pure mathematics and logic, is a man-made fabric which impinges on experience only along the edges. Or, to change the figure, total science is like a field of force whose boundary conditions are experience. A conflict with experience at the periphery occasions readjustments in the interior of the field … But the total field is so underdetermined by its boundary conditions, experience, that there is so much latitude of choice as to what statements to re-evaluate in the light of any single contrary experience. No particular experiences are linked with particular statements of the interior of the field, except indirectly through considerations of equilibrium affecting the field as a whole … Any statement can be held true come what may, if we make drastic enough adjustments elsewhere in the system. This is Quine’s famous way of expressing what has become known as the Duhem–Quine objection in philosophy to the versions of empiricism which have theories rising or falling depending on how they fare when confronted by the facts. Theory is underdetermined by the evidence because our beliefs form a ‘seamless web’ and face the ‘tribunal of experience together’. As a result, there is something other than a simple matter of empirical rectitude that governs the conduct of science. In particular, in so far as science involves empirical evidence, then there are some ineliminable prior and privileged beliefs which govern the interpretation of empirical evidence. I’m thinking of some of Frank Hahn’s defences of competitive general equilibrium theory.

References Adorno, T and M. Horkheimer (1972) The Dialectics of Enlightenment, London: Verso. Becker, G. (1996) Accounting for Tastes, Cambridge, MA: Harvard University Press. Blaug, M. (1980) The Methodology of Economics: Or How Economists Explain, Cambridge: Cambridge University Press. Boylan, T. and P. O’Gorman (1999) ‘Critical Realism and Economics: A Causal Holist Critique’, in S. Fleetwood (ed) Critical Realism in Economics: Development and Debate, London: Routledge. Coleman, J. (1990) Foundations of Social Theory, Cambridge, MA: Bellknap Press of Harvard University. Elster, J. (1983) Sour Grapes, Cambridge: Cambridge University Press. Frank, R., T. Gilovich and D. Regan (1993) ‘Does Studying Economics Inhibit Cooperation?’, Journal of Economic Perspectives, 7: 159–71. Frey, B.S. (1992) ‘Tertium Datur: Pricing, Regulating and Intrinsic Motivation’, Kyklos, 45: 161–84. Hacker, P. (1996) ‘Post-Positivism in the United States and Quine’s Apostasy’, in Wittgenstein’s Place in Twentieth Century Analytic Philosophy, Oxford: Basil Blackwell. Hahn, F. (1980) Money and Inflation, Oxford: Basil Blackwell. Hargreaves Heap, S. (1989) Rationality in Economics, Oxford: Basil Blackwell.

166

Shaun Hargreaves Heap

—— (2000) ‘Expressive Rationality: Is Self-Worth Just Another Kind of Preference?’, in U. Mäki (ed.) The Economic Realm, Cambridge: Cambridge University Press. —— and Y. Varoufakis (1995) Game Theory: A Critical Introduction, London: Routledge. Kuhn, T. (1962) The Structure of Scientific Revolutions, Chicago: University of Chicago Press. Lakatos, I. (1970) ‘Falsificationism and the Methodology of Scientific Research Programmes’, in I. Lakatos and A. Musgrave (eds) Criticism and the Growth of Knowledge, Cambridge: Cambridge University Press. Lawson, T. (1992) ‘Methodology: Non-optional and Consequential’, Royal Economic Society Newsletter, October, 2–3. —— (1997) Economics and Reality, London: Routledge. —— (1999) ‘Developments in Economics as Realist Social Theory’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London: Routledge. Lewis, P. (1999) ‘Metaphor and Critical Realism’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London: Routledge. McCloskey, D. (1983) ‘The Rhetoric of Economics’, Journal of Economic Literature, 21: 481–517. Mäki, U. (1990) ‘Practical Syllogism, Entrepreneurship and the Invisible Hand’, in D. Lavoie (ed.) Economics and Hermeneutics, London: Routledge. Quine, W. (1953) ‘Two Dogmas of Empiricism’, in W. Quine (1961) From a Logical Point of View, Cambridge, MA: Harvard University Press. Rorty, R. (1989) Contingency, Irony and Solidarity, Cambridge: Cambridge University Press. Schotter, A. (1981) The Economic Theory of Social Institutions, Cambridge: Cambridge University Press. Sugden, R. (1986) The Economics of Rights, Cooperation and Welfare, Oxford: Basil Blackwell. Weber, M. (1922) Economy and Society, edited by G. Roth and C. Wittich, New York: Bedminster Press (1968).

9

Economics as Social Theory and the New Economic Sociology Paul Lewis

Introduction One of the central aims of the project of critical realism in economics is the restoration of the study of economic life as a branch of social theory. As the manifesto of the Economics as Social Theory project found in the frontispiece of books in the eponymous series puts it: Social theory is experiencing something of a revival within economics. Critical analyses of the particular nature of the subject matter of social studies and of the types of method, categories and modes of explanation that can legitimately be endorsed for the scientific study of social objects, are re-emerging. Economists are again addressing issues such as the relationship between agency and structure, between economy and the rest of society, and between enquirer and the object of enquiry … The objective for this series is to facilitate this revival further. In contemporary economics the label ‘theory’ has been appropriated by a group that confines itself to largely asocial, ahistorical mathematical ‘modelling’. Economics as Social Theory thus reclaims the ‘theory’ label, offering a platform for alternative rigorous, but broader and more critical conceptions of theorising. (Lawson 1997: ii) The goal of advancing the social-theoretic analysis of economic life is one that critical realists share with a number of other groups of scholars, amongst whose ranks heterodox economists feature prominently. But a commitment to explaining economic phenomena using sociological theories is one that critical realists also hold in common with another prominent group of scholars drawn from outside the narrow confines of academic economics, namely economic sociologists. The past two decades have seen a surge of interest in economic sociology, associated with the rise of the so-called ‘New Economic Sociology’. Advocates of the latter contend that economic sociology involves the application to economic

168

Paul Lewis

phenomena of explanatory frameworks drawn from sociology (Smelser and Swedberg 1994: 3). This commitment to the social-theoretic analysis of economic affairs suggests that there may be a good deal of affinity between the New Economic Sociology and the project of Economics as Social Theory. However, to the best of this author’s knowledge, there has yet to be a detailed comparison of the two bodies of thought. The aim of the present chapter is to remedy that lacuna. It does so by using the vantage point provided by critical realism to examine the work of one of the most prominent contemporary economic sociologists, namely Mark Granovetter. Granovetter is an interesting candidate for such a study, for two main reasons. In the first place, his seminal article on ‘Economic Action and Social Structure: The Problem of Embeddedness’, published in the American Journal of Sociology in 1985, is widely acclaimed as a landmark contribution in the rise of the New Economic Sociology. Indeed, it was Granovetter who first coined the phrase ‘New Economic Sociology’ at an American Sociological Association talk based on his paper (Swedberg 2003: 34). Second, his account of ‘embeddedness’ has had a good deal of influence on heterodox economists, often being invoked by the latter as a way of differentiating themselves from the mainstream.1 A survey of Granovetter’s work from the vantage point provided by critical realism will therefore serve to illustrate some of the ways in which the latter can underlabour not only for economic sociology but also for heterodox economics.

Granovetter and the case for a new economic sociology Granovetter’s writings, and the ‘New Economic Sociology’ for which they provided the catalyst, are distinguished from the majority of earlier work in economic sociology by their willingness to trespass on territory that was hitherto regarded as the exclusive preserve of economists.2 The ‘Old Economic Sociology’, as Granovetter (1990a) terms it, accepted the division of labour between economics and sociology advocated by Talcott Parsons (1934: 526–9; also see Granovetter 1990a; Velthuis 1999). Parsons stipulated that the topic of the allocation of scarce resources between competing ends was to be studied by economists, with sociologists being restricted to investigating the origins of the ends in question. Consequently, practitioners of the Old Economic Sociology tended for the most part to accept that market processes lay outside the legitimate domain of sociological analysis. In stark contrast, Granovetter and other advocates of the New Economic Sociology have sought to legitimise sociological accounts of market processes by arguing that a sociological approach can yield insights into economic activity which are not captured by the analytical tools utilised by economists (most notably, rational choice theory and transactions cost economics) (Granovetter 1985: 504–5,

The New Economic Sociology

169

507; 1990a: 92, 95; 1990b: 106–7). Orthodox economics employs ‘a radically non-social approach’ which ‘will always have great difficulty in explaining what takes place in many real economic circumstances’, the consequence of which is that ‘many economic problems, which by tradition are seen as belonging to the economists’ camp, can be better analysed by taking sociological considerations into account’ (Swedberg and Granovetter 2001: 1, 14, 2). At the heart of Granovetter’s argument for a sociological analysis of economic phenomena is his belief, which closely resembles the views held by critical realists, that social-scientific research is most likely to be fruitful if it is conducted using methods which are tailored to suit the nature of the socio-economic material under investigation: It seems to me that economists should not worry so much whether something can in fact be described within a purely economic framework, but rather [should] pay attention to the phenomenon from whatever framework it can be described. I don’t see why one should be so reluctant to explain economic phenomena in styles and modes of explanation that originate from other disciplines. (Granovetter 1990b: 111) Of course, if it is indeed the case that Granovetter (like critical realists) subscribes to the methodological realist desideratum that researchers should employ explanatory frameworks that are well-suited to the nature of the socio-economic phenomena under investigation, then the question arises of his views concerning the nature of the socio-economic world. Granovetter’s critique of orthodox economics (including its transactions cost variant), together with his blueprint for the New Economic Sociology, are based on two key ontological propositions which in his view militate against rational choice theory and transactions cost economics and in favour of sociological analyses of economic phenomena: the first is that economic activity is embedded in networks of social relations; the second is that economic institutions are socially constructed. We shall consider each of Granovetter’s propositions in turn. 1 The social embeddedness of economic activity Granovetter uses the term ‘embeddedness’ to denote the fact that people are social beings whose identities, values, beliefs and goals, and consequently whose conduct, are all profoundly shaped by their location within networks of ‘concrete, ongoing personal relations’ (Granovetter 1985: 490). According to Granovetter, the way in which people interact with one another in the present is conditioned by how they behaved towards each other in the past, and more specifically by the personal social relations to which those past dealings gave rise. For instance,

170

Paul Lewis

relations of trust may develop between people who have dealt honestly with each other in the past. The key point about such relations, Granovetter argues, is that they are infused with normative content and so generate obligations and standards of behaviour (for example, expectations and norms of fairness and reciprocity) the motivational force of which can outweigh the dictates of pure self-interest: ‘[T]rust’ … refer[s] to circumstances where one enters a transaction believing that transaction partners will behave properly for reasons that transcend pure self-interest. (Granovetter 1992a: 40) On this view, relations of trust can inspire people to depart from the instrumentally rational and opportunistic forms of action emphasised by rational choice theory and transaction cost economics in favour of more cooperative modes of conduct that involve them subordinating the pursuit of their own self-interest to the attainment of collective goals and the common good (Granovetter 1985: 489–91; 1992a: 41–4; 1999: 160–1, 2002: 35–42; Swedberg and Granovetter 2001: 11–13). For Granovetter, then, the propensity to pursue one’s self-interest is not the only driving force of human action. People who are embedded in networks of social relations have the capacity to set aside the goal of satisfying their own desires in order to fulfil duties and obligations to others and to conform to norms of justice and fairness that ‘project an aura of being the right and proper way to behave, quite aside from advantages to be gained from following them’ (Granovetter 1999: 160). Hence, Granovetter agrees with the view, also promulgated by critical realists such as Lawson (1997: 106, 186–8) and Archer (2000), that people’s capacity for instrumentally rational and opportunistic modes of conduct may not always manifest itself in their actual behaviour, due to the countervailing impact of higher-order aspects of human reason (Sen 1979: 102–7). And if the self-interested drive to satisfy one’s own desires is but one of a number of distinct motivations for individual action, then approaches like rational choice theory and transactions cost economics, which assume that people behave only in an instrumentally rational and opportunistic fashion, rest on an overly simplistic and narrow account of human motivation. Granovetter’s portrayal of people as social beings whose attributes and conduct are shaped by the nexus of personal relations in which they are situated contrasts starkly with the ontological commitments of orthodox economics. The latter presupposes an undersocialised model of man ‘in which individuals are analysed as if atomised from the influence of their relations with others … and from the past history of these relations’ (Granovetter 1988: 187–8). Rational choice theory and transactions cost economics simply assume that people’s characteristics and behaviour are

The New Economic Sociology

171

fixed independently of the relations into which they enter, sponsoring a view of society as simply the additive or, to use John Stuart Mill’s term, ‘mechanical’ sum of the isolated individuals who populate it (Granovetter 1985: 483; 1992a: 29; 1993: 11). The problem with this atomistic account of socio-economic reality (and by extension with those economic models which presuppose it) is that it ignores the fact that ‘[e]conomic action (like all action) is socially situated, and cannot be explained by individual motives alone’ (Granovetter 1992b: 4). For Granovetter, as we have seen, there are sources of motivation (such as norms of fairness and reciprocity) whose very existence derives from people’s social relations and which can therefore be understood only in relational terms (Granovetter 2002: 54). Participation in society transforms homo economicus into a qualitatively different being – homo socio-economicus – who is animated by new dispositions (for example, the disposition to behave cooperatively, as distinct from the disposition to act in an opportunistic fashion) that can prompt him to pursue the common good rather than his own selfish desires. And the fact that people’s properties (such as their dispositions) are influenced by the personal relations into which they enter implies that society is (at least in part) an organic entity formed, to use Mill’s terminology once again, by the ‘chemical’ (as opposed to the mechanical) composition of its parts (Lewis and Weigert 1985: 967–8; Guillén et al. 2002: 7).3 For Granovetter, then, as for supporters of critical realism, society amounts to more than the aggregate of its individual human inhabitants (Lawson 1997: 30–2, 63–5, 160–6; Lewis 2000: 250–3, 258–9). However, the fact that orthodox economics is based on an atomistic model of man renders it incapable of doing justice to the motivations and cooperative behaviour to which people’s personal relations give rise: The extreme version of methodological individualism that dominates much of modern economics makes it difficult to recognise how economic action is constrained and shaped by the structures of social relations in which real economic actors are embedded … [T]he main difficulty [with orthodox economics] lies … in the neglect of social structure. (Granovetter 1992b: 4; also see Granovetter 1985: 504; 1999: 158 n. 2, 159, 162) Only if economists stop thinking of people as socially isolated Robinson Crusoes and instead conceptualise them as intrinsically social beings, situated in networks of relations which are infused with normative significance, will they become able to do justice to the possibility that people may be animated by loftier motivations than pure self-interest. It follows that if researchers want to understand real-world economic

172

Paul Lewis

activity, then they must devote less attention to economic models populated by isolated Robinson Crusoes and make greater use of sociological theories which attempt to conceptualise how people’s actions are shaped by (and in turn shape) networks of social relations. And it is in sociological theory that the best developed analyses of this issue are to be found (Granovetter 1985: 504–7; 1990b: 101–8; 1999: 160–2; Swedberg and Granovetter 2001: 1, 8, 11–12).4 Granovetter argues, then, that social scientists should eschew (orthodox) economic analysis, with its undersocialised model of man, in favour of sociological theories which treat people as social beings who are embedded in networks of personal relations. However, in making the case for the sociological analysis of economic activity, Granovetter is acutely aware of the need to avoid a second extreme conception of human nature, sometimes found in sociological theory, which portrays people as so unquestioningly obedient to social rules and norms that their behaviour is no more than an automatic, mechanical response to their social surroundings. Such oversocialised accounts exclude economic motives entirely and leave little if any room for genuine human choice, a lacuna that underlies James Duesenberry’s well-known comment that: ‘Economics is all about how people make choices; sociology is all about why they don’t have any choices to make’ (Granovetter 1985: 482, 485–7, 491; 1992b: 5–6). Like advocates of critical realism, then, Granovetter argues that a fruitful analysis of economic activity requires researchers to avoid both the under- and oversocialised models of man (Granovetter 1985: 487; cf. Lawson 1997: 38, 166–8; Lewis and Runde 2002: 203–5). What is required, Granovetter (2002: 41, 37) contends, is a ‘balanced account’ which ‘call[s] attention to the mixture of economic and social motives that people pursue while engaged in production, consumption and distribution’. And it is just such an account that Granovetter’s embeddedness approach purports to provide. The concept of embeddedness fulfils what critical realists would describe as an ‘underlabouring’ role, alerting social scientists to the diversity of motives driving economic activity and thereby helping them to avoid the mistake of assuming a priori that one or the other form of motivation is always and everywhere dominant (Granovetter 1993: 38; cf. Lawson 1997: 62, 106, 158–9). For Granovetter, as for critical realists, the issue of which source of motivation is the dominant influence on people’s behaviour is a contingent matter which can only be resolved ex posteriori by examining the details of individual cases: The embeddedness approach to the problem of trust and order in economic life, then, threads its way between the oversocialised approach of generalised morality and the undersocialised one of impersonal, institutional arrangements by following and analysing concrete patterns of social relations. Unlike either alternative, or the

The New Economic Sociology

173

Hobbesian position, it makes no sweeping (and thus unlikely) predictions of universal order or disorder but rather assumes that the details of social structure will determine which is found. (Granovetter 1985: 493; also see Granovetter 1985: 491; 1990a: 106, 107; 1992b: 5; Swedberg and Granovetter 2001: 11–12) The vantage points provided by the New Economic Sociology and critical realism reveal that the underlying problem with both the undersocialised and the oversocialised approaches is that each adopts an a priori position which prevents it from doing justice to the full richness of the motivations which drive economic action: the undersocialised approach simply assumes a priori that the untrammelled pursuit of self-interest is everywhere dominant, ignoring the possibility that people’s location in networks of social relations can induce them to depart from instrumentally rational and opportunistic behaviour; oversocialised accounts assume a priori that social norms and values alone motivate people, thereby exaggerating the degree to which the social rules and norms offset the pursuit of self-interest and eliminate malfeasance. The drawback of such a priori approaches is that they prejudge the issue of which source of motivation is dominant in any specific case and so offer a one-sided and therefore potentially highly misleading account of economic activity.5 2 The social construction of economic institutions The second ontological proposition that centrally informs Granovetter’s work is that economic institutions are socially constructed. Granovetter introduces this idea via a critique of orthodox economics in general and transactions cost economics in particular. The latter attempts to explain the existence of economic institutions (including firms and other forms of organisation) in terms of their (alleged) efficiency. As its most prominent exponent, Oliver Williamson (1985: 1), puts it, ‘the transactions cost approach maintains that … institutions have the main purpose and effect of economising on transactions costs’. Granovetter argues that transactions cost economics is deeply flawed. More specifically, Granovetter identifies two main defects with transactions cost economics, both of which echo critical realists’ reservations about the merits of that approach. In the first place, it assumes that the efficient institutional solution arises as an automatic response to prevailing economic conditions. In other words, transactions cost economics presupposes that the socio-economic world is a closed system in which the existence of transactions with specific features is regularly associated with the emergence of particular (efficient) modes of governance. The problem is that this leaves transactions cost economics

174

Paul Lewis

with nothing to say about the dynamic processes (or, in the lexicon of critical realism, generative mechanisms) which underpin the development of economic institutions, most notably the role of human agency in general and entrepreneurial creativity in particular. To put this point slightly differently, the claim that particular institutions exist simply because they are efficient solutions to economic problems – in this case, the problem of minimising the cost of organising various types of transaction – reeks of functionalism, for it amounts to explaining the existence of socio-economic phenomena (here, institutions) in terms of their effects (in this case, the usefulness or efficiency of their contribution to the wider economic system of which they form a part) (Granovetter 1988: 188–90; 1990a: 96, 102, 105; 1992a: 34, 47–9; 1992b: 4–5, 7; 1994: 435; cf. Foss 1994: 46–7, 52–7; Lawson 1997: 251–2; Pratten 1997).6 Second, Granovetter argues that transactions cost economics is also undermined by empirical studies which find that the efficient institutional solution to the problem of organising transactions is not always realised in practice. That is to say, contrary to the ontological presuppositions of transactions cost economics, empirical evidence reveals ex posteriori that the socioeconomic world is an open system in which the existence of particular types of transaction is not regularly conjoined with the existence of particular (efficient) governance structures (Granovetter 1990a: 106–7; 1990b: 109; 1992b: 5, 9; 1993: 23; Granovetter and McGuire 1998: 153; cf. Lawson 1997: 8–11, 70; Pratten 1997).7 Granovetter contends that the problem with both transactions cost economics and orthodox economics more generally is that by concentrating on equilibrium states and employing a comparative static approach, they deflect attention away from what should be the central focus of attention, namely the contingent, dynamic historical processes through which such institutions arise. Granovetter proposes to remedy this lacuna by offering a distinctively sociological analysis of how institutions come into existence (if, indeed, they do so at all) (Granovetter 1985: 505; 1992a: 34, 47–9; 1992b: 4–5; 1994: 453–4). Drawing on the phenomenological sociology of Berger and Luckmann (1966) and also on Padgett and Ansell’s (1993) analysis of political movements, Granovetter argues that economic institutions are socially constructed in the sense that they arise through the mobilisation of resources by people whose motivation, knowledge and endowments of resources, and therefore whose actions, are facilitated and constrained by the networks of personal relations within which they are situated. According to this open-systems approach, far from being an automatic response to the prevailing economic conditions, the emergence of economic institutions is contingent upon the capacity of socially embedded individuals to mobilise the requisite resources. Where the prevailing network of personal relations makes it difficult for people to do so, the efficient mode of organising transactions may well fail to emerge (Granovetter 1992b:

The New Economic Sociology

175

7–10; 2002: 45–6, 16; 1994: 453–4, 470; Granovetter and Swedberg 2001: 21). For Granovetter, then, the emergence of economic institutions (or lack thereof) is best explained in terms of the interplay over time between people’s actions and the networks of personal relations within which those people are embedded (Granovetter 1990a: 104; 1992a: 47–9; 2002: 49; Swedberg and Granovetter 2001: 38; cf. Archer 1995: 165–344; Lawson 1997: 56–8, 191–271; Reed 2001: 214–21).8

Granovetter on social embeddedness: a critical realist assessment Having outlined Granovetter’s approach and pointed out its affinities with critical realism, the scope for the latter usefully to augment the New Economic Sociology will now be considered. The vantage point provided by critical realism suggests that while Granovetter has performed a valuable service in drawing attention to the impact of personal relations on people’s motivation and actions, he runs the risk of ignoring other important aspects of social structure which may be of considerable practical and explanatory importance. For in conceptualising social structure solely in terms of personal relations, Granovetter subscribes to a restricted metaphysics that leaves little conceptual room for the broader macro-structural context which forms the backdrop for the construction of such micro-structural, personal relations. To see this, note that according to critical realism the social structures in which current activity is embedded are the product of actions undertaken in the past and therefore are ontologically irreducible to current human agency and interaction. More specifically, as we have seen, the structures in question are composed of a nexus of (in many cases internally related) positions, the relations between which are relatively impersonal in the sense that they typically exist independently of the particular individuals who occupy them at a particular moment in time (Archer 1995: 43, 173–6, 206; Layder 1998: 91–6). Such impersonal dimensions of social structure are ontologically distinct from the personal relations that obtain between specific individuals and so cannot satisfactorily be analysed by approaches, such as Granovetter’s, which confine their analytical attention solely to personal relations. The significance of this point can be developed through an examination of Granovetter’s analysis of trust. Recall that Granovetter emphasises the importance of adopting a dynamic approach which transcends the atemporal, comparative static and functionalist analyses of orthodox economics by examining how personal relations of trust can emerge over time out of recurrent interactions between specific individuals. It is necessary to ‘take into account the fact that people have particular relationships’. Granovetter (1990b: 100–1) argues, ‘and [that] they

176

Paul Lewis

deal with each other in ways that are conditioned by the specific history of their interaction’: It is important to avoid what might be called ‘temporal reductionism’: treating relations and structures as if they had no history that shapes the present situation. In ongoing relations, human beings do not start fresh each day, but carry the baggage of previous interactions into each new one. Built into our cognitive equipment is a remarkable capacity, depressingly little studied, to file away the details and especially the emotional tone of past relations for long periods, so that when one has not had dealings with a certain person for many years, a reactivation of the relationship does not start from scratch, but from some set of previously attained common understandings and feelings. It follows that the characteristics of structures of relations also result from processes over time, and can rarely be understood except as accretions of those processes. (Granovetter 1990a: 99; also see 1992a: 34–5) These passages reveal that because Granovetter focuses exclusively on people’s personal relations, his analysis of the legacy of the past is confined to the cognitive and emotional factors – in short, the memories – to which previous personal encounters gave rise. However, critical realists would argue that the legacy of the past also has material and discursive components that are ontologically quite distinct from the cognitive and emotional factors emphasised by Granovetter. For critical realism suggests that the impersonal social structures which form the context in which current socio-economic activity takes place embody a particular distribution of vested interests and resources. At any given moment in time, each of the locations within the nexus of positions of which (on a critical realist account) social structures are composed is associated with a particular set of vested interests. These interests provide the current cohort of people with reasons or incentives to follow particular courses of action (Lawson 1997: 264; Porpora 1998: 352–3). Historically given social structures also influence current activity because at any particular moment in time the material, authoritative and discursive resources required to prosecute various courses of action are distributed unevenly between the various positions which comprise social structures.9 These impersonal structures, and the distribution of vested interests and resources they embody, constitute the macro-structural context in which current action (including the social construction of micro-structural, personal relations) takes place. Given that they are the product of actions taken in the past, and so are external to and autonomous from the current cohort of agents, they are ontologically distinct from people’s current conduct and exert an irreducible influence on current activity (Lewis 2000: 250–3, 257–60; Reed 2001: 214–15).10

The New Economic Sociology

177

On this view, the way in which people interact with one another in the present, and (more specifically) the personal relations into which they enter, are affected not only by the history of their previous encounters but also by the broader macro-structural context provided by these already existing impersonal social structures and the distribution of interests and resources inscribed within them. The problem for Granovetter is that by focusing so intently on people’s personal relations, he diverts attention away from the way in which the material, authoritative and discursive aspects of the impersonal structures inherited from the past shape how micro-structural, personal relations are forged, renewed and break down in the present (Reed 2001: 203).11 The significance of the critical realist critique of Granovetter can be illustrated by an account of how the distribution of authoritative resources or access to positions of power laid down by pre-existing (impersonal) social structures affects the scope for developing personal relations of trust. It is to such an account that the next section of the chapter is devoted.

Trust, reciprocity and power: a critical realist perspective In order to introduce our discussion of the significance of authoritative resources (access to positions of power and authority) we need first to elaborate on the nature of trust. Recall that Granovetter conceptualises trust as a personal social relation that is governed by the principle of reciprocity; that is, the principle that one good turn deserves another. On this view, relations of trust involve a configuration of mutual rights and obligations covering a particular range of activities. (We shall return shortly to the question of what that range of activities actually is.) Significantly, in the case of reciprocity, the rights and obligations which govern the range of activities and exchanges in question are diffuse in the sense that they are not precisely specified in advance. While the principle of reciprocity stipulates that one should return favours, and so establishes the general expectation that the beneficiaries of largesse will respond in kind, the precise nature of the response is left to the discretion of the recipient. The absence of an explicit contract governing the details of the relevant activities and exchanges implies that people have no assurance other than that provided by the (perceived) integrity of their fellows that the rights and obligations in question will be honoured. And the hallmark of situations where relations of trust prevail is that people do indeed confidently expect one another to abstain from opportunism and to discharge their (diffuse, future) obligations in good faith when the time comes for them to do so.12 We can begin to discern the significance of power in this context if we consider the extent to which relations of trust exhibit what has been

178

Paul Lewis

termed a balance of reciprocity (Fox 1974: 76, 98–103, 106). The personal relation of trust between two people (A and B) displays a balance of reciprocity when the activities that A entrusts to the discretion of B are as important to A as the activities which B entrusts to A are to B. If on the other hand A enjoys a good deal of discretion in undertaking those of its activities which are important to B, while B enjoys little discretion in those aspects of its conduct which matter to A, or vice versa, then the relations of trust between A and B are characterised by an imbalance of reciprocity. The importance of power in this context arises from the fact that such imbalances of reciprocity are most likely to arise when the distribution of authoritative resources laid down by inherited, impersonal social structures gives one party the power to exert a good deal of control over the behaviour of the other, while denying the second party countervailing power over the conduct of the first (Hay, 2002: 185). For in such circumstances, people in positions of power are able closely to prescribe the behaviour of their less powerful subordinates and so minimise their discretion in those activities which the powerful deem important for their purposes, while the subordinates lack the resources required to control the activities of the powerful and so are dependent upon the latter’s goodwill in matters about which they care: Management refuses to place itself in trust to the worker, but forces the worker to place himself in trust to management. While specifically bound in what he is required to give, the employee is dependent upon grace and favour for what he hopes to receive. (Fox 1974: 97–8; also see 92–103, 106) Of course, a variety of degrees of imbalance of reciprocity is possible, depending both upon the initial distribution of authoritative resources and also upon the ingenuity of the people in making the most of the resources they have. One can imagine a continuum running from situations where there exists an exact balance of reciprocity to cases of significant imbalance such as that described in the passage from Fox quoted above. In these more extreme cases of imbalance, where one party has little choice but to submit itself to the discretion of the other without receiving any reciprocal rights, the term ‘trust’ is something of a misnomer. For what we have in such cases is a situation in which ‘the word “trust” is being used to describe a relationship from which faith and confidence are conspicuously absent, and which might be more accurately described as distrust’: [O]ne party feels constrained to submit to the discretion of the other but has only a qualified confidence in the outcome; a usage of the phrase ‘We’ve got to trust them’ means in fact: ‘We don’t trust them

The New Economic Sociology

179

but feel constrained to submit to their discretion.’ This simply describes, of course, a power relation. (Fox 1974: 67, 95; also see 66, 117) What this line of reasoning suggests, then, is that what may on superficial observation pass as relations of trust may in fact be no more than a façade of trust, to use Hardy et al.’s (1998) felicitous phrase, which conceals an imbalance of power and therefore of reciprocity in which the dominant partners can dictate the terms on which cooperation takes place (Hardy et al 1998: 65, 70–1, 82–3). Cooperative behaviour that appears to reflect trust may in actual fact be largely a consequence of domination and a lack of viable alternatives for those who ‘have to’ trust those who occupy positions of power: [T]he social embedding of economic activity often involves relations of domination, some of them based on gender, class and race … [F]orms of embedding … are not necessarily the product of a free consensus. They may represent an uneasy compromise between interests which would act differently, given the chance … [W]hat appears to indicate trust and trustworthiness may in fact be largely a consequence of domination or a lack of alternatives. (Sayer 2000: 6–8; also see Fox 1974: 99; Layder 1997: 64–5, 223–6) For critical realists, then, the key point about the distribution of authoritative resources, and the differential access to power to which it gives rise, is that it enables some people to exert a measure of control over the activities of others by imposing low-discretion roles on them, compelling them to pursue the interests and objectives of the powerful, not because they have taken them as their own (as would be the case in relations of ‘genuine’ trust) but because they have little choice in the matter if they want to retain their job. The distinction between cooperation based on genuine trust and cooperation based on a façade of trust is well worth attending to for practical (as well as explanatory) reasons. For while cooperation based on asymmetries of power may be sufficient where the straightforward, mechanical implementation of given tasks and procedures is all that is required for an organisation’s purposes, the goal of harnessing the energy and initiative of workers (or business partners) in order to create new knowledge and novel capabilities is more likely to be achieved successfully if power is widely dispersed and there is a culture of genuine trust within (or between) the relevant organisation(s) (Dodgson 1993; Nonaka and Takeuchi 1995: 73–8, 125–6, 156, 222; Sako: 1998; Hardy et al. 1998: 65–6, 71, 83; Osterloh and Frey 2000: 540, Hardy et al. 2003; Lazaric 2003). In drawing attention to the distinction between cooperation based

180

Paul Lewis

on genuine trust and cooperation based on a mere façade of trust, critical realism provides a useful reminder of an important factor influencing the scope for firms to elicit the full synergistic benefits of cooperation.

Conclusion The current chapter aims to illustrate the benefits that accrue from the sustained, explicit reflection about ontological issues facilitated by critical realism by examining a concept that has become increasingly influential in social studies over the past fifteen or so years, namely that of the ‘embeddedness’ of economic activity. The vantage point provided by critical realism suggests that while the notion of embeddedness is undoubtedly a valuable addition to the conceptual repertoire of social scientists, the most common account of embeddedness (namely that developed by the economic sociologist Mark Granovetter) may lead researchers to ignore features of socio-economic life which are potentially of considerable explanatory and practical importance (cf Fligstein 2002: 71). The arguments advanced in this chapter suggest that by concentrating on personal relations, and by confining his analysis of the legacy of the past to cognitive and emotional factors, Granovetter lapses into a form of idealism which ignores the impact of the (distribution of) non-cognitive factors inscribed in impersonal, historically given social structures. For while the dynamics of people’s personal relations may be an important part of socio-economic life, it is unwise to allow them to monopolise our attention, for a myopic concentration of attention on micro-level personal relations can lead to the neglect of impersonal, macro-structural factors that may be of considerable explanatory importance. In contrast to Granovetter, critical realism offers a metaphysics which includes the category of impersonal social structure and so provides a conceptual framework that is able to capture the impact of material, authoritative and discursive factors and, therefore, the possibility of phenomena like façades of trust. 13 The vantage point provided by critical realism, then, suggests that the notion of embeddedness should be extended beyond the networks of personal relations on which Granovetter concentrates his attention to encompass the impersonal, macro-level structures which facilitate and constrain the construction of interpersonal trust. Moving beyond the micro-level reductionism which characterises Granovetter’s work would help researchers to do justice to the macro-structural, impersonal social structures and associated distribution of resources which channel and shape the interactions through which relations of interpersonal trust and cooperation are constructed (Pratten 1997: 796–7; Reed 2001: 206, 213–23).

The New Economic Sociology

181

Notes 1 2

3

4

5

6 7 8 9

10

For some recent examples, see Coriat and Dosi ([1998] 2002: 98–101), Storper (2000), Boettke and Storr (2002) and Davis (2003). A detailed, though now slightly dated, account of both the Old and the New Economic Sociology is to be found in Swedberg (1987). Swedberg (1991) offers a shorter version. For a brief survey of the New Economic Sociology, see Swedberg (1997). Overviews of recent developments are provided by Carruthers and Uzzi (2000), Swedberg and Granovetter (2001), the essays collected in Guillén et al. (eds 2002) and Swedberg (2003: 32–52). To use a term drawn from the lexicon of critical realism, one might say that Granovetter’s argument suggests that social relations such as trust possess the emergent causal power to react back on people by changing their dispositions and capacities for various forms of behaviour (Sayer 1992: 118–21, 177; Lawson 1997: 63–5, 124, 130–3, 175–7). Granovetter’s critique of the restrictive conception of motivation presupposed by orthodox economic models is not the only argument that has been used by economic sociologists to justify the social-theoretic analysis of economic activity. An alternative rationale for the use of sociological reasoning has been offered by Beckert (1996), who argues that the standard model of economic man as an expected utility-maximiser is inapplicable to the many areas of economic life which are characterised by radical uncertainty. For a critical realist perspective on the latter approach, see Bibow et al. (Forthcoming). Practical interventions in the socio-economic world which assume a priori that instrumental rationality or opportunism are always and everywhere the dominant driving force of human behaviour may have perverse consequences. For example, performance-related pay schemes may actually decrease worker effort if the use of pecuniary incentives designed to appeal to pure self-interest actually offsets or crowds out other sources of motivation deriving from people’s commitment to social norms (Frey 1994; Ghosal and Moran 1996; Rowthorn 1996; Osterloh and Frey 2000; Fehr and Gächter 2000; Fehr and Fischbacher 2002). On the general limitations of functionalist explanations, see Elster (1979: 28–35, 133–7; 1983: 49–68) and Ingham (1996: 251–2). Also see Fligstein and Freeland (1995) and Fligstein (1996: 657–8). For a case study, see Granovetter and McGuire (1998). More specifically, the general category of ‘resources’ includes: material resources such as money and credit; authoritative resources such as access to positions of power and authority, which enable one to secure control over the actions of others; and discursive (cultural and symbolic) resources which enable one to legitimise one’s position and actions in public discourse (Layder 1997: 80–1, 122–3; Thompson 1990: 150–4). Needless to say, critical realism stops short of a reductionist approach which portrays human subjectivity and action as being completely determined by pre-existent circumstances. While the pre-existing pattern of interests might dispose people to act in certain ways, it does not compel them to do so in a deterministic fashion; in many cases it may be such as to leave people with scope to choose from a variety of actions and, of course, people are often free to act against their interests if they wish to do so. Likewise, critical realism fully acknowledges that people may creatively devise new strategies for deploying (historically given) resources so as to achieve their desired goals (Archer 1995: 70, 90–1, 205–13; Porpora 1998: 343–7, 352–3).

182

Paul Lewis

11

The claim that Granovetter’s focus on people’s personal relations leaves him unable to achieve conceptual purchase on the impersonal, material aspects of historically given social structures is reinforced by a closer examination of his analysis of the social construction of institutions. As was noted above, Granovetter relies heavily on Berger and Luckmann’s (1966) phenomenological approach to the construction of social reality. The latter perspective suggests that while people may believe that there exist social structures which are external to their own actions, and which therefore constitute an ontologically distinct stratum of socio-economic reality, the structures in question are in fact no more than the creations of those same individuals, and so are ontologically reducible to people’s current (inter)actions (Granovetter 1990b: 108; Swedberg and Granovetter 2001: 21). The problem with Berger and Luckmann’s approach, critical realists argue, is that it is incapable of doing justice to the full impact of the material aspects of historically given social structures because, as we have seen, the latter are the product of actions undertaken in the past, not the present, and therefore are ontologically irreducible to current agency and intersubjective interaction (Layder 1994: 85–90). 12 For more on the relationship between trust and reciprocity, see Gouldner (1973), Fox (1974: 66–79, 98–102) and de Vos and Wielers (2003). 13 The fact that Granovetter fails to do justice to the macro-structural, impersonal dimensions of social structure is particularly ironic, given his stated aim of resurrecting a distinctly sociological approach to economic issues, because an emphasis on the significance of such macro-social factors was a central feature of the work of classical sociologists (most notably Marx and Durkheim), who were the forebears of the New Economic Sociology (Granovetter 1992b: 4; Swedberg and Granovetter 2001: 10; Ingham 1996: 267; Layder 1997: 9–10, 108, 194; Lie 1997: 351, 354).

References Archer, M.S. (1995) Realist Social Theory: The Morphogenetic Approach, Cambridge: Cambridge University Press. —— (2000) ‘Homo Economicus, Homo Sociologicus and Homo Sentiens’, in M.S. Archer and J.Q. Tritter (eds) Rational Choice Theory: Resisting Colonization, London and New York: Routledge. Beckert, J. (1996) ‘What Is Sociological About Economic Sociology? Uncertainty and the Embeddedness of Economic Action’, Theory and Society, 25: 803–40. Berger, P.L. and T. Luckmann (1966) The Social Construction of Reality: A Treatise in the Sociology of Knowledge, London: Penguin Books. Bibow, J., P.A. Lewis and J.H. Runde (Forthcoming) ‘Uncertainty, Conventional Behaviour and Economic Sociology’, American Journal of Economics and Sociology. Boettke, P.J. and V.H. Storr (2002) ‘Post-Classical Political Economy: Polity, Society and Economy in Weber, Mises and Hayek’, American Journal of Economics and Sociology, 61: 161–91. Carruthers, B.G. and B. Uzzi (2000) ‘Economic Sociology at the New Millennium’, Contemporary Sociology, 29: 486–94. Coriat, B. and G. Dosi ([1998] 2002) ‘The Institutional Embeddedness of Economic Change: An Appraisal of the “Evolutionary” and “Regulationist” Research

The New Economic Sociology

183

Programmes’, in G. Hodgson (ed.) A Modern Reader in Institutional and Evolutionary Economics: Key Concepts, Cheltenham: Edward Elgar. Davis, J.B. (2003) The Theory of the Individual in Economics: Identity and Value, London and New York: Routledge. de Vos, H. and R. Wielers (2003) ‘Calculativeness, Trust and the Reciprocity Complex: Is the Market the Domain of Cynicism?’, in B. Nooteboom and F. Six (eds) The Trust Process in Organisations: Empirical Studies of the Determinants and the Process of Trust Development, Cheltenham: Edward Elgar. Dodgson, M. (1993) ‘Learning, Trust and Technological Collaboration’, Human Relations, 46: 77–95. Elster, J. (1979) Ulysses and the Sirens: Studies in Rationality and Irrationality, Cambridge: Cambridge University Press. —— (1983) Explaining Technical Change, Cambridge: Cambridge University Press. Fehr, E. and U. Fischbacher (2002) ‘Why Social Preferences Matter – The Impact of Non-Selfish Motives on Competition, Cooperation and Incentives’, The Economic Journal, 112: C1–C33. Fehr, E. and S. Gächter (2000) ‘Fairness and Retaliation: The Economics of Reciprocity’, Journal of Economic Perspectives, 14: 159–81. Fligstein, N. (1996) ‘Markets as Politics: A Political-Cultural Approach to Market Institutions’, American Sociological Review, 61: 656–73. —— (2002) ‘Agreements, Disagreements, and Opportunities in the “New Economic Sociology of Markets” ’, in M.F. Guillén, R. Collins, P. England and M. Meyer (eds) The New Economic Sociology: Developments in an Emerging World, New York: Russell Sage Foundation. —— and R. Freeland (1995) ‘Theoretical and Comparative Perspectives on Corporate Organization’, Annual Review of Sociology, 21: 21–43. Foss, N.J. (1994) ‘The Two Coasian Traditions’, Review of Political Economy, 6: 37–61. Fox, A. (1974) Beyond Contract: Power and Trust Relations, London: Faber and Faber Limited. Frey, B. (1994) ‘How Intrinsic Motivation is Crowded out and in’, Rationality and Society, 6: 334–52. Ghosal, S. and P. Moran (1996) ‘Bad for Practice: A Critique of the Transaction Cost Theory’, Academy of Management Review, 21: 13–47. Gouldner, A.W. (1973) ‘The Norm of Reciprocity: A Preliminary Statement’, in A.W. Gouldner, For Sociology: Renewal and Critique in Sociology Today, Harmondsworth: Penguin. Granovetter, M. (1985) ‘Economic Action and Social Structure: The Problem of Embeddedness’, American Journal of Sociology, 91: 481–510. —— (1988) ‘The Sociological and Economic Approaches to Labour Market Analysis: A Social Structural View’, in G. Farkas and P. England (eds) Industries, Firms and Jobs: Sociological and Economic Approaches, New York and London: Plenum Press. —— (1990a) ‘The Old and the New in Economic Sociology: A History and an Agenda’, in R. Friedland and A.F. Robertson (eds) Beyond the Marketplace: Rethinking Economy and Society, New York: Aldine de Gruyter. —— (1990b) ‘Mark Granovetter’, in R. Swedberg, Economics and Sociology, Princeton: Princeton University Press.

