political science research paper

ANTHONY GIDDENS

A Contemporary Critique ALEX CALLINICOS

By way of apology for the title of this article, I should point out t not merely to Anthony Giddens's recent A Contemporary Critiq ical Materialism, but also to the fact that I am discussing an en energetic and prolific writer responsible over the last few years gious outpouring of books that develop, and sometimes modify set of views on social theory. To attempt to summarize, an criticism such an oeuvre is a little like trying to catch quicksilv

My reasons for nevertheless making the attempt are twofold. Fir erudition and intellectual range of Giddens's work commands ou A writer who draws on such a wide variety of thinkers - M Durkheim, Mead, Heidegger, Wittgenstein, Schutz, Foucault name but a few - to develop his own distinctive theory dese consideration especially in the light of the deeply parochial natur British intellectual culture. The two books under review, one a reworking of social theory, the other a collection of essays that i reworking mainly through the criticism of others' views, offer opportunity to assess the direction and coherence of Giddens's

Secondly, A Contemporary Critique of Historical Materialism, suggests, itself expounds Giddens's substantive theory in the f appraisal of Marxism. The criticisms it offers represent a direct those, such as myself, of a Marxist persuasion, a challenge to w obliged to respond. But, in doing so, I am fully in agreement wi Wright that "the book is a serious engagement with Marxism, a serious reading by both Marxists and non-Marxists."' Giddens i gent and sympathetic a critic of Marx simply to be brushed as

Department of Politics, University of York.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

133

All use subject to https://about.jstor.org/terms

134

What, then, is the nub of Giddens's critique of Marxism? Brutally summar- ized, it is that Marx himself, in the bulk of his mature writings, and, much more so, his successors adopted a one-sided and economically reductive account of human conduct. They consequently developed an evolutionary theory of historical development that explained social change in terms of the growth of the productive forces, denying human agents any significant role in the reproduction and transformation of societies. As a result, Marxism has been largely insensitive to the diversity and irreducibility of power-rela- tions, blindly believing that national, racial, and sexual oppression were merely the epiphenomena of class exploitation. The hopeless inadequacy of such an intellectual framework has been decisively proved in recent years with the evolution of"really existing socialism" along lines quite unanticipat- ed by Marx, and the emergence of autonomous social movements such as feminism.

To Giddens's views thus baldly stated, one might respond, "So what else is new?" An alternative conceptualization of society to Marxism that stresses the variety and permanency of power-relations is implicit in Nietzsche's thought, was developed systematically by Weber, and has recently been revived with a fashionably "post-structuralist" twist by Foucault and

Deleuze, and rather more accessibly, by Frank Parkin in his Weberian critique of neo-Marxist class theory.2 Is not Giddens's work merely a varia- tion, no doubt erudite and sophisticated, but nevertheless still a variation on a rather well-worn theme?

Giddens thinks not. In the 1979 postscript added to the second edition of his The Class Structure of the Advanced Societies, first published in 1973, Giddens rejects descriptions of that book as "Weberian" or"neo-Weberian," insisting that "the position I developed in the book owes considerably more

to Marx than to Weber."3 He furthermore subjects Parkin's Marxism and Class Theory to virulent criticism for its negative and dismissive attitude towards Marxism.4 And he explicitly declares, "I do not seek to replace Marx by Nietzsche."5

Giddens's work can best be represented as a very ambitious attempt to effect a synthesis of distinct and opposed traditions in social theory. Thus, at the methodological level he seeks to preserve the stress laid by "interpretive sociology" (Weber, Schutz, ethnomethodology) on social relations as the result of the intentional activity of human agents, while at the same time acknowledging the importance, asserted strongly by Marxists, Parsonian sociologists, and structuralists alike, of social structures in shaping human conduct. Again, substantively, Giddens is concerned to evolve an account of

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

society that accords primacy to the phenomenon of power, maintaining that "power was never satisfactorily theorized by Marx,"6 while treating Marx's analysis of capitalism as a limited but largely correct account of one case of this general phenomenon.

The attempt to construct syntheses is undoubtedly a worthwhile intellectual activity. Indeed, Hegel's dialectic treats it as the characteristic form of theoretical progress, the concept of Aufhebung expressing how opposites are cancelled and preserved in a new unity. It is not wholly inaccurate to describe historical materialism as such a synthesis of Hegel and Ricardo. But, as this example suggests, genuinely innovative syntheses are rare and difficult to arrive at. Too often attempted syntheses amount merely to banality, inco- herence, or eclecticism.

I am reluctant to ascribe any of these three qualities to Giddens's work, brilliant and stimulating as it often is. But I wish to argue that his attempt to transcend the rival traditions he hopes to synthesize is a failure. I shall try to prove my case by considering two of the main features of Giddens's general approach, his theory of structuration, and account of power. I shall then go on to consider in detail some of his criticisms of Marxism, showing them to be largely mistaken and resting on a highly selective reading of its produc- tions. I shall conclude with some general reflections on the nature of Gid- dens's project. I am well aware that in pursuing this course I shall not pay due tribute to the many exciting and suggestive passages especially in A Contem-

porary Critique in which Giddens discusses a variety of theoretical and historical issues. My failure, however, to match his erudition and range may still lead to gains in clarity.

The Theory of Structuration

A central theme of Giddens's two main methodological treatises, New Rules of Sociological Method, and Central Problems in Social Theory, is what he calls the "dualism of agency and structure."7 By this he means the following. A variety of philosophical traditions, notably heremeneutics and post- Wittgensteinian analytical philosophy, have developed highly sophisticated

accounts of human conduct as intentional action, any explanation of which makes irreducible reference to the intentions, beliefs, and desires of the agent ascertainable primarily through the medium of language. They have, how- ever, "paid little attention to, or have found no way of coping with, concep- tions of structural explanation or social causation."8 On the other hand, those intellectual traditions that have developed such conceptions have ignored the subject. For example,

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

135

All use subject to https://about.jstor.org/terms

136

in both Althusserian Marxism and Parsonian sociology the reproduction of society occurs "behind the backs" of the agents whose conduct constitutes that society. The involvement of actors' own purposive conduct with the rationalization of action is lacking in each case ... hence the teleology of the system either governs ... or supplants ... that of the actors themselves.9

Giddens's aim is to overcome this pervasive dualism of agency and structure through the theory of structuration. His starting point is one very close to the hermeneutic and Wittgensteinian traditions:

The production of society is a skilled performance, sustained and "made to happen" by human beings. It is indeed only made possible because every (competent) member of society is a practical social theorist; in sustaining any sort of encounter he draws upon his knowledge and theories, normally in an unenforced and routine way, and the use of these practical resources is precisely the condition of the production of the encounter at all.'0

"Knowledgeable human agents" do not, however, make society in circum- stances of their own choosing. "The realm of human agency is bounded," "the production of constitution of society is a skilled accomplishment of its members, but one that does not take place under conditions that are wholly intended or wholly comprehended by them."'1

It is here that social structures enter the scene, as the "unacknowledged

conditions and unanticipated consequences" of human conduct. Giddens

distinguishes sharply between these structures, and social systems, by which

he means human collectivities persisting in time and space. The persistence of

a social system can only be accounted for by invoking structure, which is

itself "non-temporal and non-spatial, ... a virtual order of differences pro-

duced and reproduced in social interaction as its medium and outcome."12

Every social system possesses a set of "structuralprinciples," the "structural

elements that are most deeply embedded in the space-time dimensions of

social systems" and that "govern the basic institutional alignments of a

society."'3 Giddens distinguishes between three main dimensions of struc-

ture, signification, in which agents communicate and rationalize their actions

by means of interpretive schemes, domination, arising from asymmetries in

the distribution of resources, and legitimation, through which different

forms of conduct are sanctioned by means of norms. Structural principles,

according to Giddens, typically involve social contradictions, in the sense

that they "operate in terms of each other but at the same time contravene one another."14

I cannot dwell here on the many felicities of Giddens's detailed account of structure - its anti-functionalism, for example, based on the "guiding tenet: don't look for the functions social practices fulfil, look for the contradictions

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

they embody,"'5 or his conceptualization of structures as enabling as ANTHONY GIDDENS

A Contemporary Critique ALEX CALLINICOS

By way of apology for the title of this article, I should point out t not merely to Anthony Giddens's recent A Contemporary Critiq ical Materialism, but also to the fact that I am discussing an en energetic and prolific writer responsible over the last few years gious outpouring of books that develop, and sometimes modify set of views on social theory. To attempt to summarize, an criticism such an oeuvre is a little like trying to catch quicksilv

My reasons for nevertheless making the attempt are twofold. Fir erudition and intellectual range of Giddens's work commands ou A writer who draws on such a wide variety of thinkers - M Durkheim, Mead, Heidegger, Wittgenstein, Schutz, Foucault name but a few - to develop his own distinctive theory dese consideration especially in the light of the deeply parochial natur British intellectual culture. The two books under review, one a reworking of social theory, the other a collection of essays that i reworking mainly through the criticism of others' views, offer opportunity to assess the direction and coherence of Giddens's

Secondly, A Contemporary Critique of Historical Materialism, suggests, itself expounds Giddens's substantive theory in the f appraisal of Marxism. The criticisms it offers represent a direct those, such as myself, of a Marxist persuasion, a challenge to w obliged to respond. But, in doing so, I am fully in agreement wi Wright that "the book is a serious engagement with Marxism, a serious reading by both Marxists and non-Marxists."' Giddens i gent and sympathetic a critic of Marx simply to be brushed as

Department of Politics, University of York.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

133

All use subject to https://about.jstor.org/terms

134

What, then, is the nub of Giddens's critique of Marxism? Brutally summar- ized, it is that Marx himself, in the bulk of his mature writings, and, much more so, his successors adopted a one-sided and economically reductive account of human conduct. They consequently developed an evolutionary theory of historical development that explained social change in terms of the growth of the productive forces, denying human agents any significant role in the reproduction and transformation of societies. As a result, Marxism has been largely insensitive to the diversity and irreducibility of power-rela- tions, blindly believing that national, racial, and sexual oppression were merely the epiphenomena of class exploitation. The hopeless inadequacy of such an intellectual framework has been decisively proved in recent years with the evolution of"really existing socialism" along lines quite unanticipat- ed by Marx, and the emergence of autonomous social movements such as feminism.

To Giddens's views thus baldly stated, one might respond, "So what else is new?" An alternative conceptualization of society to Marxism that stresses the variety and permanency of power-relations is implicit in Nietzsche's thought, was developed systematically by Weber, and has recently been revived with a fashionably "post-structuralist" twist by Foucault and

Deleuze, and rather more accessibly, by Frank Parkin in his Weberian critique of neo-Marxist class theory.2 Is not Giddens's work merely a varia- tion, no doubt erudite and sophisticated, but nevertheless still a variation on a rather well-worn theme?

Giddens thinks not. In the 1979 postscript added to the second edition of his The Class Structure of the Advanced Societies, first published in 1973, Giddens rejects descriptions of that book as "Weberian" or"neo-Weberian," insisting that "the position I developed in the book owes considerably more

to Marx than to Weber."3 He furthermore subjects Parkin's Marxism and Class Theory to virulent criticism for its negative and dismissive attitude towards Marxism.4 And he explicitly declares, "I do not seek to replace Marx by Nietzsche."5

Giddens's work can best be represented as a very ambitious attempt to effect a synthesis of distinct and opposed traditions in social theory. Thus, at the methodological level he seeks to preserve the stress laid by "interpretive sociology" (Weber, Schutz, ethnomethodology) on social relations as the result of the intentional activity of human agents, while at the same time acknowledging the importance, asserted strongly by Marxists, Parsonian sociologists, and structuralists alike, of social structures in shaping human conduct. Again, substantively, Giddens is concerned to evolve an account of

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

society that accords primacy to the phenomenon of power, maintaining that "power was never satisfactorily theorized by Marx,"6 while treating Marx's analysis of capitalism as a limited but largely correct account of one case of this general phenomenon.

The attempt to construct syntheses is undoubtedly a worthwhile intellectual activity. Indeed, Hegel's dialectic treats it as the characteristic form of theoretical progress, the concept of Aufhebung expressing how opposites are cancelled and preserved in a new unity. It is not wholly inaccurate to describe historical materialism as such a synthesis of Hegel and Ricardo. But, as this example suggests, genuinely innovative syntheses are rare and difficult to arrive at. Too often attempted syntheses amount merely to banality, inco- herence, or eclecticism.

I am reluctant to ascribe any of these three qualities to Giddens's work, brilliant and stimulating as it often is. But I wish to argue that his attempt to transcend the rival traditions he hopes to synthesize is a failure. I shall try to prove my case by considering two of the main features of Giddens's general approach, his theory of structuration, and account of power. I shall then go on to consider in detail some of his criticisms of Marxism, showing them to be largely mistaken and resting on a highly selective reading of its produc- tions. I shall conclude with some general reflections on the nature of Gid- dens's project. I am well aware that in pursuing this course I shall not pay due tribute to the many exciting and suggestive passages especially in A Contem-

porary Critique in which Giddens discusses a variety of theoretical and historical issues. My failure, however, to match his erudition and range may still lead to gains in clarity.

The Theory of Structuration

A central theme of Giddens's two main methodological treatises, New Rules of Sociological Method, and Central Problems in Social Theory, is what he calls the "dualism of agency and structure."7 By this he means the following. A variety of philosophical traditions, notably heremeneutics and post- Wittgensteinian analytical philosophy, have developed highly sophisticated

accounts of human conduct as intentional action, any explanation of which makes irreducible reference to the intentions, beliefs, and desires of the agent ascertainable primarily through the medium of language. They have, how- ever, "paid little attention to, or have found no way of coping with, concep- tions of structural explanation or social causation."8 On the other hand, those intellectual traditions that have developed such conceptions have ignored the subject. For example,

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

135

All use subject to https://about.jstor.org/terms

136

in both Althusserian Marxism and Parsonian sociology the reproduction of society occurs "behind the backs" of the agents whose conduct constitutes that society. The involvement of actors' own purposive conduct with the rationalization of action is lacking in each case ... hence the teleology of the system either governs ... or supplants ... that of the actors themselves.9

Giddens's aim is to overcome this pervasive dualism of agency and structure through the theory of structuration. His starting point is one very close to the hermeneutic and Wittgensteinian traditions:

The production of society is a skilled performance, sustained and "made to happen" by human beings. It is indeed only made possible because every (competent) member of society is a practical social theorist; in sustaining any sort of encounter he draws upon his knowledge and theories, normally in an unenforced and routine way, and the use of these practical resources is precisely the condition of the production of the encounter at all.'0

"Knowledgeable human agents" do not, however, make society in circum- stances of their own choosing. "The realm of human agency is bounded," "the production of constitution of society is a skilled accomplishment of its members, but one that does not take place under conditions that are wholly intended or wholly comprehended by them."'1

It is here that social structures enter the scene, as the "unacknowledged

conditions and unanticipated consequences" of human conduct. Giddens

distinguishes sharply between these structures, and social systems, by which

he means human collectivities persisting in time and space. The persistence of

a social system can only be accounted for by invoking structure, which is

itself "non-temporal and non-spatial, ... a virtual order of differences pro-

duced and reproduced in social interaction as its medium and outcome."12

Every social system possesses a set of "structuralprinciples," the "structural

elements that are most deeply embedded in the space-time dimensions of

social systems" and that "govern the basic institutional alignments of a

society."'3 Giddens distinguishes between three main dimensions of struc-

ture, signification, in which agents communicate and rationalize their actions

by means of interpretive schemes, domination, arising from asymmetries in

the distribution of resources, and legitimation, through which different

forms of conduct are sanctioned by means of norms. Structural principles,

according to Giddens, typically involve social contradictions, in the sense

that they "operate in terms of each other but at the same time contravene one another."14

I cannot dwell here on the many felicities of Giddens's detailed account of structure - its anti-functionalism, for example, based on the "guiding tenet: don't look for the functions social practices fulfil, look for the contradictions

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

they embody,"'5 or his conceptualization of structures as enabling as well as constraining, so they do not simply limit the scope of human conduct, but make certain forms of activity possible. More relevant for present purposes is the fact that for Giddens the existence of structures does not in itself explain the persistence of social systems. The belief to the contrary he regards as "the decisive error in functionalism."'6 According to a thesis that Giddens calls "the duality of structure," "structure is both medium and outcome of the reproduction of practices."17 In other words, although structure shapes the conduct of human agents, it is only through that conduct that it itself possesses any effectivity, and it may itself be modified by the activity of which it is the unacknowledged condition and unanticipated consequence.

Giddens illustrates this argument by invoking the case of language. The rules of a language constitute langue, the underlying structures that governparole, speakers' actual usage. But it only by virtue of being applied in speech that these rules become operative; withoutparole langue would not exist. Further- more, it is always possible that usage will alter the rules of language; indeed, language is constantly changing as a result of shifts of usage. Similarly, it is through the conduct of "knowledgeable human agents" that structures are produced, reproduced, and transformed.

The "duality of structure" thus formulated serves to conceptualize social systems as the outcome of an interaction between underlying structures and intentional conduct. Giddens is obviously sensitive to the arguments ad- vanced especially by structuralism and post-structuralism that the subject can no longer be conceived, as it has been since Descartes, as the autonomous source of meaning. Yet he hopes to accommodate these arguments within a position that does not simply turn the subject into a mere "bearer" of anonymous structures, as Althusser, for example, did. "A de-centring of the subject must at the same time recover that subject as a reasoning, acting being."'8 As Fred R. Dallmayr observes: "Giddens has seen himself faced with a momentous challenge: ... the task of moving beyond subjectivist metaphysics without relinquishing some of its insights, and especially with- out lapsing into objectivism and determinism."19

The attractions of such a position are obvious, avoiding as it does the extremes of a functionalism or structuralism that denies any scope to the human subject, and a hermeneutics that can find no place for social struc- ture. Or rather, it would be attractive had Giddens successfully avoided these extremes. I wish to argue that he has not, and that his commitment to the notion that human agents produce society gives rise to a tendency to collapse into the heremeneutic pole of the two extremes.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

137

All use subject to https://about.jstor.org/terms

138

Part of the difficulty lies in Giddens's identification of social structures with rules. "Rules," he tells us, "generate - or are the medium of - the production and reproduction of practices."20 Such a treatment of rules as constitutive of social practice, suggested already by his invocation of an analogy with language, is, of course, far from peculiar to Giddens. It was, indeed, typical of much Anglo-American philosophy in the 1950s and 1960s under the influ- ence of a certain reading of the work of the later Wittgenstein. Winch's Idea of a Social Science represented the apotheosis of this trend, as applied to social theory. Now Giddens wishes to distance himself from social science a la Winch, if only because he believes that structure involves, as well as rules, resources whose unequal distribution gives rise to power-relations.21 Never- theless, the conception of social practice as rule-governed is crucial to his theory of structuration.

