The Anatomy of Capitalist Societies

Contemporary So<:lal Theory

General Editor: ANTHONY GIDDENS

This series aims to create a forum for debate between different theoretical and philosophical
traditions in the social sciences. The books included in the series wiu cover broad schools of
thought as well as the work of panicular thinkers whose ideas have had a major impact in
social science, under the general title of Theoretical Traditions in the Social Sciences. The
series is not limited to abstract theoretical discussion, but will also incl ude more substantive
works on contemporary capitalism, the state and politics.

PUBLISHED TITLES
Anthony Giddens: Central Problems in Social Theory

FORTHCOMING TITLES
Martin Albrow: Weber and the Construction of Social Theory
Zygmunt Bauman: Memories of Class
Zygmunt Bauman: The Pyrric Victory of the Puritan
Tony Bilton, Kevin Bonnen, Louella Hodgson, Philip Jones, Ken Sheard, Michelle
Stanwonh, Andrew Webster: Introductory Sociology
David Brown: Industrial Sociology
Simon Oarke: Marx, Marginalism and Modern Sociology
Emile Durkheim: Rules of Sociological Method
Boris Frankel: The Modern State
David Held and John Thompson (Editors): Critical Essays on Habermas
David Held: Bureaucracy, Democracy and Socialism
Oaus Offe: Structural Problems of the Capitalist State
Ali Ranansi: Marx and the Division of Labour
Gerry Rose: Deciphering Sociological Research
John Scott: The Development of the British Upper Oass
Steven Taylor: Durkheim and the Study of Suicide
John Urry: The Anatomy of Capitalist Societies

Theoredcal Tradltlom; kI the Social Sciences

FORTHCOMING TITLES
Barry Barnes: Kuhn and Social Science
David Bloor: Wittgenstein and Social Science
John Forrester: Jacques Lacan
John Heritage: Garfinkel and Ethnomethodology
Althar Hussain: Foucault
Bob Jessop: Nicos Poulantzas
Robin Williams: Erving Goffman
The Anatomy of
Capitalist Societies
The Economy, Civil Society
and the State

John Urry
Department of Sociology
University of Lancaster

M
© John Urry 1981
Softcover reprint of the hardcover I st edition 1981

All rights reserved. No part of this publication may

be reproduced or transmitted, in any form or by any
means, without permission.

First published 1981 by

THE MACMILLAN PRESS LTD
London and Basingstoke
Associated companies in Delhi Dublin
Hong Kong Johannesburg Lagos Melbourne
New York Singapore and Tokyo

ISBN 978-0-333-29431-4 ISBN 978-1-349-16506-3 (eBook)

DOI 10.1007/978-1-349-16506-3

This book is sold subject to the standard conditions of the

Net Book Agreement.

The paperback edition of this book is sold subject to the

condition that it shall not, by way of trade or otherwise, be
lent, resold, hired out, or otherwise circulated without the
publisher's prior consent, in any form of binding or cover
other than that in which it is published and without a similar
condition including this condition being imposed on the
subsequent purchaser.
For Thomas and Amy
Contents

Preface ix

1 Introduction 1

2 Civil Society 10

3 The Spheres of Production and Circulation 26

4 The Critique of Ideology 44

5 The Practices of Civil Society 63

6 The State: A Critique 80

7 The State: Some Proposals 99

8 The State, Labour-Power and Class Struggle 117

9 Capitalism and Representative Democracy 140

Notes and References 155

Bibliography 167
Index of Names 175
Index of Subjects 177
Preface

I am very grateful to the following friends who read the typescript in

whole or in part, and made extremely helpful comments and suggestions:
Nick Abercrombie, Scott Lash, Dan Shapiro, Sylvia Walby and Alan
Warde. I would also like to thank various students of mine in Lancaster
with whom I have discussed various aspects of this book. I am also
grateful to the Sociology Department at the University of Lancaster for
providing a generally supportive intellectual environment.

University of Lancaster JOHN URRY

1980
1
Introduction

In this book I shall suggest that there is a way of resolving , fairly simply
and straightforwardly, the problematic relationship between the
economic base and the superstructures of capitalist society. I shall begin
this Introduction by briefly outlining the 'reductionist' and 'autonomist'
positions, of whether the superstructures are to be 'reduced' to the
economic base, or whether they are 'autonomous' of that base. I shall
suggest that it is possible to resolve at least some of the differences
between these positions by considering in more detail the social relations
characteristic of the economy, of the state and of civil society. These
distinctions will provide the basis of my claim that the concept of civil
society enables us to theorise more satisfactorily those relationships
which exist between the entities customarily designated by the terms
'base' and 'superstructure'.
The nature of these relations between base and superstructure are
important to establish in part because of the increasing interrelationship
between the two within contemporary advanced capitalist societies. In
trying to understand these relationships I shall mainly discuss recent
contributions within Marxist debate. This is partly because this is where
many of the more interesting contributions on these issues have recently
been produced. Also, though, I shall consider these debates because
some of the arguments advanced do in fact seem to be seriously mistaken
and these greatly restrict the usefulness of the analysis. In particular, I
shall suggest that in much Marxist writing there is a failure to incorporate
adequately into the theorising, the nature, role and effectiveness of class
and other popular struggles. I shall therefore consider in some detail how
these should be studied, and in particular their significance for analysing
the state. However, it is al so the case that these issue are important within
'non-Marxist' debates, which have increasingly concerned themselves
2 The Anatomy of Capitalist Societies

with the character of the contemporary state and of the interrelations with
business and unions.! But in both cases it is important to avoid the pitfallE
of both 'reductionism' and 'autonomism'. In the following I shall suggest
that the differences between these two positions can be in part overcome
by employing the concept of civil society. 2 Let me begin then by stating
briefly and, I admit, dogmatically the nature of the reductionist and
autonomist positions which I wish to go beyond.
Reductionists 3 generally argue that the laws of motion of capital are in
themselves sufficient to account for the form ofthe state, changes in state
policy and the nature of ideology. Politics and ideology are therefore
reduced, albeit sometimes in a complex and mediated fashion, to the
nature of the economy, to the pattern of capital accumulation. The laws of
motion which govern the economy are thought to demand political and
ideological forms which thus enjoy little or no autonomy. The only kind
of non-correspondence between base and superstructure is that resulting
from 'superstructural lag', from the fact that politics and ideology may
not change directly and immediately with the changing requirements of
capital accumulation. However, it is presumed that, in the last instance,
they will change and a resulting correspondence be established between
base and superstructure. This orthodox or fundamentalist Marxist
conception is generally associated with a non-reformist view of politics.
Socialism can only be achieved through totally revolutionising capitalist
society, smashing the bourgeois state and ideology, constructing a
socialist mode of production and establishing a dictatorship of the
proletariat. If these do not occur, then capitalism remains dominant.
Capitalism cannot be transformed by reforms. 4
Autonomists, by contrast, argue that political and ideological super-
structures are autonomous of the economic base. They claim that the
attempt by Althusser and his follower's such as Poulantzas, to theorise
the unity of a given soical formation in terms ofthe relative autonomy and
partial effectivity of the political and ideological instances is in fact
incoherent. 5 It is argued by autonomists that politics and ideology, either
have to be seen as determined by the economy, as reductionists argue; or
they are not directly determined, in which case they are autonomous. So it
is said that there is a simple choice: reductionism or autonomism.
However, since Althusser and others have shown that the state and
ideology are not directly reducible to the economy, the correct position
must be autonomist. There is a necessary non-correspondence of the
levels. each one being analysable in its own terms, of separate realms or
discourses. 6 It follows from this autonomy of the levels that there is
 Introduction 3

no principle of unity of the social formation, and therefore substan-

tial transformations can be effected in one level without this
having implications for the other levels. For instance, it might be pos-
sible to change the balance between commodity and non-commodity
relations ,7 without a revolutionary transformation of the capitalist state
and the establishment of the dictatorship of the proletariat. The
dichotomy between reformism and revolution ism collapses on this
view. 8
I shall discuss and criticise specific aspects of each of these positions at
various points in this book. In doing so I shall argue for a version of the
'relative autonomy' thesis. For the moment I want to argue against the
idea that societies can be divided into the economic, political and
ideological levels. This is a distinctive feature of Althusserianism, and in
certain cases of an autonomist position. I shall try to show in Chapters 4
and 5 that ideology is not to be viewed in this way, as one level or instance
or structure in some sense analogous to the 'economic' and the 'political'.
Let me now briefly show why societies are not to be viewed in this way;
this will provide a preliminary justification for my claim that a division
into the economy, civil society and the state provides a better basis for
understanding advanced capitalist societies.
Althusser first developed the thesis that there are three instances, each
possessing a homologous structure. In each it was held that there is a
process of production in which given raw materials are transformed by
human labour into a determinate product through using definite means of
production. 9 On this view ideology is seen as embodied within a specific
set of material practices. Now this view is preferable to an alternative
account in which the economic is counterposed to the ideological, in the
sense that the former is seen as real or material, while the latter is seen as
unreal or ideational. But although this second view is incorrect because it
implies that ideology consists simply of 'free-floating ideas', it does not
follow by contrast that ideology is embedded within aspecijic set of social
practices. This is because ideology is, in a sense, everywhere and is not to
be confined to a particular level or instance, it is not just confined to the
family, schools, trade unions, etc. For example, both the social relations
of capitalist production, and the state, have massive ideological effects.
In the case of representative democracy this occurs not just through the
so-called 'ideological state apparatuses' ,10 but through the ways in which
citizens believe that they exercise self-determination; this does not serve
to legitimate a ruling class but denies its existence. Hence, ideology is not
to be seen as an instance or structure. There are many, diverse social
4 The Anatomy of Capitalist Societies

