Realism and Social Science

Andrew Sayer

• Publisher: SAGE Publications Ltd
• Year: 1999
• Print ISBN: 9780761961246, 9780761961239
• eBook ISBN: 9781446236055
• Edition: 1
• Topic: Social Sciences
• Subtopic: Sociology
• Index: Social Sciences

Realism and Social Science offers the reader an authoritative and compelling guide to critical realism and its implications for social theory and for the practice of social science. It offers an alternative both to approaches which are overly confident about the possibility of a successful social science and those which are defeatist about any possibility of progress in understanding the social world. Written by one of the leading social theorists in the field, it demonstrates the virtues of critical realism for theory and empirical research in social science, and provides a critical engagement with leading non-realist approaches.

PART I

INTRODUCING CRITICAL REALISM

---

Introduction

Realism – or at least the ‘critical realism’ that I want to defend – is not what many people think it is. Many suppose that realism claims a privileged access to the Truth and thus involves a kind of ‘foundationalism’. But such claims are inconsistent with realism, for if the defining feature of realism is the belief that there is a world existing independently of our knowledge of it, then that independence of objects from knowledge immediately undermines any complacent assumptions about the relation between them and renders it problematic. What reason have we for accepting this basic realist proposition of the mind-independence of the world? I would argue that it is the evident fallibility of our knowledge – the experience of getting things wrong, of having our expectations confounded, and of crashing into things – that justifies us in believing that the world exists regardless of what we happen to think about it. If, by contrast, the world itself was a product or construction of our knowledge, then our knowledge would surely be infallible, for how could we ever be mistaken about anything? How could it be said that things were not as we supposed? Realism is therefore necessarily a fallibilist philosophy and one which must be wary of simple correspondence concepts of truth. It must acknowledge that the world can only be known under particular descriptions, in terms of available discourses, though it does not follow from this that no description or explanation is better than any other.

In starting this book in this way, turning the tables on anti-realists, I am of course getting my retaliation in first, because I am aware that in certain quarters, ‘realism’ is synonymous with a form of naive objectivism, claiming unmediated access to the Truth. This misconception prevents realism getting a hearing. At the same time, I am also wary of naive supporters of realism who assume that it will indeed guarantee the production of true knowledge, when the independence of the world from our knowledge and the entrapment of knowledge within discourse imply the impossibility of any such guarantees.

Yet once such misconceptions are removed, I believe it can be shown that realism (particularly the critical realism pioneered by Roy Bhaskar) offers great promise for social science and theory. Critical realism provides an alternative to several philosophical and methodological positions which have been found wanting. Firstly, in the philosophy of natural science, realism offered a third way between empiricism and positivism on the one hand and the relativism that followed in the wake of Kuhn and Feyerabend’s assault on conceptions of science as a cumulative foundationalist enterprise on the other. Secondly, in the philosophy and methodology of social science, critical realism provides an alternative to both hopes of a law-finding science of society modelled on natural science methodology and the anti-naturalist or interpretivist reductions of social science to the interpretation of meaning. By simultaneously challenging common conceptions of both natural and social science, particularly as regards causation, critical realism proposes a way of combining a modified naturalism with a recognition of the necessity of interpretive understanding of meaning in social life. For realists, social science is neither nomothetic (that is, law-seeking) nor idiographic (concerned with documenting the unique). Thirdly, with respect to debates around modernism and postmodernism, it opposes the reductionism and closure of some overly confident ‘modernist’ kinds of social science, evident in the determinism and flattening of difference common in some versions of grand narratives, and in the a priorism of neoclassical economics. Such approaches radically underestimate the openness, contingency and contextually variable character of social change. On the other hand, it rejects a defeatist strain of postmodernism which assumes that the absence of certainty, regularity and closure, means that hopes of reliable knowledge claims and scientific progress must be rejected (Stones, 1996). Accordingly, critical realism seeks to avoid both scientism and ‘science-envy’ on the one hand and radical rejections of science on the other.¹

The chapters of this book are offered as realist contributions to debates on social theory and the philosophy and methodology of social science that have been prominent in the last two decades. While they are largely an attempt to apply and develop critical realist angles, at certain points I shall make criticisms of critical realism itself, particularly with regard to its account of critical social science. Although I have attempted to adopt a critical realist approach to my substantive work in social science – mainly in industrial and urban and regional studies and political economy (Sayer, 1995; Morgan and Sayer, 1988; Sayer and Walker, 1992) – I have not included any of this material in this collection, apart from drawing some examples from it in Chapter 1.