There seem to me two reasons for questioning this approach. First, it involves generalizing from the case of language to social practice as such. Such a treatment of language as paradigmatic of social life is again not peculiar to Giddens - apart from in linguistic philosophy of the post-war Oxford school, it is also to be found in Levi-Strauss's structural anthropol- ogy, in Habermas's recent work, and in deconstructionists of the Derrida sort.

Of course, these various approaches embody, on the whole, very different conceptions of language - compare the Derridean treatment of discourse as inherently heterogenous and self-subverting, and Habermas's discovery, implicit in every speech-act, of a tendency toward rational consensus. Never- theless, they share a proneness to what Perry Anderson has recently called the "exorbitation of language," the identification of the structures of lan- guage with those of society.22 But, as Anderson persuasively argues, "lan- guage is no fitting model for any other human practice": linguistic structures change very slowly, are not subject to the constraints of natural scarcity, and exist through individual rather than collective agents.23 Anderson shows how the overextension of language has led in the case of post-structuralism to a point where neither history nor indeed reality itself have any standing outside the endless play of signifiers.

These are, however, general considerations. There is a second reason for quarrelling with Giddens's identification of structure with rules. Recent discussion, notably by Saul Kripke, of the passages in the Investigations and Remarks on the Foundations of Mathematics where Wittgenstein examines the notion of following a rule suggests that in these texts, far from conceiving social practice as rule-governed, he sets out to attack the idea that, as John

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

McDowell puts it, "the rules mark out rails along which correct activity within the practice must run."25

The difficulty with rule-following thus conceived is well stated by P. F. Strawson in this limpid summary of Kripke's reading of Wittgenstein:

It is natural to suppose that when we use a word or symbol of our language, we are guided in its use by our grasp of its meaning, or of the concept which it expresses, or of the rules or instructions for its use which we have mastered; that these are what tell us that it is correct to use the expression in such-and-such a way; to apply it to this case (if it is a descriptive term) or to compute with it in this way (if it is a mathematical symbol). So we appear to invoke some fact about our mental life to explain our confidence in the correctness of our current use of the expression. But if we take this conception seriously, it seems that there can be no guarantee that what we now take ourselves to mean by the expression is the same as what we meant by it in the past. For our practice is consistent with our having meant by it something quite different... the point ... should be obvious to any reader of Nelson Goodman: perhaps by "green" in the past I meant grue (where anything green seen before now and anything seen blue from now is grue).26

The conclusion of this sort of sceptical consideration is what Wittgenstein describes as the following paradox: "no course of action could be determined by a rule, because every course of action can be made to accord with it."27 Rather than grounding social practice, rules depend for their applicability on the shared responses of members of a community. It is on the basis of this interpretation of rule-following that Wittgenstein argues, in the later parts of the Investigations, that a private language, i.e., one that in principle only one speaker could use, is impossible, since the only criterion of correct usage is that provided by the common action of members of a society.

If this argument is valid, and it has still to be shown that it is not, then Giddens cannot appeal to the notion of rules as the basis of his account of social structure. Rather than generating practices, rules collapse into them. This is not great worry if one accepts, as I do, a much stronger sense of structure than that proposed by Giddens. It merely provides persuasive backing for my opposition to generalizing from the case of language. But, for Giddens, who wishes to preserve the hermeneutic tradition's sensitivity to the interpreting and acting subject while recognizing the structural aspects of social practice, it is a serious difficulty.

Giddens shows no sign of being aware of this difficulty. This is especially surprising in the light of the fact that he is evidently acquainted with the issues raised by Wittgenstein's discussion of rule-following. Indeed, he writes that "the knowledgeability involved in practical consciousness conforms generally to the Wittgensteinian notion of'knowing a rule' or'knowing how to go on."'28 Yet, in the exchange with Dallmayr published in Profiles and

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

139

All use subject to https://about.jstor.org/terms

140

Critiques, Giddens ignores the former's demand for"further argument... to pinpoint the status of rules in the process of structuration."29

Dallmayr goes on to argue that, although Giddens has "told us repeatedly that he wants to move beyond the so-called subject-object dichotomy," in the theory of structuration, "he reverts again to the subject-object dualism of actors constituting and being constituted."30 That this is so, that there is a persisting dualism of agency and structure in Giddens's writings despite his claim to have overcome it, can be illustrated by his discussion of two issues. In each he tends to treat agents and structures as separate from, and external to each other, and to accord primacy to the former.

The first is the question of resistance by subordinate groups. Giddens attacks

Foucault's notion of "history without a subject" on the grounds that "human

beings are always and everywhere knowledgeable human agents, though

acting within historically specific bounds of the unacknowledged conditions

and unaccepted consequences of their acts."31 He proceeds to criticize

Foucault's notion of society as constituted by a pervasive and omnipresent

apparatus of "power-knowledge" by arguing that "Foucault's 'archaeology,'

in which human beings do not make their own history but are swept along by

it, does not adequately acknowledge that those subject to the power of

dominant groups are themselves knowledgeable human agents, who resist,

blunt or actively alter the conditions of life that others seek to thrust upon them."32

Now this argument is closely related to the well-known difficulty in

Foucault's recent work that, although resistance is always supposedly im-

manent in power-relations, he never provides any basis for this resistance.33

Giddens seeks to provide such a basis by invoking the concept of "knowl-

edgeable human agents." This is a completely ahistorical approach. There is no discussion of the historically specific conditions that lead oppressed

groups to resist, and that provide their resistance with sources of organiza- tion and power. It is merely a general property of"knowledgeable human agents" that they "resist, blunt or actively alter" their "conditions of life." The strategy recalls Edward Thompson's counter-position in The Poverty of

Theory of the creativity and humanity of social agents to the structural

conditions of their activity. Perry Anderson's response to Thompson seems

to me applicable to Giddens as well: the scope for agency (and hence

resistance) in history varies according to the specific circumstances in which

people find themselves.34 Giddens's preference for a philosophical invocation

of subjects' capacity to resist rather than for a historical examination of the variable conditions of action is indicative of his privileging of the pole of

agency as against that of structure in the theory of structuration.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

Support for this view is provided by a highly symptomatic misreading of Foucault on Giddens's part. He writes: "Discipline, Foucault says, disasso- ciates power from the body... At the same time, the emphasis is placed on the 'interiorization' of power."35 Foucault, of course, says no such thing. The rise of the disciplines in early nineteenth-century Europe indeed represents a shift

from the form of power characteristic of absolutism, involving the spectacu- lar use of violence in public executions. But the new form of power, in which individuals are subjected to general norms within the context of institutions such as prisons, according to Foucault, also acts on bodies. In this case, however, it serves to constitute them as subjects, providing them with a set of motives governing their actions. Giddens's talk of "the 'interiorization' of

power" reveals his own presupposition that structures act on independently formed subjects. But, for Foucault, in the disciplines, power does not pene- trate the consciousness of a pre-formed subject; rather, these disciplines, create the subject.

The same presupposition underlies Giddens's critique of Harry Braverman's

Labour and Monopoly Capital.36 Giddens quite correctly points to the

weakness of Braverman's account of scientific management and the deskil-

ling of labor, namely that it treats workers as passive objects apparently

incapable of effective organization and resistance. However, this criticism is

theoretically grounded in the assertion that "to be a human agent is to have

power," so that "the most seemingly 'powerless' individuals are able to

mobilize resources whereby they carve out 'spaces of control'" giving rise to "a dialectic of control."37

All this is true enough, but it short-circuits the analysis of different modalities of resistance in different social formations. The scope for resistance of, for example, a slave in the Athenian silver mines at Laureion, while no doubt real enough, was surely much narrower than that of, say, the auto-workers at Ford Halewood studied by Huw Beynon. A Marxist would argue that the

difference between the two can only be understood by examining what Erik Olin Wright has called the structural capacities of particular classes - in other words, the distinctive powers that they derive from their position in histori- cally specific relations of production.38 Thus the shortcomings of Braver- man's analysis of the modern labor-process is closely related to his failure to grasp that workers' position within this process endows them with a capacity

for collective organization and resistance.39 Unfortunately, Giddens's pre-

occupation with an abstract account of human subjectivity prevents him

from following through the consequences of his own insight into the way in which social structures enable as well as constrain.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

141

All use subject to https://about.jstor.org/terms

142

A second index of Giddens's underlying preference for an account of society centering on the subject is his discussion of"time-space distanciation." One of the most interesting and innovative aspects of Giddens's recent work is his insistence on integrating the spatial and temporal dimensions of human life into social theory. Recent work in human geography, for example, should be treated, not as part of a separate discipline from sociology, but as an intrinsic part of the study of social formations.40

I have neither the competence nor the desire to question Giddens's often exhilarating discussions of these matters, merely to draw attention to one feature of the conceptual framework he uses. He draws on Heidegger and Derrida in order to argue that "all social interaction intermingles presence and absence."41 Human beings have direct contact with each other as present in space and time. However, they also have access to those distant in time through oral tradition, writing, and other such storage devices, and to those distant in space by various communications systems. Social structures serve to bind practices together in time and space through their contribution to the constitution and reproduction of systems of social interaction. Giddens goes on to distinguish between societies on the basis of the degree of their "presence-availability." In "societies or communities of high presence avail- ability ... interaction is predominantly of a face-to-face kind."42 Such socie- ties are to be found, for example, among hunter-gatherers. Giddens argues that history is characterized by a process of "time-space distanciation," such that "societies are 'stretched' over longer spans of time and space."43 This occurs as both communications and storage devices become more sophisti- cated, so that people interact with those with whom they are not in face-to- face contact. The culmination of this process is the emergence of the capital- ist world-system.

I shall discuss this analysis, insofar as it supposedly contradicts Marxism, below. I wish merely to note here that Giddens's argument differs significant- ly from the conception of Being as presence-absence developed by Heidegger and Derrida. For the latter argue that the condition of "high-presence availability" that Giddens attributes to band-societies is impossible. They do so for general philosophical considerations: namely, they wish to deny that a state of parousia or absolute presence in which we have direct access to reality can exist. We can only relate to such a state as necessarily absent, something that we have lost or may still hope to (re)gain. This is the point of

Derrida's concept of differance, of a constant play of presence and absence in which we are obliged to postulate, but never to attain, presence.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

Of course, Giddens is perfectly entitled to appropriate and modify other people's concepts. But it is the nature of the modification that is interesting. One of the points of the concept of differance was precisely to dislodge from his sovereign place the Cartesian subject, having direct access to the contents of his consciousness (in other words, being "self-present"), and providing a basis on which our knowledge of the world (and indeed on some versions the world itself) could be constituted. Giddens does not reinstate the Cartesian self - rather it is the subject of social interaction, engaging in face-to-face contact with other subjects in the manner analysed by Goffman, which reigns in the condition of "high presence-availability."44 History becomes, as Derrida argues it must necessarily be, the story of the loss of presence, of the "high presence-availability" characteristic of hunter-gatherer societies. Once again, "knowledgeable human agents" triumph over structures.

These two cases - Giddens's discussion of resistance and his account of "time-space distanciation" - suggest that he has failed to overcome th duality of agency and structure. The reason for this failure is not a contingent one - it is not a matter of Giddens being inconsistent. Rather, in my view, it source lies in the very conception of structure as the unacknowledged condi- tion and unanticipated consequence of human actions.

Let us consider structure in its two aspects. Conceiving structure as th unanticipated consequence of human action does not involve any challenge to the primacy of agents over structures. On the contrary, such a conception of structure has usually been held precisely as a means of securing this primacy. This is most obviously so in the case of methodological individual ism. If social structures arise as the unintended consequences of individual actions, then it is with these actions, and the beliefs, desires, and intentions that produce them, that we must start in order to account for the existence of structures and of the practices bound up with them. Jon Elster, who recently urged in the pages of this journal that methodological individualism can provide the "microfoundations of Marxist social theory,"45 has, in line with this approach, developed an account of social contradiction as unintended consequence.46

That Giddens is committed to a stronger notion of structure than Elster's is suggested by his critique of the latter's theory of contradiction.47 And, indeed, the concept of structure as the unacknowledged condition of human action seems to go much further than any account in terms of unintended consequences, especially when Giddens talks of structure as both constrain- ing and enabling. But when we consider the detail of Giddens's account structure is once again reduced to a secondary role. Structures are, let us

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

143

All use subject to https://about.jstor.org/terms

144

recall, rules and resources. Even though Giddens does not fully grasp the import of Wittgenstein's arguments about rule-following, he does recognize rules' dependence on the practices that they supposedly govern.48 So rules only "generate" practice in a very weak sense.

The other aspect of structure is resources, which are either authoritative or allocative, depending on whether they give command over persons or objects respectively. I shall have more to say about them when I come to discuss Giddens's account of power. What I wish to stress here is that he describes resources as "media" or "means" that are "utilized" by agents seeking their own ends, and thereby (unintentionally) reproducing society.49 Structure thus understood in no way constitutes social action. It is conceived instru- mentally, as a tool used by agents in order to realize their wants.

Surprisingly, since Giddens often belabors Marx for identifying human praxis with the material work of transforming nature,50 his is a remarkably economic conception of practice, in which agents use scarce resources to achieve their ends. There seems to me no way of genuinely distinguishing this position from Popper's, whose methodological individualism is precisely a generalization of neo-classical economics's model of rational action.

Habermas's recent comments on Giddens go to the heart of the matter. Giddens's "concept of practice," Habermas writes,

is supposed to be related to the constitution of action-complexes and to their reproduction. I do not think that this choice of conceptual strategy is a fortunate one, because the basic epistemological concept of'constitution', which refers to the formation of object-domains, causes confusion in social theory. It suggests that speaking and acting subjects 'produce' their social life-context in a way similar to that in which they make products of instrumental action [i.e., labor]... The over extension of action-concepts in a theory of constitution remains stuck in metaphors; the basic concept of the philosophy of Praxis gives at best an anthropomorphistic concept of society.51

Habermas has put his finger on the central fault in Giddens's theory of structuration. The way in which Giddens conceptualizes structure means that it can only function as a secondary aspect of social practice, subject to the creative interventions of "knowledgeable human agents." The result, as Habermas points out, is "an anthropomorphistic concept of society," in which the subject reigns supreme, producing society as labor produces use-values. Far from overcoming the dualism of agency and structure, Giddens is stuck firmly at the pole of agency. The much-invoked "duality of structure" amounts to little more than a substitution of two letters - a verbal change.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

The Paralogisms of Power

Similar conceptual sleight of hand characterizes Giddens's account of power. The concept of power is central to his critique of Marxism. Thus,

a fundamental component of my arguments is that the articulation of space-time relations in social systems has to be examined in conjuction with the generation of power. A preoccupa- tion with power forms a leading thread of this book. I maintain that power was never satisfactorily theorized by Marx, and that this failure is at the origin of some of the chief limitations of his scheme of historical analysis.52

There are three main steps in Giddens's conceptualization of power. First, he argues that "the notion of 'action' ... is logically tied to that of power... Action intrinsically involves the application of 'means' to achieve outcomes, brought about through the direct intervention of an actor in a course of events."53 "Power" is understood here as "the transformative capacity of

human agency," that is, as "the capability of the actor to intervene in a series

of events so as to alter their course; as such it is the 'can' which mediates between intentions and wants and the actual realization of the outcomes sought after."54

The second step involves introducing a "narrower, relational sense" of "power," as a "property of interaction," "power as domination."55 It arises "where transformative capacity is harnessed to actors'attempts to get others to comply with their wants."56 The result is "structures of domination," which "involve asymmetries of resources employed in the sustaining of power-relations in and between systems of interaction."57 Giddens distin- guishes between two main forms of domination, allocation and authoriza- tion, depending on which type of resource is unequally distributed. Alloca- tive resources comprise "material features of the environment," "material means of production/ reproduction," and "produced goods." Giddens's account of authoritative resources, which consist in the "organization of social time-space," "production/reproduction of the human body," and "organization of human life-chances," is a good deal less perspicuous: these resources are something of a residual category.58 Nevertheless, the gist is clear enough: allocation is essentially economic domination, authorization politico-ideological.

The third step is to use these concepts to offer a theory of historical develop- ment. Giddens claims that power and domination are inherent in human social life.59 The chief variation is in the form of domination. Prior to

capitalism, Giddens argues, authorization was the prevailing form. He dis- tinguishes between class-divided and class societies. In both categories

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

145

All use subject to https://about.jstor.org/terms

146

classes, i.e., "the sectional forms of domination created by private ownership of property,"60 are to be found. However, in a class-divided society "class analysis does not serve as a basis for identifying the basic structural principles of organization of that society."61 Class-divided societies compromise all "non-capitalist civilizations" - city states, empires, and feudal societies. The structures of domination in these societies depend upon the distribution of authoritative resources. "In class-divided societies, as opposed to class society, authorization has primacy over allocation."62 Capitalism is the only case of a class society; here alone does allocation prevail. The only limited validity historical materialism possesses is in its analysis of capitalism, where economic factors do predominate over ideological and political ones.

I wish to criticize this account of power, but let me first stress its virtues. Giddens's is not a harmonistic theory of society. At its center is contradic- tion, power, and struggle. This is its merit, compared with those of Parsons and Habermas, for whom power serves primarily to integrate social systems, and to dampen down social conflict.63 Giddens's is a view of society having much in common with Weber's and, as I shall try to show, it is with Weber that he is best compared, despite his assertions to the contrary.

A crucial element in Giddens's strategy is the shift from the broad to the restricted conceptions of power, from power as transformative capacity to power as domination. He is quite explicit in insisting that it is the broad concept of power that grounds the claim that domination is inherent in social life. For example, "all social interaction involves the use of power, as a necessary implication of the logical connection between human action and transformative capacity. Power within social systems can be analyzed as relations of autonomy and dependence between actors in which these actors draw upon and reproduce structural properties of domination."64 Similarly, "transformative capacity... forms the basis of human action - the 'could have done otherwise' inherent in the concept of action - and at the same time connects action to domination and power."65

If we are to take these passages as suggesting that one may infer from the "logical connection" between transformative capacity and action to domina- tion as a persisting feature of social interaction, and that is what expressions such as "necessary implication" would lead us to believe, then Giddens is guilty of an obvious fallacy. I doubt if anyone would wish to question whether "action" is "logically tied" to "transformative capacity," but recogni- tion of the involvement of power thus understood in human action does not commit one to very much. For "transformative capacity" understood as "the capability ... to intervene in a series of events so as to alter their course" is a

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

property of all causal agents, not merely human ones. Indeed, "transforma-

tive capacity" seems in reality to be merely another expression for "causal

powers," since such powers consist pecisely in the ability to bring about some

alteration in the course of the world.66 It is true that Giddens makes much of

"the 'could have done otherwise' inherent in the concept of action," presum-

ably referring to the unique property of human beings to reflect upon and

choose between alternative interventions. But this consideration does not

bear on the nature of "transformative capacity" but rather on human agents' peculiar ability consciously to control and monitor the exercise of their causal powers.