184

Paul Lewis

—— (1992a) ‘Problems of Explanation in Economic Sociology’, in N. Nohira and R.G. Eccles (eds) Networks and Organizations: Structure, Form and Action, Boston: Harvard Business School Press. —— (1992b) ‘Economic Institutions as Social Constructions’, Acta Sociologica, 35: 3–11. —— (1993) ‘The Nature of Economic Relationships’, in R. Swedberg (ed.) Explorations in Economic Sociology, New York: Russell Sage Foundation. —— (1994) ‘Business Groups’, in N.J. Smelser and R. Swedberg (eds) The Handbook of Economic Sociology, Princeton: Princeton University Press. —— (1999) ‘Coase Encounters and Formal Models: Taking Gibbons Seriously’, Administrative Science Quarterly, 44: 158–62. —— (2002) ‘A Theoretical Agenda for Economic Sociology’, in M.F. Guillén, R. Collins, P. England and M. Meyer (eds) The New Economic Sociology: Developments in an Emerging World, New York: Russell Sage Foundation. —— and P. McGuire (1998) ‘The Making of an Industry: Electricity in the United States’, in M. Callon (ed.) The Laws of Markets, Oxford: Blackwell Publishers. Guillén, M.F., R. Collins, P. England and M. Meyer (2002) ‘The Revival of Economic Sociology’, in M.F. Guillén, R. Collins, P. England and M. Meyer (eds) The New Economic Sociology: Developments in an Emerging World, New York: Russell Sage Foundation. Hardy, C., N. Phillips and T. Lawrence (1998) ‘Distinguishing Trust and Power in Interorganizational Relations: Forms and Façades of Trust’, in C. Lane and R. Bachmann (eds) Trust Within and Between Organizations: Conceptual Issues and Empirical Applications, Oxford: Oxford University Press. —— (2003) ‘Resources, Knowledge and Influence: The Organizational Effects of Interorganizational Collaboration’, Journal of Management Studies, 40: 321–47. Hay, C. (2002) Political Analysis: A Critical Introduction, Basingstoke: Palgrave. Ingham, G. (1996) ‘Some Recent Changes in the Relationship between Economics and Sociology’, Cambridge Journal of Economics, 20: 243–75. Lawson, T. (1997) Economics and Reality, London and New York: Routledge. Layder, D. (1994) Understanding Social Theory, London: Sage Publications. —— (1997) Modern Social Theory: Key Debates and New Directions, London: UCL Press. —— (1998) ‘The Reality of Social Domains: Implications for Theory and Method’, in T. May and M. Williams (eds) Knowing the Social World, Buckingham: Open University Press. Lazaric, N. (2003) ‘Trust Building inside the “Epistemic Community”: An Investigation with an Empirical Case Study’, in B. Nooteboom and F. Six (eds) The Trust Process in Organisations: Empirical Studies of the Determinants and the Process of Trust Development, Cheltenham: Edward Elgar. Lewis, J.D. and A. Weigert (1985) ‘Trust as a Social Reality’, Social Forces, 63: 967–85. Lewis, P.A. (2000) ‘Realism, Causality and the Problem of Social Structure’, Journal for the Theory of Social Behaviour, 30: 249–68. —— and J.H. Runde (2002) ‘Intersubjectivity in the Socio-Economic World: A Critical Realist Perspective’, in E. Fullbrook (ed.) Intersubjectivity in Economics: Agents and Structures, London and New York: Routledge. Lie, J. (1997) ‘Sociology of Markets’, Annual Review of Sociology, 23: 341–60.

The New Economic Sociology

185

Nonaka, I. and H. Takeuchi (1995) The Knowledge-Creating Company, New York and Oxford: Oxford University Press. Osterloh, M. and B.F. Frey (2000) ‘Motivation, Knowledge Transfer and Organizational Forms’, Organization Science, 11: 538–50. Padgett, J.F. and C.K. Ansell (1993) ‘Robust Action and the Rise of the Medici, 1400–1434’, American Journal of Sociology, 98: 1259–1319. Parsons, T. (1934) ‘Some Reflections on “The Nature and Significance of Economics” [by Lionel Robbins]’, Quarterly Journal of Economics, 58: 511–45. Porpora, D.V. (1998) ‘Four Concepts of Social Structure’, in M.S. Archer, R. Bhaskar, A. Collier, T. Lawson and A. Norrie (eds) Critical Realism: Essential Readings, London and New York: Routledge. Pratten, S. (1997) ‘The Nature of Transaction Cost Economics’, Journal of Economic Issues, 31: 781–803. Reed, M.I. (2001) ‘Organization, Trust and Control: A Realist Analysis’, Organization Studies, 22: 201–28. Rowthorn, R.E. (1996) ‘Ethics and Economics: An Economists’ View’, in J. Groenewegen (ed.) Economics and Ethics?, London and New York: Routledge. Sako, M. (1998) ‘Does Trust Improve Business Performance?’, in C. Lane and R. Bachmann (eds) Trust Within and Between Organizations: Conceptual Issues and Empirical Applications, Oxford: Oxford University Press. Sayer, A. (1992) Method in Social Science: A Realist Approach, London and New York: Routledge. —— (2000) ‘Markets, Embeddedness and Trust: Problems of Polysemy and Idealism’, Department of Sociology, Lancaster University (http://www.comp. lancs.ac.uk/sociology/soc047as.html). Sen, A.K. (1979) ‘Rational Fools: A Critique of the Behavioural Foundations of Economic Theory’, in F.H. Hahn and M. Hollis (eds) Philosophy and Economic Theory, Oxford: Oxford University Press. Smelser, N.J. and R. Swedberg (1994) ‘The Sociological Perspective on the Economy’, in N.J. Smelser and R. Swedberg (eds) The Handbook of Economic Sociology, Princeton: Princeton University Press. Storper, M. (2000) ‘Conventions and Institutions: Rethinking Problems of State Reform, Governance and Policy’, in L. Burlamaqui, A.C. Castro and H.J. Chang (eds) Institutions and the Role of the State, Cheltenham: Edward Elgar. Swedberg, R. (1987) ‘Economic Sociology: Past and Present’, Current Sociology, 35: 1– 221. —— (1991) ‘Major Traditions of Economic Sociology’, Annual Review of Sociology, 17: 251–76. —— (1997) ‘New Economic Sociology: What Has Been Accomplished, What Is Ahead?’, Acta Sociologica, 40: 161–82. —— (2003) Principles of Economic Sociology, Princeton and Oxford: Princeton University Press. —— and M. Granovetter (2001) ‘Introduction to the Second Edition’, in M. Granovetter and R. Swedberg (eds) The Sociology of Economic Life, Second edition, Boulder: Westview Press. (http://www.soc.cornell.edu/research/ swedgranintro.html). Thompson, J.B. (1990) Ideology and Modern Culture, Cambridge: Polity Press.

186

Paul Lewis

Velthuis, O. (1999) ‘The Changing Relationship between Economic Sociology and Institutional Economics: From Talcott Parsons to Mark Granovetter’, American Journal of Economics and Sociology, 58: 629–49. Williamson, O.E. (1985) The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, New York: The Free Press.

10 The really real in economics Scott A. Beaulier and Peter J. Boettke

I Introduction1 What do we mean when we use the term ‘real’ in economics? A standard Webster’s Dictionary has 14 different definitions of real,2 using the term as an adjective, adverb, and a noun. Philosophers seldom provide a clear definition of what they mean when they talk about the real, but they often tell us what the real is not. For philosophers, the real is usually the opposite of the ideal; the real, or existent, is the opposite of the nonexistent; and the real is the opposite of the fake. Economists also weigh in with their notions of real. We can think of many ways in which economists use the term. These include their concern with the realism of assumptions in their theories, the classical dichotomy between real and nominal variables, real business cycle theories, and so on. Philosophers, economists, and the layman all have slightly different notions of what the real means, but there definitely seems to be a degree of similarity among these competing notions. The most coherent definition of realism comes from The Oxford Companion to Philosophy (1995: 746), where realism is defined as ‘the view that we perceive objects whose existence and nature are independent of our perceptions.’ In other words, a realist position is simply an ontological claim that existence exists. For the remainder of our chapter, this will be our working definition of the real. We do not intend to debate the many meanings of the real in this chapter. Instead, we will attempt to make clear the Austrian position on realism. Austrian realism (or lack thereof) has recently been the subject of criticism by critical realists, including Lawson (1997) and Lewis (2004). According to the critics, Austrian subjectivism leads to a nonfoundational ontology where reality is nothing more than the intersubjective meanings of agents in society. Since there is no reality independent from agents’ interpretations, critical realists suggest that Austrian subjectivism can easily lead to a quite problematic, ‘anything goes’ ontological position.

188

Scott A. Beaulier and Peter J. Boettke

In order to make sense of this criticism, we will first trace through several key debates regarding scientific realism in the twentieth century. Engaging in such an exercise serves to remind us of where the debate over realism has been and where the realist movement might be heading. In revisiting realist movements of the past century, we will notice that concern over realism has swung back and forth on a pendulum. Prior to Samuelson, the level of abstraction in economic science should not have raised serious objections. From Samuelson to Arrow–Hahn–Debreu to recent game-theoretic work, economic science took a turn in a decidedly anti-realist direction. This trend has turned in the opposite direction to some extent, as we are witnessing an increased demand for ‘thick’ empiricism in the form of case studies and surveys; the tide has seemingly turned towards greater realism. These issues will be taken up in Section II. In Section III, we will pause to summarize the particular strands of the critical realist movement relevant to our analysis. The works of Lawson and Lewis are works of great importance for economists. Lawson, in particular, has laid out a stern condemnation of most of what economists do. In order to grasp Lawson’s critique of Austrian economics, it is important for the reader also to understand his objections to the mainstream. Though Lawson’s argument is a dense and complicated one, a few pages summarizing his criticisms will have to suffice. Section IV will then turn to a current controversy concerning the commonalities and differences between the critical realist school and Austrian economists. A key figure in this debate has been Peter Boettke. Recently, Boettke has been criticized on the grounds that, while his approach avoids many of the problems which beset mainstream economics, the intersubjectivist socio-economic ontology upon which his approach rests fails to do justice to the full richness of the socio-economic world. According to Lewis (2004: 25): Boettke argue[s] that not all aspects of socio-economic life are reducible to mathematical formulae and quantitative data – to be is not simply to be the value of a variable – so in a similar vein critical realists argue that the material aspects of socio-economic life are irreducible to people’s interpretations thereof. For Lewis, the Austrians3 and the realists ‘part company on the issue of voluntarism and determinism in social theory’ (Lewis 2004: 24). Do the ontological positions of the critical realists and the Austrians really ‘part company’ on the issue of the social-theoretic foundations of economic analysis? Is there any way to reconcile Boettke’s position with that of the critical realists? In sum, is there more common ground between contemporary Austrians and critical realists than Lewis and Lawson realize? After outlining the criticisms, we will try to offer a defense of

The really real in economics 189 Boettke and the Austrians on the issue of ontological realism. Perhaps the ontological position of Boettke and the Austrians is much closer to that of the critical realists than Lawson and Lewis realize. In fact, in the limit, perhaps realist and Austrian ontologies map one-for-one into each other! We will attempt to provide support for this position in Section V. If our argument is supportable, then there are far more extensive gains from trade than Lawson and Lewis realize. Section VI will conclude.

II Realism revisited Philosophy of science in the twentieth century can be parsed into three distinct periods.4 The first period was from the turn of the century until approximately 1960 when antirealists like Mach, Duhem, and Poincaré carried the day. Skeptical analysis in the form of conventionalism and logical positivism was, to a large extent,5 the accepted theoretical approach. In economics, the ‘antirealist’ creed was manifested in Friedman’s (1953) classic ‘The Methodology of Positive Economics.’ In weighing in with his take on the purpose of economic science, Friedman declares that: The ultimate goal of a positive [economic] science is the development of a ‘theory’ or ‘hypothesis’ that yields valid and meaningful (i.e., not truistic) predictions about phenomena not yet observed. Such a theory is … a ‘language’ designed to promote ‘systematic and organized methods of reasoning’ [and] a body of substantive hypotheses designed to abstract essential features of complex reality. (Friedman 1953: 7) Later, Friedman makes his antirealist, instrumentalist position even more explicit. For Friedman, the realism of one’s assumptions is unimportant. The only criterion that matters when we try to distinguish the merit of different theories is the predictive power of competing theories. Friedman’s ‘as if’ argument led to the invasion of antirealism in the form of logical positivism and Popperian falsificationism into economic science. In the early 1960s, however, the philosophy of science turned in the direction of a much more realistic paradigm. As Smart (1963) and Putnam (1975) argued, the theoretical constructs utilized by scientists must actually exist. In addition, scientific theories must be decent approximations of the world. If scientific theories were not decent approximations of reality, then the success of science would have to be considered a miracle. As Stove (1982) would later state in response to Feyerabend’s (1975) antirealist movement, to argue that we cannot know whether knowledge has advanced over the last 500 years is a nonsensical position. For the realists of this generation, science is at least on the right

190

Scott A. Beaulier and Peter J. Boettke

track in bringing us closer to enlightenment. These realists acknowledge the imperfect and downright sloppy nature of science, but – on balance – its accomplishments are essentially indisputable. While a movement in the direction of greater realism was occurring in the general scientific community, economic science in particular did not experience a similar shift. Instead, economics actually moved in the opposite direction towards greater formalism and antirealism. The rise of general equilibrium theory and sophisticated macroeconomic modeling epitomized antirealism. Samuelson’s (1947) Foundations of Economic Analysis was undoubtedly the driving force behind the formalist movement. Until the contributions of Arrow and Debreu (1954), Arrow (1963) and Hahn (1978) on general equilibrium, and work on multiple equilibria (Rubinstein 1979; Fundenberg and Maskin 1986; Schelling 1960, 1978), however, the loss of reality in economics at the hands of formalism was not fully realized. While there remains a strong thrust in the direction of greater formalism in economics, there are some signs suggesting that we are currently witnessing a greater demand for realism in both the general scientific community and economic science in particular. Perhaps we are currently experiencing a unique, third period in the philosophy of science. This period can best be understood as an inflection point in time. The future direction of scientific inquiry seems to be in limbo. Perhaps we will see a move in a decidedly antirealist direction in the future. On the other hand, some signs are pointing to a future with even more realism in both the general scientific community and economic science in particular. There are at least two reasons to be optimistic that we are witnessing a move towards a greater degree of realism in economic science. First, economics has become increasingly suspicious of econometric results (McCloskey 2000). Statistical significance and large cross-sectional studies have lost some of their importance, as economists are beginning to look for ‘thick’ history and field work to support empirical arguments.6 Secondly, the focus of economic science has moved from a preoccupation with abstract, equilibrium theorizing to an extensively empirical discipline. Indeed, empiricism in the form of econometric analysis can also suffer from problems of abstraction and a loss of reality, but economists have greater relevance and a closer attachment to the ‘real’ when engaging in empirical analysis relative to formal theory.

III The critical realists We would like to think that economic science is currently experiencing a natural shift back in a more realistic direction. Whether or not this trend will persist remains an open question. The critical realist projects of Bhaskar ([1975] 1978, 1989), Fleetwood (1995) and Lawson (1997) have emerged out of this scientific context. One may wonder what all of the

The really real in economics 191 fuss is about given that we seem to be encountering a more realistic moment in economic science. This section will examine the critical realists’ position and try to determine whether they have reason to be upset. Lawson (1997) has raised the most penetrating realist objections to orthodox economic theory. While his criticisms are extensive, they can largely be reduced to three fundamental arguments: (1) contemporary econometricians are failing because they continue to follow ‘ad hoc stratagems’ while knowing full well that they are ‘sinning’ (1997: 7); (2) theorists are ultimately denying human choice because they are ‘unable to allow [it] in their formalistic modelling’ (1997: 9); and (3) the proper course for economics to follow is a more holistic, systems-oriented approach – the deductive approach produces closure and is too simplistic an approach for the complexity of the social world (1997: 17–19). We will briefly pause to unpack each of these criticisms in turn. First, let us explore Lawson’s criticism of the economist’s traditional tool for empirical study, econometric analysis. He maintains that: [t]he most telling point against [the] econometrics project is the ex posteriori result that significant invariant event regularities, whether of a probabilistic kind or otherwise, have yet to be uncovered in economics, despite the resources continually allocated to their pursuit. (Lawson 1997: 70) For Lawson, then, ‘econometric failure’ has occurred because econometricians are either implicitly or explicitly assuming that constant relationships between variables exist even though there is no evidence for this assumption in the real world. In other words, econometrics has experienced a flight from reality because the assumed relationships do not directly correspond to the actual relationships we observe in the world. Lawson’s second criticism concerns the loss of human choice in formal models. This criticism is an ontological one, which claims that formalism in economics has led to the loss of understanding of individual action (the ultimate subject matter of the science). The argument against formalism maintains that dynamic market processes, subjective individuals, and real time cannot be reduced to a system of equations or a variable in an econometric model. In addition, formal theory has also resulted in the economist becoming less relevant over time. Since the individual in formal models is often assumption driven to produce whatever result the theorist desires, the theorist is essentially able to say whatever a person wants to hear.

192

Scott A. Beaulier and Peter J. Boettke

The final, and most controversial, criticism that Lawson raises is an objection to the deductive approach. According to Lawson, economists remain bound by Cartesian constraints by continuing to adhere to the deductive method. Many heterodox economists suggest that orthodox theory is too simplistic to cope with the complexity of the real world. Klamer (1994: 52–3) and McCloskey (1983, 1986) have been critical of economists’ continued adherence to a modernist approach. Given that the intellectual and artistic tides have taken a turn away from modernism, they maintain that economics should follow suit. Yet, in spite of this, economists continue to cling to what has worked in the recent past – formalism of a very modernistic form. As Klamer (1994: 53) surmises: The question is, however, whether this vestige of modernism can survive very long, given the doubt that formalist representations will provide certain knowledge about the economy, including knowledge that enables us to effectuate desired economic outcomes. If the arts and other intellectual disciplines are a good example, economists may expect an unraveling of the current hegemonic position of the formalist approach as the ideal in their discipline, and the beginning of a time in which many different approaches will coexist, including historical, institutional, and interpretive approaches. For the purposes of this analysis, we will assume that Lawson is an adequate representative of the critical realist project. We see that the critical realist project has raised some serious challenges to orthodox economic theory. While we are sympathetic to Lawson’s ontological criticisms, we find his anti-deductivist argument less compelling. To state our position briefly: without deductivism one quickly loses tractability in one’s social theory. The burden is on Lawson and other systems theorists to offer a convincing alternative which proves to be a more powerful approach overall than what we currently enjoy under deductivism. When we engage in an analysis of the effect of a minimum wage imposition on employment of unskilled workers, our deductive method offers a powerful and predictive story. If economic science should move away from this deductive approach to a much more open-ended, systems approach, it seems crucial that the value-added question be given better treatment. That is, what is the value-added of an incredibly convoluted economic science? This section has offered a brief summary of the critical realist position vis-à-vis mainstream economists. The purpose of this chapter, however, is not to press on with further discussion regarding the specific objections Lawson raises against orthodox economic theory. Instead, we would like to turn our attention to a more specific critique that Lawson and other critical realists have raised against Austrian economists. At first glance, it seems as though the Austrian approach could succeed in offering a more

The really real in economics 193 ontological economic science. Yet the critical realists have been fairly adamant in arguing that there remains significant scope for improvement in the Austrian alternative. The next section will discuss their specific objections to the Austrian ontology.

IV The critical realists as critics In addition to the criticisms raised against the mainstream’s methodological approach, Lawson has been an ardent critic of the Austrian methodological paradigm. In focusing his critique on Hayek, Lawson (1997: 138) views the Austrian social ontology as nothing more than a ‘hermeneuticised version’ of an ‘empiricist theory of ontology founded in positivism.’ In other words, Hayek is ultimately a positivist economist with a slightly more subjective individual actor. At best, Hayek is more sophisticated than the orthodox economist, for he at least has a social ontology underlying his analysis. However, this ontology is one that ‘exist[s] apart from [individuals] being conceptualized in action’ (Lawson 1997: 139). For Lawson (and Fleetwood), Hayek’s social ontology is not nearly deep enough to warrant a genuine alternative to the orthodoxy. In the same spirit as Lawson, Fleetwood (1995: 5) also regards most of Hayek’s work7 as ‘lacking a notion of social structure in general and social rules of conduct in particular.’ Fleetwood (p. 6) also attributes Hayek’s lack of social structure to subjectivism when he writes that, ‘[Hayek’s] subjective idealist epistemology … manifests itself in his social science as hermeneutic foundationalism, that is, the view that reality is exhausted by agents’ conceptions.’ According to the critical realists, then, Hayek’s notion of the ‘really real’ does not exist independently of individual interpretations. Fleetwood (1995: 76) regards the problem with this position to be the following: Under hermeneutic foundationalism there is no notion of social structures that exist and act to a significant extent independently of agents’ identification or classification of them. On the contrary, social structures are reducible to and exist solely in virtue of the conceptions held by agents. They do not act as a (material) cause at all. There is, for Hayek II then, no domain of the deep that can sustain an elaboration of the conditions for action perceived at the level of the empirical. Hayek is thus impaled on a fork of his own making. The theoretical devices he needs for the substantive economics are ruled out of bounds by the philosophical position he adopts. In retrospect, it is quite clear that if he is to develop the institution of social rules, he has to abandon this untenable philosophical position.

194

Scott A. Beaulier and Peter J. Boettke

More recently, Lewis (2004) has raised similar criticisms against contemporary Austrians. Lewis maintains that while there is common ground among Austrian economists and critical realists on their critique of formalism, the two schools of thought part ways on issues of social ontology. Lewis follows Lawson’s critique, but focuses his attention primarily on Boettke’s (rather than Hayek’s) ontological position. He sees Boettke and other contemporary Austrians falling into the same trap as Hayek, for their ontology is a non-foundationalist one. The reason the Austrian ontology is non-foundationalist is because it is derived from the subjective interpretations of individuals in the social world. According to Lewis’s (and Lawson’s) interpretation of Austrian ontology, whenever individual interpretations of the world change, the underlying reality also changes. Hence, the Austrians have no theory of social structure, relationships, and power independent of interpretation.

V A humble response to the critics The critical realists’ primary objection to the Austrian alternative stems from the Austrians’ supposed lack of an ontological realist grounding. We find this criticism unwarranted, and we will try to explain why in this section. While an appreciation for ontological realism might not be as apparent in Hayek’s early writings (1942a, 1942b, 1943, 1944),8 a sturdy commitment to ontological realism is evident in his later writings (1960, 1967, 1973).9 In addition, a commitment to a rich social ontology also seems present in the more recent Austrian writings of both Kirzner (1982, 1992) and Boettke (1996, 1997, 1998).10 In summarizing the ontological position of the Austrians, Boettke (1996: 30–2) writes as follows: The Austrian claim is not against abstraction, but it does challenge the VHOHFWLYH UHDOLVP of modern information economics and moreover it questions whether the techniques employed in modern information economics undermine the analyst’s ability to deal with questions of knowledge and the informational role of the price system (see Thomsen 1991). As such, modern Austrian economic theory offers an alternative theory of economic life; one significantly more general in its applicability to the real world … If what we demand out of an economic theory is realism, then Austrian economics strives to deal seriously with the real social conundrum in which human actors are placed. If economic theory is also supposed to strive for universality, then Austrian economics claims to offer logical derivations of analytical propositions that meet that aspiration to universal principle of human action and interaction. By consistently and unflinchingly pursuing methodological individualism (understood in its

The really real in economics 195 phenomenological as opposed to atomistic sense), methodological subjectivism (or intersubjectivism), and process analysis, Austrian economists have developed a general theory of human action, of market processes, and of institutional evolution. By recognizing the complex problem situation which is our social world, the individual and collective coping mechanisms that allow us to live with one another are highlighted and explored. Boettke and other Austrian economists view their particular approach as a realist one, which focuses on man in a social world as he is, ‘with all his weaknesses and limitations’ (Mises 1949). Runde (2001: 7) has argued that, ‘contrary to what Lawson appears to suggest, Hayek does seem to be committed to a differentiated and largely stable reality that exists independently of how we perceive and think about it.’ In taking up Lawson’s claim that Hayek’s early work has a distinctly ‘hermeneuticised’ social reality, Runde reaches the opposite conclusion. That is, according to Runde, even the early Hayek might have actually been a critical realist! As Runde writes (2001: 11; emphasis added): Hayek regards his social structures as enduring, as defining relations between individuals, and as having an existence over and above the subjective beliefs and attitudes of the particular individuals who form the foci in the network of relations defined. On the face of it, then, it looks as if a case could well be made for maintaining that Hayek is proposing a conception of social structure that is intransitive in the sense of CR [critical realism]. Runde (2001: 12) goes so far as to suggest that Lawson completely misses a fundamental point of Hayek’s research program. For Hayek, the shared concepts and interpretations of individuals are learned. In order to have learned interpretations, Hayek must be presupposing a social structure with all of the important features of power, status, and diversity – the very features of reality which Lawson claims Hayek’s social ontology lacks. As Runde suggests, Austrians have proposed an underlying reality, which has been the product of power struggles, individual innovation, and a long evolutionary process. For Austrians, this underlying reality is not just suspended in animation and subject to change whenever individual interpretations change. Indeed, individual interpretations can provide feedback and alter the underlying reality of the social world. However, the ‘really real’ that Austrians have in mind exists independently of the minds of agents in the social world. As can be observed in Beaulier and Boettke (2000: 550), Austrians follow Warren Samuels on the issue of social ontology. They subscribe to an ontological position in which reality is ‘not entirely the product of some mysterious

196

Scott A. Beaulier and Peter J. Boettke

fate working its course, but instead the product of a multitude of decisions [and power structures].’ Austrians do, indeed, view the individual as the primary agent of change in the social nexus. Yet, Austrians are fully aware of the role noneconomic institutions and culture can play when individuals are acting and interacting in the social world. The Austrian appreciation for cultural factors does not lead to the position that social change is due to larger social forces; we cannot imagine institutional change independent of individual action. However, the individuals acting in the social world are deeply affected by the noneconomic institutions of power, culture, and history surrounding them. We believe this position is identical to the critical realist position which Lewis (2004: 13) summarizes: a satisfactory understanding of the relationship between social structures (such as the social institutions just mentioned) and human agency requires that it be conceptualized as an inherently ‘tensed’ process: at any given moment in time, social structures pre-date and so condition current human agency, which agency subsequently leads either to the reproduction or the transformation of those structures. Agents are not just the carriers of past social structures; they are also living in a social world of rights and power relations – the social structure embodied in their individuality is meeting the present social world head-on.11 If our interpretation of the Austrian ontology is correct, then Austrians seem to hold a decidedly non-hermeneuticised ontological position regarding the social world. Like the critical realists, Austrians have a much more humanistic take on individuals. Individuals are constrained and often governed by their roots – both Austrians and critical realists appreciate this. In addition to sharing common ground on the issue of embeddedness, Austrians and critical realists both maintain that a social world exists independently of individual interpretations. Yet, if Austrian economists and critical realists are essentially in agreement on issues of social ontology, why is there so much disagreement between the two schools of thought on the issue of ontology? We believe the primary source of confusion on the issue of ontology comes from the Austrian adherence to methodological individualism. As consistent methodological individualists, Austrians claim that the only way to understand the social world is through examination of individual interpretations. Interpretations are not the only factor determining the nature of the ‘really real’, but they are the only way of understanding the underlying reality. Like the critical realists, Austrians understand that individuals bring both material and historical baggage into the social nexus. These factors matter for Austrian economists, but

The really real in economics 197 the only way of sorting out how these factors affect the world is by concentrating on individual actions. This is all the ‘hermeneuticised’ Austrian ontology is saying. We have tried to provide evidence suggesting that the Austrian methodological position is one that is ontologically realist. One further point should be made regarding Austrian economists vis-à-vis the critical realists. The realist Austrian ontology, which we consider central to Austrian economics, requires a contextualist epistemological position. Ontological realism tells us that there is an objective, real, underlying world ‘out there.’ However, this social world is not an easy one to understand – it is hidden rather than objectively before us. The hidden nature of meaning in the social world thereby requires a contextualist position. Through a contextualist approach, perhaps we can get a slice of knowledge about the social world by examining the ways diverse individuals live within it. Another way to understand this position is to break the world into three dimensions. There is the ‘really real’ social world we are all trying to understand at the deepest and most hidden level. Then there are all of the individuals with their unique interpretations of what this underlying reality is really like. Finally, there are the social scientists trying to understand both the underlying social world and individual behavior within specific contexts. Some social scientists will focus exclusively on individual behavior without showing any interest in what this tells us about the broader social world. Others will skip past the intermediary step of individual experience to try to get directly at understanding the social world. Economics, when done correctly, should not be engaging in either of these forms of analysis. Instead, the method of economics should be one in which reality is understood through the medium of interpreting individual actions. We must rely on the medium of individual actions and interpretations to understand our social world. Hayek (1952: 44) best summarizes our epistemological contextualist position when he writes, ‘[s]o far as human actions are concerned the things are what the acting people think they are.’ This chapter is suggesting that Austrian methodology is on firm ground. Reality and objectivity are not just whatever comes out of the diverse and subjective interpretations of all the individuals of a community. By contrast, reality exists independently of these interpretations. A defining tenet of the Austrian position is its methodological individualism. All this tenet is saying is that the only way to understand reality is through examination of individual behavior and interpretations in the social world. This position does not translate into reality is whatever interpretations say it is; reality is made sense of through interpretation, but reality is far more than the product of individual interpretations.

198

Scott A. Beaulier and Peter J. Boettke

VI Conclusion Lawson and Lewis have both pointed out common ground between Austrian economists and critical realists. Lawson (1998: 143) finds common ground among the ‘broader features of perspective (an emphasis on uncertainty, historical processes, etc.).’ According to Lawson, however, Austrians cannot produce a fruitful alternative until the philosophical foundation of critical realism is put in place. This chapter has maintained that Austrians actually endorse and practice the kind of critical realist ontology which Lawson and Lewis are demanding. Perhaps there has been some ambiguity brought on by the Austrian emphasis on individual interpretation. However, when we look carefully at what is underlying these individual interpretations, we find a rich social structure. Austrians view individuals as embedded in a much deeper social process than the kind of crude ‘voluntarism’ depicted by Lawson and others.12 Some of the blame for this confusion falls on Austrian economists. The writings of Hayek, among others, have proven obscure in reflecting the depth of Austrian social ontology. We have attempted to reconcile these shortcomings by directing more attention to the realist ontological strand which permeates Hayek’s work as well as the work of several contemporary Austrian economists. When faced with the question of whether there is common ground between Austrian economics and critical realism, it is not an issue of one school of thought needing to ‘come around’ in their ontological or epistemological position. Instead, as we have maintained, good Austrian economics is good critical realism. Similarly, we hope that good critical realism will continue to be understood as good Austrian economics in the future.

Notes 1 2

We would like to thank Paul Lewis, David Prychitko, and Mario Rizzo for useful comments and suggestions throughout this project. The standard disclaimer applies. real, adj. 1. true; not merely ostensible, nominal, or apparent: the real reason for an act. 2. actual rather than imaginary, ideal, or fictitious: real events; a story taken from real life. 3. being actually such; not merely so-called: a real victory. 4. genuine; authentic: real pearls. 5. unfeigned or sincere: real sympathy. 6. Informal. absolute; complete; utter: She’s a real brain. 7. Philos. a. existent as opposed to non-existent. b. actual as opposed to possible or potential. 8. (of wages, income, or money) measured in purchasing power rather than in nominal value. 9. noting an optical image formed by the actual convergence of rays, as the image produced in a camera (opposed to virtual). 10. Law. of or pertaining to immovable or permanent things, as lands or buildings. 11. Math. of, pertaining to, or having the value of a real number. – adv. 12. Informal. very or extremely: You did a real nice job. – n. 13. the real, a. something that actually

The really real in economics 199

3

4 5 6 7 8 9 10

11

12

exists. b. reality in general. – Idiom. 14. for real, a. in reality; actually. b. genuine; sincere. According to Lewis’s interpretation of the Austrian school, Boettke’s position on many of these issues often translates into the Austrian position. Whether most Austrians would agree with many of Boettke’s methodological positions is an extremely relevant question regarding Lewis’s criticisms. For the purposes of this chapter, we will follow Lewis in assuming that Boettke’s methodological positions are reflective of the Austrian school. While this might seem like an overly simplistic perspective, it is a relatively adequate approximation of twentieth-century history that has some support in Fine (1986). It should be noted, however, that the work of Planck and Einstein in statistical analysis and relativity theory did turn the tide more in the direction of realism. This trend is not only evident in the heterodox writings of McCloskey, but is also appreciated by leading neoclassical economists (e.g. Bates et al. 1998). We must keep in mind Fleetwood’s Hayek I, Hayek II, and Hayek III distinction. According to Fleetwood, Hayek III’s work was much closer to the critical realist approach. For an interpretation of Hayek’s early writings as a realist exercise, see Runde (2001) and the discussion of Runde that follows in this essay. Fleetwood (1995) follows Lawson in maintaining that Hayek from 1937 to 1960 is a positivist, but suggests that Hayek post-1960 is a ‘quasitranscendental realist.’ Boettke (1998: 180) expresses his deep sympathies for an economic science grounded in realism when he writes that ‘the promise of richer explanations is worth the loss of crystalline precision, since the precision that has been achieved falsifies reality. If we continue to insist that precision is always more valuable than relevance, then I conclude with Heilbroner that “the worldly philosophy, which began with such rich promise, seems unlikely to become the source of illumination that was once its singular gift.”’ Unlike Lawson, Lewis maintains that Austrians and critical realists share a great deal of common ground on the issue of social ontology. Whereas Lawson criticizes Hayek for his lack of a deep ontology, Lewis acknowledges that Austrians recognize the existence of intersubjectively shared social structures that are ontologically distinct from individual agents’ actions. Lewis argues, however, that in focusing on intersubjective meanings, Austrians fail to do justice to the material or transsubjective dimension of the social structures within which people are situated. See Horwitz (2001: 34). Horwitz attempts to defend Hayek against the atomistic individual charge when he writes: ‘[Hodgson] is arguing against a straw man. No defender of the market, surely not Hayek at least, ever argued that every single form of human interaction should be mediated by the market.’ It should be noted, however, that among Austrians Lavoie (1994) addressed the Austrian tendency towards atomism. Like Lawson, he recognized that methodological individualism and subjectivism could breed atomism, which would ‘locate the domain of meaning in isolated individual minds’ (1994: 57).

References Arrow, K. (1963) Social Choice and Individual Values, second edition, New York: John Wiley.

200

Scott A. Beaulier and Peter J. Boettke

—— and G. Debreu (1954) ‘Existence of Equilibrium for a Competitive Economy’, Econometrica, 22: 265–90. Bates, R., A. Grief, M. Levi, J.-L. Rosenthal and B.R. Weingast (1998) Analytical Narratives, Princeton, NJ: Princeton University Press. Beaulier, S. and P. Boettke (2000) ‘Of Norms, Rules, and Markets: A Comment on Samuels’, Journal des Economistes et des Etudes Humaines, 10: 547–52. Bhaskar, R. ([1975] 1978) A Realist Theory of Science, Hemel Hempstead: Harvester Press. —— (1989) Reclaiming Reality, London: Verso. Boettke, P. (1996) ‘What Is Wrong with Neoclassical Economics (and What Is Still Wrong with Austrian Economics)?’, in F. Foldvary (ed.) Beyond Neoclassical Economics, Aldershot: Edward Elgar Publishing. —— (1997) ‘Where Did Economics Go Wrong?’, Critical Review, 11: 11–64. —— (1998) ‘Formalism and Contemporary Economics: A Reply to Hausman, Heilbroner, and Mayer’, Critical Review, 12: 173–86. Feyerabend, P.K. (1975) Against Method: Outline of an Anarchistic Theory of Knowledge, London: New Left Books. Fine, A. (1986) The Shaky Game, Chicago: University of Chicago Press. Fleetwood, S. (1995) Hayek’s Political Economy: The Socio-economics of Order, London and New York: Routledge. Friedman, M. (1953) Essays in Positive Economics, Chicago: University of Chicago Press. Fundenberg, D. and E. Maskin (1986) ‘The Folk Theorem in Repeated Games with Discounting or with Repeated Information’, Econometrica, 52: 533–54. Hahn, F.H. (1984) Equilibrium and Macroeconomics, Oxford: Basil Blackwell. Hayek, F. (1942a) ‘Scientism and the Study of Society’, Economica, 9: 267–91 —— (1942b) ‘The Facts of the Social Sciences’, in F. Hayek, Individualism and Economic Order, London: Routledge. —— (1943) ‘Scientism and the Study of Society, Part II’, Economica, 10: 34–63. —— (1944) ‘Scientism and the Study of Society, Part III’, Economica, 11: 27–39. –—— (1952) The Counter-revolution of Science: Studies in the Abuse of Reason, Glencoe, IL: The Free Press. —— (1960) The Constitution of Liberty, London: Routledge and Kegan Paul. —— (1967) Studies in Philosophy, Politics, and Economics, London: Routledge and Kegan Paul. —— (1973) ‘Rules and Order’, in F. Hayek, Law, Legislation, and Liberty, London: Routledge. Horwitz, S. (2001) ‘The Functions of the Family in the Great Society’, Working Paper, St Lawrence University. Kirzner, I. (1982) Method, Process, and Austrian Economics, New York: Lexington Books. —— (1992) The Meaning of Market Process: Essays in the Development of Modern Austrian Economics, London: Routledge. Klamer, A. (1994) ‘Formalism in Twentieth-century Economics’, in P. Boettke (ed.) The Elgar Companion to Austrian Economics, Cheltenham, UK, and Northampton, MA: Edward Elgar. Lavoie, D. (1994) ‘The Interpretive Turn’, in P. Boettke (ed.) The Elgar Companion to Austrian Economics, Cheltenham, UK, and Northampton, MA: Edward Elgar. Lawson, T. (1997) Economics and Reality, London and New York: Routledge.

The really real in economics 201 —— (1998) ‘The Nature of Post Keynesianism’, in D. Prychitko (ed.) Why Economists Disagree, Albany, NY: State University of New York Press. Lewis, P.A. (2004) ‘Boettke, the Austrian School and the Reclamation of Reality in Modern Economics’, forthcoming in the Review of Austrian Economics. McCloskey, D. (1983) ‘The Rhetoric of Economics’, Journal of Economic Literature, XXI: 481–517. —— (1986) The Rhetoric of Economics, Brighton: Wheatsheaf Books. —— (2000) How to Be Human, Ann Arbor: The University of Michigan Press. Mises, L. von (1949) Human Action: A Treatise on Economics, New Haven: Yale University Press. Putnam, H. (1975) Mathematics, Matter, and Method, Volume 1, Cambridge: Cambridge University Press. Rubinstein, A. (1979) ‘Equilibrium in Supergames with the Overtaking Criterion’, Journal of Economic Theory, 21: 1–9. Runde, J. (2001) ‘Bringing Social Structure back into Economics: On Critical Realism and Hayek’s Scientism Essay’, Review of Austrian Economics, 14: 5–24. Samuelson, P.A. (1947) Foundations of Economic Analysis, Cambridge, MA: Harvard University Press. Schelling, T.C. (1960) The Strategy of Conflict, Cambridge, MA: Harvard University Press. —— (1978) Micromotives and Macrobehaviour, New York: W.W. Norton Smart, J.J.C. (1963) Philosophy and Scientific Realism, London: Routledge and Kegan Paul. Stove, D. (1982) Popper and After: Four Modern Irrationalists, New York: Pergamon Press. Thomsen, E.F. (1991) Prices and Knowledge: A Market Process Perspective, London and New York: Routledge.

11 Addressing the critical and the real in critical realism Ben Fine

I Introduction1 It is with some anxiety that I agreed to write this chapter. Methodology does not come readily to (political) economists, and knowledge around the topic tends to be confronted and digested unsystematically. Providing some sort of order to the chaos often proves an endless task as one point of reference leads to another in tracing an analytical journey through previously uncharted waters or casual stopovers. A second misgiving in accepting the brief follows from the, questionable, principle that my enemy’s enemy is my friend. I knew that my contribution would at least in part be critical of critical realism but I hope that I have been constructive for both those who remain committed to it as well as those who share my unease. I also suspect that, without any authority other than numbers of casual conversations, my views are at least partly representative in more polished form of the stances towards critical realism taken by political economists of a similar ilk to myself. This has eased my own lingering pangs of conscience, together with an occasional personal indulgence in style and content that I hope does not detract from the finished product. The chapter is organised around two main sections together with this introduction and some concluding remarks. Sections II and III argue, respectively, that critical realism is neither critical nor realist enough. In brief, it is insufficiently critical through divorcing methodology from theory, and criticising methodology alone. This is not only methodologically questionable – no method without theory, as well as vice versa – but also strategically deficient as it is not the most effective way to go about the goal of undermining mainstream economics and ‘underlabouring’ for a more appropriate alternative. Critical realism is also insufficiently realist in failing to situate itself in relation to key economic concepts, most notably those of capital and capitalism. This is surprising given its ontological emphasis, for how can it be conscious of its own origins and meanings in the absence of such central categories?

