Concepts in Action (Studies in Critical Social Sciences, 118) 9789004314191, 9789004314207, 9004314199

Rather than treating concepts and their application in a static and iconic manner, Concepts in Action provides us with e

117 94 2MB

English Pages 344 [342] Year 2017

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Concepts in Action (Studies in Critical Social Sciences, 118)
 9789004314191, 9789004314207, 9004314199

Citation preview

Concepts in Action

Studies in Critical Social Sciences Series Editor David Fasenfest (Wayne State University) Editorial Board Eduardo Bonilla-Silva (Duke University) Chris Chase-Dunn (University of California-Riverside) William Carroll (University of Victoria) Raewyn Connell (University of Sydney) Kimberlé W. Crenshaw (University of California, la, and Columbia University) Heidi Gottfried (Wayne State University) Karin Gottschall (University of Bremen) Mary Romero (Arizona State University) Alfredo Saad-Filho (University of London) Chizuko Ueno (University of Tokyo) Sylvia Walby (Lancaster University)

volume 118

The titles published in this series are listed at brill.com/scss

Concepts in Action Conceptual Constructionism Edited by

Håkon Leiulfsrud Peter Sohlberg

leiden | boston

Cover illustration: “Text on shelf”. Source: Pixabay.com – via www.pexels.com. The Library of Congress Cataloging-in-Publication Data is available online at http://catalog.loc.gov lc record available at http://lccn.loc.gov/2017042377

Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface. issn 1573-4234 isbn 978-90-04-31419-1 (hardback) isbn 978-90-04-31420-7 (e-book) Copyright 2018 by Koninklijke Brill nv, Leiden, The Netherlands. Koninklijke Brill nv incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi, Brill Sense and Hotei Publishing. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill nv provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, ma 01923, usa. Fees are subject to change. This book is printed on acid-free paper and produced in a sustainable manner.

Contents Preface vii List of Illustrations viii Notes on Contributors ix 1 Conceptual Constructionism: An Introduction 1 Peter Sohlberg and Håkon Leiulfsrud 2 On the Near Disappearance of Concepts in Mainstream Sociology 23 Richard Swedberg

part 1 Methodological Programs and Applications 3 What Do We Do with Norms—Conform, Break, Understand or Explain? 43 Peter Sohlberg 4 Colligation 63 Richard Swedberg 5 Sensitizing Concepts in Action: Expanding the Framework 79 Anne Britt Flemmen

part 2 Culture, Nature and Consumption 6 Culture as a Sociological Concept 97 Willy Martinussen 7 Bringing Nature Back In 116 Willy Guneriussen 8 The Study of Consumption in Sociology—Beyond Utility Theory 136 Pål Strandbakken

vi

Contents

part 3 Social Structure, Organizations and Institutions 9

Social Structure 153 John Scott

10

The Organization of Action 172 Göran Ahrne

11

About Actors: An Institutional Perspective 189 Raimund Hasse

12

The Family and Interwoven Concepts 206 Håkon Leiulfsrud

13

Collective Action: Why is it so Difficult for the Social Sciences to Grasp the Rational Aspects of Collective Action? 222 Roar Hagen

part 4 Class, Gender, Race and Social Recognition 14

The Status of the Political in the Concept of Class Structure 243 Erik Olin Wright

15

Gender as Analytic, Political and Interdisciplinary Concept 264 Harriet Bjerrum Nielsen

16

Race: A Contested and Travelling Concept 284 Mette Andersson

17

Recognition: Conceptualization and Context 302 Antje Gimmler

Index 321

Preface Thanks to a generous project grant from The Norwegian Research Council, we have been able to establish a group of Nordic, European and North American sociologists working with theory and theory development. This has enabled us to meet, to have seminars, to offer joint PhD courses, and to develop a more explorative and creative approach to theory in interaction with master and PhD-students. This is a second volume published by Brill with a focus on theory and concepts in use. It is sincere hope that both volumes will inspire students and scholars to engage in a critical discussion how to deal with theory and concepts in practice. Professor Aaron. V. Cicourel’s participation in the project has been inspiring and full of joy. On behalf of all the contributors to this volume we would like to acknowledge our gratitude to professor David Fasenfest as the editor of the series in which this work is published. In addition to David, we would also like to thank Marita Løkås for superb editorial assistance and to Theodore Pride for an excellent job with the index. It has also been a real pleasure to work with Evelien van der Veer at Brill and our production editor, Judy Pereira. Trondheim 30 June 2017

List of Illustrations Figures 2.1 Percent of articles in the American Sociological Review (1936–2012) containing the words “variable” or “variable” and “concept” 32 2.2 Percent of articles in ajs (1895–2012) and Social Forces (1925–2012) containing the words “variable” or “variable” and “concept” 32 2.3 Percent of articles in all sociology journals (1936–2012) containing the words “variable” or “variable” and “concept” 32 14.1 Interconnections among core elements of class analysis 245

Tables 14.1 Typology of modes of production 260 14.2 The political dimension of class formation 261

Notes on Contributors Göran Ahrne is Professor emeritus at Department of Sociology, Stockholm University. His empirical research covers a wide variety of fields: class structure, organizations and social relations such as friendship and love. His main research interest is to integrate social theory and organization theory. His recent publications in English cover topics as meta-organizations; the significance of partial organizations and the organization of markets. Mette Andersson is Professor at the Department for Sociology and Human Geography at the University of Oslo. Her research interests are cultural and political sociology, focusing on issues related to migration, ethnicity, identity and the public sphere. Andersson’s publications include articles, books and book chapters on the meaning of religion, elite sports, identity work and political engagement among ethnic and religious minorities. Harriet Bjerrum Nielsen is Professor at the Centre for Gender Research at the University of Oslo, N ­ orway. Her main research interests are gender socialization, gender identity formation and gender constructions among children and adolescents. She works within a psychosocial perspective and has a special focus on the ­implication of ­social change. Her most recent publication Feeling gender (Palgrave-­Macmillan, 2017) is a study of gender and class in three generations of Norwegian women and men. Anne Britt Flemmen is a Professor of Sociology at the Department of Social Sciences, UiT, The Arctic University of Norway. Her research interests and publications are mainly on issues related to migrasjon and integrasjon, equality, and gendered and ethnic difference in Northern Norway and Ethiopia. Antje Gimmler is professor of Applied Philosophy at the Department of Learning and Philosophy at Aalborg University and director of the Center for Applied Philosophy at Aalborg University. Her main research fields are social and political philosophy, philosophy of health and technology as well as applied philosophy of s­ cience. In her research she focuses on the interplay of theory and practice with a theoretical framework derived from philosophical pragmatism and critical theory.

x

Notes on Contributors

Willy Guneriussen is Professor of Sociology at the Department of Social Sciences, UiT, The ­Arctic University of Norway. His main fields of research are modernization and ­globalization processes; traditions, paradigms and foundational problems in sociology/social science; social theory and the understanding of nature. He has written extensively on all of these issues. Roar Hagen is Professor of Sociology at the Department of Social Sciences, UiT, The Arctic University of Norway. His main focus of interest is theoretical sociology, ­especially theories of collective action, systems theory and functional ­differentiation as the principle of integration of modern societies. Hagen has worked with methodological issues under the label methodological ­constructivism. His publications include work on neo-liberalism in social ­science, rational solidarity and functional differentiation, and collective power. Raimund Hasse is professor of Sociology, Organization and Knowledge at the University of ­Lucerne (Switzerland). His research interests include institutional theory, organization research, and economic sociology. Recent research projects address institutional discrimination in education, the adoption of social standards, and competition. Håkon Leiulfsrud is Professor of Sociology at Norwegian University of Science and Technology (ntnu), Department of Sociology and Political Science. Before coming to Trondheim, he worked at Stockholm University and Uppsala University. His research interests and publications are mainly in the fields of social class and social stratification, industrial relations, family, and in applications of ­sociological theory. Willy Martinussen is Professor emeritus of Sociology at Norwegian University of Science and Technology (ntnu). In addition to general sociology, his research interests are democracy and political participation, social inequality and welfare distribution, culture and religion. He has published books and articles on these subjects, as well as several influential textbooks in sociology. John Scott holds Honorary Professorships at the Universities of Essex, Exeter, and Copenhagen, having previously worked as Pro-Vice Chancellor at the University of

Notes on Contributors

xi

Plymouth and Professor of Sociology at the Universities of Essex and Leicester. His research has covered social stratification and the study of elites, corporate ownership and control, social network analysis, and social theory. He is currently completing a study of the development of social theory in Britain. He is a Fellow of the British Academy, was awarded a cbe for Services to Social ­Science, and is a recipient of the Distinguished Service to British Sociology Award of the British Sociological Association. Peter Sohlberg is Professor of Philosophy of Social Science at Norwegian University of ­Science and Technology (ntnu). Before coming to Trondheim he has worked at ­Uppsala University and Stockholm University. His research interests are ­philosophy of science, with special focus on ‘knowledge-generating’ research strategies in the social sciences and sociological theory. He has published within these fields, as well as in empirical sociology and social work. Pål Strandbakken is Researcher at Consumption Research Norway and Akershus University College of Applied Sciences. His main research interests are consumption research, household energy consumption, material culture and sociological theory. His most recent publication is The consumer in society, co-edited with Gronow (Abstrakt, 2015). Richard Swedberg is Professor of Sociology at Cornell University, Department of Sociology since 2002. His two specialties are economic sociology and social theory. Before coming to Cornell, he worked at Stockholm University. His works include Max Weber and the idea of economic sociology (Princeton University Press, 1998), Principles of economic sociology (Princeton University Press, 2003) and The art of social theory (Princeton University Press, 2014). Swedberg is also known for his work on social mechanisms and has written on many of the classics, including Weber, Simmel and Tocqueville. Erik Olin Wright is Vilas Distinguished Professor of Sociology at University of Wisconsin—­ Madison. Professor Wright is a Marxist scholar world renowned for his theoretical and empirical contribution to class theory and class analysis and in egalitarian alternatives futures to capitalism. His has published 12 books, and edited six volumes within his ‘Real Utopia Project’. Professor Wright served as President of the American Sociological Association in 2011–12 and is a recipient of numerous awards and honors.

chapter 1

Conceptual Constructionism: An Introduction Peter Sohlberg and Håkon Leiulfsrud Introduction This is a book about words—words with a specific kind of power. More precisely, the topic is words loaded with theoretical meanings in the social sciences. In an earlier volume, we discussed theories, i.e., interrelated concepts or in other words conceptual structures (Sohlberg and Leiulfsrud, 2016). Concepts are the theoretically loaded elements of theories or, as the usual metaphor goes, the building blocks of theories. Paradoxically, concepts constitute the theoretical field of sociology to a high degree, but descriptions of sociology are often formulated in terms of theories rather than concepts. This formulation makes the history of concepts and conceptual development scattered, if not hidden. In this volume, the basic and underlying theme is what can be done with concepts in sociological theory. This book aims to demonstrate and discuss conceptual constructions and their uses in understanding and constructing the social world. This could be done in principal terms, but it could also be done in a more powerful way with examples and our main strategy is to use examples as argument. We will highlight the construction of concepts in s­ ociology, their variation, their functions and related controversies (cf. Swedberg, 2005a; Scott, 2006). Furthermore, we discuss the role of concepts and their importance for the living of our everyday life. It is namely, a fascinating thing that we as individuals and collectives interact with concepts. Throughout this book, we will elaborate the distinction between theoretical and social constructions. Theoretical constructions concern the construction of theoretical concepts in the social sciences, whereas social constructions concern how actors interact in social life. We thus distinguish between the overlapping aspects of lived living (social construction work) and understood living (theoretical construction work). The relationship between these constructions is basic for the complexity of social science. Our main focus is theoretical concepts as tools for understanding social life. For instance, in what way can the concept of culture make us understand what is ‘going on’ with young people? Two aspects need to be distinguished when considering concepts as heuristic and theoretical tools. One aspect concerns the very construction of the tools. How do we ‘invent’ or construct conceptual

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_002

2

Sohlberg and Leiulfsrud

tools? For instance, how do we theoretically construct the concept of culture? What properties do we give that concept? The other aspect concerns working with and applying these tools. What are the consequences of their use? What functions do they fulfil? In what way does the application of the concept of culture enlighten our understanding of the social world? By focusing on the building blocks of theory rather than theories in their totality, we also want to challenge the rigid understanding of the borders between theoretical vocabularies, traditionally associated with social structures, and the theoretical vocabularies traditionally associated with individuals and actors. These vocabularies represent separate traditions in sociology: one tradition focuses on actors, intentions, social action, and social interaction, while the other focuses on structures, systems, balance, integration and conflict. That means that as well conflict- as consensus- perspectives are represented with this kind of system vocabulary (Cuff and Payne, 1984). In social science discourse, expressions such as ‘the logic of x’ are common, (e.g., the logic of practice, the logic of power, the logic of collective action, the logic of scientific discovery). The underlying message is generally that the topics characterized by the prefix ‘logic’ not are random but systematic, and even more than, that they are a matter of logic. The idea that the interconnections between concepts in the social sciences should be a matter of formal logic is unequivocally false. The social sciences are not Euclidean; they are not characterized by axioms and theorems and purely formal relations. Attempts to build a more formal language structure in the social sciences have surely been made; however, the results are disappointing, and the cleavage between the logical purity of language and the fuzzy complexity of social reality is striking. The lack of strict, logically coherent structures in the social sciences does not imply that logic does not matter. Logical implications follow from statements, but the social sciences do not have a strict, coherent and formal logical structure. Vocabularies In using the concept of ‘vocabulary’, we want to emphasize that the constructions of the social sciences are generally formulated in a natural language and that the relations between concepts and within paradigms are substantial relations rather than logical ones. By substantial, we mean that the ‘operationalization’ or definitions of concepts are grounded in substantial interpretations rather than formal deductions. In the analytic philosophical discourse, as well as in some social science discourses, there is a tendency to over-formalize the social scientific language, implying rigorous formal relations where we actually

Conceptual Constructionism: An Introduction

3

have more loosely constituted vocabularies. These vocabularies are formed by substantial considerations as well as by traditions, but they are seldom the outcome of formal logical reasoning and purely deductive construction work.

Conceptual Spaces

Identifying the conceptual space is one way to approach the vocabulary in which a concept is embedded. Here, a conceptual space refers to how the concept discussed relates to other concepts, which can sometimes be a matter of theoretical “logic” but is more commonly a question of the contingent relations established in the sociological tradition. One example of conceptual space is Erving Goffman’s famous theatre metaphor using and transposing, e.g., the social impact on spatiality in the idea of front and backstage and extending the Shakespearean metaphor of life as performance on stage (Goffman, 1990). Another example of a conceptual space is the highly contested concept of function, which was not at all alien to Goffman, who was methodologically concerned with interrelatedness in the social world (Verhoeven, 1993; Chriss, 2003). By definition, functions are related to specific contexts and concern interrelations. Functions are generally defined in terms of ‘something’ being useful or necessary for ‘something’. Those ‘somethings’ can vary immensely in terms of content, making functional reasoning complex and diffuse in general terms (Sohlberg, 1999). By the very definition that functions are necessary/useful for something, the assumption of a function-context is implied. Therefore, discussing in terms of context-less functions is impossible. The further implication of this is that all functional reasoning could be characterized as structural-functional. To extend the conceptual space of functional reasoning, functional contexts can be labelled in abstract terms such as ‘structure’ and ‘system’, or they can be linked to more specific social aggregates (e.g., a social group, a nation, or global society). The concept of function also relates more indirectly to concepts of dysfunction, balance, integration, equilibrium and tension. In modern times, concepts such as strategy and projects are often, though not always, equivalent with functions.

Conceptual Essentialism

In this book, we are interested in variations in the use of concepts and the ways in which a specific concept can vary in different contexts. This strategy

4

Sohlberg and Leiulfsrud

of seeking variation in concepts is not self-evident. A common approach in conceptual discourses is to elaborate what concepts really are on a general level and then what respective concept is on a specific level, i.e., a convergent essentialist approach. This essentialist approach is generally implicit and has problematic implications and consequences. Beginning with the essentialist perspective applied on the general level, i.e., finding a common denominator of all concepts, this exercise is not particularly informative. Still, we will return to some of the general characterizations of concepts, as they identify some wide-ranging themes of importance for theoretical constructions. Although these themes are basic for gaining a general understanding of conceptual constructions, they are not very useful in scientific practice, i.e., in the application and understanding of concepts. Identifying the general idea or essence of the concept is a common exercise in discourses in philosophy and philosophy of science. These discourses are partly normative, providing instructions on how to construct concepts in an acceptable way. To a lesser extent, these discourses are ‘empirical’, concentrating on an analysis of the use of existing concepts. We will return to the different traditions used in the construction of concepts. In sum, identifying the essence or common denominator of all concepts does not provide much or useful substantial information. The substantial diversity of such concepts is too big to allow for any more profound findings concerning the Platonic idea of the Concept. The work executed by one concept is seldom equivalent with the job done by another concept. This is the problem with seeking the profound inner sense of concepts in general. What then about the convergent search for essences of specific concepts? When discussing specific concepts in this volume, we have attempted to provide an overview of lexical uses of a variety of concepts. Identifying the lexical uses generally means grappling with a variety of definitions. In the social sciences, an (unlikely) empirical possibility is that a specific concept has been exclusively used in one specific sense. Should it be the case that a concept has this very specific definition it is anyhow not an indication that the definition has a truth value in the essentialist understanding. Many of the theoretical debates and theoretical conflicts in the social sciences relate to different interpretations of concepts.

Relating to the Conceptual Universe of Sociology

This publication focuses on a selection of basic theoretical sociological concepts and combines a genealogical and analytical study of their constructions

Conceptual Constructionism: An Introduction

5

and applications. Presenting a selection of sociological concepts raises the question of selection principles. Does a representative sample of sociological concepts covering the main field of sociology exist? The specialization within sociology makes the search for dictionaries representing a common canon of concepts problematic. There exist certainly many sociological dictionaries dealing with sociological key-concepts. Some of these dictionaries are subdictionaries in the sense that they focus on a single concept, a single field of sociology or a single key figure in sociology, e.g., Max Weber (Swedberg, 2005b) and Pierre Bourdieu (Grenfell, 2012). Some dictionaries aim to cover sociology in general, which becomes ever more challenging (e.g., Scott, 2014; Abercrombie, Hill and Turner, 2006). As sociological discourse is developing in the direction of specialization, divided into highly specialized research traditions and tribes of researchers, it is also common to find sub-dictionaries, including those on critical theory ­(Buchanan, 2010), critical realism (Hartwig, 2007), and social movements (Chesters and Welsh, 2011), and even counter dictionaries (that are not necessarily very alternative). This volume will not contain the whole population of basic sociological concepts nor a ‘representative’ sample. We instead aim to choose some ­strategic concepts and to elaborate their variety, construction, functions and embeddedness in various contexts. This objective implies partly a de-­contextualization of the concepts, and we will also extend beyond the conventional applications of such concepts.

Theoretical and Social Constructions—Interacting with Concepts

Concepts constitute the foundation of theoretical constructions in all ­sciences; however, in the social sciences, they are uniquely characterized by their ­inclusion in everyday interactions and social constructions. The theoretical understanding of concepts such as social class, gender and identity have been formative for social action, individually and collectively in social movements (Poletta and Jasper, 2001). In recent decades, much has been written about the social construction of diverse social phenomena, following the phrasing of Peter Berger and ­Thomas Luckman (cf., e.g., Hacking, 1999; Burr, 2015). In the general discourse, the concept of social constructionism has almost been reified as a very specific standpoint in ontological and epistemological matters. Paradoxically, the discourse on social constructionism tends to be fragmented, highly programmatic and ‘chrono-centric’, lacking a clear distinction between the theoretical

6

Sohlberg and Leiulfsrud

­construction of concepts and various social constructions of social phenomena from an interactional or structural perspective. Here, social constructions refer to the actual interactions between actors. For instance, as actors, we by interaction construct friendships, seminars, parties or being parents. Theoretical constructions relate to how we construct the theoretical tools that we use to understand what is going on in a friendship, at a seminar or party, or in roles as parents. Relevant concepts in the sociological toolbox include communication, status, power, norms, culture and structure, among many others. This rather negative characterization of a general programmatic and essentialist discourse on social constructionism should be contrasted with the more varied elaboration of sociological concepts within the rich theoretical tradition in sociology (e.g., Marshall, 1992). In discussing social constructionism, it is often forgotten that in the sociological tradition from the beginning it has been self-evident that understanding social phenomena has always required theoretical constructions, such as concepts of class, social action, culture, anomie or solidarity. Therefore, the object of study is theoretically constructed and not observed in a raw sense without any preconceptions (cf., Bourdieu, Chamboredon and Passeron, 1991; Montuschi, 2003). An important contribution to the discussion of social constructionism is formulated by the philosopher Ian Hacking. One aspect of his critical discussion is basically that concepts in social science have the potential to influence the ‘objects’ of study, this in contrast with the situation in the natural sciences. Hacking describes the dynamics surrounding these kinds of concepts in terms of the ways in which they … draw attention to the principle of classification, the kind itself, which interacts with those classified. And, vice-versa, of course, it is people who interact with the classification … What was known about people of a kind may become false because people of that kind have changed in virtue of how they have been classified. hacking, 1999: 104

Hacking is in this discussion mainly concerned with concepts that classify individuals e.g., in the case of psychiatric diagnoses. One thing is different categorizations of human beings, where we can identify ourselves with the categorical labels (e.g., those embedded in systems of psychiatric diagnosis). Clearly, such categorization can influence our self-understanding and behaviour. This self-identification is also related to socially formative concepts such as social class and gender. These are social concepts in the sense that they presuppose a social structure, i.e., a gender structure or a class structure, in which

Conceptual Constructionism: An Introduction

7

the sub-categories of gender and class are respectively structurally related (see the Wright and Bjerrum-Nielsen chapters in this volume). What then about the structures relevant to Hackings’ discussion and in which diagnoses are embedded? Here, the categories in the diagnostic systems generally imply some sort of (at least statistical) deviation from normality. One can reasonably assume that the categories embedded in these diagnostic structures not are socially formative to the same extent as the social categories related to class and gender, though they can influence one’s perceived selfidentity. However, the formative potential of diagnoses is an open empirical question, not a principal matter. Characteristic for the social concepts of class and gender is that they contain respectively socially interrelated categories, which is not as much the case for diagnostic categories. To sum up, concepts that identify categories of human beings can clearly qualify as interactive concepts. In this volume, we discuss the concepts of class, race and gender as concepts identifying social categories embedded in theories. Another interesting question is whether other social science concepts, that do not identify categories of human beings, can be regarded as interactive. As human beings live in and with abstractions, this is certainly the case. To take some examples from the chapters in this volume, understandings of abstract concepts such as culture, structure, nature and norms clearly influence our self-understanding and actions. However, a homogenous understanding of these concepts in a specific group does not necessarily exist, but the concepts are acted upon by some, which is enough to make them relevant. For instance, take the understanding of indigenous cultures and the need to respect their autonomy. Here, the culture concept can be quite powerful. Another example of the self-understanding of actors being influenced by an abstract concept is the notion of racism, not (only) as a matter of individual racist actions but (also) as a structural phenomenon with all consequent implications (see A ­ ndersson’s chapter in this volume). Note that these implications are not a matter of empirical ‘consequences’ but are dependent on how we actually define the concept of structure. These implications, following from how we define structure, are surely empirical and real and experienced by people in their everyday life, but they are understood within the concept. Conversely, to attribute whatever concrete and real hardship to structural conditions is a theoretical interpretation based on the concept structure. The naïve realist assumption that such interrelations can be directly observed is highly problematic. These interrelations are instead a matter of conceptual interpretation. In this way, concepts are integrated into our everyday understanding and our social constructions in interactions. In symbolic interactionist phrasing, they are part of how we

8

Sohlberg and Leiulfsrud

define the situation, and they are also theoretically constructed in the social science discourse. When Anthony Giddens’ writes about the duality of structure and views structure not only as constraining but also as resources, active knowledge of interactive concepts can be considered a potential asset for actors (Giddens, 1976).

The Heuristic Potential of Homeless Concepts

The world of sociology is inhabited by a multitude of concepts, coined in different periods as the discipline developed. Some are regarded as obsolete; some are regarded as powerful; some are changing; and some are constant. The genealogy of concepts is the study of the different trajectories of concepts from a historical perspective. In the social sciences, the concept of genealogy is often associated with ­Michel Foucault, the development of discourses, and the understanding of their implicit meanings. Without adopting the rich metaphorical understanding of the concept implied by Foucault, it is possible to pragmatically exploit the conventional understanding of genealogy and track the historic development of concepts (see, e.g., Swedberg’s discussion of colligation and Scott’s discussion of structure in this volume). The systematization of a specific vocabulary in the social sciences is a rather late phenomenon, beginning to be intensified in the late 19th century. However, some of the concepts coined more recently may have equivalents in earlier times. The concept of function is probably this best example in this case. First used in mathematics (see a genealogy in Merton, 1968), this concept was introduced rather late in the social sciences. However, the understanding of functions in the social sciences, i.e., the idea of something being useful or necessary in a context, has its roots in the Aristotelian idea of the final cause, even though Aristotle does not use the term ‘function’. In his listing of different causes, Aristotle writes about the important final cause: ‘A fourth way in which the word [cause] is used is for the end. This is what something is for, as health, for example, may be what walking is for’ (Aristotle, 1999: Physics ii.3). In the theoretical grammar of sociology, analyses in terms of what ‘something is for’ are commonplace. The interesting point here is that the potentiality of what ‘something is for’ cannot be expressed in modern causal terms. ­Aristotle realized this limitation when he distinguished between the final cause (i.e., the function) and the efficient cause (i.e., the cause in the modern sense).

Conceptual Constructionism: An Introduction

9

In modern times, the concept of function has generally been theoretically embedded in a more abstract system-perspective than what follows from ­Aristotle’s example. Functions are given an important role as tools of analysis in Durkheim’s method (Durkheim, Lukes and Halls, 2014), and they were later systematized as a research programme, primarily in social anthropology (e.g., Malinowski, 1944). One of the problems with historic genealogies is that the embeddedness of specific concepts can be cemented by theoretical habits and traditions. For example, functional reasoning has become highly controversial in the social sciences. By contrast, in medicine and biology, among others, functional reasoning is considered an important kind of heuristic analysis. The narrative of functional reasoning in sociology is habitually formulated in terms of functionalism, the functional school, or structural functionalism. This deductive approach to ‘functional concepts’ generally implies that they exist within a narrow theoretical tradition and not beyond that. The consequence of this contextualization makes us less disposed to identify functional reasoning in other contexts. These contexts can, for example, include common reasoning in terms of quasi-intentional (unconscious) projects and strategies. If strategies and projects are inferred, without being intentionally acted upon by those who are supposed to participate in these projects and have these strategies, a functional context is constructed. The concrete behaviours of the actors are then subsumed into this function context. Beyond the traditional borders, functional analysis is relevant in the identification of dysfunctional structural contradictions, as in the Marxist tradition (Cohen, 1978). A common way of implicit functional reasoning involves identifying a functional context—though it is not referred to in those terms—and using this context as the framework for analysing processes and states of affairs. For Foucault, many concrete phenomena in society are understood as part of a disciplinary process. This assumed overarching disciplinary process is thus the functional context—in Aristotelian terms, the end for which something exists. Another example of the limitations of a deterministic context-bound understanding of sociological concepts is the concept of the social field. This concept is habitually and probably most commonly understood in the setting of Bourdieu; however, developing the concept in a wider context may be fruitful. In general terms, the idea of a social field is simply an application of a general structure-concept (Bourdieu, 1984, 2005). Concepts cannot always be fully understood or used to their full potential in the very specific theoretical context in which they were introduced. The argument of partial de-contextualization, as a way of revealing new theoretical

10

Sohlberg and Leiulfsrud

insights, is analogous to Goffman’s argument for the (partial) de-contextualization of specific interaction orders. Goffman’s idea is that different specific interaction orders (e.g., interruption rights) can be compared in ‘isolation’ from their original contexts (Goffman, 1983). The idea is that comparing the properties of interruption rights independently of their situatedness is interesting. As in the case of Goffman’s research, this comparative perspective on specific instances of the studied phenomena may uncover new insights that are otherwise concealed by their different contexts. This de-contextualization should not be regarded as the essential source of conceptual content, but it is certainly revealing insights into the potential applications of concepts, which are not accessible in or through the habitually chosen contexts or vocabularies.

The Birth of Concepts, Their Justification and the Theoretical Eye

There are several discourses on the birth of concepts. By ‘birth’, we are not referring to the long historical perspective of genealogy but to the more specific ‘invention’ and construction of concepts by scientists in their everyday ­activities. Viewed from an empirical perspective, the ample attention paid to this ‘invention’ of concepts is intriguing because it does not reflect the typical situation in most research. Researchers more commonly borrow established concepts and apply them in their specific settings. However, the borrowing of such concepts, which often means that these concepts are ‘planted’ in new contexts, is seldom discussed and problematized. The discourse on concept construction and formation covers a wide range of approaches. On one extreme, we have systematic and rather formalized instructions on how to correctly construct concepts in a scientifically rigorous way. This extreme belong more to a branch of the philosophy of science rather than being part of the actual everyday research practices in the social sciences. A classic example is Carl Hempel’s Fundamentals of concept formation in empirical science (Hempel, 1952). This discourse within philosophy of science is generally normative, i.e., prescribing the correct way to construct concepts. Therefore, the discourse is located within the context of justification, i.e., is related to the ‘correct’ or established way of argumentation within the tradition (generally the positivist tradition). However, one should be careful to not localize the normative view of concept formation exclusively within the positivist tradition. Even in qualitative traditions, there may be normative positions concerning the correct way to categorize empirical material and to construct concepts. On the other extreme are discourses that focus on the creative and heuristic aspects of concept formation, which do not have strict formal and prescriptive

Conceptual Constructionism: An Introduction

11

rules. This discourse is primarily located within the context of discovery, i.e., the actual creative processes of concept formation (cf. Swedberg’s chapters in this volume, Becker, 1998). From a traditional philosophy of science perspective, this aspect is not of interest, as this tradition has no accepted way of hand­ ling the fuzzy and idiosyncratic processes of creativity (e.g., Popper, 1992). Another less common discourse is the empirical description of the birth of actual concepts. Robert K. Merton has written such overviews of concept formation in the intersection between what we call the short-term birth of concepts and the long-term genealogy of concept formation (e.g., Merton, 1968). There is probably no way to charter in a systematic way the context of discovery concerning the invention and construction of concepts. There are no strict logical paths leading to creative concept construction. It is however to some degree possible to relate concept construction to the classic form of inferences, i.e., induction and deduction. Inductively formulated concepts would have their basis in patterns revealed by observation and generalization. Of course, inductively based concepts cannot be ‘mechanically’ established in an atheoretical way. As a mode of inference, induction is always open ended and is e.g., the basic way of forming categories in grounded theory. The inductive strategy in the social sciences generally follows a path where generalization leads to the identification of patterns. These patterns are, in turn, material for conceptual formations (e.g., in terms of categories, causal factors, and the properties of processes). This strategy is exemplified in numerous international classifications of health, including the International Classification of Functioning and Disability (icf) (who, 2001). Although icf is rooted in a scientific discourse that positions the laboratory as an ideal, the actual conceptual formation, i.e., instilling the concepts with substantial meaning, is performed within the grounded theory tradition. Therefore, patterns are programmatically identified from below, but, in practice, they are nonetheless bound by the theoretical rationale of the who classifications (Bichenbach, Cieza, Rauch and Stücki, 2012). This example illustrates the general problem with any ‘pure’ inductive path of conceptual formation; any observation is arguably dependent on a point of view or a theoretical perspective (Cicourel, 1964). The deductive way of forming concepts in the social sciences is mostly a matter of a theoretical operationalization or specification of a more general theoretical category, which is rarely a matter of formal logic. It is more of a theoretical interpretation of the most convenient ways of structuring an empirical field. In primarily non-formal sciences such as the social sciences, it is evident that the theoretical eye on concept formation is necessary as well from the inductive as the deductive perspective. In the inductive process, it is a matter

12

Sohlberg and Leiulfsrud

of identifying informative patterns in the material, which can constitute the basis for formulating a deeper understanding in terms of models and theories. These informative patterns are not a matter of pure quantitative assessments, but must be labelled in qualitative categories in order to be meaningful. In general, these qualitative categories have surplus meaning beyond their strict operationalization. This surplus meaning is a basic source of confusion as well of creativity in the theorizing process. This is the Janus-face of expanding the borders of understanding. In the deductive process, the theoretical eye has the role to make a transition from concepts integrated in a theoretical framework to corresponding concepts that are applicable in an empirical setting. Even if the basis of conceptual formation varies, there is not necessarily any reason why we should assume fundamental differences between induction and deduction in terms of how the theoretical eye functions. Independently, if the theoretical eye is based on a pre-formulated theory or observations, its function is to make understandable features of the social world, not reducible to ‘pure’ theory (for examples of the theoretical eye in action, see, e.g., Mills, 1961; Swedberg, 2016; Abbot, 2004; Becker, 1998). Concepts and their definitions are often said to lack truth value, but they should be assessed in relation to their fruitfulness. From the inductive perspective, fruitfulness means that the patterns expressed in the conceptualization identify important qualitative aspects to be further elaborated and eventually made into some form of model or theory. If so, fruitfulness from the inductive perspective equals deductive potential, i.e., what Merton refers to as the identification of new questions and avenues of research based on creative concepts (Merton, 1968). From the deductive perspective, fruitfulness means that the conceptualization based on general theory identifies important qualitative aspects to be ­investigated. Independently, if the process has a starting point in theory or observations, the assessment of the fruitfulness of concepts follows the same logic. Even if the approaches and strategies initially differ, finding an argument to distinguish between inductive and deductive fruitfulness is difficult. The deductive and inductive paths to concept formation certainly represent two different research traditions, ultimately the rationalist vs. the empiricist tradition. However, it is not logically necessarily the case that these paths are separated in the context of discovery. When Swedberg describes Weber’s idea of how to construct ideal types, he touches on the inductive as well as the deductive aspects. The ideal type ‘is essentially created through a combination of two mental operations: a synthesis of what we know about a phenomenon and an “analytical accentuation” of its key features’ (Swedberg, 2016: 6).

Conceptual Constructionism: An Introduction

13

The synthesis can be regarded as the inductive aspect, though with a theoretical eye, i.e., no mechanical generalizations. The ‘analytical accentuation’ can be regarded as the deductively established theoretical eye. This way of formulating ideal types is probably a much more realistic description of concept formation in scientific practice than the rather simplistic notion of inductive and deductive strategies. When discussing inferences, another supposedly more creative kind of inference than induction and deduction, i.e., abduction, is often mentioned. Abductive inferences are usually perceived as capturing real creativity in the research process (See e.g., Hagen, 2016). A formalization of an abductive inference is as follows: D is a collection of data. H explains D. No other hypothesis can explain D as well as H does. Therefore, H is probably true. josephsson and josephsson, 1996: 11

The authors also discuss how to evaluate an abductive inference in a rather elaborate way. However, this inference is not in any way a formal one, which is striking. There is no logical path by which to arrive at the conclusion. Seeing a concept as the outcome of an abductive inference is rather commonplace; however, this understanding is limited from a structural point of view, as a concept in itself does not have the explanatory power as an isolated tool. From a formalistic point of view, our conclusion is perhaps negative, i.e., there exist no stringent formulas or inferences by which to construct fruitful concepts. From another point of view this conclusion reflects the openness and importance of unpredictable creativity in scientific work.

Sociological Concepts in Action—The Functions of Building Blocks

The formation and construction of concepts—whether based on the exegetic exercises of the classics, spontaneous creativity, or inductive/deductive/­ abductive reasoning—is only half the story. This activity is idle if we do not consider the functions of concepts. This is generally done in a piecemeal way in research projects in line with the established practices of the tradition. The fundamental question is: What do we actually do in theoretical terms when we apply the concepts, what do we achieve? Why is science, as Herbert Blumer puts it, impossible without concepts (Blumer, 1931)?

14

Sohlberg and Leiulfsrud

This question is clearly too complex and far-reaching to address here in its totality. However, from our perspective, Blumer’s characterization of the functions of concepts is interesting: The scientific concept enables the isolation and identification of an abstracted content in experience which may become the subject of a ­separate study. Functionally, the scientific concept (1) introduces a new orientation or point of view, (2) serves as an instrument of handling one’s environment, and (3) makes possible deductive reasoning and so the anticipation of new experience. blumer, 1931: 515

This characterization of functions illustrates our thesis at the beginning of this chapter, i.e., that a fuzzy grey zone exists between theoretical constructions and social constructions. Blumer addresses both aspects, i.e., the intellectual understanding (e.g., deductive reasoning), but also the interactive aspect of the social actor—the ‘instrument of handling one’s environment’. Here, we have as well theoretical as social constructions. As part of our approach, we have also declared that little substantial information comes from discussing concepts in abstraction. From a substantial perspective, studying the specialized functions that specific concepts have is more interesting. We believe in argumentation in form of examples—not so much in general principles. A theme in this book is the cognitive, heuristic and theoretical functions of the specific concepts that we present. However, to preliminarily illustrate the idea of the functions of concepts, let us begin with some basic concepts in the social sciences. In addition to specifying their functions, we also mention two possible contexts for the use of the concepts, i.e., a vocabulary of action and a vocabulary of the system. The background for distinguishing between these two kinds of vocabularies is that they often are constructed as diametrically opposed, representing different ontological ideas of the social world. We are not here interested in participating in this debate in the abstract, but we are interested in how these perspectives are actively constructed via their respective vocabularies. When closely analysed, vocabularies of the system and vocabularies of actors tend to overlap, which is interesting in light of the polarization between actor-based and structural approaches as ontological perspectives. This fuzzy grey zone between the actor and the system can be regarded as problematic for those ­adhering to the purity of approaches. For those accepting as a social fact the enduring tension between individual reasons and structural limitations as a social reality, this grey zone is worth close attention and should not be ­neglected

Conceptual Constructionism: An Introduction

15

(Bourdieu, 1984, 2005). To analyse this grey zone based on concepts and vocabularies, may generate new and surprising insights. Notably, the concept at hand may be more or exclusively relevant in one context, but our idea here is to determine what will happen if the concept is applied in both contexts. Such investigations can be regarded as contrafactual thought experiments that aim to discern the potential of the concept in question.

The Function of Concepts in the Vocabulary of Systems and Vocabulary of Actions

In the coming chapters, we will discuss and apply different approaches to the contextualization of a variety of social science concepts. The following section is a simple and introductory exercise, in which we attempt to give a preliminary idea of what focusing on the theoretical and heuristic functions of concepts may mean in contexts of vocabularies of action and vocabularies of structure respectively. As illustrations, we have deliberately chosen concepts that are usually associated with a structural perspective. In this way, we think that the interactive grey zone between actor-based approaches and structural approaches becomes more visible.

Examples of Theoretical Functions with Concepts within the Vocabulary of Systems

Social Structure (i) To describe how elements are formed together based on some kind of relation. (ii) To specify a high-level aggregate (e.g., normative or economic) structure often with the assumption of some form of causal power following from this structure. Integration To identify some degree of compatibility between the parts and the whole. The concept can be used to describe the static outcome as well as the dynamic process. Solidarity To identify a quality of a social structure or a dimension in a systemic sense, describing forces that eliminate or hinder tension or contradictions between system parts.

16

Sohlberg and Leiulfsrud

Anomie A concept used to describe some sort of tension or conflict in a social structure, i.e., in general terms, the opposite of integration. The common use of anomie is more specifically related to the incompatibility of norms within a specific context.

Examples of Theoretical Functions with Concepts within the Vocabulary of Action

Social Structure A shorthand description of relations of interactions, i.e., the relational ­structure, e.g., formulated in sociograms. Integration To describe the degree of the actors’ participation in some kind of larger ­interaction structure. Solidarity Describing a feeling of belonging and cohesion among actors in a specific setting. Anomie A concept used by R.K. Merton to describe individual strategies based on the conflict experienced by social actors between established norms and the perceived possibility to fulfil those norms. This simple overview provides an idea of the flexibility of basic social science concepts and illustrates their functionality in different contexts. In the following chapters, a wide range of concepts will be presented and discussed more in detail.

Chapter Overview

Swedberg’s first chapter demonstrates the general lack of interest in concepts and theorizing in leading sociology journals, which is explained via hegemonic traditions and research schools that are primarily interested in research designs that are well suited to empirical research. Sohlberg’s chapter on what to do with the concept of norm illustrates a theoretical methodology and a view of concepts of an interactive kind.

Conceptual Constructionism: An Introduction

17

In line with the programme outlined in the introductory chapter, Sohlberg pays special attention to norms as heuristic devices for understanding action and actors’ ­patterns from the outside and to norm reasoning from a systems perspective. Although the concept of norms may appear fuzzy at first glance, his emphasis on its functions illustrates its crucial role in both actor and system vocabularies. Swedberg’s second essay in this volume on the concept of colligation is ­rooted in a more pragmatic approach to theory development as a craft and personal skill. Colligation, literally understood as ‘binding together through an idea’ based on facts, illustrates theorizing and concept construction used to invent or assess established or rival operationalizations of a concept. In this chapter, Swedberg provides us with a programme and some step-by-step guidelines, demonstrating the potential of colligation in the process of constructing and adjudicating concepts. In a similar vein, Flemmen argues that ‘sensitizing concepts’ in contrast to ‘definite concepts’ can be regarded as methodological devices in discussions of the relationship between data, theories and concepts. Her treatment of the concept of intersectionality illustrates the various functions of this concept in close dialogue with data and the social context in which it operates. For a sociologist, the concept of culture is of obvious interest in terms of what society includes, gives weight to, or excludes. Even more importantly, the concept of culture is a necessary source in explanations and interpretations of all human actions and interactions. Martinussen’s chapter is another illustration of a concept of an interactive kind dealing with cultural structures as theoretical abstractions, at the same time as their part and connections may be observed in people’s action and interactions. Guneriussen’s argument is that sociology generally lacks an adequate theoretical vocabulary of nature, apart from that related to the environment and natural resources. Rather than leaving ecological debates about ‘planetary boundaries’ to ecologists, climatologists and chemists, Guneriussen argues that we need to develop a sociological language and vocabulary of planetary boundaries in order to assess social and economic processes. According to Guneriussen, the added value of a broadened conceptual space is the elaboration of the social and institutional mechanisms at work in various socio-­ natural systems. Consumption is a concept that denotes a vocabulary of nature and culture in material and non-material consumption. In his chapter, Strandbakken focuses on what actors do in the realm of consumption, including the culture and norms associated with the concept. This chapter covers the symbolic value and cultural discourses of goods and services, including the fascinating ways

18

Sohlberg and Leiulfsrud

in which the idea of consumption is translated into a more mundane and integrated theoretical language and vocabulary of everyday life and lifestyles. In his chapter on the concept of structure, Scott argues that we need structural and action-orientated concepts that enable us to understand, anticipate and predict actual events. Scott refers to social structures as ‘hypothesised real structures’ that help us explore complex class relations, gender, family and kinship structures, etc. Scott’s essay exemplifies how a general concept is translated into a sociological understanding and vocabulary and emphasizes both institutional, relational and embodied structures and actors’ constructions of a sense of social structure. In line with Scott, Ahrne argues in favour of an interactive theoretical conceptualization. In this approach, the organization represents distinct ways of forming social relationships compared with the ways that institutions and networks do so. In contrast to much of organizational theory preoccupied with formal rules and structures, Ahrne examines concepts with an interest in the social mechanisms that form and transform individuals into actors. In line with Scott, one of Ahrne’s main claims is that organizations can be understood as actors and structures simultaneously. Hasse’s chapter on neo-institutionalism is another example of opening up a conceptual space to investigate how individuals, organizations and states tend to be constructed as social actors. In line with Scott and Ahrne, Hasse argues in favour of theory development of a more interactive kind, including context-bounded and socially embedded actions and interactions. In addition to discussing what constitutes an actor with decision-making qualities, Hasse’s argues that we have to pay special attention to the norms and institutional designs that govern different types of actor hoods (cf. also Sohlberg’s chapter in this volume). Leiulfsrud’s chapter on the family is an obvious example of a theoretical concept that may relate to a language that describes actors and their interactions or to an integral part of society and a vocabulary of systems; in fact, it may bridge the two. In concurrence with Swedberg’s concept of ‘colligation’, Leiulfsrud’s argument is that the family concept works as a Pandora’s box. R ­ evisiting and challenging established ideas and conceptualizations of the family and binding together various clusters of help concepts can be interesting and generate an infinite number of entries and potentially sociologically interesting questions. This concept may also be in danger of hyper-differentiation and deconstruction at the expense of a more common ground. In contrast to a more traditional sociological primed discussion, Hagen approaches the theory of social structure and action from a more meta-­theoretical

Conceptual Constructionism: An Introduction

19

perspective, showing an interest in the concept of rational ­collective action. Whereas rational collective action is believed to have the potential to transform actors and systems, collective identity is mainly regarded as a residual category by Hagen. This illustrates how a more meta- theoretical and analytical approach may leave us with different conclusions as opposed to theorizing embedded in an overall idea of for example social classes and collective interests (see also Wright’s chapter in this volume). Concepts such as social class, gender, race and recognition are examples of travelling and contested concepts, i.e., concepts that appear in several versions and theoretical universes and concepts that are heavily debated. Originally published in 1981, Wright’s chapter on the interrelationship between the Marxist understanding of class (structure) and politics is framed as a discussion of whether to address the concept of class in its most abstract form, grappling with ownership and property relations, or in tandem with the concepts of class formation and class struggle, including actors and institutions that shape political opportunities or constraints. Wright’s essay is of general interest as an example of how to work with concepts within a theoretical space. His essay is also written in alternative language to that expressing class as a variable or social fact, which highlights how the concept of class can be seen key element in a political struggle based on different interests and access to resources. Bjerrum-Nielsen’s chapter on gender provides us with an overview of a travelling and interdisciplinary concept that raises numerous questions of ontology, epistemology and politics. Despite numerous parallels with Wright’s conceptualization of social class and politics, Bjerrum-Nielsen’s chapter claims that gender in feminist theory tends to be used in a framework of social relations and an entity of meaning that is embedded in social interactions. This idea is articulated in discussions from the epistemological standpoint of women’s interests, values and identities or the attempts to break down and deconstruct normative gender and sexual practices in poststructuralist and queer theory. The link between the concepts used by social scientists and the ways in which these concepts are perceived in public life is the leitmotiv in Andersson’s discussion of how to approach the concept of race in Scandinavia. Andersson’s discussion of how to address the concept of race in Scandinavia is in line with the ideas of Wright and Bjerrum-Nielsen, but Andersson has a more ambivalent view of the potential of race as a sociological and analytical category. Andersson’s essay discusses a number of alternative interpretations of race as a social fact, including ethnicity and theoretical vocabularies, which involve actors doing and negotiating racial relations and perspectives on racialization.

20

Sohlberg and Leiulfsrud

The question of actors doing performing and negotiating social inclusion is also covered in Gimmler’s chapter on recognition. Recognition is another example of a travelling concept, in this case moving from philosophy into sociology, gender and disability research and establishing a conceptual space among welfare professionals and state policies. As with class, gender and race, the concept of recognition is intertwined with the normative spirit, diagnosis and critique of power relations in society. Although the concept of recognition is typically discussed in a theoretical vocabulary of actors, social interactions and individual/group identities, it also found in sociological discussions of ­inequality struggles and social justice. References Abbott, A.D. (2004). Methods of discovery: heuristics for the social sciences. New York: Norton. Abercrombie, N., Hill, S. and Turner, B.S. (2006). The Penguin dictionary of sociology. 5th edition. London: Penguin. Aristotle. (1999) [app. 350 b.c.]. Physics. Oxford, New York: Oxford World’s Classic. Becker, H. (1998). Tricks of the trade. Chicago and London: The University of Chicago Press. Bichenbach, J., Cieza, A., Rauch, A. and Stücki, H. (eds.). (2012). icf core sets: manual for clinical practice. Götingen: Hogrefe Publishing. Blumer, H. (1931). “Science without concepts”, American Journal of Sociology. 36(4): 515–533. Bourdieu, P. (1984). Distinction: a social critique of the judgement of taste. Cambridge, ma.: Harvard University Press. Bourdieu, P. (2005). The social structures of the economy. Cambridge: Polity Press. Bourdieu, P., Chamboredon, J. and Passeron, J.C. (1991). The craft of sociology: epistemological preliminaries. Berlin: Walter de Gruyter. Buchanan, I. (2010). A dictionary of critical theory. Oxford: Oxford University Press. Burr, V. (2015). Social constructionism. 3rd edition. London: Routledge. Chesters, G. and Welsh, I. (2011). Social movements: the key concepts. London: Routledge. Chriss, J.J. (2003). “Goffman as microfunctionalist”, Ed. A.J. Treviño. Goffman’s legacy. Lanham: Rowman & Littlefield. Cicourel, A. (1964). Method and measurement in sociology. New York: Free Press. Cohen, G.A. (1978). Karl Marx’s theory of history: a defence. Princeton, N.J.: Princeton University Press.

Conceptual Constructionism: An Introduction

21

Cuff, E.C. and Payne, G.C.F. (eds.). (1984). Perspectives in sociology. 2nd edition. ­London: George Allen & Unwin. Durkheim, E., Lukes, S. and Halls, W.D. (2014) [1895]. The rules of sociological method: and selected texts on sociology and its method. Free Press trade paperback edition. New York: Free Press. Giddens, A. (1976). New rules of sociological method: a positive critique of interpretative sociologies. New York: Basic books. Goffman, E. (1990) [1959]. The presentation of self in everyday life. London: Penguin. Goffman, E. (1983). “The interaction order. American Sociological Association, 1982 presidential address”, American Sociological Review. 48(1): 1–17. Grenfell, M. (ed.). (2012). Pierre Bourdieu: key concepts. Stocksfield: Acumen. Hacking, I. (1999). The social construction of what?. Cambridge, ma.: Harvard ­University Press. Hagen, R. (2016). “Abduction—assessing fruitfulness and the construction of scientific concepts”, Eds. P. Sohlberg and H. Leiulfsrud. Theory in action. Theoretical constructionism. Leiden: Brill. Hartwig, M. (ed.). (2007). Dictionary of critical realism. London: Routledge. Hempel, C.G. (1952). Fundamentals of concept formation in empirical science. Chicago: University of Chicago Press. Josephson, J.R. and Josephson, S.G. (eds.). (1996). Abductive inference. Computation, philosophy, technology. Cambridge: Cambridge University Press. Malinowski, B. (1944). A scientific theory of culture and other essays. Chapel Hill: ­University of North Carolina Press. Marshall, G. (1992). In praise of sociology. London: Routledge. Merton, R.K. (1968). Social theory and social structure. Enlarged edition. New York: Free Press. Mills, C.W. (1961). The sociological imagination. New York: Grove Press. Montuschi, E. (2003). The objects of social science. London: Continuum Books. Poletta, F. and Jasper, J.M. (2001). “Collective identity and social movements”, Annual Review of Sociology. 27: 283–305. Popper, K. (1992) [1959]. The logic of scientific discovery. London: Routledge. Scott, J. (ed.). (2006). Sociology: the key concepts. London: Routledge. Scott, J. (ed.). (2014). A dictionary of sociology. 4th edition. Oxford, uk.: Oxford University Press. Sohlberg, P. (1999). Mål och mening i samhället: funktionalistiska program i samhällsvetenskapen. [Ends and meanings in society: the functionalist programmes in social science]. Uppsala: Acta Universitatis Upsaliensis. Sohlberg, P. and Leiulfsrud, H. (2016). Theory in action. Theoretical constructionism. Leiden: Brill.

22

Sohlberg and Leiulfsrud

Swedberg, R. (2005a). Interest. Maidenhead: Open University Press and McGraw-Hill House. Swedberg, R. (2005b). The Max Weber dictionary: key words and central concepts. ­Stanford, ca.: Stanford Social Sciences. Swedberg, R. (2016). “On the heuristic role of concepts in theorizing”, Eds. P. Sohlberg and H. Leiulfsrud. Theory in action. Theoretical constructionism. Leiden: Brill. Verhoeven, J.C. (1993). “An interview with Erving Goffman, 1980”, Research on Language and Social Interaction. 26(3): 317–348. who (2001): International classification of functioning, disability and health. Geneva. Word Health Organization.

chapter 2

On the Near Disappearance of Concepts in Mainstream Sociology Richard Swedberg Introduction That you need to use special concepts when you do research in sociology— such as class, status, charisma and so on—is a statement that will seem obvious to most sociologists.1 But things are not that easy; and despite a voluminous literature on the importance of individual concepts in sociology, the number of empirical sociological studies that dispense with concepts is considerable.2 As will be shown later in this article, a rough estimate is that this happens in more than 50%. How this development has come about, and what consequences it has for sociology, are the two main questions that this article will try to address. One way to approach an answer to the historical question of how this state of affairs has come about, is to look at the kind of sociology that was developed at Columbia University in the 1950s and 1960s (‘Columbia Style ­Sociology’— Merton).3 The reason for proceeding in this way is that some of the sociologists who were active at this university, created what was to become the mainstream view of methodology, including the way that concepts should be looked upon and handled. After this part of the analysis, which is centered on the two-three decades after wwii, I will bring the analysis of the role of concepts in sociological research up to date. This will be done by presenting, and commenting on, some quantitative data on the use of concepts in sociological research during a much longer period, 1923–2012. 1 For help and good advice I thank Mabel Berezin, Alicia Eads and Hernan Mondani. Alicia Eads produced the figures and all the work that went into these. 2 There exist many articles, chapters and books—and even whole book series—that are devoted to discussions of single sociological concepts (e.g. Becker, 1960; Luhmann, 1992; ­Merton, 1984; Parkin, 1985; Parsons, 1951, 1967a, 1967b). To this can be added sociological and social science dictionaries, encyclopedias and similar collections, which all contain lists of concepts and/or discussions of these (e.g. Bottomore, 1992; Borgatta and Montgomery, 2002; Marshall, 1994; Boudon and Bourricaud, 1989; Ritzer, 2007; Sills, 1968; Williams, 1983; Smelser and Baltes, 2001). 3 For Merton’s preference for the term Columbia style sociology (to, say, Columbia sociology or the Columbia School of Sociology), see e.g. Clark (1996: 313–328). © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_003

24

Swedberg

To repeat, the decision to focus on the development of sociology at just Columbia University after wwii has to do with the fact that the sociologists who were active at this university played a key role in introducing quantitative methods into u.s. sociology (e.g. Platt, 1996; Sørensen, 1998; Raftery, 2001; see also Abbott and Sparrow, 2007). Through the work of Robert K. Merton, Hans Zetterberg and others, they also argued very strongly that all theories must be verified (e.g. Merton, 1949; Zetterberg, 1954). Without this step, the theory was incomplete. But it should also be pointed out that the discussion of concepts that took place at Columbia University has important roots in pre-wwii developments and was preceded by some important events in this regard. Herbert Blumer, for example, wrote several important statements on concepts, even before he launched the idea of sensitizing concepts in the early 1950s (e.g. Blumer, 1928, 1931, 1940, 1954). Before wwii some sociologists also made an attempt to import the ideas of operationalism into sociology (e.g. Alpert, 1938; Lundeberg, 1939). These ideas had become popular in the social sciences in the 1930s, mainly thanks to the work of physicist P.W. Bridgman. The central message in The logic of physics, in so far as scientific concepts are concerned, was expressed as follows: ‘we mean by any concept nothing more than a set of operations; the concept is synonymous with the corresponding set of operations’ (Bridgeman, 1927: 5; emphasis in the text). At the time, these ideas were hotly debated also among sociologists, even if they are considered dead today.4 Finally, several attempts had been made before wwii by single sociologists as well as by the American Sociological Society to once and for all decide how to define important sociological concepts (e.g. Eubank, 1932). In 1937, for example, the Committee on Conceptual Integration of the American Sociological Society was formed; and it tried as best as it could for several years to come up with what Blumer was later to call definitive concepts (e.g. Hart, 1943). Also this effort failed in a resounding way and has left few traces behind.

Columbia Sociologists on How to Handle Concepts and Variables

Concepts were mainly approached from the viewpoint of methodology, not theory, by the sociologists at Columbia University. And what they meant by 4 In its literal form, few social scientists adhere today to operationalism. ‘Almost all philosophers and social scientists reject this doctrine’, according to Gary Goertz and James Mahoney (2012a: 214–215).

On the Near Disappearance Concepts in Mainstream Sociology

25

methodology was deeply influenced by the publication in 1955 of The language of social research, a reader edited by Paul Lazarsfeld and Morris Rosenberg. There is a paradigmatic quality to this work in that it tried to formulate the basic rules for how sociological research should be conducted and what questions should be asked. Methodology should not deal with theoretical issues and the kind of topics that had traditionally been the concern of the philosophy of science. Its main task was to codify ongoing research methods.5 It was also essential to present these methods in such a way that students could ‘acquire knowledge and modes of thinking which [they] might use later’ ­(Lazarsfeld and Rosenberg, 1955: 13). The language of social research is divided into six sections, of which the first is called ‘The concept and indices’. According to Otis Dudley Duncan, who reviewed the reader for the American Sociological Review, ‘it is perhaps this section which is most likely to make a contribution to thinking about ways to conduct research’ (Duncan, 1956: 508). 5 ‘The Lazarsfeld and Rosenberg (1955) reader on The language of social research established the modern concept of methodology’ (Abbott, 1998: 159–160). According to Theodor Adorno, who worked with Paul Lazarsfeld at one point in the 1940s, methodology had a different meaning in Europe and in the United States. Referring to the empirical project he worked on, which was led by Lazarsfeld, Adorno wrote, I was disturbed … by a basic methodological problem—understanding the word ‘method’ more in its European sense of epistemology than in its American sense, in which methodology virtually signifies practical techniques for research’ (Adorno, 1969: 343). According to Alfred Schutz, the German term Wissenschaftslehre includes in English ‘both logical problems of a scientific theory and methodology in the restricted sense’ (Grathoff, 1978: 101–102). Lazarsfeld’s view of concepts was close to that of Carl Hempel in Fundamentals of concept formation in empirical science (1952a). Hempel’s main point is that science is about establishing general principles, and that concepts have to be adjusted to this task. Lazarsfeld especially liked what Hempel had to say about explication of concepts, and approvingly cites the following passage from Hempel’s work: Explication aims at reducing the limitations, ambiguities, and inconsistencies of ordinary usage of language by propounding a reinterpretation intended to enhance the clarity and precision of their meanings as well as their ability to function in the processes and theories with explanatory and predictive force (Hempel, 1952a: 12; Lazarsfeld, 1993: 236). In a symposium that took place the same year as he published his writing on concept formation, Hempel also gave a paper devoted to the concept of type, including Weber’s version of the ideal type. Hempel’s main point here was that in order for the type concept to be useful, it has to be cast in the form of testable hypotheses (Hempel, 1952b). He does not comment on Weber’s argument that the meaning of the actors has to be taken into account in the ideal type; and that this type of concept can therefore not be used in the natural sciences.

26

Swedberg

The basic message of The language of social research on how to use concepts had been formulated by Paul Lazarsfeld. For the next twenty years he would repeat these ideas, using exactly the same formulations (see Lazarsfeld and Rosenberg, 1955: 15–16; Lazarsfeld, 1958, 1966, 1973: 12–14). From this it can be concluded that Lazarsfeld had a firm set of opinions on how to deal with concepts, and that these remained the same over a period of time during which Columbia style sociology was at its most influential. Lazarsfeld’s approach to the use of concepts is imminently practical and easy to follow. He summarized it as follows: Step 1. Imagery; Step 2. Concept Specification; Step 3. Selection of Indicators; and Step 4. Formulation of Indices. You start the research by having some general idea (‘imagery’) or a concept. You may, for example, have noticed that some children are more alert and curious than others, and therefore more intelligent. Or you may be intrigued by the fact that one organization is run more efficiently than another, and probably has a better management. These are your imageries or concepts. Lazarsfeld did not believe in producing definitions of concepts. Neither was he very interested in the initial phase of concept formation: ‘I purposely use the word “imagery” in a context where other writers talk of a definition. I do not believe that concepts in the behavioral sciences can ever be defined ­precisely by words’ (Lazarsfeld, 1966: 257). In step number two you ‘take this original imagery and divide it into its components’ (Lazarsfeld, 1966: 188). Step number three consists of locating empirical indicators; and here the researcher has to be inventive. This is followed by the final step in which indices are constructed. This is where you close the distance between reality and the concept; and where you ‘put Humpty Dumpty back together again’ (Lazarsfeld, 1966: 189). As already mentioned, Lazarsfeld repeated this account, which he referred to as ‘the flow from concepts to empirical indices’, in a number of writings after The language of research (e.g. Lazarsfeld, 1958: 100, 1966: 187). He later, however, limited the applicability of his flow theory to what he called classificatory concepts. As an example of a non-classificatory concept, he mentioned the concept of role. He did not elaborate on the difference between the two kinds of concepts, but simply referred the reader to what Merton says on the concept of role, for a non-classificatory concept.

On the Near Disappearance Concepts in Mainstream Sociology

27

But Lazarsfeld was still not satisfied with his terminology, and in the 1960s he suggested that the term classificatory concept should be replaced by inferential concept (Lazarsfeld, 1966: 159). The reason for the change in terminology had much to do with Lazarsfeld’s work on latent variable analysis. The social scientist notices a phenomenon, and infers a latent variable-concept from its existence. Lazarsfeld’s view of the role of the concept in sociological analysis signaled a number of important changes. He was clearly less interested in the theoretical aspects of the concept than in its methodological aspects. It is fair to say that Lazarsfeld had an unproblematic view of the nature of concepts and how these should be used in sociological analysis. In the view of James Coleman, who clashed with him over some of these issues, ‘Lazarsfeld had a difficult time understanding sociological theory’ (Coleman, 1990: 89). Lazarsfeld helped to popularize another term besides that of the indicator in the discourse about the concept, as part of his advocacy for multivariate analysis in The language of social research. This was the variable. The term concept could in his view be replaced by that of the variable, even if he himself tended to use the two interchangeably. He also helped to make the term concept less visible and distant in another way, namely by equating it with some vague and early form of observation (‘imagery’). His general lack of interest in the theory of the concept worked in a similar direction. On the first page in the very first article in the section on concepts in The language of social research, there is a passage that gives a sense of what kind of attitude to concepts Lazarsfeld thought would be instructive for students in sociology. The article is called ‘Types of integration and their measurement’ and had been written by Werner Landecker: From the modern empirical point of view the problem of social integration is as challenging as it was from the older, more speculative point of view. However, a change has occurred as to the kind of question asked about integration. Nowadays it seems less pertinent to ask: What is integration? If this question is asked at all, then it is only in preparation for the more fruitful question: How can integration be measured? And, again this latter step is not of interest in itself but merely as a preliminary step, which leads to genuine problems of research such as these: Under what c­ onditions does social integration increase? Under what conditions does it decrease? What are the consequences of a high degree of integration? Sociology is in need of basic research oriented toward this kind of problem. landecker, 1955: 19

28

Swedberg

To get a fuller sense of the way in which concepts were viewed at Columbia University, it is helpful to look at the work of Robert K. Merton. During the late 1940s he was busy working out the basic positions for Social theory and social structure (1st edition 1949). In this enterprise Merton was very aware of Lazarsfeld’s ideas on methodology, which he on the whole was in agreement with. For one thing, Merton agreed with Lazarsfeld that the step from concept to concrete reality must be taken for a theory to be useful, and that the way was to locate indicators and create indices with their help. He wrote: In non-research speculations, it is possible to talk loosely about “morale” or “social cohesion” without any clear conceptions of what is entailed by these terms, but they must be clarified if the researcher is to go about his business of systematically observing instances of high and low morale, of social cohesion or cleavage. If he is not to be blocked at the outset, he must devise indices which are observable, fairly precise and meticulously clear. merton, 1948: 514

Like Lazarsfeld, Merton also used the terms concept and variable interchangeably. At one point in Social theory and social structure, for example, he writes that ‘concepts, then, constitute the definitions of (or propositions) of what is to be observed; they are the variables between which empirical relationships are to be sought’ (Merton, 1949: 87, 1968: 143).6 When variables or concepts are linked together, Merton added in the next sentence, you have a theory: ‘when propositions are logically interrelated, a theory has been instituted’. In the various editions of Social theory and social structure (1949, 1957, 1968), which to some extent can be viewed as the theoretical counterpart to The language of social research, Merton is primarily arguing that it is crucial to link up concepts to empirical research, while he has very little to say about the nature of concepts or their theoretical tasks.7 6 That the tendency to equate the term variable with that of concept was widespread at the time can be illustrated with the name of one of the most famous sociological concepts of the 1950s: the pattern variable (Parsons, 1951). 7 In recalling his time as a graduate student at Columbia University around 1950, Maurice Stein has described that there was a general sense that big things were about to happen in sociology. ‘Tomorrow! Tomorrow! [We felt we were] just on the verge of everything! Lazarsfeld is going to develop the perfect scale and Merton was going to develop the perfect theory about that scale’ (Dandaneau and East, 2011: 135).—In discussing Merton’s tendency to equate the concept with the variable it should be added that at one level he always seems to have loved to coin new terms and create new concepts. But it is also clear that after Columbia style

On the Near Disappearance Concepts in Mainstream Sociology

29

According to Merton, there exists a general and unfortunate tendency for concepts in sociology to lag behind social developments, but this is something that research can remedy (‘conceptual lag’; Merton, 1957: 92, 1968: 146). By being skillful in reading empirical reality, the sociologist can also improve existing concepts. As an example of this, Merton mentions Edwin Sutherland’s insight that the notion of crime also includes law-breaking activities of people in respectable professions (‘white-collar crime’). Merton does not seem to have felt that there was any problem with equating the notion of the concept with that of the variable. He never compared the two nor commented on the fact that while the concept is a very old and multi­faceted tool, which has been discussed and commented on over the centuries, this is not the case with the variable. A variable refers to the assumption that the attribute of some entity varies relatively to something else and can be modelled in various ways with the help of mathematics. A variable can be expressed in quantitative terms; and it is not an actor.8 Furthermore, it differs s­ociology had peaked and the alliance between Merton and Lazarsfeld had fallen apart, Merton felt freer to engage in this type of activity. As an example of an important statement about concepts from Merton’s later period, one can mention his introduction of the notion of proto-concept in ‘Socially expected durations: a case study of concept formation in sociology’: A proto-concept is an early, rudimentary, particularized, and largely unexplicated idea (which is put to occasional use in empirical research and, indeed, often derives from it); a concept is a general idea which, once having been tagged, substantially generalized, and explicated can effectively guide inquiry into seemingly diverse phenomena. Protoconcepts as a phase of theoretical work—a frequent phase in such work—make for early discontinuities in scientific development if only because they obscure underlying conceptual similarities in diverse substantive fields of inquiry by attending to the particularities of each substantive field; concepts make for continuities by directing our attention to similarities among substantively quite unconnected phenomena (Merton, 1984: 267). Another example of a late but general contribution to the theory of concepts from Merton’s side can be found in his discussion of the failure of many sociological concepts to adequately deal with pain and other forms of suffering. ‘Sociological euphemism’ is Merton’s term for sociological concepts that are incapable of capturing ‘social structures which are so organized as to systematically inflict pain, humiliation, suffering, and deep frustration upon particular groups and strata’ (Merton, 1973: 131). In dealing with this type of situation, Merton says, analytically useful concepts such as social stratification, social exchange, reward system, dysfunction, symbolic interaction are altogether too bland in the fairly precise sense of being unperturbing, suave, and soothing in effect (Merton, 1973: 131). 8 I have been unable to find any discussions of the variable from a general theoretical perspective, be it in the literature on the philosophy of science or elsewhere. For a detailed analysis of how the variable is often used in sociological research, especially the work of Abbott is useful. See e.g. Abbott, 2001c.

30

Swedberg

from a concept in that it is not based on a natural mental representation, as cognitive psychologists would say; it is instead the product of human ingenuity. The closest that Merton ever came to a general analysis of the variable can be found in his discussion of what he called methodological empiricism ­(Merton, 1948: 513–514). In much of the sociological research that is being conducted, Merton said, variables are not properly defined. Sociologists are often very interested in seeing how, say, A, B and C are interrelated—, but much less so in what each of these stand for. One consequence of this disregard for properly defining a variable is that it becomes hard to integrate the findings of empirical research into the discipline. Instead of a gradual cumulation, you end up with ‘a buckshot array of dispersed investigations’ (Merton, 1949: 96). The notion that researchers would treat a concept differently, once they had begun to think of it as a variable, does not seem to have occurred to Merton. All that was needed to replace a concept with a variable, in other words, were better definitions of the variables.

How Influential was the View of the Columbia Sociologists?

What was the impact of the ideas that were developed at Columbia University on the concept and the variable on the research practice of sociologists? Some of the people who were active at the time thought that it had been substantial. Twenty years after the publication of The language of social research, ­Arthur Stinchcombe, for example, reviewed the second edition of this work; and he began his account with the following words: ‘The first edition [from 1955] of The language of social research was probably the most important book in the history of methodology in the discipline’ (Stinchcombe, 1974: 126; see also ­Sørensen, 1998: 242; Platt, 1996: 29).9 It is also true that the approach to 9 To equate concepts with variables also seems to be characteristic of the first sustained attempt in u.s. sociology to explore the process of theorizing, the so-called theory construction movement that emerged in the 1960s and 1970s (e.g. Zhao, 1996). In Constructing social theories by Arthur Stinchcombe, concepts are equated with variables in an unproblematic manner. ‘A ‘variable’ in science is a concept which can have various values, and which is defined in such a way that one can tell by means of observations which value it has in a particular occurrence’ (Stinchcombe, 1968: 28–29, 38). The same tendency can also be found in Theoretical Sociology: From Verbal to Mathematical Formulations by Hubert Blalock which refers to ‘concepts or variables’ (e.g. Blalock, 1969: 28). This is similarly true for Blalock’s presidential address at asa in 1978, which was devoted to ‘measurement and conceptualization problems’, and in which the author refers to what he calls ‘theoretical variable’ (Blalock, 1979: 881). Finally, the discussion of indicators that was central to Lazarsfeld’s approach to concepts has continued till today and is still very much alive, both when it comes to how to locate the

On the Near Disappearance Concepts in Mainstream Sociology

31

concepts in The language of social research—to equate these with variables and to look for indicators for variables—is still predominant in sociology (e.g. Abbott, 2001b: 66–67; Goertz and Mahoney, 2012a, 2012b). It would of course be helpful to know in more detail, and especially at the level of research practice, what happened when Lazarsfeld, Merton and others started to advocate that concepts can be equated with variables and that the term ‘concept’ might just as well be dropped. Did it have a profound impact on the work carried out by sociologists and, if so, how? A solid historical study would be the best way to address this question. Such a study, however, does not (yet) exist. In the meantime, a preliminary and tentative kind of answer can be given by proceeding in the following way. If you look at the frequency with which the word ‘variable’, on the one hand, and ‘variable’ plus ‘concept’, on the other, have been used in all sociological articles since the 1950s, you will get a rough sense for how their authors viewed the importance of using variables and concepts in empirical sociological research. In proceeding in this way, you basically tap into the general universe of words and terms that sociologists have drawn on, when they thought about their research and how to present it to their colleagues. The general idea behind this way of proceeding, in other words, is that if you use an analysis that is based on variables, and also think that sociological concepts are important to the analysis, you will be likely to at some point use the word ‘concept’. If you do not think sociological concepts are that important and that you can do without them, you are similarly likely to not use this word.10

10

indicators and how to handle them (e.g. Sullivan and Feldman, 1979; Land and Ferris, 2008). A student who is interested in learning the technique of how to construct indicators can, for example, easily find good instructions for how to proceed (see e.g. de Vaus, 2002: 55). There also exists a number of works on various types of validity which have as their goal to ensure that the chosen indicators capture what they are supposed to capture (e.g. Adcock and Collier, 2001). There exists little discussion, in contrast, of the fact that the move from concept-to-variable-to indicator (1-2-3) is often reduced to a move from variable-to-indicator (2–3). One may ask the question if it is not possible to discuss and use a sociological concept without using the word ‘concept’. The answer to this question is that you can probably do this, even if it is very hard to refer to the key issues of any sociological concept without ­using the word ‘concept’. There is also the fact that articles that discuss concepts without using the word ‘concept’ are probably quite rare and do not affect the general trend, which is what is being discussed here. Another issue that this question raises is the following. It is impossible to write a sociological article without using any sociological concepts (type ‘society’, ‘social structure’, ‘norm’ and so on). The awareness of the author that he/ she uses concepts and that these raise special issues does, however, seem to differ quite a bit—­going all the way from the author not being aware of this fact at all, to the author who is perfectly aware of the advantages (and problems) with using concepts.

32 75

Words in Articles “Variable” “Variable and Concept”

50 25 0

19 00 19 10 19 20 19 30 19 40 19 50 19 60 19 70 19 80 19 90 20 00 20 10

Percent of Total Articles

Swedberg

Year of Publication Percent of articles in the American Sociological Review (1936–2012) containing the words “variable” or “variable” and “concept”

Percent of Total Articles

Figure 2.1

75

Words in Articles AJS “Variable” AJS “Variable and Concept” SF “Variable” SF “Variable and Concept”

50 25

19

00 19 10 19 20 19 30 19 40 19 50 19 60 19 70 19 80 19 90 20 00 20 10

0

Year of Publication Percent of articles in ajs (1895–2012) and Social Forces (1925–2012) containing the words “variable” or “variable” and “concept”

Percent of Total Articles

Figure 2.2

50 Words in Articles “Variable” “Variable and Concept”

25

Figure 2.3

20 19 30 19 40 19 50 19 60 19 70 19 80 19 90 20 00 20 10

10

19

19

19

00

0 Year of Publication Percent of articles in all sociology journals (1936–2012) containing the words “variable” or “variable” and “concept”

Figures 2.1–2.3 show the frequencies with which the words ‘variable’ and ­‘variable’ plus ‘concept’ have been used in different sociological journals since a few decades back. All of these largely largely tell the same story, but in what follows I have chosen mainly to comment on The American Sociological Review (1936–2012) since this journal is often seen as the organ of mainstream sociology.

On the Near Disappearance Concepts in Mainstream Sociology

33

Keeping in mind that this way of looking at things is merely suggestive, the results indicate the following broad conclusions. First of all, it would appear that the tendency to drop the notion of the concept from variable analysis goes much further back than to the post-war period when the main works of Lazarsfeld and Merton were produced. Given this fact, the impact of Lazarsfeld and Merton appears to have been less important than what you may have thought. It would also appear that over time the proportion of articles in asr that use the word ‘variable’ in relation to those that use both ‘variable’ and ‘concept’, has remained more or less about the same. During this period, however, the absolute number of articles that use the word ‘variable’ in relation to all articles in asr has doubled several times. The type of analysis that draws on variables has in other words become much more prevalent after World War ii— and with it also a special attitude to the concept. If one were to summarize the impact of Columbia style sociology in the 1950s, based on this exploratory analysis, it would be as follows. While the tendency to equate the concept with the variable existed well before World War ii, Lazarsfeld et al probably helped to provide an intellectual justification for it, popularize it and turn it into mainstream sociology. It is not advisable to cite exact figures, based on the material from asr. Nonetheless, looking at the period 1980–2010, it would seem that at least half of the empirical analyses (perhaps even more), dispense with referring to concepts. Again, this argument becomes even stronger if we assume that a solid sociological article would contain at least a line or two in which concepts are discussed, rather than just mentioning (or not) the word ‘concept’. The result of at least half (and perhaps even more) is also true for ajs and all sociological journals (in jstor). In the critique of variable sociology that emerged in the 1990s, it was argued that the notion of sociological theory has been impoverished and often replaced with statistical models in which the analysis of variables is central (e.g. Abbott, 1992, 1997; Hedström and Swedberg, 1998; Manzo, 2007). Many of these models operate in principle by adding up the number of variables used in the explanation of some phenomenon (‘additive models’—Sørensen, 1998: 249; see also e.g. Ragin, 2008: 112–114). Variables do the acting rather than individuals or groups (e.g. Abbott, 1992). To all of these critiques, it can now also be added that there exists a tendency in variable analysis to ignore concepts and just refer to variables. As the analysis of variables grew in popularity, so did this attitude among sociologists. The exploratory analysis of the role of concepts and variables in asr during 1936–2012 that has just been refereed to, may be able to capture some aspects

34

Swedberg

of the general development of sociologists’ use of concepts and variables, but it would clearly have been much better if it had been based on a close reading of, say, all the individual articles in asr or on a good-sized sample of these. By proceeding in this way a tentative and suggestive finding would have been replaced by a solid one. There does however exist a solid content analysis of asr; and as it so happens, some of its findings are also relevant for this paper. This is ‘The use of conceptual categories of race in American sociology, 1937–99’ by John Levi Martin and King-To Yeung (2003). What the authors of this article were interested in was not how concepts in general had developed over time in mainstream sociology, but how one specific concept had done so. This was the concept of race; and Martin and Yeung carried out their analysis through a solid content analysis of a sample of articles that had appeared in asr during 1937–1999. During this period, they note, the articles that were empirical in nature had doubled and come to totally dominate the journal (from 42% in the late 1930s, to 84% in the late 1990s; Martin and Yeung, 2003: 527). The number of studies that took race into account also rose very sharply during 1937–1999; and the method that was used in these studies was typically regression analysis or a regression-type model. These are important findings about the concept of race in sociology. But Martin and Yeung also found something else that speaks to the issue of this article. This is that while a growing number of analyses did include race in the analysis, the view of race changed. It became, to cite the authors’ summary formulation, ‘broad but shallow’ (Martin and Yeung, 2003: 538). One reason for this change, Martin and Yeung argue, was technical in nature and mainly due to the fact that race was often just added as a control variable. Martin and Yeung do not differentiate between concept and variable in their analysis, but mix the two in their discussion of what they call ‘the conceptual category of race’. In the terminology of this article, one can however say that in becoming ‘broad but shallow’, their analysis of the concept of race illustrates what may happen when the line between concept and variable becomes fuzzy or dissolved.

Concluding Remarks

The insights about concepts that can be found in the article by Martin and Yeung are important but have not led to much debate. On the whole, the interest of sociologists in the theoretical aspects of concepts is very low today and has been low ever since the days of Lazarsfeld and Merton. Some ­exceptions

On the Near Disappearance Concepts in Mainstream Sociology

35

do exist, even if they are not many and have not received much attention (e.g. Bulmer, 1979, 2001; Bulmer and Burgess, 1986; Prandy, Stewart and Blackburn, 1974; Abbott, 2001a). It has recently been suggested that there exists more of an interest in discussing concepts among qualitative sociologists than among those who use quantitative methods (e.g. Goertz and Mahoney, 2012a, 2012b). According to Gary Goertz and James Mahoney, who have written a book on the different cultures that have emerged around qualitative and quantitative studies in social science, quantitative scholars and qualitative scholars differ in the following way when it comes to concepts: For qualitative scholars, the relationship between a concept and data is one of semantics, i.e. meaning. These scholars explore how data can be used to express the meaning of a concept. For quantitative scholars, by contrast, the relationship between variable and indicator concerns the measurement of the variable. These scholars focus on how to use indicators to best measure a latent construct. goertz and mahoney, 2012b: 140

Exactly how qualitative sociologists have looked at concepts since wwii remains an interesting task to explore. My own sense is that they basically view the concept as part of methodology, as opposed to as part of theory, as advocated in this article. In this sense, qualitative and quantitative sociology are roughly in agreement. The last question to discuss, which is perhaps the most important, has to do with the consequences of not using concepts in a sociological analysis. With some exaggeration I have called this article ‘the near disappearance of concepts in mainstream sociology’, but hopefully the reader will agree that the number of empirical articles that do not even mention concepts is disturbingly large. A first point to make in this context is that it is clear that you can ignore sociological concepts in empirical research and still produce high quality work on many issues. The analysis may seem more realistic this way, being closer to reality as well as far away from artificial theory. Add to this that it makes things easier not to have to deal with the tricky and bothersome question of how to operationalize concepts such as, say, status, class and charisma. But there is also a price to pay if you choose to proceed without explicitly drawing on sociological concepts. You not only cut yourself off from a wealth of good ideas, you also run the risk of having your research being ignored since it cannot easily be integrated into the sociological tradition if it does not refer to concepts and theory, but exclusively to some topic. Whatever sense of realism

36

Swedberg

and relief that is temporarily produced by dispensing with concepts, should to my mind be weighed against the fact that research without concepts may have great difficulty in surviving and influencing future research. Science without concepts risks ending up as the famous portrait in The picture of Dorian Gray: first looking young and vital for quite some time, but ultimately collapsing as a result of a misspent life. References Abbott, A. (1992). “What do cases do? Some notes on activity in sociological analysis”, Eds. C. Ragin and H. Becker. What is a case?. Cambridge: Cambridge University Press. Abbott, A. (1997). “Of time and space: the contemporary relevance of the Chicago school”, Social Forces. 75(4): 1149–1182. Abbott, A. (1998). “The causal devolution”, Sociological Methods and Research. 27(2): 148–181. Abbott, A. (2001a). “On the concept of turning point”, Time Matters: on theory and method. Chicago: University of Chicago Press. Abbott, A. (2001b). “Seven types of ambiguity”, Time Matters: on theory and method. Chicago: University of Chicago Press. Abbott, A. (2001c). “Transcending general linear reality”, Time matters: on theory and method. Chicago: University of Chicago Press. Abbott, A. and Sparrow, J. (2007). “Hot war, cold war: the structures of sociological action, 1940–1955”, Ed. C. Calhoun. Sociology in America: a history. Chicago: University of Chicago Press. Adcock, R. and Collier, D. (2001). “Measurement validity: a shared standard for qualitative and quantitative research”, American Political Science Review. 95(3): 529–546. Adorno, T. (1969). “Scientific experiences of a European scholar in America”, Eds. D. Fleming and B. Bailyn. The intellectual migration: Europe and America, 1930–1960. Cambridge: The Belknap Press. Alpert, H. (1938). “Operational definitions in sociology”, American Journal of Sociology. 3(6): 855–861. Becker, H.S. (1960). “Notes on the concept of commitment”, American Journal of S­ ociology. 66(July): 32–40. Blalock, H. (1969). Theoretical sociology: from verbal to mathematical formulations. ­Englewood Cliffs, nj.: Prentice-Hall. Blalock, H. (1979). “The presidential address: measurement and conceptualization problems: the major obstacle to integrating theory and research”, American Sociological Review. 44(6): 881–894. Blumer, H. (1928). Method in social psychology. Ph.D. thesis. University of Chicago, ­Department of Sociology.

On the Near Disappearance Concepts in Mainstream Sociology

37

Blumer, H. (1931). “Science without concepts”, American Journal of Sociology. 36(4): 515–533. Blumer, H. (1940). “The problem of the concept in social psychology”, American Journal of Sociology. 45(5): 707–719. Blumer, H. (1954). “What is wrong with social theory?” American Sociological Review. 19(1): 3–10. Borgatta, E. and Montgomery, R. (eds.). (2002). Encyclopaedia of sociology. 2nd edition. New York: Macmillan Reference usa. Bottomore, T. (1992). A dictionary of Marxist thought. 2nd edition. London: Blackwell. Boudon, R. and Bourricaud, F. (1989). A critical dictionary of sociology. London: Routledge. Bridgman, P.W. (1927). The logic of modern physics. New York: The Macmillan Company. Bulmer, M. (1979). “Concepts in the analysis of qualitative data”, Sociological Review 27(4): 651–677. Bulmer, M. and Burgess, R. (1986). “Do concepts, variables and indicators interrelate?” Ed. R. Burgess. Key variables in social investigation. London: Routledge & Kegan Paul. Bulmer, M. (2001). “Social measurement: what stands in its way?”, Social Research. 68(2): 455–480. Clark, J. (ed.). (1996). James Coleman (Falmer Sociology Series). London: Routledge. Coleman, J. (1990). “Columbia in the 1950s”, Ed. B. Berger. Authors of their own lives. Berkeley: University of California Press. Dandaneau, S. and East, E. (2011). “Listening to sociological elders: an interview with Maurice R. Stein”, American Sociologist. 42(1): 129–144. de Vaus, D. (2002). Surveys in social research. 5th edition. London: Routledge. Duncan, O.D. (1956). “The language of social research: a reader in the methodology of social research. (Book review)”, American Sociological Review. 21(4): 508–509. Eubank, E. (1932). The concepts of sociology: a treatise presenting a suggested organization of sociological theory in terms of its major concepts. Boston: Heath. Goertz, G. and Mahoney, J. (2012a). “Concepts and measurement: ontology and epistemology”, Social Science Information. 51(2): 205–216. Goertz, G. and Mahoney, J. (2012b). A tale of two cultures: qualitative and quantitative research in the social sciences. Princeton: Princeton University Press. Grathoff, R. (1978). The theory of social action: the correspondence of Alfred Schutz and Talcott Parsons. Blomington: Indiana University Press. Hart, H. (1943). “Research notes: some methods for improving sociological definitions”, American Sociological Review. 8: 333–342. Hedström, P. and Swedberg, R. (eds.). (1998). Social mechanisms: an analytical approach to social theory. Cambridge: Cambridge University Press. Hempel, C. (1952a). Fundamentals of concept formation in empirical science. International Encyclopedia of Unified Science. Chicago: University of Chicago Press.