practices within contemporary capitalist societies and some of these exert

important ideological effects.
Let me note three implications of this argument. First, it is incor-
rect to maintain that the different practices or levels in a capitalist
society exhibit a similar structure. Indeed there is no reason whatsoever to
expect such similarity. Second, one difference between ideology and the
state, or the economy, is that the former is not unified while the latter pair
are - we will see later just what are the bases of their unity. Third, it does
not follow from the rejection of the concept of an ideological level that
capitalist societies can be analysed merely in terms of the 'political' and
the 'economic', or 'capital' and the 'state'. To do so leads either to an
over-extension of the role of the state; this can be seen in both Althusser's
conception of repressive and ideological state apparatuses and in
Poulantzas's concept of the capitalist stateY Or it leads to an over-
extension of the economic, or of the capital-relation, as is common in
much material produced within the Conference of Socialist
Economists. 12 In both views there is inadequate treatment of an essential
feature of capitalist society - namely, the existence of a multiplicity of
social classes; the variety of social groupings based on gender,
generational, racial, residential and national differences; and the impor-
tance of 'private' social relations within the family and within voluntary
associations of very many kinds. These social relations of 'civil society'
are not to be equated either with the state (as in Althusser's ISA concept)
or with the economy (as in Hegel's concept of civil society). They are not
to be equated because they involve different forms of constitutive social
relations. These are not homologous in structure. It is part of the pur-
pose of this book to illumine the different kinds of social relations which
are implied within the economy, within civil society, and within the
state.
One paradox of contemporary Marxist debates is the reproduction of
certain of the problems which have already been encountered within
orthodox sociology. There are two which are of particular importance and
indeed are inextricably connected. The first is the tendency for a
considerable amount of Marxist writing to be functionalist, either
explicitly or implicitly. This can be seen in those approaches to the study
of modes of production which presume that such modes both necessitate
certain conditions of existence and actually bring them into being. This
functionalism is also present in approaches which are apparently more
focused on class conflict and struggle, where the state, for example,
functions as the general factor of social cohesion.13 The second error
 Introduction 5

consists of the reaction to this all-embracing functionalist and struc-

turalist determinism, which is to argue that it is people, or at least social
classes, that make history, and that it is an empirical question as to
whether modes secure their conditions of existence and this depends on
the precise effectivity of social struggle. Thompson has recently
highlighted this position: The' 'mode of production" has become like a
base camp in the Arctic of Theory, which the explorers may not depart
from for more than a hundred years for fear of being lost in an ideological
blizzard. '14 But the problem in this critical position lies in failing to detail
how, when and why social struggles may make history, what is the
precise effectivity of such struggle. Thompson's position is not elabo-
rated and so we have no way of accurately analysing how it is that men do
make history; indeed the circumstances not of their choosing seriously
delimit, structure and influence the forms and the results of social
struggle.
Much sociology has consisted of rejecting functionalism and replacing
it with an action approach, of emphasising the meanings actors place
upon their experiences and, indeed, to suggest that social reality is itself
intersubjectively constituted out of these subjective meanings .15 I think
that contemporary Marxism is currently polarised around a similar
dichotomy. On the one hand, there are varieties of Marxist functionalism;
on the other, of Marxist humanism. In this book I aim to show that this
dichotomy can be transcended, and thereby also to show that the
dichotomy of orthodox sociology, between systems and action ap-
proaches, can be replaced by a preferable conceptualisation of contem-
porary capitalist societies.
It will be central to my argument that the conventional Marxist
topographical metaphors, of base and its superstructure, or the three-
layered economy, politics and ideology, are both inadequate. In neither
case can we adequately grasp the forms and effects of struggle; by which I
mean the multitude of different efforts by which both individuals and
groups of individuals struggle to maintain and expand their material
conditions oflife. I take it that these are essential to capitalist society and
that what has to be explained are their precise bases, their degree of
effectivity and their often largely unintended consequences. The two
Marxist approaches mentioned earlier both treat of these struggles
inadequately. Marxist functionalism tends to view them as merely
helping to restructure capitalist society in a manner which ensures its
more effective operation; Marxist humanism fails to explore the
particular effectivity of such struggles. I shall argue that it is necessary to
6 The Anatomy of Capitalist Societies

theorise their effectivity, to identify their place in the theory we develop

of capitalist societies.
This means that we have to return to the well-worn distinction between
function and cause. I shall try to show that the contemporary state is in
effect the consequence of social struggles of various sorts; struggles
themselves structured and delimited in certain ways. The effect of such
struggles may well be to produce changes which are 'functional for'
capitalist accumulation. However, the function we identify is not its
cause.
This emphasis on the forms of social struggle, of their effectivity and
unintended consequences, further means that much of what passes for
social class analysis is highly inadequate. Generally sociology has treated
class as an effect rather than as a cause. To the extent that it is treated as a
cause this is in relationship to the individual subject who is seen to adopt
certain beliefs or practices as a consequence of class membership.
However, I shall argue that the balance of class forces is highly significant
for state policy; and indeed that the state cannot, and must not be taken, as
perfectly functioning to reproduce capitalist relations. 16 It is crucial to
analyse the interrelations between civil society and the state, of the degree
to which they are independent of each other and of the manner in which
the dominant sphere of civil society affects the forms of struggle and the
state. I shall argue against all attempts to reduce the state directly to the
economy or to the dominant capitalist class. The state is indeed, to
varying degrees, separate from all classes and other social forces; and this
relative independence enables the state to be used against certain of these
classes and forces, incl uding the different fractions of capital. 17 How-
ever, it is also necessary to go on to consider the changing relationship
between civil society and the state; of the changing forms of represen-
tation within the state, and of the consequences that that produced, for
example, in being less able to sustain a reformist programme dependent
as that is on the clear dichotomy between state and society.
This emphasis on the relatively indeterminate yet structured civil
society also entails the rejection of the thesis that there is a realm of the
state and/or ideology which effectively produces untroubled social
cohesion. Furthermore, although civil society is that site where indi-
vidual subjects reproduce their material conditions of life, there is no
presumption that such reproduction is structured so as to be that most
appropriate to the developing needs of capital accumulation. This book is
thus focused around the need to produce a new way of theorising
contemporary capitalist societies which does not presume that such
 Introduction 7

socIeties are functionally ordered and cohesive with an effective

incorporation of the subordinate social forces through a dominant
ideology. I thus take as given the kinds of theoretical and historical
objections to the dominant ideology thesis which have recently been
advanced by Abercrombie, Hill and Turner. IS I argue that much of what is
customarily taken as ideology is properly to be viewed as part of the
practices of civil society; and hence the term 'ideology' is reserved to
apply to a fairly specific kind of effect which is present within practices of
very different sorts, in which the causes or consequences of that practice
or of some other practice are concealed. Such ideological effects (or
'functions ') are not to be taken as the explanation of the origin or the
persistence of that particular practice.
In various recent writings Hindess and Hirst have argued that the
concept of 'social formation' is the proper object for theoretical
development. '9 It is certainly true that there is no properly formulated
understanding of such capitalist social formations. And as a consequence
of this lacuna, and of the apparent sophistication of our understanding of
the capitalist mode of production, it is tempting to reduce such formations
to the laws of motion which characterise the mode. Now this is
problematic, one response to which is to develop the thesis that there is
present not merely one mode of production, but a number. 20 This problem
has been especially discussed in relationship to the 'articulation' between
modes of production within the sociology of development. 21 It has also
been discussed in connection with the relationship between the patriarc-
hal and the capitalist modes of production. 22 There are, however, two
problems with such an additive approach. First, even if we could
establish just which modes are present within contemporary capitalist
societies, there is still a great deal about these societies which is not
directly brought into existence by those modes (racial conflict, for
example). And second, it is extremely difficult to see how it is possible to
establish just how such apparently separate modes are interrelated - what
kind of space they occupy in relationship to each other. Even if we do
consider that such alternative 'modes of production' do exist, it does not
follow (a) that they are of remotely similar structure or significance,23 or
(b) that they can be understood without a general understanding of the
patterning of capitalist social formations. Hence, in the following
chapters I shall attempt to develop a theorisation of such a patterning; and
thus of the place of the various forms of production which are subordinate
to the capitalist relations of production. I shall not, however, consider
petty commodity production, domestic labour or the state as separate
8 The Anatomy of Capitalist Societies