Addressing the critical and the real

203

Section III closes with reference to the work of Paul David, especially his promotion of the notion of path dependence, an idea apparently welcome to critical realism for acknowledging that history matters and is ‘open’. In practice, however, David’s work has converged upon the mainstream economics that critical realism deplores. Why critical realism should have been guilty of a misreading of David’s work is taken up in the concluding remarks together with suggestions on how such misinterpretations might be avoided in the future.2

II Is it critical enough? For nearly thirty years as an academic economist I have attempted both to refute mainstream economics and to provide alternatives, primarily based on Marxist political economy. With a small margin to spare, my endeavours preceded the emergence of Bhaskar’s realist theory of science and, with a larger margin, the corresponding critical realist project addressed at economics. Moreover, the critical response to the mainstream arguably, if necessarily implicitly, long pre-dates the latter’s own birth given the alternative approaches that continue to be derived from classical and Marx’s political economy. Certainly, then, methodological and theoretical critique of the mainstream has not waited upon the emergence of critical realism. Further, despite claims to the contrary, critical realism is not prominent in critical responses to the mainstream. It has had some notable impact in discussion of economic methodology and in the soul-searching that now surrounds Post Keynesianism as the latter’s survival is threatened despite its considerable merits relative to the mainstream.3 But it is important to recognise that these are marginal both to mainstream economics and also to much of the criticism that it has attracted. In short, whatever its analytical merits, critical realism in practice remains limited in its impact even amongst those, such as myself, to whom it might be thought to be most attractive. Through the extraordinary energy of a small number of its most committed exponents, it has gained prominence both in absolute and relative terms, given the general decline in economic heterodoxy. Yet, its overall significance should not be exaggerated. I do not doubt that the assessment contained in the previous paragraph is open to dispute and could benefit from some sort of quantification or other evidence. This is not my intent. Rather, that is to make clear the position from which I began to assess critical realism systematically for the first time in response to the request to do so for this volume. As a dissident economist, I have long been aware of both critical realism and its origins, through both written and spoken word. But it had not seemed sufficiently important to my own work, nor to that of my opponents, to engage with it in its own right. Given my own professed dedication to undermining the mainstream, this must appear most paradoxical. For the

204

Ben Fine

slightest acquaintance with critical realism reveals that its proponents consider it the only secure basis on which to reject the mainstream and to construct alternatives.4 Characterised as deductivist, the mainstream can only be overthrown if this is mercilessly exposed as a matter of methodology. Note, though, that it is a huge and unjustified step from the notion, right or wrong, that ‘critical realist analysis … of this sort does seem to be a strategic requirement for a more relevant approach to be at all understood and accepted in the modern environment’ (Lawson 1999b: 219–20), to the conclusion that ‘if I am correct in concluding that the characteristic feature of the mainstream project in economics is its (deductivist) method, then a philosophical critique is required if that project is to be actively undermined. And given the current dominance of that project, it does seem necessary that its untenability be revealed if alternative paths are even to be seriously contemplated.’ Many political economists would, in view of their own efforts and sacrifices, reasonably take offence at such privileging of the strategic if not the logical status of such a stance. In writing this piece, I was determined to probe the gap between the claims of critical realism and my own indifference to it. Let me begin, though, by emphasising that there is much within critical realism that I admire and with which I agree. First, it demands that (economic) analysis be open and not pre-determined, not least as a result of humans making their own past, present and future rather than these being ground out by some deterministic or stochastic model. Second, analysis is to be based on a systemic understanding, incorporating a social and historical content, and including an account of tendencies, structures and relations, as well as their contingent context.5 Third, as a result, there is no immediate translation of theory into corresponding empirical observation or statistical analysis – in the same way that the law of gravity does not predict that all objects must be falling. Fourth, it follows that much of the content of mainstream economic theory is rendered redundant, not least its particular form of methodological individualism, and its emphasis on equilibrium as an organising concept. On the basis of such genuinely important insights, critical realism characterises mainstream economics as being predominantly deductivist at its core. I want to raise but not explore the question of whether this is, indeed, the main methodological feature of mainstream economics. An alternative, possible refinement, is to emphasise the eclecticism of mainstream methodology. This itself derives from the mainstream’s neglect and ignorance of methodological issues so that there is a sense in which economics does not consciously have a methodology. As such, it is chaotic and even infantile relative to the understanding to be found in other disciplines, social or otherwise. To some extent, differences over the methodology of economics will depend upon what economics and which economists are being considered, and there is a danger of circularity – lack

Addressing the critical and the real

205

of deductivism means exclusion from the mainstream. No doubt, the deductivist hypothesis is applicable to vanguard mathematical modelling, most of whose practitioners would be left floundering if confronted with the basic vocabulary in which methodology is discussed, let alone the substantive issues involved. But does it apply to the use of supply and demand in business, management and marketing, which often, as academic disciplines, equally deploy both economics and a range of more or less sophisticated methodologies that have no place within mainstream economics? As mentioned, at least for the moment I merely wish to flag the issue of what is the core methodology for mainstream economics. For the purposes of what now follows, I am prepared to accept the hypothesis that the mainstream is deductivist and, hence, fails the valid requirements of critical realism for openness. So what? Well, quite clearly, mainstream economics collapses in a heap as a result of its own inadequacies, and critical realism provides the basis for developing an alternative. Yet, nothing of the sort has happened. Mainstream economics remains as strong as ever and has become in both absolute and relative terms more, not less, savage in its methodological naivety and barbarism, and intolerant of more sophisticated and well-grounded ‘heterodoxies’. I use the term ‘barbarism’ advisedly. By chance, my own research as a political economist on (the meaning of) consumption has led me to develop a macabre interest in cannibalism because of the particularly acute issues it raises for the sources and cultural content of what we (or others) eat. The request to address my thoughts on critical realism came just as I was reading a remarkable book, not so much on cannibalism as on how it has been represented, by Lestringant (1997). I do not wish to lampoon but the image of critical realists as missionaries in the cooking pot is almost inescapable. Convinced of the virtues of their own methodology, they continue to preach for the conversion of those who are about to destroy them in the most ghastly fashion. I suspect that such lessons are not lost on political economists at Cambridge, the beleaguered home of critical realism, but such barbarism is evident everywhere, as has been documented, for example, by Hodgson and Rothman (1999) and Lee and Harley (1998) in the making of publications and appointments, respectively.6 Of course, it could be argued that the missionaries in the real world have had the last laugh, and the savages have been civilised. If so, it has been as the handmaidens of capitalism, rather than as prime movers, so the analogy should not be pushed too hard without taking full account of extraneous factors. Yet by virtue of Lewis’s (1999) generative metaphor, Lestringant’s account of how cannibals have been perceived does allow for some insight into the postures adopted towards the mainstream by critical realism. Lestringant identifies four, loosely historically sequenced, responses to cannibals. The first is to view them as sub-human savages,

206

Ben Fine

no doubt rationalising their slaughter. Unfortunately, within economics, the savages have the upper hand and are perpetrating the slaughter of heterodox economists, let alone pious students of methodology. The second is to understand cannibals as driven by necessity for survival in a Malthusian world of overpopulation relative to resources and by analogy with acute circumstances such as the shipwrecked or marooned. With historical hindsight, it is possible to understand such compulsion by natural necessity to be fundamentally flawed and, at most, socially created. By analogy, freedom from the systemic coercion attaching economics and economists to the mainstream is only a possibility in other circumstances. For, as Lawson (1997a: 281) argues in a remarkable passage that is both modest in its claims and yet unduly ambitious in raising them:7 Social emancipation, then, is an objective which critical realism, unlike positivistic orthodox economics, is able to embrace. Even so, an acceptance of critical realism is obviously nothing like sufficient for any progressive change actually to take place. A much less grand goal than social emancipation is that of combating mainstream economics, and there is the question of how effective critical realism is in allowing for progressive change (to be discussed below). A third stance towards cannibalism identified by Lestringant is based on the idea of the noble savage, eating enemies as a symbolic act of revenge, even with distaste for the meal itself. I suppose mainstream economics could see itself as exercising vengeance upon its fellow social sciences for having departed from scientific methods. Further, from the perspective of critical realism, there is hope for reform given the methodological advances in (physical) science upon which economics so shakily and falsely bases itself and from which critical realism itself originated by way of Bhaskar’s realist theory of science. Yet, heterodox economists know only too well that the mainstream’s appeal to science is merely a rhetorical device for leaving methodological issues unexamined whilst claiming kudos from the hard disciplines. Finally, cannibalism has been understood as a universal, as opposed to a sub-, human condition, and only kept under wraps by social conditioning, without which degradation results as in the case of de Sade. By the same token, deductivism is less a savage than a natural (if false) response to the world. Consequently, critical realism, and realism more generally, must be diligent in exposing and correcting aberrations. A striking example, possibly also illustrative of the previous approach, is the relationship between critical realism and Post Keynesianism. The latter is perceived as acceptable as long as it displays an appropriate openness as opposed to embracing deductivist affinities to the mainstream.

Addressing the critical and the real

207

Thus, at the level of methodology, not of substantive theory, Post Keynesianism is to be won over to critical realism. Such, in part, is the inspiration for the more general metaphor of critical realism as missionary. For it does display an extraordinary zeal for seeking converts on the basis of its methodological scriptures, otherwise serving as a critical realist police or inquisition.8 In this respect, it is highly catholic, some might say too liberal, blessing the new feminist economics, evolutionary theory, certain path dependence literature, Austrian, Marxian and (some) Post Keynesian approaches. Yet, though taking the mainstream as its point of departure, it shares three features with it, albeit as mirror images, which are liable to blunt its critical edge. First, mainstream economics in practice proceeds as if theory is everything and methodology is nothing. On the basis of the most superficial understanding, and as a seal of approval, it deploys the words scientific and rigorous, primarily in the deductivist fashion exposed by critical realism. Essentially, critical realism accepts this division between methodology and theory. The major difference is that its emphasis is upon methodology rather than theory. It sees itself as undertaking the ‘underlabour’ for economics, breaking off from methodology at the point where theory begins. However, is this dichotomy legitimate? For the time being, it is worth observing that there must be doubts over it for reasons that, ironically, appear to be intrinsic to critical realism. On the one hand, at a general level, as critical realism is historically and socially open, and perceives itself as potentially fallible, one would expect the relationship between methodology and theory to be variable and not fixed, depending upon circumstance. On the other hand, in particular, as critical realism is part of a realist discourse of wider scope, its specific application to economics might be thought to go beyond methodological imperatives and address its substantive theoretical content. Otherwise, in what respect is it distinctively about economics (on which see next point but one)? Second, as a corollary to the previous point, and in parallel with the mainstream, critical realism also exercises its own ‘policing’. Where the mainstream rejects thought and thinkers that do not conform to its own, inadequate and erroneous, notions of rigour and science, critical realism sets about delineating theory according to its own prescriptions and proscriptions. Only in its hands are methodological as opposed to theoretical instruments of tort. There are considerable dangers of reductionism and misinterpretation in such procedures. By analogy, say, with a functionalist police, all theory is reduced to, and interpreted through, a functionalist prism in order to reject it for functionalist crimes. Yet, the theory in the dock may not have been constructed nor solely be legitimately interpreted as such. If the critique, as for critical realism, remains at the level of methodology, it cannot engage with substantive theory on its own terms, a particularly important omission where theory

208

Ben Fine

is itself, as in mainstream economics, particularly carefree about methodology. In addition, at the level of methodology itself, critical realism insists that the mainstream is deductivist at core, irreducibly and solely vulnerable on this score, and thereby adopts a strategy of knocking back deductivism. The result is both to neglect much mainstream that lies outside this core and to open up lines of defence against critical realism that it misinterprets the mainstream (see, for example, Hands 1999). Whilst the extent to which the mainstream is indeed deductivist in practice is both an empirical and an analytical question (the latter especially in view of its methodological naivety and eclecticism), it is also inescapably strategic in its significance. Critical realism not only has to convince the mainstream of its deductivism, but it also then has to deploy this insight to undermine its analytical practices. Unfortunately, there is liable to be many a slip twixt the logic of critical realism, however valid, and the outcome in terms of persuasion. It is notable how little critical realism gets past first base even with, in part, sympathetic and wellinformed economic methodologists. For Hausman (1998: 205):9 My point is rather that the issues that concern Lawson are obfuscated by labelling them as questions concerning realism … what divides Lawson from those who disagree with him … is not realism – all the parties to the controversies are realists. What is at issue are the details of Lawson’s views of economic mechanisms. Third, as a further corollary of the first point, critical realism holds to an ambiguous position on the legitimacy of the economy as a separate domain of study, a stance that has been characteristic of the vast majority of mainstream economics from the marginalist revolution until recently.10 At first blush, there is no ambiguity about the status of economics as a separate (social) science. For Lawson (1997b: 5), ‘there are few legitimate grounds for regarding economics as a separate science’. Significantly, however, this contribution effectively post-dates others, not least the synthetic Lawson (1997a), and appears to have addressed ex post the issue of economics as a separate social science. As a result, there is little prominence of the negative stance in earlier work. In addition, quite apart from the qualifying ‘few’ as opposed to a definitive ‘no’ in the quotation given above, the conclusion is itself heavily qualified in other ways. For example: Economics must take its place as a branch of social science, much like high energy physics … constitutes a branch of physics. (Lawson 1997a: 31)

Addressing the critical and the real

209

I do not wish to quibble over the distinction between a social science and a branch of a social science but this does seem to be a case of having cake and eating it. Further, it is subsequently argued of economics that: Its raison d’être is not a separate domain of distinct phenomena with their own properties but a particular aspect of all social life. In consequence, the modern (forced) separation of the discipline of economics from other social sciences must be recognised as quite misguided. Indeed, this separation merely makes it difficult for economics to advance in pace with other branches of social science, and must at least be a contributing factor to the discipline’s continuing tradition of adopting methods of formalistic modelling that are hardly relevant to social scientific understanding and illumination at all. (Lawson 1997a: 31) I suspect that all but the most introverted mainstream economists would agree, the final qualification apart, and Lawson himself cites Lionel Robbins in support. Further, a similar position is essential to current attempts, discussed below, by mainstream economists to colonise the other social sciences, precisely by methods of formalistic modelling. Economic imperialism has no difficulty with the idea that economics is not a separate social science since it incorporates all (legitimate) social science (see Lazear 2000 and Fine 2000b for a critique)! Of course, it is unfair to accuse critical realism of parallels with mainstream economics simply because they share certain conclusions. It is reasonable, however, to assess critically how critical realism has reached its conclusions. Lawson (1997b: 23) proceeds by seeking a synthesis of what economics is by combining and refining the definitions of Mill, Marshall and Robbins, ultimately arriving at the following formulation: Economics is the identification and study of the factors, and in particular social relations, governing those aspects of human action most closely connected to the production, distribution and use of the material conditions of well being, along with the assessment of alternative really possible scenarios. Just because this definition is a synthesis of what Lawson has observed have usually been taken as alternative and opposed positions does not render it acceptable. Nor do two other arguments that follow as corollaries from his definition: the economic, as embedded social phenomena, does not constitute an object of study that forms a separate social science; and all human activity has an irreducible economic content:

210

Ben Fine There is no obvious basis for distinguishing economics according to the nature of its object, i.e. as a separate science. Nor does it have its own domain. I cannot think of a single sphere of human activity – from lending support to a football team, to listening to music, or even making love – that does not (or could not) have an economic aspect (given our interpretation of the economic as set out above). These and all other activities take place in space and time, both of which can have alternative uses. All human activities require material conditions and are, for example, influenced by property relations, the length of the structure of the working ‘day’ or ‘week’, opportunities forgone, and so on. But at the same time very few activities, if any, have merely an economic aspect, even including those of labouring on the shop floor or in an office or doing the shopping. Activities such as the latter can involve wearing the colours of, and so signalling support for, a particular football team, or, say, being on the lookout for an interesting companion or just being part of a crowd/community, amongst many other things. (Lawson 1997b: 31)

Gary Becker could easily have penned such a passage.11 The weakness in the account, as for Becker, derives from its transcendental foundations. For Lawson, for the economic, and hence economics, to have legitimacy as a separate social science, it must be set apart as an object of study across all history and activity. Yet, it is precisely capitalism that establishes the economy as a distinct set of practices, structures and tendencies although, if often overlooked in a grand form of commodity fetishism, these are irreducibly embedded in the social. With the economy as a separate sphere in reality as well as for analysis, the distinction between political economy and economics reflects whether the broader context is recognised and appropriately addressed. Significantly, after all else, Lawson (1997b: 32) does acknowledge, if only to reject, the argument for a separate sphere for economics by virtue of its attachment to an ‘exchange economy’. In doing so, he first mentions capitalism, the neglect of such a core concept being taken up more generally in the next section. In the context of the scope of economics as a putative discipline, Lawson’s motive in raising the term is purely polemical, arguing that economics would have to apply both to it and to socialism. In doing so, he implicitly draws upon the naive and hackneyed conclusion from the market socialism debate that neoclassical economics is the economics of socialism (1997b: 32–3).12 In short, the methodology of critical realism, and its confinement to methodology, prevent it from addressing the issue of the legitimacy of the economy as a separate entity for study according to the historical content of the economic and its relation to the social. Nor is it surprising, in this light, that it makes little progress in undertaking the task in practice!

Addressing the critical and the real

211

A number of issues are involved here, and it is important not to confuse them. Arguably, the economy is constituted as a separate entity, albeit embedded within society more generally. How it is embedded, to what extent and with what effects are historically and socially variable, with the emergence of the market and especially capitalism as decisive. No doubt, such developments lead to economics being constituted as a separate academic discipline and object of ideological discourse. From this, it does not follow that the (realist) methodological critique of economics should be discipline-specific. Indeed, one would expect it to be part and parcel of a more general methodological approach to the social sciences as a whole. In a sense, this is recognised by critical realism by containing its critique to the methodological and in a way that poses alternatives that are not specific to economics. Yet, I suspect that the vast majority of that critique would, especially in the light of postmodernism but by no means dependent upon it, be considered well known if controversial in social sciences other than economics. In short, and shifting tack slightly, it is hardly surprising that the realism project should flourish in science and economics, the two broad areas in which there tends to be a denial of the connection between the objects and subjects of study and society. Consequently, a methodological critique is most appropriate but it is hardly dramatic in its content or leverage from the perspective of other social theory (of which political economy should be an irreducible part). I can draw this section together and to a close not by addressing critical realism as being right or wrong but by simply asking whether it is critical enough. I take this to be both substantive as well as strategic in view of the strength and nature of the stranglehold of mainstream economics. There are several reasons why it is not critical enough. First, methodology from the cooking pot serves as a source for two appropriate, if entirely different, metaphors. On the one hand, the unthinking eclecticism of the mainstream is often self-confessedly derived from the methodological cookbook. On the other, from its own missionary position in the cooking pot, critical realism is right but, by its own admission, mainstream economics has practically no interest whatsoever in methodology or, it might be added, realism or the real world.13 Consequently, winning the methodological argument has no purchase whatsoever unless there be virtue alone in being right.14 In general, the mainstream simply does not engage with its methodological critics. When it does, it tends to reflect idiosyncrasies and eccentricities, with limited impact by way of repercussion and concession.15 It is highly symbolic for a range of reasons that critical realists should take Frank Hahn as a representative of the mainstream. For, as soon as he enters their world, quite apart from basking in the role of local anti-hero, he departs his own for a bit of fun at the expense of the lefties and in order to demonstrate how he is scholarly rounded. Most in his other world do not bother. Further, his exhortation for young economists not to engage with methodology is a reflection of its

212

Ben Fine

poverty within economics, and lack of intellectual integrity more generally, but also represents the most sound advice if they are to achieve any success given the publish or perish mentality of academia and the current standards of the profession. Second, the view that economists cannot necessarily be shifted by methodological arguments is supported by their failure to be shifted by logical arguments. This is well illustrated by the Cambridge critique of capital theory. Even though the use of models and empirical work, based on the as if one sector world and corresponding production function, has been shown to be invalid, they have continued to be used and have even flourished over the last decade with the emergence of the new growth theory (as if the Cambridge critique never existed).16 This is not to suggest that the Cambridge critique had no impact on mainstream economics or the emergence of, and commitment to, alternatives. Rather, it suggests that there can be no presumption that a methodological critique, despite its analytical generality, is liable to be more effective than other forms of criticism. These might be primarily logical, as with the Cambridge critique, or based on an appeal for the application of more realism to the subject matter, as in criticism of common assumptions around given preferences, optimisation based upon them, given technology and endowments, equilibrium as an organising concept, etc. Two further points need to be made here. One is that it can never be certain what critique will be effective and in what direction, if any, it will lead to change, not least because of the prevailing dynamic or logic of the discipline at the time quite apart from extraneous factors such as student radicalism in the 1970s, for example. The other is to acknowledge that mainstream economists construct theory and evidence in order to address particular real processes, such as competition, the grinding out of prices and incomes, and the allocation of resources. Their efforts should be interpreted in this light and not simply or even primarily as methodological deficiencies, not least when attempting to convince them to change their views, including their methodology. To reiterate, if mainstream economics is primarily methodologically innocent, then, conversions on the road to Damascus apart, theoretical, logical, empirical or other factors are just as, or more, likely to be effective in persuading a change in approach as the pushing of alternative methodologies. The thrust of the previous two paragraphs is that critical realism is not critical enough because it draws a sharp division between methodology and theory and confines itself to the former, a recipe for ineffectiveness in combating mainstream economics. Third, there is no case for setting these arguments aside in view of the strength of methodological critique, especially when the latter is derived from a critical realism that is marked by theoretical aloofness. For, as I have argued elsewhere (Fine 2001a: chapter 1), mainstream academic economics has become so homogeneous

Addressing the critical and the real

213

and intolerant of alternatives that it is increasingly impossible to engage fruitfully with it as the means of establishing an alternative. Further, I have also argued that currently there is a ‘revolution’ going on in or, more exactly, around economics.17 On the basis of the new informationtheoretic economics, with the capacity to offer an explanation for the social, the customary, the structural, the institutional, etc. – that is the noneconomic – economics is seeking to colonise the social sciences as never before, and not simply through Becker-type extension of utility maximisation to non-market circumstances. It does so through the perfection rather than rejection of methodological individualism. It also allows for, even positively embraces, path dependence and the marriage of formal with informal analysis, however chaotically, as in the various ‘new’ branches of the discipline addressing the household, crime, institutions, politics, development, and so on. As a result, especially as other social sciences are retreating from the extremes of postmodernism in an attempt to return to the real, the most fertile location within which alternative economics can prosper is currently to be found outside mainstream economics. It is located within other social sciences as they both respond to a colonising economics and, potentially, seek to offer an alternative. It is with reluctance that I suggest that economics has become a no-go area for political economy to the extent that it is increasingly futile to offer a critique, let alone an alternative. Yet, the debate over economics within the other social sciences offers huge opportunities and remains open, not least because of the traditional if not universal hostility to the methodology of mainstream economics across the other social sciences, the exception being where rational choice theory is present. Nonetheless, the other social sciences remain intimidated by the formalism and technical requirements of mainstream economics and, particularly in the wake of postmodernism, subject to an understandable retreat from its reductionism (and deductivism) into a preoccupation with cultural or social context and content at the expense of the economic. This might preserve (part of) their discipline from the encroachment of a colonising economics but is unhelpful in promoting an appropriate political economy both in and of itself and as an essential part in understanding the social and cultural themselves.18 In this light, the impact of critical realism on the fortunes of political economy could even be counterproductive. For, at the time when other social sciences need to be acutely conscious of the theoretical deficiencies of a colonising economics, they are liable to be presented with a methodological critique that is both unexceptional from their perspectives (and, it might be added, of the vast majority of radical political economy) and empty of alternative supportive content from traditional political economy except by way of unfulfilled potential.

214

Ben Fine

III Is it realist enough? In short, critical realism seems to be unaware of developments within and around economics that are potentially powerful levers in criticising the mainstream and in moving towards alternatives. This is because it has confined itself to methodology at the expense of substantive theory. Tracking the latter over recent years would have allowed the shifting meaning of the economic to economists to have been detected as well as their putative designs upon other disciplines. As previously mooted, this may reflect a methodological as well as a strategic deficiency on the part of critical realism, although my emphasis so far has been upon the latter. Now, though, I want to raise questions over the methodological content of critical realism by approaching it from a different angle, whether it is realist as opposed to critical enough, although the two are closely connected. The rationale for doing so is the stunningly glaring absence of key concepts in the discourse of critical realism, especially from the perspective of political economy, and especially in view of its goal of providing a critique of a mainstream economics that is itself so highly dependent on asocial and ahistorical concepts. What I particularly have in mind is the lack of any attempt to address capitalism and capital. This might be excused by the (questionable) separation between the methodology of economics and its theoretical substance, as a result of which critical realist discourse floats free from historical specificity even of the most general type. But, following the discussion of the previous section, it is very difficult to accept that a methodology of economics can be of much significance unless it engages with that mode of production or period of history that has most obviously given birth to it. Of course, economies and economists pre-date capitalism, and the latter does not exhaust all economic life today. Yet, there can surely be little doubt that an appropriate understanding of capitalism and of capital must play an important role not only in our understanding of contemporary economies but also those of the past. This proposition might be supported by critical realism itself with its emphasis on the importance of comparative study across societies and time. But what is the most important theoretical standard by which comparisons can be not so much made as constructed if it is not through an understanding of the most developed forms assumed by the economy? How otherwise are we to grasp the methodology that is appropriate to profits, commodities, money, markets, accumulation, crises and so on? It is not my intention, at least here, to foist a particular understanding of capital and capitalism upon the reader, one I would derive from Marx. It is merely to reinforce the suggestion that economic methodology is not readily, or usefully beyond a point, separated from theoretical substance, and that substance desperately needs to confront the socially and historically specific realities attached to the nature of capital and

Addressing the critical and the real

215

capitalism. Indeed, it has always been my view that the ahistorical and asocial, generally physicalist, notions of capital deployed by the mainstream are one of its greatest and inseparable methodological and theoretical weaknesses. One of the consequences of the lacuna in incorporating an understanding of capital in its economic methodology is the previously mentioned theoretical eclecticism. Essentially, all approaches that are interpreted as ‘open’ receive the critical realism stamp of approval even though these may be equally deficient in their understanding of capital and capitalism (and not least neo-Austrian thought that is so painfully inadequate in these respects). But there are also at least three areas in which the failure to relate economic methodology to the otherwise unavoidable and brutal facts of capitalism as the dominant factor in our everyday life leads to a lack of realism in critical realism. First, as previously mentioned, due deference is acknowledged to the necessity of a methodology to incorporate structure, power, conflict and tendencies. Yet these are notably underdeveloped in substance,19 and it is far from clear that they are consistently used and distinguished from one another.20 There is a sense in which this demonstrates intellectual integrity for to give these concepts substance almost inevitably requires their content to be historically and socially limited. Otherwise, critics of critical realism, such as Hausman (1998: 211), can cheekily get away with suggesting that mainstream economics deploys causal tendency or force on a par with alternatives, simply by virtue of a proposition such as agents preferring larger to smaller bundles. The germane issue is whether we can have an abstract theory of social structures, relations or of tendencies independent of the types of mode of production or social formations to which they are attached. Such language is deliberately chosen here to be reminiscent of the failed project of the Althusserian school that, in the hands of Balibar in particular, sought to construct an abstract theory of modes of production out of generalised constituent elements, such as forces and relations of production and separate spheres of economic, political and ideological practice, all wedded to relative autonomy and determination in the last instance. Thankfully, critical realism falls far short of such methodology with its use of Lego-type constructs, but only because its methodology falls short. It is not guilty of the associated functionalisms, structuralisms and idealisms but nor does it provide substance to its Lego parts for it cannot do so in the absence of its interaction with some theory, of which a stance on capital and capitalism would appear to be the most appropriate starting point. There are a number of aspects of the failure to move such methodological considerations forward. One is their being rooted in the realist theory of science, developed by Bhaskar (1975), whose appendix to his Chapter 3 on natural tendencies is entirely formalistic and

216

Ben Fine

transhistorical in content. Even a sympathetic overview of Bhaskar, such as that provided by Collier (1994), is sceptical on this score. For, whilst Collier does accept that Bhaskar distinguishes social from natural structures on the basis of the former being historically specific, nonenduring and governing laws across modes of production, he observes that even physical laws and structures are not universal – as before and after the big bang (Collier 1994: 243–4). Thus, not only the laws themselves but also the nature of the laws are physically and historically specific – not everything is like gravity, a favoured expository law for Lawson – necessarily drawing for economics, say, upon methodology and theory for which capital(ism) is an appropriate focal point.21 In addition, the methodological status of notions such as tendency and structure is rendered ambiguous. It would appear, for example, for critical realism that, until open human agency breaches them, structures condition laws or tendencies. But surely it is underlying tendencies that reproduce and are causally prior to structures? In the most general Marxist terms, there is a need to address the interaction between the forces and relations of production. This can only be done at the level of particular modes of production, carefully distinguishing between levels of abstraction (logical ordering of concepts in relation to concreteness and complexity), causation (or determination) and elaboration (or exposition). Thus, for example, the law of the tendency of the rate of profit to fall and its counteracting tendencies, appropriately interpreted, is not an empirical prediction as in many, usually dismissive, accounts in the wake of Okishio. It is a theory of underlying forces arising out of capital accumulation in the definite economic and social relations attached to capitalism (Fine 1989). Whilst presumably acceptable to critical realism as an example of open theory, both the law and the nature of the law are specific to capitalism. Is there a corresponding form of the law for feudalism?22 Ironically, a second aspect of lack of realism within critical realism, and more in practice than logically, lies in its understanding of the need for theory to be open, one of the most powerful elements in its critical armoury. For, here, it appears to be heavily influenced by its taking mainstream economics as its point of departure. Correctly, it sees the orthodoxy as unduly deterministic, particularly in view of axiomatic modelling, empirical estimation as synonymous with explanation, and undue belief in the associated practices of prediction and forecasting. There is a heavy bias, induced by mainstream practices, towards addressing openness in terms of an appropriate understanding of future uncertainty in the richest sense of the term. Given the human capacity to choose and change, to discover and innovate, the future is both unknowable and not pre-determined. But, by the same token, to deploy the cliché, both the past and the present, as well as the future, are uncertain worlds. Even though they have already happened or are

Addressing the critical and the real

217

happening, how they are understood is socially constructed not least in the conceptual apparatus by which they are addressed. Of course, such an insight is bread and butter to the other social sciences, not least in the wake of postmodernism, and is compatible with critical realism in principle since it self-consciously locates the theorist socially. Yet, the principle does not seem to be borne out in practice. For it is hard to conceive how an economic methodology can be self-consciously socially constructed without an explicit account of the contemporaneous economy and its dependence on capital and capitalism. One does not need to degenerate into the subjectivism and relativism of the extreme forms of postmodernism to accept that an economic methodology requires a critical deconstruction and reconstruction of the substantive concepts attached to the economy. As observed, these are notably absent from critical realism, thereby justifiably limiting its appeal to political economists and other social scientists, not to mention letting mainstream economists off the hook for the naturalisation and universalisation of their invalid concepts such as capital and labour. Inevitably, this aspect of critical realism is liable to reinforce the belief of non-economists that political economy and mainstream economics are much of a muchness and equally to be rejected as insufficiently conceptually critical. Third, as a particularly acute illustration of the last point, is the limited incorporation of history as evidenced by the favourable stance towards path dependence displayed by Lawson (1997a).23 This is seen, especially in the work of Paul David, to lie within the critical realist framework once it progresses to substantive analysis (Lawson 1997a: 248). Indeed, his work and the issue of path dependence is privileged in being chosen as the first such illustration:24 This ongoing, often explicitly methodological project appears to be the one that most straightforwardly overlaps with my own at the level of suggesting how social science might actually proceed … Certainly it helps to illustrate aspects of social reproduction that I take to be fundamental. (Lawson 1997a: 249) This assessment of David’s work is questionable, even from within the framework of critical realism itself.25 The alternative interpretation offered is not necessarily incompatible with critical realism. But it raises and reinforces the doubts about it that have already been expressed, and these will be taken up in the concluding remarks. In a nutshell, David’s (1975) early work appeared both dissident and ‘open’ when pitted against old versions of the then newly emerging economic history, or cliometrics, in which institutions did not matter and the world was interpreted as an as-if (deviation from) perfect competition.26 Now, in the wake of information-theoretic developments

218

Ben Fine

within and around economics, the new and improved economic history accepts that history and institutions, etc., matter. They do so in a way that merely draws upon more sophisticated, stochastic but otherwise deterministic models, with complex dynamics, multiple equilibria and systemic interdependencies. As such, David has been incorporated, possibly surpassed, and is far from being an unwilling victim.27 Thus, for him, it is a matter of linking path dependence to mathematical modelling with the result that: The core notion of path dependence has been seen to be concerned with stochastic processes of historical change – the probabilistic motions of systems through unidirectional time. (David 1997) A preferred technique is that of Markov chain analysis, for example, relying upon probabilistic movement within and between overlapping local networks, with illustrative application to probabilistic interdependence between preferences, which gives rise to different equilibria depending on the strength of demonstration effects between consumers. Significantly, in an earlier paper considering the adoption of new technology, David and Foray (1994) rely upon the Pólya urn model with random selection of coloured balls from an urn with overreplacement that, only if sufficiently strong, will lead to convergence to an urn population of single-colour balls. From there, the analysis moves through Markov chain models to ‘percolation’ to allow diffusion to be understood in terms of both connectivity and receptivity between agents. What is notably absent from David’s approach is the conscious use of history itself as the source of critical concepts as opposed to the ex post imposition of models and path dependence upon history. Indeed, particularly worrying from a critical realist perspective is the treatment of knowledge. Whilst, intriguingly, ‘context – temporal, spatial, cultural and social – becomes an important consideration in any discussion of codified knowledge’, such considerations remain shallow. For, ‘as economists, and not epistemologists, we are substantively more interested in knowledge transactions or activities … rather than where knowledge of different sorts may be said to reside’ (Cowan et al. 1999). In short, it is possible to report to the critical realist police that David’s treatment of knowledge is limited from an epistemological point of view; it certainly does not move beyond this to ontological considerations.28 No doubt, critical realism would or could accept these insights and eschew David as an exemplar. Yet the point remains that its own lack of realism, in failing to root its methodology in contemporary capitalism and engage with its categories, has prompted its initial unduly favourable stance.

Addressing the critical and the real

219

IV Concluding remarks The previous section closed with a brief exposé of David’s work and has hinted at the path dependence of the notion of path dependence itself as economic history is colonised by the new and improved cliometrics. Does this, however, constitute a criticism of critical realism? It is possible to anticipate the following response. Fair enough, David does not get past the critical realism police and thanks to local citizens for having shopped him. Indeed, Lawson’s (1997a: 253) praise for David is heavily qualified, especially where a marriage is detected with a deductivist mainstream economics.29 So perhaps Lawson was right but, possibly understandably, a little behind the times. I would not find this acceptable and I am also concerned that, in a mirror image of a Papal bull, critical realism will always prove fallible in principle but infallible in practice. I will review my earlier criticisms in light of the more concrete differences that are revealed by consideration of path dependence. First, the distinction between methodology and theory or substantive analysis cannot be sustained. How can we address the issue of path dependence, and its implications for the nature of history, for example, without simultaneously taking them together? My suspicion is that an appropriate critique and constructive alternative would depend more on an explicit theoretical than a methodological stand. Indeed, Lawson (1997a: 248–9), quoted above, accepts that path dependence is in part methodological. Second, path dependence, whether for David himself or not, has taken on different meanings over the past thirty years. It has to be situated currently relative to economics and economic history, and the colonisation of the one by the other, whereas previously it meant something different (with openings for more heterodox though informal interpretations). Third, then, from a logical point of view, it is necessary to engage with substantive economic theory both because of the nature of the object of study as well as because of its shifting meaning. In short, it is simply unacceptable for critical realism to reject the criticism that the (theoretical) proof of the pudding is in the eating, not because that is liable to be the nature of proof but because the eating and the pudding are inextricably attached to one another.30 Fourth, however, exactly the same applies for strategic reasons. Of course, it is a moot point whether David, for example, should now be considered as an ally or not against mainstream deductivism. Previously, he stood outside the mainstream; now he can be interpreted as having been colonised or even serving as a coloniser. I would not crudely draw the conclusion that he could previously have been used as a lever against the mainstream and now cannot, although it is my suspicion that such a shift in emphasis is appropriate. For the answer depends to a large extent on the stance of those economic and social historians who have not

220

Ben Fine

already been captured and who may remain open to some sort of nonreductionist political economy. Yet, how to underlabour and with what depends upon methodological, theoretical, substantive analytical and strategic choices that go far beyond the scope of critical realism. Fifth, the combination of these methodological, theoretical, substantive and analytical weaknesses is reinforced by, and reflected in, the limited realism of critical realism. This is a result of its tending to confine historical and social content to openness. This is illustrated by the favour shown for path dependence, an entirely universal term without social and historical content. Self-conscious reference to the key and judiciously chosen categories of contemporary society would allow for greater realism, for the revelation of their absence in mainstream economics, and for some purchase amongst social scientists and political economists for whom other elements of critical realism are liable to prove attractive. In short, and generalising beyond path dependence in view of earlier arguments, critical realism needs to wed its considerable methodological strengths to:31 • • •

•

An explicit confrontation with, and critique of, economic theory The construction of a political economy rooted in the categories of contemporary capitalism A careful exposition of the meanings of structure, relation and tendency, etc., their interrelationships and their historical and social scope and variability An explicit account of where it diverges from Marxism and Marxist political economy.

I am, in a fit of self-parody, acutely aware that my demands are that critical realism be led along the righteous path towards the Marxism to which I subscribe. Hopefully, such Jesuit overtones in my contribution will intrigue, challenge and inform those who remain committed to virtuous and legitimate scholarly endeavours whether in the name of critical realism or otherwise.

Notes 1

2

3

This chapter was written whilst in receipt of a Research Fellowship from the UK Economic and Social Research Council (ESRC) under award number R000271046 to study ‘The New Revolution in Economics and Its Impact upon Social Sciences’. Thanks to many commentators on an earlier draft for what have often been extensive and substantive suggestions. I am acutely aware that when I refer to critical realism, I am predominantly addressing its leading representative, Tony Lawson. In addition, informalities apart, my knowledge of critical realism comes primarily from Lawson (1997a), Fleetwood (1999), other references cited and various contributions to the Journal of Post Keynesian Economics. See King (2000).

Addressing the critical and the real

221

4 See also Lawson (1999b: 232–3) for a simplistic understanding of the strategic. In response to Hands’ (1999) question of why economics is the way it is, a precondition for understanding how to change it, he responds that it is first necessary to convince others it is this way. This parallels the idea that we have to have a convincing case that no gods exist before we can combat particular instances of religious persecution. Note also that Lawson’s answer to the question is that there is a set of factors ‘which engender the greater experience of surprise, doubt, and general curiosity on the part of the individuals involved’ (1999b: 235). Surely the opposite is the case for economists! 5 See Lawson (1999c). 6 See also Coats (1996) for ‘Americanisation’ and Fine (2001a: chapter 1) for a fuller discussion. 7 The passage is followed by a highly idealist notion (in both philosophical and utopian senses) of what would be necessary for change – shared human objectives, self-awareness, achievable goals and commitment to truth, freedom, justice and explanatory critique, to act in pursuit of real interests and needs. 8 Lawson’s (1999d: 279) response to Hausman (1998) would appear to justify the policing metaphor: It is suggested that I offer ‘economists a false dichotomy. Either they can accept a view of science as exclusively the search for exceptionless regularities … or they can accept critical realism’ … Now what is false about this dichotomy? … the choice, the dichotomy, is indeed between science being or not being ‘exclusively the search for exceptionless regularities among observable events’, between reductionist claims of deductivism and non-reductionist claims of critical realism. Surely this smacks of critical realist reductionism to methodological differences along one dimension alone? 9 Further: Lawson has embedded his central thesis – that economics should be a search for the structures and mechanisms that generate the typically irregular data that economists gather – in a controversial metaphysics, which, I suggest, distracts readers from his main concerns. (Hausman 1998: 204) And, in finishing his piece: The difficulties involved in interpreting the principles of economics are aggravated when the issues are recast as questions about realism … I believe that a realist view of causes is needed to make sense of economics, but the issues that interest economic methodologists – concerning the ‘principles’ of economics – are largely orthogonal to questions about realism and could be more clearly addressed if they were not entangled with questions about realism. (Hausman 1998: 211–12) An earlier draft was written whilst I was unaware of the subsequent exchange between Lawson (1999d) and Hausman (1999). As is or will be seen, this reinforces the conclusions drawn here. On the immediate point, see Hausman’s (1999) closing paragraph in his rejoinder. 10 In a first draft I argued that critical realism appeared to accept the notion of economics as a distinct social science. I was guided to the contrary as argued in Lawson (1997b). But, as the following discussion suggests, the initial stance is not entirely inappropriate. 11 Especially in view of his recent adoption of social capital, see Becker (1996) and Fine (2001a: chapter 3). 12 See Milonakis (2000).