It is one thing to attribute to human beings this "transformative capacity." It is quite another to infer from this true but rather trivial proposition the much more interesting, but controversial (and, in my view, false) thesis that power in the sense of domination is intrinsic to human life. There is no necessary connection between the two concepts of power. Further premises would be required tojustify the inference from transformative capacity to domination. Interestingly enough, a similar point has been made recently by Foucault. He distinguishes between "objective capacity," "which is exerted over things and gives the ability to modify, use, consume, or destroy them," and "power, which brings into play relations between individuals (or between groups)." "Objective capacities" and "power-relations," Foucault insists, "should not

... be confused."67 Although Giddens distinguishes carefully between the broad and restricted sense of power, it is difficult not to feel that when he claims that "transformative capacity ... connects action to domination and power," he is guilty of an equivocation. The result is a paralogism, in which the same word, "power," is used in different senses in the premise and conclusion of the same argument.

It does not seem to me, for reasons that I will make clear shortly, that this elementary logical mistake is necessarily fatal for Giddens's account of domination. But since he makes so much of the notion of "transformative

capacity" it is worth pointing out how little use of this concept commits one to. This brings out a common feature of Giddens's writings, namely, that they are prolific of neologisms and, more frequently, of redefinitions of familiar words, without these new concepts, and the complex typologies often accompanying them, usually offering much gain in clarity or empirical reach. Central Problems in Social Theory is especially guilty of this, an arid treatise raining down definitions and diagrams on the bewildered reader's head. From this point of view A Contemporary Critique is especially to be welcomed, since here Giddens cashes his conceptual checks, putting the typologies of Central Problems to work in order to provide an account of historical development different from Marx's.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

147

All use subject to https://about.jstor.org/terms

148

It is here that one might quite reasonably say in Giddens's defense that his analysis of domination can stand on its own legs, without the support of "transformative capacity." The test of its validity will then turn on whether it explains a series of phenomena that Marxism cannot. This argument is a cogent one, and can only be answered by turning, at last, to the concrete detail of Giddens's critique of Marx.

Marxism and Evolutionism

The Marxism that Giddens seeks to transcend is an evolutionist one, in

which social change is an outcome of the level of development of the

productive forces. Such a view of historical development is characteristic of

Marxism in general: "Marxist authors are virtually everywhere committed to

evolutionism in one guise or another."68 Only in those sections of the

Grundrisse where he discusses pre-capitalist social formations (the so-called

Formen) does Marx himself adopt a different approach: "rather than imply-

ing that the forces of production have their own internal dynamic, as in his

evolutionary view, Marx seems in the Formen to give primacy to 'ecological'

factors (dispersal or concentration of populations) and to war in stimulating social transformation."69

Giddens seeks to disprove Marx's evolutionary theory, and substantiate his

own theory of domination by providing an account of historical develop- ment in which it is the distribution of authoritative resources that has

primacy until the development of capitalism. We have already seen that he believes history to be a process of time-space distanciation in which the high presence-availability of tribal societies is progressively lost. These changes are highly significant for the manner in which societies are integrat- ed. Giddens distinguishes between societal and system integration. Societal integration involves the relationships between individual social actors; systems integration concerns the connections between social groups and collectivities.70 In tribal societies the two forms of integration are fused, so that system-integration, the relations between groups, is dependent on the face-to-face contacts between individuals, and the traditions and kinship networks with which these contacts are intimately bound.

The two forms of integration are separated with the emergence of class- divided societies. Societal integration still depends on kinship and tradition, and small localized communities continue to form the basis of these societies.

However, system-integration depends increasingly on the power of the state, and the authoritative resources that it is able to mobilize. The extraction of

surplus-labor from these communities depends on the intervention of the

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

state, and takes the form of deductions - taxation, rent, etc. - leaving much of their daily life unaffected. The necessary authoritative resources to mount such control over local communities required the development of new forms of "information storage," such as writing, that in turn depended on the emergence of the city. It is the rise of the city, "the generator of the authorita- tive resources out of which state-power is created and sustained,71 not the development of the productive forces that is the decisive factor in theemergence of classes.

Capitalism, like "class-divided societies," separates societal and system inte- gration. However, the content of these two forms of integration is radically changed. The extraction of surplus-labor comes to depend, not on the direct intervention of coercive power, but rather on economic mechanisms, the pressures, analyzed by Marx in Capital, that force workers to sell their labor-power. Capitalism is thus characterized, as Giddens has earlier stressed in The Class Structure of the Advanced Societies, by "a 'separation' between economy and polity such that there is at least a substantial scope for the play of market mechanisms independent of active political control."72 But he now accounts for the "insulation" of economy and policy on the grounds that "only with the advent of capitalism does an exploitative class relation become part of the very mechanism of the productive process."73

The other side of the separation of state and market is the transformation of the state into "a much more intrusive and comprehensive set of institutions in capitalist than in class-divided societies, so far as those subject to its adminis- tration 'internally' are concerned."74 No longer is societal integration left to kinship and tradition. The dependence of the capitalist economy upon the capacity of producers continually to increase their output requires an enor- mous expansion of state-directed apparatuses of surveillance, by which Giddens means both the accumulation of information, and what Foucault calls the disciplines, the systematic supervision and regulation of the activi- ties of those involved in a variety of institutions - factories, armies, prisons. Consequently, the "crucible of power" is no longer the city, but the nation- state, a phenomenon whose significance Giddens believes Marxism largely to have ignored.75

Nevertheless, I suspect that most people, confronted with this account of historical development, would be puzzled as to why it should be regarded, with all its stress on changing modes of surplus-extraction, as in any sense critical of, or an alternative to Marxism. Giddens's reply would presumably be that Marxism accords primacy to the productive forces, attributing to them an "internal dynamic" that is the source of all historical change. Now it

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

149

All use subject to https://about.jstor.org/terms

150

is true that such is one version of Marxism, most eloquently defended recently by G. A. Cohen in his Karl Marx's Theory of History - a Defense. It is not, however, the only version of Marxism, nor is it, as I have argued elsewhere, the most adequate.76 Furthermore, I think it can be shown that it is a view that Marx himself progressively abandons.77

However, as Erik Olin Wright persuasively argues, rejection of evolutionist Marxism does not require the abandonment of any notion of historical evolution.78 The crucial distinction, he suggests, is that between teleological versions of historical materialism that conceive social formations as evolving towards a predetermined goal, and those that adopt a weaker and unobjec- tionable evolutionary theory. Wright urges that all such a theory must claim is that "there is some process, however weak and sporadic, which imparts a directionality to movements from one form to another."79 Indeed, he sug- gests, Giddens's own account of history as a process of time-space-distancia- tion, in which time and space are progressively stretched, and "the immedi- acy of presence"80 gradually lost, meets the conditions for an evolutionary theory: "Because of the link between conflict, power, resources and distanci- ation, there will be at least some impulse for increasing space-time distancia- tion."81

Giddens cannot in any case dispense with the notion of the development of the productive forces as easily he would like. Writing, which Giddens believes to have developed "as a means of recording information relevant to the administration of societies,"82 may have in fact originated in the clay tokens used by Neolithic farmers to keep track of their food stores.83 More general- ly, time-space distanciation crucially depends on technological innovations

- improvements in transport, new systems for transmitting information, etc. The stretching of space and time is in no sense a process autonomous of the development of the productive forces.

To argue in this manner is not, however, to accord causal primacy to the productive forces. Their development is best understood, in my view, as a necessary, but not sufficient condition of changes in social relations. Furth- ermore, the scope for developing the productive forces is determined by the prevailing relations of production. These relations are best conceived as a specific mode of appropriation of surplus-labor, which in turn depends upon the distribution of the means of production. Forces and relations mutually limit one another, rather than the one exerting primacy over the other.84

Now the relations of production thus conceived seem to me perfectly capable of accounting for the shift from authoritative to allocative domination on

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

which Giddens places such stress. Marx himself provides such an explana- tion in his crucial chapter, "Genesis of Capitalist Ground-Rent," in Capital volume 3.85 There he argues that

in all forms in which the direct labourer remains the "possessor" of the means of production and labour-conditions necessary for the production of his own means of subsistence, the property relationship must simultaneously appear as a direct relation of lordship and servitude, so that the direct producer is not free; a lack of freedom which may be reduced from serfdom with enforced labour to a mere tributary relationship. The direct producer, according to our assumption, is found here in possession of his own means of production, the necessary material labour conditions required for the realization of his labour and the production of his means of subsistence... Under such conditions the surplus-labour for the nominal owner of the land can only be extorted from them [the direct producers] by other than economic means.86

As a conceptualization of feudal relations of production this account is not

wholly unproblematic.87 Nevertheless, it has provided the basis of at least one

major empirical study of late feudalism.88 And it does provide a means for

accounting for the apparent primacy of "authoritative domination" in pre-

capitalist social formations. Surplus-labor can be appropriated by a ruling

class where the direct producers themselves control the labor-process only

through the systematic use of the means of coercion. By contrast, the

complete separation of the direct producers from the means of production

under capitalism accords primacy to economic mechanisms in the extraction

of surplus-labor. As Marx puts it, "the silent compulsion of economic

relations sets the seal on the domination of the capitalist over the worker.

Direct extra-economic force is still of course used, but only in exceptional cases."89

Now Giddens is perfectly well aware of this analysis. Indeed his own account of the differences between class-divided and class societies depends crucially upon it, as he acknowledges.90 Some of the most suggestive parts of A Contemporary Critique, for example, his demonstration of how the mecha- nisms of"surveillance" described by Foucault can only be understood in the light of the mode of surplus-extraction peculiar to capitalism,91 are best taken as a development of Marx's own account. Why, then, does Giddens believe that his version of historical development is an alternative to Marx's?

There seem to be at least three reasons. The first is Giddens's persisting methodological commitment, as I have shown earlier in this article, to a version of interpretive sociology, which gives the subject primacy over structures, and treats power in the sense of domination as intrinsic to social life. I shall have a little to say about this at the end of this section. Secondly, Giddens, as we have seen, misinterprets Marxism, treating it as necessarily

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

151

All use subject to https://about.jstor.org/terms

152

evolutionist, that is, as an interpretation of history as inevitable progress toward a predetermined goal.

It is, however, the third reason to which I wish first to devote some attention. This is Giddens's tendency to identify the relations of production with market-relationships. This is most obvious in the case of The Class Structure of the Advanced Societies, where he argues, for example, that "classes are large-scale groupings" whose existence depends upon "the formation of market-relationships and a division of labour allowing the production of commodities."92 He concludes, not surprisingly, that neither feudalism nor what he chooses to call "state socialism" are class societies. Moreover, in capitalism itself "the system of exploitation operates through differentials inmarket capacity."93 On this basis, Giddens distinguishes between three classes in capitalism - the bourgeoisie, owning the means of production, manual workers, owning labor-power, and an intermediate class today composed largely of white-collar workers whose greater skill than manual workers enhances their "market capacity."

While Giddens's analysis of class-relations is considerably more sophisticat- ed in A Contemporary Critique, as well as (because?) closer to Marx, he continues to identify class with market relations. Thus he argues that in class-divided societies, class is less important than under capitalism because the main form of private property, land, is alienable only to a limited degree, if at all.94 The salience of class and the development of the market are directly related.

With Marx, however, the existence of classes depends upon the monopoliza- tion of the means of production by a minority and their consequent ability to appropriate surplus-labor. It is entirely a secondary matter whether these means of production are commodities that may be purchased and sold on the market. (Even under capitalism land is a very peculiar sort of commodity). The relations of production may well contradict the apparent organization of society, what I have called elsewhere a social formation's form of articula- tion. Thus, in the case of feudalism, the dependence of surplus-extraction on coercion means that the relations of production "appear as a direct relation of lordship and servitude." Social stratification consequently takes the form of a status hierarchy, the division of society into estates, that contradicts the underlying class-relationships. Similarly, under capitalism the relations of production take the form of market relationships, of apparently equal and reciprocal relations between commodity-owners. In the crucial chapter of

Capital volume 1, "The Sale and Purchase of Labour-Power," Marx stresses the disjunction between this apparent equality and the underlying inequality

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

between the capitalist who owns the means of production and the worker who has nothing but his or her labor-power.95

Yet when we turn to Giddens we find that the main lesson he takes from

Marx is that of "the centrality of the labour-contract to the capitalist system."96 It is not, however, the labor contract, but the distribution of the means of production underpinning it that is crucial for Marx. Giddens's discussion of class and capitalism can be said to suffer from two serious weaknesses. First, it tends to overemphasize the importance of commodity production for capitalism. This is reflected, for example, in his introduction of the notion of a commodification of time under capitalism, which raises issues too complex to be discussed here, but which seems to me to be both overstated and misleading.97 (The difficulties Giddens experiences here are in my view bound up with his rather unsteady grasp on Marx's economic theory. Of his discussion of Marx's theory of money, for example, where Giddens says that for Marx "money becomes a commodity only because it represents or symbolizes the exchange-value of all other commodities"98 - the exact reverse of Marx's own view, the less said the better.) Secondly, there is a tendency to confuse juridical property forms, such as the labor-contract, with the relations of production. What Marx is concerned with is not legal entitlement but effective possession.99

Giddens is, in my view, insufficiently aware of the fact that, for Marx, the

relations of production and class are analytical concepts rather than descrip-

tive categories. In other words, they are intended to uncover underlying

relationships precisely because the latter's nature is not directly observable.

Giddens consequently commits what Whitehead called the fallacy of mis-

placed concreteness, supposing that it is only when the division of society

into economic classes is directly visible, as under capitalism, that historical materialism can be said to be valid.

Marx had in a sense already responded to Giddens's "contemporary" cri- tique when he answered in Capital volume 1 a "German-American publica- tion" that had argued that the materialist conception of history "is all very true for our own times, in which material interests are preponderant, but not for the Middle Ages, dominated by Catholicism, nor for Athens and Rome, dominated by politics." Marx replied that "one thing is clear: the Middle Ages could not live on Catholicism, nor could the ancient world on politics. On the contrary, it is the manner in which they gained their livelihood which explains why in one case politics, in the other case Catholicism, played the chief part."'00 The role of the concept of the relations of production is precisely to explain the changes in the forms of domination.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

153

All use subject to https://about.jstor.org/terms

154

One might recast the points I have just made in slightly more formal terms, drawing on recent work in the philosophy of science. Imre Latakos distin- guished between a scientific research program's heuristic, the set of privileged propositions forming its "hard core" and specifying procedures for resolving problems, and the "protective belt" of revisable and empirically falsifiable auxiliary hypotheses surrounding the heuristic. 01 Progress within a research program depends, Lakatos argued, on the empirically successful modifica- tion and elaboration of auxiliary hypotheses along lines suggested by the heuristic. One could say, in the case of Marxism, that its heuristic consists in the basic concepts and propositions of historical materialism, while the auxiliary hypotheses are analyses of specific modes of production, social formations, and conjunctures.102 Much of the point of auxiliary hypotheses is to account for phenomena that apparently contradict the factual claims made in the heuristic. An example of such recalcitrant phenomena in the case of Marxism are the pre-capitalist societies in which authoritative rather than allocative domination seems to prevail. Giddens, because he fails to distin- guish between the different theoretical levels of a scientific research program, believes that the mere existence of these societies counts as a refutation of historical materialism.

It is an implication of this argument that the proof of the pudding is in the eating. Is historical materialism capable of providing empirically satisfactory analyses of pre-capitalist societies? The work of Marxist historians suggests that it is. One example must do. G. E. M. de Ste. Croix's The Class Struggle in the Ancient Greek World is notable, not only for the wealth of scholarship, and ethical and political passion that it displays, but because it is, as Perry Anderson says, "one of the most strenuously theoretical works of history ever to be produced" in Britain.103 Of particular importance is the manner in which de Ste. Croix uses Marx's theory of class in effect as a heuristic, that is, as the basis on which to formulate empirical propositions concerning the social structure of Antiquity. Rather than concentrate on the description of visible status differences, in the manner of M. I. Finley, de Ste. Croix uses the basic concepts of historical materialism both to anatomize classical society, and to provide a dynamic explanation of its development.104 His book is a formidable challenge to Giddens's claim that Marxism offers insight only into the workings of capitalism.

Let me now return to the two reasons for Giddens's rejection of historical

materialism other than his misunderstanding of Marx's theory of class -

namely, his methodological commitment to interpretive sociology, and his identification of Marxism with evolutionism. One might imagine a connec-

tion between the two. I have already mentioned Giddens's vigorous objection

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

to functionalism. Not unnaturally, he tends to associate evolutionism with

the use of functional explanations, since both tend to treat society as a

self-reproducing organism with its own needs.105 Now Giddens might argue

that any Marxism that continues to accord primacy to structures over

subjects must necessarily rely on the tacit or open use of functional explana-

tions that effectively reduce agents to the "supports" of structures. The

thought would be that however much Marxists such as Wright and myself

might like to repudiate evolutionist and functionalist Marxism, that is all the Marxism there is.

Giddens could appeal for support to G. A. Cohen. Rejecting Jon Elster's proposal that historical materialism be reconstructed along methodological individualist lines, Cohen correctly insists that "Marxism isfundamentally concerned not with behaviour, but with the forces and relations [of produc- tion] constraining and directing it," but asserts also that the constraining and directing role of these forces and relations can only be explicated by means of functional explanations: "if functional explanation is unacceptable in social theory then historical materialism cannot be reformed and must be rejected."06

This line of thought is not compelling. It is perfectly possible to follow

Giddens in rejecting functional explanation, while continuing to accord

primacy to structures over agents. The key lies in Giddens's own conceptuali-

zation of structures as both enabling and constraining. Thus we may accept

one of the standard accounts of action made available by contemporary

analytical philosophy, conceiving agents' behavior as proximately caused by

their intentions, beliefs, and desires, while treating the powers instantiated in

this behavior as to a large extent structurally determined. In other words,

what agents can collectively or individually do depends to a significant

degree on their position in the relations of production. The objection to the

philosophies of action formulated within the analytical or hermeneutic

traditions would then be not so much that such theories are false, or part of a

"problematic of the subject," as that they are empty, lacking as they do an

empirical account of the historically variable structural determinants of action.

Obviously these remarks are more hand-waving than a properly formulated argument. Nevertheless, they serve to suggest that the dilemma between functionalism and heremeneutics, to which Giddens offers the theory of structuration as a solution, does not exhaust the available alternatives.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

155

All use subject to https://about.jstor.org/terms

156

Capitalism, Socialism and the State

Let me turn now to consider some rather surprising criticisms of Marxism that Giddens makes towards the end of A Contemporary Critique. These center on the role of the state and violence. He writes, for example, that "Marxism has no tradition of theorizing violence either as an integral and chronic feature of repression, or of the 'world violence' of the contemporary system of power blocs and nation-states."107 This is related to the historical claim that the development of capitalism was bound up with the formation of a European power-system dominated by the great absolutist states, so that "capitalism developed within a military 'cockpit' in which the expansion of industrial production very soon came to be seen by all ruling groups as the sine qua non of national survival."108 "The conjunction between the rise of capitalism and the absolutist state system produced a system of nation-states that, far from being ephemeral, is integral to the world capitalist economy - which is at one and the same time a world military order."'09

I describe these criticisms as "surprising" because many of the statements I have just cited, which Giddens clearly believes to be damaging criticisms of Marxism, are from the standpoint of Marxism platitudinous. The notion, for example, of the state as constituted by violence, by the existence of what Lenin called "special bodies of armed men," is a central theme of such texts as

The Civil War in France, The Origins of the Family, Private Property and the State, and The State and Revolution. The more usual criticism of Marxist writing on the state is that it overestimates the role of violence as "an integral and chronic feature of repression."