modes, only as separate forms of production whose place and significance

is to be understood by relationship to, but not determination by, the
overall social patterning of capitalist social formations.
Finally, let me connect my argument to recent work which has
attempted to develop a realist conception of science. 24 There are a number
of major problems in such a position as applied to social science: these
include the ambiguous theoretical status of an 'underlying' or 'causal'
mechanism; the degree to which realism evades the problem of induction
and verification; the degree to which it provides an exhaustive conception
of science; the degree to which it legitimates what appear to be different
forms of social scientific practice (for example, individualist philosophy
of action, structuralism and Marxism); and the manner in which it appears
to justify reductionist conceptions of social structure. The last of these
can be seen from the realist interpretation of Marx's theory of modes of
production. On this account it is held that Marx postulates the existence of
a mechanism(s) which, if it were to exist, would explain the empirically
observable phenomena within certain societies. This mechanism(s) may
well be invisible to our normal sense experience, although typically there
will be indices of its existence. Clearly, within capitalism it can be held
that the production and appropriation of surplus-value constitute such a
mechanism, and Marx details how this gives rise to prices, wages,
profits, interest, rent and so on, the phenomenal forms. However, such a
conception is problematic in two respects. First, it may lead to viewing
such societies as characterised by an 'expressive totality' , that all aspects
or elements of it are merely the phenomenal forms of the inner essence or
mechanism. And second, no account is provided of the precise kinds of
practice which individuals would have to engage in, so that these
phenomenal forms are generated. Many aspects of capitalist society are
not unmediated expressions of its central mechanism; they result from the
forms of social practice and struggle in which individual subjects are
forced to engage. Thus a realist account which ignores this embodies a
form of rationalism, that the logical relations between concepts in theory
yield us knowledge of the relations between real events in practice. 25
However, as I shall now attempt to show in the next chapters, it is crucial
to establish, not just these relations between concepts in the theory of the
capitalist mode of production, but also concepts which designate the
social space in which individual subjectivities are constituted and
reproduced, the differing forms and effectivity of social struggle, and the
character of the state.
Let me now try to show that there are concepts which we can employ to
 Introduction 9

theorise the variable structure of capitalist societies, concepts which are

not logically derivable from the mode of production. Central to my
argument will be an attempt to show that such societies do not constitute a
unity as given by the underlying mechanism of surplus-value production
and appropriation. Rather such societies consist of patterned yet
heterogeneous social practices, which I shall now begin to detail.
2
Civil Society

In this chapter I shall try to show that certain attempts to reject the concept
'civil society' fail and that there are therefore good prima facie grounds
for arguing for its significance. Much of the time I shall be concerned with
Poulantzas's rejection ofthe concept. I shall attempt to show that this is
misguided and that the employment of this term would resolve certain
deficiencies in his argument. I am not claiming that the concept is
unproblematic. The chapter will end with a critique of Gramsci, pointing
out, inter alia, his inadequate grasp of the economic relations of
capitalism. In the following chapters I shall explore one particular
formulation that Gramsci provides, namely that the way to understand
civil society is as a set of social relations that lie between the economic
structure and the state. To indicate the importance of this formulation I
will begin the chapter by considering three characteristic ways within
Marxism in which the relations between the elements of capitalist society
have been viewed. In each case there are well-known deficiencies, certain
of which stem from a failure adequately to theorise the social relations of
civil society.
The first interpretation is merely the reductionist one that we
encountered before. Under capitalism there are a set of characteristic
forces and relations of production. These constitute the economic base of
society and this in turn necessitates a particular form of the state and of
ideology. The latter pair of relations are seen as necessary for the
economic base and are brought into existence so as to ensure its continued
existence. The capitalist economy is thUS. seen as necessitating and
producing a capitalist state and bourgeois forms of ideolo~y. There are a
number of well-known problems. First, it is difficult to avoid one of two
errors. On the one hand, it may be held that the state is directly governed
by, and in the interests of, the capitalist cla~s; an instrumentalist and
 Civil Society 11

historicist conception which ignores the evident complexity of civil

society and of how the state has to be situated in relation to all the
contending social forces in that society. 1 On the other hand, it may be
argued that the state is essentially functional and that the explanation of its
existence and of its changing form and policy lies in its functions for the
developing capitalist economy.2 This view clearly suffers from the many
well-known difficulties of functionalist explanation, in particular,
whether any mechanism can be provided of how ,the economy actually
generates such a state. Second, it is unclear on this reductionist position
what relationship holds between the developing contradictions between
the forces and relations of production and the social struggles of different
contending forces. Just what role does class struggle play in the
determination of the dominant relations of a capitalist society? It is
necessary to formulate a way in which such struggles can be theorised.
Third, on this view also it is difficult to see how we can understand social
relations which lie outside the state and the economy. There will be a
tendency to interpret such relations as essentially economic or to situate
them within the state. 3 Again we need to explore such relations
systematically.
In the second position, Marx advocates analysing the real relations of
the capitalist mode of production (the CMP), rather than its phenomenal
forms.4 No science, it is claimed, can re~ain on the level of appearances.
Marxist science must discover the underlying real relations of capitalist
society, its 'essence' or 'reality'. The appearances, or phenomenal
forms, are not unreal; they consist of relations within the sphere of
circulation, of distribution, exchange and consumption, as well as the
fetishised relations within politics, the state and ideology. These
relations, including for example the wage-form, derive from the essence
of capitalism, the production and appropriation of surplus-value, and
they serve to obscure that essence. Relationships are lived and experi-
enced at the level of the phenomenal forms of capitalism, in terms of the
exchange-values of commodities, rather than in terms of its underlying
real relations. The problems with this view stem from its Hegelian
legacy. It is difficult to know exactly what is meant by the contrast
essence/appearance - in particular it leads to the reductionist view that
all aspects of capitalist society are an ex pression of its 'innermost secret';5
that capitalism is an 'expressive totality' which is merely a more
sophisticated version of the first argument, of base and superstructure. It
therefore suffers from some of the same deficiencies. 6
The third type of position is one in which the economic is seen as
12 The Anatomy of Capitalist Societies

determining the causal significance of each of the different elements or

levels of a given society. It is the mode of production which determines
which ofthe structures happen to be dominant, as Catholicism was in the
Middle Ages or politics in the ancient world,6 and the causal interrela-
tions between this and the subordinate structures. The predominant mode
of production assigns rank and influence to each element; it is the
'particular ether which determines the specific gravity of every being
which has materialised within it'. 7 In capitalism it is capital, the
all-dominating power of bourgeois society, which assigns causal rank
and influence to politics, the state, ideology and culture. However, this
formulation, although allowing for the partial effectivity of the non-
economic structures, fails to show that will be the actual patterning of
such structures within any specific capitalist society. The analysis is
therefore open-ended. Instead what is needed is to demonstrate how these
different structures interrelate given different circumstances, in particular
given the different form and effectiveness of class struggle. Again we
need a way of theorising such struggles.
This does not of course exhaust Marx's (and Marxists') attempts to
formulate the relations between the 'economic' and the other levels/
structures/instances. However, these formulations are three of the most
characteristic. Yet each of them fails adequately to confront a number of
important claims which will provide the basis for my own argument.
These are claims which are central to Marx's position elsewhere,
explicitl y or implicitl y. If these are not adequately resolved then Marxism
collapses into reductionism or functionalism; one reaction to these being
to claim that political and ideological phenomena are in fact autonomous
and irreducible to manifestations of interests determined within the
structure of the economy. 8 I wish to argue that this dichotomy between
reductionism/functionalism and autonomism should not be sustained;
that there is a way offormulating the connections within capitalist social
formations that does not commit either set of errors.
Central to my argument will be a reassessment of the concept 'civil
society'. I am not maintaining that there is a readily available concept
here which can be simply inserted into a vacant conceptual space. Indeed,
it is clear that there is no such well-formulated concept - merely some
indications, especially in Gramsci's Prison Notebooks, which suggest
how a number of issues in contemporary Marxist theory might be
overcome. The general reassessment of the Althusserian position,
including the various auto-critiques,9 suggest that the tendency to write
off certain authors/texts as historicist has been, in certain cases,
 Civil Society 13

premature. Hall, Lumley and McLennan provide a convincing critique of

both Althusser's and Poulantzas 's treatment of Gramsci as historicist. 10 I
shall not maintain that to characterise Gramsci as historicist is not partly
correct; nor to argue the respective merits of these various authors on
some scale (e.g. of Marxist purity). However, I shall consider and
severely criticise Poulantzas's rejection of the concept of civil society. In
so doing we see how Poulantzas makes certain errors for which he is now
well known; and begin to see the barest outline of a more adequate
approach to civil society building upon some suggestions from Gramsci.
First, then, Poulantzas rejects 'civil society' in the following terms:

This concept of civil society ... refers exactly to the 'world of needs'
and implies the anthropological perspective of 'concrete individual'
and 'generic man' , conceived as subjects ofthe economy: which is the
correlate of the historicist problematic. The examination of the modern
state which follows from it starts from the problem of a separation
between civil society and the stateY