38

Swedberg

Hempel, C. (1952b). “Symposium: problems of concept and theory formation in the social sciences”, Science, Language, and Human Rights. American philosophical association, eastern division edition. Philadelphia: University of Pennsylvania Press. Land, K. and Ferris, A. (2008). “The sociology of social indicators”, Eds. C. Bryant and D. Peck. 21st century Sociology. London: Sage. Landecker, W. (1955). “Types of integration and their measurement”, Eds. P. Lazarsfeld and M. Rosenberg. The language of social research: a reader in the methodology of social research. Glencoe, il.: The Free Press. Lazarsfeld, P. (1958). “Evidence and inference in social research”, Daedalus. 87(4): 99–130. Lazarsfeld, P. (1966). “Concept formation and measurement in the behavioral sciences: some historical observations [followed by ‘a discussion of the issues’]”, Ed. G. DiRenzo. Concepts, theory, and explanation in the behavioral sciences. New York: Random House. Lazarsfeld, P. (1973). Main trends in sociology. London: George Allen & Unwin. Lazarsfeld, P. (1993). “The relevance of methodology”, On social research and its language. Chicago: University of Chicago Press. Lazarsfeld, P. and Rosenberg, M. (eds.). (1955). The language of social research: a reader in the methodology of social research. Glencoe, il.: The Free Press. Luhmann, N. (1992). “The concept of society”, Thesis Eleven. 31(1): 676–680. Lundberg, G. (1939). Foundations of sociology. New York: The Macmillan Company. Manzo, G. (2007). “Variables, mechanisms, and simulations: can the three methods be synthesized? A critical analysis of the literature”, Revue française de sociologie. 48(Supplement): 35–71. Marshall, G. (1994). The concise Oxford dictionary of sociology. New York: Oxford ­University Press. Martin, J.L. and Yeung, K.-To. (2003). “The use of the conceptual category of race in American sociology, 1937–1999”, Sociological Forum. 18(4): 521–543. Merton, R.K. (1948). “The bearing of empirical research upon the development of social theory”, American Sociological Review. 13(5): 505–515. Merton, R.K. (1949). Social theory and social structure: toward the codification of theory and research. Glencoe, il.: The Free Press. Merton, R.K. (1957). Social theory and social structure. 2nd revised and enlarged edition. Glencoe, il.: The Free Press. Merton, R.K. (1968). Social theory and social structure. 3rd enlarged edition. Glencoe, il.: The Free Press. Merton, R.K. (1973). “The perspectives of insiders and outsiders”, The sociology of ­science: theoretical and empirical investigations. Chicago: University of Chicago Press.

On the Near Disappearance Concepts in Mainstream Sociology

39

Merton, R.K. (1984). “Socially expected durations: a case study of concept formation in sociology”, Eds. W.W. Powell and R. Robbins. Conflict and consensus: in honor of Lewis A. Coser. New York: The Free Press. Parkin, F. (1985). Book series: Concepts in the social sciences. Maidenhead: Open ­University Press. Parsons, T. (1951). The social system. Glencoe: The Free Press. Parsons, T. (1967a). “On the concept of influence”, Sociological theory and modern society. New York: The Free Press. Parsons, T. (1967b). “On the concept of political power”, Sociological theory and modern society. New York: The Free Press. Platt, J. (1996). A history of sociological research methods in America 1920–1960. Cambridge: Cambridge University Press. Prandy, K., Stewart, A. and Blackburn, R. (1974). “Concepts and measures: the example of unionateness”, Sociology. 8(3): 427–446. Raftery, A. (2001). “Statistics in sociology, 1950–2000: a selective review”, Sociological Methodology. 31(1): 1–45. Ragin, C. (2008). Redesigning social inquiry: fuzzy sets and beyond. Chicago: University of Chicago Press. Ritzer, G. (ed.). (2007). The Blackwell encyclopedia of sociology. 11 volumes. London: Wiley-Blackwell. Sills, D. (ed.). (1968). International encyclopedia of the social sciences. 17 volumes. New York: Macmillan. Smelser, N. and Baltes, P. (eds.). (2001). International encyclopedia of the social and behavioral sciences. 26 volumes. New York: Elsevier. Sørensen, A. (1998). “Theoretical mechanisms and the empirical study of social processes”, Eds. P. Hedström and R. Swedberg. Social mechanisms: an analytical approach to social theory. Cambridge: Cambridge University Press. Stinchcombe, A. (1968). Constructing social theories. New York: Harcourt, Brace & World. Stinchcombe, A. (1974). “Review of Paul Lazarsfeld et al. (eds.). Continuities in the language of social research”, Social Forces. 53(1): 128. Sullivan, J.L. and Feldman, S. (1979). Multiple Indicators. Newbury, London and New Delhi: Sage Publications Inc. Williams, R. (1983). Keywords: a vocabulary of culture and society. Revised edition. New York: Oxford University Press. Zetterberg, H. (1954). On theory and verification in sociology. New York: The Tressler Press. Zhao, S. (1996). “The beginning of the end or the end of the beginning? The theory construction movement revisited”, Sociological Forum. 11(2): 305–318.

part 1 Methodological Programs and Applications



chapter 3

What Do We Do with Norms—Conform, Break, Understand or Explain? Peter Sohlberg Like a grammar, a system of norms specifies what is acceptable and what is not in a social group. And, analogously to a grammar, a system of norms is not the product of human design and planning. bicchieri, 2006: ix

∵ Introduction The Swedish comedian Hans Alfredson once asked counterfactually how we would regard the activities on a dance floor without music. Similarly, one could ask how we would regard the organization of human activities without norms. As constructed in social science, norms concern social forces of great complexity hidden behind a simple label. The potential flexibility of social organization is enormous. Seen as a counterfactual thought experiment, any social organization could hypothetically be constructed in an immense variety of ways (cf. Ahrne, 2016). In reality, there are firm limitations on social organization, and keeping time and space constant, we do not find this potential variety. A basic way to understand the relative orderliness of social organization and relative predictability of social behaviour in social science is to use the concept of social norms. This text develops the theme that the word ‘norm’ stands for a complex theoretical construct—a concept and not a ‘thing’ that is easily discernible. Of course, few would explicitly adhere to the view that norms are things, but if norms are treated as easily identifiable causal powers determining human behaviour and social organization, the implication is that they are examples of ‘misplaced concreteness’ (Whitehead, 1925) or, in other words, conceptual reifications. Whether we are navigating in the social landscape as participants or observers, we often assume that norms exist as brute facts, i.e., things, but the situation is actually more multifaceted than that. © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_004

44

Sohlberg

The Norm as an Interactive Concept The norm concept is fuzzy; on the one hand, it is part of everyday vocabulary, and on the other hand, it performs rather complex functions in sociological reasoning. The norm as a concept has this dual form that depends on it being an interactive concept according to Ian Hacking’s definition (Hacking, 1999). According to this perspective, no absolute cleavage in social science exists between the everyday understanding of a concept and its use in the social scientific vocabulary. This relation between tools of understanding applied to ourselves is phrased in terms of reflexivity by Anthony Giddens. Giddens identifies the reflexive activity of actors as related to the duality of structure: Human actors are not only able to monitor their activities and those of others in the regularity of day-to-day conduct; they are also able to “monitor that monitoring” in discursive consciousness. “Interpretative schemes” are the modes of typification incorporated within actors’ stocks of knowledge, applied reflexively in the sustaining of communication. giddens, 1986: 29

It is reasonable to think that the concept of norm belongs to these interpretative schemes. An interesting question to ask from a phenomenological perspective is whether we use the interpretative schemes symmetrically, i.e., whether we experience our own actions as a matter of norm-following to the same extent that we understand our fellow beings as following norms (cf. Skjervheim, 1976). The ‘objectifying’ and distanced position is not necessarily taken by a social scientist; it could just as easily be held by our partners in everyday interaction. In deconstructing the reified use of the norm concept, I will follow two paths. One line of reasoning starts with the analysis of the idea of social action and the place of norms from this perspective. In this analysis, it will become evident that from a sociological perspective, the idea of social action is more complex than instances of simple norm-following. Another line of deconstructive reasoning finds its starting point not in social actions but in a perspective of the system. Seen from the system’s point of view, norms are specific normative constituents that are necessary for the functioning of the system. However, they are not the only prerequisites for functioning. In the world of theoretical systems, we also have an ontological conflict between approaches emphasizing the normative structure as fundamental and materialist system approaches regarding normative aspects as ‘epiphenomena’. However, even if we regard normative structure as essential for social life, its ontological status remains to be determined. Few claims that absolute

What Do We Do with Norms

45

‘­normative truths’ exist in a metaphysical sense. The philosophical position that distinguishes normative statements from statements of fact is commonplace. The dominant argument is that possible normative truths cannot be observed in the same way as truths of fact. There is a rich philosophical d­ iscussion concerning the ontological status of normative statements and their relations to statements of facts (for a comprehensive discussion based on a specific historical example, see Eliaeson, Mindus and Turner, 2014). A reasonable prediction is that the question concerning the ontological status of values will never be resolved. However, this metaphysical question is of less importance for social science, as norms are enforced in social settings regardless of their basic ontological status. Independent of the ultimate nature of the statement ‘thou shall not kill’, it is a position that is forcefully sanctioned by law in civilized social settings. It is a social fact that norm systems are institutionalized to a high degree in every society or social group, covering a variety of areas and regulating how the correct, good or acceptable life should be lived. Manners vary, but they are never random in a specific social context. We also must address the question of language, as norms are supposed to be mediated by injunctions phrased in ordinary language. As referenced in the initial quotation from Cristina Bicchieri comparing systems of norms to grammar, there is no language without grammar. Similarly, there is no social system with interacting human beings without a norm system. When studying grammar, it is common to distinguish between descriptive and prescriptive rules, i.e., empirical observations of language use and prescribed rules of language use. An analogous distinction applies to norms, as they can be seen merely as statistical standards or as morally prescribed standards. This distinction between statistical and normative normality is often blurred in the social context.

Tools of Construction

Before discussing the theoretical and social construction of norms, the normative rules of the game at a meta-level, i.e., the norms of any conceptual argumentation in natural language, would be helpful to discuss. Much has been written about the ‘linguistic turn’ in the social sciences (see e.g., Delanty and Strydom, 2003). Texts—and in the 20th century, speech acts—have always been the material for analysis in philosophy. What is relatively new is that texts in different forms (narratives, discourses, etc.) have also become important objects of study in the social sciences. What can be accomplished socially with words has become an important theme from diverse social constructionist perspectives (cf. Guneriussen, 2016).

46

Sohlberg

Interestingly, the analogy between grammar and norm systems does not function perfectly when considering the argumentative rules of the game. From a strictly linguistic point of view, there seems to be no moral aspect to not following the rules of grammar. This could be considered dysfunctional and opening up for non- or misunderstanding, but a breach of the rules of grammar is seldom considered an immoral act. However, an interesting connection exists between the use of language and normativity, as expressed by Jürgen Habermas in his theory of communicative action. This is more of a normative theory formulating an ideal situation than describing the actual state of affairs. According to Habermas, the normative aspect of acceptable language use is clearly visible in his formulation of validity claims. From the perspective of communicative action, the ideal form of communication has the normative requisites of being sincere, right and true (Habermas, 1989). Thus, from this perspective, the language we use as a tool for argumentation has in itself a basic normative nature.

Genealogy—The Birth of Abstract Morality

Normative reasoning has always organized human life and has been a basic human activity. Never ending questions have addressed what fundamental human virtues should be sought (actor perspective) and what the best way to ­organize society would be (system perspective). In its various forms, religion has, of course, been a primary source of norms during most of known history. In addition to religious sources, a more secular discourse on normative matters also developed early, primarily in ancient Greece. For example, human virtues are discussed in the Socratic dialogues. Though these virtues are discussed analytically from an essential perspective in the Platonic tradition, they do not seem to be abstracted to a common concept equivalent to the abstract notion of the norm. The norm is a theoretical umbrella concept covering a range of virtues transformed to social imperatives. From a social science perspective, an extremely important theoretical shift occurs when the concept of norm becomes abstracted to this theoretical umbrella concept, which includes a variety of normative and socially enforced actions. This conceptualization in terms of the abstract norm-concept opens up for the system-theoretical embeddedness of the norm concept, which has been its primary theoretical position in classic sociology. The abstract quality of the norm concept also implies a distancing from actual guidelines for everyday conduct. In an implicit way, the application of normative thinking to systems was somehow developed during ancient times. Plato and Aristotle had elaborate

What Do We Do with Norms

47

discussions concerning the best way to organize the state. However, the idea of the state was not abstracted to a notion such as the social system in modern discourse. The structure of ancient discourse and all later examinations of utopias note that the best way to organize the state is to first identify the normatively desirable states of affairs and then determine the best way to organize society to fulfil these goals. Designing organization is always a matter of structuring functions. Here, we can see the general formula that would be labelled much later as structural-functionalism. In this formula, the goals are the functions and the specific method of organization is the way to realize these functions. Aristotle understands the origin of the state in normative terms: Every state is a community of some kind, and every community is ­established with a view to some good; for everyone always acts in order to obtain that which they think good. But, if all communities aim at some good, the state or political community, which is the highest of all, and which embraces all the rest, aims at good in a greater degree than any other, and at the highest good. aristotle, 2013: 11

Even if Aristotle’s discourse on how best to organize the state is more sophisticated than revealed in this quotation, the notion of an abstract system to be discussed in its own right is not systematically developed. However, this theoretical move is made in classic sociology, and aside from Herbert Spencer (see, e.g., Spencer, 1981), the best early example is probably Emile Durkheim. In the rules of method, Durkheim constructs the object of sociology, i.e., social facts, by stipulating the analogy between social facts and things (Durkheim, 1982). From a purely verbal perspective, this seems to be a literal reification. ­Durkheim’s idea is however rather the opposite, i.e., treating social facts as things or as a constructive element analogous to Bourdieu’s idea of constructing the object of social science (Bourdieu, Chamboredon and Passeron, 1991). According to Durkheim, one defining characteristic of social facts is their constraining nature. This is in line with my initial statement that social life could potentially be organized in an immense variety of ways that are never realized. In Durkheim’s world, constraint is not primarily a physical constraint but rather a moral one. The implication is that the normative order is central to understanding the functioning of society (cf. Parsons, 1970: 19). Durkheim is essential for the development of normative reasoning by identifying normativity as a central aspect of society and the social system. However, normativity is not a static factor in Durkheim’s understanding. Society is characterized by

48

Sohlberg

different kinds of solidarity and various strengths of the collective consciousness, which enforces morality. When the normative force of the collective consciousness gets weaker, its necessary cohesive function is replaced by functional interdependence following the division of labour. This replacement of mechanical solidarity with its functional equivalent, organic solidarity, has a teleological flavour. When the functional need (cohesion) arises, it is fulfilled by a new structural arrangement—the interdependence of parts. With the theories of Durkheim, Weber and Marx, the stage was set for an ontological battle concerning the ultimate nature of social reality. This battle is highly relevant to the position of normative reasoning in contemporary social science. The discussion concerning normative elements in general sociology is often situated within the ontological question about the first cause or ‘prime mover’ of social reality—material conditions or ideas. This is not the place for exegetic exercises, but the Marxist perspective with its materialist orientation generally regards norms as ‘epiphenomena’ and not appropriate as a basic category for understanding social life. For example: The social structure and the State are continually evolving out of the life process of definite individuals, but of individuals, not as they may appear in their own or other people’s imagination, but as they really are; i.e., as they operate, produce materially, and hence as they work under definite material limits, presuppositions and conditions independent of their will. marx and engels, 1978: 154

In a basic sense, the Marxist tradition is the most fundamental norm-critical theoretical tradition. Marxism is critical not only of specific sets of norms but also of the general idea that constraints of social organization are primarily normative. As mentioned, for Durkheim, the normative cohesion of the social system is the basic functional requisite that allows societies to prevail. However, Durkheim also introduces other aspects aside from normativity as central to understanding the continued existence of social structure. For Durkheim, solidarity is not a mental state but properties of the social structure in which organic solidarity, or functional dependence, is a crucial factor. Max Weber is often interpreted as giving an idealist alternative to the materialist understanding of the rise of capitalism. Positioning him as a clear-cut idealist has been disputed on solid grounds (see, e.g., Swedberg and Agevall, 2005). However, Weber’s strong focus on social actions and their rationales does ignore material conditions as an essential aspect of social science. I­ llustrating

What Do We Do with Norms

49

the complexity of the norm concept, value-rational actions in Weber’s understanding are obviously based on norms. However, it could also be the case that traditional actions are ultimately normatively based, though they are routinized. On a meta level, it could be a normative imperative to act according to means-end rationality. The conclusion to be drawn from this intersection of norms and action rationales is that normative aspects are not easy to isolate as causal factors in social life. Leaving the classic theorists, though not the conflicting perspectives, it is clear that the most important figure in normative reasoning in sociology during the 20th century is Talcott Parsons. The development of Parsons’ theoretical position is often described as a movement from an actor perspective to a system perspective. For example, Habermas describes Parsons’ development as a path ‘From a normativistic theory of action to a systems theory of society’ (Habermas, 1989: 204). Parsons’ starting point is not action in the commonsensical meaning of an observable action, but it is instead a generalized abstraction. With atomistic phrasing, he refers to ‘the unit act’. The idea of the unit act reveals that at this early stage, he already included other factors than norms as important for ­understanding action. The unit act has several interrelated components: an actor, a goal, and a situation as context for the action, i.e., a system in its own right. The situation is partly controllable by the actor and partly u ­ ncontrollable. ­Furthermore, we have a mode of orientation by which the actor ‘binds together’ the goal with the present situation (Parsons, 1937: 733). Norms certainly influence the goals of action and the means by which those goals are achieved, but the unit act also includes a situation with conditions and the ‘supply’ of possible means by which to choose, which is not necessarily a n ­ ormative matter. The rather dominant normative and primarily idealist tradition in classic sociology, represented by scholars such as Talcott Parsons, met its antithesis in a tradition based on an opposing ontological standpoint, i.e., the revival of the materialist ontology in the Marxist tradition. With the dominance of Parsons’ theory and the following Marxist revival, we had a rather clear-cut situation concerning the theoretical embeddedness of norms. In the functionalist tradition, the systemic importance of norms for the continuation of the social system was obvious, as was the importance of the socialization process at the individual level for the normative integration of human beings. From a Marxist perspective, the viewpoint was ‘reversed’, though not as clearly and explicitly. From this perspective, norms, thought structures and ideologies could not be separated from the basic material conditions of society and the relations of production. The critique from the materialist point of view is directed against an understanding of social systems and society on normative grounds.

50

Sohlberg

In addition to the Marxist criticism, the general system approach of Parsons was also criticized from a variety of approaches, seemingly with less universal knowledge claims. However, structurally simple theories, such as rational choice-theories, do not necessarily imply more humble knowledge claims than more complex theories. In contrast to the ontologically based criticism of Marxism, these small-scale approaches criticized the scale of Parsonian system theory rather than its normative orientation. The development of social science in the post-Parsonian and (possibly) post-Marxist era could be characterized as hyper-differentiation (cf. Sohlberg and Leiulfsrud, 2016: 4 f). A fuzzy embeddedness of normative reasoning follows this hyper-differentiation. We find small scale social-psychological theories focusing on everyday interaction in which normative aspects often have no external foundation but are seen as negotiated in concrete interactions. We also have theories focusing on general discourses in which normative elements seem to be free-floating without any clear connection to intention or agency, whether individual or collective (e.g., Foucault, 2002). Furthermore, we have a wide variety of feminist theories in which a major theme is the questioning of a naïve assumption of normative neutrality in what is regarded as male-stream research. In another camp, we have theories of an individualist orientation in which social behaviour is emphasized and norms are programmatically seen as important but not systematized in their own right (cf. Elster, 2015). This analytical discourse is methodologically characterized by isolated examples in which norms are stipulated as important in a post hoc way, but we never gain a systematic and comprehensive view of the overall landscape of norms from this perspective—a systemized view of norms that was regarded by functionalists as the mission of functional requisite analysis. Seen from the perspective of norm theory, the outcome of the hyper-differentiation in sociology since the 1970s is that we lack a systematized social science approach to norms. There are certainly ambitious overviews of the norm concept, often from a more philosophical point of view (see, e.g., Bicchieri, 2006; Hecther and Opp, 2005), but there are fewer comprehensive elaborations in which norms are systematized, taking in account as well levels of interaction as long-term systemic aspects. Another aspect of the multidimensionality of the norm concept concerns its relative importance as a social factor. This may be a remnant of the ‘ontological battle’ between the classics in which Durkheim emphasized society as a moral, normative phenomenon, whereas Marxists regarded norms as e­ piphenomena; it is seldom the case that the contemporary theory of norms elaborates on the relative importance of norms in a comprehensive model. I  have mentioned

What Do We Do with Norms

51

such a comprehensive model above when describing Parsons’ unit act in which factors other than the normative are of importance. The lack of effort spent in modern sociology to address the ontological cleavage between materialism and idealism is strange. One good and rather solitary exception is David Lockwood’s Solidarity and schism (Lockwood, 1992). Lockwood demonstrates the need for serious sociology to bring together normative and materialistic aspects. Interestingly, he does this from a structural point of view, showing that the understanding of social structure necessitates an understanding of its normativity as well as its materiality. Thus, Lockwood avoids the fruitless question of the prime mover. Thus far, I have assumed that we have a reasonably shared understanding of what a norm is, but now we will see how norms are theoretically c­ onstructed in the social science tradition. The methodological starting point for this ­discussion is that a true essential definition of norm does not exist, but we must consider the actual, lexical use of the concept in practised language. In an evaluation of lexical definitions, it is crucial to focus on the theoretical and cognitive functions alongside the concept, i.e., what do we do with norms?

The Theoretical Construction of Normativity

This chapter began with the statement that norms not are things, i.e., they are not observable entities. What we can observe are concrete instances of more general norms. In other words, the identification of a norm in concrete social situations is always a matter of interpretation based on the theoretical notion of norm. There is no instance where a concrete behaviour constitutes a norm in its totality other than the statistical sense mentioned below. One reason for confusion when discussing norms is that the concept also covers key diverse aspects with different consequences for social life. Göran Therborn describes the various understandings of norms: First of all, a norm tells us what something is. What is a meter? … What is a vegetable … Second, the norm can tell us about the distributive structure of a population, what is “normal” in it. Third, norms tell us what we ought to do. therborn, 2002: 863

Therborn’s description succeeds in presenting a condensed account of the ­basic uses of the norm concept. The first use relates to a standard or criterion to be used in a definition of something. Seen from the perspective of

52

Sohlberg

d­ efinition, this use of the norm concept replaces essential definitions. From an essentialist perspective, we do not have to stipulate requisites to be fulfilled for something qualifying as x; defining characteristics are given by Nature, by God, or are a matter of rational evidence. From a lexical norm perspective, we explicitly stipulate the qualifying requisites; in this interpretation, norms are explicit conventions, i.e., stipulated nominal definitions. This type of norm is ambiguous from a social science perspective. Seen as isolated, this type of norm is merely dependent on our use of language and the meaning of terms. However, in other settings, when norms such as these are legally or morally enforced, they are obviously relevant in a broader social science perspective. For example, this relevance could apply to legal norms in social work, specifying what is an (un-) acceptable childhood. These types of norms could also be a matter of fashion or the good taste in Bourdieu’s meaning. If aspects of fashion or good taste are heavily socially enforced in a subgroup as a matter of distinction, the norm concept becomes relevant even if it relates to a seemingly superficial convention. Bourdieu’s idea of ‘good taste’ and style as a tool for elite groups to distinguish themselves (e.g., Bourdieu, 1984) is a good example of the ambiguity within the norm concept. Within elite groups, such as the old bourgeoisie, there could be highly informal norms concerning acceptable consumption and style, marking a difference from the nouveau riche. These informal norms are vague and seldom explicit but could still be strict and enforced in some circles. The second use of the norm concept, mentioned by Therborn, concerns the statistical distribution of properties and the mean value, which defines what is normal. These types of norms are not normative in the ordinary sense, i.e., they just describe what is common. It is this criterion Durkheim uses when he tries to find a non-ideological criterion to distinguish between what is normal and what is pathological in society (Durkheim, 1982). There will always be a statistical normality in this sense. This kind of normality does not in itself say anything about enforced normativity and the imperative aspect that is so central to social norms. Of course, there will always be ‘Medelsvenssons’ (the average Swede) in the statistical meaning. It is often this statistical use of the norm concept that is applied in norm-critical discourses. The general assumption here is that normality, i.e., what is common practice, often involves a repressive attitude against what is deviant. It is the third aspect of the norm concept mentioned by Therborn that ­corresponds to the common definition of social norm, i.e., what we ought to do. In the norm discourse, this is often a matter of actions exemplified at the individual level. This is in line with a modern analytical discourse on sociological theory in which we can find reasoning oriented toward the explanation of social actions (cf. Elster, 2015). In this perspective, norms are often an

What Do We Do with Norms

53

i­ mportant explanatory factor as a kind of social force. This is the main use and function of the norm concept in social science. Most explicit definitions of norms in the social science discourse are short and seemingly simple. The aim here is not to present a conclusive list of definitions but to give some examples of definitions covering the main lexical aspects. Even though the definitions seem in line with most commonsensical understandings, the implications will obviously be complex, depending on theoretical contexts. These contexts vary from norms as random normative forces in individuals’ lives to normative components in social systems. A reasonable lexical understanding of the social norm is Jon Elster’s statement that it ‘is an injunction to act or to abstain from acting’ (Elster, 2015: 348). The rationale behind this imperative is not instrumental or dependent on the outcome; instead, ‘the action is mandatory in itself’ (ibid: 71). Elster engages in a rather complex and multidimensional classificatory exercise, identifying different kinds of norms even though he states that the dividing lines can be fluid (ibid: 351). He mentions social norms, moral norms, quasi-moral norms, legal norms and conventions. When Elster identifies different types of norms in terms of random examples, these illustrations are in line with most conventional understandings of norms. When he specifies the defining characteristics, the situation becomes more problematic. Examples of social norms, according to Elster, include taking revenge for an insult, not eating human flesh, norms of etiquette and norms against incest (ibid: 71, 91, 348). To understand the functioning of social norms, we have to realize that they are ‘triggered by the presence or behaviour of other people’ (ibid: 91), and even more concretely, they are ‘triggered when other people can observe what the agent is doing’ (ibid: 91). Furthermore, the causal e­ fficacy of social norms relates to sanctions or rather the avoidance of sanctions (ibid: 358). This detailed account of Elster’s description of social norms reflects the complexity, not to mention the ambiguity, in the definitions. When distinguishing between different kinds of norms, primarily social and moral, the differentiating criterion becomes mainly psychological. No intrinsic quality of social norms fundamentally differentiates them from moral norms; the ­psychological reactions of the actor and observer accomplish this. When a violation of a norm leads to shame in the violator and contempt in the observer, we have a social norm (ibid: 147). When a violation of a norm leads to anger in the observer and guilt in the violator, we have a moral norm (ibid). This characterization of moral norms becomes somewhat confusing when E ­ lster also states that moral norms are ‘capable of shaping behaviour even when the agent b­ elieves herself to be unobserved by others’ (ibid: 351). A possible interpretation of this ‘contradiction’ is that the staging of a norm in terms

54

Sohlberg

of the observer and ­violator is an analytic structure and does not refer to concrete situations. However, this does not solve the problem with a psychological interpretation of social norms (see, e.g., Dubreuil and Grégoire, 2012). The dichotomization of feelings in terms of the pairs—contempt-anger and shame-guilt—seems too exact to correspond to fuzzy psychological realities. We could also assume substantial empirical variation in people’s reactions to the violation of a norm, implying that one and the same (social) imperative could be both a social and a moral norm. An alternative interpretation would be that moral norms are about basic principles such as injustice, harm and rights (cf. Turiel, 1983), whereas social norms could be seen as more a matter of conventions. If psychological aspects are taken sincerely, it would also be appropriate to include the processes of internalization of norms. Staging the formation of norms analytically in terms of actors and observers in a synchronous situation misses how the collective consciousness is transmitted in complex processual forms in the lives of human beings. In Elster’s understanding of norms, they do not exist in order to fulfil a function in opposition to the functional tradition. Another approach to social norms is formulated by Elinor Ostrom. In her approach norms are seen as important for purposes such as the sustainability of cooperation in the formation of commons. At the outset, Elinor Ostrom defines norms in a way similar to Elster, but she introduces cognitive as well as normative elements in a clearer way: ‘Social norms are shared understandings about actions that are obligatory, permitted, or forbidden’ (Ostrom, 2000: ­143–144). This also follows the individual’s own valuation: ‘By norms I mean that the individual attaches an internal valuation–positive or negative–to ­taking particular types of action’ (Ostrom, 1997: 9). Elster’s definition of norms that focuses on single acts is atomistic in the sense that it does little to help understand the background of aggregated patterns of conformity. Elster’s statement about an ‘injunction to act or to abstain from acting’ does not clarify the history behind this injunction nor how this imperative is socially established and relates to a possible system of norms. Elster goes somewhat beyond the singular action by relying on the causal efficacy of sanctions, i.e., if we assume that these sanctions are systematic in some sense. A discourse focusing on random examples of singular norm-governed acts without relating these acts to their embeddedness in a wider normative context is unsatisfactory. This normative context can never be a matter of simple aggregation of norm-governed actions, as the ‘first mover’ of these actions remains hidden. Identifying this normative context has been an important part of the system-oriented perspectives in sociology.

What Do We Do with Norms

55

By focusing on ‘shared understandings’ in the quotation above, Ostrom offers something of a clue to the regularity of social actions. However, Ostrom’s norm reasoning is generally interesting as she systematizes the notion of norm in the context of the problem of how the commons could be established in a sustainable manner. Ostrom identifies several conditions for common pooled resources to be functioning in the long run, where e.g., s­ ocial norms such as ‘reciprocity, trust and fairness’ are important (Ostrom, 1990: 154). ­Methodologically, it is interesting that she identifies functional requisites based on empirical field studies and not on thought experiments. From a chronological perspective, Malinowski should be mentioned before Parsons, but in a trajectory beginning with action and ending with social systems, the right place for Malinowski is the intermediate position. ­Malinowski criticizes an abstract system perspective and argues that social science should focus on the group and not on abstract social systems. Malinowski’s criticism could be applicable to Parsons even if it primarily is directed against ­Radcliffe-Brown and the French school (Durkheim). Malinowski argues against an abstract notion of a (moral) system: The group, after all, is but the assemblage of individuals and must be thus defined-unless we fall into the fallacy of “group mind”, “collective sensorium”, or the gigantic “Moral Being” which thinks out and improvises all collective events. malinowski, 1939: 938

Even if Malinowski is programmatically critical of an abstract system perspective, he formulates a general model for institutional analysis in his theoretical practice. This scheme is meant to be a heuristic device usable for fieldwork. To conduct an institutional analysis, there are several aspects that have to be addressed. Some of these aspects include the charter of the institution (recognized purpose), the organization, norms, the material apparatus, activities, and finally the function of the institution to be identified by the researcher (cf. Sohlberg, 1999). We can also see that although Malinowski was considered the arch-functionalist, he only gives limited importance to the normative aspects of institutions. Even if the functional tradition is the main system-approach focusing on norms it is not the case that, on an empirical level, it is assumed that norm-adherence is, or even ought to be universal. For example, Robert K. Merton states, ‘Some (unknown) degree of deviation from current norms is probably functional for the basic goals of all groups’ (Merton, 1968: 236).

56

Sohlberg

The same ­applies to Malinowski that in his fieldwork is always interested in more or less systematic deviations from norms. We have seen that there seems to be a more or less continuous system dimension in which the understanding of actions can be expanded in a ‘systemic way’. This system building can be accomplished in two ways. The first way is to systematize the analytic context of the social action, as with Parsons’ unit acts. The second way is to take social actions for granted and to focus on the overall qualities of the analytic social system. Parsons’ later work is most famous for this explicit system approach in which the system seems to live its own life seemingly independent of specific social actions. The best-known example is, of course, the AGIL-scheme. The system’s theoretical angle on norms is to focus on norms’ essential functionality from the system point of view. This explicit system perspective on norms can be found in Parsons’ later writings. For example: The value system of any social system controls system processes via the motivation of individuals. In general, the value system operates through two channels: (1) the internalization of the value system (and its appropriate sub-systems) in the personalities of individual members of the system; (2) the sanctions administered by other system members which, in a changing situation, tend to stabilize the internalized orientations and adapt specific behavior to the changing exigencies. parsons and smelser, 1984: 178

The value system, in more concrete setting norms, is from this perspective regarded as functional requisite, necessary for the ‘coordination’ of individuals and where some degree of normative consensus is regarded as necessary for the system’s continued existence. However, this idea of the need for some degree of normative consensus is not restricted to the functional tradition. Anthony Giddens is a sociologist well known for his fundamental critique of functionalism. He formulates what, in other words, could be identified as the functional requisites of the cosmopolitan nation: A cosmopolitan nation needs values to which all are committed, and an identity with which citizens are comfortable, but it also has to accept ambiguity and cultural diversity. giddens, 1998: 137

Even though Giddens programmatically is critical indeed of the functional ­vocabulary, it is obvious that he here identifies conditions (the needs of a

What Do We Do with Norms

57

­ ation) that have to be fulfilled in order for a cosmopolitan nation to function n in the long run, which is a classic functional point of view.

Conclusion: So, What Do We Do with Norms?

In line with the approach in this chapter of identifying what we do with norms rather than what a norm really is, the conclusion focuses on norms as tools. The first question is whether the concept of the norm is helpful in our everyday lives as social actors. Even if it is often the case that we conform to norms as well as break them, from a phenomenological perspective, we rarely ask after norms before acting. In very specific situations, we can sometimes ask about specific behaviour or manners, but seldom do we ask after general norms. This kind of question often concerns matter of etiquette, as when a few select men wonder whether it is acceptable to wear a wrist watch with a tail-coat. In a serious social interaction, we seldom ask what norms are suitable or applicable in the situation. An indicator of the relevance of the system perspective is that we often try to identify the type of situation in which we are participating from a symbolic interactionist perspective. Norms, or rather ways to behave, are dependent on the definition of the situation. In a lexical sense, the norm concept is generally used for more general trans-situational imperatives not confined to specific situations. Thus, the norm concept is apparently not a very helpful tool in regard to guiding our everyday actions. It is more helpful in an abstract way when we try to understand action patterns from the outside. The function of the concept is generally to understand and explain patterns of human behaviour over time. This function can be seen from an actor perspective as well as a system perspective. The Norm Tool in Actor Vocabulary Social science in general has developed a specific actor vocabulary, which is often based on a methodological individualist conception of the actor as the basic ‘component’ of social life. Examples of the actor vocabulary related to norms are concepts such as preferences, intentions/goals, motivation, information and (of course) social action. The concept of the norm is generally part of the actor vocabulary in the way that actors are supposed to follow norms or at least to relate to them. Seen in this action context, the norm concept is often used in a kind of black box reasoning to explain conformity and aggregated patterns of action. This is often done in a post hoc way in which there is no systematic conceptualization of a coherent norm system. The problem

58

Sohlberg

with understanding norms as isolated causal forces influencing or causing an individual’s behaviour is that it then is vulnerable to atomistic ‘cherry picking’. Based on the observation of any behaviour, abstraction always makes it possible to infer the existence of a norm prescribing the observed behaviour. A complexity following the application of the norm concept at the actor level is that the normative element is generally only one of several aspects influencing social actions. Reasonable aspects include situational factors, goals, resources, and preferences in the understanding of social actions. The situation becomes even more complex as, for example, preferences and goals generally are influenced by normative aspects but are not reducible to them. Studying the genesis and trajectories of norms at the actor level means identifying norm sources and studying the socialization process in a broader sense. The Norm Tool in System Vocabulary The conceptual space for norm reasoning from a system perspective includes concepts such as system coordination/integration, functional requisites, the degree of normative consensus and anomie. Norms in the system perspective are approached methodologically in a radically different way than they are at the actor level. When perceived in a system perspective, norms are not seen as random aggregates of the social actions performed by individuals in a specific setting. The system perspective of norms is characterized by the assumption that any social group or social system has functional requisites that must be fulfilled for the group or system to function adequately over time. Considering the immense variety of social actions possible in any social setting as well as actual conformity, the concept of the norm is used to understand this relative order. In social science, normative reasoning has often been opposed to reasoning in terms of material conditions. This is not self-evident, and an interesting question is normative reasoning can be combined with the materiality of the social world (cf. Lockwood). It is also reasonable to see as well material limitations and normative limitations as part of restrictions on the degrees of freedom of social action. In view of the conventional contradiction between idealism and materialism in social science, it is interesting to find that a degree of normative consensus is regarded as a functional requisite in both traditions. In the Marxist tradition—and, to some extent, in the feminist tradition—this consensus is phrased in terms of false consciousness and ideology. These are factors that are regarded as requisites for the oppressive system to continue. In the functional tradition, the requisite of a degree of normative consensus is not problematized as a matter of ideology or false consciousness. Seen in dynamic terms, the trajectory of norms from a system perspective is rather fuzzy and vague. There are few accounts of the development of norms

What Do We Do with Norms

59

from a system perspective. We have Durkheim’s classic idea of a shift from mechanic to organic solidarity in which the first type of cohesion is based on kinship, and normative conformity is based on a strong collective consciousness. Due to structural differentiations in modernity, there is a change in the normative locus from a strong collective consciousness to functional dependence and contract relations, i.e., trust becomes a functional requisite. The general requisite of trust and predictability can in turn be operationalized into a diversity of norms. An analogous trajectory of normative development can be found with Michel Foucault where we have a development from external repressive normative regimes to self-disciplinary approaches over time, i.e., the internalization of norms and thought structures mediated by discourses. This chapter began with a criticism of the reification of the norm concept, i.e., the idea that norms are isolated normative units, causing conforming ­behaviour at the individual level and social cohesion at the systemic level. However, what would it mean to deconstruct this reification? It would certainly not mean that normative factors should be regarded as epiphenomena of lesser importance but rather that they are intertwined in complex and fuzzy ways with other aspects of social reality. When referring to Max Weber’s typology of social actions I mentioned that the seemingly normative kind of social action is the value-rational action, even though that Weber prophesized about the coming dominance of means-end rationality. But, it is also the case that normative aspects are important in regard to this type of ‘rational’ action. Another example of the difficulty in isolating the normative factor from other aspects or levels can be found in Nicos Mouzelis’ interesting formulation of different and complimentary layers of social interactions (Mouzelis, 1995). A basic layer in this account is explicitly normative aspects regulating what is normatively prescribed, accepted and prohibited. Mouzelis relates this level to Talcott Parsons’ theories. A ‘middle’ layer influencing the interaction, called dispositional, is based on Bourdieu’s notion of habitus. This relates to the ­specific ‘style’ based on factors such as class and gender that characterize the participant’s behaviour in the concrete action, e.g., the type of language, the worldview and references. Finally, we have an aspect of the concrete ­situation that is a more spontaneous interaction in which the participants react and interact in a process. However, normative factors can also be reasonably assumed to influence the dispositional and interactive aspects in the Mouzelis model. The distinction between these complimentary aspects of interaction is, as Mouzelis is aware, analytical. It is when we forget this analytical approach and reify norms that we begin to regard the normative aspect as a separate ‘thing’ with causal power and not as a theoretical tool for understanding normativity, always in complex

60

Sohlberg

i­ nteraction with other factors. In social science, we need a theoretical tool, i.e., the norm concept, to analytically ‘purify’ normative aspects of social reality. In lived social life, whether as actors or from a system perspective, normativity is always intertwined with other aspects of social life. References Ahrne, G. (2016). “If not, why not and what if: asking counterfactual questions”, Eds. P. Sohlberg and H. Leiulfsrud. Theory in action. Theoretical constructionism. Leiden: Brill Publishers. Aristotle. (2013). The politics and the constitutions of athens. Ed. S. Everson. Cambridge: Cambridge University Press. Bicchieri, C. (2006). The grammar of society. The nature and dynamics of social norms. Cambridge: Cambridge University Press. Bourdieu, P. (1984). Distinction: a social critique of the judgement of taste. Cambridge, ma.: Harvard University Press. Bourdieu, P., Chamboredon, J. and Passeron, J.C. (1991). The craft of sociology: epistemological preliminaries. Berlin: Walter de Gruyter. Delanty, G. and Strydom, P. (eds.). (2003). Philosophies of social science: the classic and contemporary readings. Maidenhead: Open University. Dubreuil, B. and Grégoire, J.-F. (2012). “Are moral norms distinct from social norms? A critical assessment of Jon Elster and Cristina Bicchieri”, Theory and Decision. 75(1): 137–152. Durkheim, É. (1982). The rules of sociological method: and selected texts on sociology and its method. London: Macmillan. Eliaeson, S., Mindus, P. and Turner, S.P. (eds.). (2014). Axel Hägerström and modern social thought. Oxford: Bardwell Press. Elster, J. (2015). Explaining social behavior: more nuts and bolts for the social sciences. Revised edition. Cambridge: Cambridge University Press. Foucault, M. (2002). The archaeology of knowledge and the discourse on language. London: Routledge. Giddens, A. (1986). The constitution of society. Outline of the theory of structuration. Cambridge: Polity Press. Giddens, A. (1998). The third way. The renewal of social democracy. Cambridge: Polity Press. Guneriussen, W. (2016). “Sociology and the power of (ordinary) language”, Eds. P. Sohlberg and H. Leiulfsrud. Theory in action: theoretical constructionism. Leiden: Brill Publishers.

What Do We Do with Norms

61

Habermas, J. (1989). “The theory of communicative action”, Lifeworld and system: a critique of functionalist reason. Vol. 2. Cambridge & Oxford: Polity Press. Hacking, I. (1999). The social construction of what?. Cambridge, ma.: Harvard University Press. Hechter, M. and Opp, K. (eds.). (2005) [2001]. Social norms. New York: Russell Sage Foundation. Lockwood, D. (1992). Solidarity and schism: “the problem of disorder” in Durkheimian and Marxist sociology. Oxford: Clarendon. Malinowski, B. (1939). “The group and the individual in functional analysis”, American Journal of Sociology. 44(6): 938–964. Marx, K. and Engels, F. (1978). The Marx-Engels reader. 2nd edition. New York: Norton. Merton, R.K. (1968) [1957]. Social theory and social structure. New York: Free Press. Mouzelis, N.P. (1995). Sociological theory : what went wrong?: diagnosis and remedies. London: Routledge. Ostrom, E. (1990). Governing the commons: the evolution of institutions for collective action. Cambridge: Cambridge University Press. Ostrom, E. (1997). “A behavioral approach to the rational choice theory of collective action: presidential address, American Political Science Association”, The American Political Science Review. 92(1): 1–22. Ostrom, E. (2000). “Collective action and the evolution of social norms”, The Journal of Economic Perspectives. 14(3): 137–158. Parsons, T. (1937). The structure of social action. A study in social theory with special reference to a group of recent European writers. 1st edition. New York: McGraw-Hill. Parsons, T. (1970). Social structure and personality. London: The Free Press, CollierMacmillan Ltd. Parsons, T. and Smelser, N.J. (1984). Economy and society. A study in the integration of economic and social theory. London, Boston, Melbourne and Henley: Routledge & Kegan Paul. Skjervheim, H. (1976). Deltakar og tilskodar og andre essays. [Participates and assigns and other essays]. Oslo: Tanum-Norli. Sohlberg, P. (1999). Mål och mening i samhället: funktionalistiska program i samhällsvetenskapen. [Ends and meanings in society: functionalistic programs in social science]. Uppsala: Acta Universitatis Upsaliensis. Sohlberg, P. and Leiulfsrud, H. (eds). (2016). Theory in action. Theoretical constructionism. Leiden: Brill. Spencer, H. (1981). The man versus the state. With six essays on government, society and freedom. Indianapolis: Liberty Classics. Swedberg, R. and Agevall, O. (2005). The Max Weber dictionary: key words and central concepts. Stanford, ca.: Stanford Social Sciences.

62

Sohlberg

Therborn, G. (2002). “Back to norms! On the scope and dynamics of norms and normative action”, Current Sociology. 50(6): 863–880. Turiel, E. (1983). The development of social knowledge: morality and convention. ­Cambridge: Cambridge University Press. Whitehead, A.N. (1925). Science and the modern world. Cambridge: Cambridge ­University Press.

chapter 4

Colligation Richard Swedberg Introduction The core meaning of the word colligation is ‘binding together through an idea’; and what it binds together are facts.1 The term is not very much used in contemporary social science, including sociology, but as I shall try to show, it can be quite helpful for one activity that is very important for those who are interested in theorizing, and that is concept construction. It can be of assistance both when you develop new social science concepts and when you want to make changes in existing ones. The central idea of colligation is very flexible, something that makes it both useful and interesting to work with. While it is easy to sense the potential of colligation simply by thinking about the notion of ‘binding together facts through an idea’, this concept is also extra exciting today, precisely because it has been used so little. It has not been locked into a fixed formulation; and its full potential is not known. It is an open concept. Social scientists use many different types of concepts. The ones that are best-known are those that have been created as part of an effort to analyze some important social phenomenon. ‘Class’, ‘status’ and ‘norm’ belong to this category. Social scientists sometimes also draw on the concepts that are used by the people they study, as exemplified by the way that ethnomethodologists and phenomenologists work. Alfred Schutz calls these everyday concepts firstorder constructs, saving the term second-order constructs for the concepts that social scientists create on the basis of these everyday concepts (Schutz, 1954). Colligation as a tool for concept construction does not fit either of Schutz’s two categories, and is more similar to such concepts as the ideal type (Weber) or the sensitizing concept (Blumer). What these three concepts have in common is precisely that they contain prescriptions for how to construct and/or how to use concepts in empirical research. Using Schutz’ terminology, we may call this type of concept third-order constructs.

1 For help with this paper I would first of all like to thank Mabel Berezin. I would also like to thank Raimund Hasse, Paul Lichterman, Lyn Spillman and the two editors of this volume.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_005

64

Swedberg

Besides the evocative notion that is at the core of colligation—the idea that scientific phenomena consist of facts that are bound together in an analytically useful manner through an idea—this concept is also attractive in that it has a practical dimension. As I will try to show, it can be of concrete help to the social scientist when she carries out an analysis. Note also that there are many different items that need to be bound together when you do research: facts into a concept; concepts into a theory; parts into a system; and so on. The central concepts in sociology often have histories that are fairly well known. Many of these histories are exercises in intellectual history, while those that are more ambitious fall into what has been called the historical sociology of concept formation (Somers, 1995). Concepts that are less known, in contrast, may need to be handled a bit differently, precisely because they are marginal and have developed in a more accidental manner. Here you may want to draw on what Merton calls sociological semantics, since it is often necessary to trace the way that some word or phrase has changed meaning over time. A  well-known example of this last type of analysis is the study of the term ‘serendipity’ (Merton and Barber, 2006; for a programmatic statement on sociological semantics, see Zuckerman, 2010). The rest of this paper will first be devoted to a presentation of the history of the different meanings of colligation that I have been able to locate. I will then proceed to a more general discussion of colligation, in which I suggest how it can better serve both social science and sociology than it does today. As already mentioned, this means to focus on how colligation can help us to better understand the construction of concepts and how to do this in conjunction with empirical research. The main uses of the term colligation are as follows. The first appearance of the word seems to have been in the 1500s and 1600s, at which time it meant ‘­material binding together’, ‘connection’, ‘conjunction’, ‘alliance’ and ‘union’ (oed, 2014). In the 1700s and the early 1800s it seems to have been used very ­little—that is, until it was seized on by philosopher and scientist William Whewell (1794– 1866), who assigned the term colligation a central place in his work. Whewell, who was an important academic figure in his time, argued that progress in the natural sciences comes about through a process called colligation, which he defined as the binding together of facts through scientific conceptions. Whewell was a major intellectual figure in Victorian England, and his work was well known by people such as John Stuart Mill, Charles Darwin and ­Herbert Spencer, but the term colligation fell into oblivion after his death. It was not until a century later, in the 1950s and thanks to a work in the philosophy of history, that colligation again began to be used and discussed. Soon it became a well-known word in the circles of historians, who used it to describe the way that events and historical periods can be conceptualized.

Colligation

65

A few decades later the term began to make its way into the works of sociologists as well. At first it was sociologists interested in history who came across the term and used it in their analyses. Lately, however, also cultural sociologists have incorporated it into their vocabulary. They mainly use it as a synonym for binding together facts, without any special reference to concepts. 1

Colligation as Binding Together Facts Through Conceptions in the Natural Sciences (Whewell)

The ideas of William Whewell belong to a very different discourse than that of modern science, not to speak of modern social science and sociology. Whewell’s main concern was to advance science in general, and whatever he thought of social science was part of this general ambition. The closest equivalent in the history of sociology to someone like Whewell is Auguste Comte (1798–1857), who was his contemporary and a disliked competitor. Both wanted to write a systematic history of all the sciences, and also to further the advancement of the sciences by figuring out what had made their progress possible. Whewell (pronounced weh-ell) was a polymath and a very successful academic in his days (e.g. Snyder, 2006, 2011, 2012). He first became Professor of Mineralogy in 1828 and a decade later Professor of Moral Philosophy. Both appointments were at Cambridge University, where he also held the position of Master of Trinity College. Whewell was a prolific writer and author of two giant works: History of the inductive sciences (1837, 3 vols.) and Philosophy of the inductive sciences (1840, 2 vols.). In Whewell’s view, what accounted for the impressive progress of the natural sciences since the 1500s was induction based on solid facts. So far Whewell agreed with Francis Bacon, whose works he deeply admired. What Bacon had missed, however, Whewell insisted, was that induction was as much based on ideas as it was on facts. ‘Ideas are necessary’, as he put it (Whewell, 1860: 134). Besides being a scientist and a historian of science, Whewell also had another talent that was much appreciated in his time. This was his capacity to come up with new and fitting words. His most famous accomplishment in this regard was his invention of the word ‘scientist’. He also suggested to Faraday the use of the words ‘cathode’ and ‘anode’. To cite Robert Merton, Whewell was a ‘talented word-coiner’ (Merton, 1997: 237). Whewell believed strongly that a scientist should begin by studying a phenomenon in a penetrating manner, and first when this had been done, try to formulate a theory. This way of proceeding was also reflected in Whewell’s own attempt to find out what accounted for the progress of the natural sciences. He first studied their history (in a 3-volume work of more than 1,600 pages); and

66

Swedberg

only when this had been done, did Whewell proceed to the task of drawing his conclusions (in a 2-volume work of more than 1,200 pages). The full title of the latter work describes the procedure: The philosophy of the inductive sciences, founded upon their history (1840). In Whewell’s view, two basic processes account for the progress of the natural sciences: The colligation of facts and the explication of conceptions (Whewell, 1847: 5). By colligation he meant the process ‘by which the conceptions more strictly bind together the facts’ (Whewell, 1840, 2: 170–171). The explication of conceptions referred to the gradual clarification, sharpening and generalization of these conceptions, through research. Colligation did not come about just by summing up the facts and generalizing from them. It meant adding a new element that was different from the facts: ideas or conceptions. ‘The facts are not only brought together, but seen in a new point of view. A new mental Element is superinduced’ (Whewell, 1858: 71). In an elegant metaphor Whewell likened the role of ideas in science to the thread in a pearl necklace: When the Greeks, after long observing the motions of the planets, saw that these motions might be rightly considered as produced by the motion of one wheel revolving in the inside of another wheel, these Wheels were Creations of their minds, added to the Facts which they perceived by sense … The same is the case in all other discoveries. The facts are known, but they are insulated and unconnected, till the discoverer supplies from his own stores a Principle of Connexion. The pearls are there, but they will not hang together till some one provides the String. whewell, 1847: 48

In another metaphor Whewell likens the act of a successful induction to the making of a well-tasting pie from Devonshire. This type of pie got its unique flavor from the fact that the cook put into it every well-tasting morsel that was available. ‘For the beauty of my induction’ Whewell once wrote, ‘is, that it is like the Devonshire man’s pie, into which he puts everything which he catches’ (Wheewell, 1835). As a famous example of a colligation Whewell cited the example of Kepler. In Kepler’s time, it was believed that the orbits of the planets formed a circle. Kepler, however, questioned this; and his discovery was that Mars and other planets move in the shape of an ellipse. In Whewell’s opinion, it was Kepler’s training in geometry that had made it possible for him to suggest this solution, which in a stroke explained the facts that Tycho Brahe and he himself had ­observed but did not fit the earlier theory.

Colligation

67

The emphasis on the mixture of facts and concepts that characterizes Whewell’s notion of colligation probably originated in his reading of Kant, whose ideas he admired. As opposed to Kant, however, Whewell did not assume that reality was ultimately unknowable. In Whewell’s view, the scientist can get to know reality with increasing clarity and precision, and should aim to do so. A scientific concept, according to Whewell, is based on facts tied together by an idea. If either the facts or the ideas are given too much of a place in the work of the scientist, Whewell also argued, the result will be unsatisfactory. Both are needed: When Ideas and Fats are separated from each other, the neglect of Facts gives rise to empty speculations, idle subtleties, visionary inventions, false opinions concerning the laws of phenomena, disregard of the true aspect of nature: while the want of Ideas leaves the mind overwhelmed, bewildered, and stupefied by the particular sensations, with no means of connecting the past with the future, the absent with the present, the example with the rule; open to the impressions of all appearances, but capable of appropriating none. Ideas are the Form, facts the Material of our structure. whewell, 1847: 47

Whewell was also careful to point out that scientists do not make their discoveries by accident; these are the result of careful preparation in combination with talent. The scientist has first of all to be well trained. She also has to carefully observe what is going on. And finally, she has to single out those aspects of some phenomenon that are of interest to science (‘the facts’). The last operation Whewell called the Decomposition of facts; and to him it primarily meant the kind of operations that make it possible for the scientist to measure, count and the like (Whewell, 1847: 33). The way you go about colligation, according to Whewell, is very closely related to induction. This was the method that Bacon had advocated so strongly in Novum organum (1620); and it was also the best way to proceed according to Whewell. Whewell, however, felt that Bacon had failed to emphasize the elements of ideas that goes into the process of induction. Bacon has … put prominently forwards the necessary dependence of all our knowledge upon Experience, and said little of its dependence, e­ qually necessary, upon the Conceptions which the intellect itself supplies. whewell, 1860: 135

68

Swedberg

As a consequence of the importance that Whewell attributed to the element of ideas in induction, he wanted to add to and in this way improve Bacon’s method. This is why he named the work in which he summed up his view of colligation and induction Novum organon renovatum (Whewell, 1858). Science was mainly improved through the explication of conceptions; and this typically took place through debates between scientists., according to Whewell. As science progressed, what had once been facts tied together through ideas (colligation) were often treated as facts. Whewell, in other words, was a proponent of the view that facts are ‘theory-laden’ (Hanson, 1965). It was earlier mentioned that Whewell had a special talent for coining words; and the role of naming a phenomenon should also be mentioned in relation to the process of colligation. In the more than hundred pages that Whewell devotes to this issue in one of his works, he does not explicitly state that colligation can be improved or facilitated through a good name (Whewell, 1858: 257–370). But this is where his argument is leading. A s­ cientific term that is well chosen becomes for example ‘a better instrument of thought’ (Whewell, 1858: 354). Besides making a scientific term easier to remember, the reader is told, a good name may also stimulate the imagination of the scientist. Whewell’s ideas about the role of definitions in science are also worth mentioning. Definitions, he argues, are of little value unless they are used in research. Many scientists, he notes, believe the opposite and seem possessed by a ‘craving for definitions’ (Whewell, 1858: 369). In Whewell’s view, this obsession had its origin in the very successful way in which definitions had been used in geometry. Discussing definitions or working these out through discussion, however, does not advance science. The main reason for this is that all the terms in the definition have to be defined in their turn; and this makes the whole enterprise futile. Scientific progress can only come about through research, according to Whewell, not through definitions. Whewell’s ideas on colligation were well regarded in his time but also received their share of criticism. The most important criticism came from John Stuart Mill, with whom Whewell had an ongoing debate for many years. In A system of logic (first published in 1843) Mill states that Whewell had misunderstood the nature of induction. The scientist does not add ideas to the facts, as Whewell argues; she merely sums up what exists in reality (Mill, 1952: 192). The example that Mill used to illustrate the difference between his own view of induction and that of Whewell, was the discovery by Kepler of the elliptical shape of the orbits of the planets. According to Mill, Kepler first made his observations of how the planet Mars moved, and then summed these up, ­saying

Colligation

69

that they have the form of an ellipse. The ellipse exists in reality and is not something that was added by Kepler. In Whewell’s view, in contrast, the facts of how Mars moved had been known for a long time before Kepler through the work of Tycho Brahe. But it was Kepler who succeeded in making sense of them, and he did so by conceptualizing them in the form of an ellipse, a figure that he was familiar with through his training in geometry. One commentator who was very positive to Whewell’s notion of colligation was Charles Sanders Peirce (1839–1914), another polymath as well as an accomplished historian of science. In Peirce’s mind, ‘Whewell was a most admirable reasoner, who is underestimated simply because he stands detached both from the main current of philosophy and from that of science’ (Peirce, 1998: 46). In Peirce’s view, the idea that scientists make their discoveries by tying ­together facts with the help of a new concept or idea was an important insight. Whewell, according to Peirce, was ‘the only man of philosophical power conjoined with scientific training who had made a comprehensive survey of the whole course of science’. Whewell had also confirmed Peirce’s own view ‘that progress in science depends on observation of the right facts by minds furnished with appropriate ideas’ (Peirce, 1935: 604). In fact, Peirce at times used the term colligation more or less as synonymous with his own famous notion of abduction, by which he meant the only kind of a mental operation through which something truly new is created in science. Induction and deduction were indispensable in their own right, according to Peirce. But their main tasks were to flesh out and prove the insight that could only come about through abduction. It should finally be pointed out that Whewell’s ideas about colligation were only applicable to the natural sciences. While Comte extended his ideas about positivism, which had its origin in his work on the natural sciences, to ‘physique sociale’ or ‘sociologie’ (as he later called it), Whewell did not. The term sociology cannot be found in Whewell’s work nor did he address the kind of topics that modern social scientists associate with this term. Whewell felt that a different approach from that of the natural sciences was needed to deal with the social sciences. Some of what Whewell thought of the social sciences can be read out of his view of political economy, a topic he was very interested in and also devoted quite a bit of study to (e.g. Snyder, 2006: Ch. 5). The main thrust of his views on this topic was that Ricardo was wrong in suggesting that deduction was the way to proceed. You have to work very closely with the facts, and go from there, in Whewell’s view; and this was also true for political economy.

70 2

Swedberg

Colligation as Binding Together Events into Historical Wholes (Walsh)

Even though Whewell was a major figure in his time, the term colligation did not survive his death. One reason for this, as already noted, is that John Stuart Mill severely criticized Whewell in A system of logic. Mill attacked Whewell’s notion of induction, which is closely related to his concept of colligation; and Mill’s work was very popular, going through eight editions during the author’s lifetime (e.g. Strong, 1955). The term colligation did not reappear until about a century after Whewell’s death; and when it did so, it was in a very different academic context than its original one. Colligation became popular in the 1950s and 1960s primarily among students of the philosophy of history. According to its new advocates, colligation could be applied to a different set of facts than it had originally been intended for, namely historical actions and events. As it turns out, the argument as now made that colligation should be used on historical facts precisely because these are different from scientific facts. The person who singlehandedly revived (and reinterpreted) the notion of colligation was a British philosopher at Oxford University by the name of W.S. Walsh (1913–1986). Over a period of some twenty-five years, Walsh wrote on colligation and its role in historical analysis, as part of his interest in the philosophy of history. Altogether he produced three major statements on colligation (Pompa, 2004). Walsh’s argument can be summarized as follows. While in the natural sciences you link facts together based on their ‘outer’ features, in history you also have an ‘inner’ side to take into account. When human beings act, there is some kind of thought or intention involved; and this means that the resulting events are linked together in ways that are different from the way they are linked together in the natural sciences. A historian typically analyzes a period, such as the Industrial Revolution or the Enlightenment; and this period, in its turn, consists of many events that must be pulled together or colligated in some fashion. There exist leading ideas during these periods; and it is these ideas that shape the events as well as link them together. After having summarized his view on how to use colligation in history, Walsh asks the practical question, ‘How does he [the historian] set about doing this?’ (Walsh, 1942: 133). This question shows clearly that Walsh saw colligation as a practical tool. His answer to the question was that a historian should proceed in three steps. ‘He first of all surveys the events of the period he has chosen and tries to connect them together under certain leading ideas’ (Walsh, 1942: 133).

Colligation

71

When this has been done, she should try to determine the way that the leading ideas or the dominant concepts, as he also calls them, are interrelated. The third and last step consists of constructing a narrative of the individual events, in such a way that the analysis becomes a whole. In doing so, it is important to show that the dominant concepts make the individual events intelligible. Walsh’s argument in the 1942 article is suggestive but also quite abstract. It is especially hard to see what practical consequences this approach will have for the historian. In short, how does the binding together of facts in history (colligation) differ from the way in which you bind together facts in the sciences? How exactly do you go about constructing a historical analysis? Some answers to this question can be found in Walsh’s second writing on colligation, An introduction to philosophy of history. Walsh starts out his discussion of this concept by suggesting that the idea of explanation through ­deduction—that is, along the lines of Hempel’s covering laws—does not ­always work in history. Historians do not necessarily try to explain some phenomenon by looking at other phenomena that are similar; instead they often see it as part of some larger whole. When Hitler occupied the Rhineland in 1936, for ­example, you may explain this as part of his general plan to dominate the world. This example, according to Walsh, points to the fact that it is important to include the element of intention in the analysis. ‘Every action has a thought-side, which makes the whole thing possible’ (Walsh, 1967: 59). In the case just mentioned, the intention operates from behind the action, so to speak, propelling it onwards. But the interrelatedness of events can also take other expressions, according to Walsh. There also exist factors in the future that may pull events onwards and towards them. Hitler had an overall goal for his actions; and what he did at one point in time was linked to this goal and to what he intended to do in the future. In Walsh’s last writing on colligation, ‘Colligatory concepts in history’ (1967), he backtracks a bit and states that his earlier arguments about colligation had been ‘defective in various ways’ (Walsh, 1974: 134). He especially felt that he had emphasized the role of purposive action far too much in his earlier arguments. It may well be true, for example, that events can be pursued and are pursued in a purposive manner, as exemplified by the creation of the state of Israel in 1948. But this is an exception rather than the rule. History is usually messy; and this makes it hard to follow the way intentions work, and for actions to work out as their authors want. Walsh also suggested that colligation may be more important for the interpretation of historical events rather than for their explanation. We mainly link together events, he now said, to get a sense for what they are all about. People have ideas in common; and these ideas (and their accompanying ­actions)

72

Swedberg

come together in what we may describe as a period, e.g. the Greek Age of ­Enlightenment or the rise of the bourgeoisie in England. Colligatory concepts, Walsh now argued, have to fulfill two important conditions they have to ‘fit the facts’ and ‘illuminate the facts’ (Walsh, 1974: 140). In the literature on colligation that was inspired by the writings of Walsh, his ideas have been added to as well as criticized. There exist, for example, studies of the relationship between colligation and classification; attempts to develop a typology of different kinds of (historical) colligation; and also an attempt to show that the idea of colligation can make the teaching of history more interesting for students (e.g. Thompson, 1967; Cebik, 1969; McCullagh, 1978). One of Walsh’s students also made a direct attack on Hempel and his theory of the deductive-nomological model (Dray, 1959). Hempel’s response was that Walsh’s argument about the role of intention does not contradict the idea of a covering law; in fact, it confirms it (Hempel, 1965: 470–471). 3

Sociologists on Colligation: Explanation, Sequence Analysis and More

The term colligation has never been a common term in sociology; and there are no entries for it in the many sociological dictionaries that exist. Still, from the 1980s and onwards you can find a scattering of references to colligation in the work of sociologists (e.g. Abbott, 1984; Griffin, 1993; Spillman, 2004, 2014; Wagner-Pacifici, 2010; Abel, 2011; Lichterman and Reed, forthcoming). What is clear from these is that the term colligation entered sociology through the discussion of this concept in philosophy of history in the 1950s and 1960s, and not through a reading of Whewell. This is why it has mainly been used to better understand what is meant by ‘events’ and to improve the way that causality is understood in sociology. In cultural sociology, for example, colligation has been put forward as part of an attempt to challenge the conventional causeeffect type of explanation. It has, for example, been argued that colligation can be used as an alternative to the covering law idea (Spillman, 2004: 224–229; see also Spillman, 2014). Another attempt by a sociologist to use the concept of colligation can be found in an article from 1984 by Andrew Abbott, ‘Event sequence and event duration: colligation and measurement’. This article can be described as part of the author’s effort to develop a narrative positivism in the form of sequence analysis. It represents a very ambitious attempt, not only to show the limits of mainstream sociology but also to outline an alternative approach (e.g. Abbott, 2001).

Colligation

73

How is it possible, Abbott asks, to both have a stringent theory of narratives and at the same time be able to effectively deal with the concrete details of empirical reality? In mainstream sociology, he argues, this problem is handled through the idea that there are concepts that are linked to social reality via indicators. What Abbott suggests in his 1984 article, is to replace concepts and indicators with events and occurrences. The former is general, while the latter is close to reality. What Abbott tried to accomplish, in other words, was to translate colligation into a workable empirical tool for a new type of sociological analysis. In this sense, it is clear that he contributed something new to the discussion of colligation. He especially tried to make the idea of colligation considerably more empirical than it was in Walsh’s version; and he did so by arguing that it is not enough to have good ideas, these ideas also have to make contact with reality in a serious way. In my view, Abbott and other sociologists make a much too limited use of the concept of colligation, which they basically interpret along the lines of Walsh. As I see it, colligation should primarily be used for a different task in sociology than causality and the construction of events. Its main task is to construct concepts; and this is also in accordance with Whewell’s original use of the term. As such, colligation can be of strategic assistance for the sociologist when she theorizes. It can especially help her to better handle what comes before the final formulation of a concept and its presentation to the world. The approach to colligation I advocate differs on one important point from Whewell’s version. This is that colligation should primarily be seen as a process, and that this process to a certain extent can be analyzed and understood. Whewell’s notion of colligation is more along the lines of an aha-experience that takes place when you come up with a great idea that binds together all the relevant facts in some brilliant way. There is always an element of creativity to successful theorizing, but it also consists of a series of separate actions or practices that can be taught and learned. As a process, colligation broadly consists of two distinct parts: ­de-colligation and re-colligation. The reason for the de-colligation is that social actors typically already have concepts of what is going on around them, as a result of living in society. These preconceptions, as Durkheim calls them, must be broken up or de-colligated. What people mean by, say, suicide, crime or happiness, have to be unscrambled since they have emerged for very different purposes than social science analysis. It should be noted that there is no need for de-colligation when you carry out an analysis in the natural sciences. The reason for this is that ordinary people’s concepts are not part of what is being studied.

74

Swedberg

Once the facts have been de-colligated they need to be re-assembled in a new way or re-colligated. At this stage the researcher has to select those facts that are relevant for a sociological analysis and bind them together in such a way that they form a sociological concept. If the result of tying the facts together is simply to recreate a concept that already exists—say, the concept of role—there is no novelty to the colligation (or no colligation period, to use the term in its literal meaning). It is of course very seldom that a totally new and valuable sociological concept is created. Some minor innovation is more common and it comes about by adding to an already existing concept. The concept of role, for example, has been added to over the years, resulting in such concepts as role-set, role distance and role release.

Concluding Remarks on How to Use Colligation as a Practical Tool When You Theorize and Construct Sociological Concepts Those who theorize rightly are in the end lords of the earth. —william whewell2

In the beginning of this article it was noted that colligation is not a fully established concept in modern sociology. Hopefully, what has been said so far will convince the reader that colligation merits more attention and discussion. In my view, it is Whewell’s notion of colligation that we should use and work with, rather than that of Walsh. According to Whewell, colligation should be used to construct concepts; and it is to this task that I want to return in these concluding remarks. More precisely, I would like to discuss colligation as a practical tool for theorizing, as part of the empirical research process. Whewell is particularly valuable for what he has to say on three points in my view, none of which Weber or Blumer address with their concepts of the ideal type and the sensitizing concept. The first of these points or steps is the process that Whewell calls the ­decomposition of facts. You start the research process, he says, through observation. Weber and Blumer would agree: it is only through careful and detailed ­observation that you can learn about the phenomenon you want to study. What Whewell then adds with his notion of the decomposition of facts, is that you do not only observe, and then analyze what you have observed. You now 2 (Whewell, 1836 as cited in Snyder, 2006: 33).

Colligation

75

also have to transform what you see, hear and so on, into facts that can be used in the analysis. People in Science and Technology Studies refer to this process as the creation of ‘representations’ and have shown how microscopes and other instruments are often used for this purpose (e.g. Coopmans, Vertesi and Lynch, 2014). Step 2 for Whewell is what constitutes colligation in a more narrow sense. This is where you look at the facts and try to bind some of these together through an idea, so that you then can analyze them in a meaningful way. What is involved at this stage, according to Whewell, is not only induction, in the sense of generalizing from particular instances. At this stage you first of all have to come up with an idea; and it is with the help of this idea that you can tie some facts together and show what they have in common. Say that you are observing some individuals who are staying home with their children. These persons typically have to cook, clean and take care of the children in a million ways. What exactly are these activities like, from a sociological perspective; how should they be viewed? For a long time sociologists cast them as part of the role of the housewife. They were tied together, in other words, through the idea of role. Since a few decades, however, some of these activities have been understood to constitute work; and today there exist a number of studies of household work. This latter concept has, for example, made it possible to compare work in the market place to work in the household; the hour of work that males put in, to those of women; and so on. Step 3 for Whewell is to come up with a good name for the idea that ties together the facts. A really good name, he says, should express the idea in a very clear way. Ideally, it should also inspire other researchers and fire their scientific imagination. The name ‘household work’, for example, makes it clear that some activities in the household are directly comparable, in terms of efforts and skills, to the activities that go on in the factories and offices. ‘Work’ is also a classical term in sociology and gives associations to other kinds of work (emotional work, relational work, market work, and so on). Whewell’s three steps that have been discussed so far can be seen as part of a research practice that that is associated with theorizing. This research practice can in my view also be understood as a process that is broadly centered around the two concepts of de-colligation and re-colligation. To what Whewell says about the need to transform what we observe into facts, should be added that existing folk-concepts must be broken up before any colligation can take place. This is something that natural scientists do not have to do, even if also they have to de-colligate scientific concepts, once these are no longer useful and have turned into obstacles.

76

Swedberg

Talking about de-colligation means that we also need to talk about re-colligation. And to Whewell’s argument about the need to come up with some great idea that can bind together the facts, you need to add that there is more to colligation than so. Unless a concept is simply reconstituted, the typical situation can better be described as an amendment of some existing concept, than as the discovery of a new and major concept. Cast along these lines, the concept of colligation can be very useful in sociological theorizing, more precisely in that part of theorizing process that deals with the creation of concepts. It nicely complements the ideal type and the sensitizing concept. Colligation differs from the sensitizing concept in that it discusses the process through which a concept comes into being. And it differs from the ideal type by providing the researcher with an alternative way for how to create a useful scientific concept. This alternative way singles out two aspects Weber does not address in his theory of the ideal type: de-colligation and the decomposition of facts. Finally, the reformulated concept of colligation that is suggested in this article moves away from Whewell’s view that concept creation is basically the result of an aha-experience. It suggests instead that colligation, as well as concept formation more generally, is a skill that can be learned as well as taught. References Abbott, A. (1984). “Event sequence and event duration: colligation and measurement”, Historical Methods. 17(4): 192–204. Abbott, A. (2001). Time matters: on theory and method. Chicago: University of Chicago Press. Abel, P. (2011). “Singular mechanisms and Bayesian narratives”, Ed. P. Demeulenaire. Analytical sociology and social mechanisms. Cambridge: Cambridge University Press. Cebik, L.B. (1969). “Colligation and the writing of history”, The Monist. 53(1): 40–57. Coopmans, C., Vertesi, J. and Lynch, M. (eds.). (2014). Representation in scientific practice. Cambridge, ma.: The mit Press. Dray, W. (1959). “Explaining ‘what’ in history”, Ed. P. Gardiner. Theories of history. ­Glencoe, il.: The Free Press. Griffin, L. (1993). “Narrative, event-structure analysis, and causal interpretation in historical sociology”, American Journal of Sociology. 98(5): 1094–1133. Hanson, N.R. (1965). Patterns of discovery: an inquiry into the conceptual foundations of science. Cambridge: Cambridge University Press.

Colligation

77

Hempel, C. (1965). Aspects of scientific explanation. Glencoe, il.: The Free Press. Lichterman, P. and Reed, I.A. (Forthcoming). “Theory and contrastive explanation in ethnography”, Sociological Methods and Research. McCullagh, B. (1978). “Colligation and classification in history”, History and Theory. 17(3): 267–284. Merton, R.K. (1997). “De-gendering ‘Man of science’: the genesis and epicene character of the word scientist”, Ed. K. Erikson. Sociological visions. New haven: Yale University Press. Merton R.K. and Barber, E. (2006). The travels and adventures of serendipity: a study in sociological semantics and the sociology of science. Princeton: Princeton University Press. Mill, J.S. (1952). A system of logic: ratiocinative and inductive. New York: Longmans, Green and Co. oed (Oxford English Dictionary). (2014). “Colligation”. Downloaded 25.11.2014 from: http://www.oed.com.proxy.library.cornell.edu/view/Entry/36351?redirectedFrom= colligation#ei. Peirce, C.S. (1935). Collected papers of Charles Sanders Peirce. Vol. 6. Cambridge, ma.: The Belknap Press. Peirce, C.S. (1998). “Whewell”, Writings of Charles S. Peirce. Vol. 2. Bloomington: Indiana University Press. Pompa, L. (2004). “Walsh, William Henry (1913–1986)”, Oxford Dictionary of National Biography. Oxford: Oxford University Press. Downloaded 03.12.2014 from http://www .oxforddnb.com/view/article/65673. Schutz, A. (1954). “Concept and theory formation in the social sciences”, Journal of Philosophy. 51(9): 257–273. Snyder, L. (2006). Reforming philosophy: a Victorian debate on science and society. Princeton: Princeton University Press. Snyder, L. (2011). The philosophical breakfast club: four remarkable friends who transformed science and changed the world. New York: Broadway Paperback. Snyder, L. (2012). “William Whewell”, Stanford Encyclopedia of Philosophy. Downloaded 25.11.2014 from: http://plato.stanford.edu/entries/whewell/. Somers, M. (1995). “What’s political or cultural about political culture and the public sphere? Toward an historical sociology of concept formation”, Sociological Theory. 13(2): 113–144. Spillman, L. (2004). “Causal reasoning, historical logic, and sociological explanation”, Eds. J. Alexander, G. Marx and C. Williams. Self, social structure, and beliefs. Berkeley: University of California Press. Spillman, L. (2014). “Message from the chair: thick description in causal claims”, Culture: newsletter of asa’s section on the sociology of culture. 22(3): 1–3, 5.

78

Swedberg

Strong, E.W. (1955). “William Whewell and John Stuart Mill: their controversy about scientific knowledge”, Journal of the History of Ideas. 16(2): 209–231. Thompson, D. (1967). “Colligation and history teaching”, Eds. W.H. Burston and D. Thompson. Studies in the nature and teaching of history. London: Routledge and Kegan Paul. Wagner-Pacifici, R. (2010). “Theorizing the restlessness of events”, American Journal of Sociology. 115(5): 1351–1386. Walsh, W.H. (1942). “The intelligibility of history”, Philosophy. 17(66): 128–143. Walsh, W.H. (1967). Philosophy of history: an introduction. New York: Harper Torchbooks. Walsh, W.H. (1974). “Colligatory concepts in history”, Ed. P. Gardiner. The Philosophy of history. Oxford: Oxford University Press. Whewell, W. (1835). “Letter to Jones dated May 2”, Ed. I. Todhunter. William Whewell, D.D. Vol. 2. London: Macmillan and Company. Whewell, W. (1836). An elementary treatise on mechanics. Cambridge: Deighton. Whewell, W. (1840). History of the inductive sciences. Two volumes. London: John W. Parker and Son. Whewell, W. (1847). History of the inductive sciences. Vol. 2. A new edition. London: John W. Parker and Son. Whewell, W. (1858). Novum Organon Renovatum. London: John W. Parker and Son. Whewell, W. (1860). On the philosophy of discovery, chapters historical and philosophical. London: John W. Parker and Son. Zuckerman, H. (2010). “On sociological semantics as an evolving research program”, Ed. C. Calhoun. Robert K. Merton: sociology of science and sociology as science. New York: Columbia University Press.

chapter 5

Sensitizing Concepts in Action: Expanding the Framework Anne Britt Flemmen Introduction The relationship between data and social reality is considered a common methodological problem in social research independent of research strategy (Blaikie, 2000: 120). Most researchers recognize that data are produced by the activities of social researchers on a certain version of social reality. Nevertheless, there is little consensus among scholars regarding what the solution to this challenge should be. Some scholars argue for placing greater weight on lay terms to develop a deeper understanding of laypeople. For others, this appears naïve. They argue that researchers must and should interpret the world according to theories and concepts. Certain scholars argue that because social reality has no independent existence apart from the knowledge held by the social actors, researchers’ way of gaining access to this knowledge is to enmesh themselves as deeply as possible in the everyday world of the actors. Thus, researchers can access laypeople’s understandings and interpretation. Others find this view to be overly o­ ptimistic and hold a different view of the relationship between language and reality. For still others, the task of sociological theory is to remain critical and the researcher’s role is to unveil hidden forms of oppression and power. Nevertheless, despite all of the scholarly debate regarding observation as theory dependent, the following dilemma remains: how do empirical researchers obtain the right balance? Using intersectionality as an example, I will argue for understanding sensitizing concepts as a methodology for striking this balance. Since the 1990s, gender, queer and postcolonial studies have given intersectionality many roles—so many, in fact, that it has been called feminist theory’s buzzword. As a fuzzy and contested concept, intersectionality is well suited for discussing the role of theories and concepts in social research. To approach research through a sensitizing concept is sociology’s tool to facilitate a position in which one can perceive something new. It allows empirical researchers to start with a concept to provide direction and to develop theoretical/analytical tools in close dialogue with data. Despite this potential, © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_006

80

Flemmen

I will argue for the need to expand sensitizing concepts beyond the interactionist framework. Intersectionality Gender, queer and postcolonial studies have all taken an interest in the interaction among different dimensions of inequality (Winkler and Degele, 2011). The term intersectionality was coined both to achieve this task and to emphasize the interwoven nature of categories such as gender, race/ethnicity/nationality, class and sexuality and how they mutually strengthen or weaken each other. Feminists hold different views of whether intersectionality is a theory (de los Reyes and Mulinari, 2005), a framework (McCall, 2005), a politics (Crenshaw, 1989, 1991; Carbin and Edenheim, 2013) or a discourse (McKibbin, Duncan, Hamilton, Humphreys and Kellett, 2015). The dearth of empirical studies and relative absence of discussion regarding the methodologies used to study intersectionality led a group of Norwegian researchers to explore the possibility of conducting qualitative empirical intersectional work (Berg et at., 2010). In that book, we examined the debates regarding (ethnic) diversity (‘mangfold’ in Norwegian) and (gender) equality (‘likestilling’ in Norwegian) by analysing how (gender) equality and Norwegianness are involved in different constellations of majority- and minoritymaking processes.1 Through different qualitative studies, we explored how processes that create gender intersected with processes that create ethnicity/ race, sexuality, class and nationality in Norway. In scrutinizing these processes, we found it important not to assume a preestablished hierarchy of differences. We were preoccupied with exploring how intersectionality—in the form of intersecting dimensions of d­ ifference— works and how it creates hierarchies that crystallizes certain groups as minorities. We understood gender, ethnicity, race, sexuality and nationality as something practiced or done. To us, it was important to avoid an ontological ranking of differences, that is, defining a priori which forms of differences are most relevant or most important (gender, class, ethnicity, race, sexuality etc.), instead allowing the empirical investigations to provide the answer. Western feminism has tended to privilege white heterosexual women’s experiences. This privilege operates such that the experiences of this group 1 We developed our methodological position in the introductory chapter of the edited book (Berg, Flemmen and Gullikstad, 2010). That chapter and the chapter by Kramvig and Flemmen (2010) provide the basis for my argument in this paper.

Sensitizing Concepts in Action: Expanding the Framework

81

of women appear general and set the standard for normality. Judged by this standard of ‘woman’, all other women’s challenges and problems become special interests that are not defined as questions of gender or as relevant to the women’s movement. The theoretical and political limitations of this gender perspective were scrutinized within feminist theory, and intersectionality surfaced as one possible way to problematize it. Whereas Davies (2008) celebrates the role of intersectionality as a buzzword in feminist theory, Carbin and Edenheim (2013) criticize the tendency to expand the scope of intersectionality to an all-encompassing feminist theory. In their view, intersectionality has been so successful within different feminist approaches precisely because it does not meet the requirements of a ­theory. It  does not provide us with an ontology either of the subject or of power. ­Statements such as ‘social positions intersect’, ‘categories are intertwined’ or ‘power relations are complex’ are so general that everybody can embrace them. Thus, they disguise the ontological differences between different feminist approaches, according to Carbin and Edenheim. From their poststructuralist ­position, Carbin and Edenheim are concerned that embracing intersectionality as a common feminist concept will reduce feminism to a liberal consensusbased project (for similar criticism, see Carbin and Tornhill, 2004). Debates about the fruitfulness of the concept persist, as do the different ­approaches to it. For the purpose of this paper, I will consider these differences as disagreements with regard to research tradition and to the role of concepts in research.