222

Ben Fine

13 See Blaug (1998a and b) for example. 14 Note that Lestringant points to the exquisite pain that could be experienced by missionaries, the more so as they were subject to the most horrific torture. 15 See Fine (1999b) in debate with Thompson (1997 and 1999). 16 See Harcourt (1972 and 1976), Fine (1980) and Hodgson (1997). 17 See Fine (1997, 1998a and b, 1999a and b, 2000a–f, and 2001a and b) for the general argument and for specific case studies. 18 As I have argued to historians in the context of consumption, a topic that is (un)remarkably untouched by a colonising economics, despite its popularity across the other social sciences, because of its inability to deal with the social construction of meaning (Fine 2000c). 19 Thus, in response to Hausman, Lawson (1999d: 271) makes much of the failings of economics ‘in any meaningful, systematic or sustained fashion to go beyond appearances to the (equally real – but not necessarily wholly apparent) underlying structures, powers, mechanisms and tendencies that generate or condition the surface phenomena of reality’ (also see Lawson 1999d: 274). Yet, once again, no further progress is made in this respect. 20 They are also controversial. Consider the (uncontroversial) proposition, posited as belonging to critical realism, that money is a social relation (Ingham 1999). As the debate between Fine and Lapavitsas (2000) and Zelizer (2000) indicates, what sort of social relation involves both methodological and theoretical differences? 21 Once again in response to Hausman (1998), Lawson (1999d: 280) makes explicit his stance on this issue: Hausman further asks: ‘What is gained by assimilating questions concerning the status of, for example, social norms (presumably amongst other social structures) to questions concerning the existence of electrons?’ … But as I have already indicated there is no assimilating going on; these are just the same sort of questions. 22 Joseph (1998: 102–3) purports to provide a defence of critical realism as an underlabour for Marxism and, in doing so, reveals very clearly how critical realism depends upon the transhistorical deployment of economic, social and methodological categories. Thus, in conclusion: An economic analysis must therefore be able to express the structuration and stratification of the social world and the processes involved in its reproduction and/or conservation/transformation. Through its insistence on the transformational model of social activity, critical realism points [to] the fact that it is necessary to intervene into structural processes, hence giving a key role to the concept of hegemony. And so on in similar ahistorical and asocial vein. By contrast, Roberts (1999: 43) strengthens arguments offered here at a more general level (aloof from capital). He observes that the methodological prescriptions of critical realism ‘also abet an investigation into the real world which loses contact with the sensuous nature of history’, and provides a sixteen-fold set of distinctions between critical realism and Marxism. He also suggests within critical realism a conflation between levels of abstraction (determination for him) and causation. 23 See also Lawson (1999a: 15). 24 I had expected to deal in this chapter with the second illustration – the role of workplace resistance in obstructing productivity growth in the UK economy. I am pleased not to have to come back to this (Fine and Harris 1985), although the negative reception to it is reported to have inspired the drift to critical realism (Lawson 1999c). The work of Kilpatrick and Lawson (1980) is, however, fundamentally flawed both theoretically and empirically as, I

Addressing the critical and the real

25

26

27 28

29

30

31

223

believe, has been shown by its critics. It does neither critical realism nor political economy any service to misrepresent criticisms and to fail to answer them, as is the case with Lawson (1997a). Initially, a detailed critique of David formed a separate section of this chapter but had to be omitted for lack of space. It now forms part of a more general critique of the new and improved or newer economic history, cliometrics, in the wake of information-theoretic developments within economics for which see Fine (2000d). As in economic history as measurement and interpretation of total factor productivity. Interestingly, in view of her subsequent conversion to rhetoric, McCloskey (1971) was a prominent contributor. But see Fine (1990) for a critique of her study of entrepreneurship in US and UK Victorian coal industries. Emphasis is placed on the open consequences of conditions of access to landed property so that the latter cannot be reduced to a factor input (of natural availability and quality of reserves). Not least from a Cambridge political economy perspective, in view of his collaborations with Dasgupta (Dasgupta and David 1987, 1994). See also David (1992: 222), where, in the context of ‘the growth of the stock of scientific knowledge’, epistemological connections are understood as ‘the way in which scientific knowledge is gained’. This is in order to criticise the trickle-down theory of technological progress from scientific advance, given more ‘feedbacks and interactions between advances in technology and science’. In other words, trickle-down is supplemented by trickle-up and trickle-across. For a critique of trickle-down, in the context of food knowledge, which does deal with the nature of knowledge and its origins in exploring the articulation between food and information systems, see Fine and Leopold (1993), Fine et al. (1996) and, especially, Fine (1998c). Note that Lawson’s (1997a: 251) worries about overgeneralisation due to timeless lock-in are not justified, as David and others are concerned to deploy models that allow any outcome, including opening, closing and simultaneity of systems. The response that there is no need for a proof because critical realism does not subscribe to a single pudding or recipe is bizarre to say the least (Lawson 1999a: 14). It would seem to imply that there are plenty of puddings around to be offered to sample. The failure to engage with Marxist political economy is both disappointing and surprising, particularly in view of the early start made by Keat and Urry (1975), and their discussion of critical realism in the afterword to the second edition of their volume.

References Arestis, P., G. Palma and M. Sawyer (eds) (1997) Capital Controversy, Post-Keynesian Economics and the History of Economics: Essays in Honour of Geoff Harcourt, Volume I, London: Routledge. Becker, G. (1996) Accounting for Tastes, Cambridge, MA: Harvard University Press. Bhaskar, R. (1975) A Realist Theory of Science, Leeds: Leeds Books. Blaug, M. (1998a) ‘Disturbing Currents in Modern Economics’, Challenge, 41: 11–34. —— (1998b) ‘The Problems with Formalism: Interview with Mark Blaug’, Challenge, 41: 35–45.

224

Ben Fine

Carrier, J. and D. Miller (eds) (1998) Virtualism: The New Political Economy, London: Berg. Coats, A. (ed.) (1996) The Post-1945 Internationalization of Economics, History of Political Economy, 28: Supplement. Collier, A. (1994) Critical Realism: An Introduction to Roy Bhaskar’s Philosophy, London: Verso. Cowan, R., P. David and D. Foray (1999) ‘The Explicit Economics of Knowledge Codification and Tacitness’, paper for presentation to the TIPIK Workshop, Strasbourg, April. Dasgupta, P. and P. David (1987) ‘Information Disclosure and the Economics of Science and Technology’, in G. Feiwel (ed.) Arrow and the Ascent of Modern Economic Theory, London: Macmillan. —— (1994) ‘Towards a New Economics of Science’, Research Policy, 23: 487–521. David, P. (1975) Technical Choice, Innovation and Economic Growth: Essays on American and British Experience in the Nineteenth Century, Cambridge: Cambridge University Press. —— (1992) ‘Knowledge, Property, and the System Dynamics of Technological Change’, World Bank Economic Review, Supplement: 215–48. —— (1997) ‘Path Dependence and the Quest for Historical Economics: One More Chorus of the Ballad of QWERTY’, University of Oxford Discussion Papers in Economic and Social History, no. 20 (http://www.nuff.ox.ac.uk/economics/ history/paper20/david3.pdf). —— and D. Foray (1994) ‘Dynamics of Competitive Technology Diffusion through Local Network Structures: The Case of EDI Document Standards’, in L. Leydesdorff and P. van den Besselaar (eds) (1994) Evolutionary Economics and Chaos Theory: New Directions in Technology Studies, London: Pinter. Fine, B. (1980) Economic Theory and Ideology, London: Edward Arnold. —— (1989) Marx’s ‘Capital’, third edition, London: Macmillan. —— (1990) The Coal Question: Political Economy and Industrial Change from the Nineteenth Century to the Present Day, London: Routledge. —— (1997) ‘The New Revolution in Economics’, Capital and Class, 61: 143–8. —— (1998a) Labour Market Theory: A Constructive Reassessment, London: Routledge. —— (1998b) ‘The Triumph of Economics: Or ‘Rationality’ Can Be Dangerous to Your Reasoning’, in J. Carrier and D. Miller (eds) Virtualism: The New Political Economy, London: Berg. —— (1998c) The Political Economy of Diet, Health and Food Policy, London: Routledge. —— (1999a) ‘From Becker to Bourdieu: Economics Confronts the Social Sciences’, International Papers in Political Economy, 5: 1–43. —— (1999b) ‘A Question of Economics: Is It Colonising the Social Sciences?’, Economy and Society, 28: 403–25. —— (2000a) ‘Endogenous Growth Theory: A Critical Assessment’, Cambridge Journal of Economics, 24: 245–65. —— (2000b) ‘“Economic Imperialism”: A View from the Periphery’, mimeo. —— (2000c) ‘Consumption for Historians: An Economist’s Gaze’, SOAS Working Paper in Economics, no. 91. —— (2000d) ‘New and Improved: Economics’ Contribution to Business History’, SOAS Working Paper in Economics, no. 93.

Addressing the critical and the real

225

—— (2000e) ‘Whither the Welfare State: Public versus Private Consumption?‘, SOAS Working Paper in Economics, no. 92. —— (2000f) ‘Bringing the Social back into Economics: Progress or Reductionism?’, University of Melbourne, Department of Economics Working Paper, no. 731. —— (2001a) Social Capital versus Social Theory: Political Economy and Social Science at the Turn of the Millennium, London: Routledge. —— (2001b) ‘Neither Washington nor Post-Washington Consensus: An Introduction’, in B. Fine, C. Lapavitsas and J. Pincus (eds) Development Policy in the Twenty-First Century: Beyond the Post-Washington Consensus, London: Routledge. —— and C. Lapavitsas (2000) ‘Markets and Money in Social Theory: What Role for Economics?’, Economics and Society, 29: 357–82. —— and E. Leopold (1993) The World of Consumption, London: Routledge. —— and L. Harris (1985) The Peculiarities of the British Economy, London: Lawrence and Wishart. —— M. Heasman and J. Wright (1996) Consumption in the Age of Affluence: The World of Food, London: Routledge. Fleetwood, S. (ed.) (1999) Critical Realism in Economics, London: Routledge. Hands, D. (1999) ‘Empirical Realism as Meta-Method: Tony Lawson on Neoclassical Economics’, in S. Fleetwood (ed.) Critical Realism in Economics, London: Routledge. Harcourt, G. (1972) Some Cambridge Controversies in the Theory of Capital, Cambridge: Cambridge University Press. —— (1976) ‘The Cambridge Controversies: Old Ways and New Horizons – or Dead End’, Oxford Economic Papers, 28: 25–65. Hausman, D. (1998) ‘Problems with Realism in Economics’, Economics and Philosophy, 14: 185–213. —— (1999) ’Ontology and Methodology in Economics’, Economics and Philosophy, 15: 283–8. Hodgson, G. (1997) ‘The Fate of the Cambridge Capital Controversy’, in P. Arestis, G. Palmer and M. Sawyer (eds) Capital Controversy, Post-Keynesian Economics and the History of Economics: Essays in Honour of Geoff Harcourt, Volume I, London: Routledge. —— and H. Rothman (1999) ‘The Editors and Authors of Economics Journals: A Case of Institutional Oligopoly?’, Economic Journal, 109: F165–86. Ingham, G. (1999) ‘Money Is a Social Relation’, in S. Fleetwood (ed.) Critical Realism in Economics, London: Routledge. Joseph, J. (1998) ‘In Defence of Critical Realism’, Capital and Class, 65: 73–106. Keat, R. and J. Urry (1975) Social Theory as Science, London: Routledge & Kegan Paul, second edition, 1982. Kilpatrick, A. and T. Lawson (1980) ‘On the Nature of Industrial Decline in the UK’, Cambridge Journal of Economics, 4: 85–102. King, J. (2000) ‘Has There Been Progress in Post-Keynesian Economics?’, paper presented to the conference of the European Society for the History of Economic Thought (ESHET), Graz, February, to appear in J. King (2001) The History of Post-Keynesian Economics, 1936–2000, Cheltenham: Edward Elgar. Lawson, T. (1997a) Economics and Reality, London: Routledge. —— (1997b) ‘Economics as a Distinct Social Science? On the Nature, Scope and Method of Economics’, Economie Appliqué, L: 5–35.

226

Ben Fine

—— (1999a) ‘Developments in Economics as Realist Social Theory’, in S. Fleetwood (ed.) Critical Realism in Economics, London: Routledge. —— (1999b) ‘Critical Issues in Economics as Realist Social Theory’, in S. Fleetwood (ed.) Critical Realism in Economics, London: Routledge. —— (1999c) ‘Connections and Distinctions: Post Keynesianism and Critical Realism’, Journal of Post Keynesian Economics, 22: 3–13. —— (1999d) ‘What Has Realism Got to Do with It?’, Economics and Philosophy, 15: 269–82. Lazear, E. (2000) ‘Economic Imperialism’, Quarterly Journal of Economics, 115: 99–146. Lee, F. and S. Harley (1998) ‘Peer Review, the Research Assessment Exercise and the Demise of Non-Mainstream Economics’, Capital and Class, 66: 23–51. Lestringant, F. (1997) Cannibals: The Discovery and Representation of the Cannibal from Columbus to Jules Verne, Berkeley: University of California Press. Lewis, P. (1999) ‘Metaphor and Critical Realism’, in S. Fleetwood (ed.) Critical Realism in Economics, London: Routledge. McCloskey, D. (1971) Essays on a Mature Economy: Britain after 1840, London: Methuen. Milonakis, D. (2000) ‘Market Socialism: A Case for Rejuvenation or Inspired Alchemy?’, paper to the conference of the European Society for the History of Economic Thought (ESHET), Graz, February. Roberts, J. (1999) ‘Marxism and Critical Realism: The Same, Similar, or Just Plain Different’, Capital and Class, 68: 21–49. Thompson, G. (1997) ‘Where Goes Economics and the Economies’, Economy and Society, 26: 599–610. —— (1999) ‘How Far Should We Be Afraid of Conventional Economics? A Response to Ben Fine’, Economy and Society, 28: 426–33. Zelizer, V. (2000) ‘Fine-Tuning the Zelizer View’, Economy and Society, 29: 383–9.

12 Economics as symptom Jean-Pierre Dupuy

Introduction: British critical realism versus French intersubjectivism In a review of Tony Lawson’s Economics and Reality, Edward Fullbrook mentioned the existence of a French ‘intersubjectivist’ school of economics (Fullbrook 1998). He contended that one could find in it the prolegomena to an ontological account of intersubjectivity. However, he wrote, these French economists ‘lack Lawson’s deep ontological analysis of economic reality so that the two sets of ideas, those of Lawson and the intersubjectivists, seem made for each other’ (p. 437). Fullbrook cited two scholars as being the leaders of the movement: André Orléan and myself. Since then, some connection has indeed taken place between our two groups. Tony Lawson invited me to make a presentation at his Cambridge workshop. In an important paper titled ‘Economics and Expectations’ (Lawson 1995) he elaborated on the French school’s insistence on the selfreferentiality of economic and social structures. The present contribution should be viewed as a further step in the rapprochement Fullbrook wished for. Jean-Paul Sartre once wrote that one becomes a Jew in the others’ eyes. If there is such a thing as the French intersubjectivist school, it was born in the eyes of its foes. By this I do not mean the orthodox or mainstream economists, but rather the orthodox heterodoxy, that is, various neoMarxist denominations that claim to exert a monopoly over the critique of neo-classical economics in the French intellectual landscape. The particular scholars who compose the movement have very diverse views and approaches and I have no special entitlement to represent them. However, these views and approaches have in common several crucial features which they share with the critical realist project. Other traits distinguish them from the latter, and a discussion of those differences against the background of the common beliefs and attitudes should prove constructive for both sides. This is at least my hope in writing the present chapter.

228

Jean-Pierre Dupuy

Common ground There are indeed similarities and possibilities for cross-fertilisation between the intersubjectivists1 and the critical realists. At the outset one finds on the two sides of the Channel a deep dissatisfaction with the current state of the economic discipline and the desire to transform it radically. Each side realises that this is a daunting task, considering the seemingly unassailable monopolistic position now occupied by the orthodoxy in every higher education or research institution. The two groups share fundamental assumptions about what constitutes a social system, assumptions that are flatly denied by the atomism and ahistoricism of mainstream economics. In both cases the ‘internal relationality’ of economic and social life is emphasised. Social structures are posited to be in a relation of causal circularity with human agency, the former being both condition and consequence of the latter. From this relation emerge new properties at the collective or social level, such as money or trust, that are irreducible to their very conditions of emergence. Taking seriously all the implications of this relationship of causal circularity between hierarchically distinct levels, the encompassing and the encompassed, is one of the major distinctive features of our respective approaches. The French school has been deeply influenced by, and has amply contributed to, the development and refinement of the theories of self-organising, complex systems, an important branch of what goes today by the name of cognitive science (Dupuy 2000a). These theories rest precisely on the exploration of the properties of such ‘strange loops’ or ‘organisational closure’, to take up the phrases of two of the most original proponents of cognitive science, Douglas Hofstadter (1979) and Francisco Varela (1979). To a large extent, the intersubjectivists and the critical realists appear to share the same social ontology. First of all, both schools assert that there is no social science that does not presuppose some kind of ontological commitment relative to its object. We’d better make that ontology explicit. As a consequence, it is impossible to practice economics, or carry out the critique of mainstream economics, if one remains within the confines of the discipline. It is indeed necessary to reach out to other social sciences and, above all, to philosophy. In France, if you go too far in that direction, however, sooner or later the economics profession will politely invite you to leave. That is what happened to me, and without too much regret I became a philosopher. The point remains that the philosophical critique of economics should not be considered an adventitious activity, completely separated from the practice of the profession or reserved for retired economists. It must be part and parcel of the economist’s daily concern. On this we agree too. As regards the ontology of the social realm, I’ve been striving to set the foundations for what I call a non-reductionist, complex monism. The

Economics as symptom 229 methodology I’ve been advocating is a form of methodological individualism, but, again, ‘complex’ and non-reductionist (Dupuy 1992a). I therefore applaud Tony Lawson’s advocacy of ‘a non-reductionist account of the relationship of agency and structure. Social structure is the, typically unacknowledged, condition of all our actions as well as the, usually unintended, consequence’ (Lawson 1995: 83). For all of us, it seems, this non-reductionism goes hand in hand with the recognition of the inevitability of some measure of opacity in human affairs. Tony Lawson writes: ‘While society and economy are the skilled accomplishments of active individuals, they remain to a degree opaque to the individuals upon whose activities they depend’ (Lawson 1995: 83). As we shall see, this social opacity plays a crucial role in the way the intersubjectivists see the task of social science, quite in keeping with a long tradition in French social thought. On this, we all seem to be indebted to the Scottish Enlightenment, the major lesson of which is admirably summarised by Adam Ferguson’s famous motto: social order ‘is the result of human action, but not of human design’. By and large the French intersubjectivists’ conception of science is not far removed from the critical realists’ – although it is on this issue that the two schools appear at some point to be parting company, as we shall soon see. ‘Science is primarily concerned with identifying and illuminating structures and mechanisms that underlie and govern the events of experience, but which are irreducible to them,’ Tony Lawson (1995: 81) writes, and we could not agree more. The major task of science is indeed the identification and systematisation of those deeper structures and mechanisms, rather than the elaboration of event regularities. As far as social science is concerned, by contrast with natural science, I readily acknowledge Lawson’s important point that ‘the major difference is precisely that defining characteristic of social structures that they, unlike natural ones, depend for their existence on human conceptions and actions (even if they are also a condition for such conceptions and actions)’ (Lawson 1995: 82). Nevertheless, this dependency cannot be absolute in so far as social structures are ‘intransitive’, i.e. not completely reducible to agents’ conceptualisations of them. More often than not, human beings know not what they are doing. Hence the need for an objective, ‘realist’, social science.

Reservations about critical realism: a different strategy is possible and called for Despite this common ground, I have reservations about critical realism as far as I understand its strategies, aims and style of thought. The best way to spell them out might be to bring out the differences in how the critical realists and the intersubjectivists see the epistemological, methodological

230

Jean-Pierre Dupuy

and social status of mainstream economics; and, as a consequence, in how they relate to it. According to Lawson, what constitutes the essence of the mainstream project in economics is the method it adopts along with its supporting ontology. This contention stands in sharp contrast with mainstream economic methodology, which tends to emphasise epistemology rather than ontology, as well as with the major trends in the critique of the mainstream project, which aim their attacks at the level of substantive claims. According to the French school, it is indeed because its substance is deeply flawed that mainstream economics must be criticised. However, ontology plays an important role in this critique, although it is related not so much to method as to substance itself. Actually, this might constitute the trademark of the French school’s critique by contrast with the critical realist project: ontology is taken to be inseparable from substance as well as from epistemology. Methodology appears by comparison to be a relatively minor target for the French critique. It is interesting, at the outset, to try to understand why. If mainstream economics rests on some kind of ontology, as it surely does, I do not see how that ontology can be different from what Lawson calls ‘transcendental realism’. Lawson characterises the latter as an ontology that includes ‘underlying structures of things, their powers, mechanisms and tendencies that, if triggered, act even if their effects are not directly manifest’ (Lawson 1999a: 5). Walrasian general equilibrium theory seems to be a first-rate candidate to meet this definition. As a model, despite its rhetoric, it is not interested in the least in explaining ‘event regularities’; its aim is to identify mechanisms that are sufficient to account for the kind of co-ordinated activity that prevails in a market economy.2 It is true that many economists, foremost among them the econometricians, but not only them, see their role and task differently.3 It has been one of the major achievement of intersubjectivists, I believe, to show that the economists themselves know not what they are doing. Their implicit epistemology and methodology are not at all what they take them to be. My task in what follows will be to illustrate this crucial contention. It follows that the critique of mainstream economics has a major pedagogical task, namely to enlighten the economists on the deeper meanings and alternative interpretations of their own concepts and models. For the sake of sounding less arrogant, let me hasten to add that a good number of mainstream economists have already undertaken this philosophical and hermeneutic revision. To take just one example, which I shall later elaborate, fewer and fewer are those who would still interpret the concept of economic equilibrium as Walras did – in terms of the physical balance between various forces. It is commonplace today to characterise an equilibrium as a set of representations or expectations that self-realise. We are no longer dealing with a social world modelled after

Economics as symptom 231 the natural one, but rather with a mental and cognitive universe. The change is dramatic. Obviously Lawson does not agree. I take the source of his disagreement to be of the utmost importance in revealing a major difference between the intersubjectivists and the critical realists. If, for the latter, neoclassical economics cannot be subsumed under ‘transcendental realism’, this has in the final instance much less to do with ontology than with methodology. This methodology has a name, ‘deductivism’, defined as ‘a mode of explanation which involves deducing the explanandum from a set of initial conditions plus regularities that take the form, “Whenever this event or state of affairs then that event or state of affairs” ’ (Lawson 1999b: 224). Now it might very well be disputed whether the Walrasian model does indeed meet this definition. However, Lawson seems to have a ready-made argument here: any formalistic modelling approach is of necessity deductivist. The fabrication of models entails deductivism since it presupposes it. As Lawson puts it as unambiguously as one could hope for: I suspect in fact that the deductivist method is accepted by mainstream economists because it is a precondition of mathematical modelling (for which [unthinking] enthusiasm continues unabated) and/or it is perceived (erroneously of course) to be the essential method of the successful natural sciences while no other way of proceeding has occurred to them. (Lawson 1999b: 226) Lawson writes: A condition of the successful application of science in open systems is that, against empirical realism, causal laws be analysed as expressing not event regularities, but tendencies of things that may be active but unrealised because of the action of countervailing tendencies. In other words, in citing a law one is referring to the transfactual activity of mechanisms, that is to their activity as such, and not making a claim about the actual outcome which in general will be co-determined by numerous mechanisms. (Lawson 1999b: 215) The recourse to mathematical models would render this approach impracticable: If a capacity of sorts is posited, then, if deductivist (typically formalistic) modelling is to proceed, it must be supposed that this

232

Jean-Pierre Dupuy capacity in the given situation is always exercised and its activity invariably realised or actualised. (Lawson 1999b: 224)

Now this may be true of very simple (not to say simplistic) models. But it becomes doubtful when it comes to (mathematical) models treated as complex objects. Indeed, the philosophy of modelling was radically transformed in the second half of the twentieth century with the introduction of such notions as catastrophes, attractors and bifurcations of non-linear dynamic systems, critical phenomena and symmetry breakings, self-organisation, deterministic chaos, disordered systems, and so on. It is well known today that complex systems, made up of many elements interacting in non-linear ways, possess remarkable properties – so-called emergent properties – that justify their description in terms that one should have thought had been forever banished from science in the wake of the Galilean-Newtonian revolution. Thus it is said of these systems (real-world objects as well as their mathematical representations) that they are endowed with ‘autonomy’, that their paths ‘tend’ towards ‘attractors’, that they have ‘intentionality’ and ‘directionality’ – as if their paths were guided by an end that gives meaning and direction to them even though it has not yet been reached; as if, to borrow Aristotelian categories, purely efficient causes were capable of producing effects that mimic the effects of a final cause (Dupuy 2000a). Let me take an example. One of the archetypes today of complex modelling is a connectionist (or neural) network made up of interacting elementary calculators that have been randomly connected. A very general property characteristic of such networks is that, after a transition period (often quite a short one), their aggregate behaviour stabilises to a ‘limit cycle’ – that is, a periodic spatio-temporal configuration – of weak periodicity. Everything occurs as though this stable aggregate behaviour were self-reproducing, as though it were produced by itself – hence the term ‘eigenbehaviour‘ (self-behaviour) used to characterise it – when in fact it is produced by the network. It is therefore legitimate to treat the network as an ‘autonomous’ being, as it were – autonomous in the sense that, being endowed with a spontaneity of its own, it is itself the source of its own determining characteristics and not a simple transducer converting input messages into output messages. Let us now consider an elementary automaton, or neurone, which happens never to be fired in the course of the limit cycle. It would thus appear to have no causal effect upon the outcome of the cycle. This can be said to be an illusion, and it is indeed one if one adopts the point of view of omniscience, the one that the whole system has over itself: one needs only to remove the neurone, along with its connections, from the structure of the network to realise that the aggregate behaviour is affected by its absence. However, if one chooses to treat the system as autonomous, everything occurs as if the potential

Economics as symptom 233 capacity of the given neurone to have an impact on the global behaviour through its connections to the rest of the system were not actualised. The countervailing tendencies due to the actions of all the other neurones have neutralised, as it were, a potential mechanism, preventing it from actualising itself. This example is interesting because it shows that the crucial factor is not the model per se, but rather the interpretative attitude one chooses to adopt in relation to it.4 We’ll examine below a typical illustration drawn from economics. It seems to me that mathematical modelling per se does not entail ‘deductivism’, and is perfectly compatible with transcendental realism. The fact that the critical realism critique puts so much emphasis on methodology might be the result of its distrust of mathematical modelling.5 By contrast, the French school does not see how science is possible at all without recourse to mathematics. It is the whole Western scientific tradition that led to the identification of thought with the production of models.6 From the fact that the French group downplays the role of methodology in its critique of mainstream economics, I can very well imagine that a champion of the critical realist project might conclude that we are unconscious (and unwilling) allies of the orthodoxy. That would be unfair. There is one question the critical realists do not seem interested in asking, although it occupies centre stage in the intersubjectivists’ approach. Why in the history and epidemiology of ideas has mainstream economics been such a success story? Because it was the result of a methodological mistake? When critical realism imputes the ‘crypto atomism’ of economic explanations, normally couched ‘in terms of individual (typically rational) decisions’ (Lawson 1999a: 7), to the requirements of the deductivist approach, I cannot help thinking that this is taking a narrow view of things. The ontological individualism that prevails in the methodology of the social sciences cannot be separated from the way our societies solve the problem of the social bond, the glue that holds them together and prevents them from disintegrating into violence. ‘Why is the economy the essential form of the modern world?’ was the question put by Hegel. No critique of economic reason can ignore it. These are substantial issues that involve a full-fledged social, political, and possibly moral, philosophy. It is in trying to address them that one comes to understand that mainstream economics is not there by chance, as the result of some unfortunate error in the choice of method. There exists a form of solidarity between mainstream economics and the economy-as-it-is that has to be exposed. That is at least the task the French group set for itself. The anthropological foundations of the economy (and their reflection in economics) have always seemed to us of much greater import than any discussion about methodology. Our respective attitudes towards mainstream economics are therefore quite different, not to say opposite. While critical realism is apparently

234

Jean-Pierre Dupuy

prepared to simply write it off if it could, the French school wonders whether there is not some kind of close adequacy between economics and the society that gave birth to it, a form of resonance or reflection that explains the prevalence and the permanence of the former. Economics (and the economy)7 seem to us symptoms and, as such, highly, although indirectly, informative of the world in which they developed. This account of the divergences in the ways the two groups ground their critique of the orthodoxy explains why the French ‘intersubjectivists’, unlike the British critical realists, set out to establish a conversational space with their orthodox colleagues, while remaining alert to their missteps, to the obstacles that they encounter, and, ultimately, to the limits they run up against. For any scientific paradigm, there are two kinds of critique, external and internal. External critique does not bother itself about delving into the subtleties of the paradigm it targets. It knows, from the outside as it were, that the latter does not work. The problem with external critique is that, most of the time, it has no impact whatsoever upon the champions of the targeted paradigm, who can continue to do their business as usual. Our strategy is quite distinct. It combines an internal critique – a typical example is our treatment of the paradoxes of rational choice theory – with an anthropological and philosophical analysis of the situatedness, the role and the social meaning of the economy-as-it-is as well as of mainstream economics. Philosophy, for us, is not reducible to epistemology in the French sense of the word, i.e. the police of (scientific) knowledge. It is a form of knowledge in itself. The critique of mainstream economics must include a full-blown ontological dimension, we fully agree on that, but ontology should be intimately connected with substance. As for anthropology, we draw mainly on the resources of the French tradition of social science, from Emile Durkheim and Marcel Mauss to René Girard and Louis Dumont. In what follows, I will set out to expound some (actually, a very small number) of the major issues addressed by the French group. I have chosen them as illustrations of the kind of strategy we are following by contrast with the one suggested by critical realists. Those case studies are all situated at a foundational level, which involves a philosophy of the social sciences and not only of economics.8 Obviously, I won’t be able to offer more than a summary treatment of each of them; a book-length development would hardly suffice for a full treatment. The critical realist reader shouldn’t be surprised if, contrary to the debates he or she is apparently so fond of engaging in, my approach draws on substantial case studies, examples, discussions of very specific or technical points and the like. I will consider in turn: an anthropological reading of the economy-as-it-is; the regulative role of social opacity and its consequences for the relationship between social ontology and the philosophy of the social sciences; an internal critique of the rational choice paradigm; and,

Economics as symptom 235 last but not least, the irreducibility of our social ontology to intersubjectivity.

The anthropological foundations of the primacy of the economy Max Weber once wrote: ‘A really definitive and worthwhile achievement is nowadays always a specialised achievement. Therefore, anyone who lacks the capacity to put on blinkers, so to speak, … should have nothing to do with science’ (Weber 1989: 9). It was because I refused to become a blinkered specialist that I turned to philosophy. The one question economists, orthodox and heterodox alike, will never pose, since they take the existence of their discipline for granted, is: what ‘great transformation’ did society have to incur in order to make the appearance of economics possible at all? What account can we put forward that might solve this central enigma: the economy, that most trivial part of our individual and collective existences, has become our major concern? Have we forgotten the Bard’s warning: ‘What is a man, / If his chief good and market of his time / Be but to sleep and feed? A beast, no more.’?9 These are economic questions that are too serious to be left to economics. For an answer to these queries the French group has searched the great Franco-British tradition of religious anthropology that was brought to a premature halt by decades of structuralism and post-structuralism: in particular, the French sociological school, with the works of Fustel de Coulanges, Durkheim and Mauss, and the British anthropological school, with Frazer and Robertson-Smith. Freud and the Belgo-British anthropologist Hocart gave these traditions a new momentum and, more recently, René Girard has produced an impressive synthesis. Several works authored by the French intersubjectivists have been directly influenced by Girard’s anthropology and it may be useful to recall its broad lines at the outset.10 Before dying out, religious anthropology had reached the following basic results (most of them were later reversed, or just forgotten, by the subsequent paradigms): a b c d

All non-modern social and cultural institutions are rooted in the sacred. Of the three dimensions of the sacred (myths, rituals and prohibitions) the most fundamental is the ritual. The most primitive and fundamental ritual form is sacrifice. Sacrifice is the re-enactment by the social group of a primordial event that took place spontaneously, a process of collective victimage that resulted in the murder of a member of the community. This elimination of a victim re-established peace and order. There lies the origin of the sacred. The victim is taken to be the cause or the active

236

Jean-Pierre Dupuy principle both of the violent crisis and of its violent resolution. It unites within itself opposite predicates: it is at the same time infinitely good and infinitely evil. It can only be of a divine nature.

Can a science of man exist if it abstains from posing the question of the origin of religion – if it dismisses as poorly formulated the problem of what makes all non-modern societies refer the social bond to an entity radically exterior to the world of men, the sacred? Can a science of the economy exist if it does not first ponder the major historical coincidence that characterises the modern world, the simultaneous retreat of religion and apotheosis of market value? At the heart of the Girardian ‘hypothesis’ is the proposition that the sacred is nothing other than the violence of men, expelled, exteriorised, hypostasised. The god-making machine runs on imitation. At the paroxysm of the ‘sacrificial crisis’, when a murderous frenzy has shattered the system of differences that makes up the social order and sparked a war of all against all, the contagious character of the violence produces a catastrophic convergence of every enmity upon an arbitrary member of the collectivity. Putting him to death is what abruptly restores peace. The result is religion in its three component parts. First, mythology: the interpretation of the founding event makes the victim out to be a supernatural being, capable at once of introducing disorder and of creating order. Next, ritual: always sacrificial at the outset, it begins by miming the violent decomposition of the group so that it may go on to stage the re-establishment of order through the killing of a substitute victim. Lastly, the system of prohibitions and obligations, the finality of which is to prevent a new eruption of the conflicts that previously engulfed the community. The sacred is fundamentally ambivalent: it uses violence to hold back violence. It contains violence, in both senses of the word. This is clear in the case of the sacrificial gesture that restores order: it is never other than one more murder, even if it is meant to be the last one; it is equally true of the system of prohibitions and obligations: the social structures which unify the community in normal times are the very same ones that tear it apart in times of crisis. When a prohibition is transgressed, the obligations of solidarity, leaping over the barriers of time and space (as in the mechanism of the vendetta), draw into an ever-wider conflict people who were in no way concerned by the original confrontation. These ‘things hidden since the foundation of the world’11 are not unknown to us: they have become an open secret. All one need do is glance through a newspaper: the term ‘scapegoat’ is served up at every opportunity. Just think about it: this expression declares the innocence of the victim, it reveals the mechanism of the exteriorisation of violence. Girard believes that this knowledge working through us is of divine origin – meaning that it derives from the one true God. The account of the

Economics as symptom 237 death of Jesus on the cross is, as anthropology has accurately observed, similar to the stories one finds at the heart of so many religions. If one sticks to the facts, there is no important difference between Christianity and a primitive religion. But it is the interpretation that changes everything. Here, paradoxically, Girard must pay homage to Nietzsche. The evangelical account is the first not to be narrated by the persecutors, it sides with the victim whose perfect innocence it proclaims. Which is why Nietzsche felt justified in accusing Christianity of being a slave morality. Girard is a Christian author, but this reference to Nietzsche, the most anti-Christian of all philosophers, brings out clearly that there is no need to share Girard’s religious beliefs to take his anthropology seriously. According to Girard, this knowledge has clogged up the works of the machine for making the sacred, damaging it irreparably. As it sacralises less and less well, it produces more and more violence, but a violence that has lost the power to impose order on itself. Such is the modern world, described as a ‘low-gear mimetic crisis … without catastrophic escalation or resolution of any kind’. The question that Paul Dumouchel and I tackled in this framework is the one the ‘Girardian system’ raises, but to which it offers no answer: what gives modern societies the capacity not only to resist, but even to feed on the growing indifferentiation of the world and the exacerbation of the resulting mimetic phenomena? We answered: the economy (Dumouchel and Dupuy 1979).12 Not that we claimed it constituted a substitute for the sacred, despite what facile metaphors suggest (the ‘Allmighty dollar’, etc.). But, like the sacred, the economy is ambivalent in its relationship to violence: it ‘contains’ it, in both senses of the word – thus reconciling Marx (for whom the economy is violence) and Montesquieu (according to whom the economy is the best means men could hope for in a desacralised society to keep their violence in check). As for the singularity of the economy’s relationship to mimesis, one has only to remark that the business world is the one place – poles apart from the intellectual world! – where people admit to imitating their rivals. At the heart of the economy lies what Dumouchel and I called the ‘exteriority of third parties’. This configuration is contemporary with the general weakening of the system of prohibitions and obligations of solidarity, which many authors trace to the influence of Christianity. Once the ties of solidarity among members of the community have withered, the intensification of mimetic rivalries no longer generates the polarisation against a single victim that is proper to sacrificial crises. People are more fascinated than ever by their doubles, whom they openly hate and secretly venerate, but these rivalries do not engulf the whole of the social space. The third parties are too caught up in their own fascinations not to feel themselves exterior to the rivalries of others. They do not have to take sides, and they see too clearly the truth of violence, namely its reciprocity: nothing divides the violent from each other but

238

Jean-Pierre Dupuy

their own hatred. At least, they see this truth where others are concerned, even if they never see it about themselves. People stand to one another as external third parties. Since everybody shirks their obligations of solidarity out of distraction with their private fascinations, nobody pays attention to the losers being produced all around by the antagonisms of others (the generalised competition of the market system). The economic order is the social construction of indifference to other people’s unhappiness. In this order it is not the relations between rivals that are marked by the greatest violence, but the relations of each of them with the others, that is, the relations among third parties. It is the refusal of the third parties to support the losers – much more than the blows they received from the winners – that sanctions their defeat and transforms it into a veritable social and sometimes physical death sentence. We have analysed in these terms the industrial revolution of eighteenth-century England and the resulting reorganisation of landed property. It was in that era that the question was first posed that still haunts us today: where on earth do all the poor come from, given that wealth is on the rise? In this framework, the so-called ‘individualism’ of the modern world appears to be a lie. But it is certainly not just any lie: it is a lie to oneself, a collective self-deception. The individuals’ ‘self-love’ is both a strategic lure, destined to conceal that they are more fascinated with one another than ever before (Dostoyevsky: ‘If God is dead, men will be gods for each other’), and the flip side of a real indifference to the excluded of the ‘great society’. It was a source of great intellectual enjoyment for me to be able to revisit the celebrated Das Adam Smith Problem in light of this anthropology and to show that the problem vanished into thin air once it was acknowledged that Smith’s self-love was by no means an absence of relation (as is the case with selfishness), but rather a strategy in a world of mutual fascination (Dupuy 1987). The ‘self-deceit’ of the market economy is the false belief that people seek after ‘utility’ while they are craving their spectators’ admiration. Self-love here takes the form of self-interest, of the economic motive, the desire to improve one’s material condition, to increase one’s wealth. Not because the riches acquired would be in themselves a source of satisfaction – Smith has no words harsh enough to express his scorn for this notion – but because they would have the property of attracting to their possessor the ‘sympathy’ of those who lack them. These people mistakenly attribute virtues to wealth that it does not have. But it is because they are mistaken, and because they covet it, that in the end they are not mistaken. Wealth indeed has the virtues with which it is credited, but only because it has been credited with them. It is this fool’s game that generates the Wealth of Nations and what we call the economy – but not without causing grave harm to morality. As is well known, this last point haunted Smith all his life, leading him to include, in the last edition of The Theory of Moral Sentiments, a new chapter significantly entitled ‘Of the corruption of our moral sentiments, which is

Economics as symptom 239 occasioned by this disposition to admire the rich and the great, and to despise or neglect persons of poor and mean condition’. The homine oeconomici of mainstream economics appear in this context the pitiable accomplices of this general delusion. I submit that this reading reaches further than the one that sees them as the inevitable products of a methodological mistake.