Again, to take the case of absolutism, Marx in his writings on France traces the origins of the modern capitalist state to the absolute monarchies of the ancien regime. The role of the quasi-absolutist Tudor state in promoting the development of capitalism through the systematic use of coercion, and of the world power struggle between the European states in contributing to the "primitive accumulation" of capital are well-known themes of part eight of

Capital volume 1.

Then there is the matter of the "world capitalist economy." Giddens writes as

if this were a concept discovered a few years ago by Immanuel Wallerstein,

rather than being a theme of Marx's own economic writings, which assumed

central importance for those who sought to develop the analyses they

contain. It was precisely on the capitalist world-system that the two genera- tions of Marxists after Marx focused their attention. The result was numer-

ous texts by Kautsky, Hilferding's Finance Capital, Luxemburg's The Ac-

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

cumulation of Capital, Lenin's Imperialism, and Bukharin's Imperialism and World Economy and The Economics of the Transformation Period. Lenin and Bukharin in particular stress that with the emergence of imperial- ism competition between capitals increasingly took the form of military competition between nation-states.

I do not wish to claim that this body of writing on the state and the world

system is wholly satisfactory or adequate. But it exists. It is true that the

whole body of work to which I have referred was vulgarized and integrated

into a highly evolutionist form of Marxism with the triumph of Stalinism.

Moreover, the various schools of Western Marxism were largely, though by

no means wholly, uninterested in the type of economic and political analysis characteristic of classical Marxism.

But even at the height of Stalinism's grip on Marxism evolutionism did not go unchallenged. For example, Giddens lays great stress on the concept of "time-space edges," by which he means "the simultaneous inter-connected

existence of different types of society. This helps to free us from the tendency of evolutionary thought to analyse societal development in terms of 'stages', and from the influence of 'unfolding models' of change.""0 Compare this with, for example, the opening chapters of Trotsky's History of the Russian Revolution, where central importance is given to "the law of combined development - by which I mean a drawing together of the different stages of the journey, a combining of separate steps, an amalgam of archaic with more contemporary forms."'1 l Trotsky uses the concepts of uneven and combined development to explain why socialist revolution occurred first in Russia, rather than, as evolutionist Marxism had predicted, in the more advanced capitalist countries. (Incidentally, Giddens himself had earlier offered a highly evolutionist explanation of this revolution, arguing that "the [manual] working class is most likely to achieve a high degree of revolutionary class consciousness in the initial phase of the industrialization process."112)

The revival of Marxist theory in the past decade and a half has seen a significant return to the classical focus on the world-system and the state. Giddens cites, but does not discuss Poulantzas's last book, State, Power, Socialism, which is concerned precisely with the essential role of the organ- ized violence of the state in both the constitution and the reproduction of the relations of production.113 Giddens describes Perry Anderson's Lineages of the Absolutist State as "a fundamental source,""4 but does not address the central thesis of that book, which is that the absolutist state must be under-

stood as a form of feudal class rule. Although it is true to say that much of the "state-derivation" debate among German Marxists has been sterile and

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

157

All use subject to https://about.jstor.org/terms

158

formalistic in the extreme, nevertheless a number of contributions to that and connected debates, notably those by Claudia von Braumuhl, Colin Barker, and Cesare Luporini, have focused on a theme that Giddens approp- riates as his own, namely the claim that capitalism involves a system of competing nation-states whose origins are to be found in the absolutist era.15 These texts all deal with the centrality of war to that system, as do Anderson in his discussion of absolutism, and the theorists of the permanent arms economy (T. N. Vance, Michael Kidron, and Chris Harman) in their anal- yses of contemporary capitalism.1"6

The interesting question to arise from all this is why so erudite, and, in many ways, so friendly a critic of Marxism as Giddens, should come to ignore so large and important a body of writing. There are, I think, at least two reasons. The first has to do with sociology's general failure to confront Marxism as a political tradition. With the very important exception of Marx himself, classical Marxism is not discussed by Giddens. Lenin, Hilferding, and Gramsci each appear once in the index of A Contemporary Critique, Engels, Luxemburg, Trotsky, and Bukharin not at all. Althusser and Pou- lantzas, on the other hand, receive generous attention.

Why does this matter? Because the theoretical writings of the classical Marxists have an irreducibly political orientation. Their intellectual activity bore an intrinsic connection to their participation as revolutionary socialists in working-class movements. This active involvement in politics explains why what Perry Anderson calls the "areas most central to the classical traditions of historical materialism" were "scrutiny of the economic laws of motion of capitalism as a mode of production, analysis of the political machinery of the bourgeois state, strategy of the class struggle necessary to overthrow it." 17 The political orientation of classical Marxism has, however, made it a difficult tradition for academic social science to engage with. It is, for example, only in the past few years, and largely as a result of the radicalization produced in the late 1960s and early 1970s, that serious studies of the political thought of Marx and Lenin have appeared.18 On the whole, sociologists have found it easier to address those Marxists, such as the Frankfurt school and Althusser and his followers, whose theoretical pre- occupations are largely detached from political activity, and are therefore much more readily assimilable to a purely academic discourse. Although there is a political undercurrent to Giddens's work (A Contemporary Critique is the first in a two-volume attempt to rethink socialist theory), 19 he

shares this common failure to grasp the distinctive nature of classical Marxism.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

The second reason for Giddens's highly selective reading of Marxism has to

do with his latent Weberianism. When he says that "Marxism has no

tradition of theorizing violence" he may mean, not so much that Marxism

ignores the phenomenon of violence, as that it has no satisfactory explana-

tion of it. Marxism, Giddens might say, recognizes the existence of violence;

it even seeks to explain it - for example, in the way that Engels and Lenin do,

for example, in terms of class exploitation and the specific political structures

to which it gives rise. But, Giddens might argue, these explanations will not

do. Why will they not do? In part, because of the weaknesses specific to these

explanations, and indeed Giddens subjects recent Marxist writing on the state to critical examination.120

But there is also, I think, an undercurrent to Giddens's arguments that

suggests that any attempt to explain violence in terms of other factors is

doomed to failure. Formulations such as "an overwhelmingly obvious and chronic trait of human affairs - recourse to violence and war"12' seem to

imply that this "trait" is an irreducible feature of social life. Any theory of violence that, like Marxism, treats it, at least in its more organized and destructive forms, as specific to certain forms of society (namely, those in which classes exist) is, from this perspective, necessarily mistaken. Such a line of argument, if present in Giddens's writings, is surely connected to his tendency to assimilate the two senses of power - transformative capacity and domination. For a theory of action implying that relations of domination are an essential characteristic of social formations can easily be made to issue in an account of violence that treats it as an equally "chronic" and irreducible "trait of human affairs." The kinship between such a view of society and what Giddens himself acknowledges to be Weber's Nietzschean stress on power, conflict, and violence'22 is obvious.

The Weberian drift of Giddens's argument is evident when we consider one of his other criticisms of Marxism. He argues that Marx, by focusing on the extraction of surplus-labor, offered too narrowly economic an account of exploitation, which should properly be understood simply as "domination which is harnessed to sectional interests."23 Giddens is then able to identify forms of exploitation that Marxism has failed to account for:

(a) exploitative relations between states, where these are strongly influenced by military domination; (b) exploitative relations between ethnic groups, which may or may not converge with the first; and (c) exploitative relations between the sexes, sexual inequality... These are major "absences" in Marxist theory.124

What does it mean to talk here of"'absences' in Marxist theory"? If Giddens means that Marxism has consistently ignored, or denied the existence of, say,

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

159

All use subject to https://about.jstor.org/terms

160

racial and sexual inequalities, then he is surely wrong. We can point to, for example, Engels's Origins of the Family, the debates in the Second Interna- tional on the national question, and recent Marxist work on racial oppres- sion in South Africa. If Giddens is saying that Marxism has failed to come up with a satisfactory theory of phenomena such as nationality, then I'm inclined to agree with him (though it's not clear that orthodox social science has done any better). If, however, the claim is that Marxism cannot provide such a theory, then more argument is required in its support than Giddens gives.

Let me elaborate the last point by returning to the distinction drawn above between the heuristic and the auxiliary hypotheses of a scientific research program. The heuristic of historical materialism, consisting as it does of concepts such as forces and relations of production, exploitation, class, etc., does not make any reference to sexual, ethnic, and inter-state relations. This is, however, in no sense a decisive objection to Marxism. For the work of the heuristic is done precisely in generating auxiliary hypotheses that must, inter alia, provide empirical explanations of phenomena such as racial inequality. The relative failure of classical Marxism to come up with adequate theories of racial and sexual oppression in comparison with the considerable body of economic and political analyses to which I have already referred arguably has less to do with any immanent "class-reductionist" logic than with politi- cal priorities that encouraged the development of historical materialism in other directions.

This is in no sense an invitation to Marxists complacently to ignore the many inadequacies of their theories. But it is to say that the mere existence of apparently aberrant phenomena does not refute historical materialism. What makes Marxism, despite its many "crises," still the most powerful and challenging research program in the social sciences is precisely the strength and implausibility of its basic claim, namely, that the variety and complexity of social life is to be explained ultimately in terms of the prevailing forces and relations of production. Intuitively this is no more credible a claim than was Galileo and Newton's aspiration to mathematize nature in the seventeenth century. Modern physics' success in accounting for phenomena that seem to refute it sets a model for Marxism's own ambition. Giddens, however, seems to believe that the very failure to build into the heuristic of historical materialism forms of "exploitation" other than the extraction of surplus- labor counts against it. In doing so, he implicitly counterposes to Marxism what Erik Olin Wright calls "a contingent causal pluralism,"125 in which inter-state conflict, and racial and sexual inequalities are of equal explanato- ry importance to class exploitation. His objection is thus indicative more of

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

his own presumptions about what a true social theory would be like than of any inherent defect in Marxism.

Giddens's final objection to Marxism is also more revealing of his own views

than it is of those he criticizes. The evolution of "state socialism," he claims,

shows that "Marx's thought supplies precious few clues indeed about the

continuing significance of power in socialist society."126 It all depends, of

course, on what you mean by socialism. Giddens's main discussion of state

socialism, not, to my knowledge, subsequently repudiated, is in The Class

Structure of the Advanced Societies, where it is defined as "any economic

order in which the means ofproduction isformally socialized in the hands of

the state."'27 It follows from the premises of Giddens's theory of class that

state socialism thus conceived cannot be a class society. For such a society

requires the insulation of economy and polity that will permit market forces

unrestrainedly to allocate people to their position in society. However, in

state socialism, "the forms of differentiation in market capacity which oper-

ates in capitalist societies are considerably modified."'28 Consequently, "the

institutional mediation of power created by the integration of economy and

polity ... does provide an escape route from class society."129 However, the

state socialist societies clearly involve a high degree of social and economic

inequality and political oppression. Since Marx explained phenomena of

this sort in terms of the division of classes on the basis of exploitive relations

of production, the existence of state socialism must count as a refutation of historical materialism.

An adequate reply to this argument would take too long, and would involve tramping over very well worn ground.130 Let us simply note that this conclu- sion follows from Giddens's premises, not Marx's. In the first place, as we have seen, Marx does not ground class in market-relations, but rather on the distribution of the means of production. These depend, not on juridical property-forms, but on relations of effective possession. Secondly, it is clear from The Civil War in France and his other political writings, that for Marx the decisive step in the transition to a classless society came not with the introduction of state ownership of the means of production, but with the replacement of the repressive state apparatus by new forms of workers' power based on a popular militia and the election and recall of public officials. This theory may or may not be mistaken, but it offers quite a different account of socialism to that provided by Giddens in refutation of Marx. It has been pointed out by Marxists from Engels and Williams Morris onward that state ownership of the means of production is quite compatible with the persistence of class exploitation, and that private capitalism could under certain conditions be replaced by state capitalism. Giddens is unable to

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

161

All use subject to https://about.jstor.org/terms

162

envisage such a possibility, partly I think because of his tendency to identify capitalism with the prevalence of commodity production. Once again, a misreading of Marxism prevents him from engaging with the real issues.

The Rhetoric of Synthesis

In conclusion, I wish to offer some remarks about the nature of Giddens's

project. Since the 1960s a number of younger sociologists have sought to

integrate some elements of Marxism into their thinking, prompted to do so

by a variety of reasons, notably the inability of what Giddens calls the

"orthodox consensus" to handle the phenomenon of social conflict, and the

burgeoning of such conflict in the world outside sociology departments.131

Often Such pillaging of Marxism is accompanied by its disdainful dismissal.

Giddens's own work is a version of this approach, albeit methodologically

far more sophisticated, and always respectful of, and increasingly sympathet- ic to Marxism.

Let us note what such an attempt to arrive at a synthesis of mainstream sociology and Marxism implies. It supposes that the interpretive sociology of Weber, to name the most intellectually powerful and suggestive of the mainstream traditions, itself a brilliant synthesis of hermeneutics, histori- cism, Nietzsche's genealogy of power, and marginalist economics, can be rendered compatible with elements at least of historical materialism. It seems to me that there are general grounds for doubting this to be so: the two traditions are so fundamentally opposed in method, substance, and political implications that the outcome of any attempt to combine them can only be incoherence, eclecticism, or the denaturing of one or both. As evidence for this claim I would cite the way in which Giddens continues to give primacy to certain fundamental Weberian premises - the priority of subject over struc- ture, and the omnipresence of domination.

The result is a highly unstable system, since Giddens is clearly genuinely sympathetic to Marxism, owes it many insights, and wishes to find a place for it in his synthesis. This central difficulty - Giddens's underlying commitment to interpretive sociology combined with his openness towards Marxism - helps to explain his bitter hostility to Parkin's Marxism and Class Theory.

For Parkin is concerned above all to demonstrate the incompatibility of Marxism and Weberian sociology, and to advocate the superiority of the latter. Giddens rejects this claim, yet shares the same fundamental premises as Parkin - above all, they both believe relations of domination and subordi- nation and social conflict to be ineradicable features of human existence.132

No wonder that Giddens attacks Parkin so vehemently, for the latter is

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

merely pointing out the simple but unwelcome truth that he cannot have his cake and eat it.

The attempt to strike an equipoise between Marxism and mainstream soci-

ology seems to me not only impossible, but to lead often to an intellectually rather dubious approach to empirical questions. This is most evident, not in Gidden's work itself, but in a book by John Scott, Corporation, Classes and Capitalism, which is clearly considerably influenced by him. The book is an examination of the thesis that twentieth-century capitalism is characterized

by the separation of ownership and control. Scott contrasts two opposing theories, the "theory of industrial society" (Berle and Means, Bell, Aron),

which asserts this thesis, and the "theory of capitalist society" (i.e. Marxism),

which denies it. After showing conclusively that the thesis is false, Scott

claims that both theories are mistaken! He is able to do so because he

identifies the "theory of capitalist society" with clearly inadequate Soviet accounts of state monopoly capitalism. Nevertheless the evidence he cites

demonstrates the basic Marxist proposition that an economically dominant

class continues to exist in late capitalism, and draws considerably on th

research of non-Soviet Marxists. A position is thus attained au dessus la

melee, which puts both Marxism and orthodox sociology in its place, but

only at the price of grossly misrepresenting the former. Giddens's "contem-

porary critique" seeks a similar vantage-point by means of such a misrepres- entation of Marxism.

We may say, in conclusion, that despite Giddens's often brilliant intellectual fireworks, Marxism remains to be transcended. Both its theoretical and political revival, and the economic crises and social conflicts of the past fifteen years demonstrate its continued relevance. It is, I suppose, some sort of tribute that Marxism is considered worthy of inclusion in a grander synthesis by someone as talented as Giddens. Unfortunately, it may not prove very easy to digest.

NOTES

I have used abbreviations when citing the following books by Anthony Giddens - T Structure of the Advanced Societies (2nd ed., London: Hutchinson, 1981), CSAS; New Sociological Method(London: Hutchinson, 1976), NRSM; Central Problems in Social T (London: Macmillan, 1979), CPST; A Contemporary Critique of Historical Mater (London: Macmillan, 1981), CCHM; Profiles and Critiques in Social Theory (Lond millan, 1982), PCST.

1. E. O. Wright, "Giddens's Critique of Marxism," New Left Review 138 (1983) 12 2. See, for example, M. Foucault, Power/Knowledge (Brighton: Harvester 198 Deleuze and F. Guattari, Mille Plateaux (Paris: Minuit, 1980), and F. Parkin, Mar

and Class Theory (London: Tavistock, 1979).

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

163

All use subject to https://about.jstor.org/terms

164

3. CSAS, 296.

4. A. Giddens, "Classes, Capitalism, and the State," Theory and Society, 9: 6 (1980),

reprinted in PCST The same issue contains Parkin's equally bitter "Reply to Giddens." 5. CCHM, 3. See also A. Giddens, "From Marx to Nietzsche?," in PCST, especially 229. 6. CCHM, 30.

7. See, for example, CPST, 49ff. 8. Ibid., 49.

9. Ibid., 112.

10. NRSM, 15-16.

11. Ibid., 160, 102.

12. CPST, 3.

13. Ibid., 106.

14. Ibid', 141.

15. Ibid., 131.

16. Ibid., 211.

17. Ibid., 5.

18. A. Giddens, "Hermeneutics and Social Theory," in PCST, 8.

19. F. R. Dallmayr, "The Theory of Structuration: a Critique," in PCST, 19.

20. CPST, 67.

21. For these and other criticisms of Winch, see NRSM, 44-51.

22. P. Anderson, In the Tracks of Historical Materialism (London: Verso, 1983) 40-2. See

Ibid., 60-67 for a discussion of the similarities, and differences, between Habermas and

post-structuralism. 23. Ibid., 43-45.

24. See, S. Cavell, The Claim of Reason (Oxford: Oxford University Press, 1979), S. M. Holtzmann and C. M. Leich, eds., Wittgenstein: to Follow a Rule, (London: Boston and Henley, Routledge and Kegan Paul, 1981), and S. Kripke, Wittgenstein on Rules and Private Language (Oxford: Basil Blackwell, 1982).

25. J. McDowell, "Non-Cognitivism and Rule-Following," in Holtzmann and Leich, 144. 26. P. F. Strawson, "Sense and Consensus," Times Literary Supplement 11 February 1983. 27. L. Wittgenstein, Philosophical Investigations (Oxford: Basil Blackwell, 1968) 201.

28. A. Giddens, "Action, Power and Structure," in PCST, 31.

29. Dallymayr, "Theory," 22.

30. Ibid., 27.

31. CCHM, 171. See generally ibid., 157-81, and Giddens, "From Marx to Nietzsche?" 32.

CCHM, 171.