Use of Concepts within Different Research Traditions

To contextualize the argument, I will outline how four different research traditions view the role of concepts using feminist theory and intersectionality as an example. Crenshaw (1989, 1991), the American law professor who coined the term ‘intersectionality’, described intersectionality using supporting concepts such as ‘axis’, ‘lines’ and ‘systems’. These axes were layered over each other to create forms of oppression that were double, triple or multiple. Black women are oppressed as women (by the patriarchy), as black (through racism and colonialism), etc. Crenshaw’s understanding of intersectionality can be viewed as part of an ontological research tradition. In this tradition, one is concerned with establishing the main features of social reality to create a theoretical synthesis with a strong ontological emphasis (Blaikie, 2000). Hill Collins (1998) developed this thinking further into a ‘matrix of domination’ in which gender and race are the main categories. She also linked this idea to an e­ pistemological

82

Flemmen

position, a feminist standpoint, in which certain positions based on gender, class and race are believed to provide access to a particular insight and knowledge of the world. Hill Collins correlates the ‘matrix of domination’ for various groups within larger social groupings with patterns of privilege and marginalization (Winkler and Degele, 2011). Crenshaw and Hill Collins have both been criticized for their ontological determination of the categories of woman, black woman and white woman. They introduce both categories and structures as disparate and pre-existing, independent of social and historical conditions. The categories thus come to be perceived as static and unchangeable, as additive and structural. Modelling income indicators and income differences among (fixed) social groups is an example of what is called an intercategorial analysis of intersectionality (McCall, 2005).2 This approach ‘makes a strategic use of categories and analyses relations of multiple inequalities between socially constructed groups’ (Winker and Degele, 2011: 53). Intersectionality understood as intercategorial complexity can be viewed as a work within the operationalizing research tradition. This tradition is concerned with specifying and measuring concepts to produce variables for particular research projects (Blaikie, 2000). This approach to intersectionality has been criticized both for not being sufficiently radical and for assuming and strengthening pre-existing differences (Egeland and Gressgård, 2007). Intersectionality can also be used hermeneutically. The hermeneutic research tradition is concerned with deriving technical concepts from lay concepts (Blaikie, 2000). Because intersectionality itself is not a lay concept, these types of studies are not common. There are, however, examples of theoretically oriented actor-network inspired ethnographic studies that attempt to avoid predefined categories such as gender, race, ethnicity, sexuality, etc., for as long as possible to allow social actors and the ethnographical context to actualize what is relevant to the situation (Kofoed, 2004). A fourth research tradition related to using concepts within the social sciences is sensitizing, where refining an initial flexible concept in the course of research is viewed as the optimal way to proceed (Blaikie, 2000). Our approach to intersectionality can be viewed as a work within this tradition (Berg et al., 2010). We argue that intersectionality is best approached as a concept that draws the researcher’s gaze towards ordering practices, that is, practices that produce phenomena and categories such as minority and majority, saturated 2 McCall categorizes the use of intersectionality into three approaches. The others are the anticategorial and the intracategorial (McCall, 2005). I will not discuss these other two approaches here.

Sensitizing Concepts in Action: Expanding the Framework

83

by gender, race and ethnicity. Intersectionality challenges singularity while demanding specification and actuality. To study intersections one must, in our view, explore how they are enacted and performed in specific times and places, how bodies are gendered, racialized and sexualized, made familiar or strange, comprehensible or incomprehensible. Intersectionality guides our attention in the direction of multiplicity and ambivalence, towards what is not stabilized or fixated (Berg et al., 2010). This approach to intersectionality has been criticized both as a ‘doing gender’ perspective and for not exploring contexts in which these differences are not done, or ‘the undoing’ of social differences, and hence for tending to reify the categories of difference. Orupabo (2014), inspired by Winkler and Degele (2011), advocates studying the meaning of intersectionality at different levels at which identities, representations and structures are not mutually exclusive categories. This is a critique from the ontological research tradition. In the following section, I argue that by expanding the sensitizing tradition we can open up new possibilities for theoretically inspired empirical studies. To accomplish this I will first revisit Blumer and the theoretical and methodological challenges that sensitizing concepts were developed to address. ­Second, I will show intersectionality as a sensitizing concept in action. Third, I will suggest how sensitizing concepts can be expanded to overcome certain shortcomings of interactionalism.

Sensitizing Concepts

The conceptual space from which sensitizing concepts arose was a discussion of social theorists’ preoccupation with the literature of social theory at the expense of connecting with the empirical social world (Blumer, 1969: 142). In an address to the American Sociological Society in 1953, Herbert Blumer (1969) asked, ‘What is wrong with social theory?’ His concern was threefold: first, that social theory neither in origin nor in use did appear oriented towards the empirical world; second, that social theory failed to guide research because it was divorced from empirical investigations; and third, that social theory failed to benefit from the growing accumulation of ‘facts’ issuing from research (­Blumer, 1969: 141–142). Blumer criticized researchers for relying upon a priori theoretical schemes and sets of unverified concepts. The prevailing practice at the time was to allow a theory, a model, a concept, a technique and the scientific protocol to coerce research. This resulted in a bending of analytical depictions of the empirical world to suit the form of theories (Blumer, 1969: 34).

84

Flemmen

In Blumer’s view, the difficulties in social theory were related to the fact that the concepts of our discipline in a fundamental sense are sensitizing instruments. He named them sensitizing concepts, contrasting them with definitive concepts. A definitive concept refers precisely to what is common to a class of objects by means of a clear definition in terms of attributes or fixed benchmarks … A sensitizing concept lacks such specification of attributes or benchmarks and consequently does not enable the user to move directly to the instance and its relevant content. Instead, it provides the user with a general sense of reference and guidance in approaching empirical instances. Whereas definitive concepts provide prescriptions of what to see, sensitizing concepts merely suggest directions along which to look. blumer, 1969: 147–148

What we are referring to with any given concept takes a different form in each empirical instance. When a concept is sensitizing it does not have a clear ­definition based on specific criteria or properties (see also Hammersley, 2003). ­Sensitizing concepts cannot be empirically delimited, and the point of their vagueness is that they can help us look in a specific direction without locking us into a certain understanding of the phenomenon (Sohlberg and Sohlberg, 2013: 136). In contrast to the natural sciences and their fixed semantic units, social science addresses (or should address) concepts in continuous change, which is more apt to provide direction than the iron fence created by definitive concepts (Alvesson and Sköldberg, 1994: 67). Sensitizing concepts in continuous change make it possible for the researcher to become successively acquainted with the subject of research, stimulating researchers to sense new relations, perspectives and worldviews. Therefore, sensitizing concepts may serve as heuristic devices. The main purpose of sensitizing concepts is to open the field up for exploration. The development of concepts is viewed as resulting from studying a ­subject and not as something that should structure our understanding of the  subject beforehand. To Blumer, sensitizing concepts emerge when the observer discovers something worth problematizing (Faulkner, undated), ­ which is similar to Peirce’s view on abduction. Blumer was very concerned with the importance of discussing the methodological role of concepts: Throughout the act of scientific inquiry concepts play a central role. They are significant elements in the prior scheme that the scholar has of the ­empirical world; they are likely to be the terms in which his [sic] problem

Sensitizing Concepts in Action: Expanding the Framework

85

is cast; they are usually the categories for which data are sought and in which the data are grouped; they usually become the chief means for establishing relations between the data; and they are usually the anchor points in interpretation of the findings. Because of such decisive role in scientific inquiry, concepts need especially to be subject to methodological scrutiny. blumer, 1969: 26

Opposed to the positivism that was popular at the time, Blumer was influenced by pragmatist philosophy in his methodological thinking. Pragmatists, particularly Dewey, argued that concepts are tools that should be judged according to their instrumental value in enabling people to understand and to act on the world. Theory, according to Blumer, is of value only in the empirical sciences to the extent that it connects fruitfully with the empirical world. The only means of establishing such a connection is through concepts, as they point to empirical instances. The methodological problem encountered by all social researchers is that they can collect empirical data only from a certain point of view. Observations are shaped and coloured by the researcher’s language, culture, and discipline. They are based on knowledge, past experience (professional and lay) and the expectations that result. Because all observation is theory dependent, there will always be a gap between the data and the reality they are supposed to represent (Blaikie, 2000: 120).

Intersectionality: A Sensitizing Concept in Action

Recognizing that researchers always have to collect empirical data from a particular point of view, we found sensitizing concepts to provide a good balance between a pre-established theoretical commitment and an overly open approach. The focus on intersectionality in our research is meant to make us sensitive, that is, attentive and responsive, to complex power relations without assuming an ontological hierarchy of differences. Like other social researchers, we assume that certain differences do exist—we have recruited our informants based upon, inter alia, imagined differences among gender, ethnic groups, classes or similar markers. Therefore, we have not started out in a manner that is completely open and inductive. To use a well-known phrase, we ‘started with an open, but not an empty mind’. We started with such categories, but then we opened them up. Therefore, we worked both with and against the categories. We explored which phenomena create difference and are at work in specific contexts.

86

Flemmen

In the study of multicultural and transnational families in Sápmi we explored the meaning of ethnicity with regard to how women and men do gender (Kramvig and Flemmen, 2010). Northern Norway is traditionally a multiethnic area where the indigenous Sami population, the Kvens (Finnish minority) and the Norwegian majority population have intermarried for centuries. The ethnic mix in the area is becoming more complex as a result of new international migration patterns. Immigrants from neighbouring Russia are visibly present in the local communities as are marriage migrants, labour migrants, asylum seekers and refugees from all over the world. Recognizing this complexity, the research needed to open up the categories of ethnicity and gender because it was impossible to decide theoretically beforehand how they were understood and which category was most important. A majority-inclusive perspective made it possible to include the majority (those who at any given time constitute the ‘we’ in a society) and the relationship between the majority and minorities in the analysis. Therefore, the social categories were not only a lens through which to understand the other, such as ethnic minorities or women; instead, certain people were perceived as ethnically minoritized, others as ethnically majoritized, some as ethnically marked and others as ethnically unmarked or neutral, some as privileged and others as less privileged (Kramvig and Flemmen, 2010). Our material implies that the stakes for the couples we met were n ­ egotiations regarding mutual respect (Kramvig and Flemmen, 2010). Individual dignity had to be sustained while a new union was established between the two. In couples from different ethnic backgrounds, many differences had to be seen and recognized. It was a part of everyday life and negotiations regarding which languages to use in the family, finances, and (ethnic and gender) equality in which dignity and equality were developed as key factors in the relationship between the women and the men. By focusing on the processes of negotiation, we came closer to ­understanding autonomy or what it means to be ‘in place’ (Ahmed, 2006). Whereas access to paid labour and gender equality was important to the Russian women, the Sami women underlined ethnic equality and the importance of reclaiming partially lost Sami knowledge and traditions in terms of their autonomy. The N ­ orwegian colonization of the Sami people, their language and traditions is still being repaired not only at the political level but also in peoples’ everyday lives.3 3 It is beyond the scope of this article to delve into the historical Norwegian politics of oppressing the Sami (and the Kven) population in the area, leaving generations in pain and for some a shame so strong that they have denied their ethnic heritage. As a result of this history, the ethnic negotiations in the area occur both within families and among different ethnic groups.

Sensitizing Concepts in Action: Expanding the Framework

87

Whereas equality (in Norwegian, ‘likestilling’) implicitly involves gender equality in the Norwegian public discourse and for the majority population, we found it to be linked to ethnicity by the indigenous Sami people with whom we spoke (Kramvig and Flemmen, 2010). To the Russian female marriage migrants and their husbands equality was linked to both gender and ethnicity but in a different way than for the Sami. Intersectionality thus worked as a sensitizing concept, as a concept not ­clearly defined and delimited, which made it possible to analyse the ­complexity of lived experience. Categories are continuously created, performed, quoted, reproduced and transgressed in the daily interaction between actors (Berg et al., 2010), and to us, it was important to empirically explore how this occurs. We concluded that intersectionality, as a methodological principle and not a theory, can • Help us see something that otherwise would be difficult to see (this specific complexity) • Point to a necessary work of destabilization and change • Draw our gaze towards ordering practices or what is often taken for granted • Make us pay attention to how intersections are done • Clarify how phenomena and categories such as minorities and majorities are produced (Berg et al., 2010). Sensitizing concepts are open and empirically grounded in a way that definitive concepts and abstract theories are not. To me, the value of the relatively awkward (in Norwegian) concept of intersectionality can be found here, in its sensitizing function. It provides the opportunity to discuss what this concept does, not what it is. When processes, relationships, patterns or themes are made into theories, they become more fixed and tend to lose their sensitizing function and therefore the openness to what surfaces as important empirically. In the final section of this paper, I will suggest three ways of expanding the interactionist framework to allow sensitizing concepts to perform their abovementioned bridging function.

Expanding Blumer and the Interactionist Framework

Approaching sensitizing concepts as a methodology allows us to maintain a theoretical openness that is both useful and necessary. This openness, however, is not sufficient. It is necessary to expand and refine Blumer’s original

88

Flemmen

reflections outside the interactionist framework that initially surrounded it. Three areas require expanding: • Introducing the possibility of theoretically inspired concepts (and thus recognizing the comparative element of research) • Showing the relationship with other concepts and theories (such as assisting the development of an integrated network of concepts that fits the empirical case we are studying) • Sensitizing the researcher by recognizing the situatedness of the researcher but also the limits to this critical position. Below, I will address each of these three areas. Theoretically Inspired Concepts as Comparative Element According to Bulmer (1979), Blumer’s division between sensitizing and definitive concepts is inadequate. Concepts are not theories in themselves, Bulmer notes; they are categories for organizing ideas and observations. To form a theory, concepts must be interrelated. Concepts mediate between theory and data; they form a bridge between them. Bulmer points to the much debated, inescapable and difficult paradox of categorization: where do concepts originate from in the first place, and what are the justifications for the use of particular concepts? Bulmer’s argument is that concept formation in the analysis of sociological data proceeds neither from observation to category nor from category to observation but in both directions simultaneously and in interaction (Bulmer, 1979: 652–653). Concepts are formed and modified both in the light of empirical evidence and in the context of theory. In contrast to his understanding of Blumer, Bulmer is critical of the view of quantitative social research concepts as pre-formed and fixed, while in qualitative analysis they tend to be fluid and emergent. To Bulmer, concepts such as social structure, personality and culture rest upon a general sense of what is relevant rather than on clear-cut prior specification. Although Bulmer agrees with Blumer’s view that theory, enquiry and empirical data are closely interwoven and that their interplay is the means by which an empirical science develops, he disagrees with how Blumer depicts the nature of concepts, particularly definitive concepts (Bulmer, 1979: 655). Bulmer questions the distinction between a theoretical language and an observational language. In his view, differences in the nature of concepts are not ontological. ‘The meaning of a theoretical term is fixed by horizontal as well as vertical members of the conceptual structure, and only the structure as a whole, at best, rests firmly on empirical ground’ (Bulmer, 1979: 658).

Sensitizing Concepts in Action: Expanding the Framework

89

This variety is not recognized in Blumer’s picture of scientific concepts. To Bulmer, concepts are not simply developed out of observations, but neither are they imposed a priori categories. Their use is somewhat justified in terms of their context in a particular theory and particular observations that the theory seeks to explain (Bulmer, 1979: 659). Thus, theories are not developed either deductively or inductively but instead through a process that Bulmer calls retroduction. ‘Theories put phenomena into systems. They are built “in reverse”—retroductively. A theory is a cluster of conclusions in search of a premise’ (Bulmer, 1979: 660). I would have called the process described by Bulmer abduction, but that is of no consequence to my argument. Finally, Bulmer criticizes Blumer for providing little guidance about how to develop sensitizing concepts in practice. Bulmer describes the routes to the formation of concepts in qualitative research as ‘empirically variable, according to the general sociological orientation of the researcher, the richness of the existing literature, and the nature of the phenomenon being studied’ (Bulmer, 1979: 671). In a broad sense, our approach concurs with Bulmer’s reflections. The ­concept of intersectionality was a theoretically inspired concept that we chose to bring with us into our study, not a lay concept introduced by the study’s informants. A research process always involves comparing the study that is underway with other studies. This must be recognized and made explicit by the sensitizing research tradition. Choosing to use the intersectionality concept was thus the comparative aspect of our study. We observed that intersectionality had been found appropriate in other research to explore the (gendered, ethnic, minoritized, majoritized) complexity we were studying. It allowed us to keep open which differences were most important to people in a particular place and time. Sensitizing concepts thus take inspiration from what has been found appropriate in other studies and are furthermore amplified and deepened through comparative use (Addelson, 1994: 171). As a sensitizing concept, intersectionality thus provided a link between the particular and the general. A concept cannot, however, stand alone. Instead, it must be integrated into a network of concepts or a theory. This is the second area in which the interactionist framework must be expanded. Developing an Integrated Network of Concepts The point of sensitizing concepts is not the confirmation or falsification of theories. Sensitizing concepts are suited to assisting the development of an integrated network of concepts that fits the empirical case we are studying (Addelson, 1994: 171).

90

Flemmen

In our study, we did not develop intersectionality into a theory. We used intersectionality as a methodology that allowed us to combine it with the relevant network of concepts to grasp what was occurring in the empirical material. Framed by our research question on how transnational and multicultural families contended with diversity, similarity and equality, we found the concept of gift exchange useful (Kramvig and Flemmen, 2010). Gift exchange provided an opportunity to understand the challenges people encountered in grappling with how to address their dependency on others, while s­ imultaneously demarcating the boundaries between themselves and others. The challenge in these somewhat contradictory processes was to find the balance between sustaining individual autonomy, one’s own identity, while also needing others. Differences can be transcended through the gift. We suggested that the gift is an institution than can be used to manage difference, complex situations and complex power relationships. In addition to gift exchange, we found Leigh Star’s concept of power useful. Power is defined as a question of whose metaphors are bringing the world together and are capable of keeping it together (Leigh Star, 2001: 51). Allowing ethnicity to be an open concept meant it was non-exclusionary, that is, not a question of either or, as either Sami or Kven/Norwegian/Russian, but a question of combination, both Sami and Kven/Norwegian/Russian. Thus, the important identity processes occurring in the area could be captured and recognized (Kramvig and Flemmen, 2010). Our expansion of sensitizing concepts is similar to Layder’s (1998: 115–116) orienting concepts: we do not think of sensitizing concepts as standing alone but instead view them as directive of theoretical inquiry through networks of concepts. Although I am in agreement with Layder with regard to much of his critique of the interactionist framework, I suggest retaining the term sensitizing concept, as it is an established concept within sociology and the social sciences. The Sensitized Researcher A lack of reflection on the role of researcher and the relationship between researchers and the people they study is a problem with the interactionist tradition. To expand our approach, we must address two topics: the implications of research being a collective activity (Addelson, 1994) and the role of critique (Mol, 2005; Boltanski and Thevenault, 1999). One limit of the interactionist approach is that the sensitizing concept does not consider the fact that research is a collective act (Addelson, 1994: 173). By collective act, Addelson refers to the relationship between the researcher and the people or objects of study:

Sensitizing Concepts in Action: Expanding the Framework

91

For the interactionist approach to work, the researcher must be s­ ensitized, but not only to concepts that capture the lay activity per se. Instead, the researcher needs sensitizing concepts to capture the relationship of his or her own research activity and the lay activity. The researcher needs sensitizing concepts that captured the collective activity of the researcher’s profession in relation to the vast collectivities of laypeople. In their studies, researchers need to include themselves and the collective intrusion and authority of their professions. They do offer concepts and research useful to understanding the importance of the activities of professionals, but these are for participant-observation and are not reflexive. addelson, 1994: 173–174

Interactionists therefore rely on their own professional authority and place in the social order, she concludes. To address this shortcoming in the framework, Addelson introduces the notion of ‘the sensitized researcher’. She does this to take into account the researchers’ double participation: their ­intellectual and professional work and their moral and political work; that is, she accounts for the living, sensitized, embodied being of the researcher (Addelson, 1994: 173). It is difficult to participate in politicized conversations from a position as researcher. Because topics of gender equality and ethnic diversity are controversial and high on the public agenda, this is important to understand. Our understanding of the researcher is not that of a judge who decides whether peoples’ practices are good or bad (Berg et al., 2010). Inspired by Mol (2005), as researchers we were preoccupied with investigating people’s choices and what constitutes their practices. We seek the standards employed in our empirical materials, not to bring our standards to the field. This does not mean that we considered ourselves neutral. The researcher is positioned academically through the theoretical and methodological choices made and the analysis performed. A second and related topic that must be addressed is the role of criticism and critical theory. In our approach, we wanted to position ourselves outside the traditional framework in which critical research meant either assuming a negative position or locating the only possibility for critique either in the theoretical apparatus or in the researcher. Our approach has affinities with ­Boltanski and Thévenot’s reflections: The main problem for critical sociology is its inability to understand the critical operations undertaken by the actors. A sociology that wants to study the critical operations performed by actors—a sociology of ­criticism taken as a specific object—must therefore abandon (if only

92

Flemmen

temporarily) the critical stance to recognize the normative principles that underlie the critical activity of ordinary persons. If we want to take seriously the claims of actors when they denounce social injustice, criticize power relationships or unveil their foe’s hidden motives, we must conceive of them as endowed with an ability to differentiate legitimate and illegitimate ways of rendering criticism and justifications. boltanski and thévenot, 1999: 364

Conclusion In this chapter, I have argued that sensitizing concepts are sociology’s tool for being theoretically inspired while remaining analytically open. Sensitizing concepts enable an empirically oriented approach to theorizing without denying that observations are theory dependent. By using the contested concept of intersectionality as an example, I have argued for a methodological approach that expands the interactionist framework from which sensitizing concepts emerged by addressing some of its limitations. Three areas must be incorporated. First, we must recognize the comparative element of research by explicitly stating the possibility for theoretically inspired concepts. Second, we must assist the development of an integrated network of concepts that fits the specific empirical case by showing the relationship between the sensitizing concept and other concepts and theories. Third, we must recognize the situatedness of research and the researcher by sensitizing the researcher. References Addelson, K.P. (1994). Moral passages. Toward a collectivist moral theory. New York: Routledge. Ahmed, S. (2006). Queer phenomenology; orientations, objects, others. Durham: Duke University Press. doi: http://dx.doi.org/10.1215/9780822388074. Alvesson, M. and Sköldberg, K. (1994). Tolkning och reflektion. Vetenskapsfilosofi och kvalitativ metod. [Interpretation and reflection. Philosophy of science and qualitative method]. Lund: Studentlitteratur. Berg, A.-J., Flemmen, A.B. and Gullikstad, B. (2010). “Innledning: Interseksjonalitet, flertydighet og metodologiske utfordringer”, Eds. A.-J. Berg, A.B. Flemmen and B. Gullikstad. Likestilte norskheter. Om kjønn og etnisitet. [“Introduction: Intersectionality, ambiguousness and methodological challenges”, Equalized Norwegianness. On gender and ethnicity]. Trondheim: Tapir Akademisk forlag.

Sensitizing Concepts in Action: Expanding the Framework

93

Blaikie, N. (2000). Designing social research. Cambridge: Polity Press. Blumer, H. (1969). Symbolic interactionism. Perspectives and methods. New Jersey: ­Prentice-Hall Inc. Boltanski, L. and Thévenot, L. (1999). “The sociology of critical capacity”, European Journal of Social Theory. 2(3): 359–377. doi: http://dx.doi.org/10.1177/1350506810386084. Bulmer, M. (1979). “Concepts in the analysis of qualitative data”, Sociological Review. 27(4): 651–677. Carbin, M. and Edenheim, S. (2013). “The intersectional turn in feminist theory: a dream of a common language?”, European Journal of Women’s Studies. 20(3): 233–248. Carbin, M. and Tornhill, S.S. (2004). “Intersektionalitet—ett oanvändbart begrepp?”, Kvinnovetenskaplig Tidskrift. [“Intersectionality—an inapplicable concept?”, Journal of Women’s Science]. 25(3): 111–114. doi: http://dx.doi.org/10.1177/1350506813484723. Crenshaw, K.W. (1989). Demarginalizing the intersection of race and sex: a black feminist critique of antidiscrimination doctrine, feminist theory and anti-racist politics. Chicago: University of Chicago Legal Forum. Crenshaw, K.W. (1991). “Mapping the margins: intersectionality, identity politics, and violence against women of color”, Stanford Law Review. 43(6): 1241–1299. Davies, K. (2008). “Intersectionality as a buzzword: a sociology of science perspective on what makes a feminist theory successful”, Feminist Theory. 9(1): 67–86. doi: http://dx.doi.org/10.1007/978-3-531-92555-4_3. de los Reyes, P. and Mulinari, D. (2005). Intersektionalitet. Kritiska reflektioner över (o) jämnlikhetens landskap. [Intersectionality. Critical reflections on the landscape of (in)equality]. Malmö: Liber. Egeland, C. and Gressgård, R. (2007). “The ‘will to empower’: managing complexity of the cthers”, NORA. Nordic Journal of Feminist and Gender Research. 15(4): 207–219. doi: http://dx.doi.org/10.1080/08038740701646689. Hammersley, M. (2003). “Sensitizing concept”, The SAGE encyclopedia of social science research methods. Vol. 3. London: Sage Publications. Hill Collins, P. (1998). “It’s all in the family: intersections of gender, race, and nation”, Hypatia. 13(3): 62–82. doi: http://dx.doi.org/10.2979/hyp.1998.13.3.62. Kofoed, J. (2004). Elevpli: Inklusion- og eksklusionsprocesser blandt børn i skolen. [“Elevpli: Processes of inclusion and exclusion among children in the school”]. (PhDDiss., Danmarks Pædagogiske Universitet). Kramvig, B. and Flemmen, A.B. (2010). “Mangfold, likhet og likestilling i Sápmi”, Eds. A.-J. Berg, A.B. Flemmen and B. Gullikstad. Likestilte norskheter. Om kjønn og etnisitet. [“Multiplicity, similarity and equality in Sápmi”, Equalized Norwegianness. On gender and ethnicity]. Trondheim: Tapir Akademisk forlag. Layder, D. (1998). Sociological practice. Linking theory and social research. London: Sage Publications. doi: http://dx.doi.org/10.4135/9781849209946. Leigh Star, S. (2001). “Makt, teknologier og konvensjonenes fenomenologi. Om å være allegisk mot løk”, Eds. K. Asdal, B. Brenna and I. Moser. Teknovitenskapelige kulturer.

94

Flemmen

[“Power, technologies and conventions phenomenology. About being allergic to onions”, Techno scientific culture]. Oslo: Spartacus. McCall, L. (2005). “The complexity of intersectionality”, Signs. Journal of Women in Culture and Society. 13(3): 245–258. doi: http://dx.doi.org/10.1086/426800. McKibbin, G., Duncan, R., Hamilton, B., Humphreys, C. and Kellett, C. (2015). “The intersectional turn in feminist theory: a response to Carbin and E ­ denheim”, European Journal of Women’s Studies. 22(1): 99–103. doi: http://dx.doi.org/10.1177/ 1350506814539445. Mol, A. (2005). The body multiple: ontology in medical practice. Durham: Duke University Press. doi: http://dx.doi.org/10.1086/426800. Orupabo, J. (2014). “Interseksjonalitet i praksis: utfordringer med å anvende et interseksjonalitetsperspektiv i empirisk forskning”, Sosiologisk tidsskrift. [“Intersectionality in practice: challenges in employing a intersectionality perspective in empirical research”, Journal of Sociology]. 22(4): 329–352. Sohlberg, P. and Sohlberg, B.-M. (2013). Kunskapens former. Vetenskapsteori och forskningsmetod. [Forms of knowledge. Philosophy of science and research methods]. Stockholm: Liber forlag. Winkler, G. and Degele, N. (2011). “Intersectionality as multi-level analysis: dealing with social inequality”, European Journal of Women’s Studies. 18(1): 51–66. doi: http:// dx.doi.org/10.1177/1350506810386084.

part 2 Culture, Nature and Consumption



chapter 6

Culture as a Sociological Concept Willy Martinussen Introduction ‘Culture’ is among the words from everyday language most frequently used in sociology. The reason for this—to quote David Inglis and John Hughson (2003)—is at least twofold: culture is important, and culture is everywhere. Consequently, sociology requires theories of culture. The authors emphasize both that these theories have to be informed by empirical data and that issues of social structure and social action ought to be addressed when culture is studied. Cultural phenomena are both important objects of study and important sources of explanation and understanding of human actions, interactions, relations and structures. Consequently, sociological theories of culture are an obvious meeting place for the vocabulary of action and the vocabulary of system presented in the introductory chapter of this book. Almost all human beings live in the cultures of small groups, local communities, organizations, nations, world regions, and global societies. Thus, culture is the basic universe of meaning common to small or larger collectivities. The purpose of this chapter is to present and discuss the concept of culture in sociology with the help of illustrations from our discipline’s history. After a cursory delimitation of the concept of culture in sociology and elsewhere, and a discussion of its relations to other core concepts in sociology, this chapter will concentrate on two main themes. First, I will illustrate how cultures or specific elements of cultures have been diagnosed and described from the beginning of modern sociology to the present. Second, I will present a variety of examples from research on cultural forces or the impact of culture on people’s actions and interactions and in turn on social structures of various types. My examples illustrating the array of theories in this area are primarily gleaned from classical or other well-known studies in the past. I was surprised to read Jeffrey Alexander’s (2003) account of the lunchtime reactions of his sceptical colleagues when he introduced culture as an ­autonomous perspective and a social force in its own right. Had they not, as I had, heard of the integrative effects of the rain dance rituals on their own continent? Had they not read Max Weber’s account of the Protestant ethic and the spirit of capitalism? Were they not familiar with the reports from Mexico © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_007

98

Martinussen

City regarding the reproduction of class divisions caused by a culture of poverty among the least well off (Lewis, 1959)? On a more concrete level, did one of my sociology teachers not publish a book almost 40 years ago based on a thorough empirical study of how the culture of the workers’ collectives in industrial firms contained norms regarding a fair day’s amount of work that effectively fixed the production volume (Lysgård, 1967)? Had not one of my colleagues from that era published a thesis based on his investigations of a large mental hospital, finding that many of the wardens’ and doctors’ thoughts and actions could be explained by a distinctive ‘diagnostic culture’ in a neo-­functionalist way that would have pleased Alexander (Løchen, 1965)? Had not two of my colleagues interpreted and explained how Norwegian voters behaved differently from the Danes and Swedes because of the existence of particular ‘countercultures’ in Norway (Rokkan and Valen, 1969)? Surely there were numerous ­studies of similar phenomena known to the Los Angeles sociologists? What makes the ‘new cultural sociology’ new is partly Alexander’s ‘strong program’, with its emphasis on cultural autonomy, its thick descriptions of codes, narratives and symbols, and its detailed specifications of how culture influences real life interactions and developments (Alexander, 2003: 13–14). In addition, the new cultural sociology consists of new theoretical perspectives, particularly theories of cultural repertoires and what is called the French pragmatic sociology (Larsen, 2013: 10). What is common to them is the goal of analysing the meaning of cultural objects and their consequences for social life. Interpretive sociology and naturalistic methods are important ingredients of this endeavour. The new cultural sociology partly grew out of the cultural turn in the humanities and the social sciences. One frequently cited work is Laura Desfor Edles’ (2002) book Cultural sociology in practice. The questions raised in Part i of her book could well have been posed in mainstream sociology. Part ii of her book, which focuses on cultural methodology, discusses the utility of naturalistic inquiry (ethnography, ethnomethodology, dramaturgical research), discourse analysis and frame analysis. In light of the research questions presented, this focus appears reasonable. When examining the empirical studies defined as belonging to the new cultural sociology, however, the general questions appear largely to be of a type that has been posed by sociologists as long as our discipline has existed.

The Concept of Culture in Sociology

What, then, is a universe of meaning? In the field of sociology the expression a universe of meaning is the knowledge, ideas, beliefs, values, norms and

Culture as a Sociological Concept

99

f­eelings shared by a social group, a collectivity or a society. These common ideas and values largely constitute the members’ emotions and bases for actions and simultaneously justify and make sense of those actions. They are typically ­integrated or patterned, at least to a certain degree. They are attached to symbols, artefacts and technologies of various sorts. Language is perhaps the most important of these systems of symbols. Culture as a system of meanings and symbols is transmitted among people via different means of communications, which new members of groups and collectivities learn through socialization and imitation. Cultural ideas and ­values are constantly changing as part of the ongoing interaction between people. The culture of a particular social system is thus simultaneously a social product and part of the social inheritance. Subcultures exist, and many social systems may contain cultural conflicts of varying strength. The sociological concept of culture is apparently different from two equally important concepts indicated by the word ‘culture’. One of these is the artistic or aesthetic concept. Some people use the word culture with respect to what might be considered ‘high culture’, or what is regarded as the best of literature, painting, music, theatre, etc. created in a given civilization. Others include popular music, sports and other types of entertainment. In Western polities, this part of culture is often spoken of as ‘the culture sector’. Sociologists agree that this sector is part of the field of the sociology of culture. The second and broadest concept of culture might be called the archaeological or anthropological concept. In these disciplines, culture is typically regarded as the entire way of life of a group or collectivity. It often includes not only culture as defined by sociologists but also actual social relations. In sociology we often deem it fruitful to distinguish between four sets of social phenomena. They are delimited and described in different ways, depending on the theoretical or professional identities of the author. They may be indicated by the four expressions ‘relational structure’, ‘cultural structure’, ­‘interaction patterns’, and ‘life conditions’. The last type of phenomena ­mentioned is typically—in one version or other—the point of departure of a sociological analysis or investigation and covers a vast range of how individuals or groups of people live and act. When using the concept ‘relational structure’ we typically think of the horizontal and vertical pattern of relations between social positions, social groups, organizations, nation states, etc., as these relations are defined informally or formally in agreements, laws, tractates and so on. Included in this conception are the means of action and sanction belonging to the various social units. When using role theory, for instance, we examine how the positions of father, mother, son and daughter are constituted in different societies or parts of societies. In a nation state, the positions of president, member of parliament, and

100

Martinussen

voter, along with the power resources these positions command, are clearly delimited. The universes of meaning connected to the family positions and the political positions mentioned above are part of family culture and political culture, respectively. In role theory, cultural expectations of (for instance) the mother and the president are called their roles. There would be no social life if these positions were not filled by people who send and receive the relevant expectations. The concepts of cultural and relational structure are abstractions and to make our theories empirically useful, they must be complemented by an idea related to people acting and interacting. The concept of ‘interaction patterns’ may be helpful in this respect. It denotes the role behaviour and systems of interaction—friendly, indifferent or hostile—that may be observed in and between social groups, organizations, societies, and other social units. Viewed as a totality, these three types of structures are often considered ‘social’ or ‘societal’ structures. This accords with the argument made by John Scott in his chapter on social structure in this book. The fact that ‘relational structure’ and ‘cultural structure’ are abstractions does not mean that the elements of which they consist do not exist. Their parts and connections may be observed and documented by all types of observations and questionings, including all sociological research methods. Relations and cultures are consequential because people act and interact—consciously or unconsciously—via their use of them. Certain sociologists would add a fourth type of structure, the ‘material structure’, to those indicated above because physical surroundings (nature, ecological factors, buildings, means of communication, etc.) and distribution of wealth and other resources are closely connected to the structure of social positions and mental capabilities. Material factors, including nature, are certainly important in several sociological investigations. In the following pages I intend to discuss both how culture develops and changes along with changes in relational structures and interactions and how culture influences people’s interactions and relations and thereby their life conditions and the structures around them. The first type of analysis is often referred to as ‘the sociology of culture’, whereas the second type of analysis is sometimes referred to as ‘cultural sociology’.

Cultural Structure and Change

As mentioned above, some type of culture is part of every social system, from the small group to world society. Consequently, the sociology of culture c­ overs

Culture as a Sociological Concept

101

a vast segment of sociological thinking and research. It has been with us since the beginnings of modern sociology and is constantly developing. A few examples may indicate its scope and variety. I propose that the sociology of culture consists of three main types of studies: historical-sociological accounts of cultural development, diagnoses of contemporary cultural situations, and studies of the cultural elements of social systems or parts thereof. Because the classical studies and most of the contemporary diagnoses are well known, I will only touch upon their contents before sketching the types of questions asked in some of the subfields of the sociology of culture. Accounts of Long-Term Cultural Processes Sociologists analyse social change by studying the interplay between cultural, relational, interactional and material structures. It is unsurprising that historians have covered much of the same field as those who we sociologists view as our scientific ancestors. However, there is insufficient room to discuss this interdisciplinary space here. Needless to say, the ‘fathers’ of modern sociology studied culture as a major part of their accounts of the social developments of their time. Their main ideas may be summarized by three terms: collective consciousness, ideology, and rationality. Emile Durkheim (1933) introduced the idea of a collective consciousness in smaller or larger communities. Such common worldviews comprise norms regarding preferred actions in various situations, beliefs and knowledge, and a sense of obligation to the community. Together these beliefs and values create feelings of solidarity that contribute to the continued existence of the group or society. Durkheim’s famous theory of social change exemplifies how changing material, relational and interactional structures promote cultural change. These ideas were further developed by several social scientists. Among the theorists working with the idea of culture as a common consciousness are Talcott Parsons and his followers (see, for instance, Parsons and Shils, 1951). However, Parsons’ own expositions on cultural change have not become classics. Scholars with a similar point of departure might include William F. Ogburn (1964) and Pitirim A. Sorokin (1959), along with the political sociologists Seymour M. Lipset and Stein Rokkan (1967). Lipset and Rokkan analysed ­democratization in extensive and long-term historical-sociological studies. One key concept in their work is ‘participatory culture’. Such general concepts of a more or less ­integrated collective consciousness in all types of social units dominated mainstream sociology for many years. Paired with the idea that for a large part of the population the existing culture is a false consciousness, Marxian sociologists regard culture as an ideology.

102

Martinussen

In industrial societies, this means that the values, norms and beliefs that maintain the bourgeoisie’s economic and political power are accepted by the majority of the working class (Marx and Engels, 1970). This idea has been further developed into several versions, leading to a variety of descriptions of ongoing social change or reproduction. One early thinker is Antonio Gramsci (1975), who coined the idea of cultural hegemony, i.e., that the values and beliefs that legitimate class society become self-evident for entire populations as a result of the discourses in state apparatuses, legal systems, schools, churches, mass media, etc. Karl Mannheim published several works around 1930 that were to become his famous book Ideology and utopia (1960), our first introduction to the sociology of knowledge. The so-called Frankfurt school took such ideas further in their analyses of the development of mass society and the cultural industry. Their critical theory is used in the major work Dialektik der Aufklärung (Horkheimer and Adorno, 1947), a history of Western civilization from the ancient Greeks to the modern entertainment industry. Another work highlighting the interrelations between power and culture is Norbert Elias’ (1987) analysis of the history of manners, published in his two books on Western civilization (and sometimes decivilization) processes. Whereas the ideas of collective consciousness and hegemonic culture situate the causes of cultural change in material and relational structures, Max Weber and his followers view cultural processes as part of what occurs in social interaction. Values and beliefs are constantly produced and changed both in everyday social life and in the performances that are part of it. Weber himself did not elaborate on this general idea, but several of his followers did, including Erving Goffman, Peter L. Berger, Thomas Luckmann and Fredrik Barth. That said, Max Weber did contribute two investigations of cultural change. One is the famous analysis of the background of and close connections between the developing Protestant ethic and the spirit of capitalism (Weber, 1958). This is an example of the relationship between two systems of ideas and the long-term economic consequences of that relationship. Accordingly, it also exemplifies Weber’s notion of the method for arriving at causality in the social sciences: the interpretive understanding of social action. His other contribution is a description of the increasing rationalization and demystification of social life in the (Western) world. As Dennis H. Wrong (1970) summarized it, Weber’s idea of rationalization refers to the process whereby explicit, abstract, and intellectually calculable rules and procedures increasingly replace feelings, traditions and rules of thumbs in all areas of life. It replaces religion with science, cultivated persons with experts, craftspeople with machines, and traditional wisdom with abstract rules of law.

Culture as a Sociological Concept

103

Diagnoses of Contemporary Cultures Sociologists diagnose contemporary social conditions by describing cultural traits as simultaneous causes and consequences of socio-structural trends. ­Numerous theorists have ventured their diagnoses of what characterizes ­present-day culture in the modern world. Among them we find those who attempt to provide a total picture of social life today and others who describe the development of particular aspects of culture in our time. A typical example in the first group is Anthony Giddens (1991, 2006) and his theories of modernity, individualism and globalization. The most noteworthy theories in the second group can be classified under the headings of rationalization, individualization, and democratization. Parts of Weber’s ideas on rationalization are conveyed in our own time by Richard Sennett (1998), who has described the consequences of the decreasing routinization and increasing flexibility of modern work life. This new phase of rationalization requires workers with a high level of education who have learnt useful skills and attitudes to solve a broad range of tasks at their workplace or to be prepared to change jobs continually. This flexibility hinders the development of workers’ collectives, stimulating individualization. We find similar thoughts in psychologist David McClelland’s (1976) description of the development of the achievement motive. The individualization process may be viewed as a change in the collective consciousness, perhaps as another step in the development from mechanic to organic solidarity. The idea is that values, beliefs, and actions are increasingly based on individual choices and become less dependent on traditions and old social institutions. Part of this process is the gradual secularization of the West, with its recurring periods of religious revitalization. The individual has become more important. One expression of this development is the un Declaration of Human Rights. Other signs of this process include the increasing emphasis on self-realization and narcissism, the worship of heroes and stars of various types, and the increasing worship of the human body. One example of a discussion of some of the questions and findings in this field is Robert N. Bellah et al.’s (1985) books, which has the revealing title Habits of the heart. Individualism and commitment in American life. Anthony Giddens of course is among the sociologists who have characterized this development’s underlying forces and its contents. He summarizes the forces underlying ongoing cultural globalization in five points: the spread of television (particularly programmes from the Western world), the growth of a global economy with markets and economic organizations operating across continents, a growing body of ‘global citizens’ (business leaders and people in the entertainment industry), the steady growth of international ­organizations,

104

Martinussen

and the appearance of electronic communication methods (Giddens, 2006). Ulrich Beck (1999) is another analyst of this trend, which he has labelled ­cultural liberation. Examples of more specific analyses include, among others, George Ritzer’s (1998) idea of the McDonaldization of postmodern society, Ronald Inglehart’s (1990) works on the shift from materialism to post ­materialism, and Willy Guneriussen’s (2012) analysis of the development from industrial society to the information society to the (current) adventure society. If we enter the grey zone between sociology and political science and agree that political science includes several interesting works on political culture with a sociological twist, we might mention democratization as a third element of the cultural shifts that have occurred in parts of the world. Our starting point could be Gabriel A. Almond and Sidney Verba’s (1965) book The civic culture, in which they present an elaborate scheme of the components of national political cultures and subcultures, before they compare such cultures in five Western countries on different levels of social development. The book inspired many studies of all kinds of polities, including the one by Lipset and Rokkan mentioned above. In present-day sociology, there seems to be a demand for diagnoses of contemporary societies. Obviously, an important part of such diagnoses would be research on the conflicts between religions and world views that plague parts of today’s world. Some of these conflicts exist as conflicts between subcultures within societies. Studies of Cultural Elements One major task in the more mundane and everyday sociology of culture is to study how cultural patterns are created, maintained or changed either as part of the interaction between people in various situations or by developments or changes in relational and material structures. In doing so, culture has been subdivided in several ways, including by picturing culture as an onion with several layers inside each other. From the outside we can see the symbols: things, signs and actions that express meanings are meaningful for insiders but may not be understood by those outside the social system and do not know the codes. Language and speech acts are good examples. The next layer consists of the myths and narratives common to members of the collectivity. These myths and narratives may be communicated during religious rituals, in national hymns and stories about national heroes, and in histories of organizational success by preachers, teachers, information directors or other carriers of various traditions. The innermost layers of the cultural onion are the elements of the common universe of meaning of the social unit itself. These elements constitute the frame of reference within which the group or organization or

Culture as a Sociological Concept

105

national society or world region is to be understood: its values and standards of action, its shared and basic knowledge and beliefs, and its visions and schemes of interpretation. For several reasons, sociological studies of cultural elements usually take a different division of culture as its point of departure. In a manner of speaking, they cut the onion halfway across into sectors based on typologies of beliefs and values from everyday life. Several of them are the same as the social institutions that sociologists usually see as their objects of study. In this chapter’s limited space, it is possible to mention only a few examples of studies of elements of the most important cultural institutions, which may be subdivided into core institutions and institutions belonging to the culture sector. For several practical reasons, a major portion of the studies of such institutions are conducted on a national level. Some of the existing studies are comparative, and quite a few are local. The core areas of a society’s or a group’s universe of meaning include its religion, its legal system, its language, its science, its educational system, and its mass media. These institutions are subfields of study in the international community of sociologists, as are many of the subjects belonging to the socalled culture sector. We talk about the sociology of literature, the sociology of music, the sociology of film, the sociology of theatre, the sociology of sports, and so on. Religion, of course, is a classical object of study. As we know, Karl Marx considered religion as part of the false consciousness of a society, as the ‘opium of the people’. Max Weber studied the connections in people’s minds between their religious beliefs and values and their attitudes towards economic life. Emile Durkheim considered religion as one of the forces that holds a society together. Ideas like these have been followed up in modern sociology and extended to some of the institutions mentioned above. Contemporary empirical studies of cultural phenomena address their development, their integration and disintegration, and their variations and conflicts. We find studies of the ongoing secularization and privatization of religion in the Western world, along with religion’s transformation in several countries from the worship of God to an apparatus for conducting ceremonies such as weddings and funerals. We find studies of how the legal apparatus treats people from different social classes, studies of how crime stories in books and on television contribute to social cohesion, and studies of how stigmatization of some youngsters create criminal subcultures. Conflicts over scientific methods in different parts of the academic world have been ­recorded, and communication and pressures in science laboratories are the subject of sociological analysis.

106

Martinussen

There is almost a separate field of research into how middle-class culture and codes dominate educational systems. The construction of gender roles (‘doing gender’) is studied in several contexts. Uses of the various types of mass media are heavily researched. The media contents, age, gender and class divisions, and the art school conflicts in literature, music, film, theatre and sports are extensively investigated and interpreted from a sociological point of view. One of the early contributions to the sociology of literature is Richard ­Hoggart’s (1962) study of the uses of literacy in the British working class.

Culture has Consequences

Culture is important not only as an object of study in its own right. It is important because social interaction patterns and life conditions may be interpreted and explained as results of cultural forces, which in its turn gives us a broader understanding of how social relations are kept alive or change. This insight of course is not new, as the well-known expressions ‘the power of folded hands’, ‘knowledge is power’, and ‘language is power’, indicate. From its beginnings to today, mainstream social science has theorized and investigated various types of cultural forces. This has been achieved in several ways. To keep the story short, I will present three versions, each of which emphasize somewhat different sociological ideas. Nicos Mouzelis (1995: Ch. 6) has called these versions the positional, the interactive-situational, and the dispositional approaches. Referring to Part 2 above, one could state that the point of departure of the positional approach is the relational structure of ­social systems, whereas interaction patterns are at the centre of the interactive-­situational idea and cultural structure underlies the dispositional approach. One might venture that he is referring to analyses on the structural, the ­interactional and the actor levels. Positional Power As mentioned above, both Karl Marx and Emile Durkheim launched ideas about cultural power differences between social positions and systems. Talcott Parsons and his collaborators and followers have elaborated this idea into what might be called the theory of social institutions and roles. A society is viewed as a complex set of more or less integrated institutions in which the members contribute according to values and norms linked to their positions in the institutions’ subsystems. The institutions have particular functions for the society as a whole, organizations have their specific functions in and sometimes across institutions, and positions in groups and organizations have their designated

Culture as a Sociological Concept

107

functions. What people actually do depends upon of the definition of their various roles and role expectations, which in turn and to a relatively high degree follow from specifications of the positions in the position-role-role behaviour triad. Role (i.e., cultural) conformity thus reproduces the institutional complex and contributes to social order. Parsons himself published one of the most cited and criticized analyses in this vein in his article on the American family (1955). That analysis describes a mix of stability and change in the American institution of the family in the 1950s. Functions in society as a whole and in the single nuclear family are outlined and linked to the positions of husband-provider, housewife, and child. Changes in these roles are considered to be a result of changing relations of production and consumption in the wider society. The processes by which new individuals become able to adopt both the family roles they encounter and the roles they all must eventually play in society at large (as neighbour, citizen, car driver, friend and so on) are the processes of socialization, social pressures, and social control, buttressed by sanctions of various types. ­Socialization processes range from the explicit and intended (instruction, schooling) to the ­unintended and unrecognized (imitation, unconscious adaptation of norms, etc.). The positive sanctions include community rituals, which are often studied using Durkheim’s ideas as a starting point. The family rituals include courting, wedding, baptizing, celebrating birthdays and weddings, and so on. Negative sanctions involve punishments of various types, exclusions, stigmatizations, withdrawal of love, etc. These concepts have been useful in studying both deviation from values and norms and a broad range of social conflicts. One early and famous example is Robert K. Merton’s (1957) analysis of people’s adaptations to incongruent goals and approved means in the us and the different types of deviation they produce. Among other questions, he asks why there is an inverse relationship between social status and crimes for profit. He finds the answer in a combination of individuals’ opportunity situations and the dominant cultural values concerning success. When means and ends are incongruent—as they are nearly by definition in the lower social strata—a stressful situation occurs. For some individuals, the solution to this situation is crime. This type of deviance thus represents an adaptation to the national culture. Perhaps a still more famous example of the power of culture is Emile ­Durkheim’s (1952) analysis of suicide, viewed as a consequence of a weakening of the collective consciousness. One of his general ideas is that central values and norms can be both too strong and too unclear. In both cases, the risk of suicide increases because the individual feels either too committed or too confused. The first type of suicide—the altruistic suicide—is primarily found in

108

Martinussen

special situations in which the honour or glory of the community is at stake. Suicide bombers are today’s most conspicuous examples. The more frequent type is the anomic suicide. Anomic suicides increase when the norms become unclear and social controls weaken. This may occur during economic depressions or other rapid social changes, when mobility increases and social ties are disrupted. The above examples illustrate that the cultural integration of members of groups and societies has various types of consequences. On the personal level it may promote social security, self-control and a reasonable degree of conformity. When cultural integration is too strong or weak, or when values and norms are in conflict, its consequences may become negative. At the community or society level, members’ successful socialization to their roles creates social integration, productivity and stability. The latter consequences have been particularly contested partly because they are often structural-functionalist in nature. Talcott Parsons’ ideas about the functions of the nuclear family may be outdated, but the concepts of position, role and role behaviour have been used with all types of variations in countless studies of smaller and larger social systems. The words ‘position’ and ‘role’ may have been abandoned and replaced by other formulations, but similar ideas are used extensively in empirical research in current everyday sociology. To mention only one example, the Norwegian sociologist Stein Bråten (1973) coined the term ‘model monopoly’ for the power held by leaders and experts in organizations of various types, referring to their exclusive understanding and information of how their system works and the ability to express their opinions and wants in a language common to other power holders. His idea of model power soon became popular among researchers in several subfields of sociology, perhaps particularly in studies of communication. We find the same basic line of reasoning in his ideas regarding language power, cultural capital, and the like. Interpretive Power Whereas the positional approach presupposes a relatively broad and stable cultural core in the system that is studied, the idea of culture in interpretiveinteractionist reasoning is of a more floating symbolic community based on more spontaneous interactions resulting from particular situational factors. Although a common universe of meaning exists, it is continually changing and open to influences of various types. Values, norms and beliefs are continuously defined and redefined. In the words of William Isaac Thomas (1928), if men define situations as real, they become real in their consequences. The idea is that we live in a mosaic of smaller and greater symbolic universes consisting

Culture as a Sociological Concept

109

of ideas and understandings with a certain degree of inner consistency and order. Many sociologists refer to the sum total of these universes of meaning in a society as its culture. The sociological version of this thought with regard to social construction has its roots in Max Weber’s idea of meaningful social collectivities. An ­important part of its roots stems from Durkheim (1995), with his analysis of the importance of how rituals contribute to the collective consciousness. To present a fair picture of common meanings, sociologists must be on the inside of human communities and study the subjective understandings that their members attach to particular phenomena. What types of rationality do they use, and what do their exchanges of symbols, their narratives and their joint actions look like? As we all know, Peter L. Berger and Thomas Luckmann (1967) specified a systematic point of departure for this approach. Its history in modern sociology, however, goes further back to theorists such as Alfred Schütz, Georg Herbert Mead, Charles H. Cooley, William Isaac Thomas, Herbert Blumer, and others. Following the so-called cultural turn in the humanistic and social sciences, they have been followed by a great many others. In this approach, the power of culture lies in the joint actions and interactions between people: in their negotiations, joint interpretations, rituals, selfpresentations, stigmatizations, etc. Common meanings, beliefs and ideas are ‘worked forward’ in all sorts of encounters in which we engage every day, and for various reasons we act in accordance with them. One such reason is our inclination towards obedience, as demonstrated in Stanley Milgram’s (1963) (in)famous experiments. A dramatic example from a canonized Norwegian thesis may be illustrative. Nils Christie (1972) interviewed prison guards from a prisoner of war camp in World War ii in order to understand why, as time passed, Norwegian guards recruited among ordinary people were able to commit the most hideous and violent crimes against the prisoners. He found the answer in the gradually expanding social distance between guards and prisoners. During this process the prisoners were increasingly defined and regarded as different from and less human than others—they became aliens, so to speak. This could occur because the prisoners were weak, dirty and marked by hunger and did not carry any symbols showing bonds to communities such as families, workplaces, and residential areas, as people typically do. Another example frequently used to illustrate how culture is simultaneously communicated, interpreted, elaborated and influential is the article ­‘Doing gender’ by Candace West and Don Zimmerman (1987). They observe what people do and analyse which expectations and beliefs they communicate in their countless day-to-day conversations about how boys and girls at

110

Martinussen

all ages are ­behaving and ought to behave. In all of these situations, genderspecific rules and ideals are repeated and conveyed into everyday practices. The ­mechanisms at work are both an imitation of the behaviour of others and affirmation of their expectations of what is legitimate and what is considered deviant. The result is that gender roles are confirmed and preserved. Of course, this line of reasoning could be extended to processes such as ‘doing social class’, ‘doing disability’, etc. An earlier example of the power of symbolic interaction is Howard S. ­Becker’s (1963) theory regarding recruitment to criminal subcultures. The theory is that delinquents are produced during repeated interactions in a selfreproducing process that transforms particular youth into outsiders. The first incident of importance may be to be arrested for a minor infringement or to be seen in ‘bad company’. The next step—if there is one—is to be among those who are followed, detected or investigated by the police. For the majority, this occurs only once or twice. This depends (in the us) on whether you are black or white, show that you repent, are qualified to plead your case, etc. If you are convicted, you are in real danger because you may be defined as a person against whom one must guard. Sometimes you then end up in processes of self-definition and interaction that result in a criminal career. Another example, which illustrates the power of the mass media in interaction with politicians and other influential elites, is Herbert J. Gans’ (1995) study of the definition of the American underclass and the consequences of a ­common acceptance of this definition. He finds that over the years an entire ideology has developed, with ‘undeserving’ as its key concept. An impression is created that poverty is the exclusive fault of the poor themselves, and this image is maintained by a ‘war against the poor’ that simultaneously restricts the public use of money to help the poor and monitors their morality and behaviour. On the community level—from the smallest to the largest units—people’s joint actions and deliberations are viewed as creating both value integration (Barth, 1966) and social pressures to behave in particular ways. This leads not only to social stability and social integration but also to a sense of belonging and personal security. Values and beliefs are sometimes in conflict, and such conflicts and inconsistencies may create enmity or even war. On the personal level, the consequences may be of the type described by Becker and Durkheim, as mentioned above. Several theorists have attempted to build bridges between the positional and the interactive-situational approaches, as Durkheim touched on in his time. One conspicuous example is Anthony Giddens, with his theory of structuration. His general idea is not very different from that elaborated by Berger

Culture as a Sociological Concept

111

and Luckmann. Another bridge builder is Pierre Bourdieu, who also may be regarded as a central figure in the dispositional approach. Culture and Personality The third approach to be exemplified is an amplification of the foregoing. It emphasizes how culture shapes people’s personalities and thereby makes studies of values, attitudes, ideas, skills and feelings—and the actions and interactions that follow—central objects of study. Because culture is structured in various ways and may be subdivided both horizontally and vertically, these types of studies often address subcultures and their conflicts, cultural and social inequalities, and ideologies. Karl Marx and Antonio Gramsci are among the classics that inspired this approach, which has been further developed and used by anthropologists such as Ruth Benedict and Margaret Mead and social psychologists such as Abram Kardiner and Kurt Lewin. Kardiner proposed the concept of a basic personality structure and used it in a sociological analysis in collaboration with Ralph Linton (1939). An early example from sociology is the book by Hans Gerth and C. Wright Mills (1953) titled Character and social structure. The psychology of social institutions. For many years the words ‘culture’ and ‘personality’ were frequently juxtaposed in the social sciences. The main idea of this concept has been revived by Pierre Bourdieu (1990, 1984) in his key concept of habitus. The idea of cultural power may be expressed as follows. In any social system, whether large or small, the discourse, concepts, symbols and practices in various situations are self-evident for some individuals, whereas others either are alienated in conversations or do not understand them at all. This leads to inequalities in the ability to influence and convince others. Such cultural superiority and subordination relates partially to which values and norms are legitimate in particular connections and partially to which beliefs and knowledge are correct and how they should be expressed. In a given society, this inequality is found in all spheres, and we then refer to a cultural hegemony. As suggested above, sociological studies of the reproduction of social status via the mechanism of education are almost countless. Part of this process is economic: parents in higher social strata can afford to pay for their children’s education more readily than parents in the lower classes. However, much of this explanation lies in the inequality of family cultures. This inequality has been conceptualized in several ways. Basil Bernstein viewed it as a difference of language, with an elaborated code in the middle class and a restricted code among the lower strata. As an elaborated code consists of a more detailed and varied language, with the possibility of abstraction and complex reasoning. Those who have a good command of it will have advantages in situations in

112

Martinussen

which such communication is important. Higher education consists of this type of situation (Bernstein, 1974). Pierre Bourdieu emphasizes personal development and what Max Weber would have labelled traditional action in his theory of the development of different habituses in the different social classes. This entails that children and youth from the lower classes obtain less cultural capital than those from the middle classes. The control of cultural capital is particularly decisive in determining which type of education they are able to complete and wish to pursue and thereby which occupations they may hold later in life. To a large extent, these skills, perceptions and attitudes are subconscious, habitual, and more or less consistent. However, one result of class-differentiated socialization and internalization is the strong tendency for children to select educations that keep them in the same social class as their parents (Bourdieu and Passereon, 1977). In many ways Bourdieu’s ideas are reminiscent of the concept of attitudes, which has a long history in sociology and social psychology. Habitus might be viewed as the sum total of an individual’s basic skills and attitudes and the resulting automatic bodily expressions. To mention just one example, in the sociology of politics the idea of how political beliefs, aspirations, feelings and knowledge influence citizens’ democratic participation has been used in a large number of studies. This idea probably gained momentum after the publication of The civic culture. The authors describe this culture as a composite of political cognition, feelings towards government, partisanship, the obligation to participate, and a sense of civic competence. The most central elements of political culture promoting participation have been designated political resources. From the 1960s to the present it has been shown that in Norway (and most other countries that have been studied), such mental resources are crucial for people’s voices to come to the fore. Moreover, access to this type of cultural capital is based on a person’s education, profession, social capital, etc. (Martinussen, 1977). How are the various subcultures, which are so important for people’s life conditions, reproduced? Quite a few studies indicate that the obvious sociological answer—primary and secondary socialization—tells only part of the story. More complicated spillovers by imitation, group interpretation, propaganda, mass media reception etc., are also frequent. In the 1960s, an extensive long-term study of political socialization was conducted in the us to investigate—among many other things—why parents and children were so alike in their political preferences, activity or alienation. The results indicate that the process of ‘task generalization’—the use of knowledge, skills and orientations from other areas of life (work life, leisure activities, etc.)—might be of equal

Culture as a Sociological Concept

113

weight (Martinussen, 1972). In other words, personal dispositions—a person’s habitus—reproduce social inequalities. In future sociology the probability is that cultural inequalities and their consequences will become still more important as objects of study. Phenomena such as cultural clashes and cultural fragmentation may become obvious problems for sociologists to address in our globalized world. References Alexander, J.C. (2003). The meanings of social life. A cultural sociology. Oxford: Oxford University Press. Almond, G.A. and Verba, S. (1965). The civic culture. Boston and Toronto: Little, Brown and Company. Barth, F. (1966). Models of social organization. Occasional Papers 23. London: Royal Anthropological Institute. Beck, U. (1999). World risk society. Cambridge: Polity Press. Becker, H.S. (1963). Outsiders. Studies in the sociology of deviance. New York: The Free Press. Bellah, R.N. et al. (1985). Habits of the heart. Individualism and commitment in American life. Berkeley and Los Angeles: University of California Press. Berger, P.L. and Luckmann, T. (1967). The social construction of reality. Garden City, N.Y.: Doubleday. Bernstein, B. (1974). Class, codes, and control. London: Routledge and Kegan Paul. Bourdieu, P. (1984). Distinction. A social critique of the judgement of taste. London: Routledge. Bourdieu, P. (1990). The logic of practice. Cambridge: Polity Press. Bourdieu, P. and Passeron, J.C. (1977). Reproduction in education, society and culture. London: Sage. Bråten, S. (1973). “Model monopoly and communication”, Acta Sociologica. 16(2): 98–107. Christie, N. (1972). Fangevoktere i konsentrasjonsleire. [Prison guards in concentration camps]. Oslo: Pax Forlag. Durkheim, E. (1933) [1893]. The division of labor in society. New York: The Free Press. Durkheim, E. (1952). Suicide. A study in sociology. London: Routledge and Kegan Paul. Durkheim, E. (1995). The elementary forms of religious life. New York: The Free Press. Edles, L.D. (2002). Cultural sociology in practice. Oxford: Blackwell Publishers Ltd. Elias, N. (1987). The civilizing process. Oxford: Blackwell. Gans, H. (1995). The war against the poor. New York: Basic Books.

114

Martinussen

Gerth, H. and Mills, C.W. (1953). Character and social structure. New York: Harcourt, Brace & World, Inc. Giddens, A. (1991). Modernity and self-identity. Self and society in the late modern age. Cambridge: Polity Press. Giddens, A. (2006). Sociology. 5th edition. Cambridge: Polity Press. Gramsci, A. (1975). Letters from prison: Antonio Gramsci. New York: Harper Colofon. Guneriussen, W. (2012). “Opplevelsessamfunnet—en vidunderlig ny verden? Kulturelle spenninger i det senmoderne sanmfunnet”, Sosiologisk Årbok. [“The adventure society—a wonderful new world? Culture tensions in the late modern society”, Yearbook of Sociology]. Oslo: Universitetet i Oslo. Hoggart, R. (1962). The uses of literacy: aspects of working-class life with special reference to publications and entertainment. Harmondsworth: Penguin. Horkheimer, M. and Adorno, W. (1947). Dialektik der Aufklärung. Amsterdam: Querido Verlag. Inglehart, R. (1990). Culture shift in advanced industrial society. New Jersey: Princeton University Press. Inglis, D. and Hughson, J. (2003). Confronting culture. Sociological vistas. Cambridge: Polity Press. Kardiner, A. and Linton, R. (1939). The individual and his society. New York: Columbia University Press. Larsen, H. (2013). Den nye kultursosiologien. [The new sociology of culture]. Oslo: Universitetsforlaget. Lewis, O. (1959). Five families: Mexican case studies in the culture of poverty. New York: Basic Books. Lipset, S.M. and Rokkan, S. (1967). Party systems and voter alignments. New York: The Free Press. Løchen, Y. (1965). Idealer og realiteter i et psykiatrisk sykehus. [Ideals and realities in a psychiatric hospital]. Oslo: Universitetsforlaget. Lysgård, S. (1967). Arbeiderkollektivet. [Worker’s collective]. Oslo: Universitetsforlaget. Mannheim, K. (1960) [1936]. Ideology and utopia. London: Routledge and Kegan Paul. Martinussen, W. (1972). “The development of civic competence: socialization or task generalization”, Acta Sociologica. 15(3): 213–227. Martinussen, W. (1977). The distant democracy. London: John Wiley & Sons. Marx, K. and Engels, F. (1970) [1845–1846]. The German ideology. New York: International Publication. McClelland, D.C. (1976). The achieving society. New York: Irvington Publishers Inc. Merton, R.K. (1957). Social theory and social structure. New York: The Free Press. Milgram, S. (1963). “Behavioral study of obedience”, Journal of Abnormal and Social P­ sychology. 67(4): 371–378. Mouzelis, N. (1995). Sociological theory. What went wrong?. London: Routledge.

Culture as a Sociological Concept

115

Ogburn, W.F. (1964). On culture and social change. Chicago: The University of Chicago Press. Parsons, T. and Bales, R.F. (1955). Family, socialization and interaction process. New York: The Free Press. Parsons, T. and Shils, E.A. (eds.). (1951). Towards a general theory of action. New York: Harper & Row. Ritzer, G. (1998). The McDonaldization thesis. London: Sage. Rokkan, S. and Valen, H. (1969). “Regional contrasts in Norwegian politics”, Eds. E. ­Allardt and S. Rokkan. Mass politics. New York: The Free Press. Sennett, R. (1998). The corrosion of character. The personal consequences of work in the new capitalism. New York: W.W. Norton. Sorokin, P.A. (1959). Social and cultural mobility. London: The Free Press of Glencoe. Thomas, W.I. and Thomas, D.S. (1928). The child in America: behavior problems and programs. New York: Knopf. Weber, M. (1958). The protestant ethic and the spirit of capitalism. New York: Charles Scribners Sons. West, C. and Zimmerman, D. (1987). “Doing gender”, Gender and Society. 1(2): 125–151. Wrong, D. (1970). Max Weber. Englewood Cliffs, N.J.: Prentice-Hall. Hemel Hempstead.

chapter 7

Bringing Nature Back In Willy Guneriussen Introduction Since the breakthrough of industrialism, modern society has become an important force of nature on the global scale, influencing the atmosphere, oceans, freshwater systems, soil, vegetation and animal life and thereby changing the conditions of its own development and functioning. Society is not merely social, and nature is not only natural (Castree and Braun, 2001: 5). To understand contemporary society, sociology must include an analysis of nature—both as environments and as invironments, as exterior conditions and forces and as internal dimensions of society. The imprint of anthropogenic forces on increasingly numerous aspects of the surrounding earth system has motivated natural scientists (ecologists, climatologists, chemists, etc.) to argue that over the past century, the earth has entered a new geological era: the Anthropocene (Crutzen and S­ toermer, 2000). This is the first era defined with reference to human impact on the surrounding earth system. Man is no longer ‘…a species tossed about by larger forces—now we are those larger forces’ (McKibben, 2006: xviii). The ­Anthropocene is a result of the development and expansion of modern technology, institutions, culture and social structures. Therefore, social science can contribute to the understanding of the Anthropocene by broadening the analysis of various aspects of modern society to include nature/materiality. A theory of society that focuses only on the ‘purely’ social elements— ­symbolic fields, relations, role structures, etc.—is no longer adequate to the task of understanding the epoch in which we live. ‘Humans have changed the way the world works, and now we must change the way we think about it too’ (Folke, 2013: 26).

Society-Nature: Internal and External Contradictions

For the most part, ‘standard’ sociology and social theory have left nature in the background both as the most general conditions of society and as a reservoir of resources. Conceptions of nature have been left to the natural sciences (e.g., Habermas, 1968, 1970). © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_008

Bringing Nature Back In

117

Important natural processes are now changing as a result of human impact. We must adapt to a more recalcitrant nature, a nature we can no longer control in the manner ‘promised’ by a progress-ridden modern culture, a nature that is ‘kicking back’. Nature is responding to late modern societies in the shape of risks, threats and spoiled resources (e.g., climate change and pollution of air, water and land). Can it be that the most successful societies in human history are about to reach thresholds or tipping points beyond which the natural conditions for their existence and continuation will be threatened by their very success? Most modern Northern-Western societies have been extremely successful as measured by indicators of material wealth, freedom and democracy, legal equality, education, lifespan, health, and technological capacity. This success has largely resulted from a historical moment of great luck: the discovery of and ability to use large reservoirs of energy stored in coal, oil and gas. Fossil fuel has enabled and influenced most of the structures and activities in modern society, including the complex transport system, the geography of ­settlements, urban-suburban structures, industrial areas, harbours, etc. Can humans maintain these structures and systems in a future post-fossil-fuel society? In any event, the ‘fossil fuel age’ will be an era of short duration compared with other eras in human history (Murphy, 2013: 172). Marx and Engels (1967) argued that capitalist societies would dissolve due to internal contradictions and conflicts within and between social forces and elements, namely, class conflicts and the contradictions between the forces and relations of production. Modern society is now riddled with ‘external’ contradictions resulting from its own powerful impact. The landscapes transformed by technology and all of the externalities of human/social activity are combining to create new ‘natural’ conditions on local and global scales. The impact and energy of social forces are depleting natural resources (e.g., the extinction of fish stocks), destabilizing biophysical processes (e.g., the eutrophication of water) and pushing natural systems out of homeostatic balance (e.g., climate change). The principle of (limitless?) material progress—a principle that has been at the heart of modernity since its inception—appears to be out of step with this new understanding of nature’s vulnerability. Incompatibility between the principles of (traditional) social development and its natural conditions constitutes a new ‘external’ contradiction.

Sociology and Nature

For decades, various sociological schools and traditions have been marked by a linguistic or cultural ‘turn’ in which signs, symbols and meaning have been

118

Guneriussen

of prime interest. Social realities are conceived as emergent realities of social forces and meaning realized through meaningful and meaning-generating ­actions, norms, structures, etc. The natural/material underpinning of these ­realities have receded into the background and been regarded as obvious ‘conditions and resources’. Material things have primarily been analysed as signs, as carriers of symbolic meaning (Ray and Sayer, 1999: 8; Harré, 2002). Symbolic and communicative aspects of social reality have been at the forefront of several sociological traditions (e.g., Luhmann, 1984, 1997; Habermas, 1968, 1970; Giddens, 1984). Theories of language (e.g., structuralism and post-structuralism), linguistic actions and linguistic structures, rhetoric, communication theory, discourse theory/analysis and speech act theory have been important theoretical tools for the analysis of social realities. The beginning of this turn—or ‘symbolic turn’—can be found in the classical traditions, e.g., Durkheim’s theory of religious and social symbolism, Simmel’s ideas of the sign-functions of action, artefacts and architecture, and Mead’s theory of significant symbols. Although this turn towards meaning and symbols in sociology and social theory has proven immensely fruitful, it may have contributed to an overly denaturalized conception of society and a disembodied conception of actors and actions. Well-founded critiques of radical objectivism, material determinism, and biological reductionism, among others, have sometimes evolved into extreme anti-materialism, culturalism and subjectivism.

A Material Turn

During the past two to three decades, we have witnessed the development of a new sub-discipline in sociology focusing on ‘environment and society’ (Cudworth, 2003; Barry, 2007; Gåsdal and Sande, 2009). Nevertheless, this has not resulted in nature/matter becoming more firmly established in the basic conceptions of society/social realities in sociology in general. This has remained a theme/interest for a particular sub-discipline. Ulrich Beck (1986) stands out as an exception with his epoch-making theory of risk society, as does ActorNetwork Theory (e.g., Latour, 2004, 2005; Haraway, 1991). The topic of climate change is now also drawing the attention of sociologists. Some argue that ­sociologists will have to revise their basic conceptions of society/modernity because of the new knowledge of global warming (see Giddens, 2009; Urry, 2011). That said, a number of these contributions are marked by radical constructivism—nature is re-incorporated mainly as a discursive product or ‘object’ (see Haraway, 1991: 183 ff.; Macnaughten and Urry, 1998)—and thus nature appears to disappear once again, this time behind a veil of discourses.

Bringing Nature Back In

119

Talcott Parsons: The Human Condition In the 1960s and 1970s Talcott Parsons established concepts of nature/ materiality as important dimensions in his functionalist theory of society, culminating in one of his latest and most ambitious publications, ‘A paradigm of the ­human condition’ (Parsons, 1978). Although he stated ‘…human action … consists mainly of symbolic meanings (the very antithesis of things physical)’ (Parsons, 1978: 379), he incorporated energy and matter alongside information into his new conception of the human condition. In his earlier writings, the system perspective involved nature as an external physical and biological condition without much elaboration. Nature was primarily related to the ‘adaptive subsystem’ (the economy) in his scheme of various systems. Nature had no place in the other subsystems—i.e., the political, social and cultural systems. In ‘A paradigm of …’ Parsons makes many references to the natural sciences (physics, chemistry and biology) and develops a relatively elaborate concept of nature through the further development of the four-functions scheme. This perspective is explicitly anthropocentric (Parsons, 1978: 361), and Parsons ­attempts to accentuate the various types of meaning achieved by different dimensions of nature through the ‘telic system’, i.e., the system of (cultural) symbolism that defines the meaning of various fields and levels of reality. Parsons does not describe nature merely as ‘external conditions’. Nature is included in the general scheme of the human condition, and different aspects of ­nature achieve different types of meaning, for example, instrumental vs. affective meanings. Even the physical realm (or the ‘Physico-Chemical-System’) is brought into this conception of humans/meaning/society (ibid: 357, 382), although ‘physical objects are most likely to have purely instrumental meanings’ (ibid: 365). The differentiation between physical and biological/organic dimensions is important. The human body is of particular interest to Parsons. A web of cultural symbolism defines the meanings of various parts and functions of the body—sense organs, the mouth, genitals, birth, growth, death, eating, defecation, etc. They achieve/have a variety of affective and expressive meanings. Other dimensions of the encompassing biophysical reality can also achieve various non-instrumental meanings. This is far from the traditional Cartesian conception of nature as nothing more than an external mechanical object without further ‘meaning’ and for which only a cognitive and instrumental relation is adequate. Because Parsons died a year after this publication, he never had the opportunity to develop these ideas. Ulrich Beck: Risk Society and Nature Ulrich Beck conceptualizes nature mainly through the prism of the side effects of social-technological impacts on the environment. Therefore, in light of the massive and accelerating character of this impact throughout the industrial

120

Guneriussen

era, he can argue that nature no longer exists as a separate reality beyond and outside of society. According to him, this implies ‘the end of the opposition between nature and society’ (Beck, 1986: 107, my translation). This thesis (of the end of nature, also supported by Giddens, 1991, 1994, 2009) makes sense when applied to physical, biological and geological traits of nature that have been transformed through human activity. Even certain basic trends in global climate change are ‘manufactured’. However, this thesis also appears problematic when confronted with the enduring reality of natural mechanisms, causal processes, and physical, chemical and biological principles. These forces and processes are not ‘transformed’. Instead, they are the foundation of any transformation through human or non-human influence. At a basic level, nature is still at work on both the outside and the inside of society. Actor-Network-Theory: Materiality and Social Realities Proponents of Actor-Network-Theory have performed the important task of ­re-incorporating materials/materiality into sociology/social theory (Law and Hassard, 1999; Latour, 2004, 2005). In various ways, they have contributed to a rematerialisation of concepts of social realities. As Latour has argued, things and materials of various sorts make society durable. Social networks are ­‘saturated’ with (the use or ‘enactment’ of) things and material processes—cars, roads, computers, houses, wires, radio transmissions, electricity, food, etc. However, this theory is also extremely constructivist, arguing, for example, that ‘nature’ is not a given but a scientific construction (Latour and Woolgar, 1979). It also tends to eliminate almost all ordinary concepts of action and actors (Latour, 2005). Latour has also dismissed any distinction between nature and society, calling it ‘a strange belief’ that we (Westerners) separate ‘things’ and ‘persons’ (Latour, 2004: 45). Without some type of distinction between things and persons, causality and intentionality, it is difficult to grasp how Latour himself can take up a position as an arguing subject attempting to convince other actors of the truth of his assertions. In his last work, Talcott Parsons stated that life ‘…is a mode of the organization of the physical world’ (Parsons, 1978: 372). We can expand on that idea and state that society is a mode of organization of organic, chemical, physical and psychological realities. This concerns both the environment and what I have called the invironment—the presence of nature in the internal functioning of society. Natural materials and processes are conditions of, used in and transformed through the construction of various social realities. The various social systems represent various forms of organization of various natural materials. Any materialized social system requires continuous energy support. Moreover, because energy used cannot be reused or recycled (like plastic bags), every

Bringing Nature Back In

121

­social system requires external sources of energy input. Social systems are fundamentally dependent on natural materials, processes and energy.

Co-evolution—Social-Natural Interdependency

Late modern societies are facing a new ‘socio-natural’ (or ‘social-ecological’ or even ‘socio-technical-ecological’) reality. The expansion and development of modern institutions and life forms are made possible by a historically unprecedented and massive intrusion into various natural systems in an effort to control processes from the subatomic level to wide-ranging national, regional, continental or even global geophysical systems (e.g., the damming of great rivers around the world). The totality of the side effects of these social and technological efforts of control and development appears to constitute a new reality beyond (ordinary forms of) control. We are now living in an ‘age of side effects’—a reflexive modernity in which modern institutions have to make an increasing effort to contain the unforeseen consequences of the modern techno-social institutions themselves (Beck, 1986: 14). The ‘externalities’ are adding up and coming back as global warming, threatening waste, pollution, depleted resources and runaway processes. Nature—and modern society—has turned out to be vulnerable. We can no longer conceptualize nature as unlimited materials and resources at man’s disposal. Of course, we cannot destroy nature— it will be nature regardless of what man does to it. Nevertheless, nature may change in ways that are destructive for many life forms and species, including man. Over the past several decades a new research programme has begun to study the society-nature constellation through a concept (and theories) of ‘co-evolution’—the idea that social and natural dimensions and elements are intertwined and exert a reciprocal influence on each other that—over time— may result in evolutionary change on both sides (see Weisz and Clark, 2011). This idea began with the 1960s discovery in biology that a process of reciprocal evolution appeared to occur between certain species of plants and butterflies. These discoveries led to the development of a more generalized perspective within biology and ecology that has influenced various fields, including computer science, economics and organizational studies (Weisz and Clark, 2011: 281). On the most general level, we can theorize the co-evolution of two great complex systems: nature and society. We use the following example. Some social structures and processes are made possible (afforded—not determined) and have developed because of the availability and utilization of certain

122

Guneriussen

­natural elements, processes and structures, i.e., the manner in which automobile transport and the road infrastructure are conditioned by (the discovery and use of) fossil energy carriers. A complex socio-material structure has crystallized and become a condition for further action—roads, shops, human settlements, interconnected systems of various settlement concentrations, locations of industry and residential areas, trade systems, etc. Values, interests, action goals and a range of human life forms have become ‘locked in’ to this socio-natural reality. It has evolved to become a new, emergent evolutionary plateau for social development. In addition, it affects natural processes and changes—for instance, CO2 output resulting from the extensive use of fossil fuel contributes to global warming, which then causes the release of even more CO2 and methane from melting permafrost in a vicious feedback loop. A new plateau for biological growth and decay is developing that may be the basis for later evolutionary changes—in nature and in society. This idea of the co-evolution of social and natural realities/systems implies a concept of emergence. More complex systems, new principles and mechanisms arise out of lower-level processes. Parsons (1978) used it to explain the evolving difference between chemical-physical conditions and biological processes/elements and between chemical, physical and biological systems and social realities. This does not amount to an idea of mysterious, uncaused emergence. We can instead refer to ‘weak emergence’ (Elder-Vass, 2010). New realities emerge out of the elements and how they are organized at a lower level, such as the manner in which thinking and feeling arise out of the organization of electromagnetic connections/processes in the brain without being reducible to these lower-level elements. The theory (theories) of co-evolution more specifically implies (imply) that evolutionary processes on both sides of the society-nature divide are combined, producing new emergent elements/­ processes in both fields. The idea of ‘interdependent social-ecological systems’ (Folke, 2013: 27) ­appears—in various ways and through different vocabularies—to have become a common (and diffuse) ground for a range of academic communication focused on society-nature relations. …the ultimate goal of coevolutionary research could be to identify more precisely the subsistence niche for human civilization and to explore patterns and properties of accessible, inaccessible and catastrophic socioecological coevolutionary pathways. weisz and clark, 2011: 281

The idea of ‘interdependency’ does not imply ecological determinism, namely, that natural conditions either determine or ‘allow’ only one type of social ­order.

Bringing Nature Back In

123

The concept of ‘subsistence niche’ implies the possibility of a range of social processes and practices that can subsist without undermining its natural conditions in the longer term. However, one element stands out for most of these researchers: the current state of affairs and the observable trends of change regarding the consumption and exploitation of natural resources cannot continue without eroding its own (natural) grounding. ‘… [T]he “normal” world of growth is a temporary anomaly destined to self-terminate by natural means’ (Murphy, 2013: 173). We should add ‘… and by social means’. Murphy’s point is that nature itself will put an end to human activities and societies that radically overstep planetary boundaries. In the long term, nature cannot ‘deliver’ the necessary resources and ‘ecosystem services’ to uphold these life forms. However, dramatic social change will (of course) also be driven/modified by actors (at various institutional levels) reacting to nature’s reactions to human action. The ‘self-termination’ of a life form is never only an effect of nature, except in cases of massive and rapidly evolving natural catastrophes. Social change may be ‘triggered’—not precisely determined—by changes in natural conditions, which may narrow down the ‘subsistence niche’ for future life forms.

Limits, Sustainability and Planetary Boundaries

The idea that there are ‘hard’, nature-induced limits to industrial development, material and economic growth was given forceful expression in 1972 when ‘Limits to growth’ (Meadows, Meadows, Randers and Behrens, 1972) was published. The researchers feared ‘…that industrialism itself might be unsustainable’ (Dryzek, 2013: 30) given current population growth and industrial expansion. The only way to ward off negative consequences would be to limit population growth, shut down polluting industrial production and reduce the consumption of natural resources. Various green movements actively sup­ported the report’s main thesis. Others condemned it as just another Malthusian doomsday warning. Overall, it was not well received by defenders of modern industrialism, whether right-wing defenders of capitalism or socialist and ­social-democratic defenders of industrial development. They considered social, economic and technological progress as a necessity for the system and its citizens and a core value in modern culture. It is man’s nature and duty to utilize whatever nature can offer (see Young, 2013; Dryzek, 2013). Their main concept of nature is purely instrumental. Nature is ‘resources’—whatever can be of use to humankind. We can describe this relation to nature as ‘resourcification’. The economic value of natural things and processes is the primary measure. Any limit on exploiting natural resources is viewed as a potential

124

Guneriussen

­economic loss or foregone ­income. This represents an extreme anthropocentric position: nature’s value and meaning derives (primarily) from its (economic) utility to humans. As we have noted, anthropocentrism does not always imply this type of economic instrumentalism. Parsons (1978) argued for a type of anthropocentrism that is quite different from this narrow instrumental perspective, namely, allowing for affective and emotional values. The gloomy vision of ‘Limits to growth’ according to which industrial societies’ current way of life will end in overpopulation, chaos, pollution of the water, air and land, depleted resources, etc., had little impact on mainstream politics in Western countries. On the contrary, the breakthrough of neoliberalism in the 1980s strengthened the Promethean ideology1 of unlimited growth and inexhaustible resources. Mainstream politicians greeted the report with silence and vague assurances that any environmental problem could be solved. Numerous neo-classical economists argued that the report did not understand the potential of markets and technological creativity to secure resource-­ substitution if and when a specific resource were to become either inaccessible or exhausted (Beckerman, 1974). Technologists argued that the report had neglected the potential for technological development related to pollution, energy and clean industry. However, new research has confirmed several of the original report’s prognoses (Running, 2012: 1458). In 1987, the World Commission on Environment and Development published Our common future. This report painted a far more positive picture of modern industry, technology and economy than the earlier report on limits. The idea of (absolute) limits included a message encouraging the reduction of industrial activities to help the environment and the future. The new report launched a novel and much more flexible principle: sustainable development or ‘sustainability’. This positive and reassuring message stated that economic and material ‘development’ was compatible with concern for the environment. Economic and industrial growth was possible without depleting vital ­resources in a way that would worsen future generations’ living conditions compared with our own. The report’s definition stated that sustainable development ‘meets the needs of the present without compromising the ability of future generations to meet their own needs’ (World Commission on Environment and Development, 1987: 43).

1 Prometheus (a titan and mediator between the gods and humankind) stole fire from the gods and gave it to humans, thus increasing humans’ power, including their ability to resist the gods.