Social opacity and epistemology, or why epistemology and ontology are inseparable 1 Symbolic exchange and social hypocrisy The notion of the social lie, or collective hypocrisy or self-deception, has played an essential role in French social sciences, as much in Durkheimian sociology as in the structuralism that has dethroned it. The debate on the reciprocity of mutual exchange has been one of the major controversies in French-style human sciences. The French intersubjectivists have been heavily influenced by this debate, which is still going on today in the human sciences circles. I think it worthwhile to recall it here, because it will allow me to explicate another point that keeps us apart from the British critical realists: ontology should side with substance and epistemology. In his famous work The Gift (1967), Marcel Mauss notes that in a good number of archaic societies ‘contracts are fulfilled and exchanges of goods are made by means of gifts. In theory such gifts are voluntary but in fact they are given and repaid under obligation.’ He insists the payments have a ‘voluntary character, so to speak, apparently free and without cost, and yet constrained and interested … They are endowed nearly always with the form of a present, of a gift generously offered even when in the gesture which accompanies the transaction there is only a fiction, formalism and social deception, and when there is, at bottom, obligation and economic interest.’ Separate acts (giving, receiving, returning), present themselves as so many gestures of generosity or cordiality, yet in fact they obey strict, inescapable imperatives. What then is the nature of this ‘obligation’? Once he has asked this question, Mauss adds, as if he were only repeating it in another form: ‘What force is there in the thing given which compels the recipient to make a return?’ The indigenous informer will rapidly convince him that ‘in the things exchanged … there is a certain power [the ‘soul of things’] which forces them to circulate, to be given away and repaid.’ In his equally famous Introduction à l’oeuvre de Marcel Mauss (1950) – a text which many consider to be the charter of French structuralism – LéviStrauss reproaches Mauss for allowing himself here to be ‘mystified by the native’. Mauss’s mistake, according to him, was to have remained at

240

Jean-Pierre Dupuy

phenomenological apprehension, which breaks the instances of exchange into discrete occurrences. This thus creates the need for an operator of integration to reconstruct the whole, and it is precisely the ‘soul of things’ which providentially comes in to play this role. However, this is tackling the problem from the wrong end, Lévi-Strauss asserts, because: ‘Exchange is not a complex edifice, constructed from obligations to give, to receive and to make return with the help of an emotional and mystical content. It is a synthesis immediately given to, and by, symbolic thought.’ The ‘underlying reality’ of the exchange, he explains, is to be found in ‘unconscious mental structures’, to which language can provide access. The third step: Pierre Bourdieu (1977) denounced the ‘objectivist error’ of Lévi-Strauss: ‘Even if reciprocity is the objective truth of the discrete acts which ordinary experience knows in discrete form and calls gift exchanges, it is not the whole truth of a practice which could not exist if it were consciously perceived in accordance with the model.’ In effect, consider the obligation to repay what is received and the obligation to receive. Taken together in the theoretical schema of reciprocity, they lead to a contradiction. He who immediately returns the very object he is given refuses, in fact, to receive. The exchange of gifts can only function as such on the condition that the reciprocity that would be its objective truth is hidden. All the space, or rather the time, of practice is needed to undo this contradiction. Thus in Bourdieu’s interpretation (as in Mauss’s) there is a lie here. The natives know the truth of reciprocity, but they hide it, for this truth is lethal. From whom do they hide it? From themselves, of course. LéviStrauss resorts to the unconscious, Bourdieu to the deployment of temporality (the time of the practice), but, in both cases, the subject of which one speaks, and to whom one ascribes in one case the unconscious and in the other case self-deception, is a non-subject since it is the structure or the collective. This complication is apparently insurmountable, for Lévi-Strauss in any case, whom we leave with the collective unconsciousness structured like a language. Before we proceed further, let us ask the question: here is a social structure of which the stability and the very existence demand that its ‘internal’ truth (i.e. the phenomenological apprehension) differ from its ‘external’ (i.e. objective) truth and dissimulate it. Can the latter claim any kind of priority? Any greater truth value? Realists and positivists like Lévi-Strauss and Girard answer yes. Bourdieu and many others answer no. The critical realist reader will have noticed that this issue speaks directly to a fundamental point made by Lawson. The latter writes: Because social structure is existentially dependent upon human agency (as well as vice versa) the procedures of social science are likely to differ in various ways from those of natural science. However, it does not follow from this that social structures are other

Economics as symptom 241 than intransitive. To describe certain objects or features as intransitive is merely to indicate that they exist at least in part independently of any knowledge claims of which they are the referents. (Lawson 1999b: 222) Now, if Bourdieu is right here (as it seems to me is the case), then Lawson’s contention cannot always be correct. There are cases where the social structure is completely dependent on the kind of knowledge that the participants have about it, since its very existence depends on this knowledge. In the case of symbolic or gift exchange, the ‘misrecognition’ proper to collective self-deception is not only nor mainly a lack of knowledge (epistemology), but an institution (ontology) that implies the knowledge that is dissimulated. Ontological statements are not independent of epistemological ones. Clearly, the issue involves an important ethical dimension. One of the most fearsome attacks ever to be launched against Girard has been put forward by the political philosopher Pierre Manent. Girard, like Machiavelli long before him, declares that humanity is founded on violence and that the nature of the polis is inseparable from its violent origin. But, Manent enquires, does not this ‘revelation’, which obliges men to look in the eye the role of violence in the constitution of their world, leave them even more blinded than the myth which transfigures or erases that violence? To conclude: ‘The founding “myths” are more knowing than “realistic” science.’ For men have ‘good reasons’ for repressing violence: ‘They sense […] that the ends of social life are irreducible to its violent origins … Men need to hide from themselves those origins if they want to live in accordance with their conscience’ (Manent 1982). even Bourdieu: superior value is claimed for the ‘internal’ truth, quite simply because it is this, and not the ‘external’ truth, which makes possible the very existence of society. The geometrical metaphor should moreover be inverted, for it is not the external truth that encompasses the internal truth but rather the internal truth that encompasses the external truth, since it contains it: it has it inside itself and keeps it dammed up there. 2 Rationality and opacity ‘What has all of this to do with my discipline?’, the economist may ask, not without some measure of bad faith, since exchange is the subject matter in question. Before addressing this question, let me return to Bourdieu’s analysis of collective self-deception. It suddenly becomes much clearer when he takes the example of a labourer who proclaimed the convertibility of the meal traditionally given at the end of the work into money, with which he demanded to be paid instead. Bourdieu writes that here this worker was

242

Jean-Pierre Dupuy

only ‘betraying the best-kept and the worst-kept secret: the one that is in everyone’s keeping’. Undoubtedly a brilliant formula – we also say: ‘the secret is there is no secret’; or use the oxymorons ‘public’ or ‘open’ secret, etc. It turns out that this paradoxical epistemological configuration can be analysed rigorously thanks to a concept that has come to play today a momentous role in economics, rational choice theory and game theory, as well as in a whole range of disciplines, from the philosophy of language to analytical political philosophy to artificial intelligence: the concept of common knowledge (henceforth CK).13 A proposition is CK in a given population if and only if this proposition is true; everyone knows it is the case; everyone knows that everyone knows it is the case; etc., to infinity. Extremely interesting properties have been discovered about CK. The most fascinating property is perhaps the existence of a discontinuity between CK and ‘quasi-CK’. Quasi-CK is to be understood to be the case in which the degree of reflexiveness (measured by the number of interlocking levels of the type ‘Peggy knows that John knows that Peggy knows that … ’ etc.) tends towards infinity without reaching it. Take the following example: the airline asks you to present yourself at the registration counter two hours before take-off. This rule is stated inflexibly. However, everyone knows – or at least one might assume so – that not only can this rule be transgressed, efficiency indeed requires that it be broken. Indeed, the optimum is reached when passengers present themselves at the counter in a regular flow between T2 and T-1. Everyone knows that, and one may even suppose that everyone knows that everyone knows that, etc. (such an epistemic situation is described as ‘shared knowledge’). In order for the system to operate fairly harmoniously, all that is required is that one case be excluded: the metarule according to which the rule may be broken cannot be made public. It cannot become CK. What the French sociological and anthropological tradition analyses in terms of collective self-deception (or social hypocrisy) can then be described as situations in which shared knowledge obtains but not common knowledge. In such situations, a form of negative collaboration unites individual subjects who mean to protect a collective opacity that is convenient for all. The remarkable fact is that the very same configuration lies at the core of rational choice theory, that ultimate foundation of mainstream economics. A number of leading theorists today do not shy away from acknowledging the incompleteness of the rationality paradigm. Thus, according to the game theorist R.J. Aumann (1988): ‘Pure rationality is not sufficient unto itself, but can only be meaningful in a much larger domain, where irrationality has its place.’ In a quasi-Hegelian flight of eloquence, the same author refers to ‘the interface between rationality and irrationality as the major challenge facing theorists today’. It is indeed

Economics as symptom 243 well known that, in the framework of rational choice theory, the basic elements of what makes for the social bond turn out to be irrational or even impossible: for instance, making good on one’s promises when so doing runs against one’s interests; trusting one’s partner; forming an intention to retaliate if a potential attacker executes her threat, thereby deterring her, etc. Rationality turns out to be no less paradoxical than irrationality. Rational choice theorists should indeed be credited for having brought out this fundamental weakness, by discovering a number of paradoxes the elucidation of which has generated an abundant literature: Newcomb’s paradox, the repeated prisoner’s dilemma, the chain-store paradox, the deterrence paradox, the backward induction paradox, the toxin puzzle, etc. Each of these paradoxes takes the form of a dilemma. One branch is deemed rational by orthodox rational choice theorists in spite of its sounding, as the case may be, unreasonable, unethical, absurd, selfdefeating, self-stultifying or what have you. However, these calamitous conclusions appear to be based on an assumption that so far had seemed so natural that it had never been made explicit – namely, that the players are rational, that each knows that the others are rational, etc., ad infinitum. In other terms, the player’s rationality is common knowledge. For a paradigm that equates rationality with perfect transparency (under such diverse labels as ‘perfect foresight’, ‘rational expectations’ and the like), such an assumption, again, could be taken for granted. The scandal is that it appears now to be the culprit. A major scandal, furthermore, is that it is enough to depart ever so slightly from this assumption, by positing shared knowledge of a degree of reflexiveness as high as one wishes, but not infinite, to be able to reach conclusions that are in keeping with what reasonableness, common sense, conformity with experience, ethics, etc. demand. It is as though a measure of opacity (evidently deemed ‘irrational’ by the paradigm in question) proved indispensable for reason and rationality to be in accord once again: quite an unlikely rapprochement with the French ‘sciences de l’homme’ indeed! The distinction between internal and external critique is quite clear here. In the case at hand, internal critique requires that one work out the paradox constituted by the self-defeat of the rationality paradigm – a task much more ambitious and of consequence than the mere observation that people, in real life, do not behave in accordance with the theory! Embarking on this task leads to surprising results. Firstly, it reveals that there is no serious critique of economic reason without metaphysics. I draw a distinction here between ontology, which concerns itself with the nature of being, and metaphysics, which addresses the age-old ‘big’ questions about the world, God and the self, time and causality, and so on. To be sure, metaphysics is not a (positivist) science, but it purports to be a rational discipline. It is simply wrong to think either that metaphysics is dispensable or that anything goes in that domain as far as the human

244

Jean-Pierre Dupuy

sciences are concerned. Economics, in particular, rests on a praxeology (i.e. a theory of action) and as such it involves presuppositions about the nature of time and causation that most of the time remain wholly implicit. Secondly, a metaphysical critique of mainstream economics leads one to conclude that it suffers much less from its alleged inadequacy to the real social world than from its internal flaws and inconsistencies. As far as rational choice theory is concerned, the critique that is the most upsetting for its practitioners is not that the kind of ‘disembedded’ rationality it assumes is just a void abstraction, but rather that it is incoherent. The major problem with economic rationality is not that it is too rational for the world as it is (hence the vogue for such notions as bounded, limited, situated, etc. rationality) but that it is not enough rational to be up to reason. The problem is not that it assumes too much in terms of transparency, completeness and perfection of information; the problem is that it entails an irreducible opacity to oneself. At least, this is what I trust we have been able to show in a number of cases linked to the paradoxes I referred to above. There are extensive games for which it has been proven that not only rational choice theory as we know it but also any alternative theory of rationality is incompatible with the assumption that rationality according to its standard is common knowledge, or that the theory itself is common knowledge (Reny 1993). In such cases, in other words, there is no theory of rational choice that is perfectly transparent to itself. This is a most serious blow dealt … at what exactly? I have shown that what is at stake here is the model of practical reason that rational choice theory has inherited from Leibniz. The agent faces all the possibles and chooses the one that maximises a certain index that supposedly represents her self-interest. She could have acted otherwise: herein resides her freedom. Involved here are: a conception of time (with a fixed past and an open future); a solution to the ‘compatibilist’ challenge (positing agents that are endowed with free will while they cannot help but choose the best); a theory of modalities (possibility, necessity, contingency); a theory of possible worlds and counterfactual conditionals; and more. It turns out that this model does not exhaust what we mean by practical reason. Another consistent model, situated somewhere between Spinozism and the Leibnizian account of freedom, plays a fundamental role not only in the way we find it rational to reason and act, but also in the theoretical constructions we have been using to account for social reality. I could show that the Walrasian framework, including its most recent avatars (e.g. the theory of rational expectations), corresponds to this second model. If I am right, the fact that over the last decades game theory and economics have been increasingly associated rests, as far as their grounding in a metaphysical conception of rationality is concerned, on a confusion. Game theorists operate within the framework of orthodox, ‘Leibnizian’ practical reason, along with all kinds of rational choice theorists. However, this is not the same metaphysical paradigm as

Economics as symptom 245 economic theory. Unaware of this duality, economists are prone to switch from one model to the other within a single argument, rendering it philosophically, and therefore scientifically, absurd.14 For all those contradictions, antinomies and paradoxes, there are ways out. This is not the place to explicate what they are.15 One thing is certain, on which I fully agree with Lawson: philosophy is indispensable. There is no ‘shifting the mainstream‘ of economics without a radical questioning of the philosophical foundations of the discipline. This should have tremendous consequences for the way it is taught and advanced. 3 Rationality and intersubjectivity, or why the French ‘intersubjectivists’ are not intersubjectivist As I mentioned in passing, Girard’s anthropology emphasises the role of imitation in human affairs, if only because his is a theory that gives centre stage to disruptive passions and the tragic of human life. By contrast, the vast majority of market theorists are oblivious to imitation. There are profound reasons for this. What is at stake here is both the conception of the modern individual and that of the social order. The independent and self-sufficient individual posited by economic theory is not supposed to be subject to the influence of his peers – a perfect example of collective self-deception as we saw. The collective phenomena of which the market is the framework are not supposed to have anything in common with the crowd phenomena and the contagion of sentiments and acts of which they are the theatre. And yet the three finest economists of all time, Smith, Keynes and Hayek, gave imitation a central role. No wonder the French intersubjectivists have carefully analysed their works in order to understand how and why they differed so much from the mainstream. By and large, our work on mimetic processes represents the landmark of our approach. I will here focus on Hayek and point to a contradiction that should strike any reader of his work. All through his writings, the Austrian economist marvels at that wonder of social self-regulation that is the market. It automatically finds the path of its equilibrium, and this equilibrium is an efficient social state. What gives the market its capacity for self-organisation? The answer lies in the negative feedback mechanisms that automatically go into play as soon as an agent departs from equilibrium behaviour. Simultaneously, Hayek stresses the role of imitation in market competition. Now, it is well known that imitation is eminently productive of positive feedback, a major source of dynamic instability. Hence the question: how can Hayek reconcile his unshakeable faith in the market’s capacity for self-regulation and his acute perception of the importance of imitation? Indeed, the problem is much broader since it concerns not only his theory of the market, but also his theory of

246

Jean-Pierre Dupuy

cultural evolution, which likewise associates competition, imitation and efficiency. In order to grasp clearly the nature of the problem that imitation poses for any theory of social self-regulation, consider the following elementary model. Two subjects A and B reciprocally imitate each other. The object of their mutual imitation is of indeterminate nature. But suppose that a rumour leads A to believe that B desires (seeks, wants to buy, places confidence in, hopes for, etc.) an object O. A now knows what he needs to desire (respectively: to seek, etc.): he then takes the initiative himself in such a way that his own action brings the object O to B’s attention, and when B manifests in turn his interest in O, A has proof that his initial hypothesis was correct. His representation, as implausible as it may have been a priori, has been self-realised. This phenomenon where an objectivity or exteriority emerges through the closure upon itself of a system of actions all imitating one another – let us call it a process of selfexteriorisation or self-transcendence – gathers strength as the number of actions rises. The most absurd rumours can polarise a unanimous crowd upon the most unexpected object, everyone finding proof of its value in the eyes or gestures of all the others. The process unfolds in two stages: the first is a mirror game, specular or speculative, in which everyone watches for signs of the coveted knowledge in everyone else, until they all end up being propelled in the same direction; the second stage is the stabilisation of the object that has emerged as the arbitrariness inherent in the conditions of its genesis is forgotten. The unanimity that presided at its birth projects it for a time outside the system of actors, who, all looking in the direction that it indicates, stop searching for clues in one another’s gaze. This phenomenological description of the world of imitation can be sharpened and confirmed by mathematical modelling, which proves once again to be an indispensable tool, far from the accusations levelled by the British critical realists. A very active branch of formal economics is today exploring the role of what it calls interpersonal influences in economic activity. It helps to appreciate how far removed this mimetic universe is from the ideal market. Contrary to what one might have thought a priori, and to what many authors have indeed thought, generalised imitation produces something rather than nothing. It creates self-reinforcing dynamics that converge so resolutely on their target that it is difficult to believe that this convergence is not the manifestation of an underlying necessity, in the manner of a mechanical or thermodynamic system returning invariably to its equilibrium state after straying from it under the effect of some perturbation. Yet one sees that the concept of equilibrium, which the theory of the market imported from rational mechanics, is absolutely unsuited to characterise the ‘attractors’ of mimetic dynamics. Far from expressing an implicit order, they spring from the amplification of an initial disorder, and their appearance of pre-

Economics as symptom 247 established harmony is a mere effect of unanimous polarisation. They are condensations of order and disorder. The mimetic dynamic seems to be guided by an end that pre-exists it – and that is how it is experienced from the inside – but it is in reality the dynamic itself that brings forth its own end. Perfectly arbitrary and indeterminate a priori, it acquires a quality of self-evidence as the grip of collective opinion tightens. In coming to an ‘equilibrium’, the economists’ ideal market is supposed to reflect an external reality. The prices express objective, ‘fundamental’ values that synthesise information as diverse as the availability of techniques, the scarcity of resources or the preferences of consumers. The mimetic dynamic for its part is completely closed upon itself. The attractors that it generates are not in any relationship of correspondence with an external reality, they simply reflect a condition of internal consistency: the correspondence between a priori beliefs and a posteriori results. The mimetic attractors are self-realising representations. Generalised imitation has then the power to create worlds that are perfectly disconnected from reality: at once orderly, stable and totally illusory. It is this ‘mythopoetic’ capacity that makes it so fascinating. One of the most interesting achievements of our group has probably been the following. André Orléan has shown that the same models of rational expectations that serve to formalise the progressive unveiling of accurate information, if it exists, and the unanimous convergence of the agents on the optimal behaviour, can just as easily depict an effect of polarisation on any single arbitrary value when the agents, in the grip of uncertainty, seek a remedy in reciprocal imitation. Between the equilibrium and the self-realising representation, there is no formal difference. Yet an abyss separates these two worlds as far as their meaning is concerned.16 Here again, we find that mathematical modelling per se has no ontological implications or presuppositions. Everything hinges on the interpretation. We discover furthermore that although rational expectations models stage the usual rational homo oeconomicus, everything occurs as if these models contained (again in both senses of the word) the logic of imitation. They too are symptoms of a reality that transcends them. I will now come back to my original question: how can Hayek avoid such cheerless conclusions, given that his social philosophy is based on a philosophy of mind in which imitation plays the starring role? The answer is quite simple. Consider a universe where everybody imitates everybody else, with the exception of a single individual who imitates nobody. It is easy to demonstrate that this individual will become the keystone of the system in that everyone will end up imitating him and him alone.17 Let us make one more assumption: this individual imitates nobody because he knows he is right. Then we have an evolutionary process that acts as a very efficient discoverer and propagator of information. If we consider the case of cultural evolution, the condition

248

Jean-Pierre Dupuy

for the efficiency of the process is therefore not that those who imitate another tradition are aware of what they are doing. It is that those who, by chance, stumble on to the ‘right’ tradition, stick with it and stop imitating others. It is thus not those who imitate, but those who do not imitate, who embody the consciousness of evolution. The problem is that this consciousness must come to them from the outside. The tragedy of the West, according to Hayek, is that having discovered the extended order of the market and the liberal principles that govern it, it was unable to recognise their superiority and immediately abandoned them in favour of the ‘constructivist’ illusions. We encounter once more this troubling property of imitation, namely its ambivalence. It is efficient if the correct information is present somewhere and recognised as such, but otherwise it becomes a source of illusions and waste. The problem, of course, is that it is impossible from inside the system to know in which of the two cases one finds oneself. To overcome this undecidability, it is necessary to resort to an exteriority. When the evolving path arrives at ‘truth’ or ‘efficiency’, a buzzer must go off signalling ‘look no further’ – in other words, ‘stop imitating.’ The self-exteriorisation produced by generalised imitation will only manifest its optimising virtue within the framework of a genuine exteriority. If we are talking about the cultural evolution of humanity, the question is obviously what status to give to this exteriority and who may speak in its name. There is no way to avoid prophets here and the foremost prophet, of course, goes by the name of Hayek. A last word might be fitting on the reason why the label of ‘intersubjectivists’ that Edward Fullbrook gave us is not completely satisfactory. If our social ontology could be summed up in one sentence, that would be: Society ‘infinitely transcends any intersubjectivity’ or ‘Society is irreducible to “intersubjectivity” ’.18 Again, and that might be my final word, social philosophy and ontology have an irreducible ethical dimension. Autonomy appears as the key value. An autonomous society is not a society transparent to itself and autonomy is not mastery. There is a gap, an insurmountable abyss, between individuals and the social. This gap is an ontological difference separating two modalities of being that are radically distinct and radically irreducible to one another. And yet, it is people, individuals, who create their institutions and, in an autonomous society, create them lucidly, in explicit and reflective fashion. There is no contradiction here, for society creates and alters itself in and through the individuals who compose it. No contradiction, then, but a paradox: the ‘layered’ ontology we have isolated under the label ‘selftranscendence’ recognises radically disjointed strata of being that are nevertheless capable of penetrating one another, the encompassed level being in a position to engender the encompassing level because the latter is already inside the former – even though the encompassing level ‘infinitely’ transcends the encompassed level.19

Economics as symptom 249

Notes 1 For the sake of simplicity, I will adopt the name Fullbrook gave us. This is not what we are called in France, though. The school goes by the name of ‘Économie des conventions’, where the word ‘convention’ refers to the Humean notion as conceptualised and formalised by David K. Lewis in his book Convention (1969). I will show later that ‘intersubjectivism’ is not a perfectly adequate term to describe our social ontology. 2 I therefore fully agree with D. Wade Hands (1999: 170) when, in his own critique of Lawson, he writes: ‘Neoclassical economics is not in any sense the empirical realist-inspired inquiry that Lawson makes it out to be – in fact, it seems to be more consistent with the type of transcendental realism that he himself endorses.’ 3 Lawson was too happy to quote the French economist Maurice Allais as saying, ‘The essential condition of any science is the existence of regularities which can be analysed and forecast. This is the case in celestial mechanics. But it is also true of many economic phenomena. Indeed, their thorough analysis displays the existence of regularities which are just as striking as those found in the physical sciences. This is why economics is a science, and why this science rests on the same general principles and methods of physics’ (Allais 1992: 25; quoted in Lawson 2001). Like many French economists, Maurice Allais received a thorough education in mathematics and physics at the École Polytechnique in Paris, the French MIT. Walras himself was trained as an engineer at the École des Mines de Paris. 4 A similar point could be made in the case of the property known as ‘sensitivity to initial conditions’ that characterises a system (a model) endowed with ‘weak stability’: the ‘whenever event x then event y’ form obviously does not obtain. 5 Lawson (2001) writes: ‘We have good reason to suppose that the scope of relevance of mathematics is very limited indeed in the social realm. For example, it can be demonstrated that not only the poor success rate of modern economics, but also the phenomenon of modern economists repeatedly making assumptions known to be wildly false, are due to mathematical methods being employed where they do not fit.’ 6 The ‘post-autistic economics’ movement in France seems to have caught the attention and interest of the British critical realists. One can understand why. This movement denounces ‘the pathological role’ played by mathematics in economics, notes ‘a real schizophrenia created by making modelling an end in itself and thereby cutting economics off from reality and forcing it into a state of autism’. It is indeed a scandal that economics should be taught in the French ‘Grandes Écoles’ not as a human science but as a chapter of mathematics. However, one should be wary of not transforming a wellfounded protest into a dubious combat in which mathematical modelling per se is the scapegoat. 7 In French we have only one word, ‘économie’, to designate both the economy, i.e. a sector of the social world, and economics, i.e. a viewpoint on the social world. I am still pondering whether that’s an infirmity of the language or a sign of its genius. 8 With the unfortunate consequence that I won’t say anything about the empirical part of our research programme. Although I thought that was a reasonable price to be paid given the theoretical nature of the debate launched by the British school, the reader must be aware that that absence creates a distorted view of the French ‘Économie des conventions’. 9 Hamlet, IV, 4. 10 See, in particular, Girard (1979, 1986).

250 11 12

13 14

15 16 17 18

19

Jean-Pierre Dupuy A quotation from the Gospel according to Matthew that serves as the title of another book of Girard’s. Michel Aglietta and André Orléan proposed a Girardian analysis of money, seen as a social relation brought forth by the same mechanism of exteriorisation of violence, in their important book La Violence de la Monnaie (1979). In English, see Orléan (1992). Philosophy of language is actually the field where it was born, in particular in the work of David K. Lewis, Convention (1969), which gave the French group its name (Économie des conventions). An illustration which I leave the reader to ponder: in the discussions about games put in extensive form, such as the ‘centipede’ game, one encounters arguments that can be proven equivalent to the form: an agent predicts the future and changes it. See Dupuy (1989, 1992b, 1998, 2000b). Cf. Orléan (1986, 1988a, 1990). This property is a trivial consequence of a theorem demonstrated by Orléan (1988b). These two phrases are from a political and social philosopher whose influence on our group has been important, Cornelius Castoriadis. See the chapter ‘Power, Politics, Autonomy’ in his Philosophy, Politics, Autonomy: Essays in Political Philosophy (1991: 144-5). I have been able to show that this social ontology is already present, although in an embryonic fashion, in the works of the three economists I mentioned earlier as having emphasised the role of imitation in human affairs, Smith, Keynes and Hayek. See my ‘Individualism and Self-Transcendence’ (2000c).

References Aglietta, M. and A. Orléan (1979) La Violence de la Monnaie, Paris: PUF. Allais, M. (1992) ‘The Economic Science of Today and Global Disequilibrium’, in M. Baldassarri, J. McCallum and R. Mundell (eds), Global Disequilibrium in the World Economy, Basingstoke: Macmillan. Aumann, R.J. (1988) ‘Preliminary Notes on Irrationality in Game Theory’, mimeo, Stanford University. Bourdieu, P. (1977) Outline of a Theory of Practice, translated by Richard Nice, Cambridge: Cambridge University Press. Castoriadis, C. (1991) ‘Power, Politics, Autonomy’, in Philosophy, Politics, Autonomy: Essays in Political Philosophy, edited by David Ames Curtis, New York: Oxford University Press. Dumouchel, P. and J.-P. Dupuy (1979) L’Enfer des Choses: René Girard et la logique de l’économie, Paris: Seuil. Dupuy, J.-P. (1987) ‘De l’émancipation de l’économie: Retour sur “Das Adam Smith Problem” ’, L’Année Sociologique, 37: 311–42. A short English version has been published as ‘A Reconsideration of “Das Adam Smith Problem”’, in P. Saint-Amand (ed.) (1993) Autonomy in the Age of the Enlightenment, Stanford French Review, 17.1: 45–57. —— (1989) ‘Common Knowledge, Common Sense’, Theory and Decision, 27: 37–62. —— (1992a) Introduction aux Sciences Sociales, Paris: Ellipses. —— (1992b) ‘Two Temporalities, Two Rationalities: A New Look at Newcomb’s Paradox’, in P. Bourgine and B. Walliser (eds) Economics and Cognitive Science, London: Pergamon.

Economics as symptom 251 —— (ed.) (1998) Self-Deception and Paradoxes of Rationality, Stanford University: CSLI Publications. —— (2000a) The Mechanization of the Mind: On the Origins of Cognitive Science, Princeton, NJ: Princeton University Press. —— (2000b) ‘Philosophical Foundations of a New Concept of Equilibrium in the Social Sciences: Projected Equilibrium’, Philosophical Studies, 100: 323– 45. —— (2000c) ‘Individualism and Self-Transcendence’, paper presented at the Conference: ‘Cornelius Castoriadis: Rethinking Autonomy’, Columbia University, New York, 1–3 December 2000. Fullbrook, E. (1998) ‘Shifting the Mainstream: Lawson’s Impetus’, Atlantic Economic Journal, 26: 431–40. Girard, R. (1979) Violence and the Sacred, Baltimore: The Johns Hopkins University Press. —— (1986) The Scapegoat, Baltimore: The Johns Hopkins University Press. Hands, D.W. (1997) ‘Empirical Realism as Meta-Method: Tony Lawson on Neoclassical Economics’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London: Routledge. Hofstadter, D. (1979) Gödel, Escher, Bach, New York: Basic Books. Lawson, T. (1995) ‘Economics and Expectations’, in S.C. Dow and J. Hillard (eds) Keynes, Knowledge and Uncertainty, Aldershot: Edward Elgar. —— (1997) Economics and Reality, London and New York: Routledge. —— (1999a) ‘Developments in Economics as Realist Social Theory’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. —— (1999b) ‘Critical Issues in Economics as Realist Social Theory’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. —— (2001) ‘Back to Reality’, Post-Autistic Economics Newsletter, 6. Lewis, D. (1969) Convention: A Philosophical Study, Cambridge, MA: Harvard University Press. Manent, P. (1982) ‘La Leçon de Ténèbres de René Girard’, Commentaire 19. Mauss, M. (1967) The Gift: Forms and Functions of Exchange in Archaic Societies, translated by Ian Cunnison, New York: Norton. Orléan, A. (1986) ‘Mimétisme et Anticipations Rationnelles: Une Perspective Keynésienne’, Recherches économiques de Louvain, 52. —— (1988a) ‘L’autoréférence dans la Théorie Keynésienne de la Spéculation,’ Cahiers d’économie politique: 14–15. —— (1988b) ‘Money and Mimetic Speculation’, in P. Dumouchel (ed.) Violence and Truth, Stanford: Stanford University Press. —— (1990) ‘Le Rôle des Influences Interpersonnelles dans la Détermination des Cours Boursiers,’ Revue économique, 41: 839–68. —— (1992) ‘The Origin of Money’, in F. Varela and J.-P. Dupuy (eds) Understanding Origin, Kluwer: Boston Studies in the Philosophy of Science. Reny, P. (1993) ‘Common Belief and the Theory of Games with Perfect Information’, Journal of Economic Theory 59 (2): 257–74. Varela, F. (1979) Principles of Biological Autonomy, Amsterdam: North Holland. Weber, M. (1989) Science as a Vocation, edited by P. Lassman and I. Velody with H. Martin, London: Unwin Hyman.

13 The Economics of Institutions and the Institutions of Economics Ugo Pagano

Introduction1 In my opinion, critical realism and institutionalism are two projects that share much and can be reciprocally enriching. However, the purpose of this chapter is not to list commonalities of the two projects, useful though I think such an enterprise would be. Rather, I would like to support the contention that the two projects can support each other. Institutionalism, of course, is primarily concerned with exploring the nature, manner of functioning and consequences of specific institutions. Here, I want to focus only on one institution. But it is an institution of special interest to economists, for it is the institution of (modern) Economics itself. In recent years critical realism has advanced numerous claims, two of which I want to focus on here. The first is that social reality includes social ‘structures’ that are ontologically distinct from people and influence their way of acting and thinking.2 The second is that the methodology of modern Economics is often characterized by formalisms that are inappropriate for understanding an ‘open system’ such as the economy. The purpose of this chapter is to consider a possible relationship between these two points by arguing that the ‘Economics of Institutions’ that has recently been gaining some ground among economists should apply to the ‘Institutions of Economics’ itself. We will argue that the structure of production of knowledge in the discipline of Economics has biased the economic literature towards an excessive use of formal techniques of analysis that may be inadequate for understanding (a rapidly changing) economic reality. The chapter is divided into three sections. In the first section we claim that social structures not only provide the social sciences with objects of enquiry but can also, in turn, influence the way in which people learn. In particular, the structure of production of the knowledge and the incentives faced by the individuals may bias the nature of knowledge that is produced in particular directions. Here some basic tools of modern Economics may be useful in developing an ‘Economics of knowledge’ and, in particular, some sort of ‘Economics of Economics’. In the second

Economics of Institutions and Institutions of Economics 253 section we consider the second critical realist claim noted above, namely the idea that formal modelling is often inadequate for understanding an ‘open’ economic system. Our profession may be characterized by an overproduction of formal techniques or, at least, by an unnecessary concentration on problems that are suited to this type of methodology. The emphasis placed on the assumption of maximizing behaviour will be examined as the most typical example of this type of overproduction. Finally, the last section joins together the two points. We argue that the structure of production of Economics and its incentive structure explain why the overproduction and the overapplication of some formal techniques takes place.

Structured reality and the production of social science According to critical realism reality is structured in the sense that it contains structures and mechanisms that are ontologically irreducible to the actual course of events. However, we rarely observe the mechanisms themselves, only their complex interactions. The mechanisms express themselves in tendencies that can be weakened or even overcome by the existence of other mechanisms. However, even when their manifest effects are completely eliminated, these mechanisms are still there. According to critical realism the role of science is to understand the nature of the mechanisms by whose complex interaction observable events are produced. The way in which these structures operate is particularly complex in the case of human society. The freedom of choice of human agency may, in principle, always upset the working of the structures, which in turn condition the way in which the individuals interact. However, while it is a typical mistake of ‘determinism’ to see human agency as a simple outcome of social structures, it is a typical mistake of ‘voluntarism’ to explain social structure as the outcome of the intentional choices of individuals.3 Human intentional activity does not create social structure ex nihilo. Rather, individual agents ‘draw upon social structure as a condition of acting, and through the action of individuals taken in total, social structure is reproduced or (in part at least) transformed’ (Lawson 1997: 169, emphasis in original). The production/transformation of social structure is rarely an intended project and usually individual agents are not aware of the structures upon which they are drawing. Often as an unintended consequence of their actions, individuals reproduce a set of positions that define the nature of their relations and are independent of the characteristics of the particular individuals who occupy them. Each position is defined by numerous rights, duties, liberties and exposures to liberty.4 Examples of such relations include the following: employer/employee, parent/child and teacher/student. A social system can be conceived as a set of structured processes of interaction

254

Ugo Pagano

characterized by networked, internally related positions with associated rules and practices, while an institution may be defined as those structured processes of interaction ‘that are relatively enduring and identified as such’ (Lawson 1997: 318). Thus, while institutions are reproduced through the actions of individuals, institutional change is rarely intentional; more often it is an unintended consequence of these actions. This is not only due to the bounded rationality of individuals but also to the difficulty of organizing collective action. A social group or a collectivity (including the individuals who occupy a specific set of social positions) may share the same interests. However, its members may find it difficult not only to discover and articulate the contents and the implications of their interests but also to solve the ‘free rider’ problems that are associated with collective action. Institutions influence behaviour because they provide the rules of the game and the incentives faced by individuals. Moreover, they also mould behaviour at a deeper level because they change habits and preferences (Lewis and Runde 2002: 209). This makes it difficult to become aware of the shortcomings of inefficient, or even oppressive, institutions. In the critical realist approach social scientists can help to overcome this problem. They can enable individuals to understand the mechanisms by which their interactions are structured and the nature of the institutions in which they live. In this way, the social sciences may have a progressive and emancipatory role insofar as the individuals may become aware of unsatisfactory institutions and understand some possible ways of changing them. However, in this approach, the institutions of production of social scientific knowledge must themselves be an object of study. As on the cover of Lawson’s Economics and Reality, the painter must necessarily be in the painting. The ‘Economics of Economics’ must be part of the field in the sense that the ‘Economics of Institutions’ cannot ignore the ‘Institutions of Economics’. The Institutions of Economics can help a great deal the advancement of economic studies but they can also bias research in certain directions and inhibit the development of certain topics and methods of research. It is a claim of the chapter that, in some cases, the Institutions of Economics can inhibit the progress of the Economics of Institutions.5

Economics and maximization The view of the individual which emerges from both critical realism6 and Institutional Economics suggests that individuals are made of a complex mix of instincts, emotions, habits and conscious rationality. In the same society the mix is very likely to change from individual to individual. Indeed, in real life individuals are characterized by different degrees of rationality and by different emotions. Moreover, these characteristics are

Economics of Institutions and Institutions of Economics 255 influenced by the institutions found in society and often reflect different nationalities and different moments of history. The way in which individuals with well-specified characteristics may act under and reproduce some specific institutions could be a fruitful field for economists and other social scientists. One should consider how, under some circumstances, because of their personal characteristics, the individuals tend to reproduce some institutions and, in turn, these institutions tend to reproduce their complex mix of instincts, emotions and habits. In other words, Economics could study how the interactions among individuals tend to produce ‘closures’ under which institutions tend to reproduce themselves via the characteristics of the individuals and the latter tend to reproduce themselves via the working of institutions. Even when the process of (re)production of institutions determines a closure, it is still a dynamic process and the way is always open to an end of the closure: human history is always potentially open to change in virtue of the conscious or unconscious human capacity to rebel against traditions and to innovate. By contrast, modern Economics has usually been concerned with a closed world where many characteristics of the individual are conflated in an unbounded rationality assumption.7 This assumption cancels the large majority of the differences among the individuals of the same society and among different societies. Historical context and individual characteristics lose much of their relevance in a world where all individuals show the very same degree of unbounded rationality, do not have irrational emotions and are never conditioned by habits and instincts in degrees that change from one individual to the other.8 For several reasons one can argue that the orthodox approach eliminates the most important economic problems from Economics and that, in spite of its growing formalism, it is logically contradictory. In the first place, the diversity of different economic systems is accommodated by an approach where the individuals share basically the same characteristics (which are therefore also assumed to be independent of the social and economic structures those individuals inhabit). Indeed, one should ask why economic systems should be substantially different if individuals are always perfect maximizers. In many cases, we should only observe the efficient system built by perfectly rational agents. Indeed, it is hardly surprising that much of orthodox Economics comes to this conclusion and eliminates the possibility of understanding the ‘inefficient systems’ that make a great deal of the reality of history. In the second place, if we assume equally maximizing individuals, we ignore the fundamental economic problems related to the fact that individuals have different abilities to maximize and that many resources must be invested in order to turn them into reasonably good choosers. The existence and the development of entrepreneurial ability can hardly be incorporated into modern Economics. Moreover, the orthodox

256

Ugo Pagano

approach contradicts the daily experiences of parents, teachers and other agents who are engaged with the education process. These agents regularly make choices on behalf of those being educated who do not yet have sufficient maturity to deal with these choices. Perhaps the main purpose of education is the gradual achievement of a maturity that, in a more or less distant future, can make the individuals undergoing education reasonably good rational choosers. Or, in other words, one purpose of education is to enable individuals to cope with their own instincts and emotions and to help them to become aware of their own bad habits (and possibly to get rid of them!).9 However, in order to make individuals good choosers, it may be necessary to limit their freedom of choice, especially at the early stages of their life. These major economic problems are simply removed if all the agents are born (and die) equally rational in all societies. Major aspects of the education and socialization process that characterize different societies are also similarly expelled from the field of economic analysis. At the normative level, the effects are even less satisfactory. The traditional approach ignores the fact that the State should guarantee that each citizen receives the basic inputs necessary to make her a reasonably rational chooser. In this respect the role of public education is completely undervalued in Economics. Few arguments can be applied in Economics against the freedom for parents to have the type of education that, like fish and chips, they desire for their children. In Economics one does not perceive that, in a genuinely democratic society, the inalienable right of individuals to become free choosers involves a duty to provide the education necessary for people to exercise this right.10 The inalienable right (and duty towards the other members of society) of each individual to become a decent free chooser limits the freedom of choice of education by parents and involves a form of collective responsibility. Orthodox Economics cannot examine the mechanisms of allocation of resources by which a society that gives an important role to free choices reproduces its complex structure. An approach that assumes a uniform capacity to maximize cannot be a useful way to devise the policies that can help individuals to become reasonably good masters of their own lives. Finally, when we accept that maximization is costly, either the use of this hypothesis becomes trivial and tautological or it turns out to be logically contradictory. On the one hand, one may explicitly ignore the different maximization costs faced by the different individuals. In this case the maximization hypothesis is tantamount to saying that, if tomorrow petrol will not be scarce, an unbounded quantity of petrol (rationality) may be consumed. This tautological statement is not helpful in a world where both petrol and rationality are scarce factors. On the other hand, one may try to include maximization costs in the maximization problem itself and reformulate a new (second order) maximization problem, in which one rationally decides how many

Economics of Institutions and Institutions of Economics 257 resources to allocate to the (first order) maximizing activity and how many resources to dedicate to other useful activities. Unfortunately, this involves a new (more complex) maximization problem that entails new (and greater) maximization costs. One could go on ad infinitum and reformulate a new (third order) maximization problem that includes the problem to decide how many resources to devote to a (second order) maximization problem,11 in which one decides how many resources to devote to a (first order) maximization problem and how many resources to allocate to the other productive activities. But this involves an even more complex (fourth order) maximization problem and so on. This infinite regress makes the treatment of bounded rationality within a maximization framework logically contradictory. Only an unbounded outside observer (God?) could tell us what is the best course of action for us given the constraints imposed by our own bounded rationality. By contrast, ordinary human beings will find this redefinition of the maximization problem much harder to solve than the standard optimization problem: from a formal point of view the problem is the same but for the additional complications due to the inclusion of additional constraints expressing the bounds of our own rationality! To sum up: while reality is made up by different individuals each of whom has a different mix of instincts, emotions, habits and some specific bounded capacities for intentional rationality, Economics has mainly studied undifferentiated, unbounded maximizers.12 In this situation the questions concerning how individuals’ degrees of (ir)rationality and other aspects of complex individual behaviour are influenced and, in turn, influence the structures and the institutions of society cannot be properly addressed. The dominant approach is useful only in a few limited cases when the maximization problem is not very demanding of people’s rationality or when one wants to study the situations where individual rationality is (or is not) compatible with collective rationality.13 However, in the large majority of cases the approach cannot provide insights into the ways in which society is able to reproduce itself and the cases in which this reproduction of society may break down. An open reality where new types of individuals characterized by different mixes of instincts, emotions, habits and rationality can always emerge has been replaced by a closed world (or, at least, by a more closed world) populated by individuals characterized by an identical utility maximizing behaviour. Two major topics in Economics should be the study of the multiple ways in which social structures condition and mould individuals and the analysis of the many routes through which individual actions reproduce and, sometimes, transform social structures. By contrast Economics has only been useful for studying a limited number of problems such as the possible inconsistencies between the behaviour of equal-power total maximizers and the maximum social outcome that they could achieve.

258

Ugo Pagano

Why has Economics usually dealt with such limited (and often unreal) aspects of social interaction and individual behaviour? There are many ways in which one can defend the orthodox approach. In some relatively simple cases individuals may be able to maximize. In a limited number of cases the ‘logical’ contradictions between full individual rationality and full collective rationality contain many challenging intellectual puzzles and can provide some insights into the nature of real processes. However, it is difficult to defend the abnormal growth in these problems and to ignore the irrelevance of many of the exercises even in their own terms. It is even harder to justify the simultaneous underdevelopment of the understanding of the diversity and the evolution of real economic life. In other words, it is difficult to accept the extent of the replacement of the understanding of real economic structures and real individuals by the study of an unreal world of equally perfect rational maximizers.