33. See N. Poulantzas, State, Power, Socialism (London: NLB, 1978) 146-53, and P. Dews,

"Nouvelle Philosophie and Foucault," Economy and Society 8: 2 (1979).

34. P. Anderson, Arguments within English Marxism (London: NLB, 1980) ch. 2.

35. Giddens, "From Marx to Nietzsche?," 220.

36. A. Giddens, "Power, the Dialectic of Control and Class Structuration," in PCST.

37. Ibid., 197-98.

38. E. O. Wright, Class, Crisis and the State (London: NLB, 1978) 98ff.

39. Giddens himself makes this point against Foucault: "From Marx to Nietzsche?," 222-23. 40. See CPST, ch. 6, CCHM, chs. 1, 3, 4.

41. CCHM, 38.

42. CPST, 103.

43. CCHM, 90.

44. CPST, 201-6.

45. J. Elster, "Marxism, Functionalism and Game Theory: The Case for Methodological

Individualism," Theory and Society II: 4 (1982).

46. J. Elster, Logic and Society (London: John Wiley & Sons, 1978).

47. CPST, 139-41.

48. Ibid., 59-69.

49. Ibid., 92.

50. For example, CCHM, 53.

51. J. Habermas, "A Reply to my Critics," in J. B. Thompson and D. Held, eds., Habermas:

Critical Debates (London: Macmillan, 1982) 286. 52. CCHM, 3.

53. NRSM, 110. 54. Ibid., 111. 55. Ibid.

56. CPST, 43. 57. Ibid.

58. CCHM, 51-52. 59. Ibid., 28-29. 60. Ibid., 107.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

61. Ibid., 108.

62. CPST, 162.

63. See A. Giddens, "'Power' in the Recent Writings ofTalcott Parsons," in Giddens, Studies

in Social and Political Theory (London: Hutchinson, 1977), and "Labour and Interac- tion," in PCST.

64. CCHM, 28-29.

65. Ibid., 53.

66. See, for example, R. Harre and E. Madden, Causal Powers (Oxford: Basil Blackwell,

1975).

67. M. Foucault, "The Subject and Power," afterword to H. L. Dreyfus and P. Rabinow,

Michel Foucault (Brighton: Harvester, 1982) 217-18. 68. CCHM, 20.

69. Ibid., 80.

70. Ibid., 161-66, CPST, 76ff.

71. CCHM, 145.

72. CSAS, 134.

73. Ibid., 298 (1979 Postscript).

74. CCHM, 165.

75. Ibid., 147.

76. See A. Callinicos, Is There a Futurefor Marxism?(London: Macmillan, 1982) 142-47, and

Marxism and Philosophy (Oxford: Clarendon Press 1983) 110-13.

77. See Callinicos, Marxism and Philosophy, ch. 2.

78. Wright, "Giddens's Critique of Marxism," 25-34.

79. Ibid., 26.

80. CCHM, 157.

81. Wright, "Giddens's Critique of Marxism," 31.

82. CCHM, 95.

83. D. Schmandt-Besserat, "The Earliest Precursor of Writing," in Prehistoric Times: Read-

ings from "Scientific American," ed. B. M. Fagan (San Francisco: W. H. Freeman and

Company, 1983).

84. See, Callinicos, Future, chs. 6 and 7.

85. K. Marx, Capital iii. (Moscow: Progress, 1971) 782-813.

86. Ibid., 790-91.

87. See B. Hindess and P. Hirst, Precapitalist Modes of Production (London: Routledge and

Kegan Paul, 1975).

88. G. Bois, Crise du feodalisme (Paris: Presses de la Fondation Nationale des Sciences

Politiques, 1976).

89. K. Marx, Capital i. (Harmondsworth: Penguin, 1976) 899.

90. CCHM, 122. See generally Ibid. 107-28.

91. Ibid., 157-81.

92. CSAS, 84.

93. Ibid., 131.

94. CCHM, 113-17.

95. Marx, Capital i. 270-80.

96. CSAS, 298-99 (1979 Postcript).

97. CCHM, 130ff.

98. Ibid., 116.

99. See K. Marx and F. Engels, Collected Works ((50 vols.) London: Lawrence and Wishart,

1975-) vi. 197 and xxxviii. 99, Hindes and Hirst passim, and G. A. Cohen, Karl Marx's

Theory of History - a Defence (Oxford: Clarendon Press, 1978) 217-25.

100. Marx, Capital i. 176 n.

101. I. Lakatos, Philosophical Papers (2 vols., Cambridge: Cambridge University Press, 1978), especially i. ch. 1.

102. For a similar use of the distinction between heuristic and auxiliary hypotheses, see A. Gilbert, Marx's Politics (Oxford: Martin Robinson, 1981) 5-13, and "The Storming of Heaven: Politics and Marx's Capital," in J. R. Pennock and J. W. Chapman, Marxism: Nomos XXVI (New York: New York University Press, 1983) 151-61.

103. P. Anderson, "Class Struggle in the Ancient World," History Workshop 16 (1983), 58. 104. G. E. M. de Ste. Croix, The Class Struggle in the Ancient Greek World (London:

Duckworth, 1981) ch. II.

105. See especially the critiques of functional explanation in J. Elster, Ulysses and the Sirens

(Cambridge: Cambridge University Press, 1979), and "Marxism, Functionalism and

Games Theory."

106. G. A. Cohen, "Reply to Elster on 'Marxism, Functionalism, and Game Theory,"' Theory

and Society, 11: 4 (1982) 489, 483. 107. CCHM, 250.

108. Ibid., 190.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

165

All use subject to https://about.jstor.org/terms

166

109. Ibid., 198.

110. Ibid., 83.

111. L. Trotsky, The History of the Russian Revolution (3 vols., London: Sphere, 1967), i. 23.

See G. Novack, Understanding History (New York: Monad. 1972), B. Knei-Paz, The

Social and Political Thought of Leon Trotsky (Oxford: Clarendon Press, 1978), 62-107,

and M. L6wy, The Politics of Combined and Uneven Development(London: NLB, 1981) chs. 1-3.

112. CSAS, 153.

113. CCHM,279n. 16.

114. Ibid., 275 n. 5.

115. See C. von Braumuhl, "The Analysis of the Nation-State within the World Market

Context," in State and Capital, eds. J. Holloway and S. Picciotto, (London: Arnold, 1978), C. Barker, "The State as Capital," International Socialism 2: 1 (1978), and C. Luporini, "'Politique' et 'Etatique"' in Marx et la critique de la politique, E Balibar et al.

(Paris: Maspero, 1979).

116. See T. Vance, The Permanent Arms Economy (Berkeley, 1970), M. Kidron, Western

Capitalism since the War (Harmondsworth: Penguin, 1970), and C. Harman, Explaining

the Crisis (London: Bookmarks, 1984).

117. P. Anderson, Considerations on Western Marxism (London: NLB, 1976) 44-45.

118. See, for example, H. Draper, Karl Marx's Theory of Revolution 2 vols. (New York:

Monthly Review Press, 1977 and 1978), Gilbert, Marx's Politics, T. Cliff, Lenin 4 vols. (London: 1975-79), N. Harding, Lenin's Political Thought 2 vols. (London, Macmillan:

1977 and 1981). 119. CCHM, 3, 24-25.

120. Ibid., ch. 9.

121. Ibid., 177.

122. Giddens, "From Marx to Nietzsche?," 218.

123. CCHM, 161.

124. Ibid., 242.

125. Wright, "Giddens's Critique of Marxism," 32.

126. CCHM, 244.

127. CSAS, 155.

128. Ibid., 230.

129. Ibid., 282.

130. See especially T. Cliff, State Capitalism in Russia (London: Pluto, 1975). 131. Giddens, "Hermeneutics and Social Theory," 1-3.

132. Compare CSAS, 283 and Parkin, Marxism 187-92.

Theory and Society 14 (1985) 133-166

03404-2421/85/$03.30 ? 1985 Elsevier Science Publishers B.V.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

well as constraining, so they do not simply limit the scope of human conduct, but make certain forms of activity possible. More relevant for present purposes is the fact that for Giddens the existence of structures does not in itself explain the persistence of social systems. The belief to the contrary he regards as "the decisive error in functionalism."'6 According to a thesis that Giddens calls "the duality of structure," "structure is both medium and outcome of the reproduction of practices."17 In other words, although structure shapes the conduct of human agents, it is only through that conduct that it itself possesses any effectivity, and it may itself be modified by the activity of which it is the unacknowledged condition and unanticipated consequence.

Giddens illustrates this argument by invoking the case of language. The rules of a language constitute langue, the underlying structures that governparole, speakers' actual usage. But it only by virtue of being applied in speech that these rules become operative; withoutparole langue would not exist. Further- more, it is always possible that usage will alter the rules of language; indeed, language is constantly changing as a result of shifts of usage. Similarly, it is through the conduct of "knowledgeable human agents" that structures are produced, reproduced, and transformed.

The "duality of structure" thus formulated serves to conceptualize social systems as the outcome of an interaction between underlying structures and intentional conduct. Giddens is obviously sensitive to the arguments ad- vanced especially by structuralism and post-structuralism that the subject can no longer be conceived, as it has been since Descartes, as the autonomous source of meaning. Yet he hopes to accommodate these arguments within a position that does not simply turn the subject into a mere "bearer" of anonymous structures, as Althusser, for example, did. "A de-centring of the subject must at the same time recover that subject as a reasoning, acting being."'8 As Fred R. Dallmayr observes: "Giddens has seen himself faced with a momentous challenge: ... the task of moving beyond subjectivist metaphysics without relinquishing some of its insights, and especially with- out lapsing into objectivism and determinism."19

The attractions of such a position are obvious, avoiding as it does the extremes of a functionalism or structuralism that denies any scope to the human subject, and a hermeneutics that can find no place for social struc- ture. Or rather, it would be attractive had Giddens successfully avoided these extremes. I wish to argue that he has not, and that his commitment to the notion that human agents produce society gives rise to a tendency to collapse into the heremeneutic pole of the two extremes.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

137

All use subject to https://about.jstor.org/terms

138

Part of the difficulty lies in Giddens's identification of social structures with rules. "Rules," he tells us, "generate - or are the medium of - the production and reproduction of practices."20 Such a treatment of rules as constitutive of social practice, suggested already by his invocation of an analogy with language, is, of course, far from peculiar to Giddens. It was, indeed, typical of much Anglo-American philosophy in the 1950s and 1960s under the influ- ence of a certain reading of the work of the later Wittgenstein. Winch's Idea of a Social Science represented the apotheosis of this trend, as applied to social theory. Now Giddens wishes to distance himself from social science a la Winch, if only because he believes that structure involves, as well as rules, resources whose unequal distribution gives rise to power-relations.21 Never- theless, the conception of social practice as rule-governed is crucial to his theory of structuration.

There seem to me two reasons for questioning this approach. First, it involves generalizing from the case of language to social practice as such. Such a treatment of language as paradigmatic of social life is again not peculiar to Giddens - apart from in linguistic philosophy of the post-war Oxford school, it is also to be found in Levi-Strauss's structural anthropol- ogy, in Habermas's recent work, and in deconstructionists of the Derrida sort.

Of course, these various approaches embody, on the whole, very different conceptions of language - compare the Derridean treatment of discourse as inherently heterogenous and self-subverting, and Habermas's discovery, implicit in every speech-act, of a tendency toward rational consensus. Never- theless, they share a proneness to what Perry Anderson has recently called the "exorbitation of language," the identification of the structures of lan- guage with those of society.22 But, as Anderson persuasively argues, "lan- guage is no fitting model for any other human practice": linguistic structures change very slowly, are not subject to the constraints of natural scarcity, and exist through individual rather than collective agents.23 Anderson shows how the overextension of language has led in the case of post-structuralism to a point where neither history nor indeed reality itself have any standing outside the endless play of signifiers.

These are, however, general considerations. There is a second reason for quarrelling with Giddens's identification of structure with rules. Recent discussion, notably by Saul Kripke, of the passages in the Investigations and Remarks on the Foundations of Mathematics where Wittgenstein examines the notion of following a rule suggests that in these texts, far from conceiving social practice as rule-governed, he sets out to attack the idea that, as John

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

McDowell puts it, "the rules mark out rails along which correct activity within the practice must run."25

The difficulty with rule-following thus conceived is well stated by P. F. Strawson in this limpid summary of Kripke's reading of Wittgenstein:

It is natural to suppose that when we use a word or symbol of our language, we are guided in its use by our grasp of its meaning, or of the concept which it expresses, or of the rules or instructions for its use which we have mastered; that these are what tell us that it is correct to use the expression in such-and-such a way; to apply it to this case (if it is a descriptive term) or to compute with it in this way (if it is a mathematical symbol). So we appear to invoke some fact about our mental life to explain our confidence in the correctness of our current use of the expression. But if we take this conception seriously, it seems that there can be no guarantee that what we now take ourselves to mean by the expression is the same as what we meant by it in the past. For our practice is consistent with our having meant by it something quite different... the point ... should be obvious to any reader of Nelson Goodman: perhaps by "green" in the past I meant grue (where anything green seen before now and anything seen blue from now is grue).26

The conclusion of this sort of sceptical consideration is what Wittgenstein describes as the following paradox: "no course of action could be determined by a rule, because every course of action can be made to accord with it."27 Rather than grounding social practice, rules depend for their applicability on the shared responses of members of a community. It is on the basis of this interpretation of rule-following that Wittgenstein argues, in the later parts of the Investigations, that a private language, i.e., one that in principle only one speaker could use, is impossible, since the only criterion of correct usage is that provided by the common action of members of a society.

If this argument is valid, and it has still to be shown that it is not, then Giddens cannot appeal to the notion of rules as the basis of his account of social structure. Rather than generating practices, rules collapse into them. This is not great worry if one accepts, as I do, a much stronger sense of structure than that proposed by Giddens. It merely provides persuasive backing for my opposition to generalizing from the case of language. But, for Giddens, who wishes to preserve the hermeneutic tradition's sensitivity to the interpreting and acting subject while recognizing the structural aspects of social practice, it is a serious difficulty.

Giddens shows no sign of being aware of this difficulty. This is especially surprising in the light of the fact that he is evidently acquainted with the issues raised by Wittgenstein's discussion of rule-following. Indeed, he writes that "the knowledgeability involved in practical consciousness conforms generally to the Wittgensteinian notion of'knowing a rule' or'knowing how to go on."'28 Yet, in the exchange with Dallmayr published in Profiles and

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

139

All use subject to https://about.jstor.org/terms

140

Critiques, Giddens ignores the former's demand for"further argument... to pinpoint the status of rules in the process of structuration."29

Dallmayr goes on to argue that, although Giddens has "told us repeatedly that he wants to move beyond the so-called subject-object dichotomy," in the theory of structuration, "he reverts again to the subject-object dualism of actors constituting and being constituted."30 That this is so, that there is a persisting dualism of agency and structure in Giddens's writings despite his claim to have overcome it, can be illustrated by his discussion of two issues. In each he tends to treat agents and structures as separate from, and external to each other, and to accord primacy to the former.

The first is the question of resistance by subordinate groups. Giddens attacks

Foucault's notion of "history without a subject" on the grounds that "human

beings are always and everywhere knowledgeable human agents, though

acting within historically specific bounds of the unacknowledged conditions

and unaccepted consequences of their acts."31 He proceeds to criticize

Foucault's notion of society as constituted by a pervasive and omnipresent

apparatus of "power-knowledge" by arguing that "Foucault's 'archaeology,'

in which human beings do not make their own history but are swept along by

it, does not adequately acknowledge that those subject to the power of

dominant groups are themselves knowledgeable human agents, who resist,

blunt or actively alter the conditions of life that others seek to thrust upon them."32

Now this argument is closely related to the well-known difficulty in

Foucault's recent work that, although resistance is always supposedly im-

manent in power-relations, he never provides any basis for this resistance.33

Giddens seeks to provide such a basis by invoking the concept of "knowl-

edgeable human agents." This is a completely ahistorical approach. There is no discussion of the historically specific conditions that lead oppressed

groups to resist, and that provide their resistance with sources of organiza- tion and power. It is merely a general property of"knowledgeable human agents" that they "resist, blunt or actively alter" their "conditions of life." The strategy recalls Edward Thompson's counter-position in The Poverty of

Theory of the creativity and humanity of social agents to the structural

conditions of their activity. Perry Anderson's response to Thompson seems

to me applicable to Giddens as well: the scope for agency (and hence

resistance) in history varies according to the specific circumstances in which

people find themselves.34 Giddens's preference for a philosophical invocation

of subjects' capacity to resist rather than for a historical examination of the variable conditions of action is indicative of his privileging of the pole of

agency as against that of structure in the theory of structuration.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

Support for this view is provided by a highly symptomatic misreading of Foucault on Giddens's part. He writes: "Discipline, Foucault says, disasso- ciates power from the body... At the same time, the emphasis is placed on the 'interiorization' of power."35 Foucault, of course, says no such thing. The rise of the disciplines in early nineteenth-century Europe indeed represents a shift

from the form of power characteristic of absolutism, involving the spectacu- lar use of violence in public executions. But the new form of power, in which individuals are subjected to general norms within the context of institutions such as prisons, according to Foucault, also acts on bodies. In this case, however, it serves to constitute them as subjects, providing them with a set of motives governing their actions. Giddens's talk of "the 'interiorization' of

power" reveals his own presupposition that structures act on independently formed subjects. But, for Foucault, in the disciplines, power does not pene- trate the consciousness of a pre-formed subject; rather, these disciplines, create the subject.

The same presupposition underlies Giddens's critique of Harry Braverman's

Labour and Monopoly Capital.36 Giddens quite correctly points to the

weakness of Braverman's account of scientific management and the deskil-

ling of labor, namely that it treats workers as passive objects apparently

incapable of effective organization and resistance. However, this criticism is

theoretically grounded in the assertion that "to be a human agent is to have

power," so that "the most seemingly 'powerless' individuals are able to

mobilize resources whereby they carve out 'spaces of control'" giving rise to "a dialectic of control."37

All this is true enough, but it short-circuits the analysis of different modalities of resistance in different social formations. The scope for resistance of, for example, a slave in the Athenian silver mines at Laureion, while no doubt real enough, was surely much narrower than that of, say, the auto-workers at Ford Halewood studied by Huw Beynon. A Marxist would argue that the

difference between the two can only be understood by examining what Erik Olin Wright has called the structural capacities of particular classes - in other words, the distinctive powers that they derive from their position in histori- cally specific relations of production.38 Thus the shortcomings of Braver- man's analysis of the modern labor-process is closely related to his failure to grasp that workers' position within this process endows them with a capacity

for collective organization and resistance.39 Unfortunately, Giddens's pre-

occupation with an abstract account of human subjectivity prevents him

from following through the consequences of his own insight into the way in which social structures enable as well as constrain.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

141

All use subject to https://about.jstor.org/terms

142

A second index of Giddens's underlying preference for an account of society centering on the subject is his discussion of"time-space distanciation." One of the most interesting and innovative aspects of Giddens's recent work is his insistence on integrating the spatial and temporal dimensions of human life into social theory. Recent work in human geography, for example, should be treated, not as part of a separate discipline from sociology, but as an intrinsic part of the study of social formations.40

I have neither the competence nor the desire to question Giddens's often exhilarating discussions of these matters, merely to draw attention to one feature of the conceptual framework he uses. He draws on Heidegger and Derrida in order to argue that "all social interaction intermingles presence and absence."41 Human beings have direct contact with each other as present in space and time. However, they also have access to those distant in time through oral tradition, writing, and other such storage devices, and to those distant in space by various communications systems. Social structures serve to bind practices together in time and space through their contribution to the constitution and reproduction of systems of social interaction. Giddens goes on to distinguish between societies on the basis of the degree of their "presence-availability." In "societies or communities of high presence avail- ability ... interaction is predominantly of a face-to-face kind."42 Such socie- ties are to be found, for example, among hunter-gatherers. Giddens argues that history is characterized by a process of "time-space distanciation," such that "societies are 'stretched' over longer spans of time and space."43 This occurs as both communications and storage devices become more sophisti- cated, so that people interact with those with whom they are not in face-to- face contact. The culmination of this process is the emergence of the capital- ist world-system.

I shall discuss this analysis, insofar as it supposedly contradicts Marxism, below. I wish merely to note here that Giddens's argument differs significant- ly from the conception of Being as presence-absence developed by Heidegger and Derrida. For the latter argue that the condition of "high-presence availability" that Giddens attributes to band-societies is impossible. They do so for general philosophical considerations: namely, they wish to deny that a state of parousia or absolute presence in which we have direct access to reality can exist. We can only relate to such a state as necessarily absent, something that we have lost or may still hope to (re)gain. This is the point of

Derrida's concept of differance, of a constant play of presence and absence in which we are obliged to postulate, but never to attain, presence.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

Of course, Giddens is perfectly entitled to appropriate and modify other people's concepts. But it is the nature of the modification that is interesting. One of the points of the concept of differance was precisely to dislodge from his sovereign place the Cartesian subject, having direct access to the contents of his consciousness (in other words, being "self-present"), and providing a basis on which our knowledge of the world (and indeed on some versions the world itself) could be constituted. Giddens does not reinstate the Cartesian self - rather it is the subject of social interaction, engaging in face-to-face contact with other subjects in the manner analysed by Goffman, which reigns in the condition of "high presence-availability."44 History becomes, as Derrida argues it must necessarily be, the story of the loss of presence, of the "high presence-availability" characteristic of hunter-gatherer societies. Once again, "knowledgeable human agents" triumph over structures.

These two cases - Giddens's discussion of resistance and his account of "time-space distanciation" - suggest that he has failed to overcome th duality of agency and structure. The reason for this failure is not a contingent one - it is not a matter of Giddens being inconsistent. Rather, in my view, it source lies in the very conception of structure as the unacknowledged condi- tion and unanticipated consequence of human actions.

Let us consider structure in its two aspects. Conceiving structure as th unanticipated consequence of human action does not involve any challenge to the primacy of agents over structures. On the contrary, such a conception of structure has usually been held precisely as a means of securing this primacy. This is most obviously so in the case of methodological individual ism. If social structures arise as the unintended consequences of individual actions, then it is with these actions, and the beliefs, desires, and intentions that produce them, that we must start in order to account for the existence of structures and of the practices bound up with them. Jon Elster, who recently urged in the pages of this journal that methodological individualism can provide the "microfoundations of Marxist social theory,"45 has, in line with this approach, developed an account of social contradiction as unintended consequence.46

That Giddens is committed to a stronger notion of structure than Elster's is suggested by his critique of the latter's theory of contradiction.47 And, indeed, the concept of structure as the unacknowledged condition of human action seems to go much further than any account in terms of unintended consequences, especially when Giddens talks of structure as both constrain- ing and enabling. But when we consider the detail of Giddens's account structure is once again reduced to a secondary role. Structures are, let us

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

143

All use subject to https://about.jstor.org/terms

144

recall, rules and resources. Even though Giddens does not fully grasp the import of Wittgenstein's arguments about rule-following, he does recognize rules' dependence on the practices that they supposedly govern.48 So rules only "generate" practice in a very weak sense.

The other aspect of structure is resources, which are either authoritative or allocative, depending on whether they give command over persons or objects respectively. I shall have more to say about them when I come to discuss Giddens's account of power. What I wish to stress here is that he describes resources as "media" or "means" that are "utilized" by agents seeking their own ends, and thereby (unintentionally) reproducing society.49 Structure thus understood in no way constitutes social action. It is conceived instru- mentally, as a tool used by agents in order to realize their wants.

Surprisingly, since Giddens often belabors Marx for identifying human praxis with the material work of transforming nature,50 his is a remarkably economic conception of practice, in which agents use scarce resources to achieve their ends. There seems to me no way of genuinely distinguishing this position from Popper's, whose methodological individualism is precisely a generalization of neo-classical economics's model of rational action.

Habermas's recent comments on Giddens go to the heart of the matter. Giddens's "concept of practice," Habermas writes,

is supposed to be related to the constitution of action-complexes and to their reproduction. I do not think that this choice of conceptual strategy is a fortunate one, because the basic epistemological concept of'constitution', which refers to the formation of object-domains, causes confusion in social theory. It suggests that speaking and acting subjects 'produce' their social life-context in a way similar to that in which they make products of instrumental action [i.e., labor]... The over extension of action-concepts in a theory of constitution remains stuck in metaphors; the basic concept of the philosophy of Praxis gives at best an anthropomorphistic concept of society.51

Habermas has put his finger on the central fault in Giddens's theory of structuration. The way in which Giddens conceptualizes structure means that it can only function as a secondary aspect of social practice, subject to the creative interventions of "knowledgeable human agents." The result, as Habermas points out, is "an anthropomorphistic concept of society," in which the subject reigns supreme, producing society as labor produces use-values. Far from overcoming the dualism of agency and structure, Giddens is stuck firmly at the pole of agency. The much-invoked "duality of structure" amounts to little more than a substitution of two letters - a verbal change.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

The Paralogisms of Power

Similar conceptual sleight of hand characterizes Giddens's account of power. The concept of power is central to his critique of Marxism. Thus,

a fundamental component of my arguments is that the articulation of space-time relations in social systems has to be examined in conjuction with the generation of power. A preoccupa- tion with power forms a leading thread of this book. I maintain that power was never satisfactorily theorized by Marx, and that this failure is at the origin of some of the chief limitations of his scheme of historical analysis.52

There are three main steps in Giddens's conceptualization of power. First, he argues that "the notion of 'action' ... is logically tied to that of power... Action intrinsically involves the application of 'means' to achieve outcomes, brought about through the direct intervention of an actor in a course of events."53 "Power" is understood here as "the transformative capacity of

human agency," that is, as "the capability of the actor to intervene in a series

of events so as to alter their course; as such it is the 'can' which mediates between intentions and wants and the actual realization of the outcomes sought after."54

The second step involves introducing a "narrower, relational sense" of "power," as a "property of interaction," "power as domination."55 It arises "where transformative capacity is harnessed to actors'attempts to get others to comply with their wants."56 The result is "structures of domination," which "involve asymmetries of resources employed in the sustaining of power-relations in and between systems of interaction."57 Giddens distin- guishes between two main forms of domination, allocation and authoriza- tion, depending on which type of resource is unequally distributed. Alloca- tive resources comprise "material features of the environment," "material means of production/ reproduction," and "produced goods." Giddens's account of authoritative resources, which consist in the "organization of social time-space," "production/reproduction of the human body," and "organization of human life-chances," is a good deal less perspicuous: these resources are something of a residual category.58 Nevertheless, the gist is clear enough: allocation is essentially economic domination, authorization politico-ideological.

The third step is to use these concepts to offer a theory of historical develop- ment. Giddens claims that power and domination are inherent in human social life.59 The chief variation is in the form of domination. Prior to

capitalism, Giddens argues, authorization was the prevailing form. He dis- tinguishes between class-divided and class societies. In both categories

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

145

All use subject to https://about.jstor.org/terms

146

classes, i.e., "the sectional forms of domination created by private ownership of property,"60 are to be found. However, in a class-divided society "class analysis does not serve as a basis for identifying the basic structural principles of organization of that society."61 Class-divided societies compromise all "non-capitalist civilizations" - city states, empires, and feudal societies. The structures of domination in these societies depend upon the distribution of authoritative resources. "In class-divided societies, as opposed to class society, authorization has primacy over allocation."62 Capitalism is the only case of a class society; here alone does allocation prevail. The only limited validity historical materialism possesses is in its analysis of capitalism, where economic factors do predominate over ideological and political ones.

I wish to criticize this account of power, but let me first stress its virtues. Giddens's is not a harmonistic theory of society. At its center is contradic- tion, power, and struggle. This is its merit, compared with those of Parsons and Habermas, for whom power serves primarily to integrate social systems, and to dampen down social conflict.63 Giddens's is a view of society having much in common with Weber's and, as I shall try to show, it is with Weber that he is best compared, despite his assertions to the contrary.

A crucial element in Giddens's strategy is the shift from the broad to the restricted conceptions of power, from power as transformative capacity to power as domination. He is quite explicit in insisting that it is the broad concept of power that grounds the claim that domination is inherent in social life. For example, "all social interaction involves the use of power, as a necessary implication of the logical connection between human action and transformative capacity. Power within social systems can be analyzed as relations of autonomy and dependence between actors in which these actors draw upon and reproduce structural properties of domination."64 Similarly, "transformative capacity... forms the basis of human action - the 'could have done otherwise' inherent in the concept of action - and at the same time connects action to domination and power."65

If we are to take these passages as suggesting that one may infer from the "logical connection" between transformative capacity and action to domina- tion as a persisting feature of social interaction, and that is what expressions such as "necessary implication" would lead us to believe, then Giddens is guilty of an obvious fallacy. I doubt if anyone would wish to question whether "action" is "logically tied" to "transformative capacity," but recogni- tion of the involvement of power thus understood in human action does not commit one to very much. For "transformative capacity" understood as "the capability ... to intervene in a series of events so as to alter their course" is a

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

property of all causal agents, not merely human ones. Indeed, "transforma-

tive capacity" seems in reality to be merely another expression for "causal

powers," since such powers consist pecisely in the ability to bring about some

alteration in the course of the world.66 It is true that Giddens makes much of

"the 'could have done otherwise' inherent in the concept of action," presum-

ably referring to the unique property of human beings to reflect upon and

choose between alternative interventions. But this consideration does not

bear on the nature of "transformative capacity" but rather on human agents' peculiar ability consciously to control and monitor the exercise of their causal powers.

It is one thing to attribute to human beings this "transformative capacity." It is quite another to infer from this true but rather trivial proposition the much more interesting, but controversial (and, in my view, false) thesis that power in the sense of domination is intrinsic to human life. There is no necessary connection between the two concepts of power. Further premises would be required tojustify the inference from transformative capacity to domination. Interestingly enough, a similar point has been made recently by Foucault. He distinguishes between "objective capacity," "which is exerted over things and gives the ability to modify, use, consume, or destroy them," and "power, which brings into play relations between individuals (or between groups)." "Objective capacities" and "power-relations," Foucault insists, "should not

... be confused."67 Although Giddens distinguishes carefully between the broad and restricted sense of power, it is difficult not to feel that when he claims that "transformative capacity ... connects action to domination and power," he is guilty of an equivocation. The result is a paralogism, in which the same word, "power," is used in different senses in the premise and conclusion of the same argument.

It does not seem to me, for reasons that I will make clear shortly, that this elementary logical mistake is necessarily fatal for Giddens's account of domination. But since he makes so much of the notion of "transformative

capacity" it is worth pointing out how little use of this concept commits one to. This brings out a common feature of Giddens's writings, namely, that they are prolific of neologisms and, more frequently, of redefinitions of familiar words, without these new concepts, and the complex typologies often accompanying them, usually offering much gain in clarity or empirical reach. Central Problems in Social Theory is especially guilty of this, an arid treatise raining down definitions and diagrams on the bewildered reader's head. From this point of view A Contemporary Critique is especially to be welcomed, since here Giddens cashes his conceptual checks, putting the typologies of Central Problems to work in order to provide an account of historical development different from Marx's.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

147

All use subject to https://about.jstor.org/terms

148

It is here that one might quite reasonably say in Giddens's defense that his analysis of domination can stand on its own legs, without the support of "transformative capacity." The test of its validity will then turn on whether it explains a series of phenomena that Marxism cannot. This argument is a cogent one, and can only be answered by turning, at last, to the concrete detail of Giddens's critique of Marx.

Marxism and Evolutionism

The Marxism that Giddens seeks to transcend is an evolutionist one, in

which social change is an outcome of the level of development of the

productive forces. Such a view of historical development is characteristic of

Marxism in general: "Marxist authors are virtually everywhere committed to

evolutionism in one guise or another."68 Only in those sections of the

Grundrisse where he discusses pre-capitalist social formations (the so-called

Formen) does Marx himself adopt a different approach: "rather than imply-

ing that the forces of production have their own internal dynamic, as in his

evolutionary view, Marx seems in the Formen to give primacy to 'ecological'

factors (dispersal or concentration of populations) and to war in stimulating social transformation."69

Giddens seeks to disprove Marx's evolutionary theory, and substantiate his

own theory of domination by providing an account of historical develop- ment in which it is the distribution of authoritative resources that has

primacy until the development of capitalism. We have already seen that he believes history to be a process of time-space distanciation in which the high presence-availability of tribal societies is progressively lost. These changes are highly significant for the manner in which societies are integrat- ed. Giddens distinguishes between societal and system integration. Societal integration involves the relationships between individual social actors; systems integration concerns the connections between social groups and collectivities.70 In tribal societies the two forms of integration are fused, so that system-integration, the relations between groups, is dependent on the face-to-face contacts between individuals, and the traditions and kinship networks with which these contacts are intimately bound.

The two forms of integration are separated with the emergence of class- divided societies. Societal integration still depends on kinship and tradition, and small localized communities continue to form the basis of these societies.

However, system-integration depends increasingly on the power of the state, and the authoritative resources that it is able to mobilize. The extraction of

surplus-labor from these communities depends on the intervention of the

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

state, and takes the form of deductions - taxation, rent, etc. - leaving much of their daily life unaffected. The necessary authoritative resources to mount such control over local communities required the development of new forms of "information storage," such as writing, that in turn depended on the emergence of the city. It is the rise of the city, "the generator of the authorita- tive resources out of which state-power is created and sustained,71 not the development of the productive forces that is the decisive factor in theemergence of classes.

Capitalism, like "class-divided societies," separates societal and system inte- gration. However, the content of these two forms of integration is radically changed. The extraction of surplus-labor comes to depend, not on the direct intervention of coercive power, but rather on economic mechanisms, the pressures, analyzed by Marx in Capital, that force workers to sell their labor-power. Capitalism is thus characterized, as Giddens has earlier stressed in The Class Structure of the Advanced Societies, by "a 'separation' between economy and polity such that there is at least a substantial scope for the play of market mechanisms independent of active political control."72 But he now accounts for the "insulation" of economy and policy on the grounds that "only with the advent of capitalism does an exploitative class relation become part of the very mechanism of the productive process."73

The other side of the separation of state and market is the transformation of the state into "a much more intrusive and comprehensive set of institutions in capitalist than in class-divided societies, so far as those subject to its adminis- tration 'internally' are concerned."74 No longer is societal integration left to kinship and tradition. The dependence of the capitalist economy upon the capacity of producers continually to increase their output requires an enor- mous expansion of state-directed apparatuses of surveillance, by which Giddens means both the accumulation of information, and what Foucault calls the disciplines, the systematic supervision and regulation of the activi- ties of those involved in a variety of institutions - factories, armies, prisons. Consequently, the "crucible of power" is no longer the city, but the nation- state, a phenomenon whose significance Giddens believes Marxism largely to have ignored.75

Nevertheless, I suspect that most people, confronted with this account of historical development, would be puzzled as to why it should be regarded, with all its stress on changing modes of surplus-extraction, as in any sense critical of, or an alternative to Marxism. Giddens's reply would presumably be that Marxism accords primacy to the productive forces, attributing to them an "internal dynamic" that is the source of all historical change. Now it

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

149

All use subject to https://about.jstor.org/terms

150

is true that such is one version of Marxism, most eloquently defended recently by G. A. Cohen in his Karl Marx's Theory of History - a Defense. It is not, however, the only version of Marxism, nor is it, as I have argued elsewhere, the most adequate.76 Furthermore, I think it can be shown that it is a view that Marx himself progressively abandons.77

However, as Erik Olin Wright persuasively argues, rejection of evolutionist Marxism does not require the abandonment of any notion of historical evolution.78 The crucial distinction, he suggests, is that between teleological versions of historical materialism that conceive social formations as evolving towards a predetermined goal, and those that adopt a weaker and unobjec- tionable evolutionary theory. Wright urges that all such a theory must claim is that "there is some process, however weak and sporadic, which imparts a directionality to movements from one form to another."79 Indeed, he sug- gests, Giddens's own account of history as a process of time-space-distancia- tion, in which time and space are progressively stretched, and "the immedi- acy of presence"80 gradually lost, meets the conditions for an evolutionary theory: "Because of the link between conflict, power, resources and distanci- ation, there will be at least some impulse for increasing space-time distancia- tion."81

Giddens cannot in any case dispense with the notion of the development of the productive forces as easily he would like. Writing, which Giddens believes to have developed "as a means of recording information relevant to the administration of societies,"82 may have in fact originated in the clay tokens used by Neolithic farmers to keep track of their food stores.83 More general- ly, time-space distanciation crucially depends on technological innovations

- improvements in transport, new systems for transmitting information, etc. The stretching of space and time is in no sense a process autonomous of the development of the productive forces.

To argue in this manner is not, however, to accord causal primacy to the productive forces. Their development is best understood, in my view, as a necessary, but not sufficient condition of changes in social relations. Furth- ermore, the scope for developing the productive forces is determined by the prevailing relations of production. These relations are best conceived as a specific mode of appropriation of surplus-labor, which in turn depends upon the distribution of the means of production. Forces and relations mutually limit one another, rather than the one exerting primacy over the other.84

Now the relations of production thus conceived seem to me perfectly capable of accounting for the shift from authoritative to allocative domination on

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

which Giddens places such stress. Marx himself provides such an explana- tion in his crucial chapter, "Genesis of Capitalist Ground-Rent," in Capital volume 3.85 There he argues that

in all forms in which the direct labourer remains the "possessor" of the means of production and labour-conditions necessary for the production of his own means of subsistence, the property relationship must simultaneously appear as a direct relation of lordship and servitude, so that the direct producer is not free; a lack of freedom which may be reduced from serfdom with enforced labour to a mere tributary relationship. The direct producer, according to our assumption, is found here in possession of his own means of production, the necessary material labour conditions required for the realization of his labour and the production of his means of subsistence... Under such conditions the surplus-labour for the nominal owner of the land can only be extorted from them [the direct producers] by other than economic means.86

As a conceptualization of feudal relations of production this account is not

wholly unproblematic.87 Nevertheless, it has provided the basis of at least one

major empirical study of late feudalism.88 And it does provide a means for

accounting for the apparent primacy of "authoritative domination" in pre-

capitalist social formations. Surplus-labor can be appropriated by a ruling

class where the direct producers themselves control the labor-process only

through the systematic use of the means of coercion. By contrast, the

complete separation of the direct producers from the means of production

under capitalism accords primacy to economic mechanisms in the extraction

of surplus-labor. As Marx puts it, "the silent compulsion of economic

relations sets the seal on the domination of the capitalist over the worker.