Bringing Nature Back In

125

‘Sustainability’ became an important buzzword, and it still holds a prominent place in various contemporary political and scientific discourses. The term has ‘invaded’ most fields and lost much of its original meaning along the way, although its original meaning was not very precise. Through increasingly frequent vernacular use, it seemed, the word sustainable became a synonym for the equally vague and unquantifiable adjective green, suggesting some undefined environmental value, as in green growth or green jobs. Today the term sustainable more typically lends itself to the corporate behaviour often called greenwashing. engelman, 2013: 3

Even oil drilling and coal mining can be viewed (by some) as sustainable, meaning little more than ‘less polluting than some other oil drilling or coal mining activity’. Such drilling and mining is claimed to be ‘more sustainable’, even if it adds constantly to the release of greenhouse gases, pushing the global climate system into a threatening, non-sustainable state. The production of new car models may be described as sustainable because it requires less energy and metal than the production and use of earlier models. However, the production, worldwide transportation and use of growing numbers of cars leave an increasingly large ecological ‘footprint’ on the earth. One airline company argued that its new technology saved enough aluminium in one year to build three new airplanes (Engelman, 2013: 3), in other words, it is ‘sustainable’, even though the anticipated increase in air traffic will increase environmental impact. Some refer to a sustainable economy and sustainable welfare systems as meaning nothing more than the possibility of continuing activities and institutions without financial breakdown, regardless of the long-term impact on nature. ‘Sustainability’ as a concept has been severed from the original reference to a society-nature balance. It is not only the later use or abuse of the term ‘sustainability’ that has ­diluted its meaning. Indeed, this term lacked precision from the beginning. How should we understand the ‘needs of the present’? We cannot interpret it simply as ‘preferences’ or wishes. If we did, we would have to contend with rising (sometimes limitless) expectations of the level of affluence among both the poor (peoples/countries) and the rich. It is obvious that such expectations must be curtailed. If all wishes for increased material affluence were fulfilled, humanity would consume natural resources far beyond any level of sustainability, resulting in radically reduced living conditions for future generations. Humanity’s total impact on nature has already reached or exceeded important ‘planetary boundaries’ (Steffen et al., 2015). It cannot continue at this level

126

Guneriussen

without threatening and weakening vital living conditions for humans and many other species in the long term. This raises important moral and political questions as to who can legitimately aim to increase their use of and impact on the environment and who must accept new limitations—even reducing their consumption of natural resources. I shall not take up this discussion here (see Vetlesen, 2015). The ‘needs of the present’ must somehow be balanced against the needs of future generations, and the needs of various segments of contemporary humanity must be balanced against each other. What are then our needs, and how shall we describe the needs of future generations? This is chronically unclear. Certain needs are mainly biologically conditioned and limited—e.g., the amount of freshwater a human requires to function well. However, a broad range of needs and wishes are socially and culturally conditioned, and a number of them are in principle without ‘natural’ limits—e.g., the wish for money or power. Ultimately, there is no obvious scientific answer regarding what our legitimate needs are. Through a combination of the life sciences, psychology and social science it may be possible to define a ‘bottom line’, a set of basic biological and social needs—water, food, security, group affiliation, etc. Beyond that—with regard to the wish to own a house and a car, the freedom of lifestyle choices and so forth—we need to ask where we set limits to legitimate fulfilment, and who can set them? Recently, a new framework or paradigm for analysing the impact of anthropogenic forces on nature—the framework of ‘planetary boundaries’— has moved to the forefront of the ecological debates. Its key progenitors are researchers at the Stockholm Resilience Centre (see Rockström et al., 2009; Steffen, Crutzen and McNiell, 2007; Steffen et al., 2015). This framework ‘…defines a safe operating space for humanity based on the intrinsic biophysical processes that regulate the stability of the Earth system’ (Steffen et al., 2015: 1). The new approach is also important from a sociological perspective. If it ‘survives’ scientific critique, it will have consequences for how nature is conceptualized within the social sciences.2 It will also provide measures to analyse and evaluate how well or poorly ongoing social and economic processes are adapted to a variety of planetary boundaries in the longer term and what a ‘successful society’ can mean. 2 There is also the question of political acceptance. Will it share the fate of the former ‘Limits to Growth’—being for the most part neglected or even rejected by the dominant pro-growth political parties and powerful economic and industrial actors? The new Paris consensus of 2015 on restricting CO2 emission worldwide could be viewed as a positive beginning. However, little genuine, consequential commitment is visible thus far.

Bringing Nature Back In

127

Unlike the discourse on ‘limits to growth’, this approach does not operate within fixed natural limits that cannot be crossed without ecological breakdown as the inevitable consequence. It does not identify exact tipping points beyond which biophysical systems will inevitably break down or be quickly driven into irreversible runaway processes. Neither does the paradigm rely on a relatively diffuse and often non-consequential concept of sustainability in which a wide range of ordinary, resource-demanding and polluting economic and material growth processes can be ‘greenwashed’ and thus persist (Engelman, 2013: 3). What it does, however, is to identify a range of ‘safe operating spaces’ for certain parameters of the general earth system. The concept of ‘safe operating space’ means that as long as human resource use and impact on the environment remains clearly below the upper safety threshold (for the specified parameter), it does not threaten the stability or durability of the system/processes/resources. Moving beyond this safe zone will lead to increasing ­uncertainty, risks and dangerous conditions from the perspective of human well-being (and also posing increasing threats to many other species). The farther away we move from the ‘safe space’, the more likely destructive runaway processes and irreversible deteriorating conditions for humans become. The procedures for identifying planetary boundaries are based on available knowledge/research within the natural sciences—physics, chemistry, geology and biology—combined with knowledge/research regarding the various impacts of human/social activity. The social-science side of this new conception is not very well developed compared to the natural scientific foundation. Thus far, researchers have identified nine planetary boundaries: climate change, ocean acidification, stratospheric ozone depletion, nitrogen and phosphorous cycle, global freshwater use, land system change, biodiversity loss, atmospheric aerosol loading, and chemical pollution (Rockström et al., 2009; Steffen et al., 2015). They have not been able to quantify two of them: atmospheric aerosol loading and chemical pollution. Originally, they pointed to changes within three of these dimensions as particularly threatening, clearly moving beyond a safe operating space—climate change, biodiversity loss (later renamed ‘biosphere integrity’) and the nitrogen/phosphor cycle. Nitrogen and phosphor are essential for biological growth. However, the extensive use of these ingredients in modern agriculture and industry (e.g., fertilizers) threatens water quality. Eutrophication may render the affected water inhospitable or even poisonous for animal life. In addition, phosphor is a limited resource that is now overexploited. Recently these researchers have concluded that land use—if the present trend of expansion continues—is also moving beyond the safety threshold, with possible reinforcing consequences for processes related to several other planetary boundaries.

128

Guneriussen

Steffen et al. (2015) considered freshwater use to be within the safe operating space. New research has questioned this estimate and concluded that the balance of freshwater renewal and use is moving beyond the safe zone (Jaramillo and Destouni, 2015). Several groundwater aquifers around the world have been drained much faster than they can regenerate, and many have been polluted by industrial activity. On all continents, large coastal cities are encountering difficulties caused by the draining of groundwater reservoirs. The ground level is sinking, endangering buildings and infrastructure. Compounded with rising sea levels, a number of these cities are facing severe threats of flooding. Researchers have been unable to quantify chemical pollution and choose safe operating values for this dimension. This is a serious and growing problem. Detritus from human activity is polluting the soil, air and water. Much of what is accrued is not biodegradable, and some of it is quite poisonous. It may be broken down into small parts and particles but continue to affect the biosphere for a long time. It may accumulate and threaten to push systems out of the safe zone. One particularly infamous example is what has come to be named ‘the great Pacific garbage patch’—a collection of marine debris revolving in the North Pacific Ocean that is controversially estimated to cover an area twice as big as Texas. It consists of all sorts of floating materials from large plastic objects on the surface to small, invisible particles just below the surface. Much of the material will take decades, even centuries to dissolve, and some of it never will. The small particles are particularly threatening because they inevitably affect fish and birds and spread through various food chains, eventually entering human food. Several of the great oceanic currents are slowly but steadily turning into long-distance transport systems for increasing volumes of debris. Many shores and beaches worldwide are now experiencing a process of ‘plastification’ (I will use ‘plastification’ although plastic debris is just one of several materials involved). All sorts of floating materials wash ashore and after a time are (partly) covered with sand or gravel. This affects animal life in these coastal areas, although the long-term effects are uncertain and impossible to quantify. We cannot specify a ‘safe operating space’ and a planetary boundary in this regard. However, this sort of continuous, increasing and long-term impact and pollution should be managed based on a precautionary principle that aims to lessen the impact and reduce the probability of destructive processes in the long run. Similar processes affect soil and air. The totality of the consequences of these forms of chemical influence is difficult to calculate. However, the increasing impact and accumulation of various types of debris may push local and global processes beyond the safe operating space of several planetary boundaries.

Bringing Nature Back In



129

Accelerating Impact—A Formula for Self-Destruction?

When Crutzen began to develop and popularize the theory of the Anthropocene, he argued that the industrial breakthrough in the eighteenth and ­nineteenth century signalled the epochal transition from the Holocene to the Anthropocene (Crutzen and Stoermer, 2000). Socio-technological impacts were beginning to affect the earth system in important ways. Based on their own research, along with reviews and metastudies of research on various dimensions of the earth system, Crutzen and others have argued that a distinct nonlinear increase in environmental impact began in the middle of the twentieth century, providing ‘the first indisputable evidence that human activities were affecting the environment on a global scale’ (Steffen et al., 2007: 617). They began to use the term ‘the great acceleration’ to characterize this nonlinear growth along several dimensions of society-nature relations (see also Young, 2013: 115 ff.), including energy use, population, transport, trade, water use, fertilizer use, river damming, tourism, foreign investments, urban populations, gross national product, etc. (see figures in Young, 2013: 119, 121). This means that we are not merely living in the Anthropocene but in a period in which anthropogenic forces are transforming socio-nature at increasing speed and reach. A few of these changes will—if they continue—push various local and global systems beyond safe operating spaces and into situations involving increasing uncertainty and risk of destructive runaway processes. Can these processes be controlled and slowed before they trigger irreversible deterioration in living conditions for humans and other species? Are contemporary societies able to act on this knowledge of planetary borders? In spite of man’s knowledge and technological inventiveness, it is not obvious that the complex systems and forces in modern society can be re-­programmed to achieve new goals sufficiently quickly to avoid dire consequences. These goals involve catching up with the threatening developments caused by these systems; slowing and reducing a variety of human-social-technological impacts; and restraining traditional ‘progress’—material growth, higher gdp, increasing consumption and physical mobility. Forty-five years ago, Ehrlich and Holdren (1971) constructed a formula that appears to grasp the more general mechanism underlying current trends: ‘i = pat’. This formula states that human impact (I) on the environment (or the ‘earth system’) is a function of the size of the Population (P), the level of Affluence (A) and the level/power of Technology (T). It can explain a number of accelerating processes. Population size is obviously an important factor. Increasing size (all other things being equal) means increasing impact. Since the middle of the 20th century, in the same period as ‘the great ­acceleration’,

130

Guneriussen

­ opulation size has increased sharply. However, increasing population is not p the only variable that explains increased impact. Man is the only species for which each individual’s ecological imprint is not relatively constant and ­determined by natural conditions alone. In addition to variation in natural resources and man’s natural makeup, material affluence and technological power are important determinants of the scope of impact of individuals’ activity in the society in question. Increasing material wealth and technological capacity, combined with increasing population, can explain part of the nonlinear growth in various fields. Even if population growth levels off, anthropogenic influence may still increase at high rates. This formula (i = pat) can contribute towards understanding many of the rapid changes occurring in late modern societies. Nevertheless, it requires sociological and technological refinement. Population size and growth is the ­‘easiest’ variable: growing size means increasing consumption, which implies increasing impact in a straightforward linear sense (all other things being equal). The affluence and technology elements of the equation are not straightforward. Affluence most often involves consumption capacity regarding the use of both material things and (directly or indirectly) natural resources. Higher levels of ‘material consumption’ in the general population means increased impact. However, affluence may also relate to non-material consumption. Within the service economy, the information economy and the experience economy, a broad spectrum of consumption relates to information, symbols, images, sound, social bonds, togetherness, communication, experiencing, viewing, and feeling (Jensen, 1999; Lash and Urry, 1994; McGillivray and Frew, 2007; Pine and Gilmore, 1999). Although different from traditional material consumption, there is barely any type of non-material consumption that does not use or impinge on material/natural realities. Sports and leisure facilities presuppose industrial production. Their construction and use require a large amount of energy, and many of them require new tracts of land. As a reaction to ‘bad winters’ (too little snow too late in the winter) Norwegian operators of ski tracks and alpine ski slopes use large amounts of energy to run machines that can produce snow. They are now preparing the next and even more energy-intensive step: installing machines that can produce snow when the temperature is above zero degrees Centigrade. Tourism, particularly ‘adventure tourism’, has been considered part of the new ‘green economy’. However, the expanding tourism industry presupposes an extensive transport system and facilities on the ground. As an example, representatives of this branch of the economy in Northern Norway have repeatedly called for the government to invest in roads, airports and harbours to enable an ever-growing number

Bringing Nature Back In

131

of tourists to visit and experience ‘Northern magic’ and ‘Arctic wilderness’ (Guneriussen, 2008a, 2008b, 2012). Technological development does not always result in increased impact on natural systems. New technologies that are more energy-efficient may replace old ones and lower the release of CO2 per unit produced, although the net result may be negative if the increased efficiency is converted into growing consumption. Certain technological innovations do not (primarily) result in the increased exploitation and consumption of natural resources. The problem with the ‘T’ in the ‘i = pat’ equation may not be technological development and heightened technological power in itself. In principle, technology that is more powerful may be used to reduce working hours, even to the point at which the total impact is reduced. This does not appear likely given the current global capitalist system of production and intense competition, the force of consumerism and the culture of progress and material growth.

Concluding Comments: Energy and Change

In this paper, I have paid special attention to three developments that orig­ inated in natural science: the concept and theory of co-evolution (from the 1980s onward), the concept/theory of the Anthropocene (from 2000), and the concept/theory of planetary boundaries (from 2009). Although these ideas and programmes demonstrate the interconnection of natural and social systems, they are not well developed on the social scientific side. Much work remains to elaborate on the social mechanisms that are important for the ongoing changes in various socio-natural systems. We need more developed studies of contemporary societies and their institutions to understand how various institutional mechanisms generate (increasing) impact, whereas others can limit impact. The capitalist market economy, the techno-industrial complex3 and consumerism as a lifestyle—all heavily based on fossil fuel—are the most important drivers of increasing impact (Wilenius, 1999: 35 ff). Can our type of society take a comprehensive ‘green turn’ towards maintaining human impact in various fields of activity within safe planetary boundaries? The answer to this question primarily hinges on the use of energy. The natural sources of energy, the transformation of energy sources into available energy, and the level of energy use are the key factors. In public discussion of a 3 This complex includes technological research and development organizations and relevant departments within universities.

132

Guneriussen

green turn, a green economy or a ‘low-emissions society’, it often appears as if everything depends on our willingness to press a switch button from polluting to clean energy use. In reality—and this is another lesson from natural and technological sciences, particularly from thermodynamics—a comprehensive change-over towards new energy sources can itself be very energy intensive to the point at which the environmental impact of switching far exceeds safe planetary boundaries before the alternative energy sources and technologies can replace the traditional energy carriers—mainly fossil fuel. ‘It takes energy to acquire energy’ (Zencey, 2013: 78). The rate of energy returned on energy invested (often referred to as the eroi formula) may change radically over time. In the petroleum branch, it has fallen from 100:1 100 years ago to less than 20:1 today and is steadily shrinking.4 ­Windmills may achieve a rate close to the current petroleum rate, a little under 20:1, and they release almost no atmospheric gases. Wind and other low-­ release energy sources (hydrogen, nuclear power, wave mills, etc.) appear as the obvious choice for a future society in ecological balance. However, the energy required to substitute the amount of energy from fossil fuel must ­originate from available energy sources. Today this means that we have to use enormous amounts of fossil fuel to construct new technology and energy plants. A windmill park presupposes the production, transport and use of concrete, steel and other materials. Roads must be built, land areas must be developed, etc. Wind is free and inexpensive (ecosystem service). Constructing a windmill park, maintaining it and building the necessary distribution system is not. This applies to any alternative. If we are to maintain our current level of affluence and consumption, the release of greenhouse gases will rise far beyond scientifically recognized danger levels long before—in a distant future—we have access to the required amounts of clean energy. Aside from the release of climate gases, other problematic impacts must also be managed if modern societies’ influence on natural processes is to be maintained within the safe spaces of the various planetary boundaries—land use, chemical pollution, overuse of fresh water, biodiversity loss, etc. To understand the conditions and consequences of ‘the great acceleration’ that has been occurring in late modern society, sociology must bring nature back in, which implies incorporating findings from the natural sciences. We live in a high-speed and high-consumption society that is changing the earth system and various ecological processes in ways that threaten (future ­generations of) humans and other species. This may trigger runaway processes that will undermine the conditions for a well-balanced socio-natural order. 4 This means that it took the energy equivalent of 1 barrel of petroleum to produce 100 barrels in 1920 and 1 barrel to produce less than 20 today.

Bringing Nature Back In

133

References Barry, J. (2007). Environment and social theory. London: Routledge. Beck, U. (1986). Risikogesellschaft. Auf dem Weg in eine andere modern. Frankfurt a.M.: Suhrkamp. Beckerman, W. (1974). In defence of economic growth. London: Cape. Castree, Noel and Braun, B. (eds.). (2001). Social nature. Theory, practice and politics. Oxford: Blackwell Publishing. Crutzen, P. and Stoermer, E. (2000). “The anthropocene”, Global Change Newsletter. 41(May): 17–18. Cudworth, E. (2003). Environment and society. London: Routledge. Dryzek, J. (2013). The politics of the earth. 3rd edition. Oxford: Oxford University Press. Elder-Vass, D. (2010). The causal power of social structures. Cambridge: Cambridge ­University Press. Engelman, R. (2013). “Beyond sustainababble”, State of the World 2013: is sustainability still possible?. Washington: Worldwatch Institute/Island Press. Erlich, P.R. and Holdren, J.P. (1971). “Impact of population growth”, Science. 171(1971): 1212–1217. Folke, C. (2013). “Respecting planetary boundaries and reconnecting to the biosphere”, State of the World 2013: is sustainability still possible?. Washington: Worldwatch Institute/Island Press. Gåsdal, O. and Sande, A. (2009). Miljø og samfunn. [Environment and society]. Oslo: Cappelen akademisk. Giddens, A. (1984). The constitution of society. Cambridge: Polity Press. Giddens, A. (1991). Modernity and self-identity. Cambridge: Polity Press. Giddens, A. (1994). Beyond left and right: the future of radical politics. Cambridge: Polity Press. Giddens, A. (2009). The politics of climate change. Cambridge: Polity Press. Guneriussen, W. (2008a). “Modernitet, fortrylling og magi—konstruksjon av det ‘nye’ Nord-Norge”, Sosiologisk Tidsskrift. [“Modernity, enchantment and magic—­ construction of the ‘new’ Northern Norway”, Journal of Sociology]. 16(1): 31–54. Guneriussen, W. (2008b): “Modernity re-enchanted. Making a ‘magic’ region”, Eds. J.O. Bærenholdt and B. Granås. Mobility and Place. Aldershot: Ashgate. Guneriussen, W. (2012). “Opplevelsessamfunnet—en vidunderlig ny verden? Kulturelle spenninger i det senmoderne samfunnet”, Sosiologisk Årbok. [“The ­adventure ­society—a wonderful new world? Culture tensions in the late modern society”, Yearbook of Sociology]. (2): 41–69. Habermas, J. (1968). Technik und Wissenschaft als “Ideologie”. Frankfurt a.M: Suhrkamp. Habermas, J. (1970). Erkentnis und Interesse. Frankfurt a.M: Suhrkamp. Haraway, D. (1991). Simians, cyborgs, and women. The reinvention of nature. London: Free Association Books.

134

Guneriussen

Harré, R. (2002). “Material objects in social worlds”, Theory, Culture & Society. 19(5–6): 23–33. Jaramillo, F. and Destouni, G. (2015). “Comment on planetary boundaries: Guiding ­human development on a changing planet”, Science. 348(6240): 1217. Jensen, R. (1999). The dream society. New York: McGraw-Hill. Lash, S. and Urry, J. (1994). Economies of signs and space. London: Sage. Latour, B. (2004). Politics of nature. London: Harvard University Press. Latour, B. (2005). Reassembling the social: an introduction to Actor-Network-Theory. ­Oxford: Oxford University Press. Latour, B. and Woolgar, S. (1979). Laboratory life: the social construction of scientific facts. London: Sage. Law, J. and Hassard, J. (1999). Actor Network Theory and after. Oxford: Blackwell. Luhmann, N. (1984). Soziale Systeme. Frankfurt a.M.: Suhrkamp. Luhmann, N. (1997). Die Gesellschaft der Gesellschaft. Frankfurt a.M.: Suhrkamp. Macnaughten, P. and Urry, J. (1998). Contested natures. London: Sage. Marx, K. and Engels, F. (1967). The communist manifesto. Hammondsworth: Penguin books. McGillivray, D. and Frew, M. (2007). “Capturing adventure: trading experiences in the symbolic economy”, Annals of Leisure Research. 10(1): 54–78. McKibben, B. (2006) [1989]. The end of nature. New York: Random House. Meadows, D.H., Meadows, D., Randers, J. and Behrens, W. (1972). Limits to growth. New York: Universe Books. Murphy, T.W. jr. (2013). “Beyond fossil fuels: assessing energy alternatives”, State of the World 2013: is sustainability still possible?. Washington: Worldwatch Institute/Island Press. Parsons, T. (1978). “A paradigm of the human condition”, Action theory and the human condition. New York: The Free Press. Pine, J.B. and Gilmore, J.H. (1999). The experience economy. Boston: Harvard Business School Press. Ray, L. and Sayer, A. (eds.). (1999). “Introduction”, Culture and economy after the cultural turn. London: Sage. Rockstrom, J. et al. (2009). “A safe operating space for humanity”, Nature. 461(7263): 472–475. Running, S. (2012). “A measurable planetary boundary for the biosphere”, Science. 337(6101): 1458–1459. Steffen W., et al. (2015). “Planetary boundaries: guiding human development on a changing planet”, Science. 347(6223): 1259855. doi: 10.1126/science.1259855. Steffen, W., Crutzen, P. and McNeill, J. (2007). “The anthropocene: are humans now overwhelming the great forces of nature?”, Ambio. 36(8): 614–621. Urry, J. (2011). Climate change and society. Cambridge: Polity Press.

Bringing Nature Back In

135

Vetlesen, A.J. (2015). The denial of nature: environmental philosophy in the era of global capitalism. London: Routledge. Weisz, H. and Clark, E. (2011). “Society-nature coevolution: interdisciplinary concept for sustainability”, Geografiska Annaler: Series B. 93(4): 281–287. Wilenius, M. (1999). “Sociology, modernity and the globalization of environmental change”, International Sociology. 14(1): 33–57. World Commission on Environment and Development. (1987). Our common future. Oxford: Oxford University Press. Young, O. (2013). On environmental governance. London: Paradigm Publishers. Zencey, E. (2013). “Energy as master resource”, State of the World 2013: is sustainability still possible?. Washington: Worldwatch Institute/Island Press.

chapter 8

The Study of Consumption in Sociology—Beyond Utility Theory Pål Strandbakken In the last few hundred years, the acquisition, flow and use of things—in short, consumption—has become a defining feature of our lives … In the rich world—and in the developing world increasingly, too—identities, politics, the economy and the environment are crucially shaped by what and how we consume. Taste, appearance and lifestyle define who we are (or want to be) and how others see us. trentmann, 2016: 1



Introduction: A Sociology of Consumption?

Consumption is the reason that anything is produced. Although this statement is obviously true, this truth is too often neglected by the social sciences. ­Sociology has traditionally had a strong production focus, from Spencer’s idea of an industrial society, via Marx’s economic regimes, Asiatic, ancient, feudal and bourgeois modes of production (Aron, 1991) and Durkheim’s The division of labor in society (1979) to Bell’s distinctions among agricultural, industrial and post-industrial societies (Bell, 1973). The modern sociological tradition, as it was developed from the early years of the 20th century by writers like Tönnies, Simmel, Weber and Durkheim, may be regarded partly as a critique of the shortcomings of the discipline of economics, and simultaneously as an analysis of modernity and the break with traditionalism. With the (re)introduction of economics’ models of man and behaviour to sociology (Homans, Coleman), this ‘original’ critique or attempt at transcending economics has become less clear and visible. Nevertheless, for the sociological study of consumption it appears vital to transcend economists’ definitions of demand, preferences and consumer rationality.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_009

Study of Consumption in Sociology—Beyond Utility Theory

137

It is not obvious to this author that consumption ought to be regarded as an analytical concept or whether, in the main, it is merely one phenomenon among others discussed by sociologists. There are numerous more or less formalized theories in the study of consumption, but in my opinion ­‘consumption’ itself does not necessarily qualify as a sociological concept. It does, however, bear more of a resemblance to a sociological concept when it is introduced in tandem with production. And even if consumption is not a defined concept, the twin areas of production and consumption cover a large field of sociologically relevant human activity. Furthermore, the term becomes rather more conceptual when it is juxtaposed with demand. In addition, the tem consumer is rather specific, contrasting fruitfully both with the more political term citizen and with the economists’ agent, the customer. To provide a general account of the sociological study of consumption is far beyond the scope of this (or any other) reasonably short text. The main ambition of this contribution therefore is to highlight some of the reasons that sociologists and anthropologists tend to regard economists’ perspectives as inadequate or superficial, and its main strategy is to introduce consumption in opposition to—or as a vast expansion of—economists’ demand. To this end, I start with a short detour into language, followed by a short presentation of the history of the sociology of consumption. Etymology According to The Cassel Thesaurus (1998), the transitive verb (to) consume has five specific meanings: (1) to eat and drink, (2) to use up, (3) to destroy by fire, decomposition, etc., (4) to waste, to squander, and (5) to engross or obsess. In moral and political debate, consumption is often conceived as something negative and immoral. The thesaurus contains a list of synonyms for consumption that includes, inter alia, to deplete, to expend, to destroy, to demolish, to finish up, to devastate, to throw away, to waste, to wreck, to expand and to ruin. A religious/sacrificial dimension is also visible given that synonyms such as to fascinate and to obsess will lead our thoughts in such directions. A number of these aspects have been explored by Georges Bataille (1987, 1988; Miller, 1998), best-known is probably his analysis of the potlatch rituals of Pacific North American tribes. The negative meanings or aspects of consumption are further enhanced by the historical use of the medical term ‘consumption’. According to Susan ­Sontag, ‘consumption’ was ‘in use as a synonym for pulmonary tuberculosis

138

Strandbakken

as early as 1398’ (Sontag, 2002: 10). In the 16th century it was also a common ­synonym for cancer. When social critics and environmentalists claim to be ‘against consumption’, in itself a rather absurd claim, this set of negative connotations potentially influences our evaluation of what today should essentially be treated as a value neutral and descriptive concept.

The Way We Live Now

Partly because of the division of labour between Western men and women in the 20th century, production has come to be regarded as a serious and masculine activity, whereas consumption is regarded as superficial and feminine (Strandbakken, 2007). Debates regarding consumption and over-consumption tend to focus too narrowly on the clothes shopping of young women. This sort of gendered (and age-based) critique is always blind to men’s purchase of golf equipment, all-terrain bicycles, electronic devices and so on. Far more important however, is how this perspective fails to consider the bulk of modern consumption. Lifestyle and identity shopping represent only a minuscule fraction of total consumer spending. The boring stuff, such as housing, household energy, debt, white goods, groceries, transportation, insurance and so on, is economically (and in other ways) far more essential to the analysis of modern lifestyles. This type of consumption is what we, following Gronow and Warde (2001) will call ‘ordinary consumption’. To the extent that sociology repeats a tabloid and one-dimensional vision of modern consumption as primarily involving 17-year-old girls buying clothes and cosmetics, it fails to live up to its promise to address ‘the problems of biography, of history, and of their intersections within a society’ (Mills, 1980: 12). Societies are presently as much defined by their citizens’ patterns and levels of consumption as by their production, and one of the key issues of our age— environmental unsustainability—is a challenge both to consumption and to production. In contemporary sociology, consumption, class-based lifestyles, environmental aspects of consumption patterns, etc., should be treated just as thoroughly and seriously as production. Hence, as manufacturing moves from the old, rich parts of the world (mainly Western Europe and the usa at present) to Southeast Asia, we might even regard the supermarket or the shopping mall, and not the factory, as the archetypical image of contemporary Western society. To establish a sociologically fruitful perspective on consumption, economists’ traditional emphasis on buying or ‘shopping’ should initially be expanded

Study of Consumption in Sociology—Beyond Utility Theory

139

in two directions in time. It should include planning, use and disposal, that is, essentially what occurs before and after the acquisition of a particular item or service (Stø, Throne-Holst, Strandbakken and Vittersø, 2008). Of course, in most of these phases, consumption has interesting links to production and distribution. Furthermore, the use phase can be subdivided into reshaping, co-production, ‘prosumtion’ and consuming (meaning, as noted by the thesaurus, ‘to use up’). Consumption is linked to lifestyle, culture and identity and as such is located at the centre of modern existence. The ­Canadian anthropologist and consumer researcher Grant McCracken defined his field as follows: By “culture”, I mean the ideas and activities with which we construe and construct our world. By “consumption” I broaden the conventional definition to include the processes by which consumer goods and services are created, bought and used. Culture and consumption have an unprecedented relationship in our modern world … Never has the relationship between them been so deeply complicated. mccracken, 1988: xi

This field of study should not be reduced to accounts of consumer preferences, prices and (more or less) rational decisions in the market.

Consumption in History

For the activity of consumption to become an interesting topic for sociology, society will have to reach a level of affluence that surpasses mere subsistence. An element of choice must be included to meaningfully focus on consumption as a phenomenon and on consumers as a role or group. The rural proletariat in the Middle Ages had access to some food and some clothing, but it does not appear fruitful to call them ‘consumers’. McCracken (1988: 11) has identified three ‘episodes’ or booms in the history of consumption: first, among noblemen in 16th-century England; second, a consumer explosion in the 18th century that was widened in that it included a broader class base (McCracken, 1988: 16–22; McKendrick, Brewer and Plumb, 1982); and third, a transition of consumption from noble households to the market in 19th-century France (McCracken, 1988: 22–28; Williams, 1982). All of these episodes—and probably many others—may be good candidates for the title ‘the birth of consumer society’. However, far more relevant to this discussion is the post-wwii ‘affluent society’ analysed and theorized by

140

Strandbakken

J.K. Galbraith (1958), spreading from the us and gradually reshaping large parts of the world. The modern world or modernity remains our focus, even if it can be demonstrated that a number of the mechanisms at work may be similar to those operating 400 years ago. Trentmann (2016) provides a far more ambitious account of the history of consumption, presenting and analysing a vast expanse of economic history and even comparing developments in the West with the world beyond E ­ urope and North America (Trentmann, 2016). The subtitle to his book Empire of things, how we became a world of consumers, from the fifteenth century to the twenty-first, announces a thorough investigation of the consumer side of the market economy. We may regard the development of a consumer society as a long historical process involving gradually larger segments of the population and more and more parts of the world. However, to the extent that sociology is primarily the study of modernity, it appears most convenient here to restrict our analysis of consumption to the period that started in the 20th century and expanded beyond the us after wwii.

Consumption and Lifestyle in Classical Sociology

It is not possible to write general sociology without touching upon themes of lifestyle, consumption and mass markets. Here, however, I will note only a limited number of contributions that have remained both relevant and theoretically challenging. I have chosen to focus on Thorstein Veblen, Georg Simmel and Max Weber. In addition, Marx’s differentiation between needs and wants remains with us, even if the needs concept is changing because it is—and should be—based on a social minimum/sufficiency rather than on a biological one. Veblen: Conspicuous Consumption and Status Competition In 1899, Thorstein Veblen published his famous Theory of the leisure class: an economic study of the evolution of institutions (1994). Perhaps more a ­political-polemical treatise than a work of economics or sociology, this publication nevertheless contains a set of ideas and insights that have remained interesting. Its demonstration of the irrational aspects of status competition and the comparison with what were then regarded as primitive or inferior tribes is effective and thought-provoking, and Veblen’s approach has commonalties with Bourdieu’s famous analysis of class and class culture in Distinction ­(Bourdieu, 1984).

Study of Consumption in Sociology—Beyond Utility Theory

141

Veblen’s term ‘conspicuous consumption’ has survived and is still used to exemplify the consumer’s somewhat limited rationality. At minimum, it addresses a type of rationality that does not follow directly from a use value and value-for-money logic. Simmel: Modern Living and Trickledown Effects Simmel’s essay on modern life in large cities, ‘The metropolis and mental life’ (1976), is a surprisingly readable and up to date piece of writing, analysing the transition from ‘slow rural’ to ‘fast urban’ forms of life. In terms of the heritage of the sociology of consumption, however, Simmel is even more relevant for his analysis of ‘Fashion’ (1904), which addresses trickle down effects and the interplay between differentiation by superordinate groups and imitation by subordinate groups. His description of this mechanism is brilliant: he demonstrates how the interplay of the two opposing principles of differentiation and imitation transforms fashion, in a very broad sense, into a powerful social force. Despite the use of the term ‘trickle down theory’, Simmel describes what is mainly an upward striving mechanism through which subordinate groups pursue the status symbols of their superiors. The manner in which this dynamic and interplay between differentiation and imitation works in a specific historical situation is described by the design historian Adrian Forty: A constant succession of new designs was produced in small quantities for middle-class women who wished to be dressed in patterns that had not yet been reproduced on the cheaper fabrics worn by working-class women. It is hard to pin down exactly how the patterns for the middleclass differed from those for the working class and whether they indi­cated anything about the differences that were supposed to exist between the two classes. In any case, many fashionable designs were subsequently reproduced by the manufacturers on cheap cotton, a practice which both attracted working-class customers wanting to follow the fashion, and caused the owners of dresses in the first, expensive printing of a pattern to discard them, because they had become ‘common’, and to buy new ones. forty, 1995: 74–75

With only two strata, ‘top’ and ‘bottom’, you have a situation in which the top only differentiates and the bottom only imitates. In a more realistic model that includes many status strata, all of the in-between groups will engage in both

142

Strandbakken

differentiation and imitation, whereas the top and bottom layer will have only a single motive for their behaviour. This is McCracken’s observation (1988: 94) and not Simmel’s, but it demonstrates that Simmel produced a theory (or described a mechanism) that remains relevant and interesting more than a century after its formulation. Weber: Honourable Lifestyles In Economy and Society (Weber, 1978), Weber introduced his more sophis­ ticated interpretation of stratification, differentiating riches, honour and (political) power. He showed that it was too simple to reduce stratification to a mere question of economics, highlighting the idea of honourable lifestyles (the idea of a Stand or a status group), particularly among the nobility. Weber claimed that the common basis for explaining class differences was the relationship to production and the acquisition of material goods, whereas estates (Ständer) are identified and defined based on their consumption and lifestyle. Weber developed his approach partly in opposition to Marx’s class theory. On a philosophical level, Marx’s theory might be superior—status and power ultimately result from economics—but as an empirical tool Weber’s multidimensional perspective on stratification is more useful for sociological analysis. In contemporary sociology, this multi-dimensional approach is reflected in Bourdieu’s distinctions between economic and cultural (and social) capital.

The Modern Study of Consumption

Starting in the late 1960s, the study of consumption in sociology had a very strong bias towards use value, prices and ‘value-for-money’. Consumer studies were closely connected to events inside the consumer movement. Among the early model studies, I will highlight Ralph Nader’s Unsafe at any speed (1965), which addressed automobile safety, and Vance Packard’s The waste makers (1960), which addressed planned product obsolescence. Neither writer was trained as a sociologist (Nader was a lawyer, and Packard was a journalist), but these perspectives—and these types of perspectives—informed and guided early sociological consumer research. Consumers were entitled to good products at reasonable prices, safe products, safe and nutritious food and so on. In hindsight, this sociology appears to have been rather theoretically weak and underdeveloped (typically based on simple rational actor models imported from economics). It was essentially a type of ‘useful’, pragmatic social science based on a set of value orientations such as ‘consumer rights’ (Kennedy, 1962). This research had strong societal and political relevance and was c­ haracterized

Study of Consumption in Sociology—Beyond Utility Theory

143

by the frequent un-theoretical use of rather simple quantitative empirical studies. In an article on the recent history of the field, Gronow essentially dismisses this tradition and states that ‘The sociology of consumption did not really establish itself as a recognized field of social studies until the 1980s and early 1990s’ (Gronow, 2015: 41). In the wake of the Postmodern turn in the 1980s, sociological consumer ­studies overwhelmingly and enthusiastically came to adopt the symbolic value perspectives of Baudrillard (1980) and the approaches of cultural studies ­(Kellner, 1997). The focus on the symbolic properties of products became so strong that at least one carmaker argued, somewhat tongue-in-cheek, that their vehicles might also be used for transportation (Strandbakken, 2007). Although cultural studies and research inspired by it have subsequently been subject to much critique, we should not refuse to acknowledge that this approach succeeded for a reason. In its time, it was probably a necessary reaction to the aforementioned styles of consumer research, which were more or less unilaterally concerned with use value and its relation to exchange value/ price (Strandbakken and Heidenstrøm, 2013: 10). Furthermore, we should bear in mind that this tradition produced some excellent studies of modern consumption (e.g., du Gay, Hall, Janes, Mackay and Negus, 1997). Finally, symbolic value should not be dismissed altogether, even if it was sometimes exaggerated (for some of Baudrillard’s disciples, ‘everything is a sign’). The analysis of the symbolic aspects of goods and services should be incorporated into wider perspectives that include a focus not only on materiality or use value but also on exchange value. Even in the 1990s, however, the radical postmodern approaches did not completely dominate the field. The most common consumer sociology references at that time were probably Giddens, with his theories of late or high modernity and reflexivity (1990, 1991), and Bourdieu’s analyses of class-based lifestyles and taste hierarchies (1984). Giddens and Bourdieu both shared numerous key themes and approaches with the postmodernists. Giddens’ ‘modernity theory’, with its focus on the break with traditional scripts for life and the idea of the individual who must continuously create herself, has much in common with a main theme in postmodern sociology: In the post-traditional order of modernity, and against the backdrop of new forms of mediated experience, self-identity becomes a reflexively organized endeavor. The reflexive project of the self, which consists in the sustaining of coherent, yet continuously revised biographical narratives, takes place in the context of multiple choice as filtered through abstract

144

Strandbakken

systems. In modern social life, the notion of lifestyle takes on a particular significance. giddens, 1991: 5

Like a number of his postmodern contemporaries, Giddens was criticized for overestimating the autonomy of the individual actor and therefore underestimating the force of structural constraint. His idea of reflexive modernity was often argued to be a rather biased description of the affluent and educated middle classes in the rich world. Giddens disagreed and claimed that the reflexive creation of identity ‘refers also to decisions taken and courses of action followed under conditions of severe material constraint’ (Giddens, 1991: 6). ­Although he claims to analyse the human condition, however, it appears clear that the notion of a free choice of identity becomes gradually more relevant with increasing affluence and an increasing number of opportunities. Although Bourdieu is not a postmodernist either, his work has one striking similarity to the postmodern consumer sociology: In Distinction (1984) he introduces the idea that quality is socially constructed, so that taste is regarded only as a matter of the power to define. In a sense, this places Flaubert and Ian Fleming, or Stravinsky and the Spice Girls, at the same level. On the positive side, this approach opened up and legitimized the academic study of popular culture; on the negative side, it undermined the idea of quality altogether, which was both an intellectual loss and (most likely) at odds with Bourdieu’s intentions. Unlike Giddens, Bourdieu was criticized for underestimating the autonomy of the actor and of giving too much weight to structural constraint. The 1990s dominance of cultural studies or post modern consumption ­sociology—or approaches similar to these—predictably provoked a reaction. Gronow and Warde’s book Ordinary consumption (2001)—the title essentially reveals the project—formulated a critique that interestingly was more empirical than epistemological or theoretical. They claimed that in the previous sociology of consumption, too much emphasis had been placed on extraordinary consumption, on conspicuous consumption, and on individual choice rather than on contextual or collective constraint and that too much emphasis had been placed on conscious, rational decisions rather than on routines, conventional behaviour and repetition. According to Gronow and Warde, the sociological study of consumption should address the problems and habits of ordinary people, namely, the majority that struggles to manage in their daily lives. In hindsight, it appears as if Gronow and Warde’s attempt at influencing the discipline or tradition was a success. In sum, the sociology of consumption had been overly preoccupied with the identity work of transsexual graffiti artists from Barcelona and i­nsufficiently

Study of Consumption in Sociology—Beyond Utility Theory

145

concerned with the eating habits of senior citizens in Sheffield. This does not mean that the study of spectacular and eccentric subjects is illegitimate. ­However, we must consider the balance among various types of subjects.

Expanding the Concept beyond Economics and Demand

To create an opening to emphasize consumption parallel to the common emphasis on production is only the first challenge. To develop a contemporary sociology of consumption we must also transcend economists’ reduction of consumption to demand. The basic premise of this chapter is that ­sociological perspectives on consumption and lifestyle are potentially superior to economic perspectives when one is analysing, understanding and explaining consumption. It appears that it is inadequate to reduce consumption to individual preferences, consumer rationality and sensitivity to prices, primarily because of a lack of understanding of context, an underestimation of cultural aspects and an overly simple view of human motivation. Economists will claim both that the strength of their theoretical perspective is its parsimony and that this perspective is intuitively meaningful when it describes variations in supply, demand and price based on enticements and consumer rationality. In addition, economic models are developed for more large-scale analyses. Nevertheless, we are concerned that the simplified models of consumer behaviour fail to adequately describe the field. For a more elaborate critique of the attitude-behaviour model/theory of planned behaviour, see Gronow and Holm (2015). The two earlier-presented attempts at sociological definitions of consumption highlight some of the limitations of economics. First, Stø et al. (2008) insist on temporarily expanding individual consumers’ consumption practices to include not only planning and imagining prior to acquisition but also using and disposing of the product. Second, although McCracken’s definition includes ‘processes by which consumer goods are created, bought and used’ (McCracken, 1988: xi), his approach is even more important because of his insistence on the link between culture and consumption, on how consumption is linked to themes such as identity, status, integration and belonging. Below, I attempt to demonstrate how sociology has been able to move beyond the overly narrow perspective on consumer demand and preferences. The basic idea is to highlight examples from the sociology of c­ onsumption tradition that are believed to demonstrate the fruitfulness of more s­ ociological and cultural approaches to the study of consumption and modern lifestyles.

146

Strandbakken

Communication and Meaning In 1979, Mary Douglas and Baron Isherwood jointly published The world of goods. Towards an anthropology of consumption (1996). In it, they criticize what they call utility theory by focusing on what remains unexplained in the perspective of economic theory: There is no justification in traditional utility theory for assuming anything about physical or spiritual needs, still less about envy. The theory merely assumes the individual to be acting rationally, in that his choices are consistent with another and stable over the short time that is relevant. It says that his tastes should be taken as given, that he responds to a fall in prices by readiness to buy a larger quantity and to a rise by buying less, and that he responds in consistent fashion to changes in his income. As he gets more of a particular good, his desire for additional units weakens. douglas and isherwood, 1996: 19

Echoing Marx’s critique of economists’ ‘Robinsonades’, Douglas and Isherwood are critical of the idea of defining society by adding up individual actors: ‘The individual human being, stripped of his humanity, is of no use as a conceptual base from which to make a picture of human society’ (ibid: 63). Douglas and Isherwood regard products partly as an information system that is important for making cultural categories visible and stable. The two main points in their critique, or in their attempt at transcendence of utility theory, are their insistence on consumption as an inherently social activity, ‘the joint production, with fellow consumers, of a universe of values’ (ibid: 67), and on material goods or products playing a part in society as carriers of social meaning. To approach consumption as greed or stupidity will not help us understand it. Product Consistency In economic theory, products have value as individual unities and are replaceable with other products; they are described and analysed in isolation. In the sociology of consumption, another perspective and another set of insights have been developed. A visitor from Canada stayed for some weeks. She gave me a yellow silk umbrella, a little graceful umbrella with an ivory handle. It came from an altogether different life. It leaned against the wall in my kitchen, and I thought, if I use that, I’ll have to buy different clothes, live in a different kind of flat and certainly in another part of London. lessing, 1997: 169

Study of Consumption in Sociology—Beyond Utility Theory

147

Similar to the quote above, the so-called ‘Diderot effect’ and description of ‘Diderot unities’ challenges this view and points to examples in which products go together in culturally defined bundles. The initial observation is credited to the French enlightenment philosopher Dennis Diderot, presented in a short essay called Regrets on parting with my old dressing gown in which Diderot recounts what occurred after he received a new dressing gown as a gift from a friend. The new gown makes him aware that his desk and study are not up to standard, starting a process whereby he ultimately changes and replaces most of his product portfolio. ‘Gradually, the entire study, including its chairs, engravings, bookshelf and clock, were found wanting, and replaced’ (McCracken, 1988: 119). The introduction of Diderot’s essay into the present discourse and the modern interpretation of this phenomenon stem from McCracken’s essay ‘Diderot unities and the Diderot effect. Neglected cultural aspects of consumption’ (McCracken, 1988). His point, or observation, is that bundles of very different consumer goods have a type of consistency and go together in what McCracken calls Diderot unities. These unities are culturally constructed and change over time. One example used by McCracken is the ‘correspondence’ between Rolexes and bmws in the yuppie universe. This correspondence was unknown, or rather, did not exist before it emerged in the 1980s. To differentiate between ‘new’ money and ‘old’ money, Rolex + bmw might be contrasted to Patek Philippe + Audi ­(Haram, 1994). As noted, such unities change over time, but when they are operative, they appear to be natural and more or less self-evident. In reality, the claim that things go together because of a kind of consistency is a type of empirical critique of economics in which the focus tends to be on product substitutability. Here we must regard consumption as a knowledge system, as proposed by Douglas and Isherwood (1996), and further as an activity in which it is possible to fail or succeed. Indeed, we all know this on a practical level in our daily lives. Most of us do not wear red sneakers with a dark business suit. Such knowledge has not always influenced our research, however. For environmental aspects of consumption, Diderot unities and Diderot effects are relevant for explaining levels of consumption. In certain situations, as in the Lessing quote, Diderot unities act as a constraining force on consumption; in others, as in the original Diderot essay, the Diderot effect is the driving force behind a massive product replacement. Shopping as Caring Even more than the term consumption, the term shopping triggers a set of very simplistic responses. Instead of thinking of everyday provisioning, people regularly tend to imagine:

148

Strandbakken

… an extreme form of shopping, an over-the-top activity devoted largely to indulging itself. The core connotations of this act are firstly materialism—that is an obsessional desire to buy things per se and secondly hedonism—that is a highly self-indulgent form of pleasure and leisure. miller, 1998: 68

Daniel Miller’s study of everyday shoppers in North London, with a main chapter semi-provocatively titled ‘Making love in supermarkets’, finds that on the contrary, everyday shopping is largely about caring for family, maintaining social relations and attempting to influence loved ones in certain beneficial directions. Housewives attempt to persuade their spouse to dress better or their children to eat healthier food. This perspective is far removed from the ‘traditional’ or tabloid view of shopping, and it probably also demonstrates the irrelevance of economists’ theories of motivation. Again, experience shows that consumption is essentially a social activity. ‘The theory claims that this shopping can be understood as a devotional rite’ (Miller, 1998: 9). Conclusion I have highlighted a small number of contributions from the sociology of consumption tradition to show that consumption is a thoroughly social/cultural phenomenon. By attempting to reduce consumption practices to individual choices of isolated substitutable products, economists sacrifice knowledge for clarity, simplicity and the employment of macro models. The sociology of consumption addresses a field that is blurred and confusing, with consumers not only expressing love and loyalty and defining their identities but also providing all of the unexciting necessities such as bread, potatoes, electricity and gasoline. Consumers are not irrational, but their rationality depends on context and on more complex values than price and quality, which again is not a claim that price and quality are not important. However, I am claiming that the study of consumption and lifestyle is interesting because of its complexity and its multi-tangled relations with all aspects of human existence. References Aron, R. (1991) [1965]. Main currents in sociological thought I. Montesquieu, Comte, Marx, Tocqueville. The sociologists and the revolution of 1848. London: Penguin.

Study of Consumption in Sociology—Beyond Utility Theory

149

Bataille, G. (1987). Eroticism. London: Marion Boyars. Bataille, G. (1988). The accursed share. New York: Zone Books. Baudrillard, J. (1980). For a critique of the political economy of the sign. London: Sage. Bell, D. (1973). The coming of post-industrial society. A venture in social forecasting. New York: Basic Books. Bourdieu, P. (1984). Distinction. A social critique of the judgement of taste. London: Routledge and Keegan Paul. Douglas, M. and Isherwood, B. (1996) [1979]. The world of goods. Towards an anthropology of consumption. New York: Routledge. Durkheim, E. (1979). The division of labor in society. New York: Free Press. Forty, A. (1995). Objects of desire. Design and society since 1750. London: Cameron Books. Galbraith, J.K. (1958). The affluent society. New York: Mentor. du Gay, P., Hall, S., Janes, L., Mackay, H. and Negus, K. (1997). Doing cultural studies. The story of the Sony Walkman. London: Sage. Giddens, A. (1990). The consequences of modernity. Cambridge: Polity Press. Giddens, A. (1991). Modernity and self-identity. Self and society in the late modern age. Cambridge: Polity Press. Gronow, J. (2015). “Main currents in recent consumer studies”, Eds. P. Strandbakken and J. Gronow. The consumer in society. Oslo: Abstrakt Forlag. Gronow, J. and Holm, L. (2015). “Explaining consumer choice: a critique of the theory of planned behavior”, Eds. P. Strandbakken and J. Gronow. The consumer in society. Oslo: Abstrakt Forlag. Gronow, J. and Warde, A. (eds.). (2001). Ordinary consumption. London and New York: Routledge. Haram, E. (1994). Luksusforbruk. En analyse av norske luksusforbrukere i 90-årene. ­[Luxury consumption. An analysis of Norwegian luxury consumers in the 90s]. sifo arbeidsrapport 4/1994. Lysaker. Kellner, D. (1997). “Social theory and cultural studies”, Ed. S. Owen. Sociology after postmodernism. London: Sage. Kennedy, J.F. (1962). Special message to the congress on protecting the consumer interest. March 15th 1962. Washington dc: The American Presidency. Lessing, D. (1997). Walking in the shade. Volume two of my autobiography 1949–1962. London: HarperCollins. McCracken, G. (1988). Culture and consumption. New approaches to the symbolic character of consumer goods and activities. Bloomington and Indianapolis: Indiana University Press. McKendrick, N, Brewer, J. and Plumb, J.H. (1982). The birth of a consumer society: the commercialization of eighteenth-century England. Bloomington: Indiana University Press. Miller, D. (1998). A theory of shopping. New York: Cornell University Press.

150

Strandbakken

Mills, C.W. (1980) [1959]. The sociological imagination. Harmondsworth: Penguin Books Ltd. Nader, R. (1965). Unsafe at any speed. The designed-in dangers of the American automobile. New York: Grossman Publishers. Packard, V. (1960). The waste makers. A startling revelation of planned wastefulness and obsolescence in industry today. London: Longmans, Green & Co Ltd. Simmel, G. (1976) [1903]. “The metropolis and mental life”, The sociology of Georg S­ immel. New York: Free Press. Simmel, G. (1904). “Fashion”, International Quarterly. 10(1904): 130–155. Sontag, S. (2002) [1978]. Illness as metaphor + aids and its metaphors. London: Penguin Books. Stø, E., Throne-Holst, H., Strandbakken, P. and Vittersø, G. (2008). “Review: a multidimensional approach to the study of consumption in modern societies and the potential for radical sustainable changes”, Eds. A. Tukker, M. Charter, C. Vezzoli, E. Stø and M.M. Andersen. System innovation for sustainability 1. Perspectives on radical changes to sustainable consumption and production. Sheffield: Greenleaf. Strandbakken, P. (2007). Produktlevetid og miljø—muligheter og hindringer for en refleksiv økologisk modernisering av forbruket; en teoretisk og empirisk undersøkelse. [Product durability and the environment—conditions for a reflexive ecological modernization of consumption: a theoretical and empirical exercise]. sifo fagrapport 7/2007. Oslo. Strandbakken, P. and Heidenstrøm, N. (2013). Hinsides symbolverdi. Materialiteten i forbruket. [Beyond symbolic value. The materiality of consumption]. Oslo: Novus Forlag. The Cassel Thesaurus. (1998). Ed. M. Manser. London, Wellington House: Cassel. Trentmann, F. (2016). Empire of things. How we became a world of consumers, from the fifteenth century to the twenty-first. London: Allen Lane. Veblen, T. (1994) [1899]. The theory of the leisure class: an economic study of the evolution of institutions. New York: Macmillian. Weber, M. (1978). Economy and society. An outline of interpretive sociology. Berkeley: The University of California Press. Williams, R.H. (1982). Dream worlds: mass consumption in late nineteenth century France. Berkeley: University of California Press.

part 3 Social Structure, Organizations and Institutions



chapter 9

Social Structure John Scott Introduction Social structure is well-established as one of the most useful and enduring concepts in sociological analysis. Only ‘culture’ has, perhaps, rivalled it as a central explanatory principle in social investigation. Some have questioned this reliance on structural explanation and have posited an opposition and conflict between ‘structure’ and ‘action’ as explanatory principles and, in some cases, to depict these as mutually exclusive paradigms of research. My position is that while social structure cannot explain everything, this does not mean that the concept must be abandoned. Social structure must be seen as one among a number of basic sociological concepts that have an important part to play in sociological analysis. These underpin complementary principles of explanation that may differ in relative importance and vary in significance from one situation to another but they are essential elements in viable and comprehensive sociological discourse. Sometimes social structure plays a central role, at other times it may be more subsidiary. Its relevance may be bracketed off for analytical purposes when the concern is with, for example, the more immediate features of social interaction, the impact of cultural discourse, or the effects of environmental conditions. Bracketing-off social structure and treating it as a constant background factor that does not need to be made the immediate focus of an investigation into, say, the interactional constitution of identity, does not make it irrelevant. Similarly, an investigation into changing class structure during industrialisation does not imply the irrelevance of ideas of action, interaction, and agency. Sociologists are engaged in the shared exploration of a complex reality that can be investigated in a number of different, but ultimately complementary, ways. Methodological bracketing is a necessary feature of our disciplinary practice and allows an intellectual division of labour between particular disciplinary specialisms or different explanatory approaches. Rather than engaging in wasteful discussions of action versus structure we should recognise our common participation in a powerful but diverse explanatory project (Scott, 2011). This is not simply an epistemological claim about scientific discourse. It is also an ontological claim that science aims to tell us something about the

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_010

154

Scott

­ ature of the world. We cannot follow Boudon (1968) and see social structure n as a theoretical fiction for making empirical predictions. Social structures are real constituents of the external world that is being investigated. The varying ways in which social structure has been conceptualised are alternative ways in which theorists have grasped that independently existing reality. An account of social structure must therefore begin with a discussion of the nature of theoretical investigation and the ontological status of theoretical concepts.

A Realist Approach to Basic Concepts

It was Max Weber (1904) who most clearly formulated the implications of Kant’s demonstration of the active role of the human mind in producing knowledge. Kant argued that human observers know the world only through concepts that organise impressions received through the senses. Noumenal entities and mechanisms are not—contra Spencer (1862)—‘unknowable’, but can be known only through human sensibility and conceptualisation. Weber showed that theoretical concepts depend upon the standpoint from which the sociologist or the historian approaches reality. Social scientific concepts are rooted in the particular cultural values that constitute a standpoint and provide a perspective on the world. Because the world as it exists independently of human experience can never be known without the use of value-relevant concepts, all scientific knowledge expresses a partial, one-sided experience of the world. This construction of experience means that knowledge has an indeterminate and uncertain relationship to the noumenal reality that exists independently of human observation and experience. The fact that we can only ever know the world as it seems to be from the particular cultural and historical location seems to imply an extreme relativism in human understanding. Weber argued, however, that it is possible to devise logical and precise techniques of objective description that ensure that empirical reports are technically accurate and reliable and so have a known and determinate relation to the observable events. For this reason, objective and reliable knowledge can be achieved relative to a particular value-relevant starting point. The same degree of certainty cannot, for Weber, be established about the relationship between empirical descriptions and the noumenal entities responsible for them. Talcott Parsons’ (1937) discussion of this Weberian methodology further tempered the relativistic implications of Weber’s position. Following Alfred North Whitehead (1926), he saw all empirical knowledge as relative to conceptual frameworks established through the ongoing work of a scientific

Social Structure

155

c­ ommunity. Concepts are not tied to arbitrary individual values but to the more general conceptual systems built up over a series of interrelated studies. These concepts are mutually supportive and the accumulation of empirically objective claims within an established conceptual framework warrants an assumption that the theoretical objects identified do objectively grasp aspects of the noumenal reality. Parsons referred to his methodology as ‘analytical realism’ in order to stress the greater certainty it accords to scientific knowledge of the external world. This argument has been significantly clarified in the ‘critical realism’ of Roy Bhaskar (1975, 1979; see also Harré and Madden, 1975). Bhaskar argued that culturally situated and historically changing scientific concepts produced in scientific communities can provide objectively true representations of the ‘intransitive’ noumena and the mechanisms through which they operate. Observations of the external world are poor indicators of noumenal entities because observations in real situations reflect the complex interplay of numerous unknown entities. These noumenal entities interfere with one another, each modifying the effects of the others. Observable events are ‘out of phase’ with the operations of any particular noumenal entity under investigation. Bhaskar argues that scientific understanding can be advanced through investigative methods that establish the conditions under which the effects of a postulated noumenal entity can be observed in isolation from those of others. Experimental and comparative investigation that allows appropriate methodological controls makes it possible to test scientific claims. Where the hypothesised effects are actually observed, the reality of the postulated entity can be accepted: the transitive concept can be recognised as having grasped a valid feature of the noumenal world. The development of scientific conceptualisation through empirical experimentation and testing brings about a gradual but definite growth in understanding. The pragmatic test of practical adequacy gives the confidence that a science has grasped something that is true of this particular aspect of reality. This implies that knowledge is not simply a representation of the world but also ‘a means for doing things in it’ (Sayer, 1992: 48). If a theory successfully informs practices it is, to that extent, ‘true’: it describes structures and mechanisms that adequately mimic the noumenal entities. ‘To be practically adequate, knowledge must generate expectations about the world and about the results of our actions which are actually realized’ (Sayer, 1992: 69). A theoretical structure ‘works’ and so is valid by allowing us to anticipate or predict actual events. There is an important difference between natural science and social science. While both stand apart from the noumenal entities studied, which are

156

Scott

t­herefore independent of the scientist, the objects studied by social science are not completely independent of all thought. The noumena of the social world depend for their existence on human intentionality and on the ability of human participants to conceptualise the world around them. However, actions have unintended consequences and so the entities and the mechanisms that people produce will invariably ‘escape’ their grasp and acquire properties that may be unrecognised and so unknown until subjected to scientific investigation. The social scientist must identify and understand the various ‘value relevant’ constructions that participants place on their actions but must also grasp the ways in which the variant standpoint-dependant perspectives on their common situation tessellate into a more complex system of representations that is a crucial feature of the ways in which their actions produce an independently real social world. This process, described by Karl Mannheim (1929) as producing a dynamic synthesis, does not involve a mechanical combination of viewpoints or an assumption that all points of view are equally valid. Rather, it involves reflexivity towards one’s own standpoint and a sensitivity towards participant others that recognises the authenticity of the conclusions they draw from their standpoint. The task is to construct an argument that recognises and incorporates all viewpoints but that does so reflexively—recognising also their embedded and embodied foundations—and with an overriding commitment to attaining a ‘synthesis’ that can withstand the ongoing tests of practical adequacy (Letherby, Scott and Williams, 2013). These methodological reflections establish the basis for the conceptualisation of social structure. It is legitimate to postulate social structure as an independently existing feature of the external world. Although conceptions of social structure are theoretical objects they are hypothesised real structures. This is not, of course, to say that all theoretical accounts that utilise a concept of social structure must be regarded as true. The test of practical adequacy is paramount. However, when many theories utilising social structure have passed a test of practical adequacy, it can be concluded that the concept itself has the status of a basic and necessary sociological concept. This is precisely the situation today. Theories of class structure, opportunity structure, kinship structure, corporate structure, and so on, attest to the status of social structure as a fundamental element in sociological analysis.

A Genealogy of Social Structure

The idea of social structure emerged as a central element in the conceptual framework constructed in the self-conscious development of sociology. The

Social Structure

157

new discipline was seen as a generic social science with specific theoretical objects irreducible to those of other sciences dealing with human phenomena at the mental or biological levels. It is striking, however, that very few attempts have been made to define and articulate the concept of social structure with any clarity or precision. The concept has been used as a taken-for-granted device rather than as an object of sustained discussion. Users have not spelled out the specific ontology of social structure. The analytical meaning of social structure must, therefore, be reconstructed from an exploration into the genealogy of the concept. Central to the development of sociology in France and Britain were Auguste Comte (1851–1854) and Herbert Spencer (1873–1893). They saw social structure as integral to the specific forms of complexity identified in social phenomena. This complexity resides in the organisation or arrangement of elements that are more fundamental than the social whole. In order to conceptualise this complex object and its ‘structural’ properties, Comte and Spencer used the metaphor of the social organism. In doing so, they were not claiming that the objects of the new social science depended on purely material linkages like those of the organic structures being investigated in the new science of biology. Rather, they showed that both biological organisms and human societies could be seen as ‘organised’ and that the properties of both biological and social organisms could be seen in terms of the more general processes of organisation and complexity being uncovered in the developing biological sciences. Comte and Spencer understood social ‘structure’ to refer to the arrangement of the material that forms the parts of a ‘system’. These organised parts work together, or ‘function’, to produce the organised whole. The structure of a society or social group is the skeletal ‘anatomy’ of a system, joining individuals into social groups and comprising the framework through which their activities connect them into a coherent system. A system can be described in terms of its ‘physiology’ of functional flows and the equilibrium or disequilibrium among these flows. Seeing social phenomena as organisms or systems was also the basis on which Comte and Spencer investigated change over time, much as the growth of an individual biological organism and the evolution of a species can be studied by a biologist. Biology begins from the innate properties of individual organisms co-existing in a material environment in order to construct accounts of organisms and populations developing over time. Sociology begins from the psychological properties of these individuals co-existing in a cultural environment in order to construct accounts of social structures and systems developing over time. Only Spencer gave any detailed attention to social structures and their variation from one society to another. The central elements in any social structure, he argued, are ‘social institutions’. These are the regularised, routinized, and

158

Scott

recurrent practices involved in major social functions: the domestic, ecclesiastical, ceremonial, political, economic, professional, and industrial institutions variously involved in the sustaining, regulating, and distributive ‘organs’ of a society. Institutions produce gradually more complex societies as they become ‘compounded’ into more extensive organs that integrate the activities of those involved in more differentiated activities. Spencer’s discussion of social institutions, however, remained largely descriptive, taking for granted—and so failing to spell out—their character and mode of operation. Marx (1844)—who did not, of course, define himself as a sociologist—was among the first to pursue the idea of social structure. While he took a somewhat different view of the social system, he nevertheless shared many ideas with Comte and Spencer. The particular theories of wages, prices, and profits that Marx proposed are not important here. Rather, it is the underlying system of structural concepts that must be considered. Rather than emulating the achievements of biological science, Marx drew on Hegelian philosophy and its idea of the ‘totality’. The concept of a social totality replaced that of the social organism as a way of comprehending the organised character of social and economic activity in a social formation. Hegel had seen the social spirit or mind as the key factor in social development that shapes all human relations with nature. Marx gave greater attention to nature and the material aspects of life and saw the social totality as combining both spiritual and material elements to jointly shape people’s relations to the world (Marx and Engels, 1846). Where Spencer came to see social structure as a structure of institutions, Marx emphasised its character as a structure of relations. Within the social totality, Marx gave primacy to the ‘real relations’ through which people enter into a material interchange with nature, especially as they engage in the production of their means of subsistence. At the heart of the social relations of any society, then, are its ‘relations of production’. While these have a directly material aspect in the physical arrangement of the forces of production—as fields, mines, factories, raw materials, machinery, and the distribution of labour power—they are to be understood as fully social relations that tie people into structures of interdependence. These social relations comprise the ‘economic structure’, to which Marx related the cultural forms of consciousness, or ideology, that organise and express the legal and political institutions of the ‘superstructure’ (Marx, 1859). The social relations of the economy are expressed in legal relations of ownership and employment, but it is the ability to exercise powers of control, rather than merely legal rights that constitutes them as ‘real’ relations. The real relations of a society are relations of power, of effective control and lack of control over the means of production, including labour power, and that produce ‘deeper’ and more extended ‘class’

Social Structure

159

relations of exploitation and oppression (Cohen, 1978: 63–65, 216 ff). Marx saw real economic relations as the principal determinants—the ‘base’—of the institutional relations that comprise the superstructure. The latter have, nevertheless, a secondary, reciprocal effect on the economic relations, and comprise their necessary but not their sufficient conditions. Where Comte and Spencer emphasised the relatively harmonious interdependence of parts that brings about a state of integration, Marx emphasised the ways in which the parts of a system could work against each other and undermine the stability, or equilibrium, of the totality. The key driving force in social change, Marx held, is the emergence of ‘contradictions’ between the structural elements that form the base. A contradiction exists where the logic of development inherent in the economic relations is incompatible with that behind the legal relations of ownership. Marx argued that such contradictions can be resolved only through changes in the legal relations as a result of the compelling drive of the economic forces. The contradictions within the economy give rise to a divergence of interests and conflicting action orientations and it is the conflict so produced—class conflict—that brings about revolutionary change rather than mere organic growth (Parkin, 1972). Marx’s views on social structure must be seen as complementary to those of Comte and Spencer. Stressing real relations and institutional relations, his work showed an advance on Spencer’s reliance on institutions alone. Marx’s idea of structural contradictions within a social totality complements Spencer’s view of the structural integration of a social system. This was explicitly recognised by the Marxist theorist Bukharin (1925), who made it the basis of his ‘system of sociology’. This emerging view of social structure was taken further by Durkheim. Pursuing the ideas of Comte and Spencer, he took account of developments in biology to incorporate an awareness of contradiction alongside harmony. In doing this he introduced concepts similar to those of Marx, though the two differed in the causal weight they assigned to these factors. This similarity is hardly surprising in view of the fact that Durkheim’s key contributions to social theory followed his own encounter with German idealism. Following his visit to Wilhelm Wundt and his reading of Wundt’s Ethics (1897), a study heavily influenced by Hegelian ideas, Durkheim was converted to the view that social entities could be seen as ‘organisms’ with a reality beyond that of their individual participants. Taking the idea of the social organism as formulated by Albert Schäffle (1875–1880) much more seriously than other French social theorists, Durkheim sought acceptable ways of conceptualising the ontology of the social that combined these ideas with Comtean positivism. He came to recognise that sociology deals with social systems that

160

Scott

have emergent properties sui generis and that can be described in terms of a distinct body of ‘social facts’. Durkheim saw social structures manifesting themselves in a number of ways. First he identified what he referred to as social morphology. This idea was introduced by Durkheim to conceptualise the links among the parts of a social organism as corresponding to the physical links that form the bony anatomy of a biological organism. Like Marx he looked to the material aspects of human existence but highlighted their spatial distribution. The distribution of a population and the roads, railways, and buildings that connect them into social groups comprise the morphological structure of a society. This material morphology, Durkheim held, is the ‘substratum’ of social life but is not a simple determinant as it is also shaped by human social activity. Social morphology is both a condition for and an outcome of social activity. As a condition it allowed Durkheim to comprehend the influence of the material environment and the materiality of the human body on social life. He could incorporate some of the views of Le Play (1855) and his followers, articulate the core ideas that were being developed by Vidal de la Blache (expressed later in his 1926), and lay the foundations on which Febvre (1922) would develop his model of ‘possibilism’ (see Andrews, 1993). Social morphology is, then, Durkheim’s first formulation of the idea of social structure. Powerful structural explanations can begin with a mapping and analysis of the morphology (Halbwachs, 1938). Durkheim was clear, however, that social structure comprised something more than material morphology. Physical connections and linkages are traces of the social relations that underpin them. As channels of communication they are the carriers and containers of the meanings and sentiments that comprise the proper sphere of the social. Societies are ‘moral’ phenomena with a reality sui generis, and it is this that led him to reject the strong material determinism that he saw in the geography of de la Blache. To identify this moral reality he turned to the exploration of religion, myth, and beliefs and their constitution as ‘collective representations’ and forms of a conscience collective. The representations that individuals hold and share with others are the basis of their social relations. These relations are projected out into the material world, but only partially and imperfectly. The structure of social relations itself cannot therefore be reduced to its material expression. If, then, material morphology is the ‘substratum’ of social life, representations and social relations, as moral phenomena, comprise the social structure proper. For Durkheim, the social realm comprises collective representations, the social relations that they inform, and the material morphology that is both the condition and the outcome of the representations and relations. All are conjoined inescapably in the social entity that Durkheim called ‘society’.

Social Structure

161

Durkheim was, of course, elaborating on a key insight of Spencer, who held that the ‘tissues’ through which individuals are connected into a social organism are not material, as is the case with a biological organism, but are linguistic channels that allow the communication of information from one individual to another. Social structure exists, therefore, only in and through communication. Though neither Spencer nor Durkheim spelled this out, their arguments imply that the nature of social structure is to be found in the ideas and meanings communicated among individuals and their effects on the actions of those individuals. While its effects can be observed in the material and morphological conditions that are its external expression, social structure itself has a virtual existence so far as empirical observation is concerned. Allied to this view of the virtuality of social structure was the Durkheimian contrast between ‘normal’ and ‘pathological’ system states. The normal state is an equilibrium state in which the parts function together in mutually supportive ways. This state is normal only in so far as such a system shows no tendency to change or to break down. Pathological states are those where ‘dysfunctional’ processes threaten the stability of the system and so make possible its dissolution. Durkheim’s views on this subject, as shown by Merton in his classic paper (Merton, 1949), provide a partial reconciliation of those of Spencer with those of Marx. Durkheim, however, limited his attention to how a lack of integration tends to break down the bonds that tie individuals together: when social structure dissolves and individuals are left in a Hobbesian state of anomie and egoism. This was the view elaborated by Parsons (1937, 1951), who saw ‘the structure of social action’ rooted in the values and norms of cultural systems produced and reproduced in human communities and through which individual actions are organised and constrained into definite patterns. This view of social structure as a system of ‘institutions’ implied a one-sided ‘consensus’ view of society, in so far as social structure was seen as the primary factor in ensuring social stability. It was recognised by Lockwood (1956, 1964), however, that the complementarity of Durkheim and Marx, already noted by Merton, gave Parsonian theory the potential to incorporate an analysis of contradictions as well as harmonious interdependencies. A system theory that combines the analysis of both institutional integration and institutional contradiction with an awareness of the material ‘substratum’ in which they are expressed, is able to show the limits and opportunities open to individuals and social groups and so can articulate the relations of solidarity and schism in which they are involved (Lockwood, 1992). Parsons’ reading of Durkheim became mainstream sociology through the 1950s and 1960s. This depicted the arrangement of social relations among individuals and groups, both harmonious and contradictory, as constituting the

162

Scott

social structure. It was also the most influential approach to social structure in social anthropology, finding its initial expression in the work of RadcliffeBrown (1940) and in Nadel (1957). The duality in its theoretical foundations is apparent in the much-discussed contrast between the ‘consensus’ view of Parsons (1951) and the various forms of ‘conflict theory’ (Coser, 1956; Rex, 1961, 1981). These theorists recognised two levels of social structure, which RadcliffeBrown described as the actual interactions of concrete individuals and the underlying ‘structural forms’ of relationship that shape them. Parsons presented this as a relationship between structure at the level of system integration (interdependence, harmony, and dysfunction) among institutions and structure at the level of social integration (consensus and conflict) in social relations. The importance of actual relations and interactions was especially recognised in the work of symbolic interactionists and found its clearest expression in the concept of the interaction order (Goffman, 1983). For many interactionists, however, the reality of interaction was allied with a rejection of the reality of the institutional order. The social structure of institutions, as Blumer (1962) famously put it, is simply an organising ‘framework’ held in the minds of individuals and has no independent existence apart from their interaction. Such a view, however, implies no significant departure from the mainstream, which already recognised the virtual reality of the institutional order, understood as a cultural—and so communicative—phenomenon. It was within social anthropology that the first major break with this view developed. Lévi-Strauss (1953) took a similar view to Radcliffe-Brown in his early work but combined this with linguistic models that led him to a novel and distinctive conception of the ‘underlying’ social structure. Durkheimianism provided the comprehensive starting point for Lévi-Strauss and he used ideas from geology and psychoanalysis to reconstruct the Durkheimian view. He discovered an appreciation of the ontological depth of social structure and came to emphasise the need to dig down below the ‘surface’ structure of the institutional order to the ‘deeper’ structural properties of the social system. He found in Marxism a similar idea of the relationship between superstructure and infrastructure, but Marxism also highlighted the importance of contradiction and opposition at both ontological levels. Lévi-Strauss saw social structure as the system of oppositions that are the basis of the logically possible permutations underpinning observable social and cultural relations and that give them their form. These structural oppositions and their principles of combination are the fundamental complement of the institutional order. Social structure, for Lévi-Strauss, then, is the ‘grammar’ of social life: structure underpins actions in the same way that the grammar of

Social Structure

163

a language underpins the speech of its users. Like the rules of grammar it comprises unconscious organising mechanisms that Lévi-Strauss saw as invariant properties of the way in which the human brain operates. This was the first challenge to the mainstream and became particularly influential during the 1970s (Glucksmann, 1974), with many of its most enthusiastic advocates seeing ‘structuralist’ theory as a fundamental alternative to institutional theory. Zealous elaborations of these structuralist ideas gave birth to a variety of positions that came to be seen as ‘post-structuralist’. Foucault (1972), Deleuze (1969), and Latour (1993), to varying degrees, developed such positions, though they may not always have regarded themselves as ­post-structuralists. They did not completely abandon the idea of arrangement or organisation that had been central to mainstream views of social structure. In place of ‘structure’ they posited ‘assemblies’ and in place of holistic integration they posited deep interweaving ‘rhizomes’ that appear on the surface as dispersed and discontinuous arrangements or ‘archipelagos’ of assembly. What these theorists had in common was a desire to move ‘beyond’ the holism they identified in the concepts of system and totality and that they saw as having been continued in the work of Lévi-Strauss. This critique was applied also to the forms of structuralist Marxism found in Althusser (1965; Althusser and Balibar, 1970) and his followers (for example, Poulantzas, 1973, 1975). The oppositions that operate in human communicative discourse must be seen as generating diversity and plurality in social relations and activities. This structured diversity precludes the use of such holistic concepts as ‘social structure’, they argued, as this rests on an assumption of complete integration. These theorists took to its logical extreme the position of the earlier conflict theorists. Further developments of the approach initiated by Lévi-Strauss have been proposed by Pierre Bourdieu, and Anthony Giddens. These writers, in different ways, have explored the implications of the view that the deep levels of social structure must be seen in relation to the mental operations that are involved in socially structured actions. Bourdieu (1972) utilised the ideas of Lévi-Strauss and Durkheim to explore the social morphology of Kabyle society and the structuring of social distance through opposing circuits of economic and cultural capital. Following LéviStrauss, he saw structural principles as embodied in human habitus, though he eschewed the biological determinism of Lévi-Strauss. Instead, he took up a suggestion of Durkheim’s nephew Marcel Mauss (1934) and saw bodily habitus as comprising dispositions and tendencies of action that both shape and are shaped by location within the systemic circuits of economic and cultural capital.

164

Scott

Giddens (1984) conjoined Lévi-Strauss’ analysis of structural principles with ethnomethodological accounts of rule following. He argued that the mainstream view of social structure focused on the ‘system’ of social life, and advocated the view that social structure per se comprises the virtual structure that results from following the embodied rules that generate ongoing social practices. Giddens actually referred to ‘rules and resources’ but, as I have argued elsewhere (Scott, 2007), his reference to ‘resources’ was actually a reference to the rule-bound usage of resources. Thus he combined the insights of LéviStrauss with much of the conventional account of social structure to produce a less deterministic understanding of human action and agency that owed much to Garfinkel’s (1967) account of everyday ‘methodologies’ and Cicourel’s (1972) view of generative structures of rules.

Analytics of Social Structure

My reading of the theorists considered in the genealogical development of the concept of social structure suggests that social structure is a multi-faceted concept with a complex ontology. It comprises three interrelated elements that may be termed institutional, relational, and embodied structure (see López and Scott, 2000). The cultural ideas and values of a population—transmitted from person to person through communication—comprise normative patterns that form the framework of social life. As clusters of norms or rules concerning marriage, property, citizenship, etc., these patterns are referred to as institutions, the established and expected ways of thinking, feeling, and acting. While institutional norms are held in individual minds, they are extra-individual by virtue of being shared. As such, they are objective and constraining factors in social life and are regarded as authoritative and compelling because seen by each individual as coming from ‘outside’. Institutions define standardised practices and to the extent that people act in conformity with their institutions their actual practices will correspond to the structural form. However, the ‘fully institutionalised’ social relation is simply one case of social structure. Socialisation and moral commitment are never perfect. Deviance and conflict are frequent, especially where there is no consensus or there is a pluralisation of values. Thus alongside of, and partly out of synchronisation with, the institutional structure there is a relational structure. Repeated social interactions exhibit certain formal characteristics that have a reality sui generis and that can, therefore, constrain individual acts

Social Structure

165

(Martin, 2009: 3). This is a pattern of recurrent social relations, ranging from the i­ nteractional forms studied by Simmel (1908) to more extensive patterns of inequality, heterogeneity, and differentiation that characterise cities, nations, and world systems. The virtual existence of institutional and relational structures rests upon a third aspect of social structure. This is the embodied structure that comprises the memory traces and behavioural dispositions of the habitus. E ­ mbodied structure is generative, in the same sense as the grammar embodied in linguistic competence. Institutional and relational structures are produced and reproduced through embodied structures, and embodied structures are ­sedimentations of institutional and relational structures. Institutional and relational patterns exist virtually: they have no separate physical existence but appear to do so as a result of the application of ‘programmed’ or embodied rules. The combination of institutional structure, relational structure, and embodied structure in a comprehensive conceptualisation of social structure is a clear implication of the initial insights of Spencer and Durkheim that social life is organised through the linguistic communication of mental representations. The interdependence of these aspects of social structure can be seen in a simple example of the purchasing of food in a supermarket. Cultural representations defining the social positions of shop assistant and customer, which themselves rely on representations of money, trading, fashion, and so on, involve rules that provide each participant with expectations as to how each will relate to the other in their encounters. In this way, they constitute an institutional structure of ‘shopping’ and a corresponding relational structure of buying and selling. Focused in an interaction order of face-to-face relations the practice of shopping also involves deeper relations between retailers and consumers, bankers and manufacturers, and importers and producers. In this way there are extended, non-institutionalised structural relations between, for example, European shopkeepers and African peasant farmers. Rules exist as memory traces within individual minds and as behavioural tendencies: they are embodied. Rule following and performance, may sometimes be conscious and deliberate, but are, more typically, the result of the unconscious or habitual enactment of internalised dispositions. We communicate linguistically in conversations without consciously applying rules of grammar and punctuation in each sentence we speak. Rather, we simply speak, having learned, internalised, and habitualised the rules that may be used to retrospectively reconstruct the grammaticality of our utterances. Embodied rules ‘structure’ action in the same way that the rules of grammar structure our speech. The causal properties and powers of an institutional

166

Scott

and relational structure can be explained by the rule-governed organisation of their constituent elements (Elder-Vass, 2010: 23). Thus, we undertake our daily ­shopping without consciously consulting rules of appropriate behaviour towards those who fill the shelves or operate the tills in a supermarket. We do not even consider the rules that allow us to recognise an object as a ‘till’ or that allow the shop a­ ssistant—but not the casual passer-by—to provide us with food in exchange for the swipe of a small piece of plastic. The rootedness of social structure in cultural representations underpins the argument, popularised by Giddens, that social structures must be seen as having a virtual existence and as sustained in the mentality of individuals yet expressed in the materiality of objects. This is, however, a difficult idea to understand. If social structures are not tangible physical, material objects but objects of thought, how can they be said to have any reality apart from the mind of the individual thinker? Durkheim was able to explore this question only after his research visit to Germany. Before this visit he had come to follow Espinas (1877) in recognising the existence of consciences sociales as collections of individual minds, but he was not yet able to argue for the reality sui generis of such phenomena. During his year in Germany he came to recognise the significance of the Volksgeist that Schmoller, Wagner, and von Savigny had seen as organising economic and juridical phenomena. His time with Wundt provided him with the tools that he needed to reformulate this idea as a way of grasping collective moral phenomena. Durkheim concluded that objective spirit—which Edward Tylor (1871) and the British school of anthropology had started to call ‘culture’—exists in the ideas and sentiments shared within a population through being passed from one individual to another and so from one generation to the next. This led him to reassess the organismic arguments that he first outlined himself in The division of labour in society (Durkheim, 1893) and then formalised in the Rules of the sociological method (1895). In these books Durkheim argued that the conscience collective, collective representations, and other ways of acting, thinking, and feeling are ‘external’ to individual people and constrain their behaviour. They are held in common by a plurality of individuals and are external to them because they are often manifest in observable external traces or markers of physical morphology and material artefacts and, more importantly, they form collective ‘currents’ that flow through the society and so constrain individual behaviour. Durkheim’s opposition to Tarde’s (1890) work on ‘imitation’ led him away from exploring these currents and their ‘externality’ in any detail and he

Social Structure

167

r­ emained content with a view that psychological currents are inherent in the reactions and sanctions of others to individual acts and in the internalised ­anticipations of these reactions. It was Tarde who stressed the need to focus on communication among individuals and the consequent alteration of ideas and sentiments held in each individual mind as a result of this communication (Tarde, 1894). In the work of Cooley this idea was separated from a narrow reliance on imitation and developed into a powerful analysis of symbolic communication. Cooley held that the interactional and communicative process exists collectively as ‘society’ and distributively as ‘individuals’ (Cooley, 1902: 2). Through communication the ideas and sentiments of one person are connected with those of others and so ‘act and react upon them to form a whole’ (Cooley, 1909: 10). The point was reiterated in Ellwood’s statement that a society ‘is a group of individuals carrying on a collective life by means of mental interaction’ (Ellwood, 1912: ix, 13). This whole is ‘the larger mind’, produced unintentionally and growing more differentiated: ‘An institution is simply a definite and established phase of the public mind’ (Cooley, 1909: 313). Major institutions ‘are the outcome of that organisation which human thought naturally takes on when it is directed for age after age upon a particular subject, and so gradually crystallises in definite forms’ (Cooley, 1909: 313). Thus, ‘the institution exists as a habit of mind and of action’ (Cooley, 1909: 314) and in shaping actions it shapes social relations. This is why social structures can be said to exist as real objects apart from individuals in the same sense that pictures exist apart from the individual points of colour through which they are produced or musical compositions exist apart from the sounds made by the individual instruments. Cooley concluded that ‘society is, indeed, literally, a work of art’ (Cooley, 1909: 21), a kaleidoscopic arrangement of persisting and changing patterns of cultural representations. Social structure is produced through communication and the means of communication that go beyond face-to-face communication—writing, printing, and electronic communication—allow the building of larger and more complex social structures. Virtual structures have a reality grounded in the communicative flow of discourse within a population. A final point is to clarify the solidity of social structure. The structure of a system—the arrangement of its matter, including its virtual arrangement—is what maintains the distinction between system and environment and gives the system its strength. A piece of granite, for example, has a strength that allows it to persist for millions of years, despite the fact that it is simply an arrangement of the same elementary particles that can also be arranged into a

168

Scott

cloud of vapour through which the rock can pass with ease and that may persist for only a few minutes. The strength or solidity of a system is generated by, the particular ways in which its parts are arranged. Maintenance of a boundary, for example, depends on the forces acting on it from its e­ nvironment ­being matched by ­internal processes that generate the appropriate ‘tension’ and ‘compression’ forces (Gordon, 1978). In the case of social systems it is forces such as inducement and constraint that can act on individuals to hold them in systemic patterns of arrangement. If the equilibrium of forces is not achieved, a state of disequilibrium exists as forces generate stresses and strains that disturb the arrangement of the elements and so fracture the structure once the elasticity of its boundary has been exceeded. For this fracturing to be avoided, the internal structure of a system must be such that it generates the necessary forces to meet those from outside or to move a system back into equilibrium. There is, of course, nothing inevitable about the maintenance or restoration of equilibrium. System theory has explored the processes through which a system state can be controlled and its equilibrium restored. If a system has such ‘homeostatic’—structure maintaining—processes of monitoring and feedback it is able to persist or to develop in stable ‘structured’ ways (Buckley, 1967). Archer (1995) has explored this structural development as a process of ‘morphogenesis’. Conclusion Social structure is one of a number of basic sociological concepts and grasps the operation of certain ontologically real mechanisms through which social life is organised. Social structures are formed through the articulation of institutional, relational, and embodied structures. Social structures thus comprise positions constituted by specific norms and by the relational practices that they inform. Individuals are assigned positions through the rule-governed practices of others and it is knowledge of the rules that allows those individuals to construct a sense of social structure (Cicourel, 1968, 1970, 1972), a m ­ apping of their society that allows them to act effectively as a member of that society. The society itself is produced through the practices of individuals on the basis of their sense of social structure. Only rarely is there either a consensus of individual mappings or a one-to-one relationship between any particular mapping and the actual practices of members. It is the task of the sociologist to articulate the dynamic synthesis of social imagery as an objective theory of social structure.