Poulantzas argues that such a concept makes 'impossible' the scientific

examination of the capitalist state. This is because it obscures how the
state is related to the class struggle; it makes it impossible to conceive the
specific autonomy of the political from the economic in the CMP; and it
prevents analysis of the effects of the ideological on the economy and
politics. Let us consider first why Poulantzas thinks that introducing the
concept of 'civil society' has these effects. I shall go on to argue that
Poulantzas's rejection of the concept is only possible given a number of
assumptions, each of which can be shown to be false.
First, then, Poulantzas deals with Marx's argument that the appearance
of the 'bare individual' is a historical condition of the CMP. Poulantzas
maintains that this does not refer to the real emergence of agents of
production as separate 'concrete individuals'. Rather the notion of the
'bare individual' refers to the central characteristic of capitalism, that
direct producers are separated from the means of production. Capitalism,
for Poul antzas, involves the coIl ectivisation of the labour process and not
its individualisation. So what is meant to be the basis of civil society and
of its separation from the state, namely, the existence of real, separate
individuals, is in fact merely Marx's way of characterising the separation
of direct producers from the means of production. It is this separation
which is then the basis for the 'respective autonomy' in the CMP of the
political structure and the economic structure. This further means that
14 The Anatomy of Capitalist Societies

individual agents, rather than being the basis of civil society and its
separation from the state, are to be seen as supports of the three
structures, of the economic, the political, and the ideological.
Nevertheless, Poulantzas does consider that it is essential to our
understanding of capitalism to be able to account for the existence of
individual subject-citizens. He situates their fixing institutionally but not
at the level of the economic or the relations of production. For Poulantzas
the existence of individual subject-citizens arises first from the 'juridico-
political superstructure of the capitalist state', and second from the
yuridicial and political ideology, which is dependent on the ideological
instance' .12 Thus, although within the economic level there is a
concentration of capital and the socialisation of the labour process, at the
juridico-political-ideologicallevel agents of production are constituted as
political and juridicial 'individual-subjects' deprived of their economic
determination and of their class membership. In consequence class
relations are experienced not as class relations but as fragmented and
atomised struggles between individuals. As the effect of the judicial!
political /ideological on the economic, what are in fact 'socio-economic
relations' are experienced not as social but as individual competition
between workers and between capitalists.
It is crucial for his argument that he is able to demonstrate that the
constitution of individual subjects occurs not as an effect of the
economic, of the relations of production, but of the political/ideological
instances. It is only if this can be demonstrated that he can then show that,
while Marx may employ the state/civil society couple in his later
writings, he has in fact left the problematic implied by that couple far
behind. In Marx's new problematic the use of the concepts state/civil
society indicates instead the autonomy of the capitalist state from the
relations of production, the effect of isolation upon such relations, the
nature of the state as representing the unity of the juridicially/political/
ideologically determined subject-individuals and the autonomy of the
economic and political class struggles. More specifically, Poulantzas has
to show that when Marx employs the division between the public and the
private this does not imply what one might think, namely the division
between the private sphere of economic individual/subjects and the
public sphere of politics and the state. This is because this very distinction
between the private and the public is a legal one which depends upon
specific juridico-political determinations. It is these determinations
which mean that the isolation of individUaJ.-subjects does not result from
the private sphere of economic individuals. Thus, when Marx appears to
 Civil Society 15

believe the opposite, as in The Eighteenth Brumaire, we must understand

his remarks 'quite differently'.
Poulantzas 's rejection of the concept of civil society is understandable
given the following three assumptions that he makes:

(1) that civil society is to be equated with the entire realm of the
private, of the needs, experiences, and relations of individuals , which
lie outside the public or the state - we will see below an alternative
location of civil society.
(2) that the economic is to be equated with the relations of production
and attendant labour process and that no conceptual space is allocated
to relations within the sphere of circulation.
(3) that classes and class struggles are determined by all three
structures, and that such struggles play no role in the formation of such
structures; agents are taken to be purely bearers or supports of
structures.

I shall now try to show why these three assumptions are false, considering
them in the reverse order.
(1) There are a number of problems about his treatment of classes and
class struggle. These stem from the separation he maintains between
structure on the one hand, and practice or the field of social relations on
the other. Qasses do not manifest themselves within the structure, but
exist rather as the effects of the three structures at the level of
intersubjective social relations. The structures of the economic, the
political and the ideological effect a structural determination of social
classes. Social classes are then the effect within the field of social
relations of these structures. However, curiously, the actual positions
which classes may take up do not necessarily coincide with their
structural determination. Social relations are not to be simply reduced to
the structures. Qass position need not correspond to class determination,
because there are further political and ideological factors. However,
Poulantzas fails to demonstrate how and why such factors may serve to
undermine structural determination, why a specific class situated within a
particular social formation takes on a class position at odds with its
structural class determination. In effect, Poulantzas presents us with a
sophisticated version of the in-itself/for-itself distinction, or that between
objective/subjective class. The main difference is that, for Poulantzas,
the objective determinants do not embrance merely the economic, but
also the political and the ideological structures. But this means that the
16 The Anatomy of Capitalist Societies

political and the ideological operate both to determine structurally basic

social classes and their objective interests, and at the same time to
produce class positions divergent from such objective class determi-
nations. The consequence is that classes are not precisely theorised since
no explanation is provided of why and how the political and the
ideological may in certain cases produce class positions which diverge
from their structural class determination.
A further effect of this theoretical separation between the structures
and the field of social relations is that it is theoretically impossible to
conceive of how the former are changed as the effect of class action
within the latter. I am not concerned here with the issue of revolutionary
change and with whether Poulantzas eternalises social formations
through employing a version of Althusser's conception of structural
causality. Rather, Poulantzas fails to theorise the effects of class struggle
in moderating, reforming, liberalising, transforming or making more
authoritarian the existing capitalist economic, political and ideological
structures. This is because he does not theorise social classes at the level
ofthe structures ofthe social formation. Classes exist in the field of social
relations, not at the level of the structures, and no theoretical space is
accorded to the processes by which the struggles of classes and class
fractions may in part determine those structures.
Poulantzas's approach also presupposes that there are no other
groupings external to social classes; that social classes exhaust the
division of the members of a society. This is apparently because all agents
enjoy a class membership, all such membership is nothing more than
class struggle, and such struggle only exists because of the existence of
social classes. Thus Poulantzas says that it makes no sense to argue that
there are 'social groupings' external to classes which are nevertheless
involved in class struggle. But this argument is problematic. It presup-
poses that there is only one kind of struggle in capitalist societies, namely
class struggle, focused around the dominant relations of production. It is
therefore reductionist, treating gender, racial and generational struggles
as derivative from class struggle. The fact that someone who is a member
of one particular grouping of gender, race or generation has also a
particular class membership, does not mean that the struggles of that
grouping are entirely structured by class struggle. Of course, such class
struggle will affect these other popular democratic struggles, but not so
that one can be reduced to the other.
How do these comments on Poulantzas's conception of social classes
relate to my conceptualisation of civil society? Broadly speaking the
 Civil Society 17

struggles of social classes to reproduce the material conditions of their

existence are part of, but not exhaustive of, civil society. Also present
within civil society are various other social groupings, particularly those
based on gender, race, generation and nation. These groupings are not to
be red uced to those of cl ass, although the dominant relations and forces of
production fundamentally structure the form that their struggles will take.
These groupings derive from how in civil society individual subjectivities
are constituted, of gender, race, generation, locality and nationality. The
discursive and non-discursive structures within which such constitution
occurs, particularly within the family, are in turn, related to, but not to be
reduced to, the dominant relations/forces of production. Civil society is
not then to be viewed merely as the world of individual needs, but rather,
as I hope to show, as sets of structured, institutionalised social practices.
(2) Poulantzas's second assumption necessary for his rejection of the
concept 'civil society' is that the economic instance is to be equated with
the relations of production and attendant labour processes, and that little
or no conceptual space is to be allocated to the sphere of circulation. As a
result the form in which the social relations of production are represented
within civil society is unanalysable; and therefore the concept civil
society can itself be rejected. This results from Poulantzas's 'technicist'
conceptualisation of the relations of production and a devaluation of
their val ue aspects. Thus he says that the relations of production concern
'the relations of the agents of production and the means of labour';13
nothing is said of the value basis of capitalist relations and of the
production and appropriation of surplus-value. Furthermore, he says that
the 'economic' refers to three relations, which are themselves dependent
on the combination of two economic relations 'real appropriation and
property, and so refer to the organisation of the labour process and to the
division of labour' .14 So 'property' relations are merely, on this account,
the same as the division of labour. The relations of production are
given a material, technicist interpretation. Poulantzas thus denies what is
central to Marx, namely the contradictory character of capitalism, be-
tween value and use-value, between labour-power and labour, between
variable and constant capital, and so on. Poulantzas fails to view the
relations of production as properly social.
He seems to make this mistake because he fails to see capitalism as a
system of generalised commodity exchange. He does not consider how
such a system necessitates both a sphere of production and of circulation;
and that the latter, the relations of distribution, exchange and consump-
tion, cannot be eliminated from the analysis of capitalist social
18 The Anatomy of Capitalist Societies

relations. IS If Poulantzas considers the sphere of circulation he mainly

views such relations as part of political and ideological relations, outside
the economy proper. I shall argue later that the sphere of circulation is the
crucial mediator between the sphere of production and the social relations
of civil society, and is properly seen as part of both.
(3) This relates to the third assumption that Poulantzas has to make in
order to sustain his rejection of the concept of civil society. This is the
assumption that civil society is to be equated with the entire realm of the
private, with the needs, experiences and relations of individuals, all of
which lie outside the public or the state. But this fails to distinguish the
different forms of practice, between the forces/relations of production on
one hand, and the mwtifarious 'individualising' social practices of civil
society on the other. In saying this I am following one particular
formulation offered by Gramsci. He says: 'Between the economic
structure and the state with its legislation and its coercion stands civil
society'.16 In this quotation Gramsci rejects the dichotomous view in
which the state is counterposed to civil society and in which the latter
embraces all non-state, non-pUblic relations. For Gramsci civil society
exists as a kind of intermediary, linked both with the economic structure
and with the state. This interpretation has the effect of undermining the
characteristic form of the base - superstructure relationship and of
replacing it with a more complex and plausible conception of the
dominant relations of a capitalist society. I shall therefore consider some
aspects of Gramsci's argument in order to see what further indication we
might derive as to how these relations are to be viewed.
However, before I do this it is necessary to consider one aspect of
Clarke's critique of Poulantzas, since the implication of his argument is
that it is not worth considering further the 'economic structure-civil
society-state' conceptualisation.17 He maintains that there is a tendency
in British Marxism to neo-Gramscian theories of class and the state.
Representatives of this tendency are among others Perry Anderson and
Ralph Miliband. They are neo-Gramscian in that they 'derived their
authority from a particular interpretation of Gramsci's work' .18 It is
unclear what is meant by the phase 'derived their authority', although in
Anderson's case the general interpretation is clear and not particularly
misleading. In Miliband's case, however, the position is rather unclear
and Clarke's particular characterisation enables him to interpret neo-
Gramscianism in a particularly 'instrumentalist' fashion. It is noteworthy
initially that Miliband, in The State in Capitalist Society, only refers to
Gramsci on four occasions, twice on hegemony, once on intellectuals and
 Civil Society 19