Direct extra-economic force is still of course used, but only in exceptional cases."89

Now Giddens is perfectly well aware of this analysis. Indeed his own account of the differences between class-divided and class societies depends crucially upon it, as he acknowledges.90 Some of the most suggestive parts of A Contemporary Critique, for example, his demonstration of how the mecha- nisms of"surveillance" described by Foucault can only be understood in the light of the mode of surplus-extraction peculiar to capitalism,91 are best taken as a development of Marx's own account. Why, then, does Giddens believe that his version of historical development is an alternative to Marx's?

There seem to be at least three reasons. The first is Giddens's persisting methodological commitment, as I have shown earlier in this article, to a version of interpretive sociology, which gives the subject primacy over structures, and treats power in the sense of domination as intrinsic to social life. I shall have a little to say about this at the end of this section. Secondly, Giddens, as we have seen, misinterprets Marxism, treating it as necessarily

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

151

All use subject to https://about.jstor.org/terms

152

evolutionist, that is, as an interpretation of history as inevitable progress toward a predetermined goal.

It is, however, the third reason to which I wish first to devote some attention. This is Giddens's tendency to identify the relations of production with market-relationships. This is most obvious in the case of The Class Structure of the Advanced Societies, where he argues, for example, that "classes are large-scale groupings" whose existence depends upon "the formation of market-relationships and a division of labour allowing the production of commodities."92 He concludes, not surprisingly, that neither feudalism nor what he chooses to call "state socialism" are class societies. Moreover, in capitalism itself "the system of exploitation operates through differentials inmarket capacity."93 On this basis, Giddens distinguishes between three classes in capitalism - the bourgeoisie, owning the means of production, manual workers, owning labor-power, and an intermediate class today composed largely of white-collar workers whose greater skill than manual workers enhances their "market capacity."

While Giddens's analysis of class-relations is considerably more sophisticat- ed in A Contemporary Critique, as well as (because?) closer to Marx, he continues to identify class with market relations. Thus he argues that in class-divided societies, class is less important than under capitalism because the main form of private property, land, is alienable only to a limited degree, if at all.94 The salience of class and the development of the market are directly related.

With Marx, however, the existence of classes depends upon the monopoliza- tion of the means of production by a minority and their consequent ability to appropriate surplus-labor. It is entirely a secondary matter whether these means of production are commodities that may be purchased and sold on the market. (Even under capitalism land is a very peculiar sort of commodity). The relations of production may well contradict the apparent organization of society, what I have called elsewhere a social formation's form of articula- tion. Thus, in the case of feudalism, the dependence of surplus-extraction on coercion means that the relations of production "appear as a direct relation of lordship and servitude." Social stratification consequently takes the form of a status hierarchy, the division of society into estates, that contradicts the underlying class-relationships. Similarly, under capitalism the relations of production take the form of market relationships, of apparently equal and reciprocal relations between commodity-owners. In the crucial chapter of

Capital volume 1, "The Sale and Purchase of Labour-Power," Marx stresses the disjunction between this apparent equality and the underlying inequality

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

between the capitalist who owns the means of production and the worker who has nothing but his or her labor-power.95

Yet when we turn to Giddens we find that the main lesson he takes from

Marx is that of "the centrality of the labour-contract to the capitalist system."96 It is not, however, the labor contract, but the distribution of the means of production underpinning it that is crucial for Marx. Giddens's discussion of class and capitalism can be said to suffer from two serious weaknesses. First, it tends to overemphasize the importance of commodity production for capitalism. This is reflected, for example, in his introduction of the notion of a commodification of time under capitalism, which raises issues too complex to be discussed here, but which seems to me to be both overstated and misleading.97 (The difficulties Giddens experiences here are in my view bound up with his rather unsteady grasp on Marx's economic theory. Of his discussion of Marx's theory of money, for example, where Giddens says that for Marx "money becomes a commodity only because it represents or symbolizes the exchange-value of all other commodities"98 - the exact reverse of Marx's own view, the less said the better.) Secondly, there is a tendency to confuse juridical property forms, such as the labor-contract, with the relations of production. What Marx is concerned with is not legal entitlement but effective possession.99

Giddens is, in my view, insufficiently aware of the fact that, for Marx, the

relations of production and class are analytical concepts rather than descrip-

tive categories. In other words, they are intended to uncover underlying

relationships precisely because the latter's nature is not directly observable.

Giddens consequently commits what Whitehead called the fallacy of mis-

placed concreteness, supposing that it is only when the division of society

into economic classes is directly visible, as under capitalism, that historical materialism can be said to be valid.

Marx had in a sense already responded to Giddens's "contemporary" cri- tique when he answered in Capital volume 1 a "German-American publica- tion" that had argued that the materialist conception of history "is all very true for our own times, in which material interests are preponderant, but not for the Middle Ages, dominated by Catholicism, nor for Athens and Rome, dominated by politics." Marx replied that "one thing is clear: the Middle Ages could not live on Catholicism, nor could the ancient world on politics. On the contrary, it is the manner in which they gained their livelihood which explains why in one case politics, in the other case Catholicism, played the chief part."'00 The role of the concept of the relations of production is precisely to explain the changes in the forms of domination.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

153

All use subject to https://about.jstor.org/terms

154

One might recast the points I have just made in slightly more formal terms, drawing on recent work in the philosophy of science. Imre Latakos distin- guished between a scientific research program's heuristic, the set of privileged propositions forming its "hard core" and specifying procedures for resolving problems, and the "protective belt" of revisable and empirically falsifiable auxiliary hypotheses surrounding the heuristic. 01 Progress within a research program depends, Lakatos argued, on the empirically successful modifica- tion and elaboration of auxiliary hypotheses along lines suggested by the heuristic. One could say, in the case of Marxism, that its heuristic consists in the basic concepts and propositions of historical materialism, while the auxiliary hypotheses are analyses of specific modes of production, social formations, and conjunctures.102 Much of the point of auxiliary hypotheses is to account for phenomena that apparently contradict the factual claims made in the heuristic. An example of such recalcitrant phenomena in the case of Marxism are the pre-capitalist societies in which authoritative rather than allocative domination seems to prevail. Giddens, because he fails to distin- guish between the different theoretical levels of a scientific research program, believes that the mere existence of these societies counts as a refutation of historical materialism.

It is an implication of this argument that the proof of the pudding is in the eating. Is historical materialism capable of providing empirically satisfactory analyses of pre-capitalist societies? The work of Marxist historians suggests that it is. One example must do. G. E. M. de Ste. Croix's The Class Struggle in the Ancient Greek World is notable, not only for the wealth of scholarship, and ethical and political passion that it displays, but because it is, as Perry Anderson says, "one of the most strenuously theoretical works of history ever to be produced" in Britain.103 Of particular importance is the manner in which de Ste. Croix uses Marx's theory of class in effect as a heuristic, that is, as the basis on which to formulate empirical propositions concerning the social structure of Antiquity. Rather than concentrate on the description of visible status differences, in the manner of M. I. Finley, de Ste. Croix uses the basic concepts of historical materialism both to anatomize classical society, and to provide a dynamic explanation of its development.104 His book is a formidable challenge to Giddens's claim that Marxism offers insight only into the workings of capitalism.

Let me now return to the two reasons for Giddens's rejection of historical

materialism other than his misunderstanding of Marx's theory of class -

namely, his methodological commitment to interpretive sociology, and his identification of Marxism with evolutionism. One might imagine a connec-

tion between the two. I have already mentioned Giddens's vigorous objection

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

to functionalism. Not unnaturally, he tends to associate evolutionism with

the use of functional explanations, since both tend to treat society as a

self-reproducing organism with its own needs.105 Now Giddens might argue

that any Marxism that continues to accord primacy to structures over

subjects must necessarily rely on the tacit or open use of functional explana-

tions that effectively reduce agents to the "supports" of structures. The

thought would be that however much Marxists such as Wright and myself

might like to repudiate evolutionist and functionalist Marxism, that is all the Marxism there is.

Giddens could appeal for support to G. A. Cohen. Rejecting Jon Elster's proposal that historical materialism be reconstructed along methodological individualist lines, Cohen correctly insists that "Marxism isfundamentally concerned not with behaviour, but with the forces and relations [of produc- tion] constraining and directing it," but asserts also that the constraining and directing role of these forces and relations can only be explicated by means of functional explanations: "if functional explanation is unacceptable in social theory then historical materialism cannot be reformed and must be rejected."06

This line of thought is not compelling. It is perfectly possible to follow

Giddens in rejecting functional explanation, while continuing to accord

primacy to structures over agents. The key lies in Giddens's own conceptuali-

zation of structures as both enabling and constraining. Thus we may accept

one of the standard accounts of action made available by contemporary

analytical philosophy, conceiving agents' behavior as proximately caused by

their intentions, beliefs, and desires, while treating the powers instantiated in

this behavior as to a large extent structurally determined. In other words,

what agents can collectively or individually do depends to a significant

degree on their position in the relations of production. The objection to the

philosophies of action formulated within the analytical or hermeneutic

traditions would then be not so much that such theories are false, or part of a

"problematic of the subject," as that they are empty, lacking as they do an

empirical account of the historically variable structural determinants of action.

Obviously these remarks are more hand-waving than a properly formulated argument. Nevertheless, they serve to suggest that the dilemma between functionalism and heremeneutics, to which Giddens offers the theory of structuration as a solution, does not exhaust the available alternatives.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

155

All use subject to https://about.jstor.org/terms

156

Capitalism, Socialism and the State

Let me turn now to consider some rather surprising criticisms of Marxism that Giddens makes towards the end of A Contemporary Critique. These center on the role of the state and violence. He writes, for example, that "Marxism has no tradition of theorizing violence either as an integral and chronic feature of repression, or of the 'world violence' of the contemporary system of power blocs and nation-states."107 This is related to the historical claim that the development of capitalism was bound up with the formation of a European power-system dominated by the great absolutist states, so that "capitalism developed within a military 'cockpit' in which the expansion of industrial production very soon came to be seen by all ruling groups as the sine qua non of national survival."108 "The conjunction between the rise of capitalism and the absolutist state system produced a system of nation-states that, far from being ephemeral, is integral to the world capitalist economy - which is at one and the same time a world military order."'09

I describe these criticisms as "surprising" because many of the statements I have just cited, which Giddens clearly believes to be damaging criticisms of Marxism, are from the standpoint of Marxism platitudinous. The notion, for example, of the state as constituted by violence, by the existence of what Lenin called "special bodies of armed men," is a central theme of such texts as

The Civil War in France, The Origins of the Family, Private Property and the State, and The State and Revolution. The more usual criticism of Marxist writing on the state is that it overestimates the role of violence as "an integral and chronic feature of repression."

Again, to take the case of absolutism, Marx in his writings on France traces the origins of the modern capitalist state to the absolute monarchies of the ancien regime. The role of the quasi-absolutist Tudor state in promoting the development of capitalism through the systematic use of coercion, and of the world power struggle between the European states in contributing to the "primitive accumulation" of capital are well-known themes of part eight of

Capital volume 1.

Then there is the matter of the "world capitalist economy." Giddens writes as

if this were a concept discovered a few years ago by Immanuel Wallerstein,

rather than being a theme of Marx's own economic writings, which assumed

central importance for those who sought to develop the analyses they

contain. It was precisely on the capitalist world-system that the two genera- tions of Marxists after Marx focused their attention. The result was numer-

ous texts by Kautsky, Hilferding's Finance Capital, Luxemburg's The Ac-

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

cumulation of Capital, Lenin's Imperialism, and Bukharin's Imperialism and World Economy and The Economics of the Transformation Period. Lenin and Bukharin in particular stress that with the emergence of imperial- ism competition between capitals increasingly took the form of military competition between nation-states.

I do not wish to claim that this body of writing on the state and the world

system is wholly satisfactory or adequate. But it exists. It is true that the

whole body of work to which I have referred was vulgarized and integrated

into a highly evolutionist form of Marxism with the triumph of Stalinism.

Moreover, the various schools of Western Marxism were largely, though by

no means wholly, uninterested in the type of economic and political analysis characteristic of classical Marxism.

But even at the height of Stalinism's grip on Marxism evolutionism did not go unchallenged. For example, Giddens lays great stress on the concept of "time-space edges," by which he means "the simultaneous inter-connected

existence of different types of society. This helps to free us from the tendency of evolutionary thought to analyse societal development in terms of 'stages', and from the influence of 'unfolding models' of change.""0 Compare this with, for example, the opening chapters of Trotsky's History of the Russian Revolution, where central importance is given to "the law of combined development - by which I mean a drawing together of the different stages of the journey, a combining of separate steps, an amalgam of archaic with more contemporary forms."'1 l Trotsky uses the concepts of uneven and combined development to explain why socialist revolution occurred first in Russia, rather than, as evolutionist Marxism had predicted, in the more advanced capitalist countries. (Incidentally, Giddens himself had earlier offered a highly evolutionist explanation of this revolution, arguing that "the [manual] working class is most likely to achieve a high degree of revolutionary class consciousness in the initial phase of the industrialization process."112)

The revival of Marxist theory in the past decade and a half has seen a significant return to the classical focus on the world-system and the state. Giddens cites, but does not discuss Poulantzas's last book, State, Power, Socialism, which is concerned precisely with the essential role of the organ- ized violence of the state in both the constitution and the reproduction of the relations of production.113 Giddens describes Perry Anderson's Lineages of the Absolutist State as "a fundamental source,""4 but does not address the central thesis of that book, which is that the absolutist state must be under-

stood as a form of feudal class rule. Although it is true to say that much of the "state-derivation" debate among German Marxists has been sterile and

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

157

All use subject to https://about.jstor.org/terms

158

formalistic in the extreme, nevertheless a number of contributions to that and connected debates, notably those by Claudia von Braumuhl, Colin Barker, and Cesare Luporini, have focused on a theme that Giddens approp- riates as his own, namely the claim that capitalism involves a system of competing nation-states whose origins are to be found in the absolutist era.15 These texts all deal with the centrality of war to that system, as do Anderson in his discussion of absolutism, and the theorists of the permanent arms economy (T. N. Vance, Michael Kidron, and Chris Harman) in their anal- yses of contemporary capitalism.1"6

The interesting question to arise from all this is why so erudite, and, in many ways, so friendly a critic of Marxism as Giddens, should come to ignore so large and important a body of writing. There are, I think, at least two reasons. The first has to do with sociology's general failure to confront Marxism as a political tradition. With the very important exception of Marx himself, classical Marxism is not discussed by Giddens. Lenin, Hilferding, and Gramsci each appear once in the index of A Contemporary Critique, Engels, Luxemburg, Trotsky, and Bukharin not at all. Althusser and Pou- lantzas, on the other hand, receive generous attention.

Why does this matter? Because the theoretical writings of the classical Marxists have an irreducibly political orientation. Their intellectual activity bore an intrinsic connection to their participation as revolutionary socialists in working-class movements. This active involvement in politics explains why what Perry Anderson calls the "areas most central to the classical traditions of historical materialism" were "scrutiny of the economic laws of motion of capitalism as a mode of production, analysis of the political machinery of the bourgeois state, strategy of the class struggle necessary to overthrow it." 17 The political orientation of classical Marxism has, however, made it a difficult tradition for academic social science to engage with. It is, for example, only in the past few years, and largely as a result of the radicalization produced in the late 1960s and early 1970s, that serious studies of the political thought of Marx and Lenin have appeared.18 On the whole, sociologists have found it easier to address those Marxists, such as the Frankfurt school and Althusser and his followers, whose theoretical pre- occupations are largely detached from political activity, and are therefore much more readily assimilable to a purely academic discourse. Although there is a political undercurrent to Giddens's work (A Contemporary Critique is the first in a two-volume attempt to rethink socialist theory), 19 he

shares this common failure to grasp the distinctive nature of classical Marxism.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

The second reason for Giddens's highly selective reading of Marxism has to

do with his latent Weberianism. When he says that "Marxism has no

tradition of theorizing violence" he may mean, not so much that Marxism

ignores the phenomenon of violence, as that it has no satisfactory explana-

tion of it. Marxism, Giddens might say, recognizes the existence of violence;

it even seeks to explain it - for example, in the way that Engels and Lenin do,

for example, in terms of class exploitation and the specific political structures

to which it gives rise. But, Giddens might argue, these explanations will not

do. Why will they not do? In part, because of the weaknesses specific to these

explanations, and indeed Giddens subjects recent Marxist writing on the state to critical examination.120

But there is also, I think, an undercurrent to Giddens's arguments that

suggests that any attempt to explain violence in terms of other factors is

doomed to failure. Formulations such as "an overwhelmingly obvious and chronic trait of human affairs - recourse to violence and war"12' seem to

imply that this "trait" is an irreducible feature of social life. Any theory of violence that, like Marxism, treats it, at least in its more organized and destructive forms, as specific to certain forms of society (namely, those in which classes exist) is, from this perspective, necessarily mistaken. Such a line of argument, if present in Giddens's writings, is surely connected to his tendency to assimilate the two senses of power - transformative capacity and domination. For a theory of action implying that relations of domination are an essential characteristic of social formations can easily be made to issue in an account of violence that treats it as an equally "chronic" and irreducible "trait of human affairs." The kinship between such a view of society and what Giddens himself acknowledges to be Weber's Nietzschean stress on power, conflict, and violence'22 is obvious.

The Weberian drift of Giddens's argument is evident when we consider one of his other criticisms of Marxism. He argues that Marx, by focusing on the extraction of surplus-labor, offered too narrowly economic an account of exploitation, which should properly be understood simply as "domination which is harnessed to sectional interests."23 Giddens is then able to identify forms of exploitation that Marxism has failed to account for:

(a) exploitative relations between states, where these are strongly influenced by military domination; (b) exploitative relations between ethnic groups, which may or may not converge with the first; and (c) exploitative relations between the sexes, sexual inequality... These are major "absences" in Marxist theory.124

What does it mean to talk here of"'absences' in Marxist theory"? If Giddens means that Marxism has consistently ignored, or denied the existence of, say,

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

159

All use subject to https://about.jstor.org/terms

160

racial and sexual inequalities, then he is surely wrong. We can point to, for example, Engels's Origins of the Family, the debates in the Second Interna- tional on the national question, and recent Marxist work on racial oppres- sion in South Africa. If Giddens is saying that Marxism has failed to come up with a satisfactory theory of phenomena such as nationality, then I'm inclined to agree with him (though it's not clear that orthodox social science has done any better). If, however, the claim is that Marxism cannot provide such a theory, then more argument is required in its support than Giddens gives.

Let me elaborate the last point by returning to the distinction drawn above between the heuristic and the auxiliary hypotheses of a scientific research program. The heuristic of historical materialism, consisting as it does of concepts such as forces and relations of production, exploitation, class, etc., does not make any reference to sexual, ethnic, and inter-state relations. This is, however, in no sense a decisive objection to Marxism. For the work of the heuristic is done precisely in generating auxiliary hypotheses that must, inter alia, provide empirical explanations of phenomena such as racial inequality. The relative failure of classical Marxism to come up with adequate theories of racial and sexual oppression in comparison with the considerable body of economic and political analyses to which I have already referred arguably has less to do with any immanent "class-reductionist" logic than with politi- cal priorities that encouraged the development of historical materialism in other directions.

This is in no sense an invitation to Marxists complacently to ignore the many inadequacies of their theories. But it is to say that the mere existence of apparently aberrant phenomena does not refute historical materialism. What makes Marxism, despite its many "crises," still the most powerful and challenging research program in the social sciences is precisely the strength and implausibility of its basic claim, namely, that the variety and complexity of social life is to be explained ultimately in terms of the prevailing forces and relations of production. Intuitively this is no more credible a claim than was Galileo and Newton's aspiration to mathematize nature in the seventeenth century. Modern physics' success in accounting for phenomena that seem to refute it sets a model for Marxism's own ambition. Giddens, however, seems to believe that the very failure to build into the heuristic of historical materialism forms of "exploitation" other than the extraction of surplus- labor counts against it. In doing so, he implicitly counterposes to Marxism what Erik Olin Wright calls "a contingent causal pluralism,"125 in which inter-state conflict, and racial and sexual inequalities are of equal explanato- ry importance to class exploitation. His objection is thus indicative more of

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

his own presumptions about what a true social theory would be like than of any inherent defect in Marxism.

Giddens's final objection to Marxism is also more revealing of his own views

than it is of those he criticizes. The evolution of "state socialism," he claims,

shows that "Marx's thought supplies precious few clues indeed about the

continuing significance of power in socialist society."126 It all depends, of

course, on what you mean by socialism. Giddens's main discussion of state

socialism, not, to my knowledge, subsequently repudiated, is in The Class

Structure of the Advanced Societies, where it is defined as "any economic

order in which the means ofproduction isformally socialized in the hands of

the state."'27 It follows from the premises of Giddens's theory of class that

state socialism thus conceived cannot be a class society. For such a society

requires the insulation of economy and polity that will permit market forces

unrestrainedly to allocate people to their position in society. However, in

state socialism, "the forms of differentiation in market capacity which oper-

ates in capitalist societies are considerably modified."'28 Consequently, "the

institutional mediation of power created by the integration of economy and

polity ... does provide an escape route from class society."129 However, the

state socialist societies clearly involve a high degree of social and economic

inequality and political oppression. Since Marx explained phenomena of

this sort in terms of the division of classes on the basis of exploitive relations

of production, the existence of state socialism must count as a refutation of historical materialism.

An adequate reply to this argument would take too long, and would involve tramping over very well worn ground.130 Let us simply note that this conclu- sion follows from Giddens's premises, not Marx's. In the first place, as we have seen, Marx does not ground class in market-relations, but rather on the distribution of the means of production. These depend, not on juridical property-forms, but on relations of effective possession. Secondly, it is clear from The Civil War in France and his other political writings, that for Marx the decisive step in the transition to a classless society came not with the introduction of state ownership of the means of production, but with the replacement of the repressive state apparatus by new forms of workers' power based on a popular militia and the election and recall of public officials. This theory may or may not be mistaken, but it offers quite a different account of socialism to that provided by Giddens in refutation of Marx. It has been pointed out by Marxists from Engels and Williams Morris onward that state ownership of the means of production is quite compatible with the persistence of class exploitation, and that private capitalism could under certain conditions be replaced by state capitalism. Giddens is unable to

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

161

All use subject to https://about.jstor.org/terms

162

envisage such a possibility, partly I think because of his tendency to identify capitalism with the prevalence of commodity production. Once again, a misreading of Marxism prevents him from engaging with the real issues.

The Rhetoric of Synthesis

In conclusion, I wish to offer some remarks about the nature of Giddens's

project. Since the 1960s a number of younger sociologists have sought to

integrate some elements of Marxism into their thinking, prompted to do so

by a variety of reasons, notably the inability of what Giddens calls the

"orthodox consensus" to handle the phenomenon of social conflict, and the

burgeoning of such conflict in the world outside sociology departments.131

Often Such pillaging of Marxism is accompanied by its disdainful dismissal.

Giddens's own work is a version of this approach, albeit methodologically

far more sophisticated, and always respectful of, and increasingly sympathet- ic to Marxism.

Let us note what such an attempt to arrive at a synthesis of mainstream sociology and Marxism implies. It supposes that the interpretive sociology of Weber, to name the most intellectually powerful and suggestive of the mainstream traditions, itself a brilliant synthesis of hermeneutics, histori- cism, Nietzsche's genealogy of power, and marginalist economics, can be rendered compatible with elements at least of historical materialism. It seems to me that there are general grounds for doubting this to be so: the two traditions are so fundamentally opposed in method, substance, and political implications that the outcome of any attempt to combine them can only be incoherence, eclecticism, or the denaturing of one or both. As evidence for this claim I would cite the way in which Giddens continues to give primacy to certain fundamental Weberian premises - the priority of subject over struc- ture, and the omnipresence of domination.

The result is a highly unstable system, since Giddens is clearly genuinely sympathetic to Marxism, owes it many insights, and wishes to find a place for it in his synthesis. This central difficulty - Giddens's underlying commitment to interpretive sociology combined with his openness towards Marxism - helps to explain his bitter hostility to Parkin's Marxism and Class Theory.

For Parkin is concerned above all to demonstrate the incompatibility of Marxism and Weberian sociology, and to advocate the superiority of the latter. Giddens rejects this claim, yet shares the same fundamental premises as Parkin - above all, they both believe relations of domination and subordi- nation and social conflict to be ineradicable features of human existence.132

No wonder that Giddens attacks Parkin so vehemently, for the latter is

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

merely pointing out the simple but unwelcome truth that he cannot have his cake and eat it.

The attempt to strike an equipoise between Marxism and mainstream soci-

ology seems to me not only impossible, but to lead often to an intellectually rather dubious approach to empirical questions. This is most evident, not in Gidden's work itself, but in a book by John Scott, Corporation, Classes and Capitalism, which is clearly considerably influenced by him. The book is an examination of the thesis that twentieth-century capitalism is characterized

by the separation of ownership and control. Scott contrasts two opposing theories, the "theory of industrial society" (Berle and Means, Bell, Aron),

which asserts this thesis, and the "theory of capitalist society" (i.e. Marxism),

which denies it. After showing conclusively that the thesis is false, Scott

claims that both theories are mistaken! He is able to do so because he

identifies the "theory of capitalist society" with clearly inadequate Soviet accounts of state monopoly capitalism. Nevertheless the evidence he cites

demonstrates the basic Marxist proposition that an economically dominant

class continues to exist in late capitalism, and draws considerably on th

research of non-Soviet Marxists. A position is thus attained au dessus la

melee, which puts both Marxism and orthodox sociology in its place, but

only at the price of grossly misrepresenting the former. Giddens's "contem-

porary critique" seeks a similar vantage-point by means of such a misrepres- entation of Marxism.

We may say, in conclusion, that despite Giddens's often brilliant intellectual fireworks, Marxism remains to be transcended. Both its theoretical and political revival, and the economic crises and social conflicts of the past fifteen years demonstrate its continued relevance. It is, I suppose, some sort of tribute that Marxism is considered worthy of inclusion in a grander synthesis by someone as talented as Giddens. Unfortunately, it may not prove very easy to digest.

NOTES

I have used abbreviations when citing the following books by Anthony Giddens - T Structure of the Advanced Societies (2nd ed., London: Hutchinson, 1981), CSAS; New Sociological Method(London: Hutchinson, 1976), NRSM; Central Problems in Social T (London: Macmillan, 1979), CPST; A Contemporary Critique of Historical Mater (London: Macmillan, 1981), CCHM; Profiles and Critiques in Social Theory (Lond millan, 1982), PCST.

1. E. O. Wright, "Giddens's Critique of Marxism," New Left Review 138 (1983) 12 2. See, for example, M. Foucault, Power/Knowledge (Brighton: Harvester 198 Deleuze and F. Guattari, Mille Plateaux (Paris: Minuit, 1980), and F. Parkin, Mar

and Class Theory (London: Tavistock, 1979).

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

163

All use subject to https://about.jstor.org/terms

164

3. CSAS, 296.

4. A. Giddens, "Classes, Capitalism, and the State," Theory and Society, 9: 6 (1980),

reprinted in PCST The same issue contains Parkin's equally bitter "Reply to Giddens." 5. CCHM, 3. See also A. Giddens, "From Marx to Nietzsche?," in PCST, especially 229. 6. CCHM, 30.

7. See, for example, CPST, 49ff. 8. Ibid., 49.

9. Ibid., 112.

10. NRSM, 15-16.

11. Ibid., 160, 102.

12. CPST, 3.

13. Ibid., 106.

14. Ibid', 141.

15. Ibid., 131.

16. Ibid., 211.

17. Ibid., 5.

18. A. Giddens, "Hermeneutics and Social Theory," in PCST, 8.

19. F. R. Dallmayr, "The Theory of Structuration: a Critique," in PCST, 19.

20. CPST, 67.

21. For these and other criticisms of Winch, see NRSM, 44-51.

22. P. Anderson, In the Tracks of Historical Materialism (London: Verso, 1983) 40-2. See

Ibid., 60-67 for a discussion of the similarities, and differences, between Habermas and

post-structuralism. 23. Ibid., 43-45.

24. See, S. Cavell, The Claim of Reason (Oxford: Oxford University Press, 1979), S. M. Holtzmann and C. M. Leich, eds., Wittgenstein: to Follow a Rule, (London: Boston and Henley, Routledge and Kegan Paul, 1981), and S. Kripke, Wittgenstein on Rules and Private Language (Oxford: Basil Blackwell, 1982).

25. J. McDowell, "Non-Cognitivism and Rule-Following," in Holtzmann and Leich, 144. 26. P. F. Strawson, "Sense and Consensus," Times Literary Supplement 11 February 1983. 27. L. Wittgenstein, Philosophical Investigations (Oxford: Basil Blackwell, 1968) 201.

28. A. Giddens, "Action, Power and Structure," in PCST, 31.

29. Dallymayr, "Theory," 22.

30. Ibid., 27.

31. CCHM, 171. See generally ibid., 157-81, and Giddens, "From Marx to Nietzsche?" 32.

CCHM, 171.

33. See N. Poulantzas, State, Power, Socialism (London: NLB, 1978) 146-53, and P. Dews,

"Nouvelle Philosophie and Foucault," Economy and Society 8: 2 (1979).

34. P. Anderson, Arguments within English Marxism (London: NLB, 1980) ch. 2.

35. Giddens, "From Marx to Nietzsche?," 220.

36. A. Giddens, "Power, the Dialectic of Control and Class Structuration," in PCST.

37. Ibid., 197-98.

38. E. O. Wright, Class, Crisis and the State (London: NLB, 1978) 98ff.

39. Giddens himself makes this point against Foucault: "From Marx to Nietzsche?," 222-23. 40. See CPST, ch. 6, CCHM, chs. 1, 3, 4.

41. CCHM, 38.

42. CPST, 103.

43. CCHM, 90.

44. CPST, 201-6.

45. J. Elster, "Marxism, Functionalism and Game Theory: The Case for Methodological

Individualism," Theory and Society II: 4 (1982).

46. J. Elster, Logic and Society (London: John Wiley & Sons, 1978).

47. CPST, 139-41.

48. Ibid., 59-69.

49. Ibid., 92.

50. For example, CCHM, 53.

51. J. Habermas, "A Reply to my Critics," in J. B. Thompson and D. Held, eds., Habermas:

Critical Debates (London: Macmillan, 1982) 286. 52. CCHM, 3.

53. NRSM, 110. 54. Ibid., 111. 55. Ibid.

56. CPST, 43. 57. Ibid.

58. CCHM, 51-52. 59. Ibid., 28-29. 60. Ibid., 107.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms

61. Ibid., 108.

62. CPST, 162.

63. See A. Giddens, "'Power' in the Recent Writings ofTalcott Parsons," in Giddens, Studies

in Social and Political Theory (London: Hutchinson, 1977), and "Labour and Interac- tion," in PCST.

64. CCHM, 28-29.

65. Ibid., 53.

66. See, for example, R. Harre and E. Madden, Causal Powers (Oxford: Basil Blackwell,

1975).

67. M. Foucault, "The Subject and Power," afterword to H. L. Dreyfus and P. Rabinow,

Michel Foucault (Brighton: Harvester, 1982) 217-18. 68. CCHM, 20.

69. Ibid., 80.

70. Ibid., 161-66, CPST, 76ff.

71. CCHM, 145.

72. CSAS, 134.

73. Ibid., 298 (1979 Postscript).

74. CCHM, 165.

75. Ibid., 147.

76. See A. Callinicos, Is There a Futurefor Marxism?(London: Macmillan, 1982) 142-47, and

Marxism and Philosophy (Oxford: Clarendon Press 1983) 110-13.

77. See Callinicos, Marxism and Philosophy, ch. 2.

78. Wright, "Giddens's Critique of Marxism," 25-34.

79. Ibid., 26.

80. CCHM, 157.

81. Wright, "Giddens's Critique of Marxism," 31.

82. CCHM, 95.

83. D. Schmandt-Besserat, "The Earliest Precursor of Writing," in Prehistoric Times: Read-

ings from "Scientific American," ed. B. M. Fagan (San Francisco: W. H. Freeman and

Company, 1983).

84. See, Callinicos, Future, chs. 6 and 7.

85. K. Marx, Capital iii. (Moscow: Progress, 1971) 782-813.

86. Ibid., 790-91.

87. See B. Hindess and P. Hirst, Precapitalist Modes of Production (London: Routledge and

Kegan Paul, 1975).

88. G. Bois, Crise du feodalisme (Paris: Presses de la Fondation Nationale des Sciences

Politiques, 1976).

89. K. Marx, Capital i. (Harmondsworth: Penguin, 1976) 899.

90. CCHM, 122. See generally Ibid. 107-28.

91. Ibid., 157-81.

92. CSAS, 84.

93. Ibid., 131.

94. CCHM, 113-17.

95. Marx, Capital i. 270-80.

96. CSAS, 298-99 (1979 Postcript).

97. CCHM, 130ff.

98. Ibid., 116.

99. See K. Marx and F. Engels, Collected Works ((50 vols.) London: Lawrence and Wishart,

1975-) vi. 197 and xxxviii. 99, Hindes and Hirst passim, and G. A. Cohen, Karl Marx's

Theory of History - a Defence (Oxford: Clarendon Press, 1978) 217-25.

100. Marx, Capital i. 176 n.

101. I. Lakatos, Philosophical Papers (2 vols., Cambridge: Cambridge University Press, 1978), especially i. ch. 1.

102. For a similar use of the distinction between heuristic and auxiliary hypotheses, see A. Gilbert, Marx's Politics (Oxford: Martin Robinson, 1981) 5-13, and "The Storming of Heaven: Politics and Marx's Capital," in J. R. Pennock and J. W. Chapman, Marxism: Nomos XXVI (New York: New York University Press, 1983) 151-61.

103. P. Anderson, "Class Struggle in the Ancient World," History Workshop 16 (1983), 58. 104. G. E. M. de Ste. Croix, The Class Struggle in the Ancient Greek World (London:

Duckworth, 1981) ch. II.

105. See especially the critiques of functional explanation in J. Elster, Ulysses and the Sirens

(Cambridge: Cambridge University Press, 1979), and "Marxism, Functionalism and

Games Theory."

106. G. A. Cohen, "Reply to Elster on 'Marxism, Functionalism, and Game Theory,"' Theory

and Society, 11: 4 (1982) 489, 483. 107. CCHM, 250.

108. Ibid., 190.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

165

All use subject to https://about.jstor.org/terms

166

109. Ibid., 198.

110. Ibid., 83.

111. L. Trotsky, The History of the Russian Revolution (3 vols., London: Sphere, 1967), i. 23.

See G. Novack, Understanding History (New York: Monad. 1972), B. Knei-Paz, The

Social and Political Thought of Leon Trotsky (Oxford: Clarendon Press, 1978), 62-107,

and M. L6wy, The Politics of Combined and Uneven Development(London: NLB, 1981) chs. 1-3.

112. CSAS, 153.

113. CCHM,279n. 16.

114. Ibid., 275 n. 5.

115. See C. von Braumuhl, "The Analysis of the Nation-State within the World Market

Context," in State and Capital, eds. J. Holloway and S. Picciotto, (London: Arnold, 1978), C. Barker, "The State as Capital," International Socialism 2: 1 (1978), and C. Luporini, "'Politique' et 'Etatique"' in Marx et la critique de la politique, E Balibar et al.

(Paris: Maspero, 1979).

116. See T. Vance, The Permanent Arms Economy (Berkeley, 1970), M. Kidron, Western

Capitalism since the War (Harmondsworth: Penguin, 1970), and C. Harman, Explaining

the Crisis (London: Bookmarks, 1984).

117. P. Anderson, Considerations on Western Marxism (London: NLB, 1976) 44-45.

118. See, for example, H. Draper, Karl Marx's Theory of Revolution 2 vols. (New York:

Monthly Review Press, 1977 and 1978), Gilbert, Marx's Politics, T. Cliff, Lenin 4 vols. (London: 1975-79), N. Harding, Lenin's Political Thought 2 vols. (London, Macmillan:

1977 and 1981). 119. CCHM, 3, 24-25.

120. Ibid., ch. 9.

121. Ibid., 177.

122. Giddens, "From Marx to Nietzsche?," 218.

123. CCHM, 161.

124. Ibid., 242.

125. Wright, "Giddens's Critique of Marxism," 32.

126. CCHM, 244.

127. CSAS, 155.

128. Ibid., 230.

129. Ibid., 282.

130. See especially T. Cliff, State Capitalism in Russia (London: Pluto, 1975). 131. Giddens, "Hermeneutics and Social Theory," 1-3.

132. Compare CSAS, 283 and Parkin, Marxism 187-92.

Theory and Society 14 (1985) 133-166

03404-2421/85/$03.30 ? 1985 Elsevier Science Publishers B.V.

This content downloaded from 107.77.232.140 on Thu, 15 Apr 2021 18:28:59 UTC

All use subject to https://about.jstor.org/terms