The ‘Economics of Economics’ and the ‘oversupply’ of mathematics A paradoxical consequence of the orthodox approach is that Economics is defined more in terms of its methodology than in terms of its subjectmatter or field. In this sense one can claim that whatever phenomenon has been explained in terms of rational choice it also has, by definition, been explained in terms of the economic approach. In this sense Economics becomes a methodology that can be applied well beyond its original domain of enquiry. Rational choice can be applied to law, political science, animal behaviour and so on. In this way the ‘economic approach’ can conquer the fields of law, political science and animal behaviour. However, while this sort of economic imperialism takes place, the domain of study of social and economic structures is invaded by other disciplines and methodologies. The very fact that Economics has been defined in terms of a methodology and not in terms of its subject-matter has had the undesirable consequence that economists have often lost their own field of studies. It has also had the unfortunate consequence that scientists who are usually aware of the advantages of competition should not fail to appreciate: defining a discipline in terms of a methodology, and not in terms of its subject-matter, places severe limits on competition between different methodologies. Competition among methodologies requires that they measure their ability to explain the working of their subject-matter. In this respect, while Economics (or more accurately rational choice theory) can be applied to various fields, it is completely sheltered from the competition of other methodologies. By contrast, intentional rational explanations should compete with other explanations such as those based on the evolutionary selection of unconscious habits. Moreover, each

Economics of Institutions and Institutions of Economics 259 explanation should compete with the other relative to the understanding of a given problem of a given field and in a certain context.14 If according to the orthodox definition scientists seeking different types of explanations for the same phenomenon do not count as economists, Economics becomes some sort of Methodist (or Methodological) church. Competition from alternative methodologies is ruled out by the fact that their followers are necessarily outside the church. Our questions can now be formulated in a sharper way. Why does Economics concentrate so excessively on the study of abstract interactions among individuals as rational choosers? Why does it consider as lying outside its church scholars who seek to explore other aspects of individual interactions? Why do PhD programmes tend to include increasingly sophisticated mathematics, while economic history and the history of economic thought have stopped being topics in the same programmes? Why has Economics become so biased in one direction? In my opinion the answer can only be found by analysing how the actual institutions of production of Economics work,15 in other words, by some sort of ‘Economics of Economics’.16 If orthodox economic theory is correct to maintain that, at least up to a certain degree, individuals are selfish rational choosers, then economists cannot be assumed to be the only exception to this picture of human nature. A contradiction between the interests and actions of ‘selfish economists’ and what Economics is supposed to do may, therefore, easily arise. While the welfare of society at large demands that the community of economists is supposed to study problems that are relevant to a certain field of enquiry, namely the economy, economists are (also) interested in job security, tenure and their careers. This is even more so for a community which (unlike that inhabited by Ricardo, Marx, Walras or Pareto)17 operates almost entirely within a world of incentives defined by academia. Such a community is inevitably (over)sensitive to academic screening procedures and to the rules of the competition occurring among its members. We have already seen that social structures can be defined in terms of ‘positions’ that the different members occupy with respect to one another. This implies that agents do not get their utility only from the public and private goods that they consume and the type of work activities that they carry out.18 They are also likely to get utility from their relative positions in society. In many cases, because of their positional characteristics, these relations involve the consumption of positional goods such as power and status that are characterized by the fact that their aggregate consumption is equal to zero because the positive consumption by some individuals is matched by the negative consumption of other individuals. The case of positional goods is polar to the case of public goods. Public goods differ from private goods because, while in the case of ‘pure’ private goods an individual consumes a zero amount of a good consumed by another

260

Ugo Pagano

individual, in the case of ‘pure’ public goods an individual consumes the same positive amount of the good supplied and consumed by another individual. By contrast, in the case of ‘pure’ positional goods, an individual consumes the same negative amount of the good consumed by another individual. For instance, one can consume positive amounts of status and power only if other agents consume corresponding negative amounts. The output of research – knowledge – has a strong, well-known public good aspect. However, when knowledge can be easily embodied in marketable private goods or when its usefulness for the production of private goods can be easily described and patented, then its potential public good character is not a very serious problem. In this case, the traditional market incentives can work rather well. This is considerably more difficult when knowledge is basic knowledge that is far enough from marketable goods. Here the university and, in general, incentive structures that do not rely on the sale of a good have a considerable relative advantage. This is the case when one thinks of the basic structure of society. The knowledge can be extremely useful but it is difficult or impossible to embody it in private goods and to patent it. Thus, the people involved in the production of this type of goods must either (a) be politically motivated and/or (b) enjoy the production process as such or (c) (also) be paid by some public body like a university. While some combination of these ingredients is possible and, indeed, common, in recent times much of the production of knowledge is carried out within a scientific community working at a university or in other similar institutions. In turn, these institutions rely very much on the fact that people can be motivated to do research by offering them positions and positional goods. Ranks and careers, prizes establishing the relative fame of researchers, and access to journals with different levels of prestige are all ways by which in principle a public good such as ‘understanding the economy or the society’ could be achieved. An ‘Economics of Economics’ can be very useful in investigating the real structure of scientific research. Paradoxically, parts of orthodox Economics can here become useful because, at least at first sight, the problem might be seen as one of inconsistency between collective and individual rationality. While the collective interest of society and the collective reputation of the profession imply that Economics should improve as much as possible the understanding of a field of problems, the individual economists are (also and, sometimes, mainly) interested in surviving and, possibly, winning the tough positional competition that takes place within the profession. Thus, while a mix of skills is useful to advance economic research, there is no guarantee that academic economists will acquire them with the corresponding ‘optimal weights’. In particular because of the (sometimes tough) positional competition existing within the profession, (especially young) people will try to

Economics of Institutions and Institutions of Economics 261 acquire those skills that can be easily assessed by the (often senior) screeners. Members of appointment committees and candidates have convergent interests to minimize screening and signalling costs and will privilege those skills where the ability of individuals can be easily assessed. Let us now draw the implication for Economics of this standard economic argument for the ‘Economics of Economics’. Different types of skills are characterized by different screening and signalling costs. Usually the screening and signalling costs are lower when it is possible to define problems that admit only one correct answer; by contrast they are higher when the problem is difficult to define and admits many different interpretations. Mathematics is relatively close to the first extreme, while history is close to the second (while other fields may fall in between these two cases).19 It is then possible to formulate, paradoxically also in mathematical jargon, the ‘theorem’ that mathematics is oversupplied.20 Because of the cheaper screening and signalling properties of mathematics, rational choosers will put extra effort into acquiring these types of skills (relative to a ‘first best’ solution where complete information makes the signalling and screening costs irrelevant). One could argue that this ‘second best’ solution is, in fact, a good choice for our profession. In a world of costly and incomplete information,21 comparisons with the ‘first best’ solution are somehow irrelevant and the second best is the most that can be actually achieved. However, this ‘realistic defence’ does, in turn, ignore many aspects of reality. In the first place, it is very unlikely that the ‘oversupply’ of mathematical skills relative to other types of skills is limited to the level entailed by the ‘second best’ solution. Positional competition among economists may well imply that an inflationary spiral will take place and the strength of the mathematical signal is continuously reinforced at the expense of the other skills. In the second place, while the agents will concentrate on certain skills because they are better for signalling their abilities, they will always try to pretend that this is terribly useful for a better understanding of the real structures that the scientific community is supposed to understand. Even more important, they will find that believing it (and genuinely believing it!) makes them better off. It is rather sad to believe that one acquires certain skills only to win positional competition. It is nicer to get good positions while, at the same time, believing and making others believe that what one is doing is tremendously useful. In the third place, some adverse selection takes place. Those economists who are more ready to invest in the signalling skills and have little interest in the ultimate purposes of the scientific community will tend to win the most attractive positions. Their success will, in turn, bias the average attitude of the profession.

262

Ugo Pagano

Finally, because the process is largely a tacit and unconscious one, new habits of thought emerge and the problems that one was set to explain are easily forgotten. Indeed, economists who have signalled their abilities by oversupplying mathematics will have a preference (and, in the absence of other acquired skills, a necessity) to screen even more on the basis of the skills that they have accumulated. The ethos of the profession changes because ‘not only are individuals’ choices of actions conditioned by the situated options which they perceive’ but also because the social context conditions ‘the individuals themselves, their expressions of their needs and motives, the manner in which their capacities and capabilities have been moulded, their values and interests and so forth’ (Lawson 1997: 187).22 New generations are ‘born and developed’ as economists in a different way and they (re)produce Economics without being aware of the extent to which the Institutions of Economics may have affected its contents.

Conclusion At this point one may try to answer our question. Why have economists concentrated so much on the maximization hypothesis in spite of the serious shortcomings that make this assumption so hard to justify? We may advance the hypothesis that, while it is hard to argue that the individuals described by economists have the required optimization skills, this description allows economists to show that they are the ones who can solve complicated optimization problems. In many cases the role of the maximization hypothesis is not really to portray the actual behaviour of agents but rather to showcase the mathematical abilities of the economists. While a limited amount of mathematics can be useful,23 its potential as a screening and signalling device may paradoxically be a serious problem because it causes an overapplication of mathematics and a possible degeneration of Economics into an increasingly complex and empty formalism. If social science wants to have a liberating role in helping to understand, and sometimes even change, those structures that may otherwise be hidden to individuals, it must also study how its own structures work and the real tendencies that they may generate. Critical realism has claimed that ‘structures’ are ontologically distinct from people and can influence their actions and their way of thinking. It has also maintained that the methodology of Economics has emphasized formalisms that are not really suited to an ‘open system’ such as the economy. The ‘Economics of Economics’ seems to show the interdependence of these two claims: the Institutions of Economics are ontologically distinct from individual economists and tend to generate a ‘formalistic’ propensity.24 A future task of the Economics of Institutions should be to contribute to the analysis of the possible changes that may

Economics of Institutions and Institutions of Economics 263 help to generate better Institutions of Economics and, in general, a better division of labour among social scientists.

Notes 1 2

3

4

5 6

7

I thank Massimo D’Antoni, Tony Lawson, Antonio Nicita and Paul Lewis for useful comments. I am grateful to the University of Siena PAR project for funding this research. It is because the social structures which form the context for current agency at any given juncture in time are inherited ready-made by agents that critical realists draw the ontological distinction between pre-existing social structures and current actions. The picture of social activity to which this gives rise holds that human intentional agency must be understood as acting upon (reproducing or transforming) pre-existent structures, not as creating them ex nihilo (Lewis and Runde 2002: 203). Lawson observes how ‘it is not an exaggeration to suggest that most accounts in economics that explicitly focus upon the agency–structure relation veer towards one or the other or both. That is, either (1) structure is reduced to (is conceptualised as) the mere creation of individuals, or (2) agency is reduced to (is conceptualised as being totally determined by) external, coercive, structure’ (Lawson 1997: 167). Lawson sees Hayek and Veblen as two authors who tend to make these two opposite mistakes. Pagano (2000a) argues that, while Veblen gave an impressive contribution to the understanding of human agency, he lapses into some sort of technological determinism because of his uni-linear interpretation of history. Lawson (1997: 16). Hohfeld (1919) defines the logical relations among rights, duties, liberties and exposures to liberties. He defines also the ‘second order relation’ among powers, liabilities, disabilities and immunities. Commons ([1924] 1974) defines their ‘equilibrium’ and ‘disequilibrium’ relations. On this point see Pagano (2000b). This inhibition of research into the Economics of Institutions has been counteracted by a recent surge of interest in Institutional Economics. See, for instance, Bowles (2000) and Aoki (2001). Lawson observes that rational choice must be integrated with routinized behaviour, tacit knowledge, unconscious motivation, emotions and the fact that ‘each individual is primarily the product of his or her actions and experiences within the social relations and modes of determination into which he or she is born and thereafter lives’ (Lawson 1997: 185). This integration brings about what Lawson calls a theory of situated rationality: ‘Not only are individuals’ choices of actions conditioned by the situated options which they perceive, but also the individuals themselves, their expressions of their needs and motives, the manner in which their capacities and capabilities have been moulded, their values and interests and so forth are conditioned by the context of their birth and development’ (Lawson 1997: 187). According to Lawson (1997: 106) and Runde (1999: 74) rationality is a capacity which is only sometimes actualized because it can be overcome by other forces. Pagano (2000a) makes a similar point and distinguishes among five types of bounded rationality: bounded communication skills, bounded information processing skills, bounded calculation skills, bounded preference formation skills and bounded emotional skills. In all five cases the capacity to be rational is limited by and must co-exist with other forces. Screpanti (2000: 96) argues that in some way the Marxian theory makes a similar reductive assumption: ‘As an ontology of the social being, the Marxist

264

8

9 10

11

12

13

14

15

Ugo Pagano

anthropology gives rise to a kind of humanism that is different from the neoclassical one.’ However, Screpanti observes how ‘in a fundamental sense it also is a humanism quite similar because it is based on rationality’. Moreover, a deeper point is that: ‘If rational behaviour is to be assumed, then its evolution has to be explained’ (Hodgson 1998: 189). In this sense evolutionary psychology can help to understand the emergence of conscious and intentional behaviour as well as clarify the way in which social structure can mould individual behaviour (Hodgson 1999). At the same time, rationality should not involve exaggerated forms of selfrepression! This point is related to Sen’s capability approach even if perhaps choosing is in Sen an activity that has a status different from the other capabilities. On this point see Foster and Sen (1997) and Basu and Lòpez-Calva (2003). A stimulating assessment of Sen’s capability approach is provided by Robeyns (2000). The study of the allocation of human energies between ‘first order’ and ‘second order’ choices is considered in Sunstein and Ullman-Margalit (2000). Of course the problem that they face is a ‘third order’ problem. The logical contradictions that arise when one tries to solve these problems within a maximization framework are considered by Pagano (2000a). Vanberg (2000) proposes an interesting alternative to the traditional maximizing approach based on Mayr’s (1988) programme-based behaviour. He observes how this approach could put an end to the terrible isolation of Economics from the most stimulating cognate disciplines. Cognitive science, evolutionary biology, evolutionary epistemology, evolutionary psychology, the study of adaptive complex systems could all enrich our view of decisionmaking processes if we had not assumed that individuals were all equally good maximizers. The isolation of Economics may perhaps put some pressure on economists to reconsider the limitations of the maximizing framework. The ‘closed world’ of unchanging utility maximizing individuals can tell us some stories about the conflicts between individual and collective superrationalities. However, the fact that individuals are a complicated mix of instincts, emotions, habits and intentional rationality implies that the conflict between individual behaviour and collective interest is, at the same time, more and less serious than the economic approach may imply. While the failure of individual rationality can, in many cases, make desirable collective outcomes harder to achieve, people’s reliance on habits and norms rather than pure instrumental rationality makes it less likely that (once achieved) collectively desirable outcomes will be upset by continuous calculating selfinterest. By contrast, the perfect maximizers of Economics are somehow made independent of context by their invariant perfection. For this reason if ‘laws of market are discovered, these are the same in any society, in any historical setting and independent of the past development of that society’ (Elkana 2000: 11). A similar explanation can be found in Gellner’s description of ‘ultrasubjectivism’ (Gellner 1997: 46), which, like ‘mathematical formalism’, may lead to a disregard of the structures of a given field: ‘In the world’s most developed countries, something like 50 per cent of the population receives higher education. The colleges and universities which provide it are staffed by people who are assessed in terms not merely of teaching performance but also of intellectual creativity and originality, on the model of an ever-growing natural science, and of great centres of learning where scholars find

Economics of Institutions and Institutions of Economics 265

16

17 18

19

20

21

themselves on the very frontiers of knowledge … It is all intended to resemble scientific growth. But what if there isn’t any? May this lead to a setting up of artificial obsolescence and rotation of fashion, characteristic of the consumer goods industry? In the post-war era, this demand for growth was met in American sociology by the elaboration of a scientistic jargon, which in fact had neither sharpness of definition nor any real relation to reality, nor much internal discipline, but which sounded suitably obscure and intimidating. This was followed, in anthropology, first by the “interpretative mood”, and then by its exaggerated, self-indulgent “post-modernist” continuation. Each could be presented as discovery and advance.’ In some way this can also be seen as a useful reflexivity test. If Economics works, it must also help to explain itself with a consistent approach. The importance of a ‘reflexivity test’ is considered by Mäki (1999), who argues that, in principle, the tools by which we explain the market for goods should also help the understanding of the market for ideas. However, in this case he ends up correctly arguing that the market for goods and the market for ideas may be two very different institutions and a theory that fits the former might not necessarily fit the latter. Ricardo was a professional businessman (and politician) and Marx a professional revolutionary, while Walras and Pareto became academics in Lausanne late in their lives after being employed on the railways. Work only appears in the traditional setting as ‘forgone leisure’, an unreal assumption that implies that individuals are indifferent towards work allocation. On the extension of utility to the real work activities performed by individuals, see Pagano (1985). While I have not been able to find a reasonable explanation for the exclusion of economic history from PhD programmes, the argument for the exclusion of the history of economic analysis is based on the idea that Economics is characterized by a cumulative growth of knowledge that makes it unnecessary to study past theories. Unfortunately this is not true. A more reasonable explanation is that the history of economic analysis is also a poor screening and signalling device. Moreover, in a world characterized by fierce positional competition, being a dead economist is a very serious disadvantage. There is no way in which dead economists can gain influence by offering positions to other academics! Saying that mathematics is ‘oversupplied’ does not imply that it is always useless but that in many cases it is supplied well beyond the point where it is helpful. The situation is well summarized by Peter Bauer (2000: 21) when he writes: ‘Mathematical methods often provide an effective facade or screen which covers or conceals empty formalism. They can camouflage disregard of basic propositions or simple evidence in models purporting to serve as a basis for policy. Statistics, technical jargon, and sophisticated econometric techniques can also serve as a protective screen. But the use of mathematics is particularly effective because of the language barrier it provides. What we see is an inversion of the familiar Hans Andersen story of the Emperor’s New Clothes. Here there are new clothes, and at times they are haute couture. But all too often there is no emperor within.’ A similar point is made by Lewis and Runde (1999). The argument is similar to that advanced in the numerous models where agents have to signal their abilities. In a seminar at Cambridge University, Frank Hahn once commented upon one of these models (where the best agents had to follow an inferior technique to reveal their superior abilities) by saying: ‘This is why the best economists do the worst economics!’

266

Ugo Pagano

22 The point is not made by Lawson referring to this specific context but applies rather well to this case. 23 In many cases mathematics can be one way of using as metaphors the relations existing in other fields of reality. According to the original Greek meaning, metaphors involve the transfer of concepts from some wellunderstood field of investigation to some other, less well-understood domain of inquiry. This system works insofar as it is possible to draw an analogy between the two fields and is considered as a first step towards a deeper understanding of the new field. However, the existence of such analogies can never be taken for granted and, where the relevant analogies are absent, metaphors can sometimes be misleading. They are useful only insofar as there is a sufficient symmetry between the real structure that we know well and the real structure about which we know very little. When these conditions are satisfied, metaphors can ‘provide the linguistic context in which the models that constitute the basis for scientific explanation are suggested and described’ and can allow scientists to draw upon antecedently existing cognitive resources to provide both the model and the vocabulary in terms of which the unknown mechanism etc. governing observable behaviour can be conceived and investigated’ (Lewis 1999: 97–8). 24 While this propensity is a real outcome of the structure of academic work, it may be (we do hope often!) matched by other propensities. On the meaning of propensities and their role in Popper’s work, see Runde (1999: 77) where he argues that, as Popper (1959) remarked, a good part of the usefulness of the propensity view lies in the suggestion that our theories are concerned with an unobservable reality of causal factors, generative mechanisms and so on, and that it is only through some of its more superficial effects (propensities that have been realised) that this reality can be identified.

References Aoki, M. (2001) Towards a Comparative Institutional Analysis, Cambridge, MA: MIT Press. Basu, K. and E. Lòpez-Calva (2003) ‘Functionings and Capabilities’, forthcoming in K. Arrow, A. Sen. and K. Suzumura (eds) Handbook of Social Choice and Welfare, Volume 2, Amsterdam and New York: Elsevier Science-North Holland. Bauer, P. (2000) ‘Disregard of Reality’, in P. Bauer, From Subsistence to Exchange and Other Essays, Princeton, NJ: Princeton University Press. Bowles, S. (2000) Economic Institutions and Behaviour: An Evolutionary Approach to Microeconomic Theory, unpublished manuscript. Commons, J.R. ([1924] 1974) Legal Foundations of Capitalism, Clifton: Augustus M. Kelley Publishers. Elkana, Y. (2000) Rethinking – Not Unthinking – the Enlightenment, mimeo, Budapest. Foster, J. and A. Sen (1997) ‘On Economic Inequality after a Quarter Century’, in A. Sen, On Economic Inequality, expanded edition, Oxford: Clarendon Press. Gellner, E. (1997) Postmodernism, Reason and Religion, London: Routledge. Hodgson, G.M. (1996) ‘Organizational Form and Economic Evolution: A Critique of the Williamsonian Hypothesis’, in U. Pagano and R. Rowthorn (eds) Democracy and Efficiency in the Economic Enterprise, London and New York: Routledge.

Economics of Institutions and Institutions of Economics 267 —— (1998) ‘The Approach of Institutional Economics’, Journal of Economic Literature, 36: 166–92. —— (1999) ‘Structures and Institutions: Reflections on Institutionalism, Structuration Theory and Critical Realism’, mimeo, University of Hertfordshire. Hohfeld, W.N. (1919) Fundamental Legal Conceptions, New Haven and London: Yale University Press. Lawson, T. (1997) Economics and Reality, London and New York: Routledge. Lewis, P.A. (1999) ‘Metaphor and Critical Realism’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. —— and J.H. Runde (1999) ‘A Critical Realist Perspective on Paul Davidson’s Methodological Writings on – and Rhetorical Strategy for – Post-Keynesian Economics’, Journal of Post-Keynesian Economics, 22: 35–56. —— and J.H. Runde (2002) ‘Intersubjectivity in Economics: A Critical Realist Perspective’, in E. Fullbrook (ed.) Intersubjectivity in Economics: Agents and Structures, London and New York: Routledge. Mäki, U. (1999) ‘Science as a Free Market: A Reflexitivity Test in an Economics of Economics’, Perspectives on Science, 7: 486–509. Mayr, E. (1988) Towards a New Philosophy of Biology – Observations of an Evolutionist, Harvard: Harvard University Press. Pagano, U. (1985) Work and Welfare in Economic Theory, Oxford: Basil Blackwell. —— (1995) ‘Can Economics Explain Nationalism?’, in A. Breton, G. Galeotti, P. Salmon and R. Wintrobe (eds) Nationalism and Rationality, Cambridge: Cambridge University Press. —— (1999) ‘Is Power an Economic Good? Notes on Social Scarcity and the Economics of Positional Goods’, in S. Bowles, M. Franzini and U. Pagano (eds) The Politics and the Economics of Power, London: Routledge. —— (2000a) ‘Bounded Rationality, Institutionalism and the Diversity of Economic Institutions’, in F. Louca and M. Perlman (eds) Is Economics an Evolutionary Science? The Legacy of Thorstein Veblen, Cheltenham: Edward Elgar. —— (2000b) ‘Public Markets, Private Orderings and Corporate Governance’, International Review of Law and Economics, 20: 453–77. Popper, K. (1959) ‘The Propensity Interpretation of Probability’, British Journal for the Philosophy of Science, 10: 25–42 Robeyns, I. (2000) ‘An Unworkable Idea or a Promising Alternative? Sen’s Capability Approach Re-examined’, Centre for Economic Studies Discussion Paper No. 30, University of Leuven. Runde, J.H. (1999) ‘On Popper Probabilities and Propensities’, in S. Fleetwood (ed.) Critical Realism in Economics: Development and Debate, London and New York: Routledge. Screpanti, E. (2000) ‘The Postmodern Crisis in Economics and the Revolution against Modernism’, Rethinking Marxism, 12: 87–111. Sunstein, C.R. and E. Ullman-Margalit (2000) ‘Second Order Decisions’, in C.R. Sunstein, Behavioral Law and Economics, Cambridge: Cambridge University Press. Vanberg, V. (2000) ‘Rational Choice and Rule-Based Behaviour: Alternative Heuristics’, in R. Metze, K. Mühler and K. Opp (eds) Normen und Instituzionen: Entstehung und Wirkungen, Leipzig: Leipziger Universitätsverlag.

14 A note on critical realism, scientific exegesis and Schumpeter Mário da Graça Moura

I Introduction The purpose of the present chapter is to examine whether the framework and results of critical realism, as presented in Tony Lawson’s writings (see, especially, Lawson 1994, 1997), can be of help for the practice of the history of economic thought. As an illustration of my argument, I shall focus on the interpretation of Joseph Schumpeter’s work, thus complementing other work of mine (Graça Moura 2002, 2003).1 An important characteristic of Schumpeter’s substantive writings is their protean quality. They are marked by tensions and even inconsistencies, as I suggest in section II. Whilst this may seem to imply a negative evaluation of his achievements, I believe that, on the contrary, this feature of Schumpeter’s work enhances its importance. But, of course, it also enhances its complexity. Indeed, the kaleidoscopic sweep of Schumpeter’s writings poses an obvious interpretative problem. How can we understand texts with tensions and inconsistencies therein? Two alternative rules can be followed. If such tensions and inconsistencies as have been detected do not inhibit the identification of ‘main analytical conclusions’, then the method labelled the ‘principle of scientific exegesis’ (Stigler 1965) can be accepted. This method involves discarding tensions and inconsistencies so as to arrive at a (coherent) set of concepts and relationships between concepts, with a view to maximising the value of an author’s contribution to science. Just as this method cannot be ruled out a priori, however, neither should it be taken as a generally admissible procedure. For the purpose of rendering Schumpeter’s discourse intelligible, as I suggest in section III, the principle of scientific exegesis is certainly inappropriate. The alternative is to attempt to make sense of Schumpeter by explaining why tensions and inconsistencies arise in his writings. And, as I propose in section IV, it is here that critical realism is particularly helpful. For critical realism, in highlighting ontology as a category which it is impossible to eliminate, also explains why nonetheless methods are often selected a priori, and the

Scientific exegesis and Schumpeter

269

potential for inconsistency that consequently arises. Concluding remarks follow in the final section.

II On the kaleidoscopic character of Schumpeter’s writings: an illustration A recurrent theme in Schumpeter’s work is the claim that the mechanism of innovation displays a tendency to metamorphose in the course of capitalist history. The combination of two meta-innovations, the largescale corporation and industrial R&D, would herald the supersession of the method of leadership selection through the foundation of firms which had been the hallmark of ‘competitive capitalism’. Moreover, the R&D laboratory and the corporate bureaucracy would lead to a progressive ‘automatisation’ of innovative activity, ultimately rendering the entrepreneur obsolete: Innovation in competitive capitalism is typically embodied in the foundation of new firms … The new processes do not, and generally cannot, evolve out of the old firms, but place themselves side by side with them and attack them. Furthermore, for a firm of comparatively small size, which is no power on the money market and cannot afford scientific departments or experimental production and so on, innovation in commercial or technical practice is an extremely risky and difficult thing, requiring supernormal energy and courage to embark upon … All this is different in ‘trustified’ capitalism. Innovation is, in this case, not any more embodied typically in new firms, but goes on within the big units now existing … Progress becomes ‘automatised’, increasingly impersonal and decreasingly a matter of leadership and individual initiative. This amounts to a fundamental change in many respects, some of which reach far out of the sphere of things economic. (Schumpeter [1928] 1989: 70–1) It would seem, then, that there are two Schumpeterian accounts of innovation, intended to represent different phases of the capitalist process: ‘competitive capitalism’, analysed in The Theory of Economic Development (Schumpeter 1912, 1934), and ‘trustified capitalism’, described in Capitalism, Socialism and Democracy (Schumpeter 1942) but also in much earlier works such as ‘Sozialistische Möglichkeiten von heute’ (Schumpeter 1920/21). Indeed, this perception has sometimes led to the explicit formulation of two Schumpeterian models of innovation, which have become known as Schumpeter Mark I and Schumpeter Mark II (see, e.g., Phillips 1971: 4ff. and Freeman et al. 1982: 38ff.). In the former, there exists an ‘exogenously given’ set of inventions, or pieces of knowledge, the potential of which remains unexplored. ‘In the

270

Mário da Graça Moura

beginning’ there is also a competitive equilibrium. This placid situation is disturbed by entrepreneurial pioneers: ‘outsiders’ who, with ‘venture capital’ provided by a complaisant banking system, apply new (or as yet unused) knowledge to industry and trade and (if successful) make profits. However, because this knowledge is publicly available and its application easily reproducible, imitators emerge. They set in motion a tendency to the erosion of innovational quasi-rents and to the re-establishment of a competitive equilibrium, from which new entrepreneurs restart the process of innovation. In contrast, in Schumpeter Mark II bureaucracies have replaced entrepreneur-owner-managers and invention is endogenous to the R&D laboratory of the ‘large corporation’. Dimension and market power endow large firms with advantages in the production and use of knowledge, in the appropriation of the returns to innovation, and in the acquisition of external finance. Simultaneously, a real or imaginary threat to their position provides the spur to further innovative efforts, which lead to the reproduction of their market power. Schumpeter’s writings on innovation have, of course, stimulated a huge body of research, which has made us aware of the shortcomings of Schumpeter Mark I and Schumpeter Mark II. It is clear, for instance, that neither model provides a satisfactory treatment of knowledge. Tendencies for innovators to reproduce their positions on the basis of idiosyncratic knowledge (and profits) are disregarded in Schumpeter Mark I – although locally specific knowledge makes a surreptitious appearance, to produce a sort of lock-in effect that tends to pre-empt innovative behaviour by established firms. As for Schumpeter Mark II, the relationship between corporate R&D and an exogenous science base is not specifically considered – in other words, the fact that knowledge creation is a cumulative as well as an interactive process is disregarded. In short, whilst the contrast between Schumpeter Mark I and Schumpeter Mark II suggests important features of capitalism’s transformation throughout the twentieth century – notably the fact that R&D has risen to a prominent role, and that firms have developed routines for the direction and support of innovative efforts – these models cannot be considered adequate representations of the complexity and variety of the capitalist process of innovation. But are they representations of capitalism at all? There are grounds for doubt – particularly with regard to Schumpeter’s writings on ‘competitive capitalism’. In his Theory of Economic Development, Schumpeter admittedly seeks to build on Das Wesen und der Hauptinhalt der theoretischen Nationalökonomie (Schumpeter 1908), his methodologically oriented treatise on statics, or ‘pure theory’. Indeed, his argument in the Theory is said to stand in relation to reality exactly as statics does (see Schumpeter 1912: 488). Yet the relationship between statics and reality turns out to be far from clear. In Wesen Schumpeter submits, for instance, that ‘free

Scientific exegesis and Schumpeter

271

competition not only exists never and nowhere but, taken in its theoretical sense, could not exist at all’ (Schumpeter 1908: 272).2 Rather, ‘free competition’ is an assumption devised in order to establish an equilibrium (Schumpeter 1908: 195–6), which in turn is indispensable to the ‘pure theory’ project (see, e.g., Schumpeter 1908: 33–4, 118). Schumpeter actually laments that ‘pure theory’ cannot handle the practically very important case where ‘monopolists’ confront each other, as their economic power and energy cannot be captured in an exact manner (Schumpeter 1908: 589). To the extent that Schumpeter is indeed building on his ‘pure theory’, equilibrium framework, then, it is hardly evident that his writings on ‘competitive capitalism’ refer to capitalism at all. But, to confuse things further, he also submits that ‘[c]ommon sense tells us that this mechanism for establishing or re-establishing equilibrium is not a figment devised as an exercise in the pure logic of economics but actually operative in the reality around us’ (Schumpeter 1939: 47). There is, however, yet another – perhaps related – complication. We have seen that some of Schumpeter’s statements do not have stable meanings throughout. But, in addition to this, Schumpeter Mark I is an interpretation of Schumpeter’s text. And in this text Schumpeter often contradicts substantive points made a few pages before. In the Theory, he repeatedly undermines his basic framework. For example, he represents production as a vertically integrated process where produced means of production are transitory items (Schumpeter 1934: 44), claiming that ‘the idea that somewhere there is an accumulated stock of such goods is absolutely false’ (Schumpeter 1934: 45). Nevertheless, he also observes that perfect adaptation to a changed state of affairs ‘is impossible, … especially in the case of durable producers’ goods’ (Schumpeter 1934: 33). And, of course, he repeatedly brings in industries producing means of production when presenting the business cycle implications of his theory.3 Actually, Schumpeter ends up transcending some of the limitations of (his argument as summarised in) Schumpeter Mark I. There are, for instance, many passages in which he is groping towards an adequate representation of the various flows of knowledge involved in the process of innovation. In Business Cycles he remarks, for instance, that ‘[e]lectrical technology had … become an applied science which it was possible to learn and to develop in laboratories and schools. A considerable part of the men who … rose to fill the … leading or semi-leading positions in the industrial organism … had that sort of training, at least to the extent of being able to understand and apply readily what was suggested by the scientific engineer, many being scientific engineers themselves’ (Schumpeter 1939: 440). He adds that ‘[v]igorous and varied enterprise thus went on under the almost immediate impulse of the technical departments of … concerns, the entrepreneurs being largely employees

272

Mário da Graça Moura

and monopoly positions of individual gadgets being incessantly won or lost’ (Schumpeter 1939: 440). There are ambiguities in the realm of Schumpeter Mark II as well. On the basis of Capitalism, Socialism and Democracy, it is not entirely clear how the process of innovation evolves in ‘trustified capitalism’. On the one hand, Schumpeter suggests that concentration tendencies develop as an industry matures and subsist whilst it is still ‘gaining ground’ (see Schumpeter 1942: 90 and fn.5; see also, e.g., Schumpeter 1939: 404, fn.2). This appears to point to a life-cycle mechanism, according to which (in post Nelson and Winter (1982) language) the definition of a technological regime involves a phase of radical innovation and low barriers to entry, which is then superseded as a technological trajectory is selected and established firms become the main innovative players – until the trajectory exhausts itself. On the other hand, however, Schumpeter writes that although the process of industrial concentration is less free from setbacks and compensatory tendencies than one might think, there can be little doubt about the long-run outcome (see Schumpeter 1942: 140). ‘Industrial integration is far from being complete,’ he observes (Schumpeter 1942: 163), ‘[c]ompetition, actual and potential, is still a major factor’ – but perhaps not for long. Related to this tension there is (amongst other things) a dualistic view of knowledge. In Schumpeter’s diagnosis of the process of creative destruction, uncertainty plays a major part: it entails that capitalism can only be evaluated as it unfolds over time, and that there is no rationale for a systematic anti-trust policy. The knowledge required to devise a policy that actually improves efficiency is simply unavailable: entrepreneurial decisions involve context-dependent ‘guesses’, which cannot be centralised. Despite his recognition that technological possibilities are ‘an uncharted sea’ (Schumpeter 1942: 118), however, Schumpeter argues that entrepreneurship becomes obsolescent as the capitalist process evolves. As knowledge accumulates and the R&D department takes over, calculation tends to replace imagination. And this leads more or less straightforwardly to his regrettable stance on the economic viability, indeed superiority, of centralised socialism.4 The implications of all these (barely touched upon) issues would, of course, demand a lot of elaboration. Still, the above sketch suffices to suggest that we are facing texts with tensions and inconsistencies therein. How, then, are we to understand them?

The principle of scientific exegesis George Stigler (1965), in a brief note, offers commendably clear guidelines in this respect. He is commenting on an interpretation of Ricardo. However, his comments are clearly intended to yield a general rule for ‘a problem often met in the history of economics: which passage in a man’s

Scientific exegesis and Schumpeter

273

writings do you accept when several passages are inconsistent?’ (Stigler 1965: 447).5 ‘Why should we allow the hand-picked quotation to carry an interpretation when we would reject the hand-picked fact as an empirical test of a hypothesis?’, Stigler asks. ‘In fact’, he argues, the two problems are basically the same. The substantive hypothesis seeks to isolate the main variables in the economic phenomena under consideration, and to formulate the major relationships among these variables. A successful hypothesis accounts for the important relationships in the appropriate data, but it need not account for random variation. Similarly, the textual interpretation must uncover the main concepts in the man’s work, and the major functional relationships among them. The interpretation need not account for careless writing or unintegrated knowledge … We increase our confidence in the interpretation of an author by increasing the number of his main theoretical conclusions which we can deduce from (our interpretation of) his analytical system. The test of an interpretation is its consistency with the main analytical conclusions of the system of thought under consideration. If the main conclusions of a man’s thought do not survive under one interpretation, and do under another, the latter interpretation must be preferred. (The analogy to maximum likelihood is evident.) This rule of interpretation is designed to maximize the value of a theory to the science. The man’s central theoretical position is isolated and stated in a strong form capable of contradictions by the facts. The net scientific contribution, if any, of the man’s work is thus identified, amended if necessary, and rendered capable of evaluation and possible acceptance. This rule of consistency with the main conclusions may be called the principle of scientific exegesis. Of course men make logical errors or slip into tautologies and otherwise blemish their work. One may seek to determine what the man really believed, although this search has no direct relevance to scientific progress. (Stigler 1965: 448)6 Predictably, Stigler’s principle of scientific exegesis is presented in the most straightforward and self-confident fashion – as if it were selfevident. Yet it is not so innocuous as it may superficially appear to be. Ultimately, it seems to be informed by a particular conception of science, and accordingly of progress in science, the merits of which will have to be considered. Let us start in a more common-sense manner, though. There are no doubt many cases in which the application of Stigler’s principle will do. But not always. His proposal actually turns out to be somewhat remarkable in that he is supposing ‘main conclusions’ to be readily available – or, which is the same, he is implicitly defining tensions and

274

Mário da Graça Moura

inconsistencies as uninteresting residuals. In other words, Stigler disposes of difficulties in interpreting a contradictory piece by tacitly assuming that a part of it is mere ‘noise’. The problem to be solved – what to do when certain passages are inconsistent – is ultimately sidelined rather than properly addressed. Now, is this procedure appropriate to deal with Schumpeter’s case? Clearly, this depends on the nature – peripheral or otherwise – of the tensions and inconsistencies in his writings, or on the possibility of identifying ‘main conclusions’. In the light of the previous section, I suggest, the prospects for such an endeavour do not look propitious. Is Schumpeter’s equilibrium-grounded theorising the centrepiece of his writings, as opposed, say, to his elaborations on the nature of creative destruction and on structural transformation? Or vice versa? Stigler implicitly demands that we decide, yet we have no clear basis for doing so. In order to see where the application of his principle could lead us, consider the following experiment. We take one of the pieces that purport to present Schumpeter, say, as a ‘Marxist’, on the basis of a set of supposed ‘main conclusions’7 – although his connection to Walras is acknowledged. This piece is competently done. Next, we take one of the pieces which – despite recognising some connection between Schumpeter and Marx, of course – seek to convince us that Schumpeter is essentially ‘Walrasian’. On its own terms, again, the piece will be persuasive. If, however, we confront both pieces, we are quite likely to be confused; and increasingly so as we pick more and more interpretative exercises of the same type. In juxtaposing interpretative exercises of this kind we are in fact reproducing Schumpeter’s tensions and inconsistencies in another key. Of course, there will not be many interpreters that deliberately follow Stigler’s principle of scientific exegesis. But a part of Schumpeterian exegesis nonetheless has a broadly Stiglerian flavour. This is because the mode of reasoning that Stigler is implicitly defending is, and has long been, predominant in economics – a mode of reasoning, or a conception of science, the limits of which will become apparent in the next section. Yet instead of implicitly assuming that tensions and inconsistencies are of marginal importance, or merely noting that they exist, we may attempt to make sense of Schumpeter by explaining why these tensions and inconsistencies arise.8 To that end, of course, we need a theory of inconsistencies; and it is here that critical realism is of considerable help.