once on the revolutionary party. Of course, other concepts from Gramsci

may well be implicit in Miliband's argument. Clarke, however, suggests
that all neo-Gramscians interpose a distinctive level of civil society
between the relations of production and the state. The dominance of
capital is then expressed, not at the level of the relations of production
itself, but at the level of civil society. Capital's dominance results from
the existence of a social group which has a disproportionate share of
material resources and which ensures the subordination of the state and
ideology to its interests. The dominance of capitalist relations of pro-
duction is reduced to the dominance of capitalists within the sphere of
interacting social groups in civil society. This is a reasonable summary of
Miliband's argument. However, it is not an account of the necessary
nature of a neo-Gramscian position. Oarke has interpreted it to make it
approximate to an instrumentalist view; the state, and in a way ideology
and hegemony, being seen as instruments of the economically dominant
class. But it is not necessary that a neo-Gramscian position entails
instrumentalism. Indeed the emphasis upon the contending classes and
other forces in civil society would suggest an alternative view in which a
major determinant of state form and policy consists of working-class and
popular struggle; in other words, that the state is not at all merely the
instrument of the economically dominant class. It is also interesting to see
how Clarke's own position demands a conception of civil society, in
order to resolve certain difficulties that he encounters.
Clarke argues that there is a radical distinction between a Marxist and
an Althusserian conception of the whole. The latter, the pre-given,
complex, overdetermined whole structured in dominance and determined
in the last instance by the economic, is apparently equivalent to the
conception of the whole held by structural-functionalist sociology (not
just in the tortuousness ofthe language employed!). The former is derived
from Marx's conception of the total process of social production. The
relations of production are not simple relations of the immediate labour
process (not that Poulantzas says this), but are the relations of a total
process of production. Total here apparently refers to the whole society,
since Clarke quotes Marx from Wage - Labour and Capital when he says:
The relations of production in their totality constitute what is called the
social relations, society, and, moreover, a society at a definite stage of
historic development. '19 But does Marx seriously mean that the relations
of production are equivalent to all social relations in a society? Are
relations within education, relations of production? Surely not. Wage-
Labour and Capital was originally written in 1847, comprising a number
20 The Anatomy of Capitalist Societies

oflectures delivered by Marx in Brussels. But when Engels organised its

re-publication after Marx's death he had radically to revise it. For
example, he replaced the term 'labour' with 'labour-power'. And just
before this passage Marx gives a highly technicist conception of capital (it
comprises raw materials, instruments of labour and means of subsistence
of all kinds), and a technological determinist view of historical
development. 20 Thus, the pamphlet is theoretically unsound and Clarke's
reliance on this passage from Wage-Labour and Capital cannot be
sustained without further argument.
But Oarke also maintains that his is not a reductionist (or essentialist?)
position. This is because: 'To take the relations of production as the
starting point of analysis is not, therefore, to introduce a reductionism,
for the relations of production are already social. '21 But this merely
means that it is not a technological, or a material reductionism. It does not
mean that Oarke is not presenting a reductionism in which the social
relations of capitalist production determine or dominate all other social
relations within a capitalist society. Indeed what are we to make of the
following? 'Political and ideological relations are as much relations of
production as are strictly economic relations, for they are specific forms
of the social relations within which production takes place. ' 22 This can
only be sustained by completely dissolving the specificity of the moment
of production. Yet three lines later Clarke himself distinguishes produc-
tion from the 'derivative' economic relations of distribution, circulation
and consumption. And he further suggests that it is only through a
historical-materialist analysis that one can establish concretely both the
domination of all social relations by the capital relation and the limits of
that domination. But if there are limits to that domination how are we to
think through, to theorise, such limits? Either society is given in its
totality by the relations of production (as Marx erroneously implies in
Wage-Labour and Capital); or it is not, in which case we need concepts
which enable us to theorise social relations which are not directly
relations of production. Following the second suggestion, Oarke argues
that the Marxist theories of ideology and the state will have to show 'how
and to what extent political and ideological relations are forms of the
relations of production' (my italic). 23 But Clarke provides no means of
achieving this. What is needed is a theorisation of such political and
ideological forms which brings out both their interconnection with the
circuits of production-distribution-exchange-consumption and the nature
of social classes and class struggle; in other words, with civil society. I
will now consider Gramsci's interpretation of this concept before
 Civil Society 21

proceeding to my own attempt in later chapters to resolve a number ofthe

issues so far discussed.
Gramsci, as is well known, was inconsistent in the way in which he
used the central terms of his analysis, and there is little point engaging in
an essentially academic argument about which is Gramsci's true
usage. 24 We may note that the state is contrasted with civil society;25 the
state is seen as encompassing civil society;26 and the state is seen as
identical with civil society.27 Civil society sometimes includes economic
relations, and sometimes excludes such economic relations. 28 Civil
society is seen as constituting the outer trenches of modem society which
protects the state lying inside;29 and the state is taken to be an outer ditch,
behind which lies a powerful system of 'fortresses and earthworks', of the
superstructures of civil society. 30 Hegemony likewise is conceptualised
in a number of different ways: as referring to the class alliance of the
proletariat with other exploited classes and especially with the achieve-
ment of intellectual, moral and cultural leadership;3J as the form of
intellectual, moral and social leadership and domination of the
bourgeoisie over the society as a whole as opposed to their domination
via force and violence;32 as located within civil society by contrast
with domination which is located within the state;33 and as identifi-
cation of legislature, judiciary and executive as elements of political
hegemony, hence hegemony as a feature of both the state and civil
society.34
Anderson has summarised Gramsci's different conceptions into three
models of the relations between the state, the economy, civil society and
hegemony. In the first, Gramsci takes hegemony to be the mode of
bourgeois power within civil society, while in the state, domination is the
characteristic mode. The state is seen as dominated by civil society,
which therefore means that the power of the bourgeoisie is principally
sustained through hegemony, through the institutions and associations of
civil society. The role of the state, of domination and coercion is less
significant. But Anderson points out one major deficiency of this
position; namely, that perhaps the essential ideological basis of western
capitalism is not civil society but the state or bourgeois democracy in
which all individual citizens are represented. The novelty of the consent
engendered by such democracy is that people believe that they do
exercise ultimate determination - that they can and do determine
outcomes within the political arena. In other words, it is not just a
question of a ruling class being seen as legitimate; it is rather the belief
that there is in fact no ruling class, indeed that there are no classes
22 The Anatomy of Capitalist Societies

politically. The state represents all these separate individual citizens, it

represents their totality, their equal participation formally in the
determination of policy. Consent then cannot be attributed as in
Gramsci's model simply to the institutions of civil society. Consent of a
peculiar and generalisable sort is specifically generated within and by the
bourgeois state.
In the second position there is thought to be a balance or equilibrium
between the state and civil society. Hegemony is distributed between the
two - and civil society is seen to embrace both coercion and consent.
Oearly this redresses the specific failure of the first model, because
ideology is now seen as a property of both the state and civil society.
However, in making this change Gramsci falls into an alternative error.
He sees coercion as a general characteristic of both civil society and the
state. But in fact the institutionalised legal use of coercion is not a general
feature of civil society. Although there may be semi-autonomous private
associations which exercise institutionalised coercion, these are both
exceptional and only operate under the sway of the repressive state
apparatuses. Ideology then is characteristic of both civil society and
state - but institutionalised and legitimate violence is characteristic only
of the state.
Third, Gramsci sees both political society and civil society as
comprising the state. In reality the state and civil society are the same.
There is no distinction to be made between the public and the
private - they are both part of the hegemony of bourgeois society, that
society in which bourgeois relations of production are generated and
reproduced. The power of the state is therefore as importantly cultural
and ideological as it is coercive. Thus the state embraces both the public
realm and the private associations of bourgeois society. Althusser
recently has elaborated this in his notion of ideological state apparatus, in
which churches, parties, trade unions, families, schools, newspapers, all
constitute apparatuses of the state. 35 There is however a major problem in
the collapsing of civil society into the state; namely, that it becomes
difficult or even impossible to distinguish bourgeois democracy from
Bonapartism or from fascism. Yet it is crucial to be able to determine the
precise form and effectivity of the state. Bourgeois democratic states
entail quite different relations between themselves and their respective
civil societies, than do fascist or Bonapartist regimes.
It is important to note that Gramsci was not trying to develop a general
theory of the relations between these concepts. His aim was rather to
consider the nature of the state and civil society in specific historical
 Civil Society 23

conjunctures. In particular the discussions in the Prison Notebooks

revolve around the development of imperialism and the dominance of
finance capital, and of the changes that these effect in working-class
struggle. With imperialism there is a qualitative increase in the scale and
complexity of the state and an increase in its dominance over civil
society. Civil society too has become more complex and less resistant to
the effects of immediate economic crisis. Yet, in part these changes have
resulted from the increased effectivity of mass movements and of their
role in politics and the state, and this in tum has created a crisis of
hegemony for the traditional political institutions and political parties.
This for Gramsci creates a potential basis for the working class to forge its
own unity and an alternative project, and so establish its hegemony over
bourgeois society. He sees this though as involving a long and hard
struggle, a long war of position, to wrest hegemony from the increasingly
elaborated and extended state and of the progressively complex and
unshakeable civil society. 36
The main merit of Gramsci 's approach is to have attempted to theorise
the class struggle within capitalist society. In other words, he considers
how the changing relations between the state and society, both in part
stem from class struggle and in turn affect the forms that such struggle
may take. More specifically he maintains that the dominant class must be
able to establish its hegemony and to form a historic bloc with other major
classes. Thus, the reproduction of capitalist society is crucially dependent
upon political and ideological struggle and not simply upon economic
determination. The nature of this political and ideological struggle cannot
be reduced to determination by the economy. There is a continuous
process offorming and superseding ofthe interests ofthe dominant class,
both with potential allies and with subordinate classes. The formation of
hegemony is moreover not a question of the hegemonic class simply
imposing its ideology upon upon allied and subordinate classes. Classes
are not in general the subject of history, nor are ideologies merely
nameplates carried around on the backs of particular classes. 37 The
establishment of hegemony involves the transformation of existing
ideological practice, to restructure and to reorder prevailing categories.
Ideological complexes are always the consequence of a relation of forces
between rival hegemonic principles, the winner being that principle
which is able to 'nationalise itself'. 38 Gramsci lays considerable emphasis
on the necessity for a potential hegemonic class to appropriate to its
discourse the 'national-popular' elements present within that society. It
cannot avoid struggling to become a national political movement. The
24 The Anatomy of Capitalist Societies

war of position is 'national' and it is from this crucial fact that class
struggle must proceed.
Although Gramsci's perspective is highly illuminating there are
various difficulties in his formulations, apart from that of the shifting
meaning of the key terms. Bobbio argues that he adheres to the following
four propositions: 39
(a) like Marx we should adhere to the distinction, structure/
superstructure;
(b) since civil society is not part of the structure it must be
superstructural;
(c) within the superstructure civil society dominates political society;
and (d) the superstructure dominates the structure.
Although it is doubtful that he consistently adheres to all four assump-
tions, it is true that he does not do enough to prevent such over-culturalist
interpretations. In the following chapters, therefore, although I shall
employ a conception of civil society bearing some affinities with
Gramsci, I shall try to distance myself from implications (a)- (d). Thus, I
shall argue against the dichotomy, structure/superstructure, suggesting
that Marx's own practice precluded such simplistic dichotomies. Also I
shall not imply that political society is less important than civil society.
On occasions Gramsci grossly underestimates the manner in which
ideological relations are in fact underpinned by the power of in-
stitutionalised legal coercion residing within the state. Much of the time
insurrection is an impossibility simply because of the armed power of the
state; and there is little or no need for hegemony to be established.
Gramsci seems to presume that dominant classes must establish
hegemony; it would be preferable to presume that this is only contin-
gently so. Further, Gramsci fails to analyse in any detail the disparate
social practices of civil society, except in so far as they help to establish
hegemony. He does not consider in sufficient detail the heterogeneity of
social practices, of their often contradictory and conflictual nature, and of
the fact that certain practices relate to the most personal and spontaneous
forms of human activity. 40
Finally, we should note how Gramsci neglects the economy. For
example, in the list offactors which ensure hegemony he largely ignores
capitalist enterprises. So when he suggests that civil society lies between
the economy and the state we have to understand these remarks as
metaphorical. It is necessary to start again, taking Gramsci as inspiration
but as providing us with little in the way ofsubstantive contribution. We
 Civil Society 25

have already seen that, on the relationship between the state and civil
society, Gramsci is at times unsure of the relationship and on occasions
conflates the two. In the following I shall try to avoid such conflation, but
to do so in a manner which does not deny the interconnections between
them, especially through class and popular democratic struggles.
Likewise the relationship of civil society to the 'economic structure' is
highly complex and presupposes some clear understanding of the nature
of capital (which Gramsci clearly did not possess and which seriously
flaws his analysis). In the rest of this book I attempt to formulate these
relationships more fully and adequately than is hinted at in Gramsci's
cryptic formulation. I try to support the claim that when Marx employs
the concept of 'civil society' in Capital (explicitly or implicitly) this is on
new conceptual terrain. No longer is civil society seen as comprised of
individuals and of their needs and dispositions - individuals who are the
subject of the economy. Rather, civil society refers to 'the individualizing
sphere of the circuit of capital's path to expanded reproduction' .41 We
will now see how and why. But we will also see that Marx's conception is
somewhat narrow and that it is necessary to go beyond both his
formulations, as well as those of Gramsci, in order to arrive at a
conceptualisation which is remotely sensitive to the dynamics of struggle
in late capitalist societies.
3
The Spheres of Production
and Circulation

There have been two main traditions which have conceived of social class
in terms of essentially economic criteria. On the one hand, there is the
tradition broadly characteristic of British sociology which has been
derived from Weber's cryptic formulations. This is an approach to class
in which the main distinctions drawn are based on the market, on the
differential allocation of life-chances. Different classes are able to
command different sets of resources within exchange. On the other hand,
there is the Marxist tradition in which social classes are determined by the
relations of production , by the ownership/non-ownership of the means of
production, by productive and unproductive labour, and, in certain
formulations, by the functions of supervision and management. On this
view the Weberian position is misleading since, in concentrating upon
relations within the sphere of circulation, the 'real relations' of the
capitalist mode, of the production and appropriation of surpl us-value, are
hidden from view. Gasses for the Marxist are determined at the level of
production - relations within the market are merely the phenomenal
forms of the underlying real relations of capitalist production. A theory of
class based merely on market relations cannot be correct.
In the rest of this book I support a conception of class having more
affinities to the Marxist than the Weberian position. However, there are a
number of problems of this Marxist interpretation which I will now
indicate - problems which are partly indexed by the Weberian emphasis
upon exchange relations. First, in the Marxist view the precise location of
social classes remains unclear. We have already noted the problems that
confronted Poulantzas when he argued that social classes were structur-
ally determined. In general Marxists have not adequately theorised the
place of social classes in capitalist social formations. Such a theory would
have to allow for (i) the determination of at least the working and
 The Spheres of Production and Circulation 27

capitalist classes by capitalist relations of production, and (ii) the

effectivity of the struggles of certain social classes either in directly
transforming aspects of capitalist society (e.g. welfare state), or in
indirectly causing other changes through the responses engendered by
conflicting social classes (e.g. changed managerial strategy). Second, the
form that such class struggles take is sectional; that is, classes as a whole
do not engage in struggle but only specific groups of workers or capitalists
trying to reproduce the material conditions of their existence. As such
they are forced to confront sectors of opposing social classes in the market
by attempting to improve their life-chances. Exchange relations are thus
initially where groups of workers, as well as capitalists, the middle
classes and so on, conflict. Given the nature of labour as free, they are
literally forced to struggle to improve their wage payments in the sphere
of circulation. And as such they will act sectionally, using any possible
basis by which to gain sectional advantage. This is not to say that such
struggles do not have massive implications for the overall struggle
between capital and labour, but the nature of capitalism is such, with the
division between the spheres of production and circulation, that it is
within the latter that labour and capital primarily conflict, and conflict
sectionally rather than as unified classes. I will now show this through a
discussion of the spheres of production and circulation and of the concept
of civil society. I will briefly consider that concept in The German
Ideology before proceeding to a use of it in Capital, especially in vol. I,
which is consistent with my argument.
We have already noted Poulantzas's objections to the term 'civil
society'. These are very similar to, and in fact derive from, those of
Althusser, especially in his essay on 'Contradiction and Overdetermi-
nation'. 1 Althusser says that Marx provides, in The German Ideology, a
sort of philosophico-economic phenomenology, a description and foun-
dation of economic behaviour in terms of the needs, personal interests and
relations of individuals. But since we know that, in his later work, Marx
rejects any explanation of economic life in terms of such individual
needs, in terms of homo economicus , what he does in fact is to account for
such individual economic behaviour in terms of 'a deeper, more concrete
reality: the mode of production of a determinate social formation'.2
According, then, to Althusser, the concept of civil society thus merely
signals the problem of individual economic behaviour ('dig here '), but
civil society is definitely not one of the basic Marxist concepts which
enable us to theorise its conditions of existence.
Oearly here Althusser is attempting to maximise the distance beiween
28 The Anatomy of Capitalist Societies

the Marx of The German Ideology (not forgetting that trusty Engels, of
course) and that of Capital. Yet if we look at Marx's formulations in the
former we find something curious. He says that civil society:

embraces the whole material intercourse of individuals within a

particular stage of the development of productive forces. It embraces
the whole commercial and industrial life of a given stage and, insofar,
transcends the State and the nation .... The term civil society emerged
in the eighteenth century .... Civil society as such only develops with
the bourgeoisie. 3

There are few if any implications here that economic and social life is to
be explained purely in terms of individual needs, in terms of a homo
economicus. Admittedly the argument is weak and suggests a technologi-
cal determinism, that civil society is principally determined by the level
of productive forces. However, it is not reasonable to criticise it as
Althusser does - his symptomatic reading leads him to overestimate
Marx's humanism with regard to the realm of individual needs.
What in fact is inadequate in Marx's formulation is a point only
tangentially referred to by Althusser. Consider two further quotations
from The German Ideology:

this civil society is the true source and theatre of all history, and how
absurd is the conception of history held hitherto, which neglects the
real relationships and confines itself to high-sounding dramas of
princes and states. 4

Through the emancipation of private property from the community, the

State has become a separate entity, beside and outside civil society. S

In both quotations Marx draw the contrast between the state (the
'high-sounding dramas of princes and states ') and the 'real relationships'
of civil society. There is nothing wrong about studying such civil
society, seeing it as that place where individuals operate and their real
relationships are founded. What is problematic about Marx's formula-
tions at this stage of his development are: (1) his failure to see that
individual subjects are not already pre-formed as subjects - rather
individuals are constituted as subjects by a series of interpellations within
civil society; and (2) his failure to analyse just how civil society is
structured by the relations of capitalist production, in particular by the
 The Spheres of Production and Circulation 29

relationship between the spheres of production and circulation, the latter

providing mediation between the 'economic' base and the other elements
of a social formation.
I will now set out some general propositions on the economy and civil
society before proceeding to explore the relationship between production
and circulation in more detail.
Capitalist social formations comprise a dominant capitalist mode of
production, with various subordinated forms of social and private labour.
In certain formations there will be a fully established subordinate mode of
production in which all its conditions of existence are secured. This is
unusual. Generally there will be elements of pre-existing modes or
elements of particular types of labour which continue as a result of the
reproduction of the conditions which guarantee their existence as
subordinate forms of labour. Any mode of production consists of a set of
relations of production and a set of forces of production: the former being
those relations in which one set of functionaries produces surplus-labour
which is appropriated by another set of functionaries and the associated
allocation of the means of production; the forces of production being the
form in which nature is appropriated through the organisation of the
labour process. A mode of production is given by the unity of these two
under the dominance of the former. In the capitalist mode, surplus-labour
takes the form of surpl us- val ue, the means of production are possessed by
the non-labourers, who therefore appropriate the surplus-value produced
by the labourers. Because surplus-labour takes a value form there is a
separate realm of circulation in which surplus-value is realised, a sphere
of exchange in which all commodities, including that of labour-power,
are bought and sold. The forces of production under capitalism are
organised so that progressively the direct producer is excluded from
possession or control ofthe direct instruments of production. In the stage
of machinofacture the producer is dependent upon the pace and rhythms
of the machine. Mental labour is progressively stripped from manual
labour and is embodied in differentiated and highly specialised mental
labourer functions. The labour process takes this form because of two
aspects of capitalist production which together distinguish this from other
modes of production. First, the capitalist purchases labour-power not
labour; that is, the capacity to labour for a given period of time. If set to
work satisfactorily it yields value over and above that necessary to
reproduce it. However, the achievement of this involves a continual
problem of management since in the labour process the direct producer
has to be reunited with the means of production. The worker enters a
30 The Anatomy of Capitalist Societies

normal active relationship with the means of production - a relationship

determined by the nature and purpose of the work in question. As Marx
says: 'The worker treats the hide he is tanning simply as the object of his
creative activity, and not as capital. He does not tan the hide for the
capitalist. '6 Second, the labour process is specific to particular units of
capitalist production, of capital-units which are in competition with each
other. The ability, then, of particular units to appropriate value is
governed by their competitive strength, nationally and internationally.
Capitalists compete with one another in the buying and selling of all
commodities, including that of labour-power.
The effect of both these points is that it is incorrect to view economic
relations within capitalism as a level, or a structure given simply by the
relationship between capitalists and labourers, or between capital and
labour. Capitalism is premised in fact on exchange, on the fact that
production is separated from circulation without which surplus-labour
cannot be realised in a value-form. When therefore I use the term
'economy' I mean within capitalism that structure of social relations
which embraces production and circulation. And one cannot use the term
'mode of production' in the case of capitalism without understanding that
production and circulation are distinct - indeed that capitalist relations
and forces of production cannot exist without a separate sphere of
exchange where labour-power is bought and where surplus-labour is
realised in the form of surplus"value. There is not, then, an economic
level or structure or instance - there is rather a set of circular relation-
ships, of circuits, especially that of M-C-Cl_Ml which begin and
end within the sphere of circulation. It is that circuit by which each
capital-unit, which mayor may not be owned by one or more identifiable
capitalists, competes with other capital-units to expand value, to
purchase labour-power and the means of production, to set them to work
to produce commodities which when sold in the market yield the
maximum mass of profits relative to the money advanced.
This characterisation of the eMP highlights how capitalist relations
presuppose a sphere which is relatively autonomous, the sphere of
circulation. It is this sphere which is the basis of the relative autonomy,
both of the state and of civil society. In relationship to the state this is
because, although its form is given by its function in attempting to sustain
the overall conditions for continued capital accummulation, it effects this
by working within the sphere of circulation. The state attempts to
maintain capitalist production relations by moderating, reforming and
transforming the relations of exchange, of ensuring that suitable
 The Spheres of Production and Circulation 31

proportions of different commodities are produced and reproduced which

are those necessary for capitalist production. This will be seen in
Chapters 6-9 when I discuss in more detail the nature of the state and
particularly its relationship with civil society. For the present let me
consider how the sphere of circulation is the basis for the relative
autonomy of civil society.
Civil society I define as that set of social practices outside the state and
outside the relations and forces of production in which agents both are
constituted as subjects and which presuppose the actions of such
subjects - first, in the sphere of circulation directly; second, in those
social relations within which labour-power is reproduced economically,
biologically and culturally; and third, in the resultant class and popular
democratic forces. Although I will discuss these points in detail in
Chapter 5 I will now make some qualifying comments on this for-
mulation. First, I will use the term 'social relations' of production to
embrace the unity-in-dominance of the relations and forces of pro-
duction. It is to be assumed that even if I use the term 'capitalist social
relations (of production)' this presupposes that there are subordinated
forms of social and private labour. Second, unlike the state and unlike the
social relations of production, there is no particular unity of the social
practices of civil society. Their only common characteristic is that they
are focused upon the processes by which individual subjects are
constituted. Thus, civil society embraces widely divergent practices,
from family relations to commodity markets, from trade-union organis-
ations to religious bodies. Third, the concept civil society embraces many
of the topics normally analysed by orthodox sociology - indeed such
sociology also treats of the economy and the state as though they were
roughly equivalent to each other and to the other institutions of civil
society. Sociology reduces such practices to some common form and sees
them all as varying elements of civil society. My argument, by contrast,
is that the social relations of production and of the state are distinct - they
both presume a particular unity, organisation and functional relationship
to the rest of the social formation. But while orthodox sociology errs in
failing to see the distinctive character of capitalist production and the
state, conventional Marxism tends to consider social formations as
though they were only constituted of capitalist relations of production and
the state, with ideology sometimes added as an afterthought. 7 No space is
allocated to the processes by which agents get constituted as subjects
outside the social relations of production on the one hand, and the state on
the other hand. Nor is space allocated to social classes apart from their
32 The Anatomy of Capitalist Societies

determination by the economy or their effects within the state. And

finally, such Marxism cannot analyse in a coherent non-reductionist form
political groupings and movements oriented around other interpellations
of gender, age, race, region and nationality.
Fourth, it will be noted that I have as yet not used the concept
'ideology'. Although I will discuss this in much more detail in the next
chapter, for the present it should be noted that there are two types of error
in much Marxist discussion of this concept. On the one hand, in
formulations which employ versions of the capital/state couple, ideology
is seen as a residual category referring to what is left over when the really
significant social relations have been analysed. Clearly in such formula-
tions no proper understanding can be expected although they do have one
advantage in that they highlight the heterogeneous character of ideology
in capitalist social formations. On the other hand, it is held that ideology
constitutes a structure or a level within the social formation, and indeed
that it has a unity derived from its specific function in relationship to that
formation. This view is often associated with an erroneously essentialist
conception of the human subject. By contrast I wish to maintain that the
ideology of a particular formation has no necessary unity, that it does not
constitute a distinctive level, and that it does not stem from or presuppose
anything essential about human beings. In fact there are many different
elements of capitalist social formations each of which has certain
ideological effects; including the social relations of production, the state,
the social relations within which subjects are constituted, classes and
other non-class forces. Each of these has ideological effects - none are
strictly speaking ideology per se. Each has specific conditions of
existence and involves forms of practice and social struggle.
I shall now consider that which I have argued constitutes the
fundamental condition of existence of civil society; namely, the division
between production and circulation. I shall discuss this by, in a sense,
working backward, by beginning with Marx's analysis of the sphere of
circulation and especially with how it is there that 'Freedom, Equality,
Property and Bentham'8 prevail. I shall then consider the general account
in the Grundrisse as to the relations between production and circulation,
concluding with the discussion, especially from the third volume of
Capital, as to the nature of capital and its circuits.
Surplus-value for Marx cannot arise from circulation and therefore, for
it to be so formed, something must take place, in a sense, invisible to the
circulation process. 9 And yet surplus-value cannot occur apart from that
process, from the sum total of all the mutual interactions of the
 The Spheres of Production and Circulation 33

commodity-owners. Capital can thus neither arise from circulation, nor

apart from circulation. The money-owner must buy his commodities at
their value, sell them at their value, and yet at the end must withdraw
more value from circulation that he threw in at the beginning. At this
stage he is a capitalist only in larval form - his 'emergence as a butterfly
must, and yet must not, take place in the sphere of circulation '. \0 This is
possible only because the money-owners, Mr Moneybags, must find
within the sphere of circulation a commodity whose use-value possesses
the peculiar property of being itself a source of value. There is of course
such a commodity, namely labour-power.
However, in order that the money-owner finds labour-power available
on the market certain conditions must be satisfied. First, the possessor of
labour-power must be the free proprietor of his own labour. Hence, he
and the money-owner meet in the market and enter into relations with
each other on a basis of equality. They are both owners of commodities,
one is a buyer, the other a seller, and (generally) in the eyes of the law
they are equal. Second, the possessor of labour-power must not possess
means of production which could themselves be used to produce
commodities directly. He must be separate from the means of production
so that in order to subsist the labourer sells not commodities but
the commodity of his own labour-power. So, within the market, the
labourer is free in a double sense - free to dispose of his labour-power as
his own commodity and free from the encumbrance of other commodities
to sell.
This sphere of circulation, within which the buying and selling of
labour-power is merely one especially important example, is the
exclusive realm of 'Freedom, Equality, Property and Bentham'. There is
freedom because buyers and sellers are freely able to buy and sell - no
one forces them to enter a particular market at a specific time, to sell or to
purchase a certain commodity. Buyers and sellers contract as free
persons. Relationships are equal because there is the exchange of
equivalence which results from the equalising processes of the market.
The purchasers and sellers of commodities are equal parties each of
whom has a common status as an exchanger. There is an all-sided
equality of subjects. The realm is one of property because each person
only sells what he or she possesses - what people have as property to
dispose of in one market or another. And relationships are like those
described by Bentham in that people act in terms of self-interest. The only
force which brings people together is the private interest each of them has
in the exchange, in obtaining what the other person possesses. Each
34 The Anatomy of Capitalist Societies

person thus pays attention to their own interests and not to those of the
other. Each serves the other only to serve themselves. The exchange
of exchange-values is the material basis of individual equality and fre-
edom. The freedom and equality of separate individuals stem from
these exchange relations. They entail voluntary transactions, that no
force is applied on either side and that individuals equally seek to
realise their own interests. It is a 'noisy sphere'll of individual exchange
from which arise certain freedoms of civil society and the state - in
particular the law which protects the freedom of individual juridic sub-
jects equally to make contracts, to engage in exchange, to sue and be
sued.
So far then we have seen that there is a sphere of circulation, of
individual exchange based on freedom and eqUality. I will now consider
the basis of this sphere in more detail by, first, analysing the sphere of
production. What we must do is to consider the consumption of
labour-power, which, like the consumption of every other commodity,
takes place outside exchange relations. We must 'leave this noisy sphere,
where everything takes place on the surface and in full view of everyone,
and follow them into the hidden abode of production' .12 Once we do that
we leave behind the formally free and equal relations of individual
exchange. He who was previously merely the money-owner now
becomes the capitalist; he who was merely the possessor oflabour-power
now follows as his worker:

The one smirks self-importantly and is intent on business; the other

is timid and holds back, like someone who has brought his own
hide to market and now has nothing else to expect but - a tan-
ningY

I will now consider how to theorise the relationship between the

spheres of circulation and production. Marx himself was not always clear
about this. Thus, right at the end of the third volume of Capital, Marx
says of the relationship between production and distribution (the latter
being one moment of the circulation process):

These definite forms of distribution thus presuppose definite social

characteristics of production conditions, and definite relations of
production agents. The specific distribution relations are thus merely
the expression of the specific historical production relations .... The
so-called distribution relations, then, correspond to and arise from
 The Spheres of Production and Circulation 35

historically determined specific social forms of the process of

production. 14

In one sense this is true but in another false. Clearly it is right to see the
overall patterning of distribution, in particular of payments to capital and
payments to wage-labour, as historically dependent upon the overall
social relations of capitalist production. However, it is evident that the
details of such distribution do not depend directly upon the overall
character of capitalist relations of production. For example, both
piece-rate and time-rate systems of wage-payment are compatible with
the same production relations. Also the patterns of consumption, how the
recipients of the same form of distributional payments allocate their
income to reproduce their labour-power, varies greatly and cannot be
deduced from the 'historical production relations'. In the Introduction to
the Grundrisse Marx tries to make the nature of these relationships
clearer - let us see how.
First of all, he considers and criticises two alternative accounts
provided of the relations between production, distribution, exchange and
consumption. IS First, there is the 'most shallow' account in which each of
these four elements are subject to their own general laws. Each is distinct
and they are seen as merely entering into mechanical, external relation-
ships with each other. There is no recognition that they are complexly
interconnected such that, for example, production relations imply certain
consumption relations and vice versa. Second, there is an Hegelian-
inspired approach in which there is criticism of the lack of unity implied
in the former approach. In this Hegelian formulation a conceptual identity
is posited between the four elements: of production, distribution,
exchange and consumption. But Marx criticises this because it ignores
both the particular character of production and the distinctive and unequal
forms of interdependence between the four elements.
Marx's own argument is based on a more complex conception of
production. Certain aspects of distribution, exchange and consumption
are properly part of production, what he calls 'factors' or 'moments' of
production and these cannot be seen as determined by production. Within
the factors of production there is a narrowly defined 'mutual interac-
tion' .16 Then there are other aspects of distribution, exchange and
consumption which are not properly part of production, which lie outside
it conceived in this wide sense, and which are determined by it in two
ways. First, production determines the general form that distribution,
exchange and consumption take; second, production determines the
36 The Anatomy of Capitalist Societies

relationships that each of them bears to one another and to production

itself, since the process always returns to production to begin again. They
are all members of a totality, 'distinctions within a unity',t' in which
aspects of distribution, exchange and consumption react back in a
determined form upon production. For example, if the sphere of
exchange expands and production grows in quantity, then the division
between the branches of production becomes deeper; so the element of
exchange has affected the element of production.
Let us now consider examples of those aspects which are part of
production in the wide sense and those which are not. If we consider the
sphere of distributional relations we can see that the distribution of people
among the means of production is an aspect of the relations of production,
while the distribution of commodities between the same people is an
aspect of distribution. Within exchange there is that which occurs
between production units (Le. within the sphere of production) such that
commodities can be made which are suitable for consumption, and there
is that which occurs directly for consumption by the final consumer
(departments 1 and 2 respectively). And finally within consumption itself
there is the specific aspect in which the individual worker reappears
within production as refreshed and energetic labour-power, and there is
the more general aspect in which individuals consume the entire mass of
commodities as a result oftheir differing use-values. So in each case we
have seen that the elements of distribution, exchange and consumption
can be divided into two, into those aspects which are essentially part of
production, and those which are not but which will be determined in part
by production relations. Such production relations also determine the
relationships existent between the elements of distribution, exchange,
consumption and production itself.
It is thus clear that, for Marx, production relations are central to the
capitalist economy. I will now make a few brief comments about
production before considering the degree to which especially class
relations, incl uding distributional relations, are to be explained in terms
of the social relations of capitalist production.
First, capitalist relations of production are inherently social. Capital is
thus not a thing, it is not the sum of all the material means of production
which exist. Capital only refers to those means of production which have
been transformed into a definite social production relation, belonging to a
particular historical stage of society. It is where the means of production
have been monopolised by a certain class in society which then confronts
another class; namely, those who do not possess the means of production
 The Spheres of Production and Circulation 37

and have only their labour-power to sell for money. Thus, material
objects only function as capital if they are inserted within a particular set
of social relations of production, namely those between capital and
wage-labour.
Second, this means that the typical formula of analysis employed by
classical economics is incorrect. In the 'trinity formula' it is held that
there are three factors of production - capital, land and labour - and that
each of them creates value. But Marx says that these are widely divergent
forms of revenue. (He says that they have the same relation to each other
as 'lawyer's fees, red beets and music' .18) Capital is, as we have seen, an
element of production specific to the capitalist mode, while land and
labour are both material elements of all modes of production. Further-
more, although the material characteristics of land are of importance,
land cannot be transformed into value without the exercise of human
labour. Labour indeed is the source of all value and therefore its
relationship with land or with capital is of primary importance. Indeed the
trinity formula fails to grasp the social character of production relations;
that land or capital only yields a monetary return precisely because ofthe
human labour brought to bear upon it.
Third, then, the dominant social relations within capitalism are the
relations between capital and wage-labour, relations which determine the
significance ofland, merchant-and interest-bearing capital. The relations
between capital and wage-labour are of real opposition, of a contradiction
which lies at the very heart of the CMP. There are thus two functions, one
of capital, the other of wage-labour. The functionaries of each cannot
avoid the relationship - they are each forced to act in certain ways, to
personify capital or to personify wage-labour. The industrