Human agents and social structures

Peter J. Martin, Alex Dennis

• Publisher: Manchester University Press
• Year: 2013
• ISBN:
  9781847794994
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: Political Freedom
• Index: Politics & International Relations

This is a deliberately polemical intervention into the structure/agency debate in the social sciences. It argues that central concepts in this debate – such as 'society' and the 'individual' – have been widely misconceived, and that progress in the social sciences will only occur if the real nature of the social world is respected.

PART ONE
General issues

1
Introduction: the opposition of
structure and agency

Peter J. Martin and Alex Dennis

Now, social science generally is still [at the] stage of being able to consider only the very large and clearly visible social structures and of trying to produce insight from these into social life in its totality. States and trade unions, priesthoods and family forms, guild and family structures, class formation and the industrial division of labour – these and similar organs and systems appear to constitute society and to fill out the domain of the science of society. In fact, however, these are already structures of a higher order, in which … the real concrete life of sociated individuals is crystallized … The real life of society, provided in experience, could certainly not be constructed from these large objectivized structures that constitute the traditional objects of social science. (Georg Simmel, 1997[1907]: 109–110)

There are in fact no masses; there are only ways of seeing people as masses … we mass them, and interpret them, according to some convenient formula. Within its terms, the formula will hold. Yet it is the formula, not the mass, which it is our real business to examine. (Raymond Williams, 1990[1958]: 300)

Of all the ‘turns’ taken by theory, by far the most frequent one is to turn away from the world. (D. R. Watson, 1997: viii)

The quotations above should already have provided a strong sense of what this book is all about. In the opening paragraphs of the essay from which Simmel’s remarks are drawn, he argued (as he did elsewhere) that ‘seemingly insignificant everyday mundane interactions are constitutive for sociation, society and culture’ (Frisby, 1997: 9) and contrasted ‘clearly visible social structures’ with the ‘experience’ of ‘sociated individuals’. Indeed, the apparent opposition between what has been called the ‘objective reality of institutions’ (Berger and Luckmann, 1991[1966]: 78) and the apparently subjective experience of individual human beings has given rise to a basic tension within sociological thought. Simmel’s remarks, it should be noted, are now more than a century old, yet it would seem that little progress has been made in reconciling the ‘objective’ and ‘subjective’ aspects of social life, with the concept of ‘society’ still seen in opposition to that of the ‘individual’ and social ‘structure’ opposed to human ‘agency’. Although the contributors to this book represent a variety of sociological positions (and we have not sought to impose homogeneity upon them), one thing that they do share is the belief that this tension has generated a succession of dualistic views of social life which are fundamentally misguided and a set of binary oppositions which are ultimately irreconcilable.

It has been claimed that the origins of the ‘new vocabulary of “structure and agency”’ can be dated precisely, with the publication of Giddens’ Central Problems in Social Theory in 1979 (Varela, 2007: 201). Yet, as we have already suggested, the underlying issues have a lengthy history. By the 1970s, Parsons’ notions of ‘system and voluntarism’ were well established, so Giddens’ new terms were in a sense redundant, since both oppositions referred to ‘the same theme of social determinism and individual freedom: the problem of Durkheimian social system/structure and human voluntarism/agency’ (ibid.). In the 1960s this ‘problem’ had been seen in terms of ‘two levels of social reality’ – with ‘social structure or system’ opposed by ‘social action and interaction’ (Cohen, 1968: 236). And, as Martin argues in Chapter 3, the issue is apparent in Durkheim’s and Weber’s early attempts to formulate distinctively sociological approaches. The result of all this is that ‘two theoretical perspectives which are generally regarded as antithetical’ have developed, one in which human action is held to be ‘determined by macro-level social structures’ and another in which the individual is seen as ‘consciously deciding about and governing his or her surroundings at the micro-level’ (Kirchberg, 2007: 115; see also Lukes, 1975: 19–20, n76).

Indeed, the development of these contrasting perspectives has become something of an orthodoxy for those reviewing the history of sociological thought in the twentieth century:

when the one-dimensionality of Parsonian structural functionalism (and related objectivist positions such as structuralism, Marxism, etc.) became increasingly manifest in the late sixties, a microreaction ensued which, by the end of the seventies, had shifted the metatheoretical balance to the other extreme of one-dimensional subjectivism, represented (for pedagogical reasons) by Schütz, Blumer, Garfinkel and others. It is only when the limitations of both objectivism and subjectivism were underscored that the possibility of a synthetic micro-macro link eventually emerged in the eighties. (Vandenberghe, 1999: 48)

It will become clear below that (whatever one thinks of the above characterisation of functionalism, structuralism or Marxism) we believe the portrayal of Schütz, Blumer or Garfinkel as espousing ‘extreme one-dimensional subjectivism’ to be a profound misrepresentation of their works – a misrepresentation which has, however, become tediously familiar. Rather, our contention is that the issues involved are wholly misconceived if represented in terms of such dichotomies as objective versus subjective, structure versus agency, macro versus micro, holism versus individualism, and so on, and hence cannot ever be dealt with by trying to effect some sort of reconciliation between the opposed elements, or – for example – to identify a ‘linkage’ between the ‘macro’ and the ‘micro’ (Alexander and Giesen, 1987: 1ff) so as to create a Great Unified Theory for sociology. Similarly, to place analytical emphasis on either the individual or society is to miss the point: it is true, as Durkheim put it, that in an important sense society is prior to the individual – yet at the same time no one was more aware than he that the social order is above all a moral order, depending on the beliefs and normative commitments of real people (Lukes, 1975: 227; King, 2007: 211). In short, the proper focus of sociological attention must be the achievement and maintenance of what Schütz called ‘the realm of intersubjectivity. The world … experienced by the individual as shared by his [sic] fellow creatures, in short as a social world’ (Schütz, 1972[1932]: 139). It is this realm, we suggest, that Blumer saw as ‘human group life’ (1969: 77) and which one of our contributors, Richard Jenkins, simply calls the ‘human world’ (2002a: 4). The reality of normal social life, as another contributor, Anthony King, puts it, is that people are ‘embedded in social relations with others’ (2006: 470).

The idea of social structure

Yet there are some who still wish to hold on to the idea that there must be social ‘structures’, independent of actual people and constraining them. Such a conception of structural effects underlies, for example, Bhaskar’s ‘critical realism’:

The relations into which people enter pre-exist the individuals who enter into them, and those whose activity reproduces or transforms them; so they are themselves structures. And it is to these structures of social relations that realism directs our attention – both as the explanatory key to understanding social events and trends and as the focus of social activity aimed at the emancipation of the exploited and oppressed. (Bhaskar, 1989: 4, emphasis added)

This extract raises a number of issues; we will take up four of these. First, what would the reproduction or transformation of ‘structures’ actually be like? Indeed, is the distinction a valid one? How would an actor or analyst know which was occurring? This would require a complete knowledge of all the consequences of a given action. Such matters are pursued further by Sharrock in Chapter 7; for the present the essential point is that all social activity makes a difference – change is constant, and social life must be understood in terms of process rather than ‘structures’. As Becker has put it:

There is no simple way to sum up all the changes and decide that so much change, but no less, is revolutionary. Nor is there any good reason to make the distinction so clear-cut … We need not make these distinctions definitively, since our interest is in the growth and decay of forms of collective action rather than in the development of logical typologies. (Becker, 1982: 308–310)

Second, it does not follow that because people enter into sets of social relations that precede them, these relations therefore constitute ‘structures’. No one can deny that individuals enter a world which is organised in all sorts of ways, but to call these patterns ‘structures’ and then to attribute causal powers to them amounts to reification, or committing the ‘fallacy of misplaced concreteness’ (Whitehead, 1963[1925]: 52ff).

Third, it is difficult to regard these ‘structures’ as the ‘explanatory key to understanding social events and trends’ without lapsing into some form of determinism, in which particular actions are ultimately explained as the result of general social forces.

Fourth, we are also dubious about the notion that ‘critical realism’ can contribute to the ‘emancipation of the exploited and oppressed’, and we have tried in vain to identify a single instance in which this has occurred. Lacking this, we believe the claim to be wishful thinking at best and at worst an excuse for political quiescence.

A similar ‘critical realist’ position has been developed by Margaret Archer. In response to the criticism (made by Anthony King) that ‘appeals to autonomous structures are simply errors of description’ (King, 1999: 213), Archer reiterated her view that ‘we are … talking about structural properties and powers which defy reduction and are about causal influence which no re-description can eliminate from explanatory accounts’ (Archer, 2000b: 466). Thus structures are held to ‘exert causal influences upon subsequent interactions by shaping the situations in which later generations of people find themselves … Like Bhaskar, Archer denies that the causal powers of macrostructures could be explained in terms of interwoven individual acts, done by individual people’ (Kiniven and Piiroinen, 2006a: 226).

Archer’s position, then, raises the same issues concerning reification and determinism as that of Bhaskar. In addition, like Sawyer (2001: 569), we first question the validity of proposing that patterns of social organisation give rise to ‘emergent properties’ with causal powers. Secondly, we are not persuaded that Archer’s position is either useful in the conduct of empirical research or ultimately coherent. She has argued that ‘realist social theory tackles the structure/agency problem from a position of analytical dualism … there is never a moment at which both structure and agency are not jointly in play’ (2000: 465). The dualism is to be regarded purely as an analytical and not a ‘philosophical’ one. But set against the ‘reality’ (ibid.: 464) which Archer claims for ‘structure’ – and its ‘emergent properties’ – this would seem to undermine the coherence of her position. ‘Structures’, she writes, ‘are ever relational emergents and never reified entities existing without social interaction’ (ibid.: 465). Just what is being claimed here? Thirdly, we take the view that, as Kiniven and Piiroinen (2006a and b) have argued, it is not possible to produce an objective (external, exhaustive) description of social ‘structures’, or ‘society’ (or anything else, for that matter), other than through the specific assumptions, concepts and ideas of the ‘observer’ (for a recent wide-ranging discussion of this, see Frayn, 2006). On this view, knowledge is irredeemably perspectival, and the project of ‘critical realism’ is epistemologically flawed.

The present book is not, therefore, intended as an even-handed review of a long-running controversy about the relative merits of ‘structure’ and ‘agency’ perspectives; rather, its aim is to be an intervention which seeks to make the case that structural, system, or holistic approaches to the understanding of social life and the explanation of human action are fundamentally misconceived – as, equally, are efforts which rest on individualistic assumptions. In essence, our view is that human social life is conducted in and through patterns of collaborative interaction: sociologically, our interest is thus not in the subjectivity of individuals but in the ways in which inter subjectivity is achieved and maintained.

As we have already said, the contributors to this book display a variety of orientations; however, we believe that there are a least four propositions with which they would all agree:

First, that attempts to specify what is meant by the concept of structure in sociology (and the other human sciences) have led to a bewildering variety of definitions. As Rubenstein has put it, ‘There is a striking degree of obscurity and confusion over a concept that is widely regarded as defining the sociological perspective … Definitions of structure range from a concreteness that precludes generality … to an abstractness that threatens vacuity’ (Rubenstein, 2001: 2). However, for Rubenstein there is much more agreement ‘on the type of concept sociologists are looking for. Structure is usually conceived as “external” and “objective” features of social order that are thought to have controlling power over culture and action’ (ibid.: 3).

Second, that the collective concepts (such as family, state, organisation, class and so on) – which have often been seen as fundamental to sociological analysis – have often encouraged ‘the temptation to reify collective aspects of human life’ (Jenkins, 2002a: 4); that is, to treat them as if they were real entities, independent of the human beings who constitute them.

Third, that the explanation of human action in terms of ‘structural’ phenomena ultimately, and inevitably, leads to determinism, in which conduct is held to be ‘caused’ by factors or forces independent of real people.

Fourth, that ‘structural’ explanations of social life generally rest on partial, or selective, or simply misguided versions of what have come to be called ‘micro-sociological’ perspectives (see below, pp. 9–13). Indeed when this point is pursued, the structure-agency opposition is revealed as false, one of several misleading dualisms that have proved obstacles to the productive development of sociological thought. In the present context, our contention is that only if the concept of social structure (and its associated collective entities) are presupposed can the distinctions between ‘structure’ and ‘agency’, or between ‘macro’ and ‘micro’ phenomena, be sustained. Indeed, Maines has suggested that the opposition between ‘macro’ and ‘micro’ is ‘a cropolitic distinction that does more harm than good for the understanding of social life’ (1982: 274). (For those unfamiliar with the jargon of paleontology, ‘cropolitic’ may be rendered loosely – but not inaccurately – as referring to ‘fossilised bullshit’).

The tenacity of structural explanations

Given the highly problematic nature of the concept of social structure, it is appropriate to consider why ‘structural’ explanations have exerted such a tenacious grip on sociological thought. One important reason is the longstanding commitment to the idea of sociology as – perhaps above all else – a critique of individualistic explanations. For Durkheim, indeed, this was at the heart of sociological thinking: as Lukes has put it: ‘Durkheim’s central interest was in the ways in which social and cultural factors influence, indeed largely constitute, individuals (1975: 13; see also Nisbet, 1970[1966]: 7–9). Related to this is the idea that sociological concepts must be essentially collective. As Maines has put it:

In competing for intellectual space, it was in sociology’s vested interest to trade on a social factorist formulation in which group properties were seen as exerting causal influence on some form of behaviour. In its simplest expression, the dominant proposition adopted by the field was that ‘social structures cause human action’. (Maines, 2001: 6)