Insofar as this is a book of critiques, I should perhaps point out that I am more interested in ideas than who might have authored them and more interested in their evaluation than in their history. By and large I therefore avoid questions of how particular authors are to be interpreted. In many cases the ideas are not drawn directly from major theorists but are ones I have frequently encountered in discussions. In addition, as is common in philosophy, what follows may even include some lines of argument over which no-one claims authorship, but which nevertheless are interesting possibilities in themselves. One of the main sources has been discussions with research students in social science, especially those I have taught on courses in the philosophy of social science and social theory. I realize that engaging directly with the most prestigious authors would bring me more cultural capital, but they have had plenty of attention already, and utility is more important than prestige, even where its exchange value is lower.

A recent minor theme of debate in philosophy and social theory has concerned the role of logic and metaphor and rhetoric in science and philosophy (for example, Nelson et al., 1987; Norris, 1997; Mäki, 1993). Although these elements are often presented as opposed, critical realism takes a both/and rather than an either/or position regarding them. Both lay and scientific thought requires not only logic but metaphors and associational thinking. Scientific and philosophical discourses are rhetorical in the broad sense of involving persuasion, but that does not necessarily cancel out their dependence on logic and reduce them to a form of linguistic arm-twisting. Philosophy proceeds by making connections and distinctions. In the essays that follow, I, of course, deploy metaphors, rhetoric and associational thinking but I also try to work out what entails what, what is a non-sequitur, which conditions are necessary, which sufficient; more simply I attempt to distinguish rigorously between can and must, all and some, often and always. Following a philosophical argument is like negotiating a complex, twisting route through dense networks of streets. There are many opportunities for wrong turnings – unjustified inferences. One can take a wrong turning just by misreading ‘can’ as ‘must’, or ‘some’ as ‘all’. Distinctions, especially in the form of dualisms or binaries, are regular targets of scepticism today, but whether they are good or bad is an a posteriori matter, and we can hardly avoid interrogating them logically – in terms of what they entail and don’t entail – as well as in terms of their associations and metaphorical qualities. Simply to note the presence of binaries or dualisms does not constitute an argument until one explains what is problematic about the instances in question. Similarly, that a distinction can be deconstructed does not necessarily totally undermine it; it may just elaborate and qualify it. That a distinction is fuzzy is not necessarily fatal. Some of our most useful distinctions – like that between night and day – cannot be drawn sharply, but most of the time we have little difficulty with them. In other words my main aim in these essays has been to argue, to get beyond a sound-bite approach; if they are reciprocated with counter arguments I shall be happy.

It may help to situate the essays which follow if I offer a brief personal interpretation of the context in which critical realism has developed. Inevitably it will reflect my own situation within Britain. It is not intended as an overview; others are likely to have had different experiences.

The institutional context in which I began my first research was the University of Sussex, whose commitment to interdisciplinary studies forced me out of my early disciplinary parochialism in human geography and allowed me also to teach the philosophy of social science. This gave me free rein to explore the aftermath of the major debates in the philosophy of science of the 1960s and 70s – the conventionalist assaults of authors, such as Kuhn and Feyerabend, on empiricist philosophy of science, and the anti-naturalist approaches of interpretivism or hermeneutics in the philosophy of social science. This was also the context in which critical realism first developed in Britain in the mid-1970s, where the work of Rom Harré and particularly Roy Bhaskar, and others such as Russell Keat and Ted Benton, offered an alternative to empiricism and conventionalism in the philosophy of natural science, and to positivism and interpretivism in the philosophy of social science.

I started out with critiques of positivism, especially its expectation that the social world could be shown to be a composite of a number of behavioural regularities which would eventually be described by social laws akin to those of natural science. The empirical context was the prosaic one of studies of the development of urban and regional systems (Sayer, 1976). In attempting to develop an understanding of these that was both dynamic and spatial, it slowly dawned on me that social systems were necessarily open, and that they evolved rather than equilibrated, not least because people have the capacity to learn and change their behaviour. Consequently I realized the goal of finding rough regularities, let alone laws, to describe social systems, was a pipe dream. At the same time, realist philosophy was beginning to challenge the regularity or successionist theory of causation, and to analyse the explanation of change in open systems, so that it became clear that abandoning hopes of finding regularities in no way meant abandoning explanation.

From the late 1960s through to the early 1980s, the rise of the new Left revived some distinctly grand and modernist theories of capitalist societies. The world was to be seen in terms of grand structures while pluralism was associated with the much-despised liberalism, unable to see the structural wood for the interest-group trees, a tendency now – amazingly – inverted, with pluralism being associated with the intellectual avant garde and structuralism being seen as passé (McLennan, 1996). Particularly in its Althusserian form, structuralism exuded an extraordinary scientistic arrogance, later deservedly punctured by E.P. Thompson’s withering attack in his unfortunately-titled The Poverty of Theory (Thompson, 1979). Much was excluded by those ostensibly all-embracing, all-explaining discourses –notably gender, race, sexuality and much of lived experience; here the rise of feminism, anti-racism and post-colonialism have challenged the old new Left with devastating effect.