Social Structure

169

References Althusser, L. (1965). For Marx. Harmondsworth: Allen Lane, The Penguin Press. Althusser, L. and Balibar, E. (1970). Reading capital. London: New Left Books. Andrews, H.F. (1993). “Durkheim and social morphology”, Ed. S.P. Turner. Emile Durkheim. Sociologist and moralist. London: Routledge. Archer, M.S. (1995). Realist social theory: the morphogenetic approach. Cambridge: Cambridge University Press. Bhaskar, R. (1975). The realist theory of science. Leeds: Leeds Books. Bhaskar, R. (1979). The possibility of naturalism. Brighton: Harvester. Blumer, H. (1969) [1962]. “Society as symbolic interaction”, Ed. H. Blumer. Symbolic interactionism. Englewood Cliffs: Prentice-Hall. Boudon, R. (1971) [1968]. The uses of structuralism. London: Heinemann. Bourdieu, P. (1977) [1972]. Outline of a theory of practice. Cambridge: Cambridge University Press. Buckley, W. (1967). Sociology and modern systems theory. Englewood Cliffs: PrenticeHall. Bukharin, N.I. (1925). Historical materialism: a system of sociology. New York: International Publishers. Cicourel, A.V. (1973a) [1968]. “Generative semantics and the structure of social interaction”, Ed. A.V. Cicourel. Cognitive sociology. Harmondsworth: Penguin. Cicourel, A.V. (1973b) [1970]. “Interpretive procedures and normative rules in the negotiation of status and role”, Ed. A.V. Cicourel. Cognitive sociology. Harmondsworth: Penguin. Cicourel, A.V. (1973c) [1972]. “The acquisition of social structure: Towards a developmental sociology of language and meaning”, Ed. A.V. Cicourel. Cognitive sociology. Harmondsworth: Penguin. Cohen, G.A. (1978). Karl Marx’s theory of history. Oxford: Oxford University Press. Comte, A. (2001) [1851–1854]. System of positive polity. Six volumes. Bristol: Thoemmes Press. Cooley, C.H. (1902). Human nature and the social order. New York: Scribner’s. Cooley, C.H. (1909). Social organisation. New York: Schocken. Coser, L. (1956). The functions of social conflict. London: Routledge and Kegan Paul. Deleuze, G. (1994) [1969]. Difference and repetition. London: Athlone Press. Durkheim, E. (1982) [1895]. The rules of the sociological method. London: Macmillan. Durkheim, E. (1984) [1893]. The division of labour in society. London: Macmillan. Elder-Vass, D. (2010). Causal powers. Cambridge: Cambridge University Press. Ellwood, C.A. (1912). Sociology and its psychological aspects. New York: D. Appleton. Espinas, A. (1877). Des sociétés animals. Paris: G. Baillière.

170

Scott

Febvre, L. (1974) [1922]. A geographical introduction to history. Westport: Greenwood Press. Foucault, M. (1972). The archaeology of knowledge. New York: Pantheon. Garfinkel, H. (1967). Studies in ethnomethodology. New Jersey: Prentice-Hall. Giddens, A. (1984). The constitution of society. Cambridge: Polity Press. Glucksmann, M. (1974). Structuralist analysis in contemporary social thought. London: Routledge and Kegan Paul. Gordon, J.E. (1978). Structures: or why things don’t fall down. Harmondsworth: Penguin. Goffman, E. (1983). “The interaction order”, American Sociological Review. 48(1): 1–17. Halbwachs, M. (1960) [1938]. Population and society: introduction to social morphology. Glencoe: Free Press. Harré, R. and Madden, E.H. (1975). Causal powers. Oxford: Oxford University Press. Latour, B. (1993). We have never been modern. Cambridge, ma.: Harvard Universiuty Press. Le Play, F. (1935) [1855]. “Les Ouvriers Européens”, Eds. C.C. Zimmerman and M. Frampton. Family and society. Condensed Translation. New York: D. van Nostrand. Letherby, G., Scott, J. and Williams, M. (2013). Objectivity and subjectivity in social research. London: Sage. Lévi-Strauss, C. (1968) [1953]. “Social structure”, Ed. C. Lévi-Strauss. Structural anthropology. Vol. 1. London: Allen Lane The Penguin Press. Lockwood, D. (1956). “Some remarks on the social system”, British Journal of Sociology. 7(2): 134–146. Lockwood, D. (1964). “Social integration and system integration”, Eds. G.K. Zollschan and W. Hirsch. Explorations in social change. London: Routledge and Kegan Paul. Lockwood, D. (1992). Solidarity and schism. Oxford: Clarendon Press. López, J. and Scott, J. (2000). Social structure. Buckingham: Open University Press. Mannheim, K. (1936) [1929]. “Ideology and utopia”, Ed. K. Mannheim. Ideology and utopia. Ch. 2–4. London: Routledge and Kegan Paul. Martin, J.L. (2009). Social structures. Princeton, n.j.: Princeton University Press. Marx, K. (1964) [1859]. “Preface to a contribution to the critique of political economy”, Eds. K. Marx and F. Engels. Werke. Vol. 13. Berlin: Dietz Verlag. Marx, K. (1970) [1844]. “Theses on Feuerbach”, Ed. C.J. Arthur. The German ideology. London: Lawrence and Wishart. Marx, K. and Engels, F. (1970) [1846]. The German ideology. London: Lawrence and Wishart. Mauss, M. (1973) [1934]. “Techniques of the body”, Economy and Society. 2(1): 70–87. Merton, R.K. (1949). “Manifest and latent function”, Ed. R.K. Merton. Social theory and social structure. New York: Harper and Row. Nadel, S.F. (1957). The theory of social structure. Glencoe: Free Press.

Social Structure

171

Parkin, F. (1972). “System contradiction and political transformation”, European Journal of Sociology. 13(1): 45–62. Parsons, T. (1937). The structure of social action. New York: McGraw-Hill. Parsons, T. (1951). The social system. New York: The Free Press. Poulantzas, N. (1973). Political power and social classes. London: New Left Books. Poulantzas, N. (1975). Classes in contemporary capitalism. London: New Left Books. Radcliffe-Brown, A.R. (1952) [1940]. “On social structure”, Ed. A.R. Radcliffe-Brown. Structure and function in primitive society. London: Cohen and West. Rex, J.A. (1961). Key problems of sociological theory. London: Routledge and Kegan Paul. Rex, J.A. (1981). Social conflict. Harlow: Longman. Sayer, A. (1992). Method in social science: a realist approach. 2nd edition. London: Routledge. Schäffle, A. (1906) [1875–1880]. Bau und Leben des Sozialen Körpen. Four volumes. Tübingen: Laupp. Scott, J. (2007). “Anthony Giddens and cultural analysis: absent word and central concept”, Ed. T. Edwards. Cultural theory: classical and contemporary positions. London: Sage. Scott, J. (2011). Conceptualising the social world. Principles of sociological analysis. Cambridge: Cambridge University Press. Simmel, G. (1968) [1908]. Soziologie: Untersuchungen Über die Formen der Vergesselshaftung. Berlin: Düncker und Humblot. Spencer, H. (1873–1893). Principles of sociology. Three volumes. London: Williams and Norgate. Spencer, H. (1910) [1862]. First principles. Two volumes. London: Williams and Norgate. Tarde, G. (1903) [1890]. The laws of imitation. New York: H. Holt and Co.. Tarde, G. (1969) [1894]. “Sociology, socal psychology, and sociologism”, Ed. T.N. Clark. Gabriel Tarde on communication and social influence. Chicago: Chicago University Press. Tylor, E.B. (1920) [1871]. Primitive culture. Researches into the decvelopment of mythology, philosophy, religion, language, art, and custom. Two volumes. London: John Murray. Vidal de la Blache, P. (1926). Principles of human geography. New York: H. Holt and Co. Weber, M. (1949) [1904]. “‘Objectivity’ in social science and social policy”, Ed. M. Weber. The methodology of the social sciences. New York: Free Press. Whitehead, A.N. (1938) [1926]. Science and the modern world. Harmondsworth: Penguin Books. Wundt, W. (1897) [1886]. Ethics: an investigation of the facts and laws of the moral life. Two volumes. A translated 2nd edition. New York: Macmillan.

chapter 10

The Organization of Action Göran Ahrne

A Conceptual Space: Organizations, Institutions and Networks

In the sociological vocabulary, the term ‘organization’ commonly refers to entities such as business corporations. It sometimes also refers to schools, hospitals or political parties. However, some of these entities are also often referred to as institutions. Moreover, many organizations are also viewed as networks. In the social sciences today, there is confusion about the meanings of organization, institution and network, and these concepts appear to overlap considerably. Organization, institution and network have much in common, and they form a conceptual space denoting structures of human interaction and social relationships. Like networks, organizations are relationships between people or organizations, and like institutions, organizations can produce common patterns of behaviour and shared distinctions with a certain degree of stability over time (Ahrne and Brunsson, 2011). Institution and network have probably been more widely accepted and used among general social theorists than organization, whereas the concept of organization has been strangely neglected in general social theory: ‘organizations are more important than most theories allow, and their importance has not been fully realized’ (Perrow, 2002: 12). The aim of this paper is to present a conceptualization of organization as a distinct way of forming social relationships and to introduce distinctions with specific consequences that differentiate organizations from both institutions and networks. This conceptualization will enable us to grasp how these concepts simultaneously differ from each other while conditioning each other. The uncertainty in the meaning of these concepts may be partially attributable to the processes of transformation among them in real social life. Institutions may become organized and transformed from informal norms to formal rules. Conversely, rules may turn into norms when they become routines and habits and nobody remembers when and why they were established. Moreover, networks may be transformed into organizations when members feel a need to act together on a more permanent basis. One feature that these concepts have in common is that they are generally described and regarded as a type of meso-concept: they are neither micro nor

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_011

The Organization of Action

173

macro, denoting instead an intermediate level between micro and macro. They are typically homeless concepts that have been adopted by several sociological paradigms or grand theories, that is, auxiliary concepts helping mediate between micro and macro. In this paper, I will demonstrate how organizations can be understood as important social mechanisms in themselves in terms of how they both form everyday life and transform individuals into social actors. Organizations are simultaneously actors and structures, and a focus on organizations is essential for understanding processes of stability and change in social life.

Theoretical Origins and Meanings

Organization An early attempt to introduce the concept of organization in social theory was made by Max Weber (1968) in his overview of ‘basic sociological terms’ in the first chapter of Economy and society. For Weber, organization is a special case of social relationship, which is one of the most fundamental concepts in his terminology. According to Weber, a social relationship is the behaviour of a plurality of actors for whom ‘the action of each takes account of that of the others and is oriented in these terms’ (1968: 26). Weber has a broad definition of social relationship: it may be of a ‘very fleeting character’ or of ‘varying degrees of permanence’ (1968: 28). The content of a social relationship may be of ‘the most varied nature: conflict, hostility, sexual attraction, friendship, loyalty, or economic exchange’ (ibid: 27). An important distinction that Weber makes among different types of social relationships is the distinction between open and closed relationships: A relationship will be called closed against outsiders so far as, according to its subjective meaning and its binding rules, participation of certain persons is excluded, limited or subjected to conditions. weber, 1968: 43

Weber distinguishes a large number of reasons that relationships become closed, and closure may ‘assume the most varied forms’ (1968: 45). The fact that a relationship is closed, however, does not imply that there is no fluctuation of people in the relationship, only that involvement is not unconditional and open to everyone. Weber also makes a distinction between associative relationships that generally rest on ‘a rational agreement by mutual consent’ and communal

174

Ahrne

r­ elationships that rest on ‘various types of affectual, emotional, or traditional bases’ (1968: 41). However, he also notes that in most long-lasting relationships, i.e. ‘relatively permanent social relationships between the same persons’ (ibid: 41), these two aspects are mixed. In every relationship involving the same persons, ‘it is quite possible for it to involve emotional values which transcend its utilitarian significance’ (ibid: 41). Organization is the most important type of closed relationship. However, not all closed relationships are organizations.1 Closure, furthermore, is not the only characteristic quality of an organization. A closed social relationship is regarded as an organization only ‘when its regulations are enforced by specific individuals’ (ibid: 48). The existence of an organization ‘is entirely a matter of the presence of a person in authority’. This authority is expressed via orders or rules. Thus, in Weber’s view, an organization is not only a type of closed social relationship but also requires both a form of authority and orders or rules (1968: 49–51). When the field of organization theory started to develop in the 1940s in the United States, however, only Weber’s specialized analysis of bureaucracy was analyzed, not his original interpretation of the phenomenon of organization (1968). The most influential introduction of the concept of organization that accompanied the new wave of organization theory in the United States in the second half of the 20th century was the 1958 book Organizations by James March and Herbert Simon. For them, the basic feature of an organization is decision-making. Their explicit motivation for a special theory of organizations is that organizations influence people’s behaviour in a different way than is the case outside of the organizational context. This influence enables a particularly high degree of coordination, which ‘accounts for the ability of organizations to deal in a highly coordinated way with their environments’ (March and Simon, 1958: 4). Understanding organizations as decided orders enables us to make a clear ­distinction between organizations on the one hand and institutions and ­networks on the other. In contrast with organizations, both institutions and networks are emergent social orders that merely occur rather than being decided. Organizational decisions are statements representing conscious ­choices about how other people should act or the distinctions or classifications they should operate. Decisions can be made by one person at the top of the

1 It is interesting to note that Weber did not delineate any concepts corresponding to either institutions or networks. Among the concepts outlined by Weber, the one that is closest to a concept of institution is probably ‘order’ (Swedberg, 2005: 125).

The Organization of Action

175

o­ rganization or by all members together, democratically. The important point is that in all organizations, there is a legitimate way of making decisions that the members have accepted. There is no guarantee, however, that decisions are rational. Indeed, March and Simon emphasize ‘the bounded rationality’ of decisions (March and ­Simon, 1958: Ch. 6). Decision-makers rarely have all the information they require to make a rational decision, and decisions often must be made under great stress. In his more recent work, James March discusses many of the problems and difficulties of decision-making (March, 1978; March and Olsen, 1976.). In general social theory, March and Simon’s perspective has most notably been advanced by Niklas Luhmann in his theory of communication and different types of social systems (Luhmann, 2003, 2005). The fact that organizations are based on decisions has several implications. Organizations are attempts to create a specific order, but many organizations fail (Brunsson, 2007). Decisions presuppose that there are alternatives among which to choose, and thus decisions may always be contested. A decision is not taken for granted in the same way as an institution; a decision could have been different. Making decisions also presupposes responsibility. One or several persons are responsible, and therefore, if one dislikes a decision, there is somebody to blame and somebody to whom protests may be directed. Institution A general and widely accepted definition of institution in contemporary social science is that offered by the economist Douglass C. North: ‘The major role of institutions in a society is to reduce uncertainty by establishing a stable (but not necessarily efficient) structure to human interaction’ (1990: 6). He also wrote that ‘institutions provide the structure that humans impose on human interaction in order to reduce uncertainty’ (ibid: 495). In sociology the best-known contemporary use of the concept of institution is probably found in Anthony Giddens’ theory of structuration: The most deeply embedded structural properties implicated in the reproduction of social totalities I call structural principles. Those practices which have the greatest time-space extension within such totalities can be referred to as institutions. giddens, 1984: 17, cf. 185

For both Giddens and North, institutions are primarily (and overwhelmingly) composed of rules. ‘Those types of rules which are of most significance for social theory are locked into the reproduction of institutionalized practices, that is, practices most deeply sedimented in time-space’ (Giddens, 1984: 22).

176

Ahrne

North (1993: 20) clarifies the distinction between formal and informal rules. Formal rules ‘must be complemented by informal constraints (conventions, norms of behaviour) that supplement them and reduce enforcement costs’. The difference between informal and formal constraints is regarded as one of degree: ‘a continuum from taboos, customs, and traditions on one end to written constitutions at the other’ (North, 1990: 46). According to North, however, we know very little about how informal norms evolve.2 The notion of informal rules is close to the view of institutions found in the so-called new institutionalism within organization theory. An institution, according to this school, can be defined as ‘a stable, routine-reproduced pattern of behaviour, combined with norms and conceptions that are taken for granted by larger or smaller groups of people’ (Jepperson, 1991). In this sense, ­institutions are taken for granted; people find them natural rather than constructed. Although North includes both informal and formal rules in the concept of the institutional framework, he is not blind to the existence of organizations. Indeed, he draws a clear distinction between institutions and organizations. ‘Like institutions, organizations provide a structure for human interaction’ (North, 1990: 4). Furthermore, according to North there is a mutual interdependence between institutions and organizations. North distinguishes between institutions and organizations by differentiating the rules and the players. ‘The separation of institutions from organizations is crucial if one is to get a handle on the dynamics of institutional change. Institutions are the rules of the game and organizations are the players’ (North, 1993: 12). Organizations are ‘groups of individuals bound by some common purpose to achieve objectives’ (North, 1990: 5). Organizations also have a place in Giddens’ theory, although he does not mention any differences between institutions and organizations. ‘Organizations and social movements are collectivities in which the reflexive regulation of the conditions of system reproduction looms large in the continuity of dayto-day practices’ (Giddens, 1984: 200). However, viewing organizations as players in a game is to disregard their role as independent decision-makers. Although organizations are not merely players, the rules that organizations establish for themselves or others do not have an independent place in North’s theory. He never mentions o­ rganizational 2 One difference between Giddens and North is that North places a stronger emphasis on formal rules, whereas Giddens very clearly emphasizes informal rules: ‘Most of the rules implicated in the production and reproduction of social practices are only tacitly grasped by actors: they know how to go on’ (Giddens, 1984: 22–23).

The Organization of Action

177

rules, but when he alludes to notions such as organizational strategies, he refers to institutions as the ‘underlying rules’ (North, 1990: 5). When he addresses organizational decision-making, the terminology is obscured. Changes in formal rules ‘come through the polity’ (North, 1993: 20). Among the most influential writers on the relationship between institution and organization is Richard Scott, who is generally regarded as a representative of new institutionalism in organization theory, although he is also influenced by North. In his book Institutions and organizations, Scott distinguishes among ‘three pillars of institutions’: the regulative, the normative and the cognitive. Each pillar is expressed through a certain type of social ­expression. The normative pillar is expressed through values and normative expectations of people’s behaviour. Cognitive institutional elements are expressed via categories and classifications of various types, and they are typically taken for granted. ­Regulative institutional elements are expressed in terms of rules and laws. In his discussion of the regulative pillar, Scott mainly refers to the works of North (Scott, 1995: 34–52). In this argument, however, Scott does not consider that rules and laws created through processes of decision-making are very different types of social expressions from cognitive and normative elements and ­therefore have other consequences in terms of how they shape ­social relationships. Network Compared with the institution and organization concepts, the network concept is a relative newcomer. A predecessor of the concept of network was sociometry, which was first used and applied in social psychology (Granovetter, 1973: 1360). Among the first sociologists to introduce network as a concept in general social theory was Mark Granovetter in his 1973 article ‘The strength of weak ties’. Granovetter’s aim was to view networks as one of the most fruitful bridges between micro and macro: it is ‘through networks that small-scale interaction becomes translated into large-scale patterns’ (Granovetter, 1973: 1360). Granovetter’s ambition was to demonstrate how the use of network analysis can be related to ‘various macro phenomena such as diffusion, social mobility, political organization as well as social cohesion in general’ (Granovetter, 1973: 1361). The importance of studying social bonds or ties is an insight stemming from network theory. Since the 1970s, the network concept has become increasingly prevalent and popular in social science research, particularly since around 2000 (see Borgatti and Halgin, 2011). The increasing popularity of the concept, however, has also led to confusion about its meaning, and some have noted ‘a linguistic chaos’ (Borgatti and Foster, 2003; Thompson, 2003).

178

Ahrne

The most common definition of a network is quite simple; it merely states that social networks are structures of relationships linking social actors (­Marsden, 2000: 2727). A more comprehensive definition of a network includes several other terms: A network consists of a set of ties of a specified type (such as friendship) that link them. The ties interconnect through shared end points to form paths that indirectly link nodes that are not directly tied. The pattern of ties in a network yields a particular structure, and nodes occupy positions within this structure. borgatti and halgin, 2011: 1169

A genuine network arises spontaneously. Networks are expanded through people meeting in various contexts and becoming acquainted with one another (Bommes and Tacke, 2005), but it is a slow process (Burt, 2005: 99). Such contacts may gradually develop into larger chains of relationships, and networks have a tendency to ‘ramify endlessly’ (Knoke and Kuklinski, 1982: 24). Relationships in networks are assumed to be non-hierarchical and informal, and a network is maintained through elements such as reciprocity, trust and social capital (Borgatti and Foster, 2003; Podolny and Page, 1998). The relations, or ties, within a network are generally viewed as horizontal, in contrast to the vertical and hierarchical relations in organizations. The strength of network analysis lies in its ability to describe the structure of networks. Its weakest point, however, is its description of ties, what they consist of and how they are established and reproduced (see Azarian, 2010). A common distinction is the one between strong and weak ties, which was introduced by Granovetter. Differences between strong and weak ties may be understood in terms of how tight the ties are, how often people meet and the strength of their emotional bonds (Granovetter, 1973: 1361). Networks are viewed as embedded in other social relations, for instance, among the employees of a certain corporation or among members of a political party. This is why networks can be understood as ‘parasites’: ‘They live on the organizations that feed them, while caring for their host by releasing it from residual decisional contingencies’ (Bommes and Tacke, 2005: 294). Networks are not actors in themselves, and they do not have an identity or a name. They are silent (Bommes and Tacke, 2005: 293). A developing insight from network research is that networks are less static and stable than had previously been assumed. It appears that ‘network ties are constantly being formed, resolved and renegotiated’ (Lizardo and Pirkey,

The Organization of Action

179

2014: 36). This insight places questions such as the emergence, evolution and persistence of ties on the agenda. In the 1980s there was growing interest in network theory in the study of organizations. In the preface to the edited volume Networks and organizations, the editors mention ‘the recent popularity of the term network among academics and managers who are interested in organizations’ (Nohria and Eccles, 1992: vii). Since then much has been written about network forms of organization. The general assumption is that networks have many advantages over alternative organizational forms: ‘many felicitous properties—flexibility, responsiveness, adaptability, extensive cross functional collaboration, rapid and effective decision making, highly committed employees, and so on’ (Kanter and Eccles, 1992: 525). Compared to organizations, networks are regarded as ‘faster on their feet’ (Powell, 1990). For Granovetter, however, a network is very different from an organization. In his article ‘Economic action and social structure: the problem of embeddedness’, he argues that the social relations embedded in personal networks are essential for the establishment of a high level of order. ‘I claim that both order and disorder, honesty and malfeasance have more to do with structures of such social relations than they do with organizational form’ (Granovetter, 1985: 503). The idea of the assumed network form of organization has also been critiqued. Empirical research has demonstrated that over time most organizations, even in new areas such as the it industry, adopt many of the standard bureaucratic methods of organizing. In a volume discussing new forms of organization, Paul DiMaggio concludes that what is occurring is a loosening of bureaucratic means of organizing but not a replacement of organizations with networks (DiMaggio, 2001: 218).

The Dynamics of Organization, Institution and Network

In reviewing how the concepts of organization, institution and network have been used and defined, it is obvious that they are inherently intertwined, but there is also an unfortunate tendency to use the notions of institutions and networks at the cost of organizations. However, for the interplay between organization, institution and network to offer a fruitful and interesting perspective for social analysis, it is necessary to identify the distinct ways in which they shape social bonds and relationships.

180

Ahrne

In relation to the discussion above, at least two basic mistakes are detrimental to the understanding of what these concepts represent. One is to view all types of rules as institutions, when in reality what is called a formal rule is a decided rule, whereas informal implies a rule that is not decided and has an unknown origin. I will argue that it is more fruitful to reserve the expression rule for decided, i.e. organized, rules and consider so-called informal rules as norms. The second mistake is the claim about network forms of organizations. If we succeed in maintaining these three concepts separate from each other, we are also in a better position to understand how they depend on each other and may be combined into different types of social relationships. The various ways in which organization, institution and network are constructed, shaped and reproduced imbue them with different qualities with respect to how they form a social relationship, what types of relationships they enable and what they cannot do. As organization is always based on decisions: there is no such thing as an informal organization. An organization, however, often gives rise to the formation of networks among its members that may also have connections outside the organization, and the existence and reproduction of networks often depends on a decided order. In turn, organizations depend on institutions. Not all that occurs between people in organizations can be decided. Members of organizations rely on common institutions so that they can cooperate on a daily basis. Furthermore, organizations continually import institutions from the environment so that they can adjust to other organizations. Additionally, networks depend on institutions such as birthday celebrations and other ceremonial occasions. Network ties are less visible and known to others. They are often kept secret from the outside, and they are not as explicit as, for instance, ties that result from decisions. Network ties may be both strong and weak, but they remain robust and long-lasting, although they are not always actualized. They are also elastic and stretchable. Nevertheless, they are more akin to institutions in that they are difficult to manage and may spiral out of control without anyone knowing the end game. Their autonomy is low. For their existence and reproduction, networks depend on organizations and institutions. Organizational ties are more direct and action-oriented. They either enable people to do things they otherwise would not be able to do or help people obtain things they would not otherwise obtain. One advantage of an organization is its specificity. If we wished to sum up in a single quality the distinctive characteristics of influence processes in organizations, as contrasted with many other

181

The Organization of Action

influence processes of our society, we would point to the specificity of the former as contrasted with the diffuseness of the latter. march and simon, 1958: 3

Institutions are thicker, coarser and less precise. Nevertheless, institutional forms and elements may be preconditions for organizational decisions and rules, and institutional elements may make organizations function more smoothly or softly. In a sense, organizations are more perishable than institutions. Old decisions are either forgotten or institutionalized. Organizational decisions must be made repeatedly. Organizations require new decisions and must be managed constantly. Organizations often fail, but when they succeed they have special qualities that make them robust. Many new organizations disappear after a relatively short time. If they survive and become established, however, organizations may become strong, long-lasting social actors.

...

One important thread in the development of organization theory in the 1960s was the distinction between organization and environment. The main focus of this thread was that organizational structures must be adjusted to various circumstances in their environments; thus, it has often been called ‘contingency theory’. The implication of this theory is that there is no best way to organize: everything depends on the environment. The environment can be conceptualized in different ways, for example, in terms of technology or in terms of being homogenous, heterogeneous, static or constantly changing (Lawrence and Lorsch, 1967; Thompson, 1967; Burns and Stalker, 1961). However, the environment can also be conceptualized in institutional terms. In his 1948 article ‘Foundations of the theory of organization’ the sociologist Philip Selznick concludes that the formal structures of organizations are always ‘subject to the pressure of an institutional environment to which some over-all adjustment must be made’ (Selznick, 1948: 25). For Selznick, however, institutionalization of the formal structure arises ‘from without as well as from within’ (Selznick, 1996: 273). Institutionalization from within happens as ‘deviations from the formal system tend to become institutionalized, so that “unwritten laws” and informal associations are established’ (Selznick, 1948: 27). Organizations and institutions condition each other. In a now-famous 1977 article by John Meyer and Brian Rowan, ‘Institutionalized organizations: formal structure as myth and ceremony’, the idea of the importance of the institutional environment was reintroduced. This

182

Ahrne

article gave rise to what has been called the new institutionalism in organization theory, which has become immensely popular (see also Hasse’s chapter; Greenwood, Oliver, Suddaby and Sahlin, 2008). However, in a 1996 article titled ‘Institutionalism “old” and “new”’ Selznick (1996: 273) very strongly argues in favour of observing a continuity between the old and the new institutionalism. Meyer and Rowan (1977) not only emphasize the need for organizations to adjust to social institutions to gain legitimacy but also demonstrate how an organization can protect itself from its institutional environment by decoupling its core activities from its periphery. Organizations have many options for hiding what they are doing by performing various myths and ceremonies in their periphery to project a more favourable image of themselves. To increase their legitimacy they present plans for improvement; however, it is rare for such improvements to be very successful (Brunsson, 2006). Other theories provide examples of how organizations manage their environment by strategically controlling their input and output (Thompson, 1967). The view that organizations are overly dependent upon or victims of their environment is a controversial one. Many researchers have argued that large organizations in particular have many ways to influence and create their environments. In their 1978 book The external control of organizations, Pfeffer and Salancik discuss the negotiated environment of organizations. Moreover, many organizations are able to choose their environment and move or establish new premises in a more favourably perceived environment (Ahrne and Papkostas, 2014). Charles Perrow, a well-known organization sociologist, argues strongly against giving institutions too much weight. Instead, he refers to organizations as the independent variable: I will argue that organizations increasingly make our culture including norms of family size and age of marriage, determine what is seen as efficient and for whom, and select our political values and direct our will. perrow, 2002: 10

Thus, although pressure from the organizational environment may be strong, we must conclude that ‘the social order we find in organizations is not a mere reflection of a more general social order that can be adequately understood by concepts and theories describing society in general’ (Ahrne, Brunsson and ­Seidl, 2016). Organizations are major social actors and constituents in their own right.

The Organization of Action



183

Action and Structure

Institutions cannot act, but organizations can. Institutions provide actors with ‘scripts’ and forms for action and facilitate human or organizational interaction, whereas institutions can barely be conceived as actors in themselves. Networks also have problems being actors; they are ‘silent’. Networks lack the coordination mechanisms necessary for groups of people to become actors with common strategies and goals. For networks to become actors, they must acquire at least a few organizational elements, such as arrangements for decision-making (hierarchy) and decisions about membership. Networks comprise relationships between numerous people, and institutions provide norms and values to human interaction. Organizations however, perform both of these functions, but they do so in specific ways. Organizations are unique in their ability to act while simultaneously having the ability to determine their own structure. They are players that to a large extent can make their own rules. It is the authority of an organization’s decision-making body and the specific rules that it makes that enable an organization to act on its own in a much more coordinated and lasting manner than a network. Moreover, organizations do not depend on individual members as much as networks do. Through authority and rules members are also substitutable, and thus an organization is a social entity that exists independent of particular individuals. An organizational relationship is not dependent on any particular person (Ahrne, 1994). Furthermore, organizations are also often members of organizations that have other organizations as members, so-called meta-organizations (Ahrne and Brunsson, 2008). In a global world without a world-state, these meta-­ organizations are gaining in importance, for instance, through their decisions about trans-national rules applicable to the actions of their (organizational) members and how they monitor each other. Such meta-organizations of states, corporations and voluntary associations are among the most stable structures of the global organizational landscape. Following John Scott (see Ch. 9), we can conceive of organizations as relational structures that exist ‘alongside of, and partly out of synchronisation with, the institutional structure’. Organizations are relational structures constructed for collective action. Because of their explicit relational structure, organisations may accumulate large amounts of collective resources: not only material assets such as physical capital in the shape of machinery or buildings but also symbolic ­resources such as a good reputation and contacts with other actors. Their collective

184

Ahrne

r­ esources provide them with the power to require other actors to follow their intentions. Many types of established organizations (corporations, states, voluntary ­organizations) must be regarded as among the most important ‘power containers’ that ‘store allocative and authoritative resources’ (Giddens, 1984: 262) in social life. Thus, it is misleading to understand organizations as a sort of meso-level in society, mediating between a micro- and a macro-level. To a large extent, social structures are composed of organizations. There are obvious differences between, for instance, how states do things, what states do and how capitalist enterprises do things. However, as organizations they have the same qualities as social actors in that they can coordinate the actions of large numbers of individuals while setting goals and determining rules for themselves.

Stability and Change

When we consider organizations as important parts of a social structure, it becomes possible for us to understand social stability and social change in terms of the dynamics between old and new organizations. Organizations are important means not only of stability but also of innovation and social change. Organizations provide continuity. Social revolutions often result in new leaders taking power in old organizations. However, organizations also provide change primarily through the creation and success of new organizations, often on a global scale. Examples include organizations such as Amnesty International, Greenpeace, Google and Facebook. To explain how organizations may accomplish social change, however, we have to view them in relation to old organizations. Organizations can use their considerable resources of power to enhance stability in an attempt to prevent unwanted changes, but they can also use their resources in an effort to change social conditions in various directions. Organization is a useful strategy for those who want to change behaviour, identities or status orders that are institutionalized (Ahrne and Brunsson, 2011: 96). Institutions are associated with stability. For Berger and Luckmann it is obvious that institutions ‘cannot be created instantaneously’ (Berger and Luckmann, 1991: 72). Institutions develop slowly (Czarniawska, 2009). When discussing institutional change, North is clear: ‘the overwhelming majority of change is incremental and gradual … the difficulty of fundamentally altering paths is evident’ (North, 1998: 498). Networks, however, are generally regarded as flexible and fluctuating. They are associated more with change than with stability and more with process than with structure (Emirbayer, 1997).

The Organization of Action

185

Organizations are marked by the conditions existing when they were created and tend to maintain their fundamental structural qualities throughout their existence (Stinchcombe, 1965). In the tradition of organizational ecology, it is argued that selective processes among newly established organizations favour organizations whose structures are difficult to change (Hannan and Freeman, 1984). Organizations cannot continue to change forever. There are severe cognitive limits on decision-making (March and Simon, 1958; March, Schulz and Zhou, 2000). Thus, there are strong tendencies in all established organizations to change unwillingly, as little as possible and in directions outlined in existing practices and resources. In his book Strategy and structure, Alfred Chandler (1962) considers managers’ preoccupation with daily routines to keep the enterprise going as another important source of inertia. However, the inertia of old organizations generates opportunities and space for new organizations. Because of their difficulties in doing things in new ways, old organizations leave considerable space around them that may be utilized by new organizations that can make new things in other ways. New organizations do not have to destroy the old ones to establish themselves and grow. The inherent dynamics between old and new organizations can thus generate change in the organizational landscape. Its structure changes when new organizations are established and new forms of relations between old and new organizations emerge. The landscape grows thicker and denser and the number of organizations tends to increase. Moreover, new organizations may force the older ones to change, albeit slowly. However, old established organizations may also contribute to social change if they extend or move to new places and into another environment where they may be perceived as new and innovative (Ahrne and Papakostas, 2014). Organizations are powerful social actors that cannot be reduced to institutions or networks. Organizations can do many things that institutions and networks can do, but in their own way. Organizations often break with ­institutional norms and expectations, at times secretly, but at other times when finding and encouraging new ways to do things. Compared with networks, organizations have other ways of recruiting and sanctioning members that enable them to act on a larger scale. However, organizations cannot do everything that ­institutions and networks can do. Institutions have a stronger legitimacy than organizations, and social relationships in networks have a different character and different qualities from social relationships in organizations, although organized relationships sometimes may become networks. In a global society, there is much evidence that organizations are gaining in importance. Patricia Bromley and John Meyer recently discussed

186

Ahrne

‘­hyper-organization’ and global organizational expansion as follows: ‘faced with any problematic situation, the modern impulse is to create more organizational structures’ (Bromley and Meyer, 2015: 4). To analyse and explain processes of globalization; it is time for social theory to recognize the relevance and importance of organizations. References Ahrne, G. (1994). Social organizations. Interaction, inside, outside and between organizations. London: Sage. Ahrne, G. and Brunsson, N. (2008). Meta-organizations. Cheltenham: Edward Elgar. Ahrne, G. and Brunsson, N. (2011). “Organization outside organizations: the significance of partial organization”, Organization. 18(1): 83–104. Ahrne, G., Brunsson, N. and Seidl, D. (2016). “Resurrecting organization by going beyond organizations”, European Management Journal. 34(2): 93–101. Ahrne, G. and Papakostas, A. (2014). Organisationer, samhälle och globalisering. Tröghetens mekanismer och förnyelsens förutsättningar. [Organizations, society and globalization. Inertia mechanisms and renewal conditions]. Lund: Studenlitteratur. Azarian, R. (2010). “Social ties: elements of a substantive conceptualization”, Acta Sociologica. 53(4): 323–338. Berger, P. and Luckmann, T. (1991). The social construction of reality. A treatise in the sociology of knowledge. Harmondsworth: Penguin Books. Bommes, M. and Tacke, V. (2005). “Luhmanns system’s theory and network theory”, Eds. D. Seidl and K.H. Becker. Niklas Luhmann and organization studies. Malmö: Liber. Borgatti, S.P. and Foster, P.C. (2003). “The network paradigm in organizational research: a review and typology”, Journal of Management. 29(6): 991–1013. Borgatti, S.P. and Hagin, D.S. (2011). “On network theory”, Organization Science. 22(5): 1168–1181. Bromley, P. and Meyer, J. (2015). Hyper-organization. Global organizational expansion. Oxford: Oxford University Press. Brunsson, N. (2006). Mechanisms of hope. Maintaining the dream of the rational organization. Malmö: Liber ab. Brunsson, N. (2007). The consequences of decision-making. Oxford: Oxford University Press. Burns, T. and Stalker, G.M. (1961). The management of innovation. London: Tavistock Publications. Burt, R.S. (2005). Brokerage and closure. An introduction to social capital. Oxford: Oxford University Press.

The Organization of Action

187

Chandler, A. (1962). Strategy and structure: chapters in the history of the industrial enterprise. Cambridge, ma.: M.I.T. Press. Czarniawska, B. (2009). “Emerging institutions: pyramids or anthills?”, Organization Studies. 30(4): 423–441. DiMaggio, P. (2001). “Conclusion: the futures of business organization and paradoxes of change”, Ed. P. DiMaggio. The twenty-first-century firm. Changing economic organization in international perspective. Princeton: Princeton University Press. Emirbayer, M. (1997). “Manifesto for a relational sociology”, American Journal of Sociology. 103(2): 281–317. Giddens, A. (1984). The constitution of society. Outline of the theory of structuration. Cambridge: Polity Press. Granovetter, M. (1973). “The strength of weak ties”, American Journal of Sociology. 78(6): 1360–1380. Granovetter, M. (1985). “Economic action and social structure: the problem of embeddedness”, American Journal of Sociology. 91(3): 481–510. Greenwood, R., Oliver, C., Suddaby, R. and Sahlin, K. (2008). “Introduction”, Eds. R. Greenwood, C. Oliver, R. Suddaby and K. Sahlin. The SAGE handbook of organizational institutionnalism. London: Sage. Hannan, M. and Freeman, J. (1984). “Structural inertia and organizational change”, American Sociological Review. 49(2): 149–164. Jepperson, R.L. (1991). “Institutions, institutional effects, and institutionalism”, Eds. W. Powell and P. Di Maggio. The new institutionalism in organizational analysis. Chicago: Chicago University Press. Kanter, R.M. and Eccles, R.C. (1992). “Conclusion: making network research relevant to practice”, Eds. N. Nohria and R.C. Eccles. Networks and organizations: structure, form and action. Boston, ma.: Harvard Business School Press. Knoke, D. and Kuklinski, J.H. (1982). Network analysis. Newbury Park, ca.: Sage. Lawrence, P.R. and Lorsch, J.W. (1967). Organization and environment: managing differentiation and integration. Boston: Graduate School of Business Administration and Harvard University. Lizardo, O. and Pirkey, M.F. (2014). “How organizational theory can help network theorizing: linking structure and dynamics via cross-level analogies”, Contemporary perspectives on organizational social networks. Research in the sociology of organizations. 40: 33–56. Luhmann, N. (2003). “Organization”, Eds. T. Bakken and T. Hernes. Autopoietic organization theory. Drawing on Niklas Luhmann’s social system perspective. Oslo: Abstrakt Forlag as. Luhmann, N. (2005). “The paradox of decision making”, Eds. D. Seidl and K.H. Becker. Niklas Luhmann and organization studies. Malmö: Liber.

188

Ahrne

March, J. (1978). “Bounded rationality, ambiguity and the engineering of choice”, Bell Journal of Economics. 9(2): 587–608. March, J. and Olsen, J.P. (eds.). (1976). Ambiguity and choice in organizations. Bergen: Universitetsforlaget. March, J. and Simon, H. (1958). Organizations. New York: John Wiley & Sons, Inc. March, J., Schulz, M. and Zhou, X. (2000). The dynamics of rules. Change in written organizational codes. Stanford: Stanford University Press. Marsden, P. (2000). “Social networks”, Encyclopedia of sociology. 2nd edition. Vol. 4. New York: MacMillan Reference usa. Meyer, J. and Rowan, B. (1977). “Institutionalized organizations: formal structure as myth and ceremony”, American Journal of Sociology. 83(2): 340–363. Nohria, N. and Eccles, R.G. (eds.). (1992). Networks and organizations. Structure, form and action. Boston, ma.: Harvard Business School Press. North, D.C. (1990). Institutions, institutional change and economic performance. Cambridge: Cambridge University Press. North, D.C. (1993). “Institutions and credible commitment”, Journal of Institutional and Theoretical Economics. 149(1): 11–23. North, D.C. (1998). “Where have we been and where are we going?”, Eds. A. Ben-Ner and L. Putterman. Economics, values and organization. Cambridge: Cambridge University Press. Perrow, C. (2002). Organizing America. Wealth, power, and the origins of corporate capitalism. Princeton: Princeton University Press. Pfeffer, J. and Salancik, G. (1978). The external control of organizations. New York: Harper and Row Publishers. Podolny, J.M. and Page, K.L. (1998). “Network forms of organization”, Annual Review of Sociology. 24: 57–76. Powell, W. (1990). “Neither markets nor hierarchy: network forms of organization”, Research in Organizational Behaviour. 12: 295–336. Scott, R.W. (1995). Institutions and organizations. Thousand Oaks: Sage. Selznick, P. (1948). “Foundations of the theory of organization”, American Sociological Review. 13(1): 25–35. Selznick, P. (1996). “Institutionalism ‘old’ and ‘new’”, Administrative Science Quarterly. 41(2): 270–277. Stinchcombe, A. (1965). “Social structure and organizations”, Ed. J. March. Handbook of organizations. Chicago: Rand McNally. Swedberg, R. (2005). The Max Weber dictionary. Key words and central concepts. Stanford: Stanford University Press. Thompson, G.F. (2003). Between hierarchies and markets. The logic and limits of network forms of organization. Oxford: Oxford University Press. Thompson, J. (1967). Organizations in action. New York: McGraw-Hill. Weber, M. (1968). Economy and society. Vol. 1. Berkeley: University of California Press.

chapter 11

About Actors: An Institutional Perspective Raimund Hasse Introduction This contribution will investigate the concept of new sociological institutionalism. Core assumptions and levels of analysis will be outlined, and we will specifically address the notion of actors. We will show that from this perspective, actors result from ongoing social construction processes in both the social sciences and social practice. In particular, individuals, organizations and states tend to be constructed as actors, and this construction implies that decision-making, rationality and responsibility are ascribed to—and internalized by—them. The new institutionalism can be characterized as a large—and growing— research field with roots in European social theories from the end of the 19th and beginning of the 20th century (Djelic, 2010). These forerunners include sociological classics such as Durkheim and Weber. Alexis de Tocqueville’s idea that institutions have a cultural imprint that usually remains tacit has also had a significant impact. The prime focus of the new institutionalism, however, has been organizations (Powell and DiMaggio, 1991; Scott, 1995).1 Specifically, the diffusion of novelties such as new technologies, best management practices and professional standards—and the corresponding effects of these novelties on organizational structures and practices—has attracted institutional research attention (Strang and Soule, 1998). The core finding of this strand of research has been that diffusion processes are based on various types of ­proximity (geographical, set in a regional or national context, sector-based, or attributable to the recruitment of professionals with a similar educational background; see Powell, Whittington and Owen-Smith, 2009) and can contribute to isomorphism (DiMaggio and Powell, 1983). However, it has also been emphasized that degrees of coupling of organizational structures and practices vary because formal structures can be relatively responsive to new ­expectations, whereas 1 First, organizations are viewed as (a type of) institution, which refers to the taken-forgrantedness of organizational forms, their high degree of legitimation as a means of ­coordination, and their regulating effects on practices. Second, organizations are viewed as embedded in institutional contexts with various expectations regarding appropriate forms and activities. © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_012

190

Hasse

practices are not (see Broomley, Hwang and Powell, 2012; Hasse and Krücken, 2014). Because issues of diffusion and coupling can be related to a broad range of organizational topics, the new institutionalism has attracted a great deal of ­attention since the 1990s. Most of the findings have been in line with initial contributions, but there have also been surprising empirical findings and suggestions for conceptual variations. One main issue in related debates has been the status of actors in institutional research. Another issue has been a reflection on levels of analysis. Against this background, I will first discuss the scope of institutional analysis by referring both to macro-sociological accounts and to the more recent quest for a micro-foundation. Second, an institutional perspective on actors will be described that views actors as products of social construction processes and is sensitive to status differences. Third and to further elaborate an institutional perspective on actors, differences between three types of actors—states, organizations and individuals—will be outlined. Fourth, the argument will be summarized and value implications will be addressed in the discussion section.

Core Assumptions and Levels of Analysis

Institutional analysis is based on the plain assumption that context and history matter. With regard to context, network sociologist Mark Granovetter (1985) has shown that in ongoing relations participants tend to exchange information, develop common worldviews and adapt to each other. He even argues that such relations can have a profound impact on actors’ preferences. Similarly, institutional theory emphasizes the context in which actors are embedded.2 Whereas contextual thinking emphasizes ongoing impacts, the notion of history refers to the long-lasting effects of past influences. Here, the conceptual model is socialization theory. Correspondingly, organization researchers have found that in their early years organizations are quite responsive to contextual

2 The point of departure from network sociology is that in institutional theory cultural factors are assumed to determine not only which relations are established and maintained but also, and even more importantly, what the results of such relations are. Strang and Meyer (1993) have illustrated the significance of culture as opposed to mere network relations with the example whereby, embedded in a culture of patriarchy, ongoing relations between men and women have not resulted in equalization but in institutionalized forms of division of labor and power inequalities.

About Actors: An Institutional Perspective

191

features and tend to incorporate them, whereas in later phases they tend to become more inert (Stinchcombe, 1965; Hannan and Freeman, 1977). Because both historical and ongoing contextual features predominantly involve cultural and socio-economic circumstances, it is obvious that the focused determinants are man-made. In this regard, institutional theory is devoted to the social constructivism of Berger and Luckmann (1967), who argued that social life is based on the paradox that social order in the form of norms, values and beliefs is experienced as an exterior reality, albeit one constructed by mankind. Berger and Luckmann (1967) have also profoundly shaped institutional thinking in other respects, particularly the basic understanding of institutions3 and the emphasis on issues of legitimation.4 The new institutionalism can be distinguished from two theories. Unlike functionalism, it emphasizes the symbolic meaning of social structures and actions, although the rejection of functionalism has remained more tacit with respect to the general social theory of Talcott Parsons, whereas it has been quite emphatic with respect to contingency theories in organization research (which assume that organizational structures are a functional response to work requirements and technological and/ or economic conditions). With respect to economic theories of action and sociological variants of rational choice, institutionalists emphasize non-reflected forms of behaviour such as habits and routines, they insist on the social and cultural embeddedness of action, and they consider preferences and other features of actors to require explanation. Although the isomorphic trends of organizations have been a main focus of institutional research, they are not the only focus. It has also been shown empirically that norms and ideals regarding individual life courses (e.g., health-oriented behaviour, life-long learning) and biographical phases (childhood as a protected phase, the new elderly as actively participating members of society) have become objects of similar homogenization processes (Meyer, 1986). Similarly, it has been shown that expectations addressed to nation states concerning issues such as welfare policy, education, and climate change have also resulted in more or less homogeneous developments (Meyer, 1987; Hasse, 2003). Against this background institutionalists have argued that the direction of these changes is far from arbitrary. Changes are primarily related to 3 Institutions can be defined as expectation structures and their internalization. According to Berger and Luckmann (1967), they are based on the habitualization of action (e.g., doing research) and on typifications of those who perform these actions (e.g., researchers who do research). 4 Legitimation mainly implies the provision of meaning and taken-for-grantedness (see Suchman, 1995).

192

Hasse

the principle of rationalization, which Max Weber (1972) described a century ago as a master-trend of societal modernization. Since then, proponents of the rationalization thesis have been convinced that any segment of society might be exposed to processes of rationalization. Following this line of thought, John Meyer has argued that not only organizations but also individuals and nation states have become objects of rationalization. Accordingly, the common ground among nation states, individuals and organizations is that they enact themselves as rational actors by defining goals, developing preferences and selecting alternatives. In addition to rationalization, institutionalists assume that the universalization of fairness and inclusion ideals is an integral part of the cultural context. This predominantly involves the entitlement of democratization and the strengthening of participatory elements, to norms of equal opportunity and non-discrimination, to empowerment and to the protection of minorities. This macro-sociological variant of the new institutionalism becomes most visible in John Meyer’s notion of a world polity, which highlights the early imprinting of hegemonic world culture by enlightenment and even by early Christendom. The world polity approach has been developed since the 1970s (Meyer and Hannan, 1979). It assumes a hegemonic Western culture that potentially diffuses into any domain of contemporary society. Following the theoretical scheme of Alexis de Tocqueville that views institutions as carriers of cultural principles, it is argued that modern world culture is inscribed in international organizations. Accordingly, high-status international organizations—governmental and non-governmental alike—are viewed as agents of the ideals of rationalization and fairness (Boli and Thomas, 1997). They define standards and raise expectations with regard to nation states and organizations. Whereas this macro-sociological conceptualization has stimulated empirical research since the spread of the new institutionalism, the quest for a micro-foundation entered the stage later. Nonetheless, it can be considered an important supplement. The main criticism of institutional thinking has been that it has a ‘people problem’ (Hallett and Ventresca, 2006), although in empirical investigations references to individuals are widespread. For a time, the idea of institutional entrepreneurs that establish, maintain or destabilize institutions appeared promising. Institutional entrepreneurs were conceptualized as intentional, reflexive actors that are equipped with resources. In addition to individuals and organizations, professions, standard setters and social movements could also be viewed as institutional entrepreneurs. Because this conceptualization appeared exaggerated, it was transformed into the concept of institutional work (Lawrence, Suddaby and Leca, 2009). Institutional work appeared to be particularly important for the reproduction of institutions, and

About Actors: An Institutional Perspective

193

compared to entrepreneurship it was a rather mundane concept that also included everyday activities in the office, in the courtroom, in Parliament, etc. An alternative to this action-centred micro-foundation is a stronger focus on language. Institutionalists thus also rediscovered the significance of dis­courses and distinguished between rhetoric, narrative and dialogue. In organization and management research, rhetoric in particular (e.g. in the form of public relations and marketing) attracted considerable attention. Other less conscious and less manipulative forms of language use tended to be neglected in that context. Numerous contributions did, however, focus primarily on everyday narratives and on legitimizing accounts used to attribute meaning to decisionmaking and other activities. For instance, in a frequently cited article on institutional changes in a rape crisis centre in Tel Aviv, Zilber (2002) convincingly showed that everyday narratives can be the basis for profound forms of institutional change, which in that case was a shift towards professionalization. In sum, the new institutionalism has emphasized an all-encompassing world culture that is primarily represented by international organizations, professional associations and social movements. However, this world culture is not expected to diffuse evenly, and at the micro level degrees of freedom— end requirements to translate or edit potentially global scripts—remain. Focusing on the micro level of everyday life, certain researchers have highlighted intentionality. Nonetheless, a substantial portion of micro-activities appears to be based on routines and habits that are only poorly reflected—and they are provided with meaning by language. Because of the combination of (a) considering a macro-sociological cultural framework and (b) a micro-foundation that emphasizes routines and language, the new institutionalism can claim to cover any level of sociological analysis and is not restricted to the meso-level of organizations, regardless of the fact that it has been the most successful in organization research.

Towards an Institutional Understanding of Actors

From an institutional perspective, actors in the sense of rational interest-­ driven entities are man-made; they have been initialized and remain objects of ongoing socialization processes, therapeutic treatment, consulting and management. The social sciences appear to contribute to the understanding of actors as taken-for-granted entities. The academic notion of the concept of actors can be traced to the beginning of the 20th century, but its diffusion into most spheres of the social sciences began in the 1960s (Bromley and Meyer, 2015: 125). The analytical advantage appeared to be that actor theories could be

194

Hasse

related to very different entities ranging from persons to states. With respect to individuals, however, some are more likely to be conceptualized as actors than others; for instance, adults are more likely to be considered actors than children, and the more educated and prestigious an individual, the more likely he or she is to gain actor status in social analyses (e.g., bankers in economic sociology vs. the homeless in the sociology of social problems). With respect to collective actors, it is often assumed in the social sciences that any organization ‘is’ an actor, whereas only a few decades ago even James Coleman thought that certain organizations—his example was universities—primarily serve as institutional contexts in which others (individual scientists, professions, etc.) act (Coleman, 1982). According to Coleman, universities were a type of organization lacking crucial attributes of actors. Assuming that actor status must be ascribed to and internalized by an entity, the most basic question is whether someone or something is addressed as an actor. Media coverage can serve as a good empirical indicator of the societal discourses that help constitute actors. Therefore, we may ask the following question: who acts, and who is an actor, according to their descriptions? Obviously, most individuals are portrayed as actors, as are organizations and states. Generally, media scripts tend to personalize, but occasionally social movements, religious groups or even ethnicities are also described as actors that are doing something. The actor scheme of media coverage is so dominant that a broad range of actors is taken into consideration. According to their scheme, the social world is composed of many diverse entities, all of which have the capacity to act intentionally and more or less rationally. In other social domains, in contrast, the social construction of actors is more rigid, but can be very consequential, as in the case of law. Who gains status as a legal subject, who may become an object of liability issues, and who can be held legally responsible? Although other entities such as households may also be legal subjects, we primarily find the triad of nearly all individuals, all organizations and all states. Against this background it may be argued that today there are more actors around than ever, but concurrently we find that actor status is ascribed to fewer types thereof. Collectives such as groups and networks most often cannot be viewed as actors in the strict sense; they are seldom addressed; and they can barely find their way to the courtroom. Although there is still no detailed social theory of actors, several theories have reflected on that issue. Institutional lines of thinking aside, Niklas Luhmann’s systems theory (Luhmann, 1997) and Harrison White’s network approach (White, 1992) have made suggestions in similar directions. Their common ground is the empirical observation that certain entities tend to be conceptualized or addressed as actors, whereas others are not. This

About Actors: An Institutional Perspective

195

o­ bservation is obviously related to which attributes are ascribed to individuals and ­collectives. The question of who is addressed as an actor with decisionmaking qualities and corresponding responsibilities is thus an empirical one, and its answer varies across cultures and across time. Thus, actors represent an issue of contingency and variation; certain collectives are no longer considered actors although they were once (and in certain cultures, still are) identified as driving forces (e.g., classes or clans). Against this background, institutional theory has reflected on different actors and their status in society. It has been emphasized that actors are associated with qualities that are culturally valued, and it has been shown that contemporary society is composed of a dense net of institutions that help establish and maintain an identity as actor, that is, as a goal-seeking entity that rationally selects between means to efficiently achieve ends. Some of these institutions are obvious: in the case of individuals, therapists and social workers who aim to improve the ‘agency’ of their clients; political experts and international organizations in the case of states; and consultants, managers and their professional associations in the case of organizations. These ‘helping professions’ spend considerable energy and resources on the social construction of actors and on the maintenance of that status. Their mission is not only to help establish a facade by imparting techniques of impression management to convince others but also to aim at the internalization of an actor status that makes individuals, organizations and states believe that they ‘are’ actors. Treated as actors and having internalized this status, individuals, organizations and states are objects of ongoing rationalization processes. Rationalization implies setting goals and optimizing the relationship between means and ends. Rationalization can also have procedural qualities such as transparency, accordance with institutionalized forms, or identifying causal relations to learn what and how and with what effects something can be engineered. As a consequence, and although actors are conceptualized as rational entities, further rationalization is an integral part of the normative expectation with which they are confronted. Based on the cultural script of actor rationalization, Meyer and Jepperson (2000) have emphasized that contemporary society has generated a growing number of therapists and consultants who help maintain and improve actor rationality. Typically, clients are those who find themselves in a position in which they lack actor qualities because they are viewed as lacking the ability to act in accordance with their own interests. Compulsive individuals, failed states and poorly managed organizations are typical examples of such clients. Against this background, the authors perceive status differences between rational actors and those who lack the ability to act in accordance with their interests.

196

Hasse

Furthermore, Meyer and Jepperson (2000) have drawn attention to actors who claim to act rationally but not according to their self-interest. These actors act on behalf of other actors, as in the case of professionals and their clients, or on behalf of highly legitimate abstract principles such as democracy, sustainability, human rights, nature conservation, biodiversity, cultural heritage, etc. The argument here is that as long as actors successfully maintain an image as agents of others or of abstract principles, they acquire a higher status than that of merely self-interested actors. Social movements (as agents of abstract principles) or doctors and politicians (as agents of patients or voters) thus have a high status as long as they successfully maintain their image as agents. As soon as these agents are perceived as ‘only pretending to be agents for others’ while in reality acting in accordance with their own interests (such as doctors seeking money from their patients, leaders of social movements aiming to make a career, politicians profiting from lobbyists, etc.), this high status erodes and what is otherwise considered quasi-normal economic behaviour is normatively sanctioned. In sum, we thus find a social structure that is composed of three layers: within the lowest layer we find those who are viewed as lacking actor qualities and thus require help; in the medium layer we find rational actors that in economic theories are viewed as the only model of man; and at the top we find those who are viewed as altruistically acting as agents for others or for abstract principles.

Actor Differences

States From an institutional perspective, the modern nation state can be described as a rationalized and further rationalizing entity, which, like any other type of actor, is expected to act in accordance with its interests and preferences (Bromley and Meyer, 2015). However, there are also specific features that can be taken into consideration if one aims to distinguish nation states from other actor types. One characteristic feature is that in democracies and viewed from a principal/agency perspective, the demos is considered a principal, leaving the role of agents to be played by office-holders (Moe, 1984).5

5 A principal/agency perspective has also been applied to the relationship between government and administration (Hamilton and Biggart, 1984). According to this perspective, the administration takes the position of the agent who is expected to fulfill the will of government. Therefore, it is unsurprising that public administration (of 19th-century Prussia) had

About Actors: An Institutional Perspective

197

The modern state also exhibits the specific feature that legitimation requires subordination to a set of rather restricted objectives. In contrast to empires, these objectives can be reduced to two missions. The first is the well-being of its citizens. Until the 1970s most modern states moved towards expanding their welfare programme (Pierson, 1991). It also appeared evident that advanced welfare states were economically more successful than those that lagged behind (Wilenski, 1975). Welfare states committed themselves to addressing almost anything that might harm their citizens. Social services, health care, education and compensation in the case of unemployment and various handicaps appeared on the agenda, and expenditures increased dramatically. Starting in the 1970s, expanding welfare states were attacked by both economics and sociology. It was argued that citizens suffered from high taxes and inefficient services (oecd, 1981). Concurrently, a legitimation crisis was problematized because of social control and client passivity attributable to the unfair support of better organized groups (Habermas, 1973). In the social sciences, ‘growth to the limits’ (Flora, 1986) was discussed, and in the political arena an ‘end of the welfare state as we know it’ (Ronald Reagan) was declared. As a consequence, massive reforms have been implemented since the 1980s. Many of these reforms can be viewed as cutbacks, but it may also be argued that these reforms were based on the belief that reforming the welfare state might lead to better results. In that respect, controversies regarding the future of the welfare state have not cast doubt on the objective of citizens’ well-being nor have they reduced nation states’ related responsibilities. A second objective relates to international relations. Here, protection from external military threats and territorial integrity are a primary issue. Furthermore, nation states seek to maintain or improve status in the context of other nation states. They actively embed themselves in supra-national institutions, shoulder responsibilities and accept evaluation by these institutions (Finnemore, 1996). The result is a dense net of bilateral arrangements and commitments to norms, standards and recommendations of international organizations (Meyer, Boli, Thomas and Ramirez, 1997). This has reduced the sovereignty of nation states, and a broad range of institutional research has shown that many developments result neither from functional requirements nor from national interests but from exposure to what has been labelled the world polity (Schneiberg and Clemes, 2006). Nonetheless, the ideological idea that nation states are sovereign actors has remained strong.

been the model for Max Weber’s ideal type of bureaucracy as an organizational form that is designed to leave no degree of freedom to subordinate levels.

198

Hasse

Organizations Organizations are viewed as prototypes of collective actors that have become more significant during the 20th century (Coleman, 1982). Compared with nation states, organizations are less restricted in choosing their ends, and although there is strong evidence for homogenization, formal structures vary significantly. It also appears quite legitimate for organizations to consider the expectations of a broad range of significant others. Organizations are expected to be relatively sensitive with regard to almost all of these expectations. Consequently, business firms, for instance, develop corporate responsibility strategies, although such strategies may require investments that are neither in line with a firm’s genuine missions and goals nor with the expectations of resourcerelevant environments. Furthermore, organizations are a type of actor that can be formally controlled by various principals. Members (in the case of most nonprofit organizations) or the state (public administration) can be principals, but so can owners (as in the case of business firms). In this regard, we find more variation in organizations than in nation states in which democratic ideals exclusively attribute the role of the principal to the demos. In blunt terms, we can conceive of legitimate owners of organizations but not of modern nation states. Another line of demarcation can be discerned in the consequences of organizations’ need for resources. Whereas nation states also require resources, state mortality is extremely rare. Correspondingly, and in contrast with the case of nation states (and individuals), the mortality of organizations can be accepted easily, and it may even be argued that organizations require high rates of mortality (as well as high founding rates) to adapt to new circum­ stances. Particularly in the economy, the emergence and disappearance of firms is viewed as an integral part of development, and some firms (such as high-tech start-ups) are even designed to be short-term actors. Finally, limits to growth are not an issue in the case of organizations, and this applies to both the expansion of individual organizations and the creation of further organizations. One simple reason may be that organizations are less strictly related to territories. Aside from a few rare exceptions (in the Antarctic, for instance), the expansion of existing nation states or the foundation of new ones requires other nation states to shrink or disintegrate. In the case of organizations, in contrast, there are no such pre-conditions, and it may even be argued that the formation of various organizations can be a self-reinforcing mechanism, as business firms may stimulate the founding of associations or unions or suppliers, etc. In the case of organizations, there are no limits to growth, whereas in the case of states, expansion has the character of a zerosum game.

About Actors: An Institutional Perspective

199

Individuals Individuals can be viewed as the pure model for rational actors. The normative expectation is that they are guided by their interests and (according to Hirschman, 1977), no longer by compulsive passions. There is strong pressure for individuals to be equipped with preferences and capacities to decide and act in accordance with these preferences. During their entire life course— from early childhood in kindergarten to old age in nursing homes—people are treated as decision-making individuals. The helping professions ranging from psychology to social work and education support individuals in developing and maintaining capacities to conform to this norm. Equating the institutionalized understanding of the modern individual with the ideal of a rational actor nonetheless has shortcomings. This is most crucial with respect to emotional issues. Family members, friends and lovers are not expected to hide emotions and act purely according to self-interest. Instead, we do not react with surprise and misunderstanding, nor do we sanction a person who engages in such relations with empathy and a social spirit that cannot be reduced to individualistic economic rationality. Against this background, the idea of different value spheres or institutional logics has been proposed by social scientists ranging from Max Weber (1972) to Friedland and Alford (1991) and Boltanski and Thévenot (2007). In Marxist theory, it has become popular to highlight relations between a production and a reproduction sphere, and the rational world of production has been accused of parasitically exploiting (and threatening) the sphere of reproduction. This essentially is the core of the notion of a colonialization of life worlds proposed by Jürgen Habermas (1987). In line with this argument, critical theorist Eva Illouz (2007) has argued that contemporary capitalism has developed a new mode of utilizing emotional resources. According to this line of thinking, which had similarly been proposed by the emotional labour debate (Hochschild, 1983; Sutton, 1991), many work contexts are characterized by the fact that emotional competences are required that cannot be reduced to simple self-control and ‘cooling down’. Instead, the capacity to reflect upon emotional issues and the ability to (verbally) express feelings are highly valued. In an empirical investigation of the evaluation of students by teachers at elementary schools we found that such emotional competencies were an integral part of both formal evaluation criteria and material considerations of teachers’ grading and promoting of students (Hasse and Wyss, 2015). Relevant emotional qualities, however, were not restricted to non-affection and neutrality. Instead, teachers developed an understanding of ‘emotional maturity’. First, this emotional maturity included emotional involvement. Second, an appropriate expression of emotions was viewed as an indicator of emotional

200

Hasse

maturity. Third, students were expected to reflect upon emotional impacts on others and to contribute to their well-being. Similarly, we find a demand for emotional attachment and empathy for others in many work contexts—and this is not restricted to the helping professions mentioned above. Instead, one may consider cultural industries ranging from media to movies and music, the fields of sales and distribution, which are based on a re-enchantment of consumption, work based on teamwork and close collaboration (as in R&D), and so on. It may thus be concluded that emotional issues distinguish individuals from other actor types and again, it must be emphasized that these differences do not involve the question of how individuals are but instead to normative and cognitive expectations.6

Discussion and Normative Considerations

This contribution has addressed the concept of institutions. Although this concept has a much longer history, it gained strength at the end of the 20th century. The characteristic feature of the new institutionalism is its radical social constructivist approach, which includes the constitution of modern actors. In the first step, the new institutionalism was portrayed as a cultural theory with an empirical focus on organizations, states and individuals. After referring to its macro-sociological profile, more recent reflections regarding a languagecentred micro-foundation were presented. In the second step, basic insights for an institutional understanding of actors were discussed. The emphasis was on social processes of attribution and internalization of an actor status and on status differences between actors (from actors lacking rationality at the bottom via plainly self-interested actors to rational agents for others). In the third step, the suggestion was made to be more sensitive to differences between individuals, organizations and states as genuine types of actors. Thus, specific features of the cultural constructs of nation states (welfare issues), organizations (stakeholder sensitivity) and individuals (emotional competencies) were outlined. The underlying idea was that an appropriate theorizing of actors requires to bringing differences back in. However, it should be noted again that these differences do not simply involve empirical departures from the dominant script of actors. Instead, they allude to institutionalized differences in the normative setup of state actors, organizational actors and individual actors.

6 Nevertheless, we may sometimes rhetorically attribute feelings such as ‘angst’ to states or think about paranoid organizations in a metaphorical sense.

About Actors: An Institutional Perspective

201

Instead of providing suggestions for further research in the outlined direction, one can finally address issues of valuation and normative implications of the presented concept. In sociology, there is a long tradition of placing oneself in the position of a neutral observer, and any other attitude has a poor standing in the discipline. It appears that Weber’s pathos-laden objections to so-called ‘teacher’s desk-socialism’ and his at-best ambivalent (if not fatalistic) notion of the iron cage of modernization have remained potent arguments against normative considerations. In contrast, most other social sciences have had less scruples with regard to incorporating normative issues: economists do not hesitate to argue in favour of rationality and efficiency; political scientists find it easy to evaluate based on democratic ideals; psychologists tend in the direction of the medical professions and aim to contribute to well-being and related concepts; and, albeit to a lesser extent, anthropologists raise concerns regarding cultural heritage and diversity. Against this background it may be noted that recently the French sociology of convention has re-addressed normative issues. In doing so, this field follows a tradition not only of Marxism but also of critical theory. Because the French sociology of convention has attracted remarkable attention within and beyond sociology—and considering that other social sciences have not suffered status-downgrading by addressing normative concerns—one may ask what institutional theory has to say in that respect. According to John W. Meyer, there is scarcely any reasonable alternative for sociology but to restrict itself to the role of a neutral, seemingly disinterested observer of the cultural formation of contemporary society. Nonetheless, with an opportunistic attitude one may derive from his analyses what could be a socially feasible normative standpoint. Based on Meyer’s description of a world culture that is essentially composed of norms of rationality, fairness and empowerment—and against the fact that rationality and issues of efficiency appear to be the domain of economics, which can hardly be ­challenged—one may conclude that fairness and related considerations can serve as a normative frame of reference. Accordingly, sociology may find major foci in persistent inequalities and their legitimation bases, in issues of fairness and non-discrimination, and in concerns about participation and inclusion/exclusion. The less academically prestigious field of mission-oriented research indicates that these topics are indeed those in which sociological competencies are the most demanded—in education research and welfare-state issues, and, at the level of organizations, in research on equal opportunity and diversity management. If value-laden, these research lines typically support norms, values and world views that John Meyer and related scholars would view as the

202

Hasse

dominant world culture. Very much in line with institutional reasoning, sociology might benefit from rigorously addressing social problems that are related to norms of world culture. A second strand of normative involvement is related to the social construction of rationalized and further rationalizing actors. It has been noted previously that a convincing enactment as agent for others results in high status, whereas the lack of actor qualities is viewed as problematic. In that respect, an actor’s status has normative connotations. Anything viewed as contributing to the establishment, strengthening or maintaining the actor status of organizations, states or individuals can accordingly expect societal support in terms of legitimation and resources. The normative implication here suggests a research programme that provides valuable knowledge to others—either for experts (in therapy or consulting) or directly to potential clients. Appropriate research objects, then, are failed states, irrational organizations and affectual or weakwilled individuals—and again, we find that these are major research topics in political sociology, management research and the sociology of social problems. A third normative perspective is not related to John Meyer’s world polity construct. Instead, the classical contribution of Phillip Selznick (1957) serves as a frame of reference. Recently, scholars have recalled Selznick’s approach and aimed to use this perspective in contemporary research (see Krygier, 2012). One promising entry is Selznick’s notion that to organize a given organization is to infuse it with value. Value infusion is viewed as leadership at any level. According to Selznick, value infusion begins where measuring and accounting ends. However, it must be noted that approximately 50 years after Selznick’s main contributions, little remains that has not become the object of formal procedures of accounting and evaluation. Moral issues such as equality and fairness have been transformed into diversity management programs that can be evaluated, and what used to be considered standard business ethics is now certified as corporate social responsibility. This extension of the technical domain restricts the scope of the non-technical aspects of an organization. According to Selznick, this development may raise moral concerns about technocratic tendencies and a farewell to value-oriented leadership in the sense of Selznick. Against the background of these potential entries into normative issues we conclude that institutional analysis need not restrict itself to remaining merely a neutral observer. In particular, the work of John Meyer and Phillip Selznick offers answers to the question of how an entry into normative issues might appear. Nonetheless, it remains to be seen whether adopting and strengthening values that are culturally ‘evaluated as valuable’ can be combined with rigorous analyses, and if not, what will (and, normatively, should) be viewed as the primary mission of institutional research.

About Actors: An Institutional Perspective

203

References Berger, P.L. and Luckmann, T. (1967). The social construction of reality. New York, n.y.: Doubleday. Boli, J. and Thomas, G.M. (1997). “World culture in the world polity: a century of international non-governmental organization”, American Sociological Review. 62(2): 171–190. Boltanski, L. and Thévenot, L. (2007). Über die Rechtfertigung. Hamburg: Hamburger Edition. Bromley, P., Hwang, H. and Powell, W.W. (2012). “Decoupling revisited. Common ­pressures, divergent strategies in the u.s. nonprofit sector”, Management. 15(5): 468–501. Bromley, P. and Meyer, J.W. (2015). Hyper-organization. Global organizational expansion. Oxford: Oxford University Press. Coleman, J. (1982). The asymmetric society. Syracuse, n.y.: Syracuse University Press. DiMaggio, P.J. and Powell, W.W. (1983). “The iron cage revisited: institutional isomorphism and collective rationality in organizational fields”, American Sociological R ­ eview. 48(2): 147–160. Djelic, M.-L. (2010). “Institutional perspectives—Working towards coherence or irreconcilable diversity?”, Eds. G. Morgan et al. Comparative institutional analysis. Oxford: Oxford University Press. Finnemore, M. (1996). National interests in international society. Ithaca, n.y.: Cornell University Press. Flora, P. (1986). Growth to limits: the western European welfare states since world war ii. Berlin: deGruyter. Friedland, R. and Alford, R. (1991). “Bringing society back in: symbols, practices, and institutional contradictions”, Eds. W.W. Powell and P.J. DiMaggio. The new institutionalism in organizational analysis. Chicago/London: University of Chicago Press. Granovetter, M. (1985). “Economic action and social structure. The problem of embeddedness”, American Journal of Sociology. 91(3): 481–510. Habermas, J. (1973). Legitimationsprobleme im Spätkapitalismus. Frankfurt a.M.: Suhrkamp. Habermas, J. (1987). Theory of communicative action. Vol. 2. Boston, ma.: Beacon Press. Hallett, T. and Ventresca, M.J. (2006). “Inhabited institutions. Social interaction and organizational form in Gouldner’s patterns of industrial bureaucracy”, Theory and Society. 35(2): 213–236. Hamilton, G. and Biggart, N.W. (1984). Governor Reagan, governor Brown. New York: Columbia University Press. Hannan, M.T. and Freeman, J.H. (1977). “The population ecology of organizations”, American Journal of Sociology. 82(5): 929–964.

204

Hasse

Hasse, R. (2003). Wohlfahrtspolitik und Globalisierung. Zur Diffusion der World Polity durch Organisationswandel und Wettbewerbsorientierung. Opladen: Leske und Budrich. Hasse, R. and Krücken, G. (2014). “De-coupling and coupling in education”, Eds. B. Holzer, F. Kastner and T. Werron. Isomorphism and differentiation. From globalization to world society. London: Routledge. Hasse, R. and Wyss, S. (2015). “Emotionalität als Mythos und Zeremonie. Zur Bedeutung emotionaler Ausdrucksfähigkeit und Selbstthematisierung im Kontext schulischer Beurteilung”, Ed. M.S. Maier. Organisation und Bildung. Wiesbaden: vs. Hirschman, A.O. (1977). The passions and the interests: political arguments for capitalism before its triumph. Princeton: Princeton University Press. Hochschild, A.R. (1983). The managed heart. Commercialization of human feeling. Berkeley: University of California Press. Illouz, E. (2007). Cold intimacy. The making of emotional capitalism. Polity Press: Cambridge. Krygier, M. (2012). Philip Selznick. Ideals in the world. Stanford, ca.: Stanford University Press. Lawrence, T., Suddaby, R. and Leca, B. (eds.). (2009). Institutional work. Actors and agency in institutional studies of organizations. Cambridge, ma.: Cambridge ­ ­University Press. Luhmann, N. (1997). Die Gesellschaft der Gesellschaft. Vol. 1–2. Frankfurt a.M.: Suhrkamp. Meyer, J.W. (1986). “The self and the life course: institutionalization and its effects”, Eds. A. Sorensen, F.E. Weinert and L.R. Sherrod. Human development and the life course. Hillsdale: Erlbaum. Meyer, J.W. (1987). “The world polity and the authority of the nation-state”, Eds. G.M. Thomas et al. Institutional structure. Constituting state, society, and the individual. Newbury Park, ca.: Sage. Meyer, J.W. and Hannan, M. (eds.). (1979). National development and the world system. Chicago, il.: University of Chicago Press. Meyer, J.W. and Jepperson, R. (2000). “The ‘actors’ of modern society: the cultural construction of social agency”, Sociological Theory. 18(1): 100–120. Meyer, J.W., Boli, J., Thomas, G.M. and Ramirez, F.O. (1997). “World Society and the ­Nation-State”, American Journal of Sociology. 103: 144–181. Moe, T.M. (1984). “The new economics of organization”, American Journal of Political Science. 28(4): 739–777. oecd (1981). The welfare state in crisis. Paris: oecd. Pierson, C. (1991). Beyond the welfare state?. University Park, pa.: Pennsylvania State University Press. Powell, W.W. and DiMaggio, P.J. (eds.). (1991). The new institutionalism in organizational analysis. Chicago/London: University of Chicago Press.

About Actors: An Institutional Perspective

205

Powell, W.W., Whittington, K.B. and Owen-Smith, J. (2009). “Networks, propinquity and innovation in knowledge-intensive industries”, Administrative Science Quarterly. 54(1): 90–122. Schneiberg, M. and Clemens, E.S. (2006). “The typical tools for the job: research strategies in institutional analysis”, Sociological Theory. 24: 195–227. Scott, W.R. (1995). Institutions and organizations. Thousand Oaks, ca.: Sage. Selznick, P. (1957). Leadership in administration: a sociological interpretation. Evanston, il.: Row Peterson. Stinchcombe, A.L. (1965). “Social structure and organizations”, Ed. J.G. March. Handbook of organizations. Chicago: Rand McNally. Strang, D. and Meyer, J.W. (1993). “Institutional conditions for diffusion”, Theory and Society. 22(2): 487–511. Strang, D. and Soule, S.A. (1998). “Diffusion in organizations and social movements: from hybrid corn to poison pill”, Annual Review of Sociology. 241(1): 265–290. Suchman, M. (1995). “Managing legitimacy: strategic and institutional approaches”, Academy of Management Review. 20(3): 571–610. Sutton, R.L. (1991). “Maintaining norms about expressed emotions”, Administrative Science Quarterly. 36(2): 245–268. Weber, M. (1972). Wirtschaft und Gesellschaft. Tübingen: Mohr. White, H.C. (1992). Identity and control: a structural theory of social action. Princeton, n.j.: Princeton University Press. Wilenski, H.L. (1975). The welfare state and equality. Berkeley, ca.: Berkeley University Press. Zilber, T.B. (2002). “Institutionalization as an interplay between actions, meanings and actors: the case of a rape crisis center in Israel”, Academy of Management Journal. 45(1): 234–254.

chapter 12

The Family and Interwoven Concepts Håkon Leiulfsrud Introduction The family is a fuzzy concept. It exists as an empirical reality and social practice but it remains a mystery as a social fact in all of its diversity, types and forms. It is also an organisation unlike most organisations with specific tasks and goals, having a broad range of functions and a low degree of functional differentiation. Hence, to address the family concept we must differentiate between theoretically derived conceptions and common-sense-based conceptions and narratives. Whereas most people are qualified to present us with interesting insights in terms of their own experiences, theories may help us both distance ourselves from our everyday life and ask critical questions without obvious answers. For example, what are the theoretical implications of viewing the family as ‘[t]he greatest welfare system ever devised’ (Bittman and Pixley, 1997: Ch. 9); as an arena for individuals to maximise their utility based on fundamental preferences (Becker, 1993); as a locus for patriarchy and masculine dominance (­Bourdieu, 2001); or as a locus of self-realisation, intimacy and love (Giddens, 1992)? In other words, our theoretical understanding of the family is also part of how we view society and our theoretical lenses. As illustrated by the various theoretical foci set forth above, it is problematic to operate with a universal and encompassing family concept. This is primarily because different theories and research fields pay attention to different research questions and facets of the family. By adopting a specific theoretical focus, a priori we need to pay less attention to other parts of the phenomena under investigation. Although this may enable us to theorise, it also blinds us. Almost regardless, with respect to a universal definition of the family based on functions relevant to a specific theory, the plurality of theories of potential relevance has left social scientists in general and sociologists in particular with a fragmented theoretical understanding of the family. For those seeking productive conceptualisations that are easy to use and discuss with others, this may be viewed as a genuine challenge. For others, the challenge of treating concepts not as constant but as interactive and depending on questions addressed is the key to sociology beyond common sense.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_013

The Family and Interwoven Concepts

207

Most people live in one or more family household(s) over their life span. We need not look to experts or theory to tell us what a family is: indeed, we just know based upon our personal experiences in social interaction with others. A substantial amount of modern family research is also based upon the idea that what is real for actors in terms of family identity is the family in all of its diversity and complexity. Almost regardless, whether we argue for or against a common-sense version of the family, we are left with a number of intriguing questions about the family’s theoretical status and nucleus in social sciences and it’s assumed functions in analyses of society. What interests us here is how the family concept is generally used in social science; its theoretical roots and origin; links to other clusters of concepts; and the possible implications of different views of the family institution for theoretically based diagnoses of our time and society (i.e., our ‘social construction of reality’).

Roots and Origin

‘Family’, a term that is derived from the Latin, primarily refers to the household unit (The Catholic Encyclopaedia, 2009). The English derivative in former times often referred to Roman law and to ‘the paterfamilias, his legitimate descendants and their wives’, and all persons adopted into his family and their wives (Webster’s Third New International Dictionary, 1961: 820). According to this understanding the concept is framed in a narrative of a patriarchal world order, with a strong emphasis not only on legitimate offspring and their legal rights but also on people living in the same household, including children, wives and those adopted into the family of the male(s) or female(s).1 In more recent language and history both servants and relatives are typically omitted and replaced with a focus on a stable and harmonious union between a man and a woman and their children. This is the leitmotif in both the religious doctrines of the Roman Catholic Church since the Council of Trent of 1545–1563 (The Catholic Encyclopaedia, 2009) and in Victorian family ideals based upon harmony, parenthood and love (Davidoff and Hall, 1987). It is also the backdrop of criticism in the 19th and 20th centuries of the nuclear family with a heterosexual couple and their children as the family norm. In this interpretation, it is not simply the idea of having a partner or children that is under attack but the

1 In Middle English, references were made to ‘familie’ (originating from Latin familia). To the best of our knowledge, the term ‘family’ was first used in English in the 15th century (Merriam Webster, 2016).

208

Leiulfsrud

cultural and ideological norms embedded in the family institution (see, e.g., Poster, 1978; Barrett and McIntosh, 1981; Fletcher, 1988). In modern English, the family is often linked to what is either ‘familiar’ (i.e., well known) or related to personal relationships. In everyday language, ‘familism’ is also used to illustrate a status or solidarity with the family, beyond individual interests or membership in mystical sects. Most of the words associated with family in an English dictionary still refer to its functions in terms of economy, religion, legal bonds, organisation, procreation, romance or style. If we compare Webster Dictionary’s (of the early 1960s, see above) with more recent definitions of the family, we notice a shift whereby the nuclear family is both ‘the basic unit in society traditionally consisting of two parents rearing their children’ but also ‘any of various social units differing from but ­regarded as equivalent to the traditional family’ (Merriam Webster, 2016). In other words, it appears that we now have a more open definition of ‘family’ that recognises alternative types of families. The emphasis upon basic social functions by parent(s) with regard to childrearing at first appears to be a constant but is also gradually shifting, with an increased importance placed on socialisation agents and agencies outside of the family.