Apart from purely intellectual reasons, we should also note the understandable tendency for sociologists (and academics more generally) to identify with perspectives which seemed to demonstrate – as Durkheim sought to do – that the ills of the world were not necessarily the result of the force of ‘evil’, or the machinations of malevolent individuals, but were systematically generated by the routine workings of societies themselves. Thus, for example, Collier argues in his exposition of Critical Realism that this perspective can be ‘transformative and potentially emancipatory’ since it ‘recognises that states of affairs are brought about by the workings of relatively enduring structures’. Consequently, it ‘directs the attention of people who want to make the world a better place to the task of transforming these structures’ (Collier, 1994: 15–16). (As we noted above, however, this argument would be more impressive if there was any evidence that ‘realism’ in the social sciences or humanities has done anything to ‘make the world a better place’).

For both intellectual and political reasons, then, ‘structural’ perspectives have remained dominant in sociology, while the various ‘microsociologies’ have been both traduced (Maines, 1978: 491; 2001: 249) and criticised for their alleged inability to deal with big real-world issues such as conflict and power. Thus, to take a recent example, Kirchberg suggests that ‘the “subjectivity” of existentialism, phenomenology and symbolic interactionism neglects undeniable forces like political power structures’ (Kirchberg, 2007: 118; but on this see Dennis and Martin, 2005: 192–194, or Denzin, 1992: 56–63). So it is to the so-called ‘microsociologies’ that we now turn.

‘Micro’ sociological perspectives

It is worth pausing to consider three distinctively sociological approaches which have – unjustifiably, in our view – been criticised for their alleged inability to cope with big institutional issues like power, conflict and social inequalities. These are symbolic interactionism, phenomenology and ethno-methodology: all are included, for example, in Collins’ discussion of ‘The microinteractionist tradition’ (1994: 242ff). (We are aware that rational action theory, some versions of exchange theory and similar approaches have also been described as ‘micro’ sociology, but since these rest on essentially individualistic presuppositions we will not consider them here). Indeed, Collins’ use of the term ‘microinteractionist’ is an illustration of the extent to which the macro–micro binary pervades the sociological discourse. Similarly, we are not persuaded that this opposition is transcended in Maines’ (1982) concept of ‘mesostructure’ – partly because the prefix ‘meso’ implies an ‘above’ (macro) and a ‘below’ (micro), and partly because the notion of ‘structure’ is retained.

Symbolic interactionism

The figure generally acknowledged as providing the intellectual foundations of symbolic interactionism was G. H. Mead. He, however, was ‘primarily a philosopher. He differed from the bulk of philosophers in believing that the cardinal problems of philosophy arose in the realm of human group life and not in a separate realm of an individual thinker and his [sic] universe’ (Blumer, 1981: 902). In the present context, it is this emphasis on ‘human group life’, and Mead’s rejection of individualistic assumptions, which is significant. Blumer continues: ‘In the case of the “social act”, Mead gives us a picture of human group life that exists in the form of conscious ongoing activity … This process is just not caught by the customary sociological concepts of culture, structure, values, norms, status positions, social roles, or institutions’ (ibid.: 903). We can already make three observations about Mead’s thought and its influence on Blumer (who coined the term ‘symbolic interactionism’ in 1937). First, the perspective takes social life to be essentially collective, with a focus on ‘human group life’ in real situations rather than on the unconstrained actions of individuals. Second, both Mead and Blumer respect the empirical reality of social life as process rather than structure. Third, and arising from this, ‘customary sociological concepts’ are found wanting – as abstractions, they cannot provide causal explanations for social actions other than by assuming that the latter are determined by ‘forces’ external to individuals. So Blumer emphasised that Mead’s thought ‘points to a different form of analysis’ (ibid.: 904) from that which has become conventional in sociology.

Thus Blumer ‘viewed symbolic interaction as the essential process through which all social phenomena (including structures) are created, maintained, and changed’ (Morrione, 2004: xi). Perhaps inevitably, his focus on ‘both action and the creation of meaning in an ever-emergent situated present’ (ibid.: xiii, emphasis in original), led to accusations of subjectivism and psychological reductionism, positions which Blumer himself found simply untenable (ibid.: xi). On the contrary, at the heart of Blumer’s thinking is the notion of human action as essentially collective, and he ‘represented persistent patterns, enduri...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Contents Page
• Notes on contributors
• Part One – General Issues
• Part Two – Recent Social Theorists
• Part Three – After the Debate
• Bibliography
• Index