There were also within the Marxist Left further reactions against homogenizing and reductionist tendencies, notably from Raymond Williams, of whose work nuance, complexity and sensitivity to local, lived experience were hallmarks. Many of those – myself included – who tried to apply Marxist theory empirically found that in confronting the social world in its concrete, that is, many-sided, forms, they had to develop more open, context-dependent and plural accounts, within which Marxism might have been an important ingredient but no longer a totalizing theory.² In urban and regional studies and industrial studies, this was associated with an increasing concern with differentiation, most strikingly evident in Doreen Massey’s influential Spatial Divisions of Labour (Massey, 1984). This insisted on the enormously differentiated ways in which capitalist uneven development works out in practice and was critical of the reductionism of what later became branded as ‘grand narratives’, although ‘grand analyses’ would have been a better term since they generally neglected to tell stories and instead just absorbed empirical material into their pre-existing categories and frameworks (Sayer, 1981a). The focus on differentiation and pluralism might now sound familiar in relation to postmodernism and post-structuralism, but for many of those involved in this research, all this happened largely independently of their emergence. In some ways, critical realism, with its focus on necessity and contingency rather than regularity, on open rather than closed systems, on the ways in which causal processes could produce quite different results in different contexts, fitted comfortably with these developments. Realists expect concrete open systems and discourses to be much more messy and ambiguous than our theories of them and do not consider that differentiation poses a threat to social science.

Many reacted to reductionist accounts by shifting to middle range theory and empirical studies, for example from Marxist theories of accumulation to analyses of the institutional forms present in particular capitalist societies (Sayer, 1985).³ This meant a greater openness to diverse empirical and theoretical influences, but what was and remains ambiguous about these middle range theories is whether they were intended to leave behind more abstract social theories or whether they were meant to build on them even if they don’t acknowledge them (Sayer, 1995).

Another significant current within social studies which was critical of modernist social science prior to the rise of postmodernism and the turn to discourse was hermeneutics or interpretivism. These had developed a sophisticated philosophical critique of naturalism – the doctrine that the social world could be understood in the same way as natural science – and had theorized about the interpretation of texts and ‘constitutive meanings’ (for example, Schutz, Taylor, Winch, Ricouer). These represented a different source of criticism of positivism from that associated with Kuhn and Feyerabend.⁴ At the same time there was a largely sympathetic critique of interpretivism’s tendency to reduce social life wholly to the level of meaning, ignoring material change and what happens to people, regardless of their understandings (for example, Fay, 1975; Giddens, 1976). Critical realists argued that while interpretative understanding was an important and necessary feature of any social science, it did not mean that there was no scope for causal explanation.

In my neck of the academic woods, and probably many others too, all this happened before ‘postmodernism’ began to be discussed. This arrived mainly from across the Atlantic, out of architecture and the new dialogue between literary studies and social theory. This exchange, along with the more general turn to language and discourse, was useful in exploring the similarities and relations between literature and social science. Language could not continue to be taken as transparent and unproblematic by the philosophy of science; discourse and textuality needed to be taken seriously.

Postmodernism has also encouraged a more critical view of key categories of social thought, especially the ways in which binary distinctions or dualisms typically obscure connections, hierarchy and differences, and apparently comprehensive syntheses suppress the experience of certain groups, while concealing the identity of those whose particularistic stories they actually do express. Fundamental philosophical issues have been raised regarding the character of discourse, the limits of reason, and the question of truth. What the implications are of such developments for critical realism and social science is controversial:⁵ for the ‘defeatist postmodernists’ they imply relativism, idealism and a rejection of the ambitions of social science; for others they point to a renewed social science which is conceptually cautious and more reflexive about both its implicit philosophy and methodology and its social and political coordinates. My sympathies lie with the latter view.

Differentiation is not the same as difference, interpretivism is not the same as discourse analysis, liberal pluralism is not the same as postmodern pluralism. These apparently similar concepts have different provenances and associations, but the parallels between them are intriguing. What is important to note, especially given the preoccupation in social science with novelty, is that some important critiques of modernism in social science predated postmodernism and did not share its predominantly idealist character.

If I have mixed feelings about postmodernism, there is one ‘post’ I feel committed to unreservedly – ‘post-disciplinary studies’; indeed, since the late 1970s I have identified with social science rather than with a particular discipline. Disciplinary parochialism, and its close relative disciplinary imperialism, are a recipe for reductionism, blinkered interpretations, and misattributions of causality. The essays which follow reflect this post-disciplinary outlook. Several of the chapters have appeared before in journals. For the most part I have left these largely unaltered and, with the exception of changes n...

Table of contents

• Cover Page
• Title
• Copyright
• Contents
• Part I Introducing Critical Realism
• Part II Postmodern-Realist Encounters
• Part III Social Science and Space
• Part IV Critical Realism: From Critique to Normative Theory
• References
• Index