Family: the Conceptual Space

For an anthropologist, synonyms for family may include blood, kin, clan, lineage or tribe, that is, organising principles for a unit of people to define their membership and status in terms of birth rights and duties. Family, in the conventional anthropological and sociological interpretation, is typically understood as a relatively stable intergenerational group with both biological and affinal relationships between the members of the group and bonds to a larger kinship network. Other perspectives, more concerned with actual family practice, may argue that the idea of the family as a relatively stable residential grouping is misleading, as exemplified not only by generational transitions but also by the fact that a high percentage of Western families end in divorce and new family constellations. The discussion of whether families may be viewed as relatively stable residential groupings depends on the questions asked. This is not a question of who is right and wrong but of what is theoretically and empirically relevant in analyses of actors and their interactions. The question of family as a relatively stable unit or an institution in flux is also of importance in how we view society as a social system. In the sociological tradition the family is often used as a reference point and proxy for society and social organisation more generally. This is not just

The Family and Interwoven Concepts

209

e­ xemplified in Talcott Parsons’ famous analysis of the 20th-century AngloAmerican family system as an outcome of and adjustment to industrial society (Harris, 1983) but also in more recent attempts to portray the family in the age of modernity and post-modernity (Bauman, 2003). The idea of defining the family by its structures and functions has a long ­history in both anthropology and sociology. A core assumption of both functionalist and Marxist inspired anthropology and sociology is to view the family as a ‘mirror’ or indicator of how society in general is organised. In the functionalist tradition, until the end of the 1950s, the discussion focused on u ­ niversal ­family properties and core functions related to caregiving and procreation (­Coser, 1974). In the Marxist tradition, the family resonated with the view of living arrangements reflecting and affecting the nature of class relations in society (Engels, 1986; Seccombe, 1993). In this theoretical tradition, the family is also viewed as an agent affecting the institutional design of, for example, the legal system, property relations and functional equivalents to the family (in terms of childrearing, education and safety nets). One of the main problems in both the actor and the system narrative of the family is that it is difficult to find common ground that helps view both the family in society and society in the family. To find common ground we have to expand the conceptual space and develop a theoretical vocabulary beyond that of a narrow understanding of the family.

In Search of Common Ground

The European and North American family of today is in most cases treated as a common-sense category among social scientists, i.e., either as the legitimate and dominating household arrangement in society or as an umbrella concept covering different family types. In this perspective, what is observed empirically is also likened with the family as a theoretical category serving (or failing to fulfil) various social functions in society. In principle, sociologists may agree with regard to important theoretical contributions in the field; in most cases, however, they are trapped in their own traditions and schools of thought. Systematic theory building in the field of family sociology mainly occurred after 1950, although unevenly. Serious attempts to develop formal theory in various domains of the family were made in the 1970s in the United States (see, e.g., Burr, Hill, Nye and Reis, 1979; Burr, Leigh, Day and Constantine, 1979). The ideas of formal theory were gradually replaced with a rainbow of theoretical traditions drawing from system theory, ecological approaches and the theory of family development

210

Leiulfsrud

or more interpretive, constructivist and contextual approaches (Klein and White, 1996: 55). System theory represents the most abstract and general way to consider the family complex in terms of its structures and functions in the social system. Although both anthropologists and sociologists have been preoccupied with the universal functions of the family in terms of childrearing and institutional rules regulating sexuality, it is generally believed that the family system evolves through differentiation processes pursuant to which the relationship with society as a whole changes. In Talcott Parsons’ interpretation, this notion was expressed in terms of a functional fit between industrial society and the institutional arrangement of the family of the 1950s (Parsons, 1964). Thus, Parsons is often accused of presenting a static picture of the family system, although it can also be argued that his functional differentiation approach to the family system is complex, innovative and thought-provoking (Morgan, 1975). Leaving aside Parsons’ empirical narratives, which have a strong bias towards a white middle-class practice and the prevailing ideals of the 1940s and 1950s in the us, we observe a more dynamic model of potential relevance today. In Parsons’ conceptual scheme, the family is understood as a theoretical synthesis between what is occurring inside the family and systemic factors, reinforcing each other’s influence and reciprocating each other’s internal and external effects. The sociologist’s primary task in this type of theorising is to identify how certain base functions (such as child rearing and stabilisation of family members’ everyday life) are articulated within the family sphere and between the family sphere and other societal spheres. One of Parsons’ most interesting, but often misunderstood contributions to the family system is that of ‘family connections’ (see also Morgan, 1975, 1996). Following this line of thought raises the empirical question of how childrearing and traditional family tasks such as household chores are organised. They may, for example, be performed within the family, through personal networks or voluntary organisations in which people exchange services, via market-based services, or within organisations in the realm of the ‘welfare state’. One difficulty with this type of functional reasoning, however, is that empirical discrepancies and ‘imperfections’ between what is theoretically expected and what is empirically observed are under-problematised. In Parsons’ analysis of social actions, we find an underlying model in which actors have a mode of orientation based on preferences governing their goals and actions. That said, what if these actions occur in a more complex unit, such as the family, that is larger than the aggregate of all actions? In this type of reasoning, social interaction between individuals and families (actors) are often conflated with the social logic and rationality of the social system.

The Family and Interwoven Concepts

211

This is a familiar criticism in the symbolic interactionist tradition, which not only focuses on social interaction as an on-going process between actors trapped in institutionally embedded family practices, rules and traditions but also exhibits a substantially higher degree of surprises and ‘unintended consequences’ than the standard operationalisation of the family in the functionalist tradition. Not even Parsons, however, would claim that the family might be fully reduced either to individual actions or to the aggregate of these actions. There are always layers of social interactions beyond what may be observed in any theoretical operation. In this respect, Parsons is in the same boat as most other social theorists. What is noticeable in Parsons’ conceptual space of the family is that it is primarily based on ‘ideal types’ (not to be conflated with ideals or empirical data) framed in a language of norms and roles. Based on Parsons’ writing, this may be viewed as a type of idealisation based on a static and iconic idea of the North American (‘white’) middle-class family of the mid-20th century. However, it may also be viewed as a ‘mirror’ and crystallisation of any family form and type that corresponds to the development of a post-industrial society. This understanding of the family is essentially in line with Durkheim’s status as a ‘social fact’, or even more accurately described as a ‘social thing’ (Lemert, 2005). One major problem in Parsons’ later writings is that the actors’ meanings associated with goals and interests tend to be de-coupled from the actors’ decisions about general notions of goal attainment or interest gratification (Bauman, 1978: 136). One consequence clearly expressed in Parsons’ The social system is a theoretical framework built upon actors’ more or less rational behaviour given a ‘phenomenological universality’ (Baumann, 1978: 137). This may be an issue of potential interest in terms of seeking the essence of the family and establishing a common theoretical ground. Nonetheless, it is also a way of theorising that throws out the actors and the social interactions (as opposed to systemic interactions) that constitute the family system as a concrete living arrangement. It is also a theoretical framework that tends to underemphasise conflicts and the darker aspects of family life (such as alienation, family violence and various types of deprivation) while overemphasising the avoidance of deprivations and the importance of solidarity and ‘tension management’.

Theoretical Constructions of the Family as a Locus of Equilibrium and Struggle

One of the most intriguing questions for both feminist scholars and Talcott Parsons is the question of ‘the hidden constants amnesia’ (Bourdieu, 2001)

212

Leiulfsrud

whereby the traditional gender division of work (including housework) tends to be surprisingly difficult to change in both Europe and North America. Although there has been a shift in European public opinion suggesting a strong belief in gender equality and a wish to eliminate injustices, it is scarcely a complete shift or ‘revolution’ (Esping-Andersen, 2009) with regard to the objective and cognitive structures associated with the genders. The historical mechanisms that potentially maintain this stability and neutralise the importance of history and political struggles are obviously of interest in any discussion of gender, but equally so in theorising the family as a living arrangement. Esping-Andersen’s (2009) argument is that gender inequality both inside and outside the family sphere is suspended between two types of equilibrium manifested in the traditional nuclear family (and the bread-winner model) and the ‘gender-equality equilibrium’. Whereas the first equilibrium may be regarded either as the historical result of a modernisation process (see Parsons’ argument above) or as an efficient way of organising family welfare (Becker, 1993), the second equilibrium may be explained both by increased female labour market participation and by males’ greater involvement in household work and childrearing. The sociological problem, as it is presented by Esping-Andersen (2009, 2011), resides both in how to explain the prevalence of the female-housewife equilibrium and in the fact that many family households represent a combination of the two equilibria. Feminist sociology and scholarship is often formulated in opposition to and as an alternative to a functional analysis of the family. It is also presented as a ‘rethinking’ of the family (Thorne, 1992) and as addressing critical questions about how we conceptualise the family as an institution and organisation. Whereas Parsons constructs a family concept, asking why the family system is relatively stable in modern industrial societies, feminist scholars often focus both on potential struggles and conflicts within the family and on the family connections that ultimately constitute the family in society. The argument is that as individuals, we ‘enter into productive, reproductive, and consumptions relations’ through families and that ‘the two genders enter them differently’ (Rapp, 1992: 51). Whereas Parsons, Becker and EspingAndersen are focused on the complementarity of family roles and the symmetrical family, one key assumption in feminist thought is that the family system is rooted in patriarchal power relations, subordination, ‘relations of ruling’ (Smith, 2005), and everyday practices that systematically favour male dominance in society. Of course, the feminist entry is rooted in personal experiences and a history of living inside or outside the dominant family system. Although the legacy

The Family and Interwoven Concepts

213

of pioneers such as Simone de Beauvoir (1972) is profound in terms of how to view gender as the mirror of the Other, the main contribution of feminist scholarship surpasses conventional sociology and family studies, generating a broader discussion of family types and family arrangements, including samesex unions. This has resulted in a more critical view of hegemonic family practices and ‘gender contracts’ (Davidoff and Hall, 1987; Hirdmann, 1990) in which gender roles and normative rules referring to expectations regarding rights, duties, responsibilities and properties are negotiated and enacted. Feminist family sociology has also contributed to a shift in focus on the family complex with respect to our understanding of power, subordination and oppression. In contrast to the previous discussions in which family life was viewed though the prisms of class, gender or ethnicity, it is the interplay of these stratification mechanisms that operates and defines gender-based family contracts.

Why, How and for Whom is a Family Concept Important?

At a fundamental level, we have to ask ourselves the following question: why do we need a family concept? Do we need it to explain social behaviour related to established family roles, norms and practices, or is it used as a proxy to explain human behaviour more generally? Are there any concepts that could potentially replace the family and if so, what is gained and lost? Is the concept of family used as metaphor, an empirical category, or merely as shorthand for what is supposed to be well known and obvious? Whereas certain feminist scholars, based on the diversity in form and family practices, pose the question of ‘Is there a family?’ (Collier, Rosaldo and Yanagisako, 1992), it would be foolish to abolish the family concept in social theory. Despite its variations in form and content, the family is both a ‘social fact’ and an organising principle for its members and other members of society, with ramifications for organisations, social networks and individuals (see Ahrne’s chapter in this volume). It is also an interesting example of social organising that is not confined to individual and particular family practices but instead of deeply embedded in how societies are organised. For both ordinary people living in various types of family constellations and legal scholars, it may be counterproductive to ask why we need a family concept in the first place. Families consist of individuals who may perceive their family relations differently, but it is equally obvious that the family is a living social arrangement in its own right with legal implications for family

214

Leiulfsrud

­ embers, including taxation status and rights to social benefits. In the case of m the law and the legal system, we find official views of how to regulate marriage and the relations between parents and children, including adoption and surrogacy. Family law is also a good proxy for how societies incorporate and regulate both non-marital family relations and same-sex marriage. One of the main functions of family law is to regulate the formal establishment of the family and secure the rights of family members and inheritors when a marriage or civil union is dissolved. The law, and what counts as a family in the legal system, is therefore relevant to how we define formal family affiliations and membership. The fact that contemporary family legislation in several European nations—including the Nordic countries—is less concerned with how we live and moral duties and more concerned with the contractual dimension is relevant to how we view the family as a ‘social fact’. In other European nations, marriage is still institutionalised as a living arrangement in which the partners are obliged to consult each other and take the other partner’s interest into consideration in the event of a new job or a change of residence (Lødrup and Sverdrup, 2009: 23). Both types of family legislation are of theoretical interest because they involve different family narratives and questions related to rights and moral duties. These differences have inspired both feminists and comparative welfare researchers (Esping-Andersen, 1999) to use the degree of ‘de-familisation’ as a theoretical proxy for different types of welfare regimes and welfare arrangements. This is a good example of how a family narrative, based on its functions as both a welfare agency and a ‘gender contract’, is used to theorise with regard to even larger narratives such as ‘conservative’, ‘liberal’ and ‘social democratic welfare regimes’. The caring function of the family and varieties of ‘familism’ have also been used by comparative welfare researchers in an attempt to bring gender relations into the core of the theoretical discussion of the nexus between the family, the state and the market (Lister, 1994; Sainsbury, 1999; Leitner, 2003; Bambra, 2007; Cho, 2014). This theoretical turn has broadened the discussion of the family and hegemonic family models in different societies with a sharper focus on gender roles (Yodanis, 2005). Nevertheless, we also observe a tendency among many welfare researchers to treat hegemonic family arrangements and gender contracts as an object wherein the family concept is closer to an icon than an interactive tool (for a useful discussion, see Cho, 2014). As shown in the examples above, we find that numerous research fields depend on a productive family concept to study or address the family system in practice. Because of the type of questions that are relevant, what is viewed as productive by certain researchers may be viewed as a weakness by others. References to different schools of thought confirm the view that the family is an important concept that serves different purposes.

The Family and Interwoven Concepts



215

Missing Links and Questions

Based on the discussion above, Talcott Parsons’ view of the family as a ‘system unit’ remains of interest to how we perceive gender and family contracts and organisations that can perform similar or supplementary functions. What befalls traditional roles and value systems, as perceived by welfare-regime researchers, when they are fully or partly (re) organised in voluntary organisations, in market-based transactions or by agencies of the welfare state? This question also leads us to ask how this reorganisation is manifested in the family in terms of affective and solidarity bonds, and how it may potentially shape other organisations based on universalism and functional specificity. It is of further interest to ask questions about external family connections. For example, kindergartens and schools seldom operate in social isolation: they both affect family practices and childrearing standards and are affected by the families and other agents they serve. One limitation of a family concept based on ideal typical and functional reasoning is that it misses the fact that the question of whether families are organised in ‘joint’ or ‘segregated’ role relationships is empirical in nature (Bott, 1957). This fact is of importance in how we traditionally discuss gender equality, namely, based on ideals of joint roles with a high degree of common activities and a low degree of task differentiation. Elisabeth Bott’s (1957) analysis of family organisation is also important because it proposes at least two, and possibly three family models. The first model, based on joint role relations, corresponds to an idea of the family arrangement as one unit with a common understanding of tasks and duties beyond individual interests. The second model, based on segregation, favours the feminist argument that men and women enter and perceive the family differently but also notes a potentially higher degree of individuation and social distance. A third option, found in Bott’s data but not described in detail, is to perceive family organisation as both joint and segregated depending on the questions addressed.

Built-in Family Ambivalences

One serious problem in standard sociological conceptualisations of the f­ amily is that the family tends to be discussed using two relatively ­one-dimensional narratives. The first narrative is commonly based on the family system or marriage as the ‘ideal type’, whose family members and various roles and tasks exhibit a number of common attributes. The second narrative is one of ­diversity and difference in family types, which are not perceived as a functional difference.

216

Leiulfsrud

The idea that family may be perceived and lived in flux between harmony and conflict, solidarity and schism, loyalty and betrayal, a sacred heaven or a prison is well documented in biographies and fiction. These dualities are also familiar in film and art but more rarely are treated as social ambiva­ lences among sociologists in their professional capacity. To paraphrase Charles Lemert (2005), as sociologists we are trained to spot and analyse the ‘mysterious power of social structures’, but in professional practice ‘the lively subjects’ tend to be either conflated with what we observe (in a naïve type of realism) or forgotten in highly abstracted theoretical reasoning (Lemert, 2005: x). To better grasp these built-in tensions and contradictions, we must develop a theoretical vocabulary that has the ability to observe and recognise tensions and contradictions beyond stereotypical reasoning and conventional truths. This is also the ambition of scholars who argue that we must look at concrete family practices to develop a theoretical vocabulary of how social structures and norms operate in everyday life and everyday interactions. This may help us enhance our understanding of gender relations inside and outside of the family sphere beyond stereotypes (Morgan, 1996; Smith, 2005). Although theorising based on conceptual fixation may help sharpen our ­focus, it also increases the risk of under-communicating contradictions and ambivalences. The goal of paying more attention to theoretical a­ mbivalences is that it may open up for question, for example, why women in family ­relationships based on violence, sexual abuse and social misery remain in these relationships. The concept of love illustrates how both social scientists and philosophers base their theorising on a narrative of personal connections rather than on a pendulum between connections and disconnections, togetherness and loneliness, love and hate, intimacy and the fear of losing either the person you love or the illusion of love. It is also a story of how our ­theoretical language of love has been trapped in an idealisation of mutual reciprocity between two or more persons, without a theoretical vocabulary of potential dysfunctions such as physical assault, rape, stalking or a lack of mutual recognition. Concepts interwoven with the family such as fatherhood, motherhood and childhood have received a great deal of attention since the 1990s, revealing hegemonic practices and different types of fatherhood, motherhood and ­childhood. The same may also be viewed with respect to the concepts of masculinity and femininity and how they are treated as socially constructed. What is more surprising is how little attention has been devoted to social constructions or deconstructions of established family concepts in sociology. Researchers in the symbolic interactionist tradition may show us how people perceive their family members but it is rare to find systematic analyses that perform explorative analyses of theory of the family or theoretically based

The Family and Interwoven Concepts

217

family narratives revealed in white papers or official reports (for a notable exception and a sociologist who has explored the family from these angles, see Morgan, 1975). Among the best illustrations of the complexity of the family are found in personal diaries and correspondence, biographies and fiction, not in theory or empirical research. Because our ideas about the family are largely taken for granted or rooted in our own perceptions or family biographies, it may very well be that we do not ask the interesting and critical questions to propose alternative interpretations and new ways of theorising the family. Discussion Sociology is at its best in studies of the family as a relational structure, in terms of internal roles and social positions, or as a social system interwoven in a wider relational web of groups, networks, organisations or the state (see, for example, Ahrne’s chapter in this volume). Empirically informed research may also potentially enable us to develop a language of family-specific interaction patterns and living conditions (see Martinussen’s chapter in this volume). Sociology is more poorly equipped theoretically in studies of the family as a cultural structure with acting and interacting members sharing a history, an idea of uniqueness and sameness, paired with a language and vocabulary of its own. Sociologists from Durkheim onwards have obviously been aware of the family system in terms of not only its positive and integrative functions but also its dysfunctions and pathologies (Durkheim, 1984). Sociologists preoccupied with the social organisation of the family have also addressed interesting questions about social disorganisation in terms of inadequacies or failures to live up to the expectations associated with the family as a living arrangement (Komarowsky, 1967: Ch. 15) or more specifically, about the question of divorce (Yodanis, 2005; Kalmijn, 2013). Sociologists such as Francesco Alberoni (1983) have developed theoretically interesting distinctions between ‘love’ (as regulated and institutionalised) and ‘infatuation’ (as a phase of social disorder), but sociological theory appears to exhibit a surprisingly large number of holes in how we treat the family as a social system. Although the distinction between organisation and disorganisation may help us illuminate inadequacies and failures in role performances, it may also leave us with a static view of the family as a living arrangement and life project in which the fulfilment of roles or family obligations or measures of success and failure may vary considerably over time.

218

Leiulfsrud

Measures of success from a family-policy perspective built on egalitarian norms and a sense of mutual recognition for work inside and outside of the household are particularly important in a Nordic context. This is good illustration of how the family is manufactured by both family members and familypolicy measures to break traditional gender-based barriers. This is obviously an issue of relevance in attempts to steer gender and family roles towards increasing paternal involvement in childcare. It is also a nice illustration of how active policy interventions may change the conventional view of mutually exclusive yet complementary gender roles (see Parsons, 1980; Komarowski, 1967), forcing us to develop a new theoretical vocabulary of family interaction. As long as the dominating culture embedded in the Western and AngloSaxon family system continues to be built on the norms of freedom to choose your partner and live your life in privacy, it is not merely the official or wellknown norms that are of interest but also the more latent dimensions that are not always revealed to the public. This includes an increased interest in ­difficult-to-steer topics such as the functions of joint family stories and memories, ­family secrets—including family taboos that are not to be discussed outside the family sphere—or the meaning and function of love beyond public or heavily politicised clichés. Among the most fascinating aspects of both Parsons’ family concept and the critique of his functional approach to the family institution and its organising is that they appear to be grounded in seemingly conflicting utopian ideals regarding society and the possibility for social actions to occur, including ‘the transcendental conditions of that possibility’ (Bauman, 1978: 130). In the case of Parsons, the conventional view shared by many sociologists is that his is a static and overly deterministic theoretical framework with an emphasis on passive agents adjusting themselves to the established social order (Morgan, 1975; Cheal, 1991). This view is also a criticism of the idea of family convergence at the expense of family divergence (Cheal, 1991). Instead of having a family concept that potentially explains possible links between the family and the society in which it operates, the idea is to liberate the family from the norm and the power structures traditionally associated with it. This may lead us to ask interesting questions about family members with overlapping or different interests as members of various family arrangements. It may also, in terms of social citizenship and human rights, lead to questions related to seemingly universal actors and interests, blurring gender and other built-in power relations. The literature on the family, particularly in the field of family therapy, has been good at developing client metaphors and metaphors to change people’s behaviour. Metaphors may help us not only theoretically condense important properties of what we believe is at the core of family but also obscure

The Family and Interwoven Concepts

219

­ npredictable family behaviour or properties outside of the family system. u One possible way to theorise the family system further is to depart from some of the metaphors used in this chapter such as the family as a ‘social thing’ or ‘social fact’ (Durkheim, 1984; Lemert, 2005), a welfare system (Bittman and Pixley, 1997), a locus for patriarchy and dominance (Bourdieu), a locus of selfrealisation, intimacy and love (Giddens) or a primary relationship. Other metaphors at the core of system theory may view the family as a container or a machine (Roseblatt, 1994). The fact that a concept such as the family is complex and fuzzy and ideal for numerous types of metaphors is not a good reason to abolish it. If the fear of fussy concepts and sociological ambivalences were a valid argument, we would have to give up being sociologists. To work with a fuzzy concept is not to be fuzzy or sloppy in how we actually approach the concept, but it does require a theoretical imagination and eye beyond naïve realism in which the family is reduced to what we observe empirically. References Alberoni, F. (1983). Falling in love and loving. New York: Random House Inc. Bambra, C. (2007). “Going beyond the three world of welfare capitalism: regime theory and public health research”, Journal of Epidemiology & Community Health. 61(112): 1098–1101. Barrett, M. and McIntosh, M. (1981). The anti-social family. London: Verso. Bauman, Z. (1978). Hermeneutics and social science: approaches to understanding. London: Hutchinson. Bauman, Z. (2003). Liquid love: on the frailty of human bond. Cambridge: Polity. Beauvoir, S. de. (1972) [1948]. The second sex. Harmondsworth, Penguin. Becker, G.S. (1993). A treatise on the family. Enlarged edition. Harvard: Harvard University Press. Bittman, M. and Pixley, J. (1997). The double life of the family. St. Leonards: Allen & Unwin. Bott, E. (1957). Family and social network. London: Tavistock Publications Ltd. Bourdieu, P. (2001). Masculine domination. Stanford: Stanford University Press. Burr, W., Hill, R., Nye, F.I. and Reiss, I. (eds.). (1979). Contemporary theories about the family. Vol. 1–2. New York: Free Press. Burr, W., Leigh, G., Day, R. and Constantine, J. (1979). “Symbolic interaction and the family”, Eds. W. Burr, R. Hill, F.I. Nye and I. Reiss. Contemporary theories about the family. Vol. 2. New York. Free Press. Cheal, D. (1991). Family and the state of theory. New York: Harwester-Wheatsheaf.

220

Leiulfsrud

Cho, E.Y.-N. (2014). “Defamilization typology re-examined: Re-measuring the economic independence of women in welfare states”, Journal of European Social Policy. 24(5): 201–212. Collier, J., Rosaldo, M. and Yanagisako, S. (1992). “Is there a family?: new anthropological views”, Eds. B. Thorne and M. Yalom. Rethinking the family: some feminist questions. Boston, ma.: Northeastern University Press. Coser, R.L. (ed.). (1974). The family, its structures and functions. 2nd edition. London: The Macmillan Press. Davidoff, L. and Hall, C. (1987). Family fortunes: men and women of the English middle class 1780–1850. London: Hutchinson. Durkheim, E. (1984) [1893]. The division of labor in society. New York: Free Press. Engels, F. (1986) [1884]. The origin of the family, private property and the state. London: Penguin. Esping-Andersen, G. (1999). Social foundations of postindustrial societies. Oxford: Oxford University Press. Esping-Andersen, G. (2009). The incomplete revolution: adapting to women’s new roles. Cambridge: Polity Press. Esping-Andersen, G. (2011). Family formation and family dilemmas in contemporary Europe. Bilbao-Spain: Fundación bbva. Fletcher, R. (1988). The abolitionists: the family and marriage under attack. London: Routledge. Giddens, A. (1992). The transformation of intimacy: sexuality, love and eroticism in modern societies. Cambridge: Polity Press. Harris, C.C. (1983). The family and industrial society. London: George Allen & Unwin. Hirdman, Y. (1990). The gender system: theoretical reflections on the social subordination. The study of power and democracy in Sweden. Report no. 40. Uppsala: Maktutredningen. Kalmijn, M. (2013). “Long-term effects of divorce on parent–child relationships: within—family comparisons of fathers and mothers”, European Sociological Review. 29(5): 888–898. Klein, D. and White, J. (1996). Family theories: an introduction. Thousand Oaks: Sage. Komarowsky, M. (1967). Blue-collar marriage. New York: Vintage Books/Random House. Leitner, S. (2003). “Varieties of familiaism. The caring function of the family in comparative perspective”, European Societies. 5(3): 353–375. Lemert, C. (2005). Social things. Oxford: Rowman & Littlefield Publishers. Lister, R. (1994). Citizenship: Feminist Perspectives. London: Macmillan Education. Lødrup, P. and Sverdrup, T. (2009). Familieretten. [Family law]. Department of Law. University of Oslo. Merriam Webster (2016). “Family”. Downloaded 02.05.2016 from: www.merriam-­ webster.com/dictionary/family?show.

The Family and Interwoven Concepts

221

Morgan, D. (1975). Social theory and the family. London: Routledge & Kegan Paul. Morgan, D. (1996). Family connections. Cambridge: Polity Press. Parsons, T. (1964) [1943]. “The kinship system of contemporary United States”, Essays in sociological theory. New York: The Free Press. Parsons, T. (1980) [1955]. “The isolated conjugal family”, Ed. M. Anderson. Sociology of the family: selected readings. Harmondsworth-Middlesex: Penguin Books. Poster, M. (1978). Critical theory of the family. London: Pluto Press. Rapp, R. (1992). “Family and class in contemporary America: notes towards an understanding of ideology”, Eds. B. Thorne and M. Yalom. Rethinking the family: some feminist questions. Boston, ma.: Northeastern University Press. Rosenblatt, P.C. (1994). Metaphors of family systems theory: towards new constructions. The Guildford Press: New York and London. Sainsbury, D. (ed.). (1999). Gender and welfare state regimes. Oxford: Oxford University Press. Seccombe, W. (1993). Weathering the storm: working-class families from the industrial revolution to the fertility decline. London: Verso. Smith, D.E. (2005). Institutional ethnography: a sociology for people. Lanham: Altamira Press. The Catholic Encyclopaedia (2009). “Family”. Downloaded 02.05.2016 from: www. newadvent.org/cathen/05782a.htm. Thorne, B. (1992). “Feminism and the family: two decades of thought”, Eds. B. Thorne and M. Yalom. Rethinking the family: some feminist questions. Boston, ma.: Northeastern University Press. Webster’s Third New International Dictionary (1961). “Family”. Vol. 1. Chicago: Encyclopedia Britannica Inc. Yodanis, C. (2005). “Divorce culture and marital gender equality: a cross national study”, Gender & Society. 19(5): 644–659.

chapter 13

Collective Action: Why is it so Difficult for the Social Sciences to Grasp the Rational Aspects of Collective Action? Roar Hagen Introduction For some time, I have explored the idea that the conflict between collective and individual rationality is a vehicle for societal integration (Hagen, 1999, 2000, 2006). As with other important concepts, sociology contextualizes integration in the vocabulary of both the system and action. As outlined in the book introduction, the systems approach to societal integration identifies some degree of compatibility between the parts and whole. An action theoretical account describes the degree of participation of actors in some larger whole. A contentious approach is to combine the two vocabularies. Functions relate whole and part and belong to vocabularies of systems, while collective action looms large in vocabularies of action. Moreover, there is a close association between the two concepts, as functions are collective goods, and accordingly, only collective action can provide systems with functions. My suggestion is that the concept of collective rationality can connect the two perspectives. The rationale behind this proposal takes off from the general assumption that collective action constitute societies. Actions are the causes of imbalances between the parts and whole and create function problems. To achieve integration, individuals must change their behaviour, and rational actors should be willing to give up their dysfunctional behaviour and act in accordance with the needs of society, if they benefit from the conversion. The idea is that rational collective action has the capacity to transform both individuals and systems and eventually societies. Collective rationality is, however, presently a residual category and not a scientific concept. The prospect of collective action confronts participants with a conflict between collective and individual rationality, between strategies that benefit everybody in the group and strategies that are most beneficial to individuals acting alone. The common solution to this problem of collective action is for the social sciences to give up on the concept of a collective rationality. Collective benefits are of no avail in explaining collective action, and

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_014

Collective Action

223

­vocabularies of action have proved to be unable to establish an actor’s perspective on collective action. Without micro foundations, vocabularies of systems resort to functionalism, and they assume that social systems exist because they are useful to society but are unable to explain how societal usefulness become motives for actions. To develop collective rationality from a residual category to scientific concept, we must change the present vocabularies of system and action through the following steps: First, I elaborate on the close connection between functions, collective goods and rational collective action in theories of modern society. Second, I show how sociology reduces collective rationality to either individual rationality or normative commitment. In both cases, the subjective perspective to collective action is lost and compensated by functionalist reasoning. Third, I locate the reason for these theoretical shortcomings in certain ontological assumptions about the social. Fourth, I establish an alternative social ontology that enables us to reconstruct collective rationality as the subjective perspective of social systems of collective action, which solve functions problems for society.

Functional Differentiation

Sociological theories of modern society agree on some form of functional differentiation, where society is divided into several more or less autonomous functions systems (Luhmann), fields (Bourdieu) or discourses (Foucault, Habermas). The concepts are different, but they refer to about the same entities and activities, such as the economy, politics, science, education, the mass media, art, sports, and religion. Each system fulfils a particular task or function for society. For instance, education gives society the capacity to supply the other systems with qualified labour; science furnishes society with the capacity to produce new knowledge; the task of the economy is to secure the efficient allocation of scarce resources; and the political system provides society with the capacity to make collectively binding decisions. The more or less explicit assumption is that this separation of capabilities is useful to society. Specialization speeds up the development of each sector and enables greater overall societal capacities and complexity; it also creates dependencies. Each part becomes dependent on the performances of the other parts, and individuals become dependent on participating in most if not all the function systems for their own life. Somehow, the societal process of differentiation, specialization and inclusion must be balanced. The question then

224

Hagen

arises if, how and eventually to what extent modern society can regulate its parts. To this question, sociology offers different and contradictory answers. The response from the new systems theory of Niklas Luhmann (2012, 2013) is negative. Modern society consists of autonomous function systems that interpret and react to irritations in the environment according to their internal logics, and there is no additional entity within society that can act as a super system and manage the operations of the subsystems. Foucault counters Luhmann’s theory of negative integration with the concept of hyperintegration, where modern society is considered a matrix of power relations emanating from the need to defend society. There is, however, no actor, individual or collective at the controls, and the ‘dispositif’ governs behaviour without anybody’s intention to do so (Foucault and Gordon, 1980). Between these extremes, we find Bourdieu and Habermas, who both believe in the self-steering of society through collective action. Both authors consider the political system to be the nexus of modern society, and political power regulates operations in the other systems or fields. The two theories nevertheless differ in how they understand the relation between society and power. Societal integration through capital and power is a central tenet of the Marxist tradition. Bourdieu (1996) and Wacquant (2005) adopt the class m ­ odel of societal integration but extend the concept of capital, where each field has a particular form of capital. Capital accumulated in the different fields is then converted to political power through social capital, and the struggle for capital in the different fields is regulated to the advantage of those with the most capital. Jürgen Habermas (1996) is a renowned theorists on societal integration through democratic means. The political systems belong to the democratic circle of power, and people react to the negative effects of function systems through public discourse and create social movements that develop arguments, which career-oriented politicians convert to social reform.

Rational Collective Action

Differences in the answers to the question concerning steering in modern society highlight differences in how theories appreciate the possibility for rational collective action. Both Habermas and Bourdieu operate with groups of people who act together to improve their societal conditions. The group might be a small section of the upper class or a large group that embraces the ‘people’.

Collective Action

225

In both cases, the goal of concerted action is a common or collective good, as in the case of political reform, for instance. According to Luhmann (1995: 198–201), the capacity for collective action is a central problem for all complex systems, as a kind of threshold in the evolution of greater complexity. Greater capacity for action toward the environment demands internal restrictions, and Luhmann surmises some kind of calculation of collective benefits, when improvements in the system’s position relative to the environment explain the development of the necessary internal restrictions. How exactly do societal necessities motivate action? The problem of collective action creates uncertainty with respect to the support of others that suppresses every impulse to collectivize the willingness to act. ‘Everyone waits, and the longer nothing happens, the greater the probability that nothing will’ (Luhmann, 1995: 199). An action needs to be specifically designated by symbols that clearly show that the entire system is bound by it. ‘Collective action always implies collective binding and this means that collective action is included as a premise in the meaning of the system’s other actions’ (Luhmann, 1995: 200). Hierarchy is a common form where the top decides about actions of subordinates and craves legitimacy, for example, by referring to the will of God. Modern society finds a solution with the political system, which makes collective decisions by majority rule. There are scarcely any traces of this transformative conflict between collective and individual rationality in Luhmann’s (2012, 2013) theory of society. With increased societal differentiation, collective action becomes too complex for individuals to handle. Indeed, human beings do not possess the cognitive capacity to grasp how vast numbers of actors and actions connect, and the complexity of modern society makes collective action irrelevant, where it becomes like solidarity, something of the past. Helmut Willke (1992, 1995, 2007) attempts to bridge these conflicting impulses in Luhmann’s theory and finds a solution to both problems of collective action and increasing societal complexity in a theory of context steering by experts emanating from and representing the different function systems. ­Functions are collective goods, and Willke ascertains the social dilemma involved in their production. Nevertheless, he sees an exception for the strategy of what he terms rational solidarity, when actors are willing to sacrifice their immediate interest to defect in exchange for the greater long-term benefits from cooperation—on the condition that the others do the same. Willke’s main argument is nonetheless that the calculation of the relationship between individual and collective benefits is too complicated for most people, and similar to Luhmann, he resorts to societal integration by symbolic

226

Hagen

representation. In modern society, such collective symbols are at the disposal of politicians and corporate actors representing the different function systems relatively free of context. Building on Etzioni’s The active society, Willke (1992: 252) defines collectivity as ‘a macroscopic unit that has a potential capacity to act by drawing on a set of macroscopic normative bonds which tie members of a stratification category’. People feel bound by the decisions of corporate actors, but how do organizations supervised by the state sort out problems regarding their interaction relative to the needs of society as a whole? On this higher level, Willke (1992: 329) reintroduces the principle of enlightened egotism, where the ­participating systems realize the costs of unregulated interaction, and the emergent utility from cooperation motivates the systems to regulate their own operations (ibid: 349). Similar problems and theoretical solutions arise in the works of Bourdieu and Habermas. Capital is the individual appropriation of collective labour, and people with similar amounts and compositions of capital are potential social groups that are capable of collective action. To explain the genesis of classes, despite warnings against the common Marxist flaw of confusing ‘class on paper’ with class action, Bourdieu implies that having a collective interest in a particular political reform is a reason to act on it (Bourdieu, 2005; Lane, 2006). His more theoretically elaborated analysis highlights the role of symbolic representation and corporate actors such as labour unions, famous artists, and popular politicians, who command individual and collective sentiments (Bourdieu, 1985). Habermas (1984) seeks to establish a rational, not emotional, solution to the problem of collective action. In explicating the integrative powers of communicative rationality, Habermas (1996: 336–351) takes on Willke’s revision of systems theory, in which the concept of autonomous systems, each with its special codes and language, recreates the Hobbesian problem of order on a higher cognitive level. The systems lack a common language that enables them to communicate with each other. Only communication rooted in natural language has the power to create societal integration. By making statements, the speaker raises claims to the factual truth and normative rightness of the utterance. As a sincere person, the speaker must defend his utterances against objections. A public discourse among all those affected enables the development of a shared understanding of common problems and obliges participants to act on this understanding. According to Habermas, communicative action is also collective action, but the collective good plays no role in the motivation for action. The motivating force exclusively comes from the procedural aspects of the use of language.

Collective Action



227

Collective Rationality as a Residual Category

In all cases, the theoretical logic is the same. The analysis deploys two competing concepts of collective action: rational action, where collective benefits motivate action, and a non-rational concept of collective action, where the motivation is emotional and normative. The first postulation is that collective benefits motivate action. On reflection, a conflict between collective and individual rationality becomes manifest. Now, either collective rationality is reduced to individual rationality, or the theoretical analysis ends in a symbolic representation of the collective interest. To understand the reasons for this reductionism, we take a closer look at rational choice theory, where the argument is most elaborated. At first glance, collective action seems to be an easy concept. Why would someone contribute to the production of a collective good and participate in collective action? For the less theoretically versed, the question has a straightforward answer: the individual benefits from the cooperation. The rational actor cooperates if the value of the collective good exceeds the cost of producing it. Individual gain thus motivates collective action. For a long time, this was a good theory—until Mancur Olson (1965) and game theory revealed the theoretical flaw. While individual goods are exclusive and private, collective goods are inclusive and public. Everybody in the group has access to the collective good irrespective of whether he or she contributes to its production. It is thus possible to free ride. Hence, while it pays to cooperate, it pays even more to defect. Consequently, collective action confronts the participating individuals with a conflict between collective and individual rationality—between actions that are best for everybody and actions that are best for individuals acting alone (Rapoport, Orwant and Chammah, 1965). An extensive body of literature demonstrates that rational calculation leads to defection because this is the only secure alternative irrespective of what the others do (Kollock, 1998). However, when everybody defects, the collective good will not come about, and the individuals are worse off than if they had all selected cooperation. Hence, individual rationality is collectively irrational, but collective rationality is impossible. The problem of collective action is a reformulation of the Hobbesian problem of order. The theoretical solution is also the same. To explain social order and successful cases of collective action, social science removes the rational actor and modifies the concept of action. Both Parsons (1968), working on the older formulation, and Elster (1989), dealing with the newer formulation, came to the same conclusion. Individuals have internalized social norms that create strong emotions that prohibit them from acting selfishly. While norms

228

Hagen

­motivate individuals from within, power is an external source that can sanction selfish behaviour negatively and contributions to the collective good positively. In both cases, the explanation severs the connection between the collective good and intentions. Normatively oriented individuals act only on what feels right and wrong, while those abiding to power act on the consequences for themselves. Acting individuals do not consider the collective consequences of their actions, and the collective or common good is no reason for action. Moreover, introducing power is a temporary fix, as instances that solve problems of collective action are themselves collective goods and raise second-order problems of collective action. The only logically consistent alternative is to remove the rational actor, and the unfortunate result of this theoretical operation is that the subjective perspective to collective action is lost. The social sciences do not have a concept of the conflict between collective and individual rationality from the actor’s point of view. The problem of collective action does not belong to social reality; rather, it is a pure theoretical construct on behalf of the social scientist. Collective rationality becomes a ­residual category, and when introduced to account for the problem of collective action, it disappears with the solution. Consequently, the social sciences lack the conceptual apparatus to make empirical observations of the rational aspects of collective action, and the lack of such an apparatus inhibits present theories of societal differentiation and integration. Luhmann, Bourdieu, Habermas and Foucault all solve problems of collective action and explain societal integration by symbols that generate emotional and normative commitments. Willke, Bourdieu, Habermas and Foucault also keep the backdoor open that individual rationality could motive collective action. However, the concept of enlightened or long-term self-interest does not solve problems of collective action, and rational solidarity is not some kind of generalized exchange. The tit-for-tat strategy, where I do, as you want, if you do, as I want applies only to two individuals interacting forever. Even the longest chain of mutual exchanges, positive or negative, as in Foucault’s ‘conduct of conduct’, presupposes third-party involvement. Collective action thus comprises at least three people with limited times spans. Habermas does not distinguish between strategic action as either individual or collective. He is correct about the close association between collective and public goods and arguments. However, public arguments and the purpose of reaching understanding alone cannot mediate conflicts between collective and individual rationality. Consequently, Habermas’ analysis resorts to an ­ideal speech situation where the actors have no purpose other than reaching an understanding.

Collective Action



229

From a Residual Category to a Scientific Concept

The idea behind this project was that collective rationality is a vehicle for ­societal integration. The unequivocal conclusion from rational choice theory, however, is that there is no such thing! The social sciences ‘must emphatically not try to explain successful collective action in terms of the benefits it brings to the group’ (Elster, 1989: 34). If the stance that ‘group rationality’ cannot explain collective action is logically impeccable, it is also counterintuitive and contradicts common sense. Do we not believe that there might be good reason to pay taxes and accept other constraints on our self-interest to solve problems for society and have access to collective goods? Perhaps the theoretical logic is wrong? My point now is not to follow either of the two traditional trajectories and claim that individual human beings are self-interested and prone to defect and that collective action will therefore fail or that individuals are normatively committed to the common good and will therefore succeed. Quite the contrary, the distinction between collective and individual rationality is an attribution of the social system of collective action, which says to the social actors that it is stupid, selfish and morally wrong to act on their self-interest and not in accordance with the best interest of society. Egotism prohibits collective problem solving and societal development. The normative claim is that the individuals should sacrifice their self-interest to achieve the capacity for collective action, and the reward for this act of rational solidary is access to the collective good produced by collective action. The transformative capacity of the social system of collective action resides in the distinction between collective and individual rationality. From moment to moment, individuals must shift between two position and perspectives— between what they should do to contribute to the common good and the possibility of defection. Collective rationality can also be a unifying force in public arguments about function problems. For instance, there might be several or different interpretations of the collective good and the ways to enable functions, and to sway the opinion of the others and acquire their support, one must argue that the interpretation is in the interests of all of us—that it is the collectively rational choice of action. Collective actions are social dilemmas and therefore social systems in the sense that the consequences of the choice of action depends on the choice of others. To study possible paths to either integration or anomie, we need scientific concepts that distinguish between collective and individually rational strategies of action to understand how function problems arise and become problems of collective action. Eventually, one side of the distinction ­dominates the other, and society will be able to solve its function problems

230

Hagen

and achieve integration, or collective action will fail, and disintegration will ensue.

System and Action

Why is it so difficult for the social sciences to grasp the rational aspects of ­collective action? I suggest that it is because of the present ontology of the social. The general assumption of the social sciences is that society consists of ­individual human beings and their actions, which creates a particular rationale for the scientific construction of concepts. One can make a second assumption that individuals are rational in the basic sense that they select actions depending on the consequences. This concept of rational action immediately leads to the two problems of complexity and social order. As the number of actors, actions and interdependencies increases, it becomes increasingly difficult to calculate the consequences. To the extent that individuals are able to do so, they confront the conflict between collective and individually rational courses of action. The established answer to the second problem is that social order and integration are possible because individuals are already committed to the ­collective or common good. They have internalized the necessary socially conditioned expectations for their own behaviour and the behaviour of others. Problems of social order thus do not arise. This solution to the problem of order nevertheless creates a secondary problem for theory construction. To explain successful cases of collective action, the social sciences must explain how the necessary norms, values, identities and symbolic representations came about. To achieve this, one creates a new collective level ‘above’ individual human beings. Durkheim has already introduced collective consciousness in addition to individual consciousness. More commonly, it is the social system that transforms societal needs to intentions for action. Parsons (1951) introduces cultural systems as the cybernetic device that both registers negative consequences of actions for society and creates motivation for actions with consequences more useful to society. Parsons is nevertheless not able to explicate how systems transform consequences to intentions, and he assumes that the necessary social norms come about because society needs them. Luhmann (1995) recognizes the problem and locates the causes in the present ontology of the social. The statement that society consists of individual human beings creates a paradox for theory construction. Society consist of individual human beings plus something else, namely, society. Society is

Collective Action

231

s­ omething different and distinct from the individuals comprising it. Within this distinction, social science can explain human behaviour with properties of the social. However, when the social scientist seeks to explain the social causes that explain human action, the ontology directs the analysis back to the individual human being. Ultimately, social sciences must find everything social inside the individual human, because society comprises individual humans. Based in ontological individualism, Luhmann concludes that sociological theory can make no progress. Luhmann therefore seeks to remove the individual and claims that society consists of neither individual human beings nor even of actions but only of c­ ommunication. This ontology offers a new rationale for theory construction: communication is the basic elements of self-referential and self-producing ­social systems. Luhmann underscores his point by the statement that only communication and not individual human beings can communicate. Individual human beings and their actions are thus on the outside and belong to the environment of society. Can sociology do without action and actors? Apparently not. Luhmann deliberately applies concepts such as intention, expectation, and experience to denote properties of both psychic and social systems and thus make social system into actors. He also cannot do without the concept of action. The full definition of his new social ontology is that society consists of communication and action (Luhmann, 1995: 137–175). To avoid functionalist reasoning associated with the vocabulary of systems, social sciences can fall back on the vocabulary of action. The limitation of this strategy seems to be that it is impossible to develop a theory of society, or only a very thin theory of society can be developed, as in the case of Elster (2007). Sociology obviously has some kind of people problem (see Hasse’s chapter in this volume).

Object Individualism

The theoretical discourse on societal integration is characterized by a perpetual oscillation between action and system. Both approaches presuppose a contribution from the other side that it cannot deliver. Since the failure applies to both sides, the epistemic obstacle should be something that they have in common. Both vocabularies are necessary, so the problem cannot be with action, the human individual or the social system per se but eventually with how sociology conceptualizes these phenomena. My hypothesis is that the real obstacle to theory development is the habit of thinking of the social in spatial terms.

232

Hagen

It is common in popular communication to use space and linear time as references. The same holds for social scientists who place social phenomena up or down, left or right, inside or outside, in the centre or at the periphery. Silber (1995) demonstrates how sociologists use spatial metaphors consciously, as in the case of Giddens, and in a theoretically strong way, as when Bourdieu resembles social fields with magnetic fields. I, however, have a deeper and more influential use of the spatial in mind. The most general assumption of the social sciences is that society consists of human beings and actions as things. This perhaps implicit assumption creates a rationale for theory construction where action becomes something that literally stretches out between and connects human bodies. Social relations are actions and include the physical prolongation—the forces emanating from bodily movements and having a causal effect on other things. The real epistemic obstacle to theoretical development is that individual human beings are associated with the body and action, with bodily movements. To discern this general assumption about the social from ontological individualism, I will label it object individualism (Hagen, 2006). Object ­individualism leads to the distinction between individuals and society as being understood as physical entities existing in different places—or with society as a metaphysical space of ideas hovering over the physical realm. It is this assumption that necessitates both the separation of individuals and society and that ultimately situates the social in the human body. We can now make a more precise formulation of the theoretical challenge for sociology, the obstacle and the means to overcome it. The task of sociology is to establish the connection between human action and societal needs. This relationship between individuals and society actualizes the two problems of collective action—social order and complexity. Object individualism solves the conflict between collective and individual rationality on the level of the individual as a physical entity, and every such solution removes the problem and conflict from the actor’s point of view. Likewise, there cannot be any meaningful account of the conflict if sociology expels the individual human being from society, as in the case of modern systems theory. Only human subjects can experience the conflict between collective and individual rationality. Indeed, even higher level primates are incapable of taking a group perspective to behaviour (Tomasello, 2014). What we need is a new ontology that provides a rationale for constructing concepts that can handle the problem of collective action from the actor’s point of view. The raw material for these abductions in scientific creativity is the residual category of a collective rationality (Hagen, 2016). The residual serves as a clue to the new ontology and rationale, and the clue is action as

Collective Action

233

meaning and not a thing. John R. Searle (2010) therefore makes an important contribution to general sociological theory when he establishes a clear-cut distinction between action as bodily movements (hereinafter bm) and social action.

Individual Action and Social Action

Actions are intentional, and in bm, the intention is to manipulate the own body, for instance, when an individual formulates intentions in his mind to lift his arm. Searle denotes this kind of motivation intentions-in-action. bm might stand alone, but usually bm are anticipated by what Searle designates pre-­ intentions. For example, I may decide to lift my arm in fifteen seconds. bm are associated with trying, while another word for pre-intentions is decision. Both actions can fail and are causally self-referent. The fifteen seconds are gone, but I did not do as I decided. I tried, but my arm failed to lift because I lacked the effort. Together, decisions and bm constitute the total action. How does this vocabulary to describe individual action carry over to social action, where the outcome of my bm depends on the bm of other individuals? Searle’s project is to create a new ontology for the social sciences that enables this transition from individuals to society. His proposal for a general explanation of what he calls collective action is therefore disappointing. Cooperation becomes possible because actors assume that others will do their part of the collective bm (Searle, 2010: 52). Searle’s sociology falls within the established pattern. Individuals do not consider the alternative that the others will not follow suit and exploit the situation to their own benefit. They do not address the consequences of the interaction and therefore escape both problems of collective action. The subjective perspective to collective action as a meaningful relation is thus lost. Searle does not say how the actors acquired this pre-intention for cooperation. There are no social dynamics, no inter-action and no social processes in Searle’ theory of collective action. Moreover, Searle does not distinguish between collective and social action and neither seem to have causal self-­ reference. In its place, Searle introduces a theory of language, where language is the source of meaning that enables different types of speech acts. Nevertheless, the distinctions between bm, intentions and pre-intentions are theoretically significant. They enlighten Luhmann’s conundrum. Actions as bm surely are no part of and take place on the ‘outside’ of society, while decisions that consider the decisions and actions of others are a shorthand for communication. This is the true realm of the social.

234

Hagen

The New Ontology of Distinction

According to Searle, his conceptualization logically follows from the basic facts established by the natural sciences, namely, that the only place for intentions is within individual heads. Intentions are thoughts, and thinking is a property of the mind. Realistically, one thought cannot enter the mind of another human being, and the causal chain between intention and action is located in the body. In her own thinking, what an individual human being can do is distinguish between different kinds of thoughts, such as thoughts referring to an external world of actions, as bm and internal states, and within this internal state, she can further distinguish between pre-intentions and intentions. These distinctions are real on the molecular level. In addition, they are real because they attribute meaning to bm. According to Searle, both action as decision and action as trying are causally self-referent. If they are, it is not as physical but psychic and social events. The individual distinguishes between intention and bm and disappoints herself, since she did not do as she intended. The failure matters because the action was a means to achieve something important, and the bm, the intention and even herself as an actor, i.e., an individual capable of action, attain value relative to this goal. To account for the self-reference of action, we must infer not only the mind but ultimately a psychic system. This system observes itself and acquires self-awareness relative to the environment. As an actor, she knows the internal distinctions are real because she can experience failure. It is straightforward to continue this line of reasoning and assume that psychic systems think about how the decisions and bm of others interfere with their own decisions and bm. On this basis, they will make assumptions about what others will do, and they might act on these assumptions. The contentious question is whether this interaction among psychic systems and their actions is also self-referent and constitutes a social reality with emergent properties. In the theoretical discourse, it is now common to introduce an extra-­ individual entity as the collective consciousness, language or social system, which replaces the subjective perspective and coordinates action on behalf of society. Parson and Luhmann use the vocabulary of system in this sense. Both, however, offer another account of the social system that does not do away with but constitutes a subjective perspective. Addressing the problem of the double contingency in interaction, Luhmann (1995: 104, 109–110) proposes the following solution: Two people choose their action dependent on the choice of others. How do they get on? They require help of the social system of interaction. One just has to suggest something, for

Collective Action

235

instance, a smile, and in relation to the reaction, a social system with a past and future emerges. The participating individuals use their knowledge about the systemic properties of their interaction to understand the intentions of others. This rationale for conceptual analysis and construction is enabled by a new ontology of the social. Luhmann (1995) claims that the social consists not of actions but of communication—and action. He defines communication as the threefold distinction between information, utterance and understanding. A person distinguishes between what was said and by whom and understands. In other words, this sums up what we already said of thinking about self and others. Understanding concludes the communicative event and creates an open situation of either acceptance or rejection, and this selection is an action (Luhmann, 1995: 48). Another word is decision, as used by both Luhmann and Searle. Action as the distinction between acceptance and rejection is both the end and the beginning of communication. What then is it? It certainly is not a thing and not only a psychic event. It is a distinction in social systems of meaning. Social systems provide action with consequences and enable an individual choice of action. This is possible because individuals have at least a rudimentary understanding of the system. A system is always ‘an interaction between what is out “there” and how we organize it “in here”. “System” denotes an interaction between the objective world and how it is looked at or thought about; it denotes a mode of perceptuo-cognito organization’. jordan, 2013: 52

Social systems enable the human individual to perceive herself and others as social actors. Individuals and actions are not bodies and things but distinctions in the media of meaning provided by social and psychic systems. This new ontology gives us the theoretical rationale to address the conceptual shortcomings of present theories of modern society.

Rational Collective Action and Functional Differentiation

Science, education, politics, the market economy and the other important parts of modern society are social systems. They consist of communication that creates an open but structured choice between acceptance and rejection. Participating in the communication, individual human beings become social

236

Hagen

actors as scientists, teachers or students, and politicians or voters. The systems are autonomous and autopoietic. Only science can decide whether a contribution is scientific. Individuals are also free to choose their own values. The system continues to exist only if enough people make the right kind of choice. Specialized autonomy implies interdependence, and imbalances do occur. The system of education does not deliver the qualifications needed by medicine or science; the economy is sluggish and restricts political spending; a terrorist attack reveals that national security is not up to the task; and global warming raises new demands on perhaps all the subsystems of a modern society (Luhmann, 1989; Guneriussen’s chapter in this volume). People experience problems and react with public argument and debate. Some problems solve themselves, whereas others require the attention of society. In the latter case, we will speak of function problems. Function problems highlight a future state where the problem does not exist, since a social system provides society with a certain capacity. Function problems are common problems. As they are created by people, they must be fixed by the concerted action of those concerned. The prospect of collective action, however, confronts actors with conflicts between strategies that are best for everybody and actions that are best for the individual acting alone. The conflict between collective and individual rationality is a distinction, meaning that the con­ certed action acquires meaning as collectively rational relative to the possibility of defecting on the collective effort. The distinction is an attribution, meaning that egotism is not the reason people drive cars that pollute the earth or evade taxation. From their ­personal perspective, their choice of action is practical and perhaps fair and just. ­Generalized, this line of actions is not sustainable, and to solve common problems, we must sacrifice self-interest and select collective rational lines of actions. This choice is a rational motivated solidarity because the individual gives up an individually beneficial course of action to share in the collective good that results from collective action. The social system of collective action gives a particular meaning to behaviour that causes problems that society needs to solve. The conflict plays out in the public discourse, and here, collective rationality triumphs. One cannot ­literally claim that we should all defect. To change public opinion, the argument must point to benefits for society. Even individual rationality may gain support as collectively rational—for instance, as an argument for deregulation and for society to rely on markets to solve function problems in the public sector. The public discourse, however, binds arguments, not collective action. ­Further, to solve the problem of collective action, modern society relies on

Collective Action

237

power, not sincerity. The political system applies collective power to make collective decisions collectively binding (Hagen, 2010), and the use of power is collective to the extent it suppress individual rational courses of action and enhances society’s capacity for collective action. It demands collective recognition, signifying that the use of power is necessary to solve problems for society and that those opposing power must argue with collective and not merely individual consequences. Thus, collective action is included as a premise in the meaning of the system’s other actions. The decision binds the collective and not simply everybody with the majority rule. More often, collective action does not interfere directly with the system that provides society with functions; rather, it builds institutions that guides its operations. An institution constitutes any combination of rules, regulation and organizations that has this purpose and effect (cf. Ahrne’s and Hasse’s chapters in this volume). For instance, universities and several other organizations are set up to frame the decisions of researchers to select certain topics and to be more productive. Many collective actors take an interest in making such institutional environments and attempts to influence the democratic circle of power. To become collective actors, they certainly must solve their own problems of collective action. If the use of power then is distributive and serves special interest and not society, it is for the political system or eventually the empirical analysis to decide.

Sociological Theory and Societal Integration

Modern society is a double level system. It is both part and whole. It is differentiated into several systems, and it has the capacity to observe and relate to the functions of these subsystems as a collective. The conflict between collective and individual rationality is then a vehicle for societal integration because the distinction is both a subjective tension and a social dilemma. It forces actors to heed the collective consequences of their actions and dismiss actions that go against the interest of society. To conceptualize processes of societal self-regulation, the social sciences must to do away with object individualism because it confounds the two types of action and creates a rationale for concept construction that reduces collective phenomena to properties of the individual human body. It is then difficult to distinguish between social and collective action, and problems of complexity become too complex. Society as a community with the capacity for collective action, however, is never analogous to the whole of society. As a totality, society is a vast network

238

Hagen

of institutions and systems that is impossible for the human mind to grasp. It is also not necessary to do so. Society needs only to react on tension in the societal structure. To the extent that society as a community is capable of solving function problems through collective actions of larger or smaller scope, society appears to be a fine-tuned matrix of power relations—without any collective or individual actor ever intending it to be exactly that. We do not need functionalism to explain this wonder. Processes of collective self-regulation create institutions that acquire a life of their own. The purpose is forgotten and eventually revived. By reintroducing the subjective perspective on collective action, we gained new possibilities for exploring how collective rationality is a source of normative commitment, symbolic orders and power that transforms societal needs to motives for action. To appreciate this, we had to change existing vocabularies of system and of action. References Bourdieu, P. (1985). “The social space and the genesis of groups”, Theory and Society. 14(6), 723–744. doi:10.2307/657373. Bourdieu, P. (1996). The state nobility : elite schools in the field of power. Cambridge: Polity Press. Bourdieu, P. (2005). The social structures of the economy. Cambridge: Polity. Elster, J. (1989). The cement of society : a study of social order. Cambridge: Cambridge University Press. Elster, J. (2007). Explaining social behavior : more nuts and bolts for the social sciences. Cambridge: Cambridge University Press. Foucault, M. and Gordon, C. (1980). Power/knowledge : selected interviews and other writings 1972–1977. Brighton: Harvester Press. Habermas, J. (1984). The theory of communicative action. Vol. 1. Boston, ma.: Beacon Press. Habermas, J. (1996). Between facts and norms : contributions to a discourse theory of law and democracy. Cambridge, ma.: MIT Press. Hagen, R. (1999). Rasjonell solidaritet. [Rational solidarity]. Oslo: Universitetsforlaget. Hagen, R. (2000). “Rational solidarity and functional differentiation”, Acta Sociologica. 43(1): 27–42. Hagen, R. (2006). Nyliberalismen og samfunnsvitenskapene : refleksjonsteorier for det moderne samfunnet. [Neoliberalism and social sciences: reflection theories of the modern society]. Oslo: Universitetsforlaget.

Collective Action

239

Hagen, R. (2010). “Collective power: reception and prospect of a scientific concept”, Ed. W. Østreng. Transference: interdisciplinary communications 2008/2009. Oslo: cas. Hagen, R. (2016). “Abduction—assessing fruitfulness and the construction of scientific concepts”, Eds. P. Sohlberg and H. Leiulfsrud. Theory in action. Theoretical constructionism. Leiden: Brill. Jordan, N. (2013). Themes in speculative psychology. Florence: Taylor and Francis. Kollock, P. (1998). “Social dilemmas: the anatomy of cooperation”, Annual Review of Sociology. 24: 183–214. Lane, J.F. (2006). Bourdieu’s politics : problems and possibilities. London: Routledge. Luhmann, N. (1989). Ecological communication. Cambridge: Polity Press. Luhmann, N. (1995). Social systems. Stanford, ca.: Stanford University Press. Luhmann, N. (2012). Theory of society. Vol. 1. Stanford, ca.: Stanford University Press. Luhmann, N. (2013). Theory of society. Vol. 2. Stanford, ca.: Stanford University Press. Olson, M. (1965). The logic of collective action: public goods and the theory of groups. Cambridge, ma.: Harvard University Press. Parsons, T. (1951). The social system. New York: The Free Press of Glencoe. Parsons, T. (1968). The structure of social action: a study in social theory with special reference to a group of recent European writers. New York: Free Press. Rapoport, A., Orwant, C.J. and Chammah, A.M. (1965). Prisoner’s dilemma : a study in conflict and cooperation. Ann Arbor: University of Michigan Press. Searle, J.R. (2010). Making the social world / the structure of human civilization. Oxford: Oxford University Press. Silber, I.F. (1995). “Space, fields, boundaries—the rise of spatial metaphors in contemporary sociological-theory”, Social Research. 62(2): 323–355. Tomasello, M. (2014). Natural history of human thinking. Cambridge, ma.: Harvard ­University Press. Wacquant, L.J.D. (2005). Pierre Bourdieu and democratic politics : the mystery of ministry. Cambridge: Polity. Willke, H. (1992). Ironie des Staates : Grundlinien einer Staatstheorie polyzentrisher G ­ esellschaft. Frankfurt a.M: Suhrkamp. Willke, H. (1995). Systemtheorie iii: Steuerungstheorie : Grundzüge einer Theorie der Steuerung komplexer Sozialsysteme. Stuttgart: Fischer, utb. Willke, H. (2007). Smart governance : governing the global knowledge society. Frankfurt a.M, New York: Campus-Verlag.

part 4 Class, Gender, Race and Social Recognition



chapter 14

The Status of the Political in the Concept of Class Structure Erik Olin Wright In rethinking the basic categories within Marxist theory over the past fifteen years, Marxists have devoted considerable attention to the concept of class. They have both reconceptualized the place of ‘class’ in the over-all Marxist theory of society and social change and transformed the concept itself (for an overview of alternative perspectives on class within current Marxist debates see Wright, 1981). Many of the attempts at reconceptualizing have revolved around the relationship between the political and the economic in class relations. Traditionally, Marxists have regarded class structure as an economic category. Whether defined by property relations or by production relations class structure was understood in strictly economic terms. Capitalists appropriated surplus value because of their location within economic relations; workers produced surplus value because they did not own their own means of production and had to sell their labor power to capitalists. In this notion of a ‘class-in-itself’, politics entered the analysis explicitly in only two ways: first, the state was seen as essential for reproducing this structure of economic class relations and for setting its legal presuppositions (guaranteeing contracts, enforcing property rights, and so forth); and second, politics was seen as central to how classes became organized in the class struggle. Indeed, the transition from a ‘class-in-itself’ to a ‘class-for-itself’ was traditionally viewed as a movement from the purely economic existence of classes to their political existence. More recent Marxist analyses have stressed the importance of political relations in the very definition of class relations. Not only does the state establish the legal preconditions of property relations, but in a deep sense those relations themselves have a political dimension. Different theorists express that dimension in different ways—as power relations, relations of domination and subordination, relations of control—but in all cases they assert a notion of class relations that necessarily embodies a political aspect. Even at the most abstract level, they have argued, a purely economic understanding of class ­relations is unsatisfactory.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_015

244

Wright

John Roemer challenges this recent trend in class analysis in his article, ‘New directions in the Marxian theory of class and exploitation’. His central argument is that, at the most abstract level, classes can be defined strictly in terms of ownership relations. Political factors enter into the story only at ‘lower’ levels of abstraction, particularly in the institutional conditions necessary for maintaining the basic property relations. In this paper I will critically examine Roemer’s arguments in support of this thesis. I will argue that while, as Roemer argues, exploitation can be defined in purely economic terms, class cannot. Class is an intrinsically political concept and for it to serve its explanatory purposes it must have its political dimensions systematically represented within the concept itself (Roemer, 1982: 253–287).1 Before making these arguments, however, I will briefly situate the theoretical object of this discussion— class structure—within a broader context of class analysis and discuss what is meant by ‘political practices’ and ‘political relations’.

Class Structure as an Element in Class Analysis

It is useful in discussing the concept of class to distinguish three separate elements in a class analysis: class structure, class formation, and class struggle. While each of these presupposes the other two and can be defined only in terms of its connection with the other elements, it is nevertheless important to make the distinctions. Class struggle refers to the practices of individuals and collectivities in pursuit of class interests; class formation designates the social relations within each class that determine its capacity to pursue its interests; and class structure is the social relations between classes that determine or shape basic interests over which classes-in-formation struggle (see Figure 14.1). These three elements, then, are related in the following manner:2

1 This paper will not address Roemer’s innovative strategy for defining exploitation using game-theory models, nor his development and defense of the Class-Exploitation Correspondence Principle. I consider both of these to be extremely important contributions to the Marxist theory of exploitation and class. My Critique is limited to the way Roemer deals with politics in his analysis. His claims about domination could be modified without any fundamental change in his general argument. 2 This model is, of course, a radically incomplete picture. The state, ideology, nonclass relations and interests, and many other factors have been left out. It is not meant to show how all the aspects of class are determined but simply to explain the interrelationships among them. For a discussion of the precise meaning of ‘limits’, ‘selection’ and ‘transformation’ in this diagram and for further elaboration on what constitutes class, see Wright (1978).

The Status of Political in the Concept of Class Structure

245

class struggle

TS

C LE

SE

T class structure

Figure 14.1

S

LIM

AN

M OR

SF

N RA

TR

SF

ITS

OR

MS

LIMITS

class formation

Interconnections among core elements of class analysis

The underlying structure of class relations limits the possible forms of collective class organization, which in turn limit the possible forms of class struggle. Within these limits, class struggle transforms both class structure and class formation. These transformations imply that the limits on class struggle (and on class formation) are not permanently fixed but change in response to the struggles themselves. It is in this sense that the model an be seen as ‘dialectical’: struggles transform the conditions of their own determination. This model is, of course, purely formal in character. There is no specific content given to any of the terms and no concrete propositions about the nature of the limits and transformations involved. The model provides a framework to specify a theory of class, but does not itself constitute such a theory. One of the critical steps in developing a theory is to elaborate the logic of each of the elements in the model. In this paper I will focus on the concept class structure, particularly on the role of political relations. I will not, except in passing, discuss the role of the political in class formation and class struggle. This is not to suggest that explicating the concept of class structure is somehow the key to the entire analysis, but simply that it is a necessary starting point.

The Concept of the ‘Political’

In order to define the political, it is first necessary to define social practice. Following Althusser, ‘practice’ can be defined as human activity that transforms some raw material, using specific means of production, into some product (­Althusser, 1970: 166 ff). Practices are thus human activities viewed in terms of their transformative effects in the world. Different practices are distinguished by the nature of the transformation (the nature of the raw material, of the means of production, of the formative activity, and of the product). Economic practices can thus be defined as those activities that produce and transform

246

Wright

use-values; political can be defined as those activities that produce and transform social relations; and ideological practices can be defined as those activities that produce and transform the subjective experience of those relations. Concrete, observed activities of people typically involve each of these types of practice. When workers work on an assembly line, they simultaneously transform nature into useful products (an economic practice) and produce and reproduce a particular structure of social relations (a political practice) and particular forms of subjectivity (an ideological practice). These distinctions among practices correspond to distinctions among social relations. Thus, economic relations can be defined as social relations that shape or limit the activities of transforming nature; political relations can be defined as those social relations that shape or limit the activities of transforming social relations; and ideological relations can be defined as those social relations that shape or limit the activities of transforming subjectivity. Again, any concrete social relation may involve all three types. Using the example of the factory, we might say that the technical division of labor is primarily an economic relation in that it systematically shapes the activities that transform nature; the authority structure is primarily a political relation in that it systematically limits the capacities of workers to transform the relations within which they work; and the job structure (seniority, competition in internal labor markets) is primarily an ideological relation in that it systematically shapes the subjectivity of workers on the job.3 When we speak of ‘political practices’ or ‘political relations’ the terms should be understood as a shorthand for practices or social relations within which the political aspect is the most important. This may be quite difficult to determine empirically in specific cases, as in the debate over whether educational ­institutions should be viewed as primarily ideological (producing forms of subjectivity) or primarily economic (producing skilled labor power). We must nonetheless acknowledge this complexity of practices and relations and set the agenda for investigating the relationships of the various types. The focus here is on the political dimension of class structure (the structure of class relations). As already stated, a political relation is a relation (or that aspect of a relation) that shapes the practices of transforming social relations. In these terms, the relations of domination and subordination are quintessentially political. To say that A dominates B is to say that A not only 3 It would be incorrect, however, to say that such labels exhaust the character of actual ­social relations within the factory. The technical division of labor also influences capacities to transform social relations; authority relations also shape subjectivity; and so on.

The Status of Political in the Concept of Class Structure

247

tells B what to do or in other ways directs B’s activities, but also that A has the capacity to constrain B’s attempts at transforming the relationship between A and B. To be a subordinate is not simply to be in a position in which one is given orders, but to be unable to transform the relationship of commandobedience. This is what distinguishes following instructions or suggestions in a reciprocal relationship and following orders in a hierarchical relation. They may be behaviorally equivalent in a given instance, but they are structurally quite distinct. The question at hand, then, is whether this particular political aspect of social relations—domination and subordination—is essential in defining class relations. I will show in the next section that as John Roemer argues, such relations of domination are not necessary for a definition of exploitation, but that they are necessary for a definition of class relations.

Roemer’s Treatment of Domination in the Concepts of Class and Exploitation

In his discussions of class and exploitation, Roemer adopts two rather different stances toward domination. In the first part of his analysis he argues that both class and exploitation can be specified strictly in terms of the distribution of property rights, without any reference to domination relations. At the end of the paper, when he introduces a game-theory analysis of exploitation, he argues that there is an implied relation of domination in the concept of exploitation and thus in class as well. What I will argue is that each of Roemer’s formulations is half right: class does require domination relations; exploitation does not. Let us first examine the strategy Roemer employs to investigate exploitation and class as direct consequences of the distribution of property rights. His strategy is to examine several different economies that differ only in the kinds of markets that are allowed in them and in the character of the distribution of productive assets. In the course of these investigations he proves two propositions, both of which may at first glance seem quite surprising. First, he shows that exploitation can occur in situations in which all producers own their own means of production, and thus there is no domination whatsoever within the actual process of production; and, second, he shows that there is complete symmetry in the structure of exploitation in a system in which capital hires wage laborers and in a system in which workers rent capital. Let us look at each of these in turn.

248

Wright

Roemer demonstrates that exploitation can exist in an economy in which every producer owns his or her own means of production and in which there is consequently no market in either labor power or means of production; the only things that are traded are final products of various sorts, but different producers own different amounts of productive assets. The result is that some producers have to work more hours than other producers to produce the exchange-equivalent of their own subsistence. What Roemer shows in this simple economy is that the result of trade among producers is not only that some producers work less than others for the same subsistence, but that the producers who work less are able to do so because the less-endowed producers have to work more. That is, an actual transfer of labor occurs from the assetpoor to the asset-rich. (The critical proof is that if the asset-poor person simply stopped producing—died—the asset-rich producer would be worse off than before and have to work longer hours). Since in this economy the exploiter clearly does not in any way directly dominate the exploited—they both own their own means of production and use them as they please—this example shows that exploitation does not presuppose immediate domination relations. Of course, a repressive apparatus may be needed to guarantee the property rights themselves—to protect the asset-rich from theft of assets by the assetpoor—but no domination directly between the rich and poor is implied. The second analysis is more complex. It compares the class structures on what Roemer calls a ‘labor market island’ and a ‘credit market island’. On both islands some people own no means of production, and other people own varying amounts of the means of production. The distribution of these assets is identical on the two islands. And on both islands people have the same motivations: they all are labor-time minimizers for a common level of subsistence. The two islands differ in only one respect: on the labor-market island, people are allowed to sell their labor power, whereas on the credit-market island, people are not allowed to sell labor power but are allowed to borrow, at some interest rate, the means of production. Roemer then demonstrates two things. First, that on each island there is a strict correspondence between class location (ownership of differing amounts of the means of production, including no means of production) and exploitation status (having one’s surplus labor appropriated by someone else). This is important ‘class-exploitation correspondence principle’. Second, he demonstrates that the two class structures are completely isomorphic: every individual on one island would be in exactly the same class on the other island. It is because of this strict functional equivalence of the labor-market island and the credit-market island that Roemer concludes that domination plays no essential role in the most abstract definition of classes. Roemer writes:

The Status of Political in the Concept of Class Structure

249

Exploitation can be mediated entirely through the exchange of produced commodities, and classes can exist with respect to a credit market instead of a labor market—at least at this level of abstraction. In this analysis, coercion is still necessary to produce Marxian exploitation and class. However, it suffices for the coercion to be at the point of maintaining property relations and not at the point of extracting surplus labor directly from the worker … These results thus force some re-evaluation of the classical belief that the labor process is at the center of the Marxian analysis of exploitation and class … I have demonstrated that the entire constellation of Marxian ‘welfare’ concepts can be generated with no institution for the exchange of labor. Furthermore, this has been done at the level of abstraction at which Marxian value theory is customarily performed. roemer, 1982: 266

Political relations are important for institutionally reproducing class and exploitation, but they are not essential in the very definitions of these concepts. This is not, however, the only assessment of domination made in Roemer’s analysis. Toward the end of the paper, when a game-theory approach is introduced, domination re-enters the analysis as a central feature. The idea is to compare the different systems of exploitation by treating the production system as a kind of game and asking if a coalition of players would be better off if they withdrew from the game under certain specified procedures. Different types of exploitation are defined by the withdrawal rules that would make certain kinds of agents better off. ‘Feudal exploitation’ is defined as the situation in which agents would be better off if they withdrew from the game with only their personal assets (that is, if they were freed from relations of personal bondage). Capitalist exploitation is defined as the situation in which agents would be better off if they left the game with their per capita share of total social assets (not just personal assets). Roemer’s game is a clever and insightful device, but it immediately runs into problems without additional specifications. For example, under the rules laid out so far, the handicapped could be said to exploit the healthy feudalistically, since the healthy would be better off if they withdrew with their personal assets from the game in which the handicapped are aided. Even more damaging, perhaps, if two islands, one rich and one poor, are arbitrarily grouped together even though they have no relations with each other, the poor island would be considered ‘exploited’ capitalistically by the richer one (that is, it would be better off if it withdrew from the game with its per capita share of the combined assets of the two islands).

250

Wright

It is to avoid these and related problems that Roemer added a number of further specifications of the game-theory approach in footnote 9 to the paper. There he states: a coalition S is said to be exploited at an allocation if two conditions hold: (1) that S does better than at the current allocation by taking its payoff as specified by the characteristics function of the game; and (2) that the complement of S (called S’) does worse than at the current allocation by taking its payoff … One way to pre-empt the invalid example might be to require a third condition for exploitation, namely, (3) that S’ be in a relation of dominance to S. Since dominance is undefined and is as elusive a concept exploitation, the addition of (3) is ad hoc … and reduces the sharpness of the game-theoretic characterization. roemer, 1982: 277

This final criterion, Roemer’s reluctance to include it notwithstanding, implies that a relationship of domination in some sense or other is required for the definition of exploitation and class. The handicapped do not dominate the healthy—indeed, if anything, the relations of domination are in the opposite direction—and thus even if they receive benefits from the assets of the healthy, they cannot be considered exploiters. Similarly, the poor island is not exploited by the rich one, since even though it would benefit from getting its per capita share of the two islands’ combined assets, there is no social relationship between the people of the two islands.4 Why is it that, in the discussion of the game-theory strategy of analyzing class and exploitation, Roemer was compelled to introduce relations of domination into the basic definition of class, whereas in his earlier discussion 4 I would want to add a fourth criterion to Roemer’s three: the two groups not only must exist in a relationship of domination and subordination, but this relationship must in some sense causally the explain the inequalities between the two groups. Prison guards, for example, dominate prisoners, and the prisoners would be better off materially (and in other respects) if they withdrew from the prison with their per capita share of the combined assets of guards and prisoners (or indeed with just their personal assets), but they not necessarily exploited by the guards, since the income of the guards is not gained by virtue of their domination of prisoners (that is, they do not appropriate any surplus labor from prisoners). Roemer’s second criterion—that S’ be worse off—touches on this issue, but it is possible for S’ to be worse off even if its position in the initial game did not explain the initial inequalities. (The situation in the prison example would be quite different, of course, if the guards obtained services from prisoners. Then part of the inequality between guards and prisoners would be causally explained by a relationship of domination and subordination.)

The Status of Political in the Concept of Class Structure

251

he was not? It is because, I think, the initial discussion was confined to the problem of exploitation and class within commodity-producing economic systems, whereas the game-theory discussion was designed to explicate the problem across more fundamentally different economic systems, including ­non-commodity-producing economies. Since feudalism, for example, revolves around relations of bondage and since this is at the heart of the definition of feudal class relations, it is impossible to generate a purely economic definition of feudal classes. So-called extra-economic coercion must be considered part of the definition of class relations in feudalism, not simply an institutional boundary-setting political process. Within commodity-producing societies, however, it appears that political relations are separated from economic relations, and it becomes possible to talk about classes and property rights as if they did not imply domination. This view of the relationship between class exploitation in commodityproducing systems is, I believe, incorrect. Let us return to Roemer’s d­ iscussion of simple commodity production and the two market ‘islands’. In each of these analyses Roemer convincingly shows that exploitation can be specified strictly in terms of property rights and their distributions. Domination enters the story of exploitation only externally, in the enforcement of property rights themselves. But what about class relations? Here we notice that there is a critical difference between the analysis of simple commodity production and the two islands. In the simple commodity-producing case there are, in Roemer’s view, no classes properly speaking, since all actors have the same relationship to the means of production, whereas in the two islands we do have classes: a class of owners and a class of nonowners. But why does owning matter to such an extent as to warrant the designation ‘class’? In the simple commodity-producing society depicted by Roemer there are people who may live a life of relative leisure because of the heavy toil of others. With relatively little modification of the conditions of his story we could also have people with very different levels of final consumption—rich and poor standards of living (rather than just high and low levels of toil). Why is not the distinction ‘rich’ and ‘poor’ itself a class distinction? The reason is that the rich do not dominate the poor in the simple commodity-producing society. No social relationship binds them directly to each other in a relation of domination and subordination. In both the credit-market island and the labor-market island, however, the owners and nonowners are directly bound together in relations of domination and subordination. There is thus a crucial difference between having few assets, but still enough to produce one’s own means of subsistence, and having no assets, and thus having

252

Wright

to either sell one’s labor power or rent the assets of others. Rich asset-owners do not directly tell the poor asset-owners what to do—they do not directly dominate them. However, a new kind of social relation is generated between the asset-owner and the nonowner: owners do dominate nonowners. This implies that property rights has a different social content in the two cases. In the simple commodity-producing economy, property rights only specify a set of effective powers over things—productive assets. While of course such effective powers imply that one has the right to exclude other people from using those assets (or to prevent them from taking them), the right itself implies no ongoing relationship in which effective powers over people are exercised. In the credit-market and labor-market economies, property rights imply a set of effective powers over both things and people. The owner of assets not only has the right to use those assets but the right to control in specific ways the behavior of people who have no assets but who desire to gain access to assets. The labor contract and the credit contract both imply a relation of domination—an agreement on the part of those without assets to follow certain orders from those with assets. Because property rights in the labor-market and credit-market islands entail such relations of domination and subordination, the exploitation relations in this case constitute a class relation and not simply a basis of inequality (Cohen (1979) makes a similar point in his discussion of class relations only when combined with relations of subordination and domination). Exploitation without domination, or domination without exploitation, does not constitute class relations. Domination by itself, such as that of prison guards over prisoners, may be a form of oppression, but not class oppression. Similarly, exploitation without domination is not a form of class relations. Children certainly appropriate the surplus labor of their parents, but do not (at least in the normal sense of the term) dominate them and thus cannot be considered a ‘ruling class’ within a family.5 Roemer is thus correct when he asserts that analyzing the labor process is not essential to specify the minimum conditions for capitalist exploitation. But he is wrong when he asserts that the labor process is also not essential for an abstract understanding of class relations in capitalism. At a very minimum, the capitalist labor process must be understood as a structure of relations within which capitalists have the capacity to dominate workers. For ownership to be the basis of a class relation, ownership rights must imply domination over the activity of workers. And this is indeed what the analyses of the labor process 5 But note: where fathers both exploit and dominate their children, as is true in some societies, then the father-child relation could be considered a form of class relation.

The Status of Political in the Concept of Class Structure

253

are concerned with: the forms of domination that govern laboring activity within production.6

Implications for Class Analysis

The argument that the political is intrinsic to the concept of class at even the highest level of abstraction has a number of important implications for class analysis. I will discuss several of these: implications for the labor theory of value; implications for the defense of the Marxist concept of class against its various bourgeois rivals; implications for the more concrete elaboration of the concept of class in capitalist societies; implications for the problem of classes in socialism; and implications for the general Marxist analysis of modes of production. The Labor Theory of Value Roemer argues that the justification for choosing labor power as the numeraire commodity for defining value and exploitation is that it is uniformly distributed throughout the population. This property is essential for a ‘proper’ theory of exploitation, that is, a theory that classifies the poor as exploited and the rich as exploiters. Only labor power, Roemer argues, has this property since ‘no 6 In an earlier version of this paper I attempted to justify the concern with domination on the grounds that workers, being labor-time minimizers (one of the assumptions in Roemer’s models), would try to work as little as possible once employed by owners. Domination was thus needed to get them actually to perform labor, or in traditional Marxist terms, to transform labor power into labor. Roemer and others pointed out in discussions of that initial analysis that this was quite parallel to the problem of capitalists’ ‘cheating’ each other in their exchanges, that is, violating the terms of the exchange contract. The problem of cheating, resistance, and so forth can thus be considered a problem at a ‘lower level of abstraction’ than the specification of the formal exchange relation itself. And thus, just as Marx ignored the problem of cheating among capitalists and the need for institutional safeguards against such cheating (although much contract law is concerned with this) in his abstract analysis of the capitalist mode of production, so, it was argued, we can ignore the problem of cheating by workers (failing to perform the contracted amount of labor). The problem of domination, however, is not simply one of responding to resistance on the part of workers; it is also a ­question of exercising effective powers over workers’ activities in order to get them to do what the capitalist wants them to do. While this is directly entailed by capitalist property rights, it is a mistake to talk as if it is the property rights as such and not the domination relations implied by them that constitute class relations. Property rights that do not entail relations of domination do not define class relations, as is the case in the simple commodityproducing economy.

254

Wright

produced commodity is uniformly distributed, since proletarians are dispossessed of all productive assets’ (Roemer, 1982: 274). Furthermore, the purpose of the theory is to explain class struggle between capitalists and workers, and the use of labor power does indeed generate a theory of exploitation that corresponds to the polarization between capitalists and workers. Once we add domination relations directly into our idea of class, a different kind of argument can be built for the use of labor power as the numeraire ­commodity or, equivalently, for the use of labor time as the metric for exploitation. Labor time, as opposed to any other metric for the surplus product, is simultaneously a measure of appropriation relations and domination relations. It is a measure of how much product is appropriated and how much human time is dominated through that appropriation. As appropriators, exploiting classes appropriate surplus products in one way or another, and if the appropriation relation was sufficient to define class relations, any basic good could provide a satisfactory metric for the quantitative aspect of class relations. But, as I have argued, the concept of class is intrinsically a political concept as well. The ideal metric of exploitation, therefore, should capture both aspects of class relations. Labor time does do this, for it identifies how much laboring activity is dominated in production.7 To justify the choice of labor time as the metric of exploitation we must argue that domination relations are as central to class relations as are appropriation relations. While I have shown that domination relations are implied in Roemer’s analysis even though he relegates them to secondary importance, I have not yet provided a general argument in support for their importance in a class analysis. To do so, I will turn to a comparison of Marxist and non-Marxist concepts of class. Marxist versus Non-Marxist Concepts of Class Non-Marxist concepts of class typically take one of two forms: either they are structured around categories of distribution without reference to domination, or they are structured around categories of domination without reference to distribution. In the first of these tendencies, class is defined either directly in terms of distributional outcomes (incomes) or in terms of the proximate determinants of those outcomes (occupation or ‘market capacity’—the Weberian approach). In either case, relations of domination are either absent from or 7 Labor time is, of course, only a quantitative measure of domination relations, not qualitative. Labor time by itself does not provide an adequate way of analyzing domination, but it is the one metric of value that expresses both the magnitude of the product and the magnitude of domination.