IV A critical realist framework The theory of inconsistencies that I want to propose is based on the methodological framework developed by Lawson (1994, 1997), which constitutes the core of critical realism in economics. But this theory can be seen as a particular case of a more general framework, which it is useful

Scientific exegesis and Schumpeter

275

to introduce first. Indeed, this more general framework can be argued to build, if only up to a point, on Schumpeter’s own metatheoretical reflections. I shall therefore draw on Schumpeter as is convenient for my purposes. In fact, I start from his famous conjectures on ‘vision’, ‘analysis’ and the evolution of science.9 1 Vision and analysis – plus the irreducibility of ontology Schumpeter writes in his History of Economic Analysis that at the beginning of scientific efforts in any field there is a ‘vision’, a preliminary ‘picture of things as we see them’ (Schumpeter 1954: 42). Without some notion of a set of phenomena as a meaningful object there would be nothing to analyse. This vision is treated according to certain rules and, in the process of its elaboration into a scientific piece, is modified, whereas the methods by which it is processed and scrutinised tend to change, too. Yet scientific development is not a smooth process. On the contrary, it is punctuated by violent eruptions: ‘[o]wing to the resistance that an existing scientific structure offers, major changes in outlook and methods, at first retarded, then come about by way of revolution’ (Schumpeter 1954: 46). And vision in the sense defined above ‘re-enter[s] the history of every established science each time somebody teaches us to see things in a light of which the source is not to be found in the facts, methods, and results of the pre-existing state of the science’ (Schumpeter 1954: 41).10 Scientific analysis, then, ‘is not simply a logically consistent process that starts with some primitive notions and then adds to the stock in a straightline fashion’. It is ‘an incessant struggle with creations of our own and our predecessors’ minds and it “progresses”, if at all, in a criss-cross fashion, not as logic, but as the impact of new ideas or observations or needs, and also as the bents and temperaments of new men, dictate’ (Schumpeter 1954: 4). From Schumpeter we get, in sum, a picture of the scientific process that places considerable emphasis upon tensions and discontinuities. What, though, lies at the bottom of the scientific ‘struggles’ that he rightly underlines? In this regard – significantly, I suggest – Schumpeter is not sufficiently explicit. I want to argue that these ‘struggles’ ultimately reflect a conflict of ontologies. For a Schumpeterian vision entails a preliminary theory of (the nature of) what exists;11 and so does any method by which this vision is explored. This follows from the fact that every method presupposes a theory of knowledge, and that knowledge is necessarily knowledge of something: a theory of knowledge implicitly presupposes an ontology. This does not imply, of course, that theorists first reflect on the nature of reality and then go on to choose methods adapted to that reality. In any case, their efforts always start from the work of their predecessors as well. And, certainly, established scientific structures have various ways of

276

Mário da Graça Moura

reproducing themselves – which, in turn, leads to tensions between vintages of researchers, or rival ‘schools’. But, in a similar fashion, tensions may also arise within the work of individual theorists. For, clearly, theorists cannot discard their basic framework – whatever its origin and coherence – as a matter of course. In fact, it is inadequate to use the notion of choice here in that the recognition of an alternative framework as such already supposes a ‘conversion’ process. But, in any case, no framework can be definitive. Theorists may be ‘forced’ to devise ‘auxiliary hypotheses’ in order to cope with unforeseen problems – and, in so doing, they may (further) endanger the coherence of their framework.12 In short, it is not altogether surprising that there should be tensions and inconsistencies of a certain kind in scientific writings. Rather than mere ‘noise’, some tensions and inconsistencies could be a microcosmic reflection of clashes between ‘schools’. They could be attributable to the resilience of discourses with different ontological presuppositions – a resilience that is inherent to the evolution of science. 2 From philosophical empiricism to deductivism Explaining Schumpeter’s tensions and inconsistencies, I now want to suggest, requires us to introduce a particular pattern of ontological inconsistency – a pattern that results from a commitment to a certain mode of explanation. Following Lawson (1997: 16ff.), I shall label this mode of explanation deductivism. Its intellectual origin can be located in ‘philosophical empiricism’. Typically, Schumpeter provides an impeccable definition of this theory of knowledge – which, he submits, has not worn well as a philosophy (see Schumpeter 1954: 121). Philosophical empiricism, he writes, is the doctrine … developed mainly by English thinkers of the seventeenth and eighteenth centuries (especially by Hobbes, Locke, and Hume), that (a) all knowledge of the individual is derived through experience during his own life; (b) that his experience may be equated to the sense impressions to which his mind is exposed; (c) that prior to this experience his mind is not only a complete blank but … as such, ‘mind’ does not exist at all; (d) that the impressions are the ultimate elements into which all mental phenomena may be resolved, not only remembrance, attention, reasoning – including the construction of causal sequences – but also the affective ones, the ‘passions’: all these are but agglomerations of elemental impressions and produced by their random ‘associations’. This resolution of the human ‘mind’ or ‘soul’ into atomic impressions may be likened to the reduction of all physical phenomena to atomistic mechanics, a much

Scientific exegesis and Schumpeter

277

employed analogy that made empiricism popular to some and hateful to other people. (Schumpeter 1954: 120–1) From this definition and the previous considerations on the necessary dependence of theories of knowledge upon theories of reality, we can see, then, that ‘philosophical empiricism’ yields an ontology constituted by the category of experience. It posits that existential claims can be made only about the objects of experience. Accordingly, science must seek to identify constant conjunctions of atomistic events as they are experienced. Regularities of the ‘whenever event x then event y’ type are the only conceivable generalisations or ‘laws’ (Lawson 1997: 19). The widespread acceptance of ‘philosophical empiricism’ in social science seems to have been inspired by the success of natural science. Curiously, though, this theory of knowledge cannot explain why experimental results in natural science are successfully applied in situations where the ‘whenever x then y’ regularities experimentally obtained are not empirically manifest. In fact, to the extent that constant conjunctions of events are restricted to experimental conditions, ‘philosophical empiricism’ entails that laws of nature depend on the intervention of scientists, and that science is isolated from the empirical world (Lawson 1997: 27ff.). These paradoxes disappear, however, if being ceases to be implicitly defined by sense experience. It is only because of this illicit conflation that science is identified with constant conjunctions of events. Systems in which such conjunctions are ubiquitous are said to be closed. If, in contrast, the natural world is viewed as open and structured, in the sense that various, possibly countervailing forces govern, but are irreducible to, empirical phenomena – so that constant conjunctions of events do not typically obtain – then experimental activity, and its usefulness, are intelligible (Lawson 1997: 27ff.). Rather than constituting a law, a scientific experiment can be understood as an intervention designed to isolate, and so facilitate the empirical identification of, a mechanism that is at work both in the closed systems which scientists attempt to engineer in their experiments and in the open natural world. Now, if the social world, too, is open – which it must be for human choice to be real – then it follows that ‘philosophical empiricism’ cannot be an adequate conception for social science. It cannot be the case that a set of conditions x typically determines y, as this would mean the absence of choice. Indeed, choice implies not only that the social world is open but also that it is structured. For, as Lawson (1997: 30ff.) argues, choice supposes intentionality, and therefore some degree of knowledgeability, which in turn requires some endurability in the objects of knowledge. If constant conjunctions of events are rare, these objects of knowledge must lie at the level of structures which underlie, but do not pre-determine,

278

Mário da Graça Moura

human action – and which human action in turn reproduces or transforms. If choice exists, then, there is no reason why constant conjunctions of events should be pervasive in the social world; and significant conjunctions of this kind have not been found (see, e.g., Lawson 1997: 70). In fact, the programme of ‘philosophical empiricism’, rather than being strictly empiricist,13 underwent a partial change. The endeavour to establish constant conjunctions eventually acquired a purely formal character, in that only the structure of explanation of ‘philosophical empiricism’ survived: deductivism. In other words, the lack of empirically observable constant conjunctions of events was sublimated into the creation of constant conjunctions of the form ‘whenever conditions x then outcome y’ in a fictitious world. Simultaneously, efforts were made to interconnect the regularities so constructed, with a view to establishing their global coherence – an objective which, in turn, shaped the axiomatic structure from which regularities were to be deduced. But, like every mode of explanation,14 deductivism can only provide knowledge of a world of the sort which it implicitly presupposes. And the latter sort of world entails the impossibility of genuine choice. In so far as theorists acknowledge the capacity of choice, there is, then, a potential for inconsistency in their substantive work. Significantly, much of contemporary economics is marked by a schism between a discourse geared to analytical rigour, which excludes what cannot be subsumed within deductivism, and a contrasting, more fragmentary and informal discourse intended to address worldly matters. But this means that this particular co-existence of conflictual discourses does not primarily reflect the fact that science is necessarily work-inprogress, that it is impossible to achieve a theory of things ‘as they really are’. First and foremost, it reflects a commitment to a certain mode of explanation that is accepted on the basis of a (mis)conception of natural science tacitly taken for granted – i.e., independently of sustained reflection about its (prospective) empirical relevance for social science. This is why we are entitled to speak of a particular pattern of ontological inconsistency. 3 Returning to Schumpeter This pattern deserves to be highlighted because deductivism marked the intellectual atmosphere in which Schumpeter worked. Consider the following statement: The pioneers of modern academic sociology, anthropology and economics, Weber, Durkheim, Radcliffe-Brown, Malinowski, Menger, Jevons and Walras, all regarded subsumption under general laws as essential to scientific explanation, and took the discovery of such laws

Scientific exegesis and Schumpeter

279

to be the means for making the social sciences truly scientific. Commitment to the covering-law model has kept such goals alive in the face of continual disappointment. For example, when social anthropologists discovered that their fieldwork yielded few interesting general laws involving relatively concrete phenomena, such as motherhood or farming, they did not abandon the pursuit of general laws. Many responded by seeking such general relationships among more abstract structural characteristics, such as ‘binary opposition’. Many economists elaborate the internal logic of some general model, serenely accepting that their work makes no appreciable contribution to explaining specific episodes of inflation or unemployment, or specific international economic relations. The intellectual justification is, basically, that the elaboration of general models is the most promising route to the discovery of general laws, an essential aspect, in turn, of explanations. (Miller 1987, quoted in Lawson 1997: 292) But there are further prima facie reasons to suspect that the tensions and inconsistencies in Schumpeter’s substantive work may be attributable to an ex ante commitment to deductivism. In his History, the centrepiece of which is an unsurpassably erudite dramatisation of the gradual ascent of the neoclassical paradigm, he repeatedly considers the latter’s epistemic status. And these considerations, many of which can be found as far back as Wesen,15 match what one would expect from an author unable to discard deductivism – even though Schumpeter rejects the vision that goes with it. His contempt for ‘philosophical empiricism’, and the meta-theoretical traditions to which it is related, is apparent throughout. Utilitarianism, its bedfellow in social science, is dismissed as ‘the shallowest of all conceivable philosophies of life that stands indeed in a position of irreconcilable antagonism to the rest of them’ (Schumpeter 1954: 133). Economists are repeatedly charged with a lamentable proclivity to fall into rationalist errors. They are accused, for example, of having a metasociological framework that turns on individual rationality – one in which social classes are ‘mere labels affixed to economic functions’ and individuals, rather than ‘living and fighting beings’, are ‘mere clotheslines on which to hang propositions of economic logic’ (Schumpeter 1954: 886–7). But their analytical work, Schumpeter insists, can be disentangled from mistakes such as these. The history of economic analysis is the history of the gradual emergence of the conception of general equilibrium into the light of consciousness (Schumpeter 1954: 242); and the analytical value of this achievement is not materially affected by the semi-philosophical foliage that surrounds it. In other words, deductivism can be salvaged.

280

Mário da Graça Moura

And this assessment Schumpeter justifies as follows: ‘Economics, like every other science, started with the investigation of “local” relations between … economic quantities, such as the relation between the price of a commodity and the quantity of it that is available in a market’ (Schumpeter 1954: 242). Gradually, however, it was realised that such regularities are not independent of one another but related, which made it possible to ascend to a higher level of ‘abstraction’. Walras’s work represents the culmination of this process, and is ‘the only work by an economist that will stand comparison with the achievements of theoretical physics’ (Schumpeter 1954: 827). For, ‘from the standpoint of any exact science, the existence of a “uniquely determined equilibrium (set of values)” is, of course, of the utmost importance, even if proof has to be purchased at the price of very restrictive assumptions; without any possibility of proving the existence of uniquely determined equilibrium – or, at all events, of a small number of possible equilibria – at however high a level of abstraction, a field of phenomena is really a chaos that is not under analytic control’ (Schumpeter 1954: 969). Accordingly, Walras’s system is ‘the Magna Carta of economic theory’ (Schumpeter 1954: 242). Of course, Schumpeter continues, ‘the technical imperfections of that monument of constitutional law [are] an essential part of the analogy’ (Schumpeter 1954: 242). In the last analysis, the Walrasian system may be ‘nothing but a huge research program’ (Schumpeter 1954: 1026). But it would spell misunderstanding to ‘question the value of a theory that holds only under conditions, the mere statement of which seems to amount to refuting it’ (Schumpeter 1954: 1012). ‘Lest [the reader] should … turn away from Walras’ construction on the ground of its hopeless discrepancy from any process of real life’, Schumpeter submits after discussing the shortcomings of the Walrasian tâtonnement: ‘I wish to ask him whether he ever saw elastic strings that do not increase in length when pulled, or frictionless movements, or any other of the constructs commonly used in theoretical physics; and whether, on the strength of this, he believes theoretical physics to be useless’ (Schumpeter 1954: 1015). It is apparent, in short, that Schumpeter is committed to deductivism. And it is equally apparent, I think, that this commitment is not a product of sustained reflection on the nature of the social world.16 Rather, Schumpeter’s acceptance of deductivism is a consequence of a misunderstanding regarding natural science. In as much as his vision turns on entrepreneurship, innovation and structural change – hence on change, choice, novelty and openness – the potential for tensions and inconsistencies in his substantive writings is quite clear; and yet these tensions and inconsistencies are not necessarily an obstacle to understanding, as they are conceivably explicable. And to understand a contradictory text is to explain why it is contradictory.

Scientific exegesis and Schumpeter

281

V Concluding remarks I have discussed Stigler’s principle of scientific exegesis and Lawson’s critical realism. The former supposes the identification of ‘main conclusions’ in a text. If they are not readily available, scientific exegesis is likely to produce a considerable amount of confusion. There is, of course, a way out of this: to decide (a priori) what sort of conclusions are scientifically relevant. If one is committed to a deductivist mode of explanation, it is obvious what sort of ‘analytical conclusions’ matter. I think that Stigler, in presenting the principle of scientific exegesis, is ultimately taking this stance – a stance which Schumpeter shares where economic analysis is concerned. Lawson, however, demonstrates that the acceptance of deductivism rests on a misunderstanding. And, on the basis of this demonstration, it is possible to construct a theory of inconsistencies in economic writings – a theory that, prima facie, could be much more useful than Stigler’s principle for the practice of the history of economic thought and, in particular, for understanding Schumpeter’s writings. Throughout Schumpeter’s work, as I see it, there is a repeated attempt to express a world view that turns on openness – innovation and structural transformation are his main concerns – within a deductivist framework that is accepted largely a priori and just cannot capture these features. In the course of his writings, to be sure, this tension manifests itself in various guises (and this fact demands interpretation). In any case, though, Schumpeter never succeeds in overcoming his initial commitment to deductivism, which – I only suggest – is at the source of all sorts of inconsistencies in his work.17 No attempt to assess any hypothesis on Schumpeter has been made here, of course, as this would require a detailed analysis of his work as a whole. Nevertheless, I want to end by providing some brief suggestions for such an analysis.18 Recall Schumpeter’s Theory of Economic Development. From the beginning of this book, the oscillation between an open system ‘vision’ and equilibrium ‘analysis’ is, I think, quite apparent. Schumpeter presents two types of action – everyday behaviour within the ‘circular flow’, and entrepreneurial activity – which involve different degrees of conscious rationality. The former type of behaviour presupposes a set of rules or traditions or structures, not consciously known, upon which individuals draw. Decisions are made and could have been different – behaviour is not pre-determined – but they are made within an institutional setting that is tacitly taken for granted. Entrepreneurial action means transcending some of those rules of behaviour. This in turn presupposes that individuals become consciously aware of the traditional ways of doing things. It involves creativity, and much more conscious rationality than everyday behaviour.

282

Mário da Graça Moura

Now, whilst making this distinction, Schumpeter also identifies the ‘circular flow’ with an equilibrium system. And, in importing a deductivist framework, he fatally undermines the possibility of coherently bringing in entrepreneurship and of satisfactorily analysing innovative action. He cannot properly integrate into this framework the various flows of knowledge involved in innovation. Neither can he adequately elaborate on the role of produced means of production in the process of innovation. Business Cycles, as is well known, is an attempt to develop the argument of the Theory so as to illuminate the history of capitalism. But a deductivist framework cannot be adapted to the extent that it captures historical features. In a world of the sort which deductivism implicitly presupposes, there would be no history. Thus, the theoretical chapters of Business Cycles do not offer a coherent model – in fact, Schumpeter ends up undermining his equilibrium framework – and have only a superficial connection with the historical chapters. In Capitalism, Socialism and Democracy the commitment to deductivism may, at first glance, seem less clear. But the tension between openness and closure surfaces once again. Schumpeter emphasises uncertainty, and ridicules the utilitarian belief that capitalism can be systematically improved by government policies. Yet he also argues that entrepreneurship tends to obsolescence. The latter view – which implicitly supposes closure – leads him (reluctantly) to the conclusion that centralised socialism will function better than capitalism. His antirationalist proclivities – which are underpinned by the vision of an open world – lead him to predict that centralised socialism is likely to be a ludicrous (and possibly ‘fascist’) failure.

Notes 1

2 3 4 5

CEMPRE, Faculdade de Economia, Universidade do Porto. CEMPRE – Centro de Estudos Macroeconómicos e Previsão – is supported by the Fundação para a Ciência e a Tecnologia, Portugal, through the Programa Operacional Ciência, Tecnologia e Inovação (POCTI) of the Quadro Comunitário de Apoio III, which is financed by FEDER and Portuguese funds. This paper was presented at the ‘Workshop on Realism and Economics’, Newnham College, Cambridge, in November 2002. I am grateful to the participants at this seminar, and especially to António Almodovar, Tony Lawson and Paul Lewis, for their comments. The argument builds on my 1997 PhD dissertation, for which I gratefully acknowledge financial support from FEP (UP) and INVOTAN. Translations from Wesen are my own. For a comprehensive analysis of this tension in Schumpeter’s work, and of its implications, see Oakley (1990). A stance that can be described as Walrasian (Keizer 1997). Whether Stigler as a historian of economic thought consistently practises what he preaches is a different matter, beyond the scope of this chapter.

Scientific exegesis and Schumpeter 6

7 8

9

10 11

12 13

283

To determine ‘what the man really believed’ leads, in Stigler’s terminology, to a ‘principle of personal exegesis’. In the case of the Ricardian problem with which Stigler is concerned, both principles are deemed to lead to the same interpretation. It is immaterial for this experiment whether these putative ‘main conclusions’ are inferred from Schumpeter’s work as a whole or from one of his main texts only. Another possibility would be to argue that Schumpeter is not inconsistent. This is the project carried out in Shionoya (1997). I have criticised it in Graça Moura (2002). To the extent that it can be taken to mean that Schumpeter’s inconsistencies do not matter, Shionoya’s interpretation could be a Stiglerian exegesis. Schumpeter’s best-known discussion of these matters is in his posthumously published History of Economic Analysis (Schumpeter 1954). As is always the case with Schumpeter, however, these valedictory reflections are prefigured in his early work. Consider, for instance, the following statement: ‘Natura non facit saltum – this phrase Marshall has put as a motto at the beginning of his work; and, indeed, the phrase does express the work’s character. But I want to hold against him that at least the evolution of human culture, and specifically the evolution of knowledge, are marked by leaps … Though the new work may rest upon the old, progress is not continuous’ (Schumpeter 1908: 8). As was the case with Keynes (see Schumpeter 1954: 41–2, 1171–2). Conversely, the transition from Mill to Jevons and the other Marginalists is attributed to a new, superior analysis (see Schumpeter 1954: 892). Where, of course, the word ‘preliminary’ must be underlined – as must the word ‘theory’. As Schumpeter insists, it is not just that a theorist ‘is the product of a given social environment – and of his particular location in this environment – that conditions him to see certain things rather than others, and to see them in a certain light’ (Schumpeter 1954: 34). Rather, ‘environmental factors may even endow [the theorist] with a subconscious craving to see things in a certain light’ (Schumpeter 1954: 34). In Schumpeterian terminology, as is well known, this leads to the problem of ‘ideology’. Nevertheless, a ‘conversion’ may not happen. At any event, the catalyst for discarding a Weltanschauung is probably not a rational process. As Schumpeter puts it in the context of a discussion of the meta-sociology of Condillac, Hartley and Hume, ‘[t]he way in which that fundamental science of man – of human nature, human knowledge, and human behavior – produced all sorts of “natural laws” may be illustrated best by what may be described as the English “natural aesthetics” and the “natural ethics” of the eighteenth century … [T]he analytic task was tackled in the spirit of … psychologism: … psychology … was to explain all there was to explain … [T]he psychology actually used, though not always strictly associationist, was always individual psychology, introspective, and of the most primitive kind, rarely if ever involving anything beyond some simple hypothesis about the reactions of the individual psyche from which everything else followed by deduction. Aesthetical and ethical values were thus explained in a manner suggestive of that in which Italian and French economists in the eighteenth century, and the majority of economists of all countries in the nineteenth, explained economic values. This procedure was called empirical and in a sense it was, but only in the sense in which, for example, the Jevons–Menger–Walras theory of marginal utility is. There was nothing “experimental” or inductive about it and it was in fact not very realistic,

284

Mário da Graça Moura

notwithstanding all the programmatic utterances, war cries, and appeals to Francis Bacon’ (Schumpeter 1954: 126–7). 14 Of course, many theorists committed to deductivism do not claim to ‘explain’ (or to predict) empirical phenomena. Rather, their theories are argued to serve other, somewhat ambiguous purposes. This shifts the discussion to a different ground, but it does not solve the problem touched upon in the next few sentences. For elaborations see Lawson (1997). 15 Methodological statements extracted from Schumpeter’s German works can be found in Graça Moura (2002, 2003). See also Shionoya (1997). 16 For further evidence of this, see, e.g., his comments on Hayek’s Scientism papers (Schumpeter 1954: 17–18). 17 This means, of course, that he never manages to suppress his ‘vision’ either. And this makes his work much more interesting, if less coherent, than it would otherwise be. 18 For an elaboration of some of these suggestions, see Graça Moura (2002, 2003).

References Freeman, C., J. Clark and L. Soete (1982) Unemployment and Technical Innovation: A Study of Long Waves and Economic Development, London: Frances Pinter Publishers. Graça Moura, M. da (2002) ‘Metatheory as the Key to Understanding: Schumpeter after Shionoya’, Cambridge Journal of Economics, 26: 805–21. Graça Moura, M. da (2003) ‘Schumpeter on the Integration of Theory and History’, The European Journal of the History of Economic Thought, 10: 279–301. Keizer, W. (1997) ‘Schumpeter’s Walrasian Stand in the Socialist Calculation Debate’, in W. Keizer, B. Tieben and R. van Zijp (eds) Austrian Economics in Debate, London and New York: Routledge. Lawson, T. (1994) ‘Critical Realism and the Analysis of Choice, Explanation and Change’, Advances in Austrian Economics, 1: 3–30. Lawson, T. (1997) Economics and Reality, London and New York: Routledge. Miller, R.W. (1987) Fact and Method: Explanation, Confirmation and Reality in the Natural and the Social Sciences, Princeton, NJ: Princeton University Press. Nelson, R.R. and S.G. Winter (1982) An Evolutionary Theory of Economic Change, Cambridge, MA, and London: The Bellknap Press of Harvard University Press. Oakley, A. (1990) Schumpeter’s Theory of Capitalist Motion: A Critical Exposition and Reassessment, Aldershot: Edward Elgar. Phillips, A. (1971) Technology and Market Structure: A Study of the Aircraft Industry, Lexington, MA: Heath Lexington Books. Schumpeter, J.A. (1908) Das Wesen und der Hauptinhalt der theoretischen Nationalökonomie, Munich and Leipzig: Duncker & Humbolt. —— (1912) Theorie der wirtschaftlichen Entwicklung, Leipzig: Duncker & Humblot. —— (1920/21) ‘Sozialistische Möglichkeiten von heute’, Archiv für Sozialwissenschaft und Sozialpolitik, 48: 305–60. —— ([1928] 1989) ‘The Instability of Capitalism’, Economic Journal, 38. Reprinted in R.V. Clemence (ed.) (1989) Essays on Entrepreneurs, Innovations, Business Cycles, and the Evolution of Capitalism, New Brunswick and London: Transaction Publishers.

Scientific exegesis and Schumpeter

285

—— (1934) The Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest, and the Business Cycle, Cambridge, MA: Harvard University Press. —— (1939) Business Cycles: A Theoretical, Historical and Statistical Analysis of the Capitalist Process (2 volumes), New York: McGraw-Hill. —— (1942) Capitalism, Socialism and Democracy, 5th edition (1976), London: Allen & Unwin. —— (1954) History of Economic Analysis, London: Allen & Unwin. Shionoya, Y. (1997) Schumpeter and the Idea of Social Science: A Metatheoretical Study, Cambridge: Cambridge University Press. Stigler, G.J. (1965) ‘Textual Exegesis as a Scientific Problem’, Economica, 32: 447–50.

15 Transforming methodology Critical realism and recent economic methodology D. Wade Hands

Introduction Although critical realism is often classified as a particular type of ‘economic methodology,’ this classification seems inappropriate given the traditional definition of methodology. Critical realism is broadly ‘methodological’ – concerned with philosophical investigation into the scientific character of economics – but it is not literally a ‘methodology’ in the way that methodology has traditionally been defined within general science theory. Under the traditional definition, methodology is strictly concerned with ‘method’ – methodological rules that specify the proper relationship between empirical evidence and scientific theories for a wide range of different scientific domains. Notice that, according to this traditional definition, methodology focuses strictly on the form of the scientific investigation: the form that scientific inquiry must take in order to guarantee the cognitive/epistemic value of the products produced by the inquiry. Critical realism is not a ‘method’ in this sense. The goal of critical realism is to philosophically ‘underlabour’ for the social science of economics (Lawson 1997a: 61), and that underlabouring is not restricted to methodological rules or the form of the scientific inquiry. For example, critical realists argue that since intentionality is a fundamental feature of human action, accommodating such intentionality must be an essential feature of economics or any social science. This is more than a claim about the proper ‘method’ of science, it is a claim about the content of a particular science, specifically the prerequisite ontological posits of that science. While critical realism certainly overlaps with traditional methodological concerns, it is also clear that it is more than just one particular type of economic methodology. The purpose of this chapter is to examine the relationship between the critical realist literature and a number of recent developments within the philosophy of science and economic methodology. I will argue that critical realism has much in common with these recent developments – recent methodological trends – and that critical realism shares a common philosophical backdrop with these trends: the backdrop of recognizing

Recent economic methodology

287

and responding to problems associated with the core ideas of the previous generation of the philosophy of science and the economic methodology derived from it. The argument that critical realism shares a family resemblance, or at least a common point of departure, with recent methodological trends is not, prima facie, an argument either for or against critical realism. I have criticized critical realism in the past – particularly its interpretation of mainstream economics (Hands 1997a) – and I have also discussed the criticisms offered by others (Hands 2001b), but this paper is not about criticism. This paper is about finding methodological common ground. To this end I will discuss a number of the key problems associated with the traditional approach to economic methodology – what I will call generalrules-based methodology – and make the case that critical realism also recognizes these difficulties and begins from effectively the same philosophical starting point as most contributions to the diverse and rapidly growing methodological literature. One of the main themes in my recent methodological work (Hands 2001a, 2001b) has been to redefine the field of economic methodology in such a way that it includes a number of current research projects that are outside (sometimes way outside) of the traditional methodological rubric. I have proposed the term ‘the new economic methodology’ to capture the vast literature that combines economics and science theory in some way. In a nutshell my argument is that if one sticks with the traditional narrow definition of economic methodology – as the search for a few very general rules for the proper conduct of economic science – then economic methodology is effectively dead, since it has systematically failed to produce such rules, and no such rules seem to be anywhere on the horizon. On the other hand, there currently exists a vast, expanding, and quite interesting, literature that combines science theory and economics in some way, and, under that broader notion of economic methodology, the field is not only alive, but alive and flourishing. The purpose of this chapter is to make the case that critical realism is one of the approaches that falls under the general rubric of the new economic methodology. Critical realism is not like traditional economic methodology in that it does not search for a few general methodological rules, but it is like much of the new methodological literature in that it starts from the same difficulties of the rules-seeking approach in its effort to transform mainstream economic theory. While I do think it is useful to substantially broaden the definition of economic methodology – to include things like the economics of scientific knowledge and various philosophies of natural science that employ the tools of economic analysis – such radical broadening is not really necessary to make the main point of this chapter. For the current purposes it is only necessary to discuss the problems of rules-based methodology and explain how critical realism acknowledges these difficulties and

288

D. Wade Hands

attempts to move forward with its own underlabouring endeavors in response to these problems. Thus when I refer to ‘recent economic methodology’ in this chapter, I do not necessarily mean all of the variety of different projects that I have lumped under the label of the new economic methodology; I simply mean the recent work by economists and philosophers that is responding to the various difficulties associated with the previous generation of rules-based methodological inquiry and is self-consciously engaged in methodological discourse about economic science. Similarly, it is not necessary to refer to all of the various fields within contemporary science theory (such as the sociology of science, anthropology of science, rhetoric of science, etc.) when discussing the problems associated with the general theory of scientific knowledge; for my purposes here contemporary philosophy of science will serve quite effectively as the main reference point within general science theory. The paper has two main sections. The first examines the recent trials and tribulations of rules-based economic methodology and how those difficulties are related to more general problems within contemporary philosophy of science. The discussion in this first section will be rather brief and focus on only a few of the many controversial issues involved in the literature. For those interested in a more comprehensive treatment I suggest the book-length discussion in Hands (2001b). The second section turns to critical realism and a few of the ways in which it is consistent with, and starts from the same philosophical problem situation as, the literature discussed in the first section. The conclusion provides a brief response to some of the potential criticisms that might be raised about the claims made in the first two sections.

Economic methodology and the philosophy of science Critical realist books and papers do not generally start with the philosophy of science, or even problems within the philosophy of science; they generally start with the problems of mainstream economics. For example, the first chapter of Tony Lawson’s Economics and Reality opens with the sentence: ‘Contemporary academic economics is not in a healthy state’ (Lawson 1997a). Lawson’s real concern is economic theory; its difficulties may stem from the fact that the discipline has (explicitly or implicitly) adopted an inadequate positivist conception of scientific knowledge and its associated characterization of explanation and concept of a scientific law, but problems in the philosophy of science are only part of the diagnosis, not the subject of the inquiry. The original starting point for Lawson is the sad state of economic science. Of course this immediately aligns critical realism with the interests and concerns of various heterodox economists (Austrian, Marxian, Sraffian, Post Keynesian, feminist, etc.), who also start from the position that mainstream economics is generally in bad shape and needs to be replaced.

Recent economic methodology

289

Such criticism of mainstream economics does not automatically align critical realism with other approaches to economic methodology – even recent approaches – since there is not any general agreement about the current ‘state’ of mainstream theory within the methodological literature. For some methodologists, the current state is, as with critical realism, generally quite bad; for others it is globally just fine; for still others it is fine in certain sub-fields and problematic in others; for still others the whole project of methodological appraisal is a philosophical dead end and should be replaced by historical, sociological, or rhetorical inquiry; and there are of course many other different views. Since my purpose is to emphasize the ways in which critical realism is ‘like’ other recent methodological discourse, I will not begin by discussing the current ‘state’ of mainstream economic theory, but rather by considering the current ‘state’ of the philosophy of science, a subject where there seems to be much more agreement. While the science theory literature of the last forty or so years has been both expansive and extremely diverse, it is possible to distill a few points of general agreement from among the wide range of different commentators. First, the positivist-inspired ‘Received View’ (Suppe 1977) that was dominant within mid-twentieth century Anglo-American philosophy of science – what Philip Kitcher (1993) calls ‘Legend’ – has irrevocably broken down. The causes of this breakdown remain hotly debated, but it is possible to discern a few particularly influential elements in the overall set of destructive forces. Thomas Kuhn’s Structure of Scientific Revolutions (1970), Paul Feyerabend’s Against Method (1975), and two key papers by W.V.O. Quine (1951, 1969) clearly played important roles in the eventual breakdown of the Received View. Of course this is not to suggest that the philosophical establishment was felled by the blows of a few key individuals, only that a wide range of disparate critical forces coalesced around a few key texts. Recent studies of the positivist movement such as those by Cartwright, Cat, Fleck, and Uebel (1996) and Friedman (1999), as well as recent re-examinations of Kuhn’s life and work such as Fuller (2000), make it quite clear that the issues surrounding the demise of the Received View are extremely complex. Second, while the Received View is discredited, there has not been any single approach (or even set of approaches) that has emerged to replace the once dominant positivist-inspired framework. Third, while there is certainly not any consensus about a replacement program, there is a partial consensus about the general features that any adequate framework must have. These general features are, in no particular order, that: it must be sensitive to the social character of the scientific endeavor; it must recognize, and in some way defuse, both the problem of the theory-ladenness of empirical observations and the Duhem–Quine underdetermination problem; it must view science as more contingent and less methodical than Legend would have it; it must to some degree be

290

D. Wade Hands

consistent with the actual history and practice of successful science (it cannot be just arm-chair philosophizing); it must be naturalistic in spirit in that it must build on the best contemporary science and approach the subject of scientific knowledge in a relatively scientific way; and finally it must steer some kind of middle ground between, on one hand, the old narrow (epistemically) normative philosophy of science of the Received View and, on the other hand, total relativism about the cognitive status of scientific knowledge. Fourth and finally, philosophy of science must respond to the sociological literature, inspired in part by Kuhn (although not endorsed by him), that emphasizes the social character of science in a way that tends to debunk or undermine the cognitive status of science. This literature – such as the early influential contributions by Bloor (1991), Collins (1985), Latour and Woolgar (1986), Latour (1987), and Knorr Cetina (1981) – reduces the activities and beliefs of scientists to the same type of social forces that determine the activities and beliefs of any other social agents, resulting in either the ‘mere description’ of scientific behavior without a normative-epistemic evaluation, or in the negative evaluation that scientists are not epistemically ‘special’ at all. Much of the recent work in the philosophy of natural science has focused on recovering some (positive) version of epistemic normativity from such sociological accounts, while at the same time accepting many of the assumptions of the sociological literature (particularly the social character of scientific knowledge). So the Received View is gone and its demise has left a particular philosophical problem situation within the philosophy of natural science, but how have these changes affected the recent literature in economic methodology? Well, there are certainly many effects, but the most obvious is that it is simply not possible to continue to do economic methodology in the same way that it was done during most of the twentieth century. The traditional approach to rules-based methodology is best characterized as the ‘shelf-of-scientific-philosophy’ (Hands 1994) view of economic methodology. The philosophy of natural science was viewed as a ‘shelf’ of potential methodological rules, and economic methodology, according to this view, consisted of taking things off the shelf and then ‘applying’ them to the particular science of economics. There are at least two features that need to be emphasized about this shelf approach to economic methodology. First, the items on the shelf were generally taken off fully assembled; the ideas were seldom reconfigured or reconditioned for their use in economic science. The general attitude seemed to be that if the items didn’t fit very well, the problem was with economics – it was economics that needed to be reconfigured, not the philosophy of science. In a few cases – say Austrian economic methodology – the items on the shelf were rejected, but even here they were still considered to be fully assembled (just not suitable). Second, the shelf was a philosophy of science shelf – primarily products produced by Positivist & Popperian (P&P)

Recent economic methodology

291

philosophical industries – in particular, it was not a general Philosophy shelf and it was not an actual Science shelf. It was not a general Philosophy shelf in that the items were manufactured exclusively by only a few philosophical firms – P&P philosophers of natural science – and not by the discipline of philosophy more generally; the shelf was bare of Hegelian, pragmatist, Aristotelean, or other broad philosophical ideas. It was also not an actual Science shelf in that the items on the shelf were not based on the activities of flesh-and-blood natural scientists – Kuhn’s concern – they were based on what philosophers of science thought good scientists ought to do. The point is that rejecting the shelf-of-scientificphilosophy view of economic methodology does not necessarily mean rejecting all philosophical ideas or rejecting all consideration of how natural scientists go about their day-to-day activities; rejecting the shelf only means rejecting the idea that all there is to the field of economic methodology is to take prepackaged ideas off the philosophy of natural science shelf – 3' x 5' card philosophy of science (McCloskey 1994) – and ‘apply’ those ideas to economics. The fall of the Received View and the corresponding disruption within Legend philosophy of science have necessitated a move away from the shelf-of-scientific-philosophy approach to the methodology of economics. The cupboard is simply bare; there is nothing on the shelf that even the majority of philosophers of science would endorse as the proper scientific method. Given this, those working in the field of economic methodology have been forced to look elsewhere for ideas and inspiration. Where have they turned? Well, they did not turn to just one new shelf, but rather to a dizzyingly wide array of different sources and resources. Those writing in the field of economic methodology have looked outside of the philosophy of science to areas of science studies such as the sociology of scientific knowledge and science and technology studies (see Hands 1997b for a discussion and references); they have turned back to nineteenth science theorists such as Mill (Hausman 1992), to contemporary philosophy of mind (Rosenberg 1992), or to post-positivist-empiricism (Boylan and O’Gorman 1995); they have approached directly the question of how practicing economists employ various concepts such as idealizations (Mäki 1994) or models (Morrison and Morgan 1999); they have turned to classical rhetoric and the literature on the rhetoric of science (McCloskey 1994, 1998); they have even looked to economic theory itself for inspiration (Sent 1999; Wible 1998). There is no formula for doing economic methodology today; the shelf-of-scientific-philosophy is gone and in its stead is a flurry of innovative and exciting new activities.

Critical realism and recent methodological trends This section will discuss a few of the many ways in which critical realism is consistent with the recent developments in economic methodology and

292

D. Wade Hands

post-positivist philosophy of science discussed in the previous section. Critical realism certainly starts from the breakdown of the Received View and the demise of the shelf-of-scientific-philosophy, and turns to other resources in its underlabouring efforts, but the consistency runs much deeper and is more broad-based than merely this initial starting point. Since there are so many different facets to the critical realist research program, and since these different facets often relate to the demise of the shelf-of-scientific-philosophy and the new approaches to economic methodology in very different ways, the most effective way to proceed seems to be to simply list a number of significant features that critical realism shares with recent non-rules-based methodological literature. The list is weakly ordered – from general to specific – but the ordering is far less important than the overall consistency that emerges from the ensemble of the various items discussed. (i) First of all, critical realism draws on certain philosophical resources – specifically the transcendental realism of Roy Bhaskar (1978, 1987, 1989) – but not on resources from mainstream philosophy of science. Bhaskarian realism, like most general philosophical frameworks, involves a particular characterization of science, but it is not just a philosophy of natural science (and certainly not a P&P-based philosophy of natural science). Also in keeping with the spirit of contemporary methodological developments, critical realism has systematically reconfigured the transcendental realist program in various ways to better fit the needs of those interested in economics and economic methodology. While much of the early work in critical realism did take ideas directly off the Bhaskarian shelf (a philosophical, but not philosophy-of-science, shelf) and apply them to social science in a relatively prepackaged way, this pattern has changed in the more recent literature; as the program has matured, critical realists have increasingly modified the Bhaskarian concepts in order to make them better suited to the particular concerns of contemporary economics and other social sciences. The bottom line is that critical realism is not a shelf-of-scientific-philosophy-based approach to the philosophy of social science; when philosophical ideas are borrowed, they are not taken from the philosophy-of-science portion of the shelf, and the philosophical ideas that are used have increasingly been reconfigured to meet the particular needs of contemporary social science. (ii) Perhaps the most obvious point of common ground is that critical realism is self-consciously anti-positivist and begins its analysis from the recognition of many of the problems associated with Legend philosophy of science discussed above. While critical realists focus their attention on underlying objects, structures, and forces that exist independently of scientific investigators, the knowledge produced by such scientists is fundamentally social, theory-and-interest laden, situated, and underdetermined:

Recent economic methodology

293

[A]lthough the emphasis throughout has been realist, … this stance is complemented not by an absolutist or foundational position in knowledge but by an epistemological relativism, by the thesis that we can only know these things under particular (historically and socially relative), and potentially transformable, descriptions. Clearly, if the (intransitive) objects of knowledge exist (largely) independently of our knowledge of them, such knowledge as we actually possess cannot be identified, or be said to be in correspondence, with such objects; it is not reducible or equivalent to them. Knowledge, rather, exists in a historically specific, symbolically mediated and expressed, practice-dependent, form. (Lawson 1997a: 58–9) Thus while critical realists would defend the independence of the intransitive domain of natural and social objects/forces, the scientific knowledge we possess about such objects/forces is socially conditioned, contingent, and potentially interest-laden. This is more than just the rejection of the Received View; it is also an endorsement of many of the generally accepted features of post-Legend philosophy of science. It is also important to note that critical realists similarly reject another cornerstone of the Received View: the deductive-nomological (D-N) model of scientific explanation. Critical realists characterize explanation in abductive terms – the identification of the deep underlying (intransitive) causal forces responsible for the phenomenon in question – and while this is just one of many competing explanatory schemes, it begins from the same problems of the D-N model as many other contemporary approaches. (iii) Critical realism is not about methodological rules. The key idea behind any rules-based approach to science theory (P&P-inspired or any other) is that there are proper rules that individual scientists need to follow in order to produce legitimate scientific knowledge and that those rules are independent of the particular subject matter, social context, interests, or pragmatic concerns of the science in question. Critical realism is not rules-based in this sense. Perhaps the most important point of critical realism is that, regardless of how effective empirical realism has been in the natural sciences, it is not appropriate in social science, and, in particular, there are a number of subject-matter-specific features of the social world – specifically human intentional agency and the causal power of social structures – that must be part of the ontological backdrop of any adequate social science. Of course critical realism remains a normative philosophical approach – it does specify the ontological features that an adequate social science ought to include – and in that sense it is contrary to various approaches (particularly sociological approaches) within contemporary science theory that have abandoned normative considerations in favor of a purely descriptive approach to scientific

294

D. Wade Hands

knowledge, but this normativity is substantially different than the normativity that emerges from rules-based philosophy of science. There are normative aspects to critical realism, but it does not specify methodological rules that individual scientists must follow in order to obtain epistemically justified scientific knowledge. (iv) Critical realism is not a priori arm-chair philosophy of science in the sense of the Received View, and in the sense opposed by Kuhn, Feyerabend, and recent historians and sociologists of science. Critical realism, like most of the work in contemporary science theory, attempts to ground its view of scientific knowledge on scientific theory and the actual conditions of scientific practice. In this sense critical realism is more naturalistic than the Received View and most of the economic methodology predicated on it. Since this is a rather controversial claim, I will defend it in two separate ways. First, even though critical realism draws its philosophical inspiration from transcendental philosophy of the Kantian tradition, and even though this is not the philosophical inspiration for most contemporary naturalisms (see, for example, Callebaut 1993a; Kitcher 1992; Kornblith 1985; Laudan 1990; Rosenberg 1996; Solomon 1995), it is possible to characterize the conditions for possible scientific knowledge in transcendental naturalist terms. Bhaskar has even been cited as a ‘well-known defender of transcendental arguments within a naturalistic framework’ (Callebaut 1993b: 2, n.3). Second, there are naturalistic philosophers of science – Nancy Cartwright (1989, 1999) in particular – who advocate philosophical positions that are quite similar to transcendental realism. This is not to say that Cartwright’s philosophy of science (or Bhaskar’s) is entirely unproblematic, but it does suggest that it is possible to start from what is essentially a naturalistic position and end up with a philosophical perspective that looks very much like critical realism. While critical realism does not support the ‘naturalistic’ view that social and natural science share the same basic explanatory framework, it is broadly consistent with the naturalistic turn that characterizes much of contemporary science theory. (v) Critical realists start from the position that human intentionality is an essential feature of any social science, and that such intentionality is prima facie in conflict with P&P notions of proper scientific inquiry. Now while many contemporary philosophers of mind (Churchland 1992), and certain philosophers writing about economics (Rosenberg 1992), use this same conflict to argue for the elimination of all intentional ‘folkpsychological’ concepts from psychological and economic science – precisely the opposite of the stance taken by critical realists such as Lawson (1997a) – the point of departure is exactly the same. Both begin from the position that explaining human action in terms of intentionality (beliefs and desires) is in conflict with the D-N model of explanation and the associated notion of scientific laws, and therefore that one of these things – either intentionality or D-N explanations and P&P justified

Recent economic methodology

295

scientific laws – needs to go. It should also be noted that the tension between D-N and social science explanatory schemes has also been identified within the Popperian wing of the P&P tradition (see Caldwell 1991; Hands 1991a, 1991b), and while various Popperian solutions have been offered, it has recently been suggested that critical realism might be a useful resource for reconciling these difficulties as well (see van Eeghen 1996). (vi) Finally, and focusing more explicitly on economics, critical realists have systematically emphasized the role of tendency laws in economic science. The explicit discussion of tendency laws in economics of course goes back to at least Mill (1874), but the importance has recently been reemphasized in influential works by Hausman (1992) and Cartwright (1994, 1995). While critical realists such as Lawson employ a slightly different (and less empiricist) notion of tendency laws than either Hausman or Cartwright, the point is not that critical realism is identical to other recent approaches in economic methodology; it is simply that critical realism is motivated by a similar set of issues and concerns (in economics as well as in science theory). The nature of the tendency laws that are at work in economic science is an important methodological question that has been identified by critical realists as well as by a number of other recent methodological commentators.

Conclusion The purpose of this chapter was threefold. I have tried to explain the major developments that have taken place within the philosophy of science since the breakdown of the Received View, to explain the recent turn away from both the shelf-of-scientific-philosophy and the search for general rules within the recent methodological literature, and to make the case that critical realism is broadly consistent with these changes in philosophy of science and economic methodology. While more could obviously be said about any one of these subjects, I believe I have provided an effective defense of all three parts of the argument. Critical realism clearly has much in common with other recent developments in science theory and economic methodology. In closing I would like to make a few comments (predictions?) about how the above argument might be received among various readers within the methodological community. I fear the argument might succumb to the old adage that ‘man who walks in middle of street gets hit by cars going both ways.’ It is possible that both defenders and critics of critical realism will find my argument to be unsatisfactory. Consider critical realists first. Most critical realists perceive their efforts in fairly grand-historical terms; the goal is essentially the transformation of mainstream economics and its replacement by an alternative (or alternatives) that is (are) committed to an adequate social ontology and

296

D. Wade Hands

devoid of the many difficulties currently associated with mainstream economic theory. Recruits for this transformative effort will come from heterodox economists and others profoundly displeased with current economic theory; it is a grass-roots movement of critical economists and other social scientists, and while it is broadly ‘philosophical,’ it is not really for philosophers of science or other science theorists, and I suspect that many critical realists will have little interest in the question of whether their transformative underlabouring is, or is not, consistent with recent developments within the philosophy of natural science (or even economic methodology). Critical realism is about substantive transformation, not fashionable philosophical chit-chat. Thus it is possible that most critical realists – even if they do not disagree with what I have said about philosophy of science, economic methodology, or critical realism – will find my argument to be essentially beside the point. So what if critical realism is consistent with a number of recent developments within philosophy of science and economic methodology: who cares? Now consider the response of critical realism’s philosophical and methodological critics. While some critics will undoubtedly challenge my claim that critical realism is consistent with these recent developments, a more likely response will be: ‘Okay, critical realism may start from the same set of problems, but its solutions (and the whole transcendental approach) to those problems are radically at odds with the answers offered elsewhere within contemporary science theory and economic methodology.’ So what if critical realism starts from the same philosophical problem situation as much of the recent literature, their answers are systematically (perhaps blithely) contrary to that literature? While it may not be possible to answer either defenders or critics in a way they will find satisfactory, I nonetheless conclude by offering a response to both sides of the debate. For the critical realists my response is simply that, interesting or not, critical realism is one aspect of broader intellectual trends within the study of scientific knowledge. This may not be the reason that most critical realists support the program, but it is a fact of intellectual life, and it is also one of the reasons that critical realism has attracted so much attention within the methodological literature. There is a substantial literature – in both philosophy of science and economic methodology – in which arguments are presented that have much in common with the arguments made by critical realists (without either side recognizing the similarity of the positions). Recognition of these commonalities would simultaneously make new intellectual resources available to critical realists, and potentially open the door for the wider dissemination of critical realist ideas. Critical realists have increasingly been willing to modify the Bhaskarian transcendental realist framework to suit the problems of the social sciences; recognizing where overlap exists with other philosophical positions provides a wealth of additional resources for these efforts.

Recent economic methodology

297

For the methodological critics of critical realism my response is simply to reiterate what I said in the introduction. My point is not that critical realism is right or wrong about all of the solutions it offers for the problems in either science theory or economics – I have previously criticized aspects of the critical realist literature – my point is simply that critical realism starts from many of the same issues and concerns as other contemporary approaches. One does not need to agree with the contents of Debreu’s Theory of Value in order to recognize it as an economics text, and one does not need to agree with everything written by critical realists in order to recognize it as literature consistent with contemporary methodological trends. The title of this chapter is ‘transforming methodology’; it is neither ‘transforming economics’ as some critical realists might have it, nor ‘transforming critical realism’ as some critics might have it. Critical realism is one element within a broad set of different literatures that share a common philosophical point of departure and are actively engaged in trying to transform the way we think about science in general and economic science in particular. While this fact may not be appreciated by everyone on both sides of the debate, I believe that I have successfully demonstrated that it is nonetheless an important feature of contemporary discussion about scientific knowledge and economic science.

Notes 1

2

3

4

One needs to be careful about the use of ‘traditional’ here. The vision of economic methodology that is being replaced – the traditional view – is endemic to mid-twentieth-century economics; those writing on economic methodology during the nineteenth and early twentieth centuries seem much less likely to exhibit this ‘traditional’ tendency. I will not provide a survey/summary of critical realism. I recommend the books by Lawson (1997a), Collier (1994), and Fleetwood (1999); or the many papers where the position has recently been summarized and defended, for example Jackson (1995), Lawson (1994a, 1994b, 1995, 1996, 1997b, 1999), or Fleetwood (1997); or my own summary in Hands (1997a or 2001b). Since Karl Popper and the Popperian philosophical school were never as influential in general philosophy of science as in economic methodology, most philosophers would not include Popperian falsificationism within the Received View, but given its role in economics it seems appropriate to include it. Perhaps I should say: ‘that make them better suited to the particular concerns of contemporary non-Marxist economics and other social sciences.’ Bhaskarian transcendental realism was explicitly a Marxist philosophical framework and much of the early debate about critical realism was confined to debates inside the Marxist tradition. Over time the work of Tony Lawson and others has moved critical realism out of the strictly Marxist sphere of discourse and into the wider realm of economic methodology and general philosophy of social science. Since these authors employ critical realism to

298

5

6

D. Wade Hands criticize neoclassical economics, the recent critical realist literature may still have a certain Marxist appeal; the difference is that it is not exclusively a Marxist appeal. Many institutionalists, feminists, Post Keynesians, Austrians, and countless (often non-aligned) others who believe the disabilities of mainstream economics are ontological in origin have been attracted to the recent critical realist literature. While this is not the place for a detailed discussion of what I mean by ‘naturalism’ and the associated ‘naturalistic turn’ in recent philosophy of science – these subjects are main themes in Hands (2001b) – it is important to note that my use of the term is different from the way it is generally used within the critical realist literature. Most critical realists use the term ‘naturalism’ in the way that it has traditionally been used by philosophers of social science – that ‘social science phenomena are susceptible to explanation in essentially the same sense as are natural phenomena’ (Lawson 1997a: 45–6) – while I use it in the way that it is used in contemporary philosophy of natural science – roughly, that ‘scientific knowledge’ is a natural object that can be investigated using scientific resources in the same way that any other natural object can be investigated. These two uses of ‘naturalism’ or ‘naturalistic’ are obviously related, but they are not the same thing. It is important to emphasize the difference since I am crediting critical realism with participation in the broad naturalistic turn within contemporary philosophy of science, while at the same time most critical realists would stress that the program is not naturalistic; given the different uses of the term, these two assessments of critical realism need not be in conflict. The similarities and differences among these three notions of tendency laws (and the associated ceteris paribus conditions) are examined in detail in chapter 7 of Hands (2001b).

References Bhaskar, R. (1978) A Realist Theory of Science, Second edition, Brighton: Harvester Wheatsheaf. —— (1987) Scientific Realism and Human Emancipation, London: Verso. —— (1989) Reclaiming Reality, London: Verso. Bloor, D. (1991) Knowledge and Social Imagery, Second edition, Chicago: University of Chicago Press. Boylan, T.A. and P.F. O’Gorman (1995) Beyond Rhetoric and Realism in Economics: Towards a Reformulation of Economic Methodology, London: Routledge. Caldwell, B.J. (1991) ‘Clarifying Popper’, Journal of Economic Literature, 29: 1–33. Callebaut, W. (ed.) (1993a) Taking the Naturalistic Turn, Chicago: University of Chicago Press. —— (1993b) ‘Turning Naturalistic: An Introduction’, in W. Callebaut (ed.) Taking the Naturalistic Turn, Chicago: University of Chicago Press. Cartwright, N. (1989) Nature’s Capacities and Their Measurement, Oxford: Clarendon Press. —— (1994) ‘Mill and Menger: Ideal Elements and Stable Tendencies’, in B. Hamminga and N. De Marchi (eds) Idealization VI: Idealization in Economics, Amsterdam: Rodopi.

Recent economic methodology

299

—— (1995) ‘Ceteris Paribus Laws and Socio-Economic Machines’, The Monist, 78: 276–94. —— (1999) The Dappled World: A Study of the Boundaries of Science, Cambridge: Cambridge University Press. ——, J. Cat, L. Fleck and T. Uebel (1996) Between Science and Politics: The Philosophy of Otto Neurath, Cambridge: Cambridge University Press. Churchland, P.M. (1992) A Neurocomputational Perspective: The Nature of Mind and the Structure of Science, Cambridge, MA: MIT Press. Collier, A. (1994) Critical Realism: An Introduction to Roy Bhaskar’s Philosophy of Science, London: Verso. Collins, H.M. (1985) Changing Order: Replication and Induction in Scientific Practice, Beverly Hills, CA: Sage. van Eeghen, P.-H. (1996) ‘Towards a Methodology of Tendencies’, Journal of Economic Methodology, 3: 261–84. Feyerabend, Paul K. (1975) Against Method, London: New Left Books. Fleetwood, S. (1997) ‘Situating Critical Realism in Economics’, Ekonomia, 1: 1–8. Reprinted with minor revisions in S. Fleetwood (ed.) (1999) Critical Realism in Economics: Development and Debate, London: Routledge. —— (ed.) (1999) Critical Realism in Economics: Development and Debate, London: Routledge. Friedman, M. (1999) Reconsidering Logical Positivism, Cambridge: Cambridge University Press. Fuller, S. (2000) Thomas Kuhn: A Philosophical History for Our Times, Chicago: University of Chicago Press. Hands, D.W. (1991a) ‘Popper, the Rationality Principle and Economic Explanation’, in G.K. Shaw (ed.) Economics, Culture and Education: Essays in Honour of Mark Blaug, Aldershot: Edward Elgar. —— (1991b) ‘The Problem of Excess Content: Economics, Novelty, and a Long Popperian Tale’, in M. Blaug and N. De Marchi (eds) Appraising Economic Theories: Studies in the Methodology of Scientific Research Programs, Aldershot: Edward Elgar. —— (1994) ‘Blurred Boundaries: Recent Changes in the Relationship between Economics and the Philosophy of Natural Science’, Studies in History and Philosophy of Science, 25: 751–72. —— (1997a) ‘Empirical Realism as Meta-Method: Tony Lawson on Neoclassical Economics’, Ekonomia, 1: 39–53. Reprinted with minor revisions in S. Fleetwood (ed.) (1999) Critical Realism in Economics: Development and Debate, London: Routledge. —— (1997b) ‘Conjectures and Reputations: The Sociology of Scientific Knowledge and the History of Economic Thought’, History of Political Economy, 29: 695–739. —— (2001a), ‘Economic Methodology Is Dead – Long Live Economic Methodology: Thirteen Theses on the New Economic Methodology’, Journal of Economic Methodology, 8: 49–65. —— (2001b) Reflection Without Rules: Economic Methodology and Contemporary Science Theory, Cambridge: Cambridge University Press. Hausman, D.M. (1992) The Inexact and Separate Science of Economics, Cambridge: Cambridge University Press. Jackson, W.A. (1995) ‘Naturalism in Economics’, Journal of Economic Issues, 39: 761– 80.

300

D. Wade Hands

Kitcher, P. (1992) ‘The Naturalists Return’, The Philosophical Review, 101: 53–114. —— (1993) The Advancement of Science: Science Without Legend, Objectivity Without Illusions, Oxford: Oxford University Press. Knorr Cetina, K. (1981) The Manufacture of Knowledge: An Essay on the Constructivist and Contextual Nature of Science, New York: Pergamon. Kornblith, H. (ed.) (1985) Naturalizing Epistemology, Cambridge, MA: MIT Press. Kuhn, T.S. (1970) The Structure of Scientific Revolutions, Second edition, Chicago: University of Chicago Press. Latour, B. (1987) Science in Action, Cambridge, MA: Harvard University Press. —— and S. Woolgar (1986) Laboratory Life: The Construction of Scientific Facts, Second edition, Princeton, NJ: Princeton University Press. Laudan, L. (1990) ‘Normative Naturalism’, Philosophy of Science, 57: 44–59. Lawson, Tony (1994a) ‘A Realist Theory for Economics’, in R. Backhouse (ed.) New Directions in Economic Methodology, London: Routledge. —— (1994b) ‘Why Are So Many Economists So Opposed to Methodology?’, Journal of Economic Methodology, 1: 105–33. —— (1995) ‘A Realist Perspective on Contemporary “Economic Theory” ’, Journal of Economic Issues, 29: 1–32. —— (1996) ‘Developments in Economics as Realist Social Theory’, Review of Social Economy, 54: 405–22. Reprinted with minor revisions in S. Fleetwood (ed.) (1999) Critical Realism in Economics: Development and Debate, London: Routledge. —— (1997a) Economics and Reality, London: Routledge. —— (1997b) ‘Critical Issues in Economics as Realist Social Theory’, Ekonomia, 1: 75–117. Reprinted with minor revisions in S. Fleetwood (ed.) (1999) Critical Realism in Economics: Development and Debate, London: Routledge. —— (1999) ‘What Has Realism Got To Do with It?’, Economics and Philosophy, 15: 269–82. McCloskey, D.N. (1994) Knowledge and Persuasion in Economics, Cambridge: Cambridge University Press. —— (1998) The Rhetoric of Economics, Second edition, Madison, WI: University of Wisconsin Press. Mäki, U. (1994) ‘Isolation, Idealization and Truth in Economics,’ in B. Hamminga and N. De Marchi (eds) Idealization VI: Idealization in Economics, Amsterdam: Rodopi. Mill, J.S. (1874) ‘On the Definition of Political Economy; and on the Method of investigation Proper to It’, in Essays on Some Unsettled Questions of Political Economy, Second edition, London: Longmans, Green, Reader & Dyer (Augustus M. Kelley reprint 1968, 1st edition 1844). Morrison, M. and Morgan, M.S. (eds) (1999) Models as Mediators, Cambridge: Cambridge University Press. Quine, W.V.O. (1951) ‘Two Dogmas of Empiricism’, Philosophical Review, 60: 20–43. —— (1969) ‘Epistemology Naturalized’, in Ontological Relativity and Other Essays, New York: Columbia University Press. Rosenberg, A. (1992) Economics – Mathematical Politics or Science of Diminishing Returns?, Chicago: University of Chicago Press. —— (1996) ‘A Field Guide to Recent Species of Naturalism’, British Journal for the Philosophy of Science, 47: 1–29. Sent, E.-M. (1999) ‘Economics of Science: Survey and Suggestions’, Journal of Economic Methodology, 6: 95–124.

Recent economic methodology

301

Solomon, M. (1995) ‘The Pragmatic Turn in Naturalistic Philosophy of Science’, Perspectives on Science, 3: 206–30. Suppe, F. (1977) The Structure of Scientific Theories, Second edition, Urbana, IL: University of Illinois Press. Wible, J.R. (1998) The Economics of Science: Methodology and Epistemology as if Economics Really Mattered, London: Routledge.

Index

Abolitionists 146 abstraction 82, 216 active mind 87 actual domain 4, 39 actual realm 111 Adorno, T. 161 age 135 agency-structure model and embeddedness in heterodox economics 132–50; heterodox schools 148–50; individual and society 136–42; individuals’ influence 142–3; institutions’ influences 144; overall model 145–8; self-concept 133–6 Althusser, L. 215 Amadeo, E.J. 56 analysis 275–6 Ansell, C.K. 174 anthropological foundations 235–9 antirealism 189, 190 antitrust laws 142–3 anxiety avoidance 159 Archer, M.S. 3, 5, 6, 7, 10, 13; heterodox economics and ontology 3, 5, 6, 7, 9, 10, 13 Arestis, P. 34, 36, 37, 38, 42, 46, 47 Aristotle 87, 232, 291 Arrow, K.J. 44, 45, 103, 188, 190 atomism 69 Aumann, R.J. 242 Austrian school of economics 2, 15, 22–3, 57, 207, 288, 290; really real in economics 187, 188, 189, 192–3, 194, 195, 196, 197, 198 authority 23 autonomy 248 Bachelier, L. 104 Bagehot, W. 63

Balibar, E. 215 Barker, D.K. 118, 119 Bauer, P.T. 15 Beaulier, S. 22–3, 187–99 Becker, G.S. 124, 159, 210, 213 behaviourism 153 beliefs 158–9, 169 Berger, P.L. 174 Bhaskar, R. 9, 16, 68, 111, 137, 190; critical and real in critical realism 203, 206, 215–16; recent economic methodology 292, 294, 296 Blaug, M. 103, 160 Block, T. 90 Bloor, D. 290 Boettke, P.J. 15, 22–3, 187–99 bounded rationality 257 Bourdieu, P. 240, 241–2 Boylan, T.A. 103, 291 Brouns, M. 121 business class 65, 66 business firm 71 Caldwell, B.J. 295 Callebaut, W. 294 Cambridge economists 35, 205, 212 cannibalism 205–6 capital 214–17 capitalism 210–11, 214–17, 220; competitive 269, 270, 271; trustified 269, 272; see also macroeconomic theory and capitalism Cartwright, N. 289, 294, 295 Cat, J. 289 causality 100 causation 153, 216 centralised socialism 282 change 81–4, 118 Chick, V. 15, 36, 63 choice 153

Index 303 Churchland, P.M. 294 circular flow 281–2 civil law of contracts 45 Classical economics 38, 87, 100, 101, 103 closed-system models 11, 19, 38, 68, 79, 173; Post Keynesian economics 39–41, 42, 44, 47, 50, 51 Colander, D. 55 Coleman, J. 159 collective action 254 collective rationality 258 Collier, A. 137, 216 Collins, H.M. 290 common knowledge 242, 243, 244; quasi 242 competitive capitalism 269, 270, 271 conceptual analysis 88 conceptual reference 78 conduct 170 connectionist (neural) network 232 consumption 88 contextualism 114–15 contracts 44, 45, 51; long-term 4 conventionalism 141, 189 cooperation 22, 179–80 Coulanges, F. de 235 Cowan, R. 218 critical and real in critical realism 202–23; Cambridge economists 205, 212; cannibalism 205–6; capital 214–17; capitalism 210–11, 214–17, 220; deductivism 204–8, 213, 219; mainstream economics 203–6, 208–9, 211–17, 219; methodological issues 204–8, 210–16, 219–20; path dependence 217–20; Post Keynesianism 203, 206–7; postmodernism 211, 213, 217 Critical Realism in Economics: Development and Debate 96, 103 Crotty, J. 61 Dannefer, D. 135 data generating mechanism 97–8 David, P. 203, 217–18, 219–20 Davidson, P. 36–7, 38, 40, 41–2, 44, 45, 46, 57 Davis, J. 20–1, 132–50 Davis, K. 114 de Medici, L. 63 Debreu, G. 188, 297 decision sciences 98 decision-making, crucial 41

deductive-nomological model 293, 294–5 deductivism 11–12, 18, 23–5, 38, 204–8, 213, 219, 276–8; economics as symptom 231, 233; heterodox tradition 160, 163; macroeconomic theory and capitalism 58, 67; Post Keynesian economics 38, 50; really real in economics 192; scientific exegesis and Schumpeter 279–81, 282 deep domain 39 deep variable 102 demand 46; see also effective demand principle Demos, D. 133, 134, 135 Dennett, D. 80 Descartes, R. 192 determinism 4, 60, 70, 78–81, 139, 140, 148, 188, 253 development economics 15 developmental age 135 Dillard, D. 63, 64 discursive consciousness 140 disposable income 88 Dodgson, M. 179 domains 4, 39 Dostaler, G. 62 Dostoyevsky, F.M. 238 Dow, S.C. 36, 37 Duesenberry, J. 172 Duhem, P. 189, 190, 289 Dumont, L. 234 Dumouchel, P. 237 Dunn, S.P. 15, 18–19, 33–52 Dupuy, J.-P. 24, 227–50 Durkheim, E. 234, 235, 239, 278 Dutt, A.K. 56 earner 65 ecological economics 148–9 econometrics 76–7, 96–106, 191; conventional ideas 98–9; definition 103–5; economic regularities 100–1; evaluation 105–6; modern ideas 99–100; reality, definition of 96–8; viewpoints 101–3 economic agent 107 economic methodology 286–98; and critical realism 291–5; and philosophy of science 288–91 economic regularities 100–1 economic science 107, 108, 110, 111, 121–3, 190

304

Index

economic texts as empirical material 123–4 ‘Economics of Economics’ 258–62 economics of institutions and institutions of economics 252–66; ‘Economics of Economics’ and ‘oversupply’ of mathematics 258–62; maximization 254–8; structured reality and social science production 253–4 Economics and Reality 96, 100, 101, 103 Economics as Social Theory and New Economic Sociology 22, 167–82; social construction of economic institutions 173–5; social embeddedness 169–73, 175–7; trust, reciprocity and power 177–80 economics as symptom 227–50; British critical realism and French intersubjectivism 227; common ground 228–9; primacy of economy, anthropological foundations of 235–9; rationality and intersubjectivity 245–8; rationality and opacity 241–5; reservations about critical realism 229–34; symbolic exchange and social hypocrisy 239–41 ‘Économie des Conventions’ 24 education 256; institutions 88 effective demand principle 36, 44, 47, 51, 69–70 efficient market hypothesis 104 Eichner, A. 36, 37 elaboration 216 Elder, G. 135 electoral process 143 Elson, D. 108 Elster, J. 161 embedded rationality 139 embeddedness 168, 169–73, 175–7, 196; see also agency-structure model and embeddedness in heterodox economics emergence 47–8, 232 empirical: domain 4, 39; feminists 112; material 123–5; realism 101; realm 111 empiricism 190, 276–8, 279 English language 60 enterprise 63 entrepreneurial activity 71, 281, 282 entrepreneurial system 44

epistemology 126, 230; feminist 115–17 equilibrium 38, 204 ergodicity 40, 41, 42, 50, 51 Eriksson, R. 86 essentialism 118 Europe 146–7 everyday behaviour 281 evolutionary economics 2, 84, 207 existence 78, 84–6 experimental results 68, 69 experimentation 16 exteriority 248; of third parties 237–8 external: critique 234; relations 8; truth 240, 241 extrinsic reason 163 femininity 108, 113, 115, 119–20, 121, 125, 126 feminist economics 2, 15, 20, 107–28, 148–9, 207; central issues 108–11; definition 114–15; empiricism 109, 110, 112; epistemology and ontology 115–17; gender and economic science research project 120–5; gender, theorisation of 112–14, 117–18; methodology 288; postmodernism 110; power, conceptualisation of 119–20; standpoint 109–10; Western subject, impact of 118–19 Ferber, M.A. 108 Ferguson, A. 229 feudal system 146–7 Feyerabend, P.K. 189, 289, 294 field work 88, 89 Figart, D.M. 115 Fine, B. 23–4, 202–23 firm 63 first-order model 80 Fleck, L. 289 Fleetwood, S. 2, 13, 34, 96, 190, 193; macroeconomic theory and capitalism 55, 67, 68, 69 Fligstein, N. 180 Folbre, N. 108 Foray, D. 218 forecastability 100, 104, 105 foresight, perfect 82 formalism 190, 191, 194, 213, 252, 262 Foss, N.J. 174 Fox, A. 178, 179 Frank, R. 161 Frazer, J. 235 free rider problem 254

Index 305 free will 80 freedom of choice 256 Freeman, C. 269 French intersubjectivism 24, 227, 245–8 French School 24 Freud, S. 62, 156, 235 Frey, B.F. 179 Frey, B.S. 163 Friedman, M. 189, 289 Fullbrook, E. 227, 248 Fuller, S. 289 functionalism 207 Fundenberg, D. 190 Galileo 232 game theory 24 Gecas, V. 133, 134 gender 20, 107–11, 119, 123–4, 126; concepts 125; differences 121; identity 113, 125; individual 113; language 125; notions, normative 113; perceptions 121; structure 113, 114, 125; symbols 113, 125; theorisation 112–14, 117–18; see also gender and economic science research project gender and economic science research project 120–5; aim and scope of project 120–1; economic texts as empirical material 123–4; empirical and theoretical results 124–5; theorising gender, economic science and society 121–3 general equilibrium theory 190, 230 general-rules-based methodology 287, 288 generalisation 107, 114 Giddens, A. 48, 136–7, 141 gifts 239, 240, 241 Girard, R. 234, 235–7, 240, 241, 245 goals 169 Goffman, I. 135 Graça Moura, M. da 25, 268–84 Granger causality 76–7 Granger, C.W.J. 19–20, 96–106 Granovetter, M. 22, 168–77, 180 Guillén, M.F. 171 Hahn, F.H. 36, 40, 44, 45, 103, 180, 190, 211–12 Hamouda, O. 36, 37 Hands, D.W. 25–6, 103, 208, 286–98 Haraway, D. 117 Harcourt, G.C. 36, 37, 56, 68, 69, 71

Harding, S. 15; feminist economics 107, 109–10, 112, 113, 115, 116, 117, 119, 121 Hardy, C. 179 Hargreaves Heap, S. 21–2, 152–65 Harley, S. 205 Hausman, D.M. 208, 215, 291, 295 Hawk-Dove game 158 Hay, C. 178 Hayek, F.A. 57, 59, 193, 194, 195, 196, 197, 246–6, 247–8 Hegel, G.W.F. 233, 242, 291 Hendry, D. 104 heterodox economics 152–65, 168; mainstream economics, attack on 160–3; ontology: situated rationality 157–9; positivism and idealism 153–7; see also agency-structure model and embeddedness in heterodox economics; heterodox economics and ontology heterodox economics and ontology 1–28; socio-economic world 3–26 Hewitson, G. 108, 112, 115 Hicks, J.R. 42, 63 higher-order model 80 historical age 135 historical time 45 Hobbes, T. 173, 276 Hocart, A. 235 Hodgson, G.M. 15, 33, 137, 205 Hofstadter, D. 228 Hollis, M. 79 Holt, R. 36 Horkheimer, M. 161 human agency 4–10, 13, 18–19, 21, 23, 216; economics of institutions and institutions of economics 253; Economics as Social Theory and New Economic Sociology 167, 174, 175; economics as symptom 228, 240; feminist economics 111, 119; heterodox tradition 152, 155, 159, 161, 162; macroeconomic theory and capitalism 60; methodology 293; Post Keynesian economics 38, 39, 41, 42, 43, 47, 48; really real in economics 196 human rights 145, 146 Hume, D. 68, 76–7, 153, 276 idealism 153–7 idealization 82 identity 135, 169

306

Index

imitation 245–6, 247–8 immutability 42 inconsistencies 268, 269, 274, 276, 279, 281 individual rationality 258 individualism 23, 81, 196, 204, 238 inflation 65, 66, 100 information, perfect 82 Ingham, G. 59, 62, 103 innovation 269, 271, 272, 281 input-output approach 87 institutional development 86–91 institutionalism 86 Institutionalist economics 148, 149, 150 institutions 84–6, 133, 142–3, 144, 145, 150 institutions of economics see economics of institutions and institutions of economics intentionality 139, 140, 286, 293, 294 interaction, individual 147 interest rates 66, 67 internal: critique 234; relations 8; truth 240, 241 intersubjectivism 2, 227, 245–8 intrinsic reason 163 investment 65, 88 irreducibility 158, 159 Jacobsen, J. 108 Jevons, W.S. 124, 125, 278 Journal of Post Keynesian Economics 36 Kaldor, N. 35, 57 Kalecki, M. 37, 38, 57, 59, 71 Kant, I. 91, 294 Keuzenkamp, S. 121 Keynes, J.M. 37, 38, 41, 44, 46, 47, 245; macroeconomic theory and capitalism 57, 59, 62, 63, 65–7, 69–70, 71 Keynesianism 35, 46, 103 Kindleberger, C.P. 62 Kirman, A.P. 57 Kirzner, I. 194 Kitcher, P. 289, 294 Klamer, A. 36, 55, 192 Knapp, G.F. 65 Knorr Cetina, K. 290 knowledge 107, 218; common 242, 243, 244; shared 242, 243; theory 275, 276–7 Kornblith, H. 294

Kregel, J. 36, 37 Kuhn, T.S. 156, 160, 289, 290, 291, 294 Kuiper, E. 20, 107–28 labor 64, 66, 71 Labour, B. 290 Laing, R.D. 141 Lakatos, I. 156 Laudan, L. 294 Lavoie, D. 17 Lavoie, M. 37, 68 Lawson, T.: agency-structure model and embedded individual 132–3, 136, 137–41, 142–3; critical and real in critical realism 204, 206, 208–10, 217–18, 219; econometrics 96, 98–103; economics of institutions and institutions of economics 253–4, 262; Economics as Social Theory and New Economic Sociology 167, 170, 171, 172, 174, 175, 176; economics as sypmtom 227, 229, 230–1, 233, 240–1, 245; feminist economics 107, 111, 112, 114, 115, 117, 118–19, 120, 124; heterodox economics and ontological turn in methodology 1–2, 3, 4, 6–7, 8, 10, 11–2, 13, 14, 16–17, 18; heterodox tradition 153–6, 157, 158–9, 160, 161, 162; macroeconomic theory and capitalism 55, 56, 58, 59–61, 68, 70; Post Keynesian economics 34, 35, 36, 38–9, 40, 42, 43, 48, 50; really real in economics 187, 188–9, 190–2, 193, 194, 195, 198; recent economic methodology 286, 288, 293, 294, 295; scientific exegesis and Schumpeter 268, 274–5, 276, 277–8, 279, 281; transformational growth 77–8, 81, 82–3 Layder, D. 5, 7, 8, 9–10, 175, 179 Lazaric, N. 179 Lazear, E. 209 Learner, A. 104 Lee, F.S. 35, 70, 205 Leibniz, G. 244 Leijenaar, M. 114 Lerner, A. 57 Leslie, T.E.C. 49 Lestringant, F. 205–6 Lévi-Strauss, C. 239–40 Lewis, J.D. 171

Index 307 Lewis, P.A. 1–28, 78, 167–82, 205, 253; macroeconomic theory and capitalism 55, 56, 58, 67; really real in critical realism 187, 188–9, 194, 196, 198 life cycle 84 lifestyles 86 Locke, J. 276 Longino, H. 108 Lucas, R.E. 59, 70, 105 Luckmann, T. 174 M-C-C1-M164 McCloskey, D.N. 160, 190, 192, 291 McGuire, P. 174 Mach, E. 189 Machiavelli, N. 241 McKenna, E.J. 4 macroeconomic theory and capitalism 19, 55–72; methodology 58–61, 67–70; microfoundations literature 56–8; monetary theory of production 62–5; political economy 65–7 macroeconomics 50, 97, 190 mainstream economics 132, 160–3, 203–6, 208–9, 211–17, 219; economics as symptom 228, 230, 231, 233, 234, 239, 242, 244, 245; methodology 287, 288–9, 295–6 Majewski, R. 90 Mäki, U. 291 Malinowski, B. 278 Malthus, T. 206 Manent, P. 241 Maris, G. 62 Markov chain analysis 218 Marsh, D. 8 Marshall, A. 59, 84, 89, 110, 124, 125, 209 Marterbauer, M. 65 Marx, K. 59, 62, 64, 71, 214, 237, 259, 274 Marxist economics 2, 23, 148–9, 203, 207, 216, 220, 274, 288 masculinity 108, 109, 113, 115, 116, 119–20, 121, 124, 126 Maskin, E. 190 mathematics 67–8, 69, 70, 71, 125, 231–3, 258–62 Mauss, M. 234, 235, 239, 240 maximization 254–8, 262 means 133 mechanisms 77, 154 Meijer, M. 124

Menger, C. 57, 63, 278 metaphysics 243–5 methodological: debate 58–61; distinctiveness 36; individualism 23, 81; institutionalism 86; issues 204–8, 210–16, 219–20; structuralism 81 methodology 39, 230, 231, 233, 287–8; see also economic methodology microeconomic theory 61, 97 microfoundations literature 56–8 Mill, J.S. 110, 171, 209, 291, 295 Miller, R.W. 279 Mises, L. von 57, 195 modernism 118, 192 monetary theory of production 62–5 money 45, 71 Montesquieu, C. de S. 237 Moore, H.L. 79 Morgan, M.S. 291 Morrison, M. 291 motivation 172, 173 Mutari, E. 115 mythology 236 Nash equilibria 162 nation 85 natural sciences 16–18, 97, 98 Nell, E.J. 19, 57, 76–94 Nelson, J.A. 108, 113, 115, 118 Nelson, R.R. 272 neo-Marxism 227 neoclassical economics 36, 51, 152, 210; economics as symptom 227, 231; feminist economics 108, 109, 110, 114; macroeconomic theory and capitalism 57, 59, 61, 63, 65, 67, 69; transformational growth 76, 77, 79 new economic methodology 287, 288 New Economic Sociology see Economics as Social Theory and New Economic Sociology Newton, I. 232 Nietzsche, F.W. 237 non-reductionism 228–9 Nonaka, I. 179 nonergodicity 41 norms 172, 173 objectivity 110, 153, 157 obligations 86–7, 88, 139, 177 O’Gorman, P.F. 103, 291 Okishio, N. 216 Old Economic Sociology 168

308

Index

old institutionalist economics 2, 15 Oldersma, J. 114 O’Neill, O. 82 ontology 39, 91, 268; economics as symptom 228, 230, 231, 234; feminist economics 107, 108, 109, 110, 111, 114, 115–17, 126; heterodox tradition 161, 162; irreducibility of 275–6; macroeconomic theory and capitalism 59–60; philosophical 77; really real in economics 188, 189, 193, 194, 195–6, 197; situated rationality 157–9 opacity 241–5 open-system models 14, 18–19, 39–47, 80–1; economics of institutions and institutions of economics 252–3; Economics as Social Theory and New Economic Sociology 174; heterodox tradition 154; macroeconomic theory and capitalism 61, 68, 71; Post Keynesian economics 47, 51; really real in economics 192; scientific exegesis and Schumpeter 281 ordering axiom 41 organicism 69 Orléan, A. 227, 247 orthodox economics 1–2, 36, 157; Economics as Social Theory and New Economic Sociology 169, 171, 174 Osterloh, M. 179 output 88 Padgett, J.F. 174 Pagano, U. 24–5, 252–66 Pareto, V. 259 Parsons, S. 86 Parsons, T. 168 path dependence 207, 217–20 Pencavel, J. 124, 125 personal values 143, 145–6, 147 Peter, F. 116, 118–19 Petty, W. 124, 125 Phillips, A. 269 philosophical empiricism 276–8, 279 philosophical foundations 152 philosophy of science 288–91 Poincaré, J.H. 189 political economy 23, 65–7, 214, 217, 220 Pólya urn model 218 Popper, K. 25, 189, 295

population size 102 Porpora, D.V. 7, 8, 176 positional goods 259, 260 positive autocorrelations 104 positivism 152, 153–7, 160–1, 163, 189, 193, 240, 289 Positivist and Popperian 290–1, 294–5 Post Keynesian economics 2, 15, 18, 33–52, 148, 149, 203, 206–7; critical realism 38–9; methodology 288; open systems and transmutability 39–43; open-systems approach 43–7; uncertainty as emergent property 47–9 post-modernism 152, 153, 156, 211, 213, 217 post-structuralism 235 power 23, 107, 111, 119–20, 123, 139, 154, 177–80 Power, M. 115 practical reference 78 practices 85 pragmatism 156, 163, 291 Pratten, S. 13, 174, 180 Pressman, S. 36 primacy of economy, anthropological foundations of 235–9 prisoner’s dilemma 161 private goods 259–60 producing institutions 88 property rights 158–9 psychological motives 62 public goods 259–60 pure theory 270–1 purposive level 159 Putnam, H. 189 quasi-common knowledge 242 Quine, W.V.O. 156, 289 Radcliffe-Brown, E. 278 radical economics 148, 149 random walk model 104 rational choice theory 24, 161, 162–3, 242–3, 244, 258; Economics as Social Theory and New Economic Sociology 168, 169, 170 Rational Economic Man 118 rational expectations model 79, 247 rationality 38, 241–8, 260; bounded 257; collective 258; embedded 139; individual 258; situated 138, 157–9, 161, 162; unbounded 255 real domain 39

Index 309 real realm 111 ‘Realist Theory of Economics’ 96, 100, 101, 103 reality, definition of 96–8 really real in economics 187–99; critical realists 190–3; critical realists as critics 193–4; critics, response to 194–7; realism revisited 189–90 realms 111 Received View 289–90, 291, 292, 293, 294, 295 recent economic methodology 288 reciprocity 147, 177–80, 240 Redman, D. 103 reducibility 158, 159 reductionism 60, 70, 213 Reed, M.I. 175, 176–7, 180 referring 78 reflexive monitoring level 159 regularities 104 reification 139, 140, 148 relativism 152, 153, 154, 155, 156, 157, 163, 290 religion 235–6, 237 Reny, P. 244 representative agent 82 reproduction 112, 139, 140, 141, 142, 144, 146 retroduction 155, 156 Ricardo, D. 59, 259, 272 rights 177 risk 48–9 rituals 235, 236 Rizzo, M.J. 15 Robbins, L. 209 Robertston-Smith, W. 235 Robinson, J.V. 35, 37, 46, 57, 64, 67 Rogers, C. 62 role 85 Rorty, R. 156 Rosenberg, A. 291, 294 Rosenberg, M. 133, 134 Rotheim, R.J. 44, 45, 56, 62 Rothman, H. 205 Rubinstein, A. 190 rules 87, 139, 140, 141, 157–8, 172, 173, 193 Runde, J.H. 4, 13, 78, 81–2, 172, 195, 253; macroeconomic theory and capitalism 55, 56, 58, 67 Rutherford, M. 144 sacred 236, 237 sacrifice 235, 236

Sade, Marquis de 206 Sako, M. 179 Samuels, W.J. 15, 195 Samuelson, P.A. 59, 79, 188, 190 Sargent, T.J. 57 Sartre, J.-P. 227 Savage, L. 41 Sawyer, M. 34, 36, 38, 42, 46, 47 Sayer, A. 5, 179 Say’s Law 36 Scarth, W.M. 56 Schaapman, M. 121 Schelling, T.C. 190 Schoeneman, T. 135 Schotter, A. 157–9 Schulz, H. 79 Schumpeter, J.A. 25, 49, 62, 64, 71, 268–84 Schumpeter Mark I 269, 270, 271 Schumpeter Mark II 269–70, 272 science 109, 206, 215, 229; see also economic science scientfic exegesis and Schumpeter, J. 268–84; kaleidoscopic character of writings 269–72; philosophical empiricism and deductivism 276–8; principle of exegesis 272–4; vision, analysis and ontology 275–6 scientific changes 117–18 scientific theories 189 Scott, J. 107, 113–14, 117, 120, 122, 124 Scottish Enlightenment 229 seasonality 100 second-order model 80 self-appraisal 134, 135 self-concept 132–6, 138–41, 143–4, 146–7, 149 self-deception 240, 241, 242 self-esteem 134, 135 self-evaluation 134, 135 self-interest 170, 171, 173 self-regulation 245–6 self-understandings 138 Sen, A.K. 170 Sent, E.-M. 291 service institutions 88 Shackle, G.L.S. 41, 42, 49 shared knowledge 242, 243 shelf-of-scientific-philosophy 290–1, 292, 295 Shrauger, J. 135 Sims, C. 104 situated rationality 138, 157–9, 161, 162 slavery 146, 147

310

Index

Smart, J.J.C. 189 Smelser, N.J. 168 Smith, A. 59, 61, 110, 124, 125, 238, 245 Smithin, J. 19, 55–72 Smolensky, E. 57 social activity see transformational model of social activity social age 135 social construction of economic institutions 173–5 social economics 2, 148, 149, 150 social hypocrisy 239–41 social life 161 social order 13 social psychology 21 social science 97, 253–4 social structure 4, 5–10, 13, 17, 18–19, 21–4; economics of institutions and institutions of economics 252, 253, 257, 259; Economics as Social Theory and New Economic Sociology 167, 175, 176, 177, 180; economics as symptom 228, 229, 236, 240, 241; feminist economics 111; heterodox tradition 155, 157; macroeconomic theory and capitalism 56, 60, 61, 62, 65, 67, 69, 70; methodology 293; Post Keynesian economics 38, 39, 49; really real in economics 193, 195, 196; transformational growth 76, 83, 85 social values 133, 142–3, 144, 145, 149 socialism 210; centralised 282 socialization institutions 88 society 107, 121–3 socio-economic world 3–26; closedsystem models 11, 19; deductivism 11–16, 18, 23–5; Economics as Social Theory and New Economic Sociology 169, 171, 173, 174, 176, 180; embeddedness 20–1; human agency 4–10, 13, 18–19, 21, 23; natural sciences 16–18; open-system models 14, 18–19; orthodox economics 1–2; really real in economics 188; social structure 5–10, 13, 17, 18–19, 21–4; transformational model of social activity 4, 6, 8–9, 19, 20–1, 24 Solomon, M. 294 Solow, R. 36 Spinozism 244

Sraffian economics 288 state, role of 64–5 Stigler, G.J. 268, 272–4, 281 Stove, D. 189 Strassmann, D.L. 14–15, 17–18 Strober, M. 115 structural transformation 281 structuralism 81, 235 structuration theory 137 structure 77, 216 structured reality 253–4 structures 154 subjectivism 187, 193, 195 subjectivity 143, 144, 146, 148 Sugden, R. 157–9 Suppe, F. 289 Swedberg, R. 168, 169–70, 172, 173, 175 symbolic exchange 239–41 Sztompka, P. 5 tacit consciousness 140 Takeuchi, H. 179 tendencies 78–9, 216, 295 tensions 268, 274, 276, 279 Thatcher, M. 58 third-order model 80 Thomsen, E.F. 194 Thornton, H. 63, 71 time 45, 118 Tobin, J. 55, 57 Tonkens, E. 114 training institutions 88 transactions cost economics 168, 169, 170, 173–4 transcendental realism 101, 111, 230, 231, 233 transformation 112, 139–42, 144, 146, 281, 295–6 transformational growth 76–94; institutional development 86–91; institutions and existence 84–6; outline of critical realism 76–8; realism and determinism 78–81; uniqueness and change 81–4 transformational life cycle 84 transformational model of social activity 4, 6, 8–9, 19, 20–1, 24 transmutability 39–43 trust 22, 170, 172, 175, 177–80 trustified capitalism 269, 272 truth 240, 241 Turnovsky, S.J. 56 type 85

Index 311 Uebel, T. 289 unbounded rationality assumption 255 uncertainty 45, 47–9, 50, 51, 282; radical 18 unconscious level 159 unemployment 46, 100 uniqueness 81–4 United States 97, 147; Constitution 145–6 universalisation 107, 114 utilitarianism 279 value-neutrality 125 values 150, 169; personal 143, 145–6, 147; social 133, 142–3, 144, 145, 149 van Eeghen, P.-H. 295 Varela, F. 228 Varoufakis, Y. 162 Velthuis, O. 168 violence 236–7, 241 vision 275–6 volatility measures 104 voluntarism 4, 139, 148, 188, 253

wage labor 64, 71 Walras, L. 59, 230–1, 244, 259, 274, 278, 280 Weber, M. 48, 71, 235, 278 Weigert, A. 171 Weintraub, S. 57 Western Male Subject 125, 126 Western Modernist Subject 110 Western subject, impact of 118–19 Wible, J.R. 291 Wicksell, K. 71 Williamson, O.E. 173 Wilson, E.O. 80 Winslow, T. 62 Winter, S.G. 272 Wittgenstein, L.J.J. 157 Woolgar, S. 290 Working, E. 79 Working, H. 79 Wray, L.R. 65 Zannoni, D.C. 4 Zarnowitz, V. 103