The Status of Political in the Concept of Class Structure

255

incidental to the discussion. The second tendency, most explicitly found in the work of Ralph Dahrendorf, defines classes solely in terms of power, or authority relations. There are ‘command classes’ and ‘obey classes’ in every institutional sphere of the society, with no special status being given to economic institutions. The Marxist account of class subsumes both of these images of class relations through the concept of exploitation. Class relations are the unity of appropriation relations (the Marxist way of theorizing categories of distribution) and domination. The justification for this view of class relations rests on two arguments. First, within production relations, domination without appropriation and appropriation without domination are unreproducible structures of social relations.8 Second, the coincidence of domination and appropriation within production relations provides the basis for understanding collective actors in the epochal processes of social conflict and social change.9 The first of these can be termed the ‘conditions of existence’ argument, the second, the ‘historical materialism’ argument. Let us briefly examine each in turn. The first thesis states that within the social relations of production any time the relations of domination and appropriation cease to correspond with each other the situation would be highly unstable and tend toward a restoration of correspondence. Imagine, for example, that as the result of a series of labor reforms, workers organized in militant trade unions won the capacity to collectively organize the process of work, including the capacity to allocate labor and the means of production to different purposes, but that the rights to the products produced with these means of production, and thus the appropriation of the surplus product, remained in private hands. Capitalists could not tell workers what to do or to fire them, but because they owned the means of production and appropriated the surplus product they could effectively veto any investment decision made by workers (for example, they could decide to consume their surplus rather than let it be used productively). This would be 8 This does not mean that in every social position, domination and appropriation perfectly coincide but that a complete noncorrespondence cannot be stable. It is entirely possible in capitalist production for certain positions—middle and lower management for example—to be in a domination relation to workers without being in an appropriation relation. This kind of noncoincidence in the heart of the idea of ‘contradictory locations within class r­ elations’, a concept developed to decode the class logic of ‘middle strata’ (see Wright, 1978: Ch. 2). What I am excluding as a possible structure of production relations is one in which power is completely divorced from appropriation. 9 ‘Epochal social change’ refers to fundamental, qualitative transformations of a society’s social structure. In the Marxist tradition this revolves around a transformation from one mode of production to another.

256

Wright

a situation in which appropriation brought with it no immediate power of domination, and domination was unaccompanied by appropriation. In such a situation it seems likely that either workers would attempt to extend their powers to include actual appropriation or that the appropriators’ capacity to block investments would become new means of domination, thus undermining or limiting the apparent domination of production by workers. A radical noncorrespondence between appropriation and domination within the relations of production cannot endure for long periods of time. There are two basic reasons for this. First, the appropriation of surplus products ­requires power. Direct producers usually do not like to toil for the benefit of exploiting classes, and unless there are coercive mechanisms at the disposal of the e­ xploiting class to force them to do so, the level of exploitation is likely to decline. S­ econd, unless relations of domination enable people in positions of domination to command resources, that domination quickly reaches severe limits. In the end, it is the capacity to command the use of the social surplus that provides the material basis for effective domination within the relations of production. A concept of class that unites the relations of domination and appropriation, therefore, is structured around the necessary conditions of existence of both domination and appropriation. But Marxist theoretical claims go beyond this kind of functional or reproductive argument. Historical materialism, in its various incarnations, is an attempt to understand the conditions and dynamics of epochal social change and social conflict, not simply the conditions for the reproduction of stable structures of social relations. To define class as the unity of domination and appropriation is meant also to provide a way of understanding these problems. A defense of historical materialism (or more accurately, of a modified version of historical materialism) lies outside the scope of this paper, but I will offer a few comments on the suitability of the concept of class being discussed here for the theoretical ambitions of historical materialism (for my views on classical historical materialism and its weaknesses, see Levine and Wright, 1980). The heart of social change necessarily revolves around the transformations of the social use and allocation of productive time and resources. This has two important implications for the present discussion. First, since class struggles are structured by the social relations within which laboring time and resources are allocated and used, such struggles are always implicated in epochal social change (although this does not imply that all such change is reducible to class struggles; other kinds of conflict, involving other sorts of actors and determinations, may also be of great importance in specific historical circumstances).

The Status of Political in the Concept of Class Structure

257

Second, any social movement, whatever its social base and whatever its logic of development, that pursues projects of fundamental social change ultimately faces the problems of reorganizing how time and resources are controlled. If the system of class relations is left intact, then there are clear limits to the range of possible social changes. If those limits are to be surpassed, then the social movement must be a movement for the transformation of class relations. Ethnic, religious, nationalist, and other nonclass movements are thus forced to engage in class-like struggles, struggles that systematically transform basic class relations. To summarize: Non-Marxist accounts of class stress either distribution (appropriation) or domination, but not the unity of these two within a concept of class exploitation. The Marxist attempt to combine these two elements within a single concept produces a much more powerful theoretical tool, both in terms of analyzing the conditions of the existence of classes (the relational requirements of their reproduction) and in terms of analyzing the conditions for epochal social transformation. For both of these purposes it is essential in analyzing class structures that classes be understood as having a political dimension even at the highest level of abstraction. Implications for the Concrete Investigation of Class Structures Abstract concepts are to be evaluated not only for their logical presuppositions and coherence, but for their usefulness in more concrete investigations. One of the advantages of a concept of class that is defined explicitly in terms of the unity of exploitation and domination relations is that it provides a strategy for examining capitalist class relations at more concrete levels. For example, such a concept provides a way of understanding the class character of managerial positions within capitalist production. Managers can generally be understood as locations within the social relations of production that (1) dominate the working class, (2) are dominated by the bourgeoisie, and (3) are exploited by capital, but (4) are exploited to a lesser extent than are workers. Whereas the capitalist class and the working class are perfectly polarized on both the domination and exploitation dimensions, managers occupy what I have termed elsewhere a ‘contradictory location within class relations’ (Wright, 1978: Ch. 2). They are simultaneously in the capitalist class and in the working class, occupying class locations that have some of the relational characteristics of each class. If capitalist class relations are defined exclusively in terms of exploitation relations, then most managers would fall into the working class. The specification of class in terms of both exploitation and domination thus provides a strategy for more concrete analyses of class.

258

Wright

The Analysis of Socialist Exploitation and Class One of the most promising lines of investigation opened up in Roemer’s work is the strategy for analyzing exploitation in socialist societies. Roemer suggests that socialist exploitation should be understood in terms of inequalities generated by the distribution of ‘inalienable assets’, that is, skills. The exploiters in socialism are those who possess skills; the exploited are the unskilled. Given Roemer’s formal criteria for exploitation, this would be a reasonable way of characterizing the distributional outcomes of skill inequalities in socialist societies. The question, however, is whether or not this kind of exploitation can be considered a class relationship. If, in addition to benefiting from an exploitive redistribution, the skilled also dominated the unskilled, then this relationship would constitute a class relation. However, unlike the possession of alienable assets, the sheer possession of skills does not logically entail domination of the skilled over the unskilled. Thus, it is possible to imagine a situation in which the skilled still received an exploitive redistribution of income, even though production was controlled by democratic bodies of workers that decide on production priorities and procedures and that give orders to both skilled and unskilled workers. This would be the case, for example, if the only way of inducing people to acquire skills is through heavy incentives that effectively redistribute income from the unskilled to the skilled. This does not imply, however, that within the actual organization of ongoing production it was the skilled workers who dominated the unskilled workers. In such a situation, the skilled could reasonably be regarded as a privileged stratum of workers, but not as a different class. The two kinds of ‘socialist’ societies we have described are likely to have very different forms of social conflict, even though they may share a similar pattern of distribution. If the skilled actually dominated the unskilled as well as exploited them, social conflicts would be likely to crystallize between the unskilled and skilled. If the skilled received exploitive redistributive benefits, but did not dominate the unskilled, conflicts would be less likely to take on a classlike character. Conflicts might develop over the motivational underpinnings of the incentive structures, but they would not necessarily develop between the skilled and unskilled. If, however, we fail to distinguish these two situations by failing to incorporate the notion of domination into the specification of class relations, then in both cases skilled and unskilled would have to be regarded as antagonistic classes.10 10

The conditions under which socialist ‘exploitation’ crystallized as a new form of class structure bears directly on what Roemer terms ‘status exploitation’. Although not

The Status of Political in the Concept of Class Structure

259

Modes of Production I argued earlier that to include the political in the structural definition of class would facilitate a class analysis of the social conflicts implicated in epochal social change. In this final section I will examine how this definition of classes affects the theoretical specification of the ‘epochs’ themselves, that is, of ‘modes of production’. In distinguishing capitalism and feudalism as modes of production, M ­ arxists have usually stressed that feudal exploitation required ‘extra-economic’ coercion whereas capitalist exploitation was purely ‘economic’. This formulation was typically accompanied by the claim that in feudalism politics and economics (or the state and production) were institutionally fused in the social organization of the feudal manor, whereas in capitalism the political and the economic are institutionally separated. The argument of this paper challenges this traditional view of the modes of production. Classes in both capitalism and feudalism imply domination, and not simply system-preserving coercion, but domination directly within the social organization of production itself. The issue is where the coercion is located, how it is organized, and how it is articulated to other aspects of the system of production (technical, ideological, and so forth). Instead of seeing the contrast between capitalism and feudalism as economic exploitation versus extra-economic coercion, the contrast should be formulated as follows: class exploitation based on noncoercion outside the labor process and on coercion inside the labor process versus class exploitation based on coercion outside ­analyzed extensively in his ‘New Directions’, Roemer’s status exploitation refers to situations in which a person receives exploitive net redistributions not by virtue of ownership of private property or skills, but by virtue of incumbency in some office, typically of a bureaucratic character. I would argue that when socialist exploitation as defined by Roemer becomes a form of class relations, that is, when it coincides with relations of domination, it will also tend to generate what Roemer calls status exploitation. After all, if people with skills also control bureaucratic centers of domination (so that they dominate unskilled workers as well as receive exploitative transfers from them), why wouldn’t they use their bureaucratic positions to extract additional surplus? ‘Socialist exploitation’ in and of itself is thus not likely to become the central principle of exploitation in any form of class structure. Status exploitation—or perhaps more appropriately in this context, ‘bureaucratic exploitation’—however, is intrinsically linked to a relation of domination and can therefore be considered a basic principle of class relation. While according to traditional Marxism, socialism is not a new mode of production but rather a transition from class society (capitalism) to a classless one (communism), the concept of bureaucratic or status exploitation suggests the existence of a form of postcapitalist, a new mode of production altogether. For a further elaboration of these issues, see Wright (forthcoming).

260

Wright

Table 14.1 Typology of modes of production

Mode of production

Political dimension outside of the labor process

Political dimension inside of the labor process

Slavery Feudalism Capitalism Communism

Domination Domination Self-determination Self-determination

Domination Self-determination Domination Self-determination

the labor process and on self-determination inside. The issue, then, is how the political dimension of the production relations is linked to the economic dimension of those same relations.11 This way of understanding production suggests the following simple typology of modes of production (see Table 14.1).12 This way of conceptualizing the modes of production has important implications for class analysis. I will briefly discuss three: the problem of class formation, the analysis of politics in general, and the transition between modes of production, particularly from capitalism to socialism and communism.

Class Formation

Traditionally, Marxists have understood the process of class formation as a transition from a ‘class-in-itself’, which was seen as an economic category, to a ‘class-for-itself’, which was seen as a political category. The analysis ­presented 11

12

Ellen Meiksins Wood (1981: 66–95) made a similar argument in her article. She characterizes capitalism as a social system in which politics are made private (that is, removed from the ‘public sphere’) through the organization of ‘politics of production’ within the private factory. In feudalism, the political dimension of production coincided with the political dimension of the stare—both were united in the feudal lord, and thus the politics of production had a ‘public’ character. In capitalism, it is not that the political and the economic are institutionally separated, but that the political dimensions of production are institutionally separated from the state. For related arguments on politics of production see Burawoy, 1979. This typology is only a first approximation. To deal effectively with such modes of production as the ‘Asiatic mode of production’ or (if it is a legitimate concept) the ‘state ­bureaucratic mode of production’, various distinctions within the category ‘coercion outside of the labor process’ would have to be made. I will not explore these issues here.

The Status of Political in the Concept of Class Structure

261

Table 14.2 The political dimension of class formation

Political formation of working class

Formation centered on production politics

Formation centered on state politics

Syndicalist Reformist social democratic Revolutionary

Yes No

No Yes

Yes

Yes

here suggests that classes can never be seen as purely economic categories, even at their most disorganized and atomized. They are always political (see Table  14.2). This suggests that instead of seeing class formation as a one-­ dimensional process of political formation, we should develop a typology of class formations. Classes can be formed around the political dimensions of production relations, around the political dimensions of the state, or around both. Without attempting to defend the argument, I suggest the following simple typology of working-class formations. Politics Politics cannot be analyzed simply as state-centered political processes and practices, that is, as politics oriented toward and structured by the state apparatuses. Instead, political analysis should focus on the articulation of what Michael Burawoy has called ‘global politics’ and ‘production politics’—politics organized around the state and politics organized within the process of production (Burawoy, 1981). Burawoy, for example, analyzes the relationship between these two sites of politics in the transition from the colonial to the independent state in Zambia, paying particular attention to the politics of production in the mining sector (Burawoy, 1982). The mode of production in mining, Burawoy argues, can best be characterized as a ‘colonial mode of p ­ roduction’, a variant of capitalism that depended upon coercive forms of labor control and on certain forms of extra-economic coercion outside of the labor process. The whole social organization of the mines was built around this particular form of production during the colonial period. Burawoy then observes what happens when there is a drastic change in the form of the state and an accompanying change in the character of global politics, while the structure of the production system in the mines remains relatively unchanged. This meant in Zambia

262

Wright

that global politics and production politics no longer corresponded with but instead contradicted each other.

Transitions between Modes of Production

The classic Leninist position on the transition between capitalism and socialism was that the proletariat had to smash the capitalist state apparatus and construct a new kind of state—a proletarian form of the state—that would enable the working class to be stabilized as a ruling class. Expropriating the means of production from the capitalist class plus restructuring the state were sufficient to consolidate socialism and accordingly to ensure the transition to communism. As became clear in Lenin’s praise of Taylorism (scientific management), one-man management, and so on, no fundamental restructuring of production politics was deemed necessary. The argument of this paper suggests that the transition from capitalism to socialism requires a change in production politics as well as in global politics. If workers are dominated within production relations, it is hard to see how they could become a dominant class in any meaningful sense of the word, even if private ownership of the means of production were abolished. Under such conditions a new class system is likely to emerge in which public appropriation of the surplus product would combine with new forms of domination over direct producers. It is only when the political dimension of production relations and thus of class relations is recognized that such a new class system can be adequately theorized. If forms of appropriation of surplus labor are the only criterion for class, and if modes of production are understood in purely economic terms, then the public appropriation of the surplus product becomes ipso facto socialist production. Conclusion John Roemer’s work is one of the few genuinely novel contributions to the ­Marxist theory of exploitation and class to be produced in recent years. It opens up possibilities not only for deepening our understanding of exploitation within a Marxist perspective but for critically assessing the competing claims made by the different theoretical traditions. His analysis is less s­atisfactory when he extends his idea of exploitation to the problem of class. The value of the Marxist concept of class lies in the way it links together economic and political relations within a single category. Classes are not determined solely by

The Status of Political in the Concept of Class Structure

263

relations of exploitation or by relations of domination, but by the two together. If domination is ignored or made marginal, as it is in some of Roemer’s analysis, the concept of class loses much of its power in explaining social conflict and historical transformation. References Althusser, L. (1970). “On the materialist dialectic”, For Marx. New York: Vintage. Burawoy, M. (1979). “The politics of production and the production of politics”, Ed. M. Zeitlin. Political power and social theory. Vol. 1. Greenwich, ct.: JAI Press. Burawoy, M. (1981). “Terrains of contest”, Socialist Review. 58. Burawoy, M. (1982). “The hidden abode of underdevelopment”, Politics & Society. 11(2): 123–166. Cohen, G.A. (1979). Karl Marx’s theory of history: a defense. Princeton, n.j.: Princeton University press. Levine, A. and Wright, E.O. (1980). “Rationality and class struggle”, New Left Review. 123. Roemer, J. (1982). “New directions on Marxian theory of exploitation and class”, Politics & Society. 11(3): 253–287. Wood, E.M. (1981). “The separation of the economic and political in capitalism”, New Left Review. 127: 66–95. Wright, E.O. (1978). Class, crisis and the state. London: New Left Books. Wright, E.O. (1981). “Varieties of Marxist conceptions of class structure”, Politics & S­ ociety. 9(3): 323–370. Wright, E.O. (forthcoming). “Capitalism’s futures: a preliminary reconceptualization of post-capitalist modes of productions”, Socialist Review.

chapter 15

Gender as Analytic, Political and Interdisciplinary Concept Harriet Bjerrum Nielsen Introduction Gender is a term that is both ambiguous and ambitious. Sometimes it refers to a categorical distinction between women and men (either/or). Sometimes it designates a distributive difference between groups of women and groups of men (more or less of something, for instance, a capacity, attitude, behaviour, or salary). Finally, it may indicate the cultural discourses of gender, which work as mostly unacknowledged frames of interpretation in our perception of the world. These three analytic meanings of gender—as distinction, as distribution and as discourse—may lead to considerable confusion about what we are truly discussing when we refer to gender and gender differences. It is not unusual either in everyday conversations or in research for distributive gender differences to be interpreted through the lens of common gender discourse and understood as distinctions. Gender has many facets and may be studied from numerous different theoretical perspectives and academic disciplines. Gender is central to division of labour and the structuring of institutions such as family, school, workplace, market and state. It is also a profound cultural system of meaning, norms, conventions, symbols and myths. It is a dimension of bodies and physical reproduction, individual identities and personal experience, social relations and everyday interaction. The impact and meaning of gender are complex, contextual, and changing over time. Moreover, the dimensions of gender are deeply entangled with other lines of difference and inequality, including age, sexuality, social class, nationality, race and ethnicity. These entanglements, or articulations, contribute to shaping the organisation, salience, and meanings of gender in various spheres of life. What we currently refer to as gender studies emerged with the feminist movement in the 1970s, implying that the perspective has a built-in critical edge: it both interrogates how gender is produced, legitimised, maintained and changed and considers this academic knowledge a contribution to the struggle against discrimination and injustice. This engagement with social justice © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_016

Gender as Analytic, Political and Interdisciplinary Concept

265

raises important issues about the relationship between knowledge, power and politics.1

Gender as Distinction, Distribution and Discourse

In everyday language, gender is a distinction between women and men, girls and boys. The distinction is based on the close-to-dichotomous anatomical differences between male and female bodies (genes, gonads and genitals) that are often considered the core of biological sex. A small number of babies are born with ambiguous genitals or intersex conditions (Fausto-Sterling, 2000; Hines, 2004). There also appears to be an increasing number of people (transgender, transsexual) who do not identify with the sex they were assigned at birth. Some of these people choose to correct their bodies via hormonal treatment or surgery. Nevertheless, for many practical purposes the gender distinction works unequivocally. As Toril Moi argues from a Wittgensteinian perspective, all categories are blurred at the edges, and the existence of difficult cases does not prove that there are no easy ones (Moi, 1999: 40). All other empirical gender differences, whether they are biological (hormone levels, secondary sex attributes, brain structure, height or muscular strength), psychological (differences in motivations, attitudes or cognitive capacities) or behavioural (differences in preferences and ways of being and behaving), are of distributive character, namely, they involve complex variation, not dichotomy. The variation within each gender group is often larger than the average difference between the two groups. Most distributive patterns depend on social and historical context and therefore are neither stable nor timeless. Because gender has many facets and different dimensions do not always 1 References to feminist theory and gender studies are often used interchangeably, but political and philosophical issues tend to be predominant in feminist theory, whereas the term gender studies covers all types of (critical) research on gender. Because of the connection to women’s liberation the field was first called women’s studies, but starting in the 1990s gender studies has increasingly been used as the general term. This acknowledges both that men also have gender and may suffer from the gender system and that gender may be at work in spaces without women present. Gender studies also now includes studies of sexuality, often referred to as queer studies or lgbtq studies. I write this chapter from the perspective of a social researcher working with qualitative methods, situated in an interdisciplinary centre for gender studies of which I was the director for many years (1993–2009). My research topic is gender and childhood/youth studies from a perspective of social change, with a special interest in the interconnection between psychological and socio-cultural gender.

266

Nielsen

o­ ccur together within a single person’s experiences and behaviour, it is also the case that a man or a woman may be ‘typical’ in some respects and ‘atypical’ in others. Distributive patterns do not only involve statistically significant differ­ ences but may also be understood within the framework of qualitative studies as gendered patterns identified in a sample. Gender stereotyping involves interpreting a gendered distribution or pattern as a categorical distinction. This ignores the social and historical limitations of the pattern, variation within each group, and the overlap between women and men. Consequently, that which characterises a majority of women will be equated with femininity, whereas that which characterises a majority of men will be equated with masculinity—and this may well be true even in cases in which almost half of a gender group does not fit a given characteristic (Fine, 2010). The mechanism underlying this slip is that we tend to notice behaviour that confirms gender stereotypes, whereas we marginalise as exceptional behaviour that deviates from the stereotypes and overlook more gender-neutral behaviour. Because masculinity and femininity—whatever they are used to designate—neither form a clear dichotomy nor are isomorphic with gendered bodies, the result is often a conceptual confusion in which some women are viewed not as ‘real’ women and some men are viewed not as ‘real men’. This confusion may also arise in research in which gender is used as a variable and average differences between women and men are circularly explained by ‘gender’. This systematically ignores how the meaning of survey or interview questions posed may be different for women and men because of their different positions in the social order and different frames of interpretation. As the Norwegian psychologist Hanne Haavind once wrote (before the legalisation of gay marriage): If women and men on average express the same degree of satisfaction with their marriages, are they then satisfied with the same things? Have the women and the men who have used the scale for self-measurement compared themselves with others in the same gender position in marriage as themselves, or have they assessed their satisfaction in comparison with those in the other gender position?… The method misses the significance of the fact that women are married to men, whereas men are married to women. haavind, 2000: 165, translated from Norwegian

Here, it is important to note that focusing on only gender difference tends to limit the analysis of gender to one that is an individual characteristic only. However, gender can also be described as a social relation and an entity of

Gender as Analytic, Political and Interdisciplinary Concept

267

meaning that emerges in processes of interaction. We do not always understand the dynamics of a classroom or a workplace more clearly by being aware of small average differences in cognitive skills and behaviours. This perspective calls attention to the dynamics of power and negotiations in social constructions of meaning. It also opens up towards understanding multiple forms of femininity and masculinity (inflected by dimensions such as social class, age, or racialised and ethnic status) and the hegemonic position that a particular type of femininity or masculinity may attain in a given context (Halberstam, 1998; Connell, 2000). Gender is empirically present both in bodily appearance and experiences and in the patterning of social structures, interactions and identities. However, gender is also a forceful frame of interpretation in our minds. Such gendered discourses are applied not only to people but also to non-human things such as colours, nations, ships, bombs, and tornadoes (Scott 1988). Gender as a frame of interpretation may involve double standards and attributions in which behaviours are interpreted and valued differently according to a person’s sex. When a boy performs well in school, his performance is often considered the result of intelligence, whereas if a boy performs poorly in school, he might be considered lazy or merely bored. When a girl performs well in school, her performance it is often deemed the result of her dutifulness and hard work, whereas if a girl performs poorly in school, she might be considered to lack intelligence (Walkerdine, 1990). Such double standards and attributions seldom work on a conscious level; they slide in as taken-for-granted dimensions of ways of thinking about and practicing gender. In the 1980s-era debate about the problems experienced by girls in school, the proposed solutions focused on individual changes, such as finding ways to strengthen girls’ self-confidence. When the problems boys experienced in school came into focus 20 years later, the analysis and solutions were framed in a structural way: the school system did not meet the needs of boys and therefore ought to be changed (Öhrn, 2000). What appears to characterise gender as a historical discourse is the tendency to split and dichotomise phenomena into two distinct groups, with the resulting dichotomy read as a hierarchy. Things defined as feminine tend to be viewed as secondary or even inferior to things defined as masculine. Simone de Beauvoir writes in her landmark 1949 book that women are perceived and treated as ‘the second sex’ (Beauvoir, 1989). According to de Beauvoir, men represent the universal human, the unmarked category of mankind, whereas women constitute a special gendered sub-category whose gender explains their deviation from universal humanity. This way of thinking may persist even in situations in which women have achieved positions of power or girls’ educational achievements exceed those of boys. Thus, there is no automatic

268

Nielsen

­connection between gender as it is present empirically in the world and gender as a frame of interpretation. However, interpretive models of gender interact with gender in the world as symbolic gender enters into experiences and practices related to gender. Gender in our heads and gender in the world thus continually feed into one another.

Sex and Gender

The degree to which measures of psychological gender differences are actually related to gender differences found in genes, brain structure and hormone levels, or whether such measures depend on learning and experience or a mixture of the two (Hines, 2004; Fine, 2010), is largely unknown. The whole idea of a one-way causal route from biology to behaviour has been questioned by recent research documenting the remarkable flexibility of the human brain, the contextual contingency of bodily processes, and the ability even of genes to adjust their effects to individual life circumstances (Rose, 2005; Keller, 2010). This does not mean that gendered patterns of behaviour are a mirage or that the patterns that do exist have no biological basis (even if we do not know exactly what that basis is). The point is that there is no clear or straightforward connection between the near-dichotomous dimensions of anatomical sex and the complex, multi-dimensional and context-dependent nature of gender differences and gendered patterns of behaviour. The distinction between biological and social gender—sex and gender,2 in English usage—has performed important political work in feminist theory, but it has also been an issue of heated theoretical debate in the past few decades. In the 1960s, the distinction came into use in the English language, 2 There are different reasons for the inconsistency in the use of the term “gender” either as a general category or as covering only socially constructed aspects. One is that unlike modern English, not all languages have two words for sexual difference. The French term “sexe”, the Norwegian “kjønn”, the Danish “køn” and the Swedish “kön” cover both the biological and social aspects of sexual difference in a single word. This was also the case in English before the 1960s, when “sex” was the general term for sexual difference and “gender” was merely a grammatical term. Since the 1990s, gender has gradually taken over from sex as the general term (Nicholson, 1994: 80; Moi, 1999: 31; Cassin, Apter, Lezra and Wood, 2014: 376). After the sex/ gender division gained influence in gender studies, Swedish gender researchers introduced the term “genus” to designate social gender in Swedish, and this term has achieved a more general use, including in Swedish politics. Consequently, the original Swedish word “kön” is now associated more with biological gender and sexuality. Danish and Norwegian gender researchers have not followed suit and continue to prefer the integrated term “kjønn” or “køn”.

Gender as Analytic, Political and Interdisciplinary Concept

269

with ‘sex’ referring to the biological differences between women and men and ‘gender’ referring to psychological identities and cultural attributions (Cassin et al., 2014: 375). This distinction between sex and gender soon spread to the international community of feminist researchers. In her influential essay ‘What is a woman?’, the literary scholar Toril Moi analyses the development of the terms within feminist theory: ‘Gender may be pictured as a barricade thrown up against the insidious pervasiveness of sex’, she writes (Moi, 1999: 15). The point is that when sex is perceived as pervasive, the difference between male and masculine or female and feminine disappears. The result is indeed a case in which distribution is interpreted as distinction. During the nineteenth century, gender difference increasingly became understood in terms of reproductive biology. Previously, gender worked primarily as an axis of divisions of labour along with the distribution of power, authority and privilege. With an enhanced division between public and private life in the nineteenth century and the rise of modern medical science, the difference between males and females came to be understood as fundamental, dictated by their biology and inborn psychological capacities (Laqueur, 1990; ­Fausto-Sterling, 2000). Thus, biology was from then on used to naturalise and justify social gender norms. Women were reduced to their biological reproductive capacity and the consequences this was assumed to have for their mental capacities (Rosenbeck, 1987; Moi, 1999). This gender polarisation emerged in the historical period that introduced a completely new view of human beings based on the ideas of the European Enlightenment. Women were viewed as not fully human, but instead as a subgroup defined and confined by their bodies. Although the more extreme version of this biological determinism dissi­ pated during the first decades of the twentieth century as women gained access to education and the right to vote, observed gender differences and gender hierarchies were still largely explained and justified by biological difference. Thus, when the second feminist movement emerged in the 1970s, there was an urgent need to build an argumentative barrier, as Moi notes, to refute biological arguments that could be used against expanding women’s rights. The response to this need was to separate the concept of social and historical ‘gender’ from the concept of biological ‘sex’. It was not the feminists, however, who invented this distinction. In 1963, the psychiatrist Robert Stoller, who worked with intersex and transsexual patients, introduced the concept of ‘gender identity’ as something related only to the psychological experience of belonging to one sex or the other (Moi, 1999: 22). The concept of ‘gender role’ was conceived in the same period and was used in Talcott Parsons’ work on the nuclear family. When feminist scholars started to use the sex/gender distinction in the early 1970s (see, for instance, Oakley,

270

Nielsen

1993), however, they emphasised not psychological gender identities or the functional order of the nuclear family but the conflict, power and suppression inherent in the gender order. In 1975, social anthropologist Gayle Rubin published the important essay ‘The traffic in women’, in which she used Stoller’s categories for feminist purposes and viewed sexuality and bodily sexual difference as the ‘raw material’ for the production of gender: … a “sex/gender system” is the set of arrangements by which a society transforms biological sexuality into products of human activity, and in which these transformed sexual needs are satisfied. rubin, 1975: 159

According to Rubin’s definition, gender is always oppressive because it is shaped by patriarchal power systems: the social organisation of sex rests upon gender, obligatory heterosexuality, and the constraint of female sexuality (Rubin, 1975: 179, 204; see also Moi, 1999: 24). In the following decades most gender research was based on this distinction between sex and gender, but eventually the idea of a biological ‘nature’ working as a universal, mute and passive matter was challenged. The sex/­gender distinction may have helped women escape biological determinism, but it still provided a biological foundationalism for cultural gender dichotomies in which biology was assumed to be an unchangeable and neutral ‘coat rack’ for different cultural gender garments (Nicholson, 1994: 81–82). Drawing on Laqueur and Foucault, Linda Nicholson argued that our conceptions of the body are also always cultural and historical. She mentioned the distributive character of most biological differences, which do not fit into a binary model of sex. Judith Butler (1990) argued that our ideas of biological sex—and, indeed, the conception of heterosexuality as natural—should be regarded as an effect of the gender system, not its foundation. Butler also suggested that the materiality of the body is not universally given, but instead are produced by ‘materializing effects of regulatory power in the Foucaultian sense’ (Butler, 1993: 10). From a sociological perspective, Pierre Bourdieu wrote in 1990 (in the first draft of La domination masculine) that the sexed body should be regarded as a social construction produced by the symbolic violence in gender domination, employing his concept of habitus as learned bodily dispositions.3 Another 3 Bourdieu’s concepts of habitus, field and capital have inspired many feminist thinkers (see later in this chapter). However, the publication (Bourdieu, 1998, including the prior version from 1990) in which Bourdieu specifically addresses gender never became an important text in gender theory. Judith Butler (1999), Toril Moi (1991) and Bourdieu-inspired feminists such

Gender as Analytic, Political and Interdisciplinary Concept

271

­critique originated from the emerging field of feminist science and technology studies, opposing the view of the Western philosophical tradition in which nature is viewed as passive matter that humans can control and dominate. Haraway held that ‘the “body” is an agent, not a resource’ and suggested viewing it as an active, meaning-generating ‘material-semiotic actor’ (Haraway, 1988: 594–595). From the phenomenological perspective, Toril Moi developed Simone de Beauvoir’s claim that the body should not be understood as a reified thing but as ‘a situation’ related to a person’s lived experience and subjectivity. In contrast to Butler and perhaps Bourdieu, Haraway and Moi did not contest bodily structures as facts, but as facts with no absolute meaning: ‘… the relationship between body and subjectivity is neither necessary nor arbitrary, but contingent’ (Moi, 1999: 40, 114). The meaning of a body can be decided only in a historical analysis of individual women or men. Like Gayle Rubin, Moi finds the concept of gender no less reductive than the concept of sex because myths of cultural gender identities may work just as deterministically as fixed ideas of sex.

Equality and Difference—Gender, Power and Politics

A different way of discussing the meaning of the concept of gender is to frame it as a political question of gender equality and gender difference. Is the goal of feminism to make women and men equal or to dispose of the hierarchical relation between them? Is it the dichotomy or the hierarchy of the gender order that is the source of evil? Carol Patemann coined the problem as ‘Wollstonecraft’s dilemma’ (Pateman, 1989: 196–197), after one of the first European feminists, the British Mary Wollstonecraft, who published A vindication of the rights of woman in 1792. The dilemma operates such that either women can insist on obtaining the same social rights as citizens as men but thereby take men’s lives and values as models and become like them, or women can insist on gaining respect and recognition as women and mothers, with the consequence that they will not attain recognition as full citizens. Thus, the tension between equality and difference is a product of the contradiction between as Lois McNay (2004) and Lisa Adkins (2004) have disregarded it and typically relegated it to a footnote. It is dismissed partly as a simplistic structuralist account with no understanding of the complexities, contradictions, ambivalences and gradual changes that characterise gender in modern society (Moi, 1999: 286; Adkins, 2004: 191–192; McNay, 2004: 189, note 4) and partly for kicking in doors that had been opened long before by feminist scholars whom Bourdieu apparently had not read (Moi, 1999: 283, note 21).

272

Nielsen

universal human rights and gender polarisation that first emerged in the late eighteenth century. The various feminist responses to Wollstonecraft’s dilemma rely on different conceptions of gender, power and knowledge, and these positions may be understood both genealogically and theoretically. Iris Marion Young identifies a tension between a ‘humanist position’ favouring equality and a more ‘gynocentric position’ favouring difference (Young, 1985). The humanist ­position— which ties back to Mary Wollstonecraft, John Stuart Mill and Simone de Beauvoir—views women as human beings whose potential is inhibited and distorted by a society that allows self-development only for men. Women’s confinement to femininity is what impedes the development of their full human potential. Thus, the primary vehicle of women’s oppression is the idea of femininity as an essence, which does not have a parallel in a similar idea of masculinity (Young, 1985: 173). There is no celebration of gender differences in this position. The feminist political project here involves individual freedom: women have an objective interest in challenging the cultural mind-set both in themselves and in others that reduces their freedom. Gynocentric feminism disagrees that the problem is that women are not allowed to fully participate in humanity/being human. Rather, it lies ‘in the denial and devaluation of specifically feminine virtues and activities by an overly instrumentalised and authoritarian masculinist culture’ (Young, 1985: 176). Second-wave feminism incorporates both positions: the fight for women’s rights is also a fight for a better society. Socialist and Marxist feminists of the 1970s focused on the structural and economic power that oppressed women and on the necessity for collective actions to challenge this. In the late 1970s, the gynocentric position gained influence, partly as a result of the many new empirical studies conducted within different academic disciplines that revealed women’s lives and valuable activities and contributions. Simone de Beauvoir came under heavy criticism for her devaluation of women’s bodies, reproductive labour and activities and for mirroring a patriarchal culture in her idealisation of men’s situation in society. The gynocentric position of the 1980s lead to a somewhat exalted celebration of women as linked with nature, nurturing, cooperation and peace-loving. It was also used to engage in a more radical critique of the West’s capitalist and patriarchal society not only for oppressing women but also for engaging in the gendered separation of production and reproduction. As a result, care and human relations are devalued and a destructive economic logic, oppressive to both men and women, comes to dominate society. Whereas the conception of knowledge in humanist feminism relied on Enlightenment ideas of objectivity and critical-rational thinking, the gynocentric position became more preoccupied with the subjective

Gender as Analytic, Political and Interdisciplinary Concept

273

aspects of knowledge production and the links between power and the construction of knowledge. Nancy Hartsock (1983) formulated the epistemology of the ‘feminist standpoint’, which claims that women’s position and experiences in the sexual division of labour produce not only a different but also a superior understanding of nature and social phenomena. Humanist and gynocentric feminism proposed different solutions to the equality/difference dilemma. However, neither had questioned the category of women, and it was only to a limited degree that they questioned whether women have common political interests. The notion of a universal female subject was first challenged in the 1980s by Black feminists in the us, who argued that race and class were overlooked dimensions in feminism: Black women were oppressed as much by white middle-class women as by Black men (see, for instance, Davis, 1981; and Hull, Scott and Smith, 1982, the latter with the telling title All the women are white, all the blacks are men, but some of us are brave). This idea complicated the question of gender equality, gender difference, and the visions of women’s collective action against oppression. The emergence of postcolonial and poststructuralist theories in the 1990s introduced new ways to consider power, knowledge and politics. The Marxist or patriarchal concept of power as a top-down model of domination was challenged by Foucault’s discursive understanding of power as something that is everywhere, something that nobody can escape, a social force that creates and regulates which categories of meaning and subject positions are possible. Both women and men contribute to maintaining the gender order through their actions and language. Thus, a crucial difference from the concept of power found in humanist and gynocentric feminism was that women were viewed neither as victims nor as angels but as co-producers of the gender order. The epistemological relativism introduced by standpoint theory was radicalised in the poststructuralist version, in which not only is knowledge entangled with power, knowledge is power. From the poststructuralist perspective, the equality/difference dilemma tends to lose meaning. When the focus is on women primarily as an internally diverse group, the concept of feminist politics is challenged, whether as a struggle for gender equality or as a struggle for the acknowledgement of women’s special values and identities. Because identity categories are viewed both as constituted by processes of exclusion and sites of power struggles and as internally dependent on what is externalised, any category will be internally unstable and the target of continual resignification (Butler, 1990; Hall, 1996). Thus, the approach to politics and change in poststructuralist and queer theory tends to work on the level of cultural categories and representations: normative categories of gender and sexuality must be deconstructed and destabilised

274

Nielsen

to reveal that they are neither natural nor innocent identities. The inquiries begin from the margins with people engaged in non-normative gender and sexual practices who do not feel like they fit into whatever ‘we’ is articulated as a norm (Stormhøj, 2013: 65). The queer perspective gives priority to nonnormative sexualities because heteronormativity is viewed as inherent in the meaning of gender. The queer perspective introduced a different understanding of gender politics, which carried great importance in an increasingly mediatised and sexualised society. However, it also created new problems. For instance, it is difficult to discern how to effectively struggle against discrimination and inequality among existing groups if all identity categories are viewed as unstable. In addition, the focus on cultural categories and borders may diminish attention to economic and material inequality and weaken the critique of a society that devaluates care and interdependence (Fraser, 1997, 2009). A further issue was the claim that change must stem from people located at the margins, which tends to conceptualise the dominant norm as monolithic and undynamic (McNay, 2004; Stormhøj, 2013). The dominant norm emerged as a static background to non-normative performances, not taking into account that gender norms and practices are constantly experiencing a process of reconfiguration that may sometimes contribute to increased gender equality—for instance, when more women take up paid work and more men participate in child care (Adkins, 2004; McNay, 2004; Ellingsæter and Leira, 2006; Aarseth, 2009; Nielsen, 2017). This brings us back to the dilemma between equality and difference in a new way: does a destabilisation of the binary structure of gender lead to fewer social inequalities between people of different genders or different sexual identities? Or is it the other way around: does increased social equality between women and men or between people with different sexual identities make the category of gender less important, constraining and exclusionary? Is it gender as experience or gender as representation that is the main problem? Or can these two perspectives be combined? I will return to this question in my conclusion, but first I will accentuate the difference between these perspectives to illustrate how the dichotomy has led to further conceptual polarisations. ​

Being, Doing, Performing Gender: Subjectivity and Agency

Viewing gender primarily as a cultural representation or primarily as lived experience has consequences for the understanding of how gender identities and gender norms are produced, maintained and transformed. Various theories have emphasised different aspects of the process of learning gender

Gender as Analytic, Political and Interdisciplinary Concept

275

throughout life: gender as ways of being and relating; gender as embodied dispositions; gender as a dimension of interaction, play and negotiation; gender as a norm and a process of normalising. These approaches are also based on different notions of subjectivity and agency. There has generally been an increased focus on gender as something that is produced and maintained through practices. Instead of viewing gender as something one ‘is’, the processual perspective emphasises gender as something one ‘does’ (West and Zimmerman, 1987). West and Zimmerman’s sociological version of ‘doing gender’ (1987) relies on symbolic interactionism in which social meaning is created in interaction. Similarly, in Simone de Beauvoir and Toril Moi’s philosophical version, the philosophical distinction between existence and essence is central: it is our acts that define who we are (Moi, 1999: 55). In both versions, there is an acting subject who evolves through agency and in response to others’ agency. Experiences of interaction are sedimented over time in a way that means that how a person ‘does’ gender also becomes an expression of who she has become through these actions, and thus in that sense, ‘is’. Hence, studies of ‘being’ gendered and ‘doing’ gender could here be understood as functionally related, revealing different aspects of social processes involved in constructing gendered subjectivities. Agency and subjectivity in this case become mutually intertwined. This is also a perspective we find in the work of feminist sociologists who have developed Bourdieu’s concept of habitus but placed a stronger emphasis on agency and reflexivity than we find in Bourdieu’s own work. Lois McNay, for instance, understands gender as a ‘lived relation’ instead of as a structural location, whether in terms of materialist or cultural thinking (McNay, 2004: 175). Lisa Adkins connects modern gender detraditionalisation with gender arrangements in late modern society that contribute to increase reflexivity as ‘a habit of gender’ in a continuous interchange with pre-reflexive or embodied meanings connected to gender (Adkins, 2004: 192; see also Aarseth, 2009; Nielsen, 2017). In the poststructuralist version of ‘doing gender’, however, agency and subjectivity are split. In Butler’s account, the theory of the internal instability of norms and categories is combined with a theory of performativity, which is defined as ‘the reiterative and citational practice by which discourse produces the effect that it names’ (Butler, 1993: 2). Gender exists only as performance, and there is ‘no doer behind the deed’. The instability of the category of gender, in combination with its performative character, makes the category open to change, as the norms may unintentionally be wrongly cited. However, given that these failed performances or resignifications are merely coincidental consequences of the indeterminacy of linguistic categories triggered by unconscious processes in the speaker, their effects are also undetermined (McNay,

276

Nielsen

2004). Within this particular philosophical line of thinking, subjective agency is mainly understood as unconscious defensive reactions that guard the illusion of gender. Compared with West and Zimmerman, there is little focus on social interaction in Butler’s version of doing gender. She subsequently modified her ideas in this regard (for instance, in her 2004 book Undoing Gender), but in her two books from the early 1990s, which had and still have a significant impact on gender studies, gender performance appears to be a rather solitary show. The relational psychoanalyst Lynne Layton holds that if identities are theoretically reduced to always and only being defensive formations, we lose not only the relational meaning and the biographical motives underlying the performance of gender but also the possibility of understanding identities as more or less conflicted and relations as more or less defensive: What is missing from Butler’s account, even in its most psychoanalytic form, is an understanding of what motivates people’s relation to norms. Indeed, one reason why it is so difficult to grasp what Butler means by agency is that her system has no place for the mediating power of relationships, for longing for love, approval, and recognition. layton, 1998: 150

This also leaves us in a rather timeless universe from which the historical contingencies of subject formations and agencies are more or less absent (Benhabib, 1995; Fraser, 1997; Moi, 1999; Stormhøj, 2013). In the words of Lois McNay, the lack of a temporal dimension in Judith Butler’s theoretical work implies that gender ‘tends to be construed as a relatively atemporal system of dominant norms’ in which the possibility of change is unmediated by praxis or agency (McNay, 1999: 102). Raewyn Connell has argued that Butler’s idea of generalised categorical instability cannot take into account that in certain historical periods gender identities and relations change quickly, whereas in others, they change slowly. Nor does it explain why some people want to change gender arrangements, whereas others resist (Connell, 2009: 90).

Realism and Nominalism

The relationship between gendered subjectivities and gender performativity can also be connected to the opposition between realism and nominalism, that is, the question of whether words and concepts refer to empirical phenomena or only to the linguistic and discursive systems themselves. This distinction is

Gender as Analytic, Political and Interdisciplinary Concept

277

often confused with the distinction between essentialism and constructionism, but it is not the same thing. Essentialism means viewing gender as a fixed, ahistorical and coherent essence in individuals. Hardly any feminists or gender researchers would subscribe to such an idea. Instead, the essentialism debate concerns what counts as essence and what is implied in construction (Hacking, 1999). Most social researchers are constructionists in the sense of Bourdieu or Berger and Luckmann, meaning that they think that gender, body and sexuality vary historically and culturally and that such ideas and practices have an impact on what women and men are allowed to become. This form of social constructionism typically relies on critical realist, interactionist or interpretative paradigms in which the concept of the ‘reality’ of gender is neither abolished nor used in a linguistically unmediated, positivistic sense. The various poststructuralist positions identify as nominalist instead of constructionist in the social and historical sense. The nominalist position moves the focus of interest towards how gender is constituted for our minds. The claim is that the linguistic binarity between man/woman or between ­homosexuality/ heterosexuality and repetitions of those binaries in discursive practices unavoidably create gender as an essentialist category of meaning. From this ­perspective, there are no identities or essences in the world to discover or describe, only differences and borders that are composed of complex power ­dynamics and result in on-going social processes of inclusion and exclusion. A problem that arises here, not least for social researchers, is that if there are only differences and no identities to study, meta-communication will tend to be substituted for communication in research and politics: we cannot speak about the world, only about how we speak of the world (Nielsen, 1995, 2000; Hacking, 1999; Knapp, 2005). To analyse how we speak of gender is obviously important, and feminists have made important critical interventions into mainstream science in this regard. One example is Donna Haraway, who convincingly demonstrated how biological studies of animals were based on human gender stereotypes, legitimising these gender norms as natural (Haraway, 1992). Another example is Hanne Haavind’s above-mentioned intervention, in which she criticises the use of gender as a self-explaining variable (Haavind, 2000). The words we use are not innocent. When we talk about men and women or allow specific groups to embody a general category, we also contribute to the idea that gender is a given and dichotomous phenomenon. However, this point does not consider that men and women act in the world and that they carry the gendered structures of society and culture both with them and in them. If describing them—even if only in the distributive sense—­automatically means essentialising them, feminist social sciences and historical studies lose ground. Gender studies can hardly survive as an ­interdisciplinary research field only by

278

Nielsen

offering critiques of others’ use of language; it must also offer new and better descriptions of the world. Describing the world implies dialoguing with others, not only unveiling their perspectives—and that means listening to what people are attempting to say, not just how they say it. This observation applies to theories, politics and informants alike. A meta-perspective on a theory, concept or utterance is not the same as addressing the relevance or validity of that theory, concept or utterance. To some extent, the split between nominalism and critical/interpretative realism may follow a split between the humanities and the social sciences. However, it is fully possible to conduct deconstructive work in the social sciences and to study historical ideas of gender and sexuality without viewing them solely as discursive positions. Toril Moi, for instance, takes a realist stand when she criticises the anti-realist and anti-dialogical position inherent in philosophical difference thinking. Departing from ordinary language philosophy (Wittgenstein instead of Saussure and Derrida) she claims that the meaning of words arise from what we use them for, not primarely from their oppositions to other words. This places language use in a historical context and embeds it in specific social relations, which is what is missing from Butler’s early account of performativity. Thus, it is not the word woman that is the problem but what is claimed, directly or indirectly, regarding women and femininity in a given context. However, Moi also appears to fear certain words when she adamantly refuses to discuss gender identities in any other sense than as myths and stereotypes. In contrast to Butler, she thinks individual gendered subjectivities exist and suggests that one should study the variation in women’s lived experiences. However, Moi rejects the idea that it could also be important to study social patterns, e.g., the similarities in women’s subjectivities: ‘To speak of a generalised “gender identity” is to impose a reifying or objectifying closure on our steadily changing and fluctuating experience of ourselves in the world’ (Moi, 1999: 81). The question is, what is Moi implying with the word ‘generalised’ here? If it means a claim regarding an essential feminine identity, any gender researcher would agree—but what about smaller and more limited patterns of similarities between certain women in a specified context (see also Bordo, 1990)? Other gender scholars from the humanities are more open to such limited generalisations of gender patterns, employing, for instance, Wittgenstein’s idea of family resemblances (Nicholson, 1994) or Sartre’s concept of seriality (Young, 1994). These questions hold particular importance for empirical gender studies because the ultimate goal here is seldom the study of language or of individual or literary cases of experience. For gender researchers working empirically with women and men’s lives currently or historically, it is indeed often possible to

Gender as Analytic, Political and Interdisciplinary Concept

279

identify distributive gender patterns. Recognising such patterns is equivalent neither to essentialising cultural forms nor to imposing stereotypes on people, which ignore their freedom and agency. We can have gendered patterns even if they do not apply to everybody in one gender group, if they apply to some in the other gender group as well, or if they are limited to certain classes, contexts or periods. It is important to identify such patterns, which represent a social practice that not only is part of a social order but also provides some of the conditions for change. Thus, for gender researchers working empirically, gender in the world can neither be reduced to nor radially separated from the gender categories through which it is constituted and experienced. The epistemological space that social researchers inhabit is characterised by continuous mediation among the characteristics of the objects studied, the objects’ interpretation of themselves, and the researcher’s interpretation of how she interprets the objects and herself. Conclusion As an analytical, interdisciplinary and inherently critically concept, gender has multiple dimensions and raises intricate questions of ontology, epistemology and politics. It is not possible to establish a simple overview of positions in the field because the theoretical issues work on many and partly intersecting levels. Their meaning may also change depending on the academic field and the research questions raised. The dimensions discussed in this article in­ cluded the non-reducible, analytic meaning of gender as a simultaneous distinction, distribution and discourse; different conceptions of sex and gender and how they relate to assumptions about the mind-nature dichotomy; the tension between gender equality and gender difference with wider impacts on epistemology and the conception of power and politics; and gender as something one is, does or performs, and the consequences of this notion for the understanding of subjectivity and agency, reproduction and change. Finally, I discussed the issues of essentialism and constructionism and of nominalism and realism as different epistemological points of departure. I emphasised a line of division between gender researchers working with gender mainly as a representation and those working with gender mainly as an experience. These two points of departure have had further consequences not only for the understanding of power and politics but also for the conceptions of biological and social gender and how the concept of gender is understood. In practice, however, gender researchers often draw on both positions but in different proportions. For critical realists, it is evident both that gender is a representation

280

Nielsen

and that this representation plays a part in what men and women in the world can become. Gender as a frame of interpretation cannot be separated from the distributional patterns found or the manner in which gender as distinction is employed. Thus, the nominalist position may be part of the epistemological considerations of the ‘realists’ but not their ontological foundation. Most of the ‘nominalists’ would admit that they are not attempting to describe all aspects of how gender works. For instance, in Judith Butler’s later works, she attempts to clarify that of course she understands that there are things that exist outside language and that the empirical doers of gender obviously have a psychological and social history that precedes the deed, noting that this is not the issue she is addressing in her theoretical work (see, for instance, Butler, 2004: Ch. 10). The differences may thus boil down to the fact that there are many and different phenomena to study and many and different questions to ask precisely because gender studies is constituted as a broad and interdisciplinary field. The advantage of this is that gender studies has access to a broad set of knowledge and theoretical approaches. For this reason, gender studies may occupy a privileged position in terms of academic and political border crossing. Instead of viewing the relationship between nominalism and critical/ interpretative realism as a split, it can also be put to work as a methodological challenge and possible integration. However, the interdisciplinary character of the field also raises the risk that one discipline, or a specific theory from this discipline, may at times dominate the field, leading to uniformity instead of to multiplicity. This may result in a reductive approach (everybody poses the same questions) or theoretical claims with unclear limits (when claims are moved from one field of inquiry into another without asking whether the object of study is the same). We may observe an example of this in the debate on being, doing and performing gender, in which the disagreement appears to stem from a confusion of different identity concepts and the work they are intended to perform in different ­disciplines or in different research inquiries. Butler’s concept of the subject is positioned within a longstanding philosophical dispute about whether ‘the subject’ can be claimed to be the ultimate foundation of reason. In sociology and psychology, the concept of ‘subject’ performs different work. In a sociological context, it will often refer to shifting social and political identities. In psychology, it refers to the realm of individual experience, including the experience of being ‘oneself’ even in the midst of social and personal change. If the meaning of words is what they are used for, this is unproblematic. However, it becomes a problem if the meaning of a concept in one context is uncritically imported into another. If the philosophical distinction between identity and difference enters the social sciences or historical studies as a general ‘theory’—without Butler’s reservations or later amendments—it may become

Gender as Analytic, Political and Interdisciplinary Concept

281

two distinctly ­different methodological approaches. Then, one suddenly must choose between either studying people’s identities as unitary, coherent and universal, or studying how their identities are linguistically constituted by the marginalisation of other identities. This dichotomy makes it problematic to speak both of living subjects and lived identities and of variations, relations and historical formations because these words and their reference to social phenomena are interpreted as essentialist thinking. However, the real problem here is the dichotomy of identity versus difference that does not fit many of the inquiries in social sciences or historical studies. As Lynne Layton (1998) has argued, from a psychological perspective, identity and difference, coherence and incoherence, and stability and change in identities and ­subject ­formations are simply not an either/or question. Thus, interdisciplinarity may sometimes result in unhelpful theoretical dualisms. Rachel Falmagne (2009) claims that ­theoretical differences and tensions are productive when they are used c­ onstructively to transform reconceptualisations and selective syntheses but not when they are constructed within a dualistic frame that remains within a particular theoretical territory and its corresponding community of ­scholars. Falmagne argues that theoretical tensions often signal that each account ­excludes important features of the phenomenon studied, which its language is ill equipped to ­address. However, if theoretical tensions are not reduced to a question of either/or, they can ­represent productive tensions for feminist scholars both to explore and to find ever-new ways to hold. References Aarseth, H. (2009). “From modernized masculinity to degendered lifestyle projects”, Men and Masculinities. 11(4): 424–440. Adkins, L. (2004). “Reflexivity”, Eds. L. Adkins and B. Skeggs. Feminism after Bourdieu. Oxford: Blackwell. Beauvoir, S. de. (1989) [1949]. The second sex. New York: Vintage books. Benhabib, S. (1995). “Subjectivity, historiography, and politics”, Ed. L. Nicholson. Feminist contentions. New York and London: Routledge. Bordo, S. (1990). “Feminism, postmodernism, and gender-scepticism”, Ed. L.J. Nicholson. Feminism/Postmodernism. New York-London: Routledge. Bourdieu, P. (1998). La domination masculine. Paris: Seuil. Butler, J. (1990). Gender trouble. New York and London: Routledge. Butler, J. (1993). Bodies that matter. New York and London: Routledge. Butler, J. (1999). “Perfomativity’s social magic”, Ed. R. Schusterman. Bourdieu: a critical reader. Oxford: Blackwell. Butler, J. (2004). Undoing gender. New York and London: Routledge.

282

Nielsen

Cassin, B., Apter, E., Lezra, J. and Wood, M. (2014). Dictionary of untranslatables: a philosophical lexicon. Princeton: Princeton University Press. Connell, R. (2009). Gender. In a world perspective. Cambridge: Polity Press. Connell, R.W. (2000). The men and the boys. Sydney: Allen & Unwin. Davis, A. (1981). Women, race and class. New York: Random House. Ellingsæter, A.L. and Leira, A. (eds.). (2006). Politicising parenthood in Scandinavia. Bristol: Policy Press. Falmagne, R.J. (2009). “Subverting theoretical dualisms”, Theory & Psychology. 19(6): 795–815. Fausto-Sterling, A. (2000). Sexing the body. New York: Basic Books. Fine, C. (2010). The delusions of gender. New York: W.W. Norton. Fraser, N. (1997). “Heterosexism, misrecognition, and capitalism: a response to Judith Butler”, Social Text. 15(3–4): 279–289. Fraser, N. (2009). “Feminism, capitalism and the cunning of history”, New Left Review. 56(March–April): 97–117. Haavind, H. (2000). Kjønn og fortolkende metode: metodiske muligheter i kvalitativ forskning. [Gender and interpretative method: possible methods for qualitative research]. Oslo: Gyldendal Akademisk. Hacking, I. (1999). The social construction of what?. Cambridge, ma. and London: ­Harvard University Press. Halberstam, J. (1998). Female masculinity. Durham: Duke University Press. Hall, S. (1996). “Introduction: who needs ‘identity’?”, Eds. S. Hall and P. Du Gay. Questions of cultural identity. London: Sage. Haraway, D. (1988). “Situated knowledges: the science question in feminism and the privilege of partial perspective”, Feminist Studies. 14(3): 575–599. Haraway, D. (1992). Primate visions: gender, race, and nature in the world of modern science. London and New York: Verso. Hartsock, N. (1983). Money, sex and power: towards a feminist historical feminism. New York: Longman. Hines, M. (2004). Brain gender. Oxford: Oxford University Press. Hull, G.T., Scott, P.B. and Smith, B. (eds.). (1982). All the women are white, all the blacks are men, but some of us are brave. New York: The Feminist Press. Keller, E.F. (2010). The mirage of a space between nature and nurture. Durham and ­London: Duke University Press. Knapp, G.-A. (2005). “Race, class, gender: reclaiming baggage in fast travelling theories”, European Journal of Women’s Studies. 12(3): 249–265. Laqueur, T. (1990). Making sex: body and gender from the Greeks to Freud. Cambridge, ma.: Harvard University Press. Layton, L. (1998). Who’s that girl, who’s that boy? Clinical practice meets postmodern gender theory. Northvale, New Jersey-London: Jason Aronson.

Gender as Analytic, Political and Interdisciplinary Concept

283

McNay, L. (1999). “Gender, habitus and the Fields. Pierre Bourdieu and the limits of reflexivity”, Theory, Culture and Society. 16(1): 95–117. McNay, L. (2004). “Agency and experience: gender as lived relation”, Eds. L. Adkins and B. Skeggs. Feminism after Bourdieu. Oxford: Blackwell. Moi, T. (1991). “Appropriating Bourdieu: feminist theory and Pierre Bourdieu’s sociology of culture”, New Literary History. 22(4): 1017–1049. Moi, T. (1999). What is a woman? And other essays. Oxford: Oxford University Press. Nicholson, L. (1994). “Interpreting gender”, Signs. 20(1): 79–105. Nielsen, H.B. (1995). “Seductive texts with serious intentions”, Educational Researcher. 24(1): 4–12. Nielsen, H.B. (2000). “Rommet mellom sitering og dekonstruksjon”, Kvinneforskning. [“The space between citation and deconstruction”, Women’s studies]. 24(1): 16–30. Nielsen, H.B. (2017). Feeling gender. A generational and psychosocial approach. London: Palgrave Macmillan. Oakley, A. (1993) [1972]. Sex, gender and society. Aldershot: Arena. Öhrn, E. (2000). “Changing patterns? Reflections on contemporary Swedish research and debate in gender and education”, nora. 8(3): 128–136. Pateman, C. (1989). The disorder of women: democracy, feminism and political theory. Cambridge: Polity Press. Rose, S. (2005). The 21st-century brain: explaining, mending and manipulating the mind. London: Jonathan Cape. Rosenbeck, B. (1987). Kvindekøn: den moderne kvindeligheds historie 1880–1980. [Female gender: the history of modern femininity 1880–1980]. Copenhagen: Gyldendal. Rubin, G. (1975). “The traffic in women: notes on the ‘political economy’ of sex”, Ed. R.R. Reiter. Towards an anthropology of women. New York: Monthly Review Press. Scott, J.W. (1988). Gender and the politics of history. New York: Columbia University Press. Stormhøj, C. (2013). “Queer theories, critiques and beyond”, Kvinder, køn og forskning [Women, gender and research]. 13(1): 61–72. Walkerdine, V. (1990). Schoolgirl fictions. London: Verso. West, C. and Zimmerman, D.H. (1987). “Doing gender”, Gender & Society. 1(2): 125–151. Young, I.M. (1985). “Humanism, gynocentrism and feminist politics”, Women’s Studies International Forum. 8(3): 173–183. Young, I.M. (1994). “Gender as seriality: thinking about women as a social collective”, Signs. 19(3): 713–738.

chapter 16

Race: A Contested and Travelling Concept Mette Andersson Introduction In a recent article Howard Winant (2015) asks ‘why our heads don’t explode under the pressures of cognitive dissonance’ regarding the present situation of race and racism. Whereas old versions of established racism and white supremacy have been officially discredited, he argues, racially informed action and social organization, racial identity and race-consciousness continue to influence contemporary life. In this situation, he continues, the very existence of race itself is often denied. His words speak directly to the situation in Scandinavian societies, which avoid race both in common parlance and as a sociological concept. In this chapter, I examine arguments for and against importing the race concept into the standard conceptual vocabulary of Scandinavian sociology. One central theme of the discussion will be the relationship between commonsense interpretations and scientific interpretations of the race concept. In the 1940s, a period dominated by increasing critique of the biological definition of race, Gunnar Myrdal, a Scandinavian social scientist commissioned to lead a major project on race in the United States, highlighted the difficult task faced by social science when attempting to impose its definition of racial groups onto the us public: The correct observation that the Negro is inferior (socially, economically, politically) was tied up to the correct belief that man belongs to the biological universe, and by twisting logic, the incorrect deduction was made that the inferiority is biological in nature. myrdal, 2009: 120

Myrdal argues that it is difficult for ordinary people to understand complex historical and sociological definitions and explanations, quoting another social scientist, Reuter, stating that ‘when color differences coincide with difference in cultural levels, then color becomes symbolic and each individual is automatically classified by the racial uniform he wears’ (Myrdal, 2009: 120).

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_017

Race

285

The difference between social science in the us in the 1940s and contemporary social science is that although researchers continue to view race as a social construct, the term itself is more disputed. Because of past negative uses of the concept of race and the lingering association with racial hierarchies in contemporary common parlance, sociologists debate whether the concept should be used at all. Certain scholars prefer to place the concept in quotation marks (‘race’) both to accentuate the social construction of race and to distinguish it from its earlier biological references. Others prefer not to use it at all, arguing that continuous scientific use of the race concept contributes to manifesting a classification with no valid reference to boundaries between human beings. Still others argue that we should continue to use it because ideas regarding racial difference—referring to bodily markers such as skin colour/ phenotype—contribute to creating and upholding social hierarchies. The latter authors argue that a concept such as ethnicity—denoting cultural heritage or social boundary construction—does not capture the force of the visual in creating exclusion and social hierarchies. Furthermore, the race concept evokes different associations and histories in different places and across time. In contemporary German sociology, for instance, the concept leads by association to the Holocaust and is avoided by most social scientists, who prefer to use the ethnicity concept to analyse social stratification and the position of immigrant minorities in social institutions and micro-interactions. In us sociology, the concept refers back to the institution of slavery and to the political struggles of African Americans in the 1950s and 1960s. There, as in the uk, race is still used as a category with which people are asked to identify in censuses and in this respect is sociologically relevant as an administrative and identity category (Rex, 2009; Saperstein and Penner, 2012). In France, the concept is avoided, primarily because of France’s assimilation policy that forbids the acknowledgement of group differences in the public space. French sociologists may refer to the concept of racism but seldom to race (Amiraux and Simon, 2006). In Norway and in the Scandinavian countries more generally, the concept prompts associations with the Nazi occupation during the Second World War. It also evokes memories of a shameful history of phrenology and racial hygiene (Skorgen, 2002; Andreassen, 2014; Garner, 2014) and the harsh treatment of groups such as the Sami, Travellers and Jews in earlier Scandinavian history. Although a few Scandinavian sociologists and anthropologists have started to discuss and analyse racism in their respective countries (Fangen, 1997; Gullestad, 2002; Keskinen et al., 2009; Svendsen, 2014; McIntosh, 2014; Bangstad and Døving, 2015), there have been few debates about whether to use the race concept in Scandinavian sociology. When the

286

Andersson

race concept has been used, it has typically been in recent work on intersectionality (in tandem with gender and class) (Orupabo, 2014; Berg, Flemmen and Gullikstad, 2010) and more indirectly in studies focusing on whiteness as a source of privilege (Andersson, 2007). One reason such discussion has not yet occurred is that in both Norway and Denmark (less so in Sweden) there has been a certain resistance to analyses of racism both within and outside academia. Racism is often viewed as belonging to a negative past and race to a history of biological racism long forgone. Race has therefore become a hidden concept, which in the sociological analysis of integration, exclusion and social stratification in regard to first- and second-generation immigrants has been replaced with the ethnicity concept. This chapter starts with a presentation of the general history of the concept of race in the natural sciences and sociology and continues with a more specific discussion of whether to use the concept. A range of alternative or substitute concepts is presented before moving to the last section of the chapter, in which I discuss whether we need the concept of race in Nordic sociology today. It is from the vanguard point of the last question—namely, how this concept intervenes in the Scandinavian context—that this chapter evolves.

A Genealogy of the Race Concept

To understand the debate regarding the concept of race in sociology, we must go back in time and examine how the concept was used and understood before sociologists’ 20th-century intervention in the debate. Most theorists on race and racism argue that the concept of race is a modern invention, first used in the English language in the 15th or 16th centuries (Jordan, 2009; Banton, 2009; Hirschman, 2004; Winant, 2000). When the concept was used at that time, it did not refer to ideas of intrinsically different human types. Although from the earliest trade relations between Europeans and Africans there had been speculation about why Africans were black-skinned and Europeans white-skinned, these speculations drew on other sources. When the English discussed why the Africans were black and the Europeans white in the 15th and 16th centuries, religion, tradition and literature were the primary sources of ‘evidence’ (Banton, 2009; Jordan, 2009). Speculations based on religious myths regarding Noah’s son, who was condemned to blackness because he beheld his father naked, and on conjectures about pigmentation caused by the strength of the sun were common. Also increasingly common were discussions of whether all humans had a common or different origins. However, to argue in favour of polygenesis in this period, implying that humans derived from different origins, would

Race

287

contradict the Bible and the story of Adam as the father of humanity, constituting heresy. With the advent of the natural sciences in the 18th century and the focus on categorization of biological diversity in animals and plants, human differences became a topic of natural scientific examination. Carl von Linné was among the first to categorize humans according to distinct biological traits, and others such as the French anatomist Cuvier followed in further developing such typologies. Cuvier distinguished between ‘Caucasian’, ‘Mongolian’ and ‘Ethiopian’ types, each of which contained further sub-categories (Banton, 2009: 59). The idea of inherently different racial types of a biological character, namely, racial typology, increasingly took hold in the natural sciences and spread to the cultural and social sciences in the 19th century. Whereas Cuvier at the end of the 18th century had believed in monogenesis but theorized that a great natural catastrophe had led to the development of different racial types, both polygenesis and monogenesis defenders used racial typology as an explanatory factor in the course of the 19th century (exemplified, for instance, in the work of De Gobinau (1856) and Knox (1850)). Moral and intellectual traits were increasingly viewed as dependent on biological inheritance, and a hierarchy of superior and inferior races was constructed. Although Darwin’s Origin of the species was published in 1859, it was well into the 20th century before his evolutionary theory replaced racial typology (Banton, 2009: 61). Whereas Darwin stated that the term race could be used for subspecies, the whole idea of permanence in nature was under attack in his theory of natural selection and primacy of adaptation to habitat. Banton argues that the reason racial typology remained dominant until the early 20th century was not only the difficulty involved in understanding Darwin’s complex theory but also the popularity of the racial hierarchy hypothesis in this specific historical period. Both European imperialism and American racial segregation required a theory to defend their treatment of non-white peoples, and the scientific idea of a permanent racial hierarchy was suitable for that purpose. In addition, the race concept was commonly used among the general population as a synonym for the concept of nation in the late 19th and early 20th centuries. Many scholars of racism also note that social Darwinism, that is, using Darwin’s theory to defend the thesis that certain races were more adaptable than others and thus reproducing the earlier idea of racial hierarchy, was responsible for the continued scientific support of racial typology and the development and popularity of phrenology and the eugenics movement (e.g., Hirschman, 2004). Although both natural and social scientists criticized racial typology in the 19th century, this critique became stronger in the early 20th century.

288

Andersson

­ nthropologists such as Franz Boas and Ruth Benedict and sociologists such as A W.E.B. Du Bois (1995) and Robert Park (2009) criticized the biological reference to the race concept and focused on social, historical and cultural ­explanations of why the social groups that were called races were inferior in material and symbolic ways. For these social scientists, races were understood as malleable social groups developing through migration, conflict and resource competition across history and across space, such as in the multi-ethnic cities of the us of that period. Thus, between the early 20th century and the 1950s, a ­social-constructionist definition of race started to compete with the biological definition. After the Second World War and the establishment of the United Nations, unesco organized three conferences for natural and social scientists, all of which concluded that the race concept no longer had scientific relevance in terms of explaining biological differences between human beings (Rex, 2009). Biologists argued that racial typology and references to race as subspecies should be dismissed and that the concept of population should replace the race concept. Monogenesis was re-established as scientific truth and interbreeding viewed as positive for the development of the human species. Social scientists upheld the biologists’ argument that the aggregates called races were social or cultural groups evolving through patterns of conquest, interaction, migration and conflict over time. Whereas most anthropologists in this period chose to dismiss the concept of race in favour of the concept of ethnicity, many sociologists in the us and in the uk continued to use the race concept that had come to be defined as a social-constructionist concept. One of their reasons for continuing to use a discredited concept was that social hierarchies and social interaction were still marked by racism and the exclusion of people of colour. Another reason was that minority groups themselves mobilized on the basis of their racial differences and social inferiority compared to majority populations. Therefore, the main argument for upholding the race concept was that although race had no scientific validity as a biological concept, people still ranked others according to phenotype associated with a hierarchy of cultural and moral values. Various definitions of racism ranging from racism being an ideology of racial hierarchy based on biology to broader definitions emphasizing institutional practices of exclusion and cultural hierarchies were developed, pointing both to different aspects of the social (structure or agency) and to various links between the concepts of race and racism. In the 1960s and 1970s, distinct race research paradigms were developed in both us and British sociology. Although these paradigms were largely influenced by Weberian and Marxist traditions focusing mainly on the macro- or meso-level of ­society,

Race

289

there were also strong empirical research programmes in the us based on methodological individualism and the rational choice perspective on action. The concept of race was still in use but defined and understood differently in the different programmes. In the 1980s, critiques of earlier race research both in Britain and the us centred on how 1960s and 1970s research on race had reproduced an essentialist perspective in which race was viewed as a stable social category in a manner similar to the biological race concept. When race was used without problematizing the political and historical references to the concept, lay people would easily believe that race was a fixed social category and not a category that through history had been the continuous object of political and scientific struggle. The contingent character of the concept and power struggles between states and social movements became more central. Certain theorists started to place the race concept in quotes in this period as a way of reminding the audience that the concept had no real social or biological reference. From the 1990s until the present, theorizing on race has been more attentive to globalization and transnational migration, and inspirations from ­post-structuralism, post-colonialism and various post-modern approaches have been influential. One example of this is intersectionality research, which focuses on how several dimensions of identity or difference influence each other in patterns of interaction and inequality. Whiteness studies and critical race studies are other traditions that have developed since the 1990s. In addition, although there have long been comparative studies showing how racial histories and racial classifications distinguish countries such as the us, South Africa and Brazil, comparisons among different regional histories and patterns of racism and the relationships between them appear to be more frequent (Goldberg, 2006, 2009; Saperstein and Penner, 2012; Garner, 2014). The purpose of this cursory review of the history of the race concept has been to focus on the most important tendencies and to provide background for the discussion below. As Howard Winant (2015) claims, over time the race concept has moved from being a socio-political construct to a biological construct back to a socio-political construct. Race is still understood as a theoretical concept referring to bodily markers of difference, but the idea of biology determining intellect and culture has been definitively discarded. The race concept is used in various ways in different theoretical and empirical research traditions. There is no agreed-upon definition of the concept in sociology, and there is not, as in gender studies, a parallel concept referring to a biological distinction. Although the relationship between the concepts of gender and sex is certainly debated in gender studies, one cannot dispute the fact that only women give birth to children and that men’s and women’s bodies in g­ eneral

290

Andersson

are differently constituted. The sociological race concept has no similar biological reference. The body is certainly in focus, but it is phenotype and colour (that is, the bodily exterior), not bodily function, intelligence or morality that are discussed. Thus, and specifically regarding how race is viewed both as a relational category referring to other races and as a category that can function as a basis for political mobilization among people belonging to less powerful groups, the concept resembles the sociological concept of class. Race is a general theoretical concept with no definite empirical references, a concept that, like class, is debated from the perspectives of various research traditions. Nevertheless, there has been less debate about how to theorize and operationalize race in quantitative empirical research (Martin and Yeung, 2003; Hirschman, 2004) than in quantitative class studies and more debate about whether the concept should be used at all, and if not, which concepts can be used instead.

Conceptual Space

Discussions of the destiny of the race concept in sociology are marked by the histories of immigration and racism evoked by the concept in different countries. In the British context of the 1980s and 1990s, Robert Miles stands out as the main critic of the continued use of the race concept. His argument, like others before him, is that we must stop using a concept that has no scientific value, is discredited, and contributes to reproducing false ideas among the general population. Because races do not exist, we cannot continue to speak of race relations or racial groups (Miles, 2009). Miles argues that we should focus instead on the process of ‘racialization’, a concept inspired by Franz Fanon (1986) and later developed within sociology by Michael Banton in the late 1970s (Murji and Solomos, 2005). Banton viewed racialization as a way of naming the historical process of categorization by which people and nations became viewed as races (ibid: 9). Miles, however, offers a more general definition of racialization as those instances where social relations between people have been structured by the signification of human biological characteristics in such a way as to define and construct differentiated social collectivities. murji and solomos, 2005: 11

To Miles, this implies interior or exterior references to biology and sensitivity to the shifting practice of which groups have been racialized and the references that have been used to do so.

Race

291

The critique of the race concept and the suggestion of new concepts such as racialization naturally lead to new conceptual debates and problems. In this section, concepts parallel to and overlapping with the race concept will be briefly presented. Because the use of the concept of racialization has increased since the 1990s, and because it is used and defined in various ways (as categorization, discourse, practice of exclusion, etc.), certain scholars have argued that this concept has become useless for sociological analysis (Goldberg, 2006). Others have argued that the concept of racialization continues to rely on the concept of race (Anthias, 1992; Mason, 1994) and that it would be better to maintain the race concept provided it is defined in a specific way. In response to Miles’ argument that race exists because of racism, Mason (1994) responds that this is the linkage justifying race as a sociological concept and that race and racism may be viewed as mutually constitutive. The debate on how to define racism has also influenced discussions of the race concept. Beginning in the early 1980s, allegedly new theoretical perspectives on racism that extend references to the concept from biological markers to include cultural markers have been advanced under headers such as ‘new racism’, ‘differentialist racism’ and ‘culture racism’. One rationale for these concepts is the observation that very few political actors who pursue exclusionary politics against immigrants or ethnic and religious minorities refer to race, instead citing cultural or religious differences as arguments in public discourse or political programmes. As in the 1980s, both empirical research and theorizing in the field began to focus on the position of immigrants and ethnic and religious minorities more generally: researchers have argued that we need either a wider definition of racism or an approach to racism acknowledging that there are several forms of racisms and that racism must be studied in specific empirical contexts (Solomos and Back, 1996). However, arguments about the new-ness of cultural racism are often disputed among historically minded sociologists, who state that in the heyday of biological racism, references to cultural difference and hierarchy also flourished (Gilroy, 2000: 299). Along with the increasing focus on migration and immigrants in European sociological research, the concept of ethnicity has become the central concept addressing human difference. Whereas Max Weber (1968: 389) offered a definition of ethnic groups as based on a subjective belief in common descent already in the 1920s, it is anthropologists who subsequently and until recently have been the most occupied with definitional discussions of ethnic groups and ethnicity. There are several competing definitions of ethnicity in social anthropology (Eriksen, 1993; Jenkins, 1994; Fenton, 2010). Central elements of these definitions are usually that ethnicity involves ideas of common descent or boundary construction between in-group and out-groups. However, certain

292

Andersson

subtleties of the anthropological debate, which to a large degree is based on empirical research on multi-ethnic countries in the South and the East, are easily lost in the translation to sociological empirical research on migrant minority groups. When categorizing immigrant minorities in Western Europe, the concept of ethnic group has often been used as a synonym for national groups, omitting the often-complex patterns of ethnic distinction and social inequality in the countries from which many immigrants originate. In the migration research context, ethnic groups thus tend to be understood as minority groups from different nation-states, and ethnic relations tend to be understood as relations between ethnic majority and minority groups of immigrant descent. Whereas the ethnicity concept has become standard in migration and integration research, for instance, in analyses of everyday interaction, social mobility, discrimination and marginalization, the boundaries of this concept in regard to the sociological race concept are often unclear. Many textbooks and summary articles on ethnicity or race note these concepts’ overlapping dimensions (Knowles, 2003; Back and Solomos, 2009; Brubaker, 2009; Fenton, 2010; Guibernau and Rex, 2010; Wimmer, 2015). In definitions of ethnicity, colour or phenotype is not a basic defining criterion, although certain definitions (see Jenkins, 1994) note the strong effect of the ascription of ethnic identity from the outside. It is precisely this latter idea that underlines the power dimension inherent in the ascription of difference by others, visible in social interaction and mediated in public debate and politics, that most resembles the dimension captured by a sociological race concept. Many researchers have noted the similarity of identity ascription in politics and public debate with regard to race and ethnicity, and some have developed concepts such as ‘ethnification’ to accentuate similar processes to those captured by the concept of racialization. Here, the main focus is on how ethnicity is constructed as an effect of discourse and how such discourse tends to create aggregates of people as more alike and sharing more traits than they do in reality. In macro-theories of racial formation and racialization, racial classification and inequality are typically viewed as end-products of processes that originally had other explanations, such as the capitalist exploitation of freed slaves or immigrant workers in the labour market or white nationalism or the long-term effects of the history of slavery and segregation, for instance. Macro-theories of racialization focus on materialist or cultural explanations of why social relations and hierarchies become racialized. Here, racial or ethnic hierarchies are considered effects of discourse, politics, or socioeconomic inequality, typically with an eye for how processes of racialization vary across time and space. Micro-theories of race and racialization tend to focus on how social interaction and identity become imprinted with ideas of racial meaning, difference

293

Race

and marking. Based on perspectives from symbolic interactionism, pragmatism, phenomenology and psychoanalysis, such theories focus on how difficult it is for individuals to avoid racial ascription and on how racial categorization leads to different reactions, for instance, through internalization of the idea of racial inferiority and shame among those exposed to it. The concept of ‘double consciousness’ coined by W.E.B. Du Bois in the late 19th century (Du Bois, 1995) refers to how people exposed to double standards develop a reflexivity in which they perceive themselves both as national citizens and as inferior minorities. Franz Fanon continued this tradition, accentuating how racist images, imperialism and colonialism made it difficult to escape one’s blackness (Fanon, 1986). Within this tradition one also finds work illustrating how race acquires meaning as a category of solidarity, organization and political mobilization. Theories that view racial background as a main explanation for social inequality or social mobility are rare in current sociology. In their article on the use of the race concept in mainstream empirical American sociology from 1937 to 1999, however, Martin and Yeung (2003) find an increasing likelihood that sociologists will take race into account as an explanatory factor without problematizing the concept itself. They argue not only that researchers seldom critically discuss what the race concept refers to but also that the increase in the use of race as a control variable largely relates to advances in methodological techniques that enable more variables to be considered. Such use of race as a control variable in quantitative research recalls how ethnicity is used similarly in quantitative studies of social stratification and inequality in diverse societies. Martin and Yeung’s study therefore draws attention to how empirical studies often employ more conventional and essentialist ideas of what race and ethnicity are than the ideas referenced in theoretical discussions of those same concepts.

To Use or Not to Use in the Scandinavian Context

In this section I will address possible benefits and drawbacks of introducing the race concept into Scandinavian sociology. The dilemma in Scandinavia, as elsewhere, is whether it is possible to use a theoretical concept of race in sociological analysis without reproducing the idea that there are discrete races or racial groups. This question is a specific version of the more general theme of the relationship between sociological concepts and common-sense concepts. To address this relationship, we must address not only the frequency of reference to and adaption of scientific concepts in a general population but also how scientific concepts interact with collective representations of history and

294

Andersson

national identity. The theoretical definitions of concepts, the reality-references of such definitions and the ethical and political challenges resulting from the diffusion of sociological concepts are important and interrelated themes here. I will start with the problems of definition and reality-reference and continue with the political and ethical challenges resulting from diffusion to the general public. In theories informed by post-structuralism, phenomenology or pragmatism, race is defined either as an effect of history and power struggles or as a cultural and material boundary produced and reproduced through politics, societal discourse and interaction. The main focus in poststructuralist theories is how objects, ideas and people become ‘raced’ through politics and discourse. Operationalization of the race concept into indicators resembling human aggregates or groups in the real world is not deemed a problem for the relevance of theory and analysis. Instead, race is perceived to exist because it is implicitly or explicitly referenced in discourse and politics or because social hierarchies are marked by racial attributes. Micro- and meso-theories inspired by traditions such as phenomenology, symbolic interactionism and pragmatism offer similar arguments in regard to the reality-reference. Here, the realness of the category of race is typically viewed as a variant of the Thomas-theorem postulating that because people perceive race as real, it becomes real in its consequences. Researchers working from social-constructionist perspectives (which constitute the bulk of perspectives on race and racialization in sociology today) may believe they are escaping from the reality-reference of race and racial groups. Research based on epistemological traditions other than social constructionism, such as certain versions of realism or deductive research aiming to test hypotheses in which race is used as a control variable, is confronted by tougher challenges. In deductive research, for instance, indicators of race as a control variable must be operationalized in ways that more closely relate to the empirical world of human diversity. Although social-constructionist research may escape this specific reality-reference problem, researchers must still relate to how laypeople unaware of the theoretical paradigms in social science interpret their use of a discredited concept. Another challenge is that because race here is defined rather vaguely—i.e., as an effect of discourse, history or social interaction—the concept itself can expand to include almost everything that can be understood as referring to race in a cultural or material way. This is also a major challenge in regard to the concept of racialization, which is sometimes used in such a broad way that almost nothing escapes the racialization process. The political and ethical challenges stemming from using the race concept relate both to the specific histories and to the later construction of these histories in the Scandinavian countries. They also relate to how Scandinavian

Race

295

citizens today may react when they read about themselves as ‘raced’ or as members of racial groups. As mentioned in the introduction, phrenology, eugenics and social Darwinism were influential movements in Scandinavia and constitute dark spots in a history from which today’s Scandinavian political authorities want to distance themselves. In addition, repeated attention to and memory of the Nazi occupation of Denmark and Norway in the Second World War remind people of the Holocaust, in which Jews were viewed as a separate race that would pollute the racial hierarchy (Bauman, 1989). These contemporary historical references bear on the biological race concept and the atrocities following the political implementation of biological racism. One major difference between the situation in Scandinavia and the situation in the us and the uk is that race in the latter countries has continued to be used as a concept with which citizens are asked to identify in censuses. In the us and the uk, the concept of race is naturalized in the sense that it refers to broader spatial histories that continue to inform not only political and social debate but also research on difference and social hierarchies. Another difference in Scandinavia is that in both the us and the uk, race has been adopted as a category for resistance movements among minorities themselves. Stuart Hall (2009) reflects, for instance, on how the category of black (then adopted from the us by British anti-racists) had no meaning for him as an immigrant from Jamaica, where a fine-tuned colour hierarchy interacted with other status categories. He describes how he had to ‘learn to be black’ and how people of different backgrounds could unite when black was interpreted as a political category in the uk in the 1970s and 1980s. In Scandinavia, similar minority histories exist among native minority groups such as the Sami, but they have not been dominant during the political mobilization among non-white immigrant minorities arriving after the 1960s. Although the Scandinavian countries were also involved in colonialism and the slave trade, albeit on a lesser scale, immigrants did not bring memories of Scandinavian colonial rule when arriving. Nevertheless, visible minorities living in Scandinavia today often refer to world history and to earlier social struggles among non-white minorities as a means of discussing and understanding their own situation as minorities in Scandinavia (Andersson, Jacobsen, Rogstad and Vestel, 2012). Furthermore, both qualitative and quantitative research shows high frequencies of ­self-reported experiences of discrimination and racism among non-white Scandinavians (Løwe, 2008; Prieur, 2010; Sandset, 2012). Research also shows that people with backgrounds from different nation-states may identify with each other as non-white minorities vis-à-vis the white majority and that race is loaded with meaning in social institutions such as sport, in which athletes may view their superiority and inferiority in racial terms (Andersson, 2007).

296

Andersson

One central theme here is therefore the potential mismatch between the spatial scale through which Scandinavian nation-states interpret their history and the spatial scale adopted by contemporary minorities within Scandinavia to interpret their situation. A Norwegian public debate about the word Negro in the late 1990s showed, for instance, that many, including a professor of Norwegian, viewed the word as innocent in the Norwegian setting. These arguments were based on a sharp distinction between Norway and the rest of the world. The minority activists who brought the debate to the media, however, viewed Norway as part of a broader spatial and temporal landscape in which more often than not, references to the word Negro have been negatively ­loaded. Among the benefits of introducing the concept of race to Scandinavian sociology would be to widen the spatial and temporal lens, drawing attention to Scandinavia’s place in a broader world in which race across the centuries was an effective social and political category influencing how non-white people were viewed and treated. An introduction of the race concept to Scandinavian sociology could therefore challenge methodological nationalism in sociology. If societies are viewed only as synonymous with nation-states, attention to nationalism and to broader spatial and historical patterns influencing social interaction and culture are often left out of sociological analysis (Wimmer and Schiller, 2003). The most important benefit of using the concept of race would be to more precisely examine categorization and exclusion based on colour or phenotype. As noted in earlier sections of this chapter, the concept of ethnicity only implicitly and in certain instances refers to bodily markers of difference. For instance, when we study social interaction or political mobilization among people with one parent from a sub-Saharan African country and one white Scandinavian parent (mixed-race people), the ethnicity category is not the most obvious choice for analysis. The same applies when we study non-white adopted young people’s identity work. Although ideas of common descent/ history and practices of boundary-making between groups may be salient factors in both of these cases, skin colour and phenotype matter more for the ascription of difference by others and for self-identification in these cases. Conclusion One can argue, as many do in the British context, that the silencing of race would make it difficult to formulate racism (Mason, 1994; Smaje, 1997; Harries, 2014). In Scandinavia, the sociological analysis of racism is in its early days, and the concept of discrimination is still the preferred concept in studies of

Race

297

exclusion of non-white minorities. However, the argument that the silencing of race makes it difficult to formulate racism is also an important argument in the Scandinavian context. This argument is only partly addressed by researchers who argue against the race concept by pointing to various other reasons for discrimination, such as prejudice or institutionalized practices of differentiation in regard to religion, culture and/or ethnicity. As argued above, there are several potential benefits of using the race concept in Scandinavian sociology. The most obvious benefit is that race, in contrast to ethnicity, points to bodily markers of difference. A second benefit is that through a focus on race as a difference category we draw attention to how the present is informed by a world history in which the ordering of human value was informed by phenotype and genotype. A third benefit is that we acknowledge that non-white people in Scandinavia often identify across ethnic and cultural boundaries as non-white in contrast to the white majority populations. If we had a concept that accentuated the centrality of visible bodily difference and related ideas that was less infected by its earlier innate biological references, it would have been easier to argue that such a concept should be used more extensively in Scandinavian sociology. Although social science should develop theory based on how contemporary peoples and societies construct ideas of difference and inequality (and here visible difference obviously matters), it is difficult to sidestep the former history of racial typology and social Darwinism. This argument is more prevalent in Scandinavia than in countries such as the us and the uk, where the concept is still used in censuses and where the shifting definitions of race from socio-political to biological and back to socio-political are part of the sociological discussion and are shared by many citizens. In Scandinavia, use of the race concept diminished in the wake of the Nazi occupation of Norway and Denmark and public awareness of state racism against indigenous populations. A contemporary re-introduction of race as a sociological concept might contribute to re-solidify the idea that race depicts distinct social groups—or worse, distinct racial groups—in which outer biological traits correspond to inner qualities, such as moral character and intelligence. If we argue that the race concept should be avoided in Scandinavian sociology, we would therefore require other concepts pointing to phenotype and colour as effective distinction categories. The concept of a ‘visible minority’ is one such concept that is increasingly used in research on migration and integration. This concept is less rooted in the past than the race concept. It refers to an abstract category of individuals or to the relationship between majority and minorities and not to ideas and objects related to a history in which race was a major distinguishing category of human diversity. It is more

298

Andersson

of a technical concept that is useful in studies of contemporary social interaction and discrimination. Whiteness theories focusing on the unmarked position of white people are another option. However, as several authors have argued, whiteness theories viewing whiteness as a locus of power and as the normative position of privilege for judging others may also fall into the trap of objectifying both white and non-white groups (see, for instance, Ware and Back, 2002). The concept of racialization, a concept that was supposed to correct the potential essentialism of the race concept, is another option. In Robert Miles’ general definition, the concept points to a process in which social relations between people have been structured by the signification of human biological characteristics in such a way as to define and construct differentiated social collectives (Murji and Solomos, 2005: 9). The question is whether this matches the current situation in Scandinavia and the degree to which social relations between minority groups of immigrant background and white majorities are structured more by the signification of human biological characteristics than by the signification of difference in terms of culture, nation, ethnicity, class, gender and religion. Empirical studies would probably show that this depends on the domains, arenas and interactions of investigation and that in certain arenas neither of these distinction categories matter for social interaction or inequality. Intersectionality theory, which looks at how hierarchies of gender, race and class mutually interact and produce processes of inclusion and exclusion, is yet another option. In this theory, which grew out of black feminists’ critique of white feminism in the late 1970s, the concept of race remains intact, but it is argued that race always works with and through other categories such as class and gender. This perspective may be helpful in empirical studies, but the theoretical discussion regarding intersectionality is unhelpful for examining the specific question of the relationship between social science concepts and lay concepts. The discussion of how to capture processes signified by the race concept elsewhere will undoubtedly continue in Scandinavian sociology. One important criterion for this discussion is that the Scandinavian countries should not be viewed as cases apart from the rest of the world in terms of either history or the spatial scope of analysis. People will continue to use phenotype as a marker when categorizing each other; in all likelihood, phenotype and ethnicity will continue to mark positions in status hierarchies and identity work. Sociologists must develop methodologies and theories that interpret and explain such processes. Theoretical concepts always risk being contaminated, misinterpreted and misused when disseminated to people, institutions and organizations outside of academia. However, there is also a risk of misuse within the

299

Race

academy. Although no guarantees can be made, part of our work as researchers is to make every effort to reduce such misuse and misinterpretation. That is the main reason we should think twice before include the race concept in the Scandinavian sociological vocabulary. References Amiraux, V. and Simon, P. (2006). “There are no minorities here. Cultures of scholarship and public debate on immigrants and integration in France”, International Journal of Comparative Sociology. 47(3–4): 191–215. Andersson, M. (2007). “The relevance of the black Atlantic in contemporary sport. Racial imaginaries in Norway”, International Review for the Sociology of Sport. 42(1): 63–79. Andersson, M., Jacobsen, C.M., Rogstad, J. and Vestel, V. (2012). Kritiske hendelser—nye stemmer. Politisk engasjement og transnasjonal orientering i det nye Norge. [Critical events—new voices. Political commitment and transnational orientations in the new Norway]. Oslo: Universitetsforlaget. Andreassen, R. (2014). “Response: the Nordic discomfort with ‘race’”, Nordic Journal of Migration Research. 4(1): 42–44. doi: 10.2478/nmjr-2014-0004. Anthias, F. (1992). “Connecting ‘race’ and ethnic phenomena”, Sociology. 26(3): 421–438. Back, L. and Solomos, J. (2009). Theories of race and racism. A reader. London: Routledge. Bangstad, S. and Døving, C.A. (2015). Hva er rasisme? [What is racism?]. Oslo: Universitetsforlaget. Banton, M. (2009). “The idiom of race”, Eds. L. Back and J. Solomos. Theories of race and racism. A reader. London: Routledge. Bauman, Z. (1989). Modernity and the Holocaust. Cambridge: Polity Press. Berg, A.-J., Flemmen, A.B. and Gullikstad, B. (2010). Likestilte norskheter. Om kjønn og ­etnisitet. [Equalized Norwegianness. On gender and ethnicity]. Bergen: Fagbokforlaget. Brubaker, R. (2009). “Ethnicity, race, and nationalism”, Annual Review of Sociology. 35: 21–42. DuBois, W.E.B. (1995). The souls of black folk. 100th anniversary edition. New York: ­Signet Classic. Eriksen, T.H. (1993). Ethnicity and nationalism. Anthropological perspectives. London: Pluto Press. Fangen, K. (1997). “Rasisme i plural form. Paradigmer og begreper om rasisme”. [“Racism in plural form. Paradigms and concepts of racism”]. Oslo: Institutt for sosiologi og samfunnsgeografi UiO, Rapport Nr. 7 1997.

300

Andersson

Fanon, F. (1986). Black skin, white masks. London: Pluto Classics. Fenton, S. (2010). Ethnicity. Cambridge: Polity Press. Garner, S. (2014). “Injured nations, racializing states and repressed histories: making whiteness visible in the Nordic countries”, Social Identities. 20(6): 407–422. Gilroy, P. (2000). Between camps. Nations, cultures and the allure of race. London: Allan Lane. The Penguin Press. Gobinau, A.D. (1856). The moral and intellectual diversity of races, with particular ­reference to their specific influence in the civil and political history of mankind. Philadelphia: J.B. Lippencott & Co. Goldberg, D.T. (2006). “Racial europeanization”, Ethnic and Racial Studies. 29(2): 331–364. Goldberg, D.T. (2009). “Racial comparisons, relational racisms: some thoughts on method”, Ethnic and Racial Studies. 32(7): 1271–1282. Guibernau, M. and Rex, J. (2010). The ethnicity reader. Nationalism, multiculturalism and Migration. 2nd edition. Cambridge: Polity Press. Gullestad, M. (2002). Det norske sett med nye øyne. [The Norwegian seen with new eyes]. Oslo: Universitetsforlaget. Hall, S. (2009). “Old and new identities, old and new ethnicities”, Eds. L. Back. and J. Solomos. Theories of race and racism. A reader. London: Routledge. Harries, B. (2014). “We need to talk about race”, Sociology. 48(6): 1107–1122. Hirschman, C. (2004). “The origins and demise of the concept of race”, Population and Development Review. 30(3): 385–415. Jenkins, R. (1994). “Rethinking ethnicity: identity, categorization and power”, Ethnic and Racial Studies. 17(2): 197–223. Jordan, W. (2009). “First impressions”, Eds. L. Back. and J. Solomos. Theories of race and racism. A reader. London: Routledge. Keskinen, S. et al. (eds.). (2009). Complying with colonialism. Gender, race and ethnicity in the Nordic region. Surrey: Ashgate. Knowles, C. (2003). Race and social analysis. London: Sage. Knox, R. (1850). The races of men: A fragment. Philadelphia: Lea and Blanchard. Løwe, T. (2008). Levekår blant unge med innvandrerbakgrunn. [Living conditions among immigrant youth]. Oslo: Statistisk Sentralbyrå, Rapport 14/2008. Martin, J.L. and Yeung, K.-T. (2003). “The use of the conceptual category of race in American sociology, 1937–99”, Sociological Forum. 18(4): 521–543. Mason, D. (1994). “On the dangers of disconnecting race and racism”, Sociology. 28(4): 845–858. McIntosh, L. (2014). “Impossible presence: race, nation and the cultural politics of being Norwegian”, Ethnic and Racial Studies. 38(2): 309–325. Miles, R. (2009). “Apropos the idea of race …. again”, Eds. L. Back. and J. Solomos. Theories of race and racism. A reader. London: Routledge.

Race

301

Murji, K. and Solomos, J. (2005). Racialization. Studies in theory and practice. Oxford: Oxford University Press. Myrdal, G. (2009). “Racial beliefs in America”, Eds. L. Back. and J. Solomos. Theories of race and racism. A reader. London: Routledge. Orupabo, J. (2014). “Interseksjonalitet i praksis. Utfordringer med å anvende et interseksjonalitetsperspektiv i empirisk forsking”, Sosiologisk Tidsskrift. [“Intersectionality in practice. Challenges in employing an intersectionality perspective in empirical research”, Journal of Sociology]. 22(4): 329–351. Park, R.E. (2009). “The nature of race relations”, Eds. L. Back. and J. Solomos. Theories of race and racism. A reader. London: Routledge. Prieur, A. (2010). “Fargens betydning. Om rasisme og konstruksjon av etniske identiteter”, Sosiologi i Dag. [“The meaning of color. On racism and the construction of ethnic identities”, Sociology Today]. 40(3): 69–91. Rex, J. (2009). “Race relations in sociological theory”, Eds. L. Back. and J. Solomos. Theories of race and racism. A reader. London: Routledge. Sandset, T. (2012). Color as matter. Narratives of race, ethnicity, and the deployment of color. (Master Thesis, Diss., University of Oslo). Saperstein, A. and Penner, A.M. (2012). “Racial fluidity and inequality in the United States”, American Journal of Sociology. 118(3): 676–727. Skorgen, T. (2002). Rasenes oppfinnelse. Rasetenkningens historie. [The origin of race. The history of race-thinking]. Oslo: Spartacus. Smaje, C. (1997). “Not just a social construct: theorising race and ethnicity”, Sociology. 31(2): 307–327. Solomos, J. and Back. L. (1996). Racism and society. London: MacMillan Press ltd. Svendsen, S.H.B. (2014). “Learning racism in the absence of ‘race’”, European Journal of Women’s Studies. 21(1): 9–24. Ware, V. and Back, L. (2002). Out of whiteness. Color, politics, and culture. Chicago: University of Chicago Press. Weber, M. (1968). Economy and society. Vol. 1. New York: Bedminster Press. Wimmer, A. (2015). “Race-centrism: a critique and a research agenda”, Ethnic and Racial Studies. 38(13): 2186–2205. Wimmer, A. and Schiller, N.G. (2003). “Methodological nationalism, the social sciences, and the study of migration: an essay in historical epistemology”, International Migration Review. 37(3): 576–610. Winant, H. (2000). “Race and race theory”, Annual Review Sociology. 2000(26): 169–185. Winant, H. (2015). “The dark matter: race and racism in the 21th century”, Critical Sociology. 41(2): 313–324. doi: 10.1177/0896920513501353.

chapter 17

Recognition: Conceptualization and Context Antje Gimmler Introduction The concept of recognition has had a remarkable and successful history ever since the philosopher G.W.F. Hegel introduced it in his Jenaer writings as a basic principle of social integration and subsequently discussed it in his famous Phenomenology of spirit as the main principle for achieving s­ elf-consciousness.1 Whereas recognition also appeared in pragmatism (e.g., G.H. Mead) and in Alexandre Kojève’s Marxist-Heideggerian philosophy of history, it was not until the 1980s and 1990s that it became a genuine master concept of social theorizing used in different theoretical traditions. The most recent prominent conceptualization of recognition as a general social theory can be found in Axel Honneth’s theory of recognition. Other theories of recognition include Charles Taylor’s well-known political theory of recognition and the sociological approach to respect by Richard Sennett. Whereas Taylor and Honneth view recognition as a basic human need and therefore favour a normative and anthropological theory design, Sennett uses as a starting point the notion of respect that forms a distinguished social gesture against the background of the inherent inequalities governing modern societies. Certainly, critical stances towards recognition as a master concept of social philosophy are also found. Nancy Fraser’s critique of Honneth’s theory of recognition is a case in point. The historical and political context is also an interesting dimension in understanding the concept of recognition. Recognition became prominent precisely when Marxism appeared to be defeated and the ideological clashes between capitalism and communism no longer resonated in passionate intellectual or real struggles. In this article, I shall examine the cognitive, heuristic and theoretical functions of the concept of recognition. To evaluate both the explanatory power and the limitations of a sociological concept, the theory construction must be analysed and its actual productivity for sociological theory must be evaluated. In the first section, I will introduce the concept of recognition as a travelling 1 Hegel was not the first to use the notion of recognition; in his philosophy of law, the philosopher J.G. Fichte was the first to do so.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004314207_018

Recognition

303

concept playing a role both on the intellectual stage and in real life. In the second section, I will concentrate on the presentation of Honneth’s theory of recognition, emphasizing the construction of the concept and its explanatory power. Finally, I will discuss Honneth’s concept in relation to the critique that has been raised, addressing the debate between Honneth and Fraser. In a short conclusion, I will return to the question of the explanatory power of the concept of recognition.

Recognition as a Travelling Concept

As mentioned above, the concept of recognition became prominent after being conceptualized by Axel Honneth, who understood the introduction of a theory of recognition as a renewal of Critical Theory. The theory of recognition combines a social and moral theory with a formal anthropological theory, analysing and explaining social reality from a normatively justified point of view. From Honneth’s perspective, this is a timely and consequent renewal of the normative spirit of Critical Theory. In his view, the concept of recognition in Critical Theory perfectly addresses the task of the critical project, which can be described as the justification of its critical investigation of society using a normative concept rooted in social reality. The concept of recognition thus enables the diagnosis and critique of the so-called social pathologies. Critical Theory has adopted, so to speak, the concept of recognition, although it should also be mentioned that there are significant differences among different theorists within Critical Theory (e.g., Habermas, 1993; Fraser, 2003) regarding the outlook on and the role played by recognition in Critical Theory. Today, Critical Theory can be characterized as being a heterogeneous intellectual and academic movement with an overlapping consensus regarding the linkage ­between normative theory and social reality (see Cooke, 2000). Although ­Honneth himself has recently developed an even broader and more comprehensive theory of modern societies in his book ‘Freedom’s right’ (2014), both the concept of recognition and the identification of social pathologies continue to play a decisive role in his theory. Another reason for the success of the concept of recognition can be found in the debate about multiculturalism. Here, the approach to a politics of recognition has been developed by Charles Taylor (1992), who has become most influential and extensively discussed. The neo-Aristotelian philosopher Taylor highlights the modern dimension of the concept of recognition. Recognition is an intrinsically modern concept because it presupposes individuality as a value. From Taylor’s perspective, recognition does not merely replace rival

304

Gimmler

concepts such as rank, status or honour; instead, it indicates a fundamental change in modern societies regarding how individual identity and social cohesion can be achieved. Recognition, Taylor states, is a fundamental human need and is realized in collectives and groups that form the horizon of individual identity. Multiculturalism, then, creates severe problems for this understanding of identity, and in Taylor’s approach the politics of recognition should safeguard spheres of collective identity building. Because of the primacy of the cultural identity of collectives, Taylor’s approach has often been criticized for being culturally conservative and failing to address the tensions and contradictions of multiculturalist societies (e.g., Benhabib, 2002). Both Taylor’s and Honneth’s theories of recognition have elicited critical reactions. The discussion between Taylor and Habermas (Taylor, 1992) and between Honneth and Nancy Fraser (Fraser and Honneth, 2003) are cases in point. Recently, a new topic of discussion has been initiated with the debate between Honneth and the philosopher Jacques Rancière (Genel and Deranty, 2016). Again, the notion of recognition is central, although Rancière introduces a way of circumventing and denouncing the normativity of the concept of recognition, which is a crucial point for both Honneth and Critical Theory. ­Numerous critiques have been levelled against the concept of recognition, questioning either the fruitfulness of the concept, the societal presupposition that it involves or the problematic application and ideological use of recognition as a political instrument (e.g., McNay, 2008; Patchen, 2003; Oliver, 2001; Zurn, 2005). A different thread of argumentation regarding recognition can be identified in the work of the sociologist Richard Sennett (2003), whose book Respect in a world of inequality employs the concept of respect in his analysis of how inequality penetrates and taints social relations. Surprisingly, he never refers to Honneth’s theory of recognition. This omission of Honneth’s influential theory may stem from Sennett’s goal of abstaining from a normative theory of respect and instead focusing on actual practices of respect. Sennett explores respect as part of the social fabric. His investigation of different types of inequalities ranging from personal competences to social and political stratifications leads to different ways of showing respect. The concept of recognition not only has garnered remarkable attention on the intellectual scene but also plays a decisive role in practical policies and other professions. In Denmark, for instance, the concept of recognition has often been used both as a theoretical framing of the reform of the public service sector and as a model for new and fairer relations between the citizen or social client and various administrative systems and social institutions. No leaflet distributed by the public service sector, from district psychiatric services to the

Recognition

305

job centre, misses the opportunity to declare a commitment to the principles of recognition (see e.g., Personalestyrelsen, 2008). This may be related to the Scandinavian welfare state and the values of equality and participation, which are deeply entrenched in institutions and administrative systems in the Scandinavian countries (see Jørgensen, 2015). Leadership theories and courses use the concept of recognition (e.g., Haselbo and Lyndgaard, 2007); internationally, recognition has also entered the caring professions and social work (Hugman, 2005; Houston, 2009; Juul, 2009). Two Danish intellectuals criticize what they call the ‘recognition industry’ in a polemical book with the catchy title Krigeren, borgeren og taberen (The warrior, the bourgeois and the loser) (Thyssen and Dahl, 2006). The authors diagnose an unhealthy alliance between the Scandinavian welfare state and the concept of recognition. These are only a few examples of how recognition became a catchy phrase, both evoking positive connotations and causing mistrust related to ideological misuse. Based on the discussion thus far, it should be clear that the concept of recognition has been successfully adapted to different theoretical traditions. A  presentation of the concept of recognition should address both the construction of the concept and its consequences for social analysis. It appears that the ­explanatory power of recognition is most obvious for the understanding and explanation of social relations, social integration and social cohesion. However, a theory of recognition should also address systems of autonomous function such as law, the economy or political power. Another aspect of the explanatory power of a concept arises in the context of its application. The distinction between a theoretical concept and its application in particular contexts provides a more realistic assessment of its explanatory power; it also provides an understanding of its genealogy. In the Scandinavian welfare-state context, the function of the concept of recognition will differ from its function in the context of a liberal country such as the us. However, one might ask the following question: does the analytical distinction between a theoretical concept and practical application merely ignore the intertwinement between theory and reality? The concept of recognition has been critiqued based on this heuristic mistrust. Surprisingly, Nancy Fraser (Fraser and Naples, 2004) and Slavoj Zizek (2003) voice the same suspicion regarding the boom of recognition in the 1980s and 1990s. They tie the success of the concept of recognition to the downfall of communism, the rise of neoliberalism and the acceptance of a global capitalist world order. From this perspective, the concept of recognition flourishes precisely because it fits in with the new political and economic landscape. As Nancy Fraser phrases it, ‘In the fin de siècle context, the turn to recognition has dovetailed all too neatly with a hegemonic neoliberalism that wants nothing more than to repress all memory of social egalitarianism’

306

Gimmler

(Fraser and Naples, 2004: 1111). The question is whether the concept of recognition has become a replacement concept, substituting for the political fight by interests (with all its nasty implications of violence, injustices, discontent and structural inequality) with a much softer so-called ‘struggle for recognition’ that might not address structural inequalities.

Honneth’s Theory of Recognition

The concept of recognition that Honneth has developed over the past 30 years constitutes the centre of a comprehensive social theory that is both normative and empirically substantial. Based on the interpretation of Hegel’s notion of recognition in his early Jenaer works and making reference to the pragmatic understanding of the concept of intersubjectivity of G.H. Mead, Honneth aims at a social theory that analyses and singles out social pathologies while remaining empirically rooted. The theory of recognition thus intertwines different theoretical layers into a comprehensive social theory that combines moral theory with developmental psychology, philosophical anthropology, a theory of social relations and a historical reconstruction of the process of modernization. The basic idea underlying the concept of recognition can be summarized as follows: real experiences of neglect and disrespect along with the demand for recognition expressed by individuals and groups direct the social philosopher towards a normative ideal of social relations that enables successful self-realization and in turn enables criticism of societies’ social pathologies. The focus on the social pathologies that inhibit humans’ self-realization also involves a turn away from an abstract theory of justice (e.g., Rawls’ theory of justice). Below, I shall first examine Honneth’s relation to Critical Theory and the normativity claim that is central both to Critical Theory and to Honneth. Second, I shall present the theory of recognition conceptualized by Honneth, including the three distinct spheres of recognition. The theory of recognition was developed by Honneth as a contribution to social philosophy, following the tradition of Rousseau, Marx and the older Critical Theorists (Honneth, 1999). In this understanding, social philosophy is not distinct from social theory because both are interested in empirical backing and empirical application of the theory to diagnose and analyse social reality. The goal of the tradition of social philosophy in which Honneth situates himself has always been both to describe and to normatively evaluate social reality, or even further, to unfold the normative conditions for a good life and ­therefore for a just society. Hence, the theoretical construction of Honneth’s concept of recognition presents the empirical explanatory power and

Recognition

307

­ ormative ­foundation of social theories as being mutually dependent: from n the experience of neglect and disrespect that can be empirically identified in different layers of social relations follows the idealization of those social relations that would support undistorted social relations and the self-realization of human beings. In this conceptualization, identification of the moral grammar that constitutes the infrastructure of the social life-world is based upon the real experiences of neglect or unequal treatment. As Honneth outlines, ‘the recognition-theoretical turn’ (Honneth, 2003: 134) enables the critical philosopher to detect the ‘trace of an intramundane transcendence in the social culture of everyday life’ (Honneth, 1999: 323). In continuity with the tradition of Critical Theory, Honneth aims to reconstruct a theoretical framework that enables an understanding of social reality in terms of the idealization of inherent claims. As he states with reference to Hegel, ‘we should generalize our knowledge of the social preconditions of personal identity-formation into a concept that has the character of a theory of egalitarian ethical life (Sittlichkeit)’ (Honneth, 2003: 177). This intimate linkage of theory and practice, from actual experiences of misrecognition to an ideal of social relations, is not a direct one. Neither deductive argumentation nor inductive inference can lead to the justification of the inherent normativity of Critical Theory. Instead, the structure of the argument takes the form of a weak transcendental argument that identifies the conditions that enable human beings to live ‘the good life’. In Honneth’s case, negative experiences of disrespect indicate the legitimate expectations of individuals and conditions that ‘to the best of our knowledge’ (Honneth, 2003: 177) are necessary conditions for a just and good society. Here, a problem arises that Honneth and Fraser have debated extensively: What types of violations of recognition should form the starting point of the critical analysis? Whereas Fraser appears to favour the politics of identity and specific social movements as an empirical background, Honneth attempts, through his anthropological approach, to avoid what he calls the ‘short circuit between social movement and social discontent’ (Honneth, 2003: 128). However, the types of experiences that qualify to represent the inherent normative claims of social relations are also a crucial point in Honneth’s anthropological approach. The relationship between Honneth’s theory of recognition and the tradition of Critical Theory in general is a complex one (Basaure 2011). In several articles, Honneth himself has marked the continuities and differences of his position in relation to the older Critical Theory and to Habermas (e.g., 1987, 1989, 1999). I will concentrate on Honneth’s relationship with Habermas. At least three elements of Habermas’ reformulation of Critical Theory are decisive for Honneth’s conceptualization of recognition. First, the main achievement of Habermas’ work can be found in the intersubjective turn that he inaugurated

308

Gimmler

in The theory of communicative action. From Honnneth’s perspective, the turn from a philosophy of subjectivity that hypostatizes the subject as engaged in a purely instrumental relationship with an object to a philosophy of intersubjectivity allows a conceptualization of social relations that goes beyond instrumentality. Moreover, this approach includes the formulation of necessary conditions that constitute individuals’ social fabric and can function as a guide for a normative ideal. In Habermas’ version, these validity claims are constitutive for what he calls communicative action. Honneth transforms the normative ideal that (according to Habermas) could be found in undistorted communication into the much broader ideal of recognition. The third element adopted by Honneth from Habermas is the identification of so-called pathologies.2 Habermas had characterized the reactions to the colonization of the lifeworld as the ‘pathological de-formation of the communicative infrastructure of the lifeworld’ (Habermas, 1989: 375). The political and social struggles of the 1970s and 1980s mark the existence of life-world pathologies, identifying a new type of social indignation. They do not follow the traditional logic of class struggle but instead are an expression of a life-world under pressure stemming from the imperatives of systems. In line with Habermas and others (e.g., Ulrich Beck), Honneth clearly recognizes the expressions of resistance that occur under such pressure, e.g., the ecological or feminist movement, as new forms of social movements. As Honneth (1999) outlined in his inaugural lecture in Berlin, a renewed focus on identity is necessary to explain the full range of social pathologies in modern societies. With this theoretical move, Honneth departs from Habermas’ focus on communication as the social glue that holds society together. From Honneth’s perspective, Habermas’ theory is too thin because it focuses on only communication, interaction and the possible violation of the communicative presuppositions. Habermas’ theory cannot encompass the thickness of the social fabric with its different layers of individual experiences, identity claims and relations of mutual dependency. This is why Honneth advocates not only the intersubjective turn but also a recognition turn that consequently includes a focus on identity building and the social conditions of individuation. Honneth wants to provide a broad understanding of social relations in which bodily gestures, emotions and material symbols also play an important 2 The use of the medical terminology of ‘pathology’ and ‘diagnosis’ is interesting in itself (see Susan Sonntag’s essay on Illness as metaphor, 1978). Laitinen, Särkelä and Ikäheimo (2015) outline that with the use of the term pathology, Honneth wants to demarcate his own position from standard political and moral theories and particularly from other current theories of justice (e.g., Rawls (2003)).

Recognition

309

role in the analysis of society. His almost phenomenological description of the experience of ‘not being seen’ as a form of disrespect is a case in point, showing that the type of social relations that should be addressed by recognition theory function far below the cognitive threshold (Honneth, 2001). The theory of recognition is conceptualized by Honneth as the differentiation of three realms or spheres of recognition, following Hegel’s distinction in his Jenaer works. The distinction also resembles societal differentiation into the private, legal and civic spheres. All three realms are ideally governed by a relationship of mutual recognition, which is a necessary prerequisite both for human beings’ self-realization and for a just social order. Although recognition is the main principle that governs all three spheres, the theory of recognition encompasses ‘a plurality of three equal-ranking principles of social justice’ (Honneth, 2004: 358). The theory of recognition informs what counts as social justice: the principles of social justice are formed along the line of spheres of recognition. With the principle of love in the private sphere, equality in the legal sphere and social esteem in the civic sphere, relations of mutual recognition are identified that qualify as categories both to conceptualize the ideal of social justice and to identify social injustice. The first realm of recognition is the fundamental sphere of becoming an individual and thereby developing self-confidence. Honneth refers to psychoanalytic and psychological developmental theories (e.g., Donald Winnicott, Daniel Stern, Jessica Benjamin) to support his claim that mutual recognition is a necessary prerequisite for becoming a person. In an ontologically fundamental sense, what is at stake in this intimate sphere is a human being’s capacity to relate to others with confidence and trust. With respect to individuation, confidence appears only if a child’s needs and feelings are recognized. Although the infant is deeply dependent upon significant others, the relationship must transcend dependency and power relations. It must be a relationship of mutual recognition. In psychoanalytical terms, one can state that the strength of the Ego is achieved only in reciprocal relations of recognition. In addition, friendship and love relationships are part of this sphere of intimate relations. The emotional ties involved in such intimate relations are based upon the normative expectation of being recognized, not because of a certain interest, strategic goal or other selfish motivation, but because of one person’s intrinsic value to another. From a historical perspective, Honneth mentions ‘the emergence of the “bourgeois” love-marriage’ (Honneth, 2003: 139) as an example of the institutionalization of the principle of love. A lack of mutual recognition in the intimate sphere causes impaired identity, psychological disorders and a lack of self-confidence. This type of impairment has consequences, not only for how an individual is able to execute his or her life plans or even to project his or

310

Gimmler

her needs towards future realization but also for how an individual uses the other spheres of recognition. It is no surprise that theorists of social work are attracted to the concept of recognition because it allows them to link a client’s personality and socialization to an actual social problem. In the second realm, an individual’s autonomy is recognized as his or her formal capacity for autonomous moral action. The person is subject to the law in the sense that the person has rights and liberties. Individuals are recognized both as equal and as authors of their own will. Self-respect is Honneth’s term for the outcome of this form of institutionalized recognition, which is mirrored in the moral authority of the person. Again, from a historical perspective, recognition in the shape of equality and being a citizen with rights and liberties has been institutionalized in modern law, constitutions and human rights. In an almost Weberian manner, Honneth notes that the establishment of legal equality depended upon a differentiation process in which a person’s legal and social statuses were separated (Honneth, 2003: 140). Legal equality is independent of social esteem, class, rank and prestige. A person’s autonomy encompasses not only the execution of his or her rights but also the liberty to express his or her wishes for and understandings of a good life. As in the first sphere, the lack of recognition of a person’s legal equality can be found in the disrespect of that person’s equal rights. Struggles for citizenship rights, e.g., the feminist movement or the civil rights movement in the us, bear witness not only to experiences of injustice and disrespect but also to the insistence on being treated as equals. In the third realm of recognition, the civic sphere, human beings achieve social self-esteem by contributing as members of a group. Here, Honneth ­refers to the fact that we are recognized as members of a collective for our contributions. Self-esteem in the civic sphere is again closely connected to selfrealization. Individuals’ self-realization depends on the evaluation of others. Self-esteem cannot be achieved by merely doing something in a self-sufficient way. Attributing value to each other’s achievements is the form of recognition exchanged between the members of a collective. However, norms with regard to what carries sufficient value to receive social esteem are neither stable nor objective. Although different types of groups and collectives form the civic sphere (e.g., non-governmental organizations (ngos) or sports clubs), Honneth often identifies the pre-political civic sphere with work relations. In this case, too, Honneth (2003: 141 ff, 2010a) reconstructs the principle of mutual recognition as present and identifiable, even in the capitalistic work order. This is indeed a difficult endeavour, and Honneth must admit that from its very beginning the achievement principle, although a product of the emancipation of achievement from a person’s rank and status, was dominated by the capitalist

Recognition

311

organization of work and income. The achievements of particular groups— e.g., women in their typical professions, non-professional blue-collar workers, temporary workers, the unemployed or those in precarious positions—to name but a few, are systematically devalued in the capitalist work order. Honneth argues that the struggles for the equal payment of women or the working class’s struggle for safe working conditions and fair payment were never pure struggles of interest; they were instigated by the experience of being treated unjustly. Honneth’s concept of recognition is normative in the sense that it presupposes a standard for the evaluation of society in terms of social pathologies and injustices. This standard or normative core of the theory of recognition must be justified and linked with actual experiences, and the linkage must be sufficiently general to identify all types of experiences of injustice. Therefore, Honneth insists upon a formal philosophical-anthropological theory that can identify conditions of human self-realization. Concurrently, the linkage must be sufficiently precise both to identify historical developments of recognition struggles and to distinguish legitimate demands for recognition from illegitimate ones. What may at first appear to be a very plausible approach to social injustice turns out to be relatively complex and not without challenges if scrutinized more closely.

Critical Questions to Honneth’s Theory of Recognition

Honneth’s theory of recognition has produced followers and critics, some more sympathetic to his theoretical endeavour than others. I shall focus my discussion of Honneth’s concept of recognition on the subject of its explanatory power with regard to the reconstruction of recognition as an underlying normative standard of bourgeois-capitalistic society. This has been one of the main topics of discussion between Fraser and Honneth. Before I address this subject I shall mention two other critical points: the function of the private sphere in the argument that recognition is a basic human need and the dependency of achievements upon the evaluation of others. Within Honneth’s theoretical framework, the task of adapting studies in developmental psychology is to show that human beings are ontogenetically dependent upon a relationship of mutual recognition. The primal scene is that of the relationship between the child and the significant other (typically the mother), which ideally is one of mutual recognition. Honneth refers to D ­ onald Winnicott’s and Daniel Stern’s developmental psychology and to the psychoanalytical studies of Jessica Benjamin. However, when empirical studies or

312

Gimmler

­particular traditions in an empirical discipline are used to support a theoretical claim, it always arouses the suspicion of that the study providing the best fit is the one that has been chosen. As noted by Lois McNay, one can also reference scholars other than object relations theorists. She accuses Honneth of ‘naïve psychologism’ (McNay, 2008: 276) and problematizes the mutuality of recognition in the dyad of mother and child. McNay asks whether the mother and child dyad is not a typical relation of asymmetry, considering the lack of capabilities and power of the child. Honneth argues that the ontogenetic priority of love as a principle of recognition does not only rely upon the ontogenetic primacy of individuation. The theory construction is not that of an inductive inference from empirical studies to a generalized idea. Instead, these very basic conditions of becoming a person indicate that they are indeed necessary, universal conditions to which no plausible alternative can be found. This is what Honneth calls ‘to the best of our knowledge’ (Honneth, 2003: 177). Although this solution would appear to settle the problem, it is not clear whether the empirical findings in developmental psychology that Honneth uses are merely an indication of recognition as a basic and universal need or should be viewed as an empirical warrant for the normative standard. Related to this question is the focus on intersubjectivity. Honneth thinks of recognition as the nucleus of social interaction. Recognition is bestowed by one individual upon the other and ideally is mutual. Self-confidence, ­self-respect and self-esteem are the results of intersubjective relations. In his approach to respect, Richard Sennett introduces a third relationship, the relationship that the self maintains with itself via the achievement of something.3 Sennett ­distinguishes craft and mastery: The sense of craft requires investment in the object of one’s labor as an end in itself: as Auden says in his poem, forgetting oneself in a function. Whereas in mastery the object is a means to another end, that of displaying what you have done, what you have become, to another. sennett, 2003: 84

Strictly speaking, there is no space for the self-worth produced by craft in Honneth’s theory. From Sennett’s perspective, the relationship to an object is a 3 Indeed, the relationship of the self with the product is also Hegel’s solution to the dilemma of recognition. It is in the work and the product of work that the self achieves s­ elf-consciousness, not in the per se asymmetric relationship between master and slave (Gimmler, 2004). In his interpretation of Hegel, Honneth focuses on the early Jenaer formulations in which mutual recognition is central. See also Ikäheimo and Laitinen (2007) for a philosophical analysis of the different dimensions of recognition.

Recognition

313

necessary third relationship that breaks individuals’ mutual dependency. For Sennett, the focus on mastery and on how others perceive achievements is part of a narcissist society with an exaggerated demand for recognition instead of the outcome of an ideal society guided by the normative infrastructure of recognition. The question that arises is this: By whom does one want to be recognized? A case in point could be the pride of the working class based on a clear distinction from the lifestyle and values of the upper and middle class. Sennett (and e.g., Skeggs, 1997) points to the fact that recognition is not blind to class, gender or ethnicity but instead is part of the constitution of these economic and cultural differences. If the concept of recognition is indeed able to function as a normative standard to identify violations of social justice in society in general, Honneth must show that the normative standard can be found by reconstructing the necessary conditions of successful and satisfying self-realization while also demonstrating that the normative standard of legitimate recognition is already at work in the development of society and can be applied in empirical studies. Recognition, then, is not a utopian ideal but a necessary function of social relations. This is a strong thesis proposed by Honneth with the aim of reconstructing the ‘bourgeois-capitalist society as an institutionalised recognition order’ (Honneth, 2003: 138). That said, do the three realms of recognition really cover all aspects of ‘bourgeois-capitalist society’? In one sense Honneth is right: bourgeois-capitalist society produces or fosters not only asymmetrical recognition but also symmetrical recognition, if not in the positive sense of fulfilment of recognition claims then in the negative sense of marking the lack of recognition. However, as Honneth also admits in his discussion with Nancy Fraser, capitalism constitutes a general framework of economic inequality, and all layers of social relations are liable to be affected by this inequality and the power relations that accompany it. We need consider only the economic inequality between a university cleaning staff and the professor whose office they clean. Obviously, we typically respect the cleaning staff, although to be honest, they tend to become invisible, that is, according to Honneth’s theory, they are not recognized. This is precisely the point made by Nancy Fraser against Honneth’s theory of recognition, namely, that misrecognition and maldistribution are often intertwined. It was Nancy Fraser who identified what she thinks is a problematic reductionist consequence of Honneth’s reconstruction. If recognition is the only analytical category available to analyse social injustice, the struggle of groups that both lack social recognition and are economically underprivileged is reduced to a mere cultural struggle. Collectivities, Fraser (2003) argues, are often ‘bi-valent’; that is, they are affected by both misrecognition and maldistribution, e.g., women’s unequal wages and the low level of respect for the typical

314

Gimmler

professions in which they work. This is why Fraser attacks Honneth for reducing recognition to a lifestyle problem and for ‘assum[ing] a reductive culturalist view of distribution’ (Fraser, 2003: 34). Fraser introduces as a countermove ‘a “two-dimensional” conception of justice’ (Fraser, 2003: 35). Here, distribution and recognition are two distinct dimensions of social justice, and one cannot be reduced to the other. Fraser’s main critique is that Honneth overgeneralizes recognition and ends up with a universalistic normative theory that does not grasp the differences and subtleties of societies structured along the lines of both unequal material distribution and unequal cultural representation. At least one result of the discussion between Honneth and Fraser is that in his rejoinders, Honneth must systematically integrate the issue of material distribution into his theory of recognition. To achieve this theoretical task, one must distinguish among Honneth’s various arguments and the different levels of theory construction that they address. The first level is the reconstruction of an argument regarding historical development. Here Honneth attempts to show, with the aid of contemporary and historical examples, that both the institutionalization of equality in law and the establishment of the capitalistic economy are to be understood as ­institutionalizations of the norm of recognition; even the capitalistic organization of labour follows the implicit norms of recognition (Honneth, 2010a: 95). Strikes for higher wages or safer working conditions then become part of a struggle for recognition and the actualization of an implicit norm that is regulative for the organization of the economy. Honneth might find sociological backing for this thesis in Weber’s analysis of the spirit of capitalism and in contemporary studies on the incorporation of normative claims by Luc Boltanski and Eve Chiapello. With respect to theory construction, although Honneth needs empirical examples to aid his reconstruction of the social order as an echo of legitimate claims of recognition, he also must distance himself from an overly narrow and concrete identification of the normative core with particular social movements. The reason for this inconsistency in Honneth’s argumentation is to be found at the second level of his argumentation. The concept of recognition relies upon a formal anthropology and moral theory. The normative claim of recognition, Honneth argues, cannot be bypassed without society losing its moral infrastructure. The alternative to the recognition paradigm for social theory is a thoroughly utilitarian understanding of social cohesion and relations. Honneth does not outline this alternative, but it is clear that a utilitarian or rational-choice oriented social theory would by no means fit with Critical Theory. These two arguments together—the problematic identification of the inherent normative claims of bourgeois-capitalist society and the general speculative claim of a formal anthropological theory of

Recognition

315

recognition—constitute what Zurn (2005) has called a ‘generality/concretion dilemma’ (Zurn, 2005: 105). One could reformulate the dilemma as follows: either the theory is situated at a high level of generality and consequently argues for recognition as the implicit norm for all social relations, or the theory is sufficiently concrete for the identification of specific experiences and social movements that realize the normative concept of recognition. If the theory is overly general and abstract, it tends to be empirically empty. In the case of Honneth’s recognition theory, all types of real experiences might support the thesis of recognition as the normative core of modern Western societies, and all types of experiences of neglect might qualify as social pathology. An analysis of society based on these premises would be hopelessly fuzzy. However, if the theory is as concrete as possible, the empirical evidence might not back up the speculative claim or even contradict the theory. This is a serious problem of the theory of recognition that directly affects both the sociological productivity and applicability of the concept of recognition. This dilemma is also apparent when we explore the criteria for legitimate and illegitimate experiences of recognition. Honneth’s concept of recognition must distinguish between illegitimate and legitimate recognition; otherwise, we end up with a concept that can be applied to all types of social upheavals or even all types of social action. The obvious examples of such discrimination between legitimate and illegitimate recognition are racist or homophobic groups. Honneth makes the following observation: It is obvious that we cannot approve of every political uprising as such, nor hold every demand for recognition to be morally legitimate or defensible. Rather, in general we judge the objectives of such struggles to be positive only when they point in the direction of a societal development that we can grasp as coming closer to our notions of a good or just society. honneth, 2004: 353

Honneth appears to have a strong argument, noting that modern Western societies are built upon implicit normative principles such as equality, inclusion and fairness. Groups that do not work towards these norms cannot be viewed as making legitimate demands. However, the normative ideal of a good or just society is initially constructed by idealizing experiences of disrespect and neglect. Unfortunately, this is a circular argument. Honneth also admits that in empirical cases it could be difficult to evaluate whether a demand is legitimate or illegitimate. The idea of a ‘good society’ upheld by a significant number of people might conflict with this act of recognition as regards the rights of gay and lesbian people.

316

Gimmler

Honneth attempts to solve this problem by introducing a moderate form of value realism (Honneth, 2010b). Recognition is actualized as the act of affirming the positive characteristics and qualities of individuals and groups. Candidates for qualities that are entitled to recognition must be part of a plausible and reasonable set of values. People can be recognized for their achievements, their contributions to a group, their characters and the values they hold, as bearers of rights or simply as human beings. It is necessary to restrict recognition to a plausible and reasonable set of values. Racist groups, for example, demand recognition for their values but are unwilling to provide reciprocal recognition to other groups (Honneth, 2010b: 118). What counts as more or less plausible might differ in societies and subgroups. Honneth acknowledges this point by arguing for a cultural and historical contextualization of values while simultaneously dismissing value relativism. At this point, one might wonder whether the intended rich phenomenology of recognition is sufficiently supported by the normative limitations inherent to Honneth’s notion of recognition. Does the normatively loaded notion of recognition capture all of the contradictions and subtleties of acts of recognition and misrecognition?

Conclusion: Recognition in Action?

The theory of recognition comes in different flavours, so to speak. In this article, I have focused on Honneth’s version of the theory of recognition. Although the basic idea of grounding a social theory in the immanent claims of social reality itself—thereby developing not only a descriptive vocabulary but also a normative vocabulary that constitutes the moral infrastructure of both social action and systems—appears very attractive, the actual realization of such a normative theory is not without difficulties. The problem of how the normative core of the theory can be justified is a major problem in Honneth’s theory of recognition. Oscillating between universalism and contextualism, he attempts to keep the theory of recognition applicable and backed by social reality; concurrently, the concept of recognition must attain a compelling normative authority in order to function as a standard for the diagnosis of what Honneth calls social pathologies. For Honneth and Critical Theory, the explanatory power of the concept of recognition is bound to this normative justification. Is this really necessary? Another problem with the theory is the claim that recognition functions as a concept that enables analyses of society as such. However, how does the concept of recognition address systems and the different forms of ­integration?

Recognition

317

Strictly speaking, system integration and its consequences for social integration are difficult to grasp using the concept of recognition. An example is the practical inequalities that arise in welfare-state policies in which certain citizens are more apt to use the system than others (see Lockwood, 1996). Equality is realized on paper but not in reality. A theory of recognition would have to address the fact that even ‘good’ principles might have ‘bad’ results. Another point that cannot easily be dismissed is the heuristic mistrust mentioned in the introduction to this chapter. The concept of recognition might be allied with precisely those political and social realities that it wishes to criticize. ­Somewhat polemically, but in accordance with Fraser’s critique, the ­neo-pragmatist Richard Rorty voices the same suspicion: ‘My feeling is that there’s been a tacit collaboration between right and left in changing the subject from money to culture’ (Rorty, Nystrom and Puckett, 2002: 32). Rorty is not criticizing Honneth directly, but as a pragmatist he would prefer to see more real social reform and fewer theoretical battles about normative justifications for moral principles. References Basaure, M. (2011). “Continuity through rupture with the Frankfurt school: Axel Honneth’s theory of recognition”, Eds. G. Delanty and S.P. Turner. Routledge international handbook of contemporary social and political theory. London: Routledge. Benhabib, S. (2002). The claims of culture. Equality and diversity in the global era. Prince­ ton: Princeton University Press. Cooke, M. (2000). “Between ‘objectivism’ and ‘contextualism’: the normative foundations of social philosophy”, Critical Horizons. 1(2): 193–227. Fraser, N. (2003). “Social justice in the age of identity politics: redistribution, recognition, and participation”, Eds. N. Fraser and A. Honneth. Redistribution or recognition. A political-philosophical exchange. London and New York: Verso Books. Fraser, N. and Honneth, A. (2003). Redistribution or recognition. A ­political-philosophical exchange. London and New York: Verso Books. Fraser, N. and Naples, N.A. (2004). “To interpret the world and to change it: an interview with Nancy Fraser”, Signs. 29(4): 1103–1124. Genel, K. and Deranty, J.-P. (2016). Recognition or disagreement. New York: Columbia University Press. Gimmler, A. (2004). “Pragmatic aspects of Hegel’s thought”, Eds. W. Eddinton and M. Sandbothe. The pragmatic turn in philosophy. Albany: State University of New York Press.

318

Gimmler

Habermas, J. (1989). Theory of communicative action. Vol. 2. Cambridge: Polity Press. Habermas, J. (1993). “Struggles for recognition in constitutional States”, European Journal of Philosophy. 1(2): 128–155. Haslebo, M.L. and Lyndgaard, D.B. (2007). Anerkendende ledelse. [Appreciative leadership]. København: Dansk Psykologisk Forlag. Honneth, A. (1987). “Critical theory”, Eds. A. Giddens and J.H. Turner. Social theory today. Polity Press: Cambridge. Honneth, A. (1989). Kritik der Macht. Reflexionsstufen einer kritischen Gesellschaftstheorie. Frankfurt a.M.: Suhrkamp. Honneth, A. (1999). “The social dynamics of disrespect”, Ed. P. Dews. Habermas. A critical reader. Cambridge: Blackwell. Honneth, A. (2001). “Invisibility: on the epistemology of ‘Recognition’”, The Aristotelian Society. Supplementary Volume lxxv. 75(1): 111–126. Honneth, A. (2003). “Redistribution as recognition: a response to Nancy Fraser”, Eds. N. Fraser and A. Honneth. Redistribution or recognition. A political-philosophical exchange. London and New York: Verso Books. Honneth, A. (2004). “Recognition and justice: outline of a plural theory of justice”, Acta Sociologica. 47(4): 352–364. Honneth, A. (2010a). “Arbeit und Anerkennung. Versuch einer theoretischen Neubestimmung”, Das Ich im Wir. Frankfurt a.M.: Suhrkamp. Honneth, A. (2010b). “Anerkennung als Ideologie. Zum Zusammenhang von Moral und Macht”, Das Ich im Wir. Frankfurt a.M.: Suhrkamp. Honneth, A. (2014), Freedom’s right. The social foundations of Democratic Life, New York: Columbia University Press. Houston, S. (2009). “Communication, recognition and social work: aligning the ethical theories of Habermas and Honneth”, The British Journal of Social Work. 39(7): 1274–1290. Hugman, R. (2005). New approaches in ethics for the caring professions. Houndsmill, Basingstok: Macmillan Palgrave. Ikäheimo, H. and Laitinen, A. (2007). “Analzying recognition: identification, acknowledgement, and recognitive attitude towards persons”, Eds. B. van den Brink and D. Owen. Recognition and power. Cambridge: Cambridge University Press. Jørgensen, S.L. (2015). “Recognition, education, and civic equality: uncovering the normative ideals of the welfare state”, Eds. J. Jakobsen and O. Lysaker. Social and critical theory: recognition and freedom: Axel Honneth’s political thought. Leiden: Brill. Juul, S. (2009). “Recognition and judgment in social work”, European Journal of Social Work. 12(4): 403–417. Laitinen, A., Särkelä, A. and Ikäheimo, H. (2015). “Pathologies of recognition: an introduction”, Studies in Social & Political Thought. 25: 3–24.

Recognition

319

Lockwood, D. (1996). “Civic integration and class formation”, British Journal of Sociology. 47(3): 531–550. McNay, L. (2008). “The trouble with recognition: subjectivity, suffering and agency”, Sociological Theory. 26(3): 271–296. Oliver, K. (2001). Witnessing. Beyond recognition. Minneapolis: University of Minneapolis Press. Patchen, M. (2003). Bound by recognition. Princeton and Oxford: Princeton University Press. Personalestyrelsen (2008). “Anerkendende ledelse i staten”. [Appreciative leadership in the state]. Downloaded from http://perst.dk/~/media/Publications/2008/Anerkendende%20ledelse%20i%20staten/Anerkendende%20ledelse%20endelig.ashx. Rorty, R., Nystrom, D. and Puckett, K. (2002). Against bosses, against oligarchies. Chicago: Prickly Paradigm Press. Sennett, R. (2003). Respect in a world of inequality. New York and London: W.W. Norton. Skeggs, B. (1997). Formations of class and gender: becoming respectable. London: Sage. Sonntag, S. (1978). Illness as metaphor. New York: Farrar, Straus & Giroux. Taylor, C. (1992). “The politics of recognition”, Multiculturalism and ‘The politics of recognition’. Princeton: Princeton University Press. Thyssen, O. and Dahl, H. (2006). Krigeren, borgeren og taberen. [The warrior, the bourgeois and the loser]. København: Gyldendal. Zizek, S. (2003). Ein Plädoyer für die Intoleranz. Wien: Passagen Verlag. Zurn, C.F. (2005). “Recognition, redistribution, and democracy: dilemmas of Honneth’s critical social theory”, European Journal of Philosophy. 13(1): 89–126.

Index Abductive inferences 13 action communicative 46, 226, 308 individual 161, 233 joint 109–10 Actor-Network-Theory 120 actor qualities 195–196, 202 actor 183–184, 189–190, 193–202, 211 collective 198, 237, 255 institutional understanding of 193–196, 200 rational 192, 195–196, 199, 222, 227–228 social construction of 194–195 status 194–196, 202 vocabulary 57–58 adequacy, practical 155–156 affluence 129–130 agency 50, 153, 164, 195, 204, 208, 215, 274–276, 279, 283, 288, 319 agents 53, 137, 192, 196, 202, 209, 215, 249, 271 Althusser 163, 245 ambiguity 25, 52–53, 56, 187 Anthropocene 116, 129, 131 application 2, 4, 5, 9, 41, 46, 58, 165, 304, 305 approaches 14–15, 18–19, 71–73, 81–82, 91, 109, 111, 143–144, 219 appropriation 254–257, 262 appropriation relations 254–255 Aristotle 8–9, 46–47 asset 248, 250–252 combined 249–250 personal 249–250 productive 247–248, 252, 254 attitude 27, 33, 103, 105, 111–112, 201, 264–265 attribute 7, 29, 84, 194–195, 198 authority 91, 174, 183, 204, 269 Beck 104, 119–120 Becker 11–12, 20, 23, 36, 110, 113, 186–187, 206, 212, 219 behaviour 57–58, 110, 142, 144, 166, 172–173, 176, 222, 224, 230, 266–268 belief 77, 98, 101–103, 105, 108–111, 160, 191, 197

benefit 83, 202, 222, 229, 233, 236, 250, 256, 293, 296–297 collective 222, 225, 227 Berger 37, 110, 113, 184, 186, 191, 203, 277 Bhaskar 155 biological reference 288–290 biology 9, 119, 121, 127, 157, 159, 268, 269–270, 288–290 Blumer 13–14, 20, 24, 37, 63, 74, 83–85, 87–89, 93, 162, 169 bodies 83, 227, 232–235, 264–265, 269, 270–271, 277, 281–282, 290 Boltanski 90–93, 199, 203 Bourdieu 5, 9, 20–21, 47, 52, 111–113, 140, 144, 163, 219, 224, 226, 228, 270–271 Bulmer 88–89 Burawoy 260, 261, 263 Butler 270–271, 273, 275–276, 278, 280–282 capacity 65, 194, 199, 222–223, 225, 229, 236–237, 244, 246–247, 252, 255–256, 264 capital 142, 169, 224, 226, 247, 257, 270 cultural 108, 112, 163 capitalism 48, 97, 102, 115, 123, 134, 252–263, 282, 313–314 capitalist 243, 252–255, 310 Carbin 80–81 categories 7, 11, 63, 82–83, 85–88, 273, 275, 289–290, 293, 295 cultural 146, 273–274 economic 243, 260–261 political 260, 295–296 categorization 6, 88, 287, 290–291, 296, 300 causal power 15, 59, 133, 169–170 change 27, 100, 103–104, 106–107, 121, 123, 147, 184–185, 191, 261, 276, 279 cultural 101–102 institutional 176, 184, 188, 193 children 26, 75, 93, 107, 112, 115, 148, 194, 207–208, 214, 252, 289, 309, 311–312 choice 139, 146, 187, 229, 234, 236, 254 citizen 56, 107, 123, 138, 197, 271, 295, 297, 304, 310, 317

322 claim 144–145, 147–148, 179–180, 235–236, 273–274, 277–278, 280–281, 316 normative 229, 307, 314 class 7, 19, 80, 140, 243–244, 247–251, 253–259, 262–263, 298 analysis 140, 244–245, 253–254, 259–260 capitalist 257, 262 exploitation 251, 254, 256–257, 259 formation 19, 244–245, 260–261, 319 lower 111–112 structure 6, 156, 243–249, 251, 253, 255, 257–259, 261, 263 struggle 19, 243–245, 254, 256, 308 working 102, 141, 257, 261–262, 313 classification 6, 11, 72, 77, 174, 177, 285 class-in-itself 243, 260 class relation 209, 243, 246, 251–259, 262 definition of 243, 247, 251 climate change 120 coercion 249, 259–260 extra-economic 251, 259, 261 co-evolution 121–122, 131 collective action 61, 183, 222–233, 235–239, 272–273 rational 19, 222–224, 235 social system of 223, 229, 236 collective consciousness 48, 54, 101–103, 107, 109, 230, 234 collectives 1, 98, 103, 194–195, 304 colligation 8, 17–18, 63–77 colligatory concepts in history 71 communication 44, 46, 99–100, 105, 108, 112, 160–161, 164, 167, 231, 233, 235, 308 communism 259–260, 262, 302, 305 community 47, 101, 108–109, 155, 237–238, 281 Comte 69, 148, 157–159, 169 concept construction 10–11, 17, 63, 237 concept formation 10–13, 21, 25–26, 38, 76, 88 conception 65–68, 99, 116, 119, 156, 176, 270, 272, 279, 318 concept 1–39, 63, 73–74, 76, 83–85, 88–93, 285–286, 289–291, 297, 303–307, 316–317 abstract 7, 257 basic 14, 154 central 22, 61, 64, 188, 291 classificatory 26–27

Index definitive 84, 87–88 ethnicity 285–286, 292 general 18, 101 homeless 8, 173 interactive 7, 8, 44 key 20–21, 101, 110–111 network 177 non-classificatory 26 sensitizing 17, 24, 63, 76, 79–92 social 6–7 sociological 4–6, 29, 31, 35, 74, 137, 153, 168, 293–294 conceptualization 12, 18, 46, 154, 156, 172, 192, 234, 302, 307–308 conceptualizing 69, 159, 171, 260 conceptual space 3, 18, 20, 58, 83, 172, 208–209, 211, 290 condition 27, 48–49, 55–56, 120, 132, 159–160, 250–251, 256–258, 307 human 119, 134, 144 natural 117, 122–123, 130 conflict 2, 16, 104–105, 108, 110–111, 222, 227–228, 232, 236–237, 258, 288 social 107, 169, 171, 255–256, 258–259, 263 construction 1–2, 4–6, 10–11, 13, 18, 21, 130, 133, 235, 301, 303, 305 conceptual 1, 4 consumer 137, 139, 141–142, 148–149, 165 society 140, 149 consumption 17–18, 95, 107, 123, 130, 132, 138–150, 200 history of 139–140 modern 138, 143 non-material 17, 130 sociology of 136–137, 141, 143–144, 146, 148 context 3, 8–10, 14–16, 88–89, 190, 225–226, 278–279, 305 normative 54 social 17, 45 theoretical 9, 53 contradiction 15, 53, 58, 116–117, 159, 161–162, 216, 271, 304, 316 control 112–113, 117, 121, 158, 180, 205, 243, 252, 259, 271 convention 52–54, 62, 176, 201, 264 Cooley 167, 169 cooperation 54, 225–227, 233, 239, 272 craft 312

Index

323

Crenshaw 81–82 critique 33, 90–91, 136, 138, 143–146, 149, 287, 289, 291, 303, 304 Crutzen 129, 133–134 cultural representation 165–167, 274 cultural sociology 72, 100, 113 new 98 cultural studies 143–144 culture 1–2, 6–7, 17, 97–109, 111, 113–115, 139, 190, 195, 283, 317 civic 104, 112–113 power of 107, 109 sociology of 78, 99–101

disrespect 306–307, 309–310, 315, 318 distribution 100, 158, 160, 247–248, 251, 254–255, 257–258, 264–265, 269, 279, 314 domination 243, 247–248, 250–260, 263 matrix of 81–82 relation 243, 246–248, 250, 252–257, 259, 263 Douglas 146–147 Durkheim 47–48, 59, 101, 105–107, 109–110, 136, 159–161, 165–166, 217 dynamics 6, 60, 62, 176, 184–185, 187–188, 256, 267

de Beauvoir 267, 272, 275, 281 decision 24, 60, 144, 174–175, 180, 183, 211, 226, 233–235, 237 organizational 174, 181 de-colligation 73, 75–76 deduction 11–13, 69, 71 Degele 80, 82–83 development 23–24, 49–50, 58–59, 88–89, 103–104, 116, 121, 124, 140, 287 sustainable 124 technological 124, 131 diagnose 7, 101, 103–104, 303, 305–306, 308, 316 dichotomy 265, 267, 271, 274, 281 differences 80, 83, 85, 90, 176, 200, 265, 269, 271, 274, 280–281, 295–297 average 265–267 biological 269–270, 288 cultural 291, 313 differentiation, functional 206, 223, 235, 238 diffusion 177, 189, 190, 193, 204–205, 294 dilemma, equality/difference 273 DiMaggio 179 dimension 15, 80, 119, 127–129, 243, 264–265, 267, 273, 275, 279, 289, 292 discipline 8, 30, 84–85, 98–99, 136, 144, 201, 280 discourse 4–5, 8, 10–11, 54, 59, 60, 65, 264–265, 275, 279, 291–292, 294 public 87, 224, 226, 236, 291 societal 194, 294 theoretical 231, 234 discovery 20, 36, 66–69, 76, 78, 117, 121–122 discrimination 264, 274, 292, 295–298, 315

economic 121, 136, 140, 142, 145, 147, 187–188, 201, 259 relation 159, 243, 246, 251 theory 146 economists 137–138, 145–146, 148, 201 economy 119, 130, 134, 136, 158–159, 170, 173, 223, 236, 238, 247–248, 301, 305 political 69, 149, 170, 283 Edenheim 80–81, 93–94 education 103, 111–113, 117, 191, 197, 204, 209, 223, 235–236, 269, 283, 318 elements, cultural 101, 104–105 ellipse 66, 69 Elster 50, 52–54, 60, 227, 229, 231, 238 empirical research 16, 28–30, 35, 38–39, 63–64, 80, 94, 98, 108, 179, 217, 291–293, 301, 311–312 energy 117, 120–121, 124–125, 131–132, 195 environment 14, 119–120, 124, 126–127, 129, 133, 135–136, 180–182, 224–225 epochal social change 255–256 equality 80, 86–87, 90, 93, 205, 271, 274, 305, 309–310, 314–315, 317 equilibrium 3, 157–159, 168, 211–212 Esping-Andersen 212 essence 4, 211, 272, 275, 277 ethnicity 80, 82–83, 86–87, 90, 92–93, 291–293, 296–298, 299–301 events 64, 70–73, 78, 117, 142, 214 exchange 109, 166, 225, 249, 253 political-philosophical 317–318 exclusion 93, 107, 273, 277, 285–286, 288, 291, 296–298 expansion, global organizational 185–186, 203

324 expectation 85, 109–110, 125, 155, 165, 185, 189, 191–192, 198, 213, 217, 231 experience 154, 231–232, 267–268, 273–275, 278–279, 307, 309–311, 315 explanatory power 13, 302–303, 305, 311, 316 exploitation 123, 159, 244, 247–259, 262–263 fact decomposition of 67, 74, 76 social 47, 160, 211, 213–214, 219 factories 75, 138, 158, 246 failure 29, 217, 231 fairness 55, 192, 201, 315 family 18, 86, 93, 107, 109, 114–115, 148, 170, 206, 207–221, 252, 264 cultures 100, 111 institution 207–208, 218 law 214, 220 practice 213, 215 roles 107, 212, 218 sphere 210, 212, 216, 218 system 210–212, 214–215, 217, 219, 221 types 209, 213, 215 femininity 266–267, 272, 278 feminism 81, 221, 271, 273, 281–283 gynocentric 272–273 feminist theory 19, 81, 93–94, 265, 268–269, 283 feudalism 251, 259–260 fields 98–99, 121–122, 130–131, 139, 143, 200–201, 209, 223–224, 238, 280 social 9, 232 forces 15, 103, 105, 116–117, 120, 129, 134, 158, 168, 195, 232 cultural 97, 106 social 32, 36, 39, 43, 117–118 formal rules 18, 172, 176–177, 180 formation 10, 13, 54, 89, 180, 198, 261 Foucault 8–9, 50, 60, 163, 170, 223–224, 228, 238, 270 framework 9, 19, 79–80, 91, 126, 157, 162, 164 interactionist 80, 87–90, 92 theoretical 12, 211, 218, 307, 311 Fraser 274, 276, 282, 302–7, 311, 313–314, 317–318 fruitfulness 12 functional context 3, 9 functionalism 9, 56, 191, 223, 238 functional reasoning 3, 9, 210, 215

Index function problem 229, 236 function 1–3, 8–9, 13–14, 17, 47, 57, 106–108, 171, 206, 218, 222–223, 237, 305 theoretical 15–16 systems 223–226 game theory 227, 249–251 gender 7, 19, 80, 212–213, 264–283, 298 contracts 213 differences 264, 266, 268–269, 271–273, 279 doing 109, 275 equality 86–87, 91, 212, 215, 271, 273, 279 group 265–266, 279 hierarchies 269, 298 identity 269, 274, 276, 278 roles 106, 110, 213–214, 269 patterns 268, 278–279 social 268, 279 studies 264–265, 268, 276–277, 280, 289 system 270 genealogy 8, 10, 46, 157, 286, 305 generations 86, 124–126, 132, 166 Giddens 56, 103–104, 143–144, 149, 164, 175–176 gift exchange 90 Goertz 31, 35, 37 Goffman 3, 10, 21, 162, 170 goods, collective 222–223, 225, 227–229 grammar 43, 45–46, 60, 162, 165 Granovetter 177–179 Gronow 143–145 groups 55, 61, 80, 99–101, 103, 176, 224, 239, 250, 264, 290, 310, 316 ethnic 85–86, 291–292 guards 109–110, 250, 276 Habermas 46, 49, 116, 133, 203, 223–224, 226, 228, 238, 303–304, 307–308, 318 habitus 59, 111–112, 165, 270, 275, 283 Hacking 5–6, 21, 44, 61, 277, 282 Hannan 185, 187, 191–192, 203–204 Haraway 118, 133, 271, 277, 282 Hempel 10, 21, 25, 38, 71–72, 77 hierarchies 80, 183, 188, 225, 267, 271, 287–288, 291–292 social 285, 288, 294–295 Hill Collins 81–82 historical materialism 256

325

Index history 64–66, 70–72, 76–78, 102, 137, 212, 283, 285, 289, 294–298, 302 Hobbesian problem 226–227 Honneth 302–304, 306–319 human beings 6–7, 49, 54, 225, 232, 269, 272, 285, 288, 307, 309–311, 316 individual 229–232, 235 humanities 98, 125–126, 134, 146, 278, 287 identity 144, 273, 276–277, 280–281, 292, 304, 307–308 ideal type 12–13, 25, 63, 74, 76, 197, 211, 215 imitation 99, 107, 110, 112, 141–142, 166–167, 171 indicators 26–28, 30–31, 35, 57, 73, 117, 199, 209, 294 individualism, object 231–232 individuals 1–2, 56, 161, 166–168, 192, 194–195, 199–200, 213, 222, 227–233, 235 induction 11–12, 65–70, 75 industrial society 102 inequality 111, 250, 252, 258, 264, 289, 292–293, 297–298, 301–302, 304, 313, 319 social 94, 111, 113, 274, 292–293 institutional analysis 55, 190, 202 entrepreneurs 192 perspective 163, 189–193, 195–197, 199, 201, 203, 205 work 192 institutionalism, new 189–193, 200 institutions 55, 105–106, 158, 162, 164, 167, 172, 174–177, 179–185, 188–189 integration 2–3, 15–16, 27, 38, 159, 161, 222, 228–230, 280, 286, 297, 299 cultural 108 intention 230–231, 233–235 interaction patterns 100, 106 interaction 6–7, 16–18, 59–60, 97, 100–111, 162, 234–235, 267, 275 human 172, 175–176, 183 interests 16, 18, 19, 27, 34–35, 195–196, 211–212, 215, 218, 237, 244, 309 intersectionality 17, 79–83, 85, 87, 89–90, 92–94, 286, 301 island credit-market 248, 251–252 labor-marked 248, 251

justice 306, 308, 314, 318 social 20, 264, 309, 313–314, 317 Kepler 66, 68–69 knowledge 79, 82, 85, 94, 98, 101–102, 111–112, 129, 154–155, 168, 170, 272–273 scientific 78, 154–155 labor 113, 136, 149, 190, 220, 246, 248–249, 253, 255, 312 power 243, 248, 252–254 process 249, 252, 259–261 language 18–19, 25–26, 37–38, 44–46, 60, 86, 193, 217, 226, 233, 278 Latour 118, 120, 134, 163, 170 laws 45, 67, 71–72, 99, 102, 120, 171, 177, 214, 220, 302, 305, 310, 314 Lazarsfeld 25–29, 31, 33–34, 38–39 legitimation 189, 191, 197, 202 Lévi-Strauss 162–164, 170 lifestyle 18, 131, 136, 138–140, 142, 144–145, 148, 313 Lockwood 51, 58, 61, 161, 170, 317, 319 logic 2–3, 12, 24, 37, 39, 188, 245 Luckmann 111, 113, 184, 186, 191, 203, 277 Luhmann 23, 38, 134, 175, 186–187, 194, 204, 223–225, 228, 230–231, 234–236, 239 Malinowski 9, 21, 55–56, 61 managers 257 March 174–175, 185, 187–188, 205 markets 103, 124, 139, 188, 214, 236, 247–248, 251, 264 marriage 164, 182, 214–215, 220, 266 Martin 34, 38, 165, 170, 290, 293, 300 Marx 20, 48, 61, 114, 117, 134, 136, 148, 158–161, 169–170, 253, 263 Marxism 48, 50, 162, 201, 302 Marxist 50, 209, 243, 254–257, 259–260, 273 conceptions of class structure 263 concept of class 253, 262 perspective 48–49, 262 Marxist theory 199, 243–244, 262 masculinity 216, 266–267, 272 material 11–12, 17, 33, 118, 120, 123–124, 128, 130, 132, 157–158, 160–161 material structures 100–101, 104 McCall 82 McCracken 139, 145, 147, 149

326 McNay 271, 274–276, 283, 304, 312, 319 measurement 20, 27, 30, 35, 37–38, 72, 76 mechanisms 110–111, 122, 140–142, 154–156, 186, 266 Merton 11–12, 21, 23–24, 28–31, 33–34, 38–39, 64–65, 77, 161, 170 metaphors 1, 90, 150, 157, 213, 218–219, 221, 308, 319 methodology 23–25, 28, 30, 35, 79–80, 87, 90, 155, 164, 171, 298 Meyer 182, 186, 188, 190–193, 195–197, 203–205 micro 172–173, 177, 184, 294 middle class 111–112, 141, 313 migration 288, 291–292, 297, 300–301 minorities 80, 82, 86–87, 192, 295, 297, 299 non-white 295 modernity 59, 103, 140, 143 modern society 116–117, 129, 223–226, 235–237, 303–304 late 117, 121, 132, 275 modes of production 259–262 Mouzelis 59, 106 multiculturalism 303–304 mutual recognition 309–312 nationalism 296, 299, 300 nations, cosmopolitan 56–57 nation states 192, 197–198, 292 natural resources 123, 125–126 natural sciences 6, 65–66, 69–70, 73, 287 nature 116–126, 158, 245–246 negotiations 86 network 172, 174, 177–180, 183–185, 194 analysis 177–178 integrated of concepts 88–90, 92 theory 179 ties 178, 180 Nicholson 270 nominalism 276–280 Non-Marxist concepts of class 254 normative core 311, 314–316 theory 46, 303–304, 314, 316 normativity 46–48, 51, 60, 304, 307 norm 16–18, 43–60, 98, 107–108, 191, 218, 275 concept 44, 46, 49–53, 57–60 general 51, 57

Index implicit 314–315 informal 52, 172, 176 institutional 164, 185 legal 52–53 moral 53–54, 60 systems 45–46 nuclear family 108, 207–208, 269–270 object individualism 231–232, 237 ontological status 44–45 ontology, new 232–235 order institutional 162 social 182, 191, 227, 230 organizational analysis 187, 203 institutionalism 187 theory 18, 187 organization 18, 120, 157, 172–186, 189–195, 198, 200, 202, 206, 210 theory 174, 176–177, 181–182 Ostrom 54–55 parents 111–112, 208, 214, 252 Parsons 49, 55–56, 101, 106–108, 119–120, 161–162, 209–211, 218 Peirce 69 personality 111 perspectives 14, 19, 44, 46, 49–52, 137–138, 142–143, 146, 265, 274 principal/agency 196 queer 274 subjective 223, 234, 238 phenotype 288, 298 philosophy 45, 308 social 306 planetary boundaries 17, 123, 125–128, 131–132 political dimensions 246, 257, 260–262 mobilization 290 practices 246 politics 19, 243–244, 260–263, 273, 292, 294 global 261–262 population 5, 288 size 129–130 positions 45, 48, 82, 99–100, 106–8, 163, 195–196, 255–256, 272, 279, 307–08 gynocentric 272

Index power 1–2, 90, 110, 142, 184, 224, 228, 237–238, 255–256, 267, 269–270, 273, 279 effective 252–253 labour 158 political 102, 171, 224, 263, 305 pre-intentions 233–234 principles, structural 163–164, 175 processes 12, 73, 75–76, 80, 102–103, 110, 120, 121–123, 128, 168, 274–275, 292, 298 production 137–138, 199, 245, 247–248, 255–256, 259–263 politics 260–263 relations 158, 255–256, 260–262 product, surplus 254–256, 262 property rights 252–253 race 19, 34, 82, 284–299 race concept 284–299 biological 289, 295 sociological 290, 292 racial groups 284, 290, 293–295 typology 287–288, 297 racialization 290–292, 294, 298 racism 7, 81, 284–291, 295–297 rational choice theory 227, 229 rational solidarity 225, 228 rationality individual 222–223, 225, 227–229, 232, 236–237 collective 203, 222–223, 227–229, 232, 238 rationalization 102–103, 192, 195 reality-reference 294 reasoning, normative 46–50, 58 recognition 19–20, 271, 276, 302–319 notion of 302, 304 politics of 303–304, 319 spheres of 309–310 theory of 303, 305–306, 309, 311, 314–318 re-colligation 73, 75–76 relational structure 99–100, 102, 106, 164–166, 183, 217 relation 15–16, 33, 44, 97, 99–100, 158, 160, 190, 243, 246, 249, 251–252 ideological 246 institutional 159

327 intimate 309 legal 158–159 political 243–246, 249, 251, 262 real 158–159 relationship 172–174, 178, 247, 250–251, 312–313 closed 173–174 religion 46, 102, 104–105 representations 75, 83, 155–156, 160, 165–166 reproduction 98, 102, 111, 113, 175–176, 180, 192, 199, 256–257, 272, 279 research 22–23, 25–27, 36–38, 68, 81–82, 92, 129, 191, 201, 295 institutional 190–191, 197, 202 practice 30–31, 75 traditions 12, 81–82, 290 quantitative 36–37, 293, 295 residual category 222–223, 227–229, 232 resources 8, 19, 116, 121, 123–124, 164, 184–185, 192, 195, 198, 202, 256–257 rights 271–272, 310 Roemer 244, 247–248, 250–254, 258–259, 263 Rokkan 101, 104 role 26–27, 74–75, 79, 81, 107–108 theory 99–100 Rosenberg 25–26 rules 67, 71, 164, 165–166, 168–169, 172, 174–177, 180–181, 183–184, 237 embodied 164–165 informal 176, 180 Safe zone 127–128 Sami 86–87, 295 sanction 53–56, 99, 107, 167, 199 scheme, interpretative 44 Schutz 63 science 4–5, 38, 65–69, 75–78, 134, 235–236 behavioral 26, 38–39 inductive 65–66, 78 philosophy of 10, 25, 29 political 104 Searle 233–235, 239 self-interest 196, 199, 229 self-realization 306–307, 309–310, 313 Selznick 181–182, 202 Sennett 302, 304, 312–313 sex 93, 219, 265, 268–271, 279, 281–283, 289 sex and gender 268–270, 279

328 sex/gender distinction 269–270 sexual difference 268 sexuality 80, 82, 220, 264–265, 268, 273, 277–278 Simmel 141–142, 165 skills 111–112, 258–259 social action 44, 48, 52, 55–59, 233, 315–316 actors 14, 16, 18, 57, 73, 79, 82, 173, 178, 182, 184–185, 229, 235 change 101, 123, 159, 184–185, 243, 255–257, 259, 265 class 105, 110, 112 constructionism 5–6, 20, 277, 294 construction 1, 5–7, 14, 45, 109, 202, 267, 270, 285 group 58, 82, 157, 160 institutions 105–106, 111, 157–158, 182, 285, 295, 304 interaction 19–20, 57, 59, 102, 210–211, 288, 292, 294, 296, 298 life 47–49, 51, 57, 98, 100, 102–103, 160, 162, 164–165 morphology 160, 163 movements 176, 192–194, 196, 224, 257, 289, 307–308, 315 norms 43, 52–55, 60–61 organism 157–161 pathologies 303, 306, 308, 311, 315–316 reality 48, 59–60, 73, 79, 81, 118, 228, 303, 306–307, 316 relation 158, 160–163, 178–179, 232, 244, 246–247, 256, 298, 306–308 structure 15–16, 18, 29, 48, 97, 153–154, 156–171, 184, 216 system 47–49, 99–101, 120–121, 158–159, 210, 223, 231, 234–236 social relationships 18, 172–173, 177, 180, 185, 250–251 closed 174 social research 38–39, 79, 93, 239 language of 25–28, 30–31, 37–39 methodology of 37–38 socialism 253, 258–260, 262 socialization 107–108 societal differentiation 228, 309 integration 222, 224–226, 228–229, 231, 237

Index society 47, 49, 61, 99, 105, 107, 120, 149, 157–158, 209, 223–224, 229–234, 236–238 affluent 139 bourgeois-capitalist 313–314 capitalist 117, 253 contemporary 104, 116, 129, 131, 192, 195, 201 global 185 good 315 human 146, 157 socialist 258 society-nature 116, 122 coevolution 135 sociological analysis 73–74, 99, 105, 153, 156, 286, 291, 293 dictionaries 5, 72 interpretation 205, 208 perspective 44, 75, 145, 270 semantics 64 theory 33, 52, 79, 217, 231, 233 tradition 3, 6, 35, 118, 208 sociologists 23–24, 30–31, 34, 72–73, 99, 101, 105, 109, 210, 216–17, 288 qualitative 35 works of 65, 72 sociology 1–2, 4–6, 23–24, 29, 105, 201 classic 46–47, 49 contemporary 138, 142, 145 culture in 97–98 general 48, 140 modern 51, 74, 97, 101, 105, 109 Spencer 47, 136, 157–159, 161, 165 stability 107–108, 126–127, 159, 161, 172–173, 184, 212, 281 states 47–48, 133–134, 183–184, 189, 194–195, 198, 200, 243, 260–262 internal 234 modern 197 status 156, 194, 195–197, 208 exploitation 258–259 Steffen 125–129, 134 stereotypes 216, 266, 278–379 Stinchcombe 30, 39, 185, 188, 191, 205 strategies 3, 9, 11–12, 186, 198, 222, 225, 231, 236, 247, 257–258 structural approaches 14–15 structural inertia and organizational change 187

329

Index structure-concept, general 9 structures 7–8, 15, 18, 100, 117, 157–158, 162, 168, 171, 175, 178, 183–188 cultural 17, 99–100, 106, 217 embodied 18, 164–165, 168 formal 181, 188–189, 198 institutional 164–165, 183, 204 normative 44 organizational 181, 186, 189, 191 societal 100, 238 subcultures 99, 104, 111–112 subject 14, 81, 84–85, 105, 120, 143, 145, 161, 275, 280, 308, 310–311, 317 subjectivity 170, 246, 271, 274–275, 279, 281, 308, 319 subordination 111, 197, 212–213, 243, 246–247, 250–252 subsistence niche 122–123 subsystems 106, 119, 224, 236–237 suicide 107–108 superstructure 158–159, 162 sustainability 54, 124–125, 127, 133–135, 150, 196 symbolic meanings 118–119, 191 symbols 98–99, 104, 109, 111, 117–118, 130, 203, 225, 228, 264 system perspective 57–60, 119 abstract 55 system 14–15, 43–44, 56, 99, 157, 161, 167–169, 222–223, 235–237, 316–317 cultural 119, 161, 230, 264 legal 102, 105, 209, 214 natural 17, 117, 121, 131 political 223–225, 237 psychic 234–235 theory 50, 161, 168, 209–210, 219 value 56, 215 vocabularies 2, 17, 58–60, 222–223, 231, 234, 238 Taylor 239, 302, 304, 319 tensions 3, 16, 168, 216, 238, 271–272, 279, 281, 304 terms 3, 8, 9, 12, 27, 53–54, 63, 65, 125, 173, 207, 210, 217, 257–258, 268 economic 243–244, 262 general 3, 9, 16, 265, 268

scientific 68 theoretical 13, 88 theoretical constructionism 21–22, 60–61, 239 constructions 1, 4–6, 14, 51, 211, 306 eye 10–13 language 88, 216 perspectives 11, 98, 145, 264, 291 tools 1, 6, 59–60, 257 traditions 6, 209, 262, 302, 305 vocabularies 2, 17, 19, 20, 209, 216 theorize 73–74, 121, 290 theory 12, 21, 28, 35–36, 38, 60–61, 76–77, 88–89, 110, 149, 245, 315 construction 230–232, 302, 312, 314 cultural 171, 200 formal 209 labor 253 political 283, 302, 317–318 Thomas 115, 192, 203–04 Thompson 72, 78, 177, 181–182, 188 ties 75, 177–180, 272, 305 totality, social 158, 175 traditions 2–4, 9–11, 81–82, 86, 102–104, 143, 293, 306 functional 54–56, 58 umbrella concept, theoretical 46 universes 98, 100, 109, 146 use value 141–143, 246 utility theory 146 value infusion 202 values 30, 98–99, 101–103, 105–108, 111, 146, 164, 202, 253, 316 exchange 143 values and beliefs 102, 110, 191 variable analysis 33 variable and concept 32 variables 19, 24, 27, 28, 29–35, 37, 82, 89, 130, 266, 293 control 34, 293–294 violator 53–54 vocabularies 2–3, 8, 10, 14–15, 17–18, 39, 65, 122, 217, 222, 231, 233 Walsh 70–74 Weber 48, 59, 74, 76, 102, 136, 142, 154, 173–174

330 welfare state 197 Whewell 64–70, 72–76 whiteness 286, 298 Willke 225–226, 228 women 86, 220, 264–267, 269, 271–274, 277–278, 282–283 workers 98, 103, 114, 243, 246, 252–258, 262

Index world 85–86, 104, 133–134, 138–140, 154–156, 268, 277–278, 296 culture 193, 201 empirical 83–85, 294 external 154–156, 234 polity 192, 197 Yeung 34, 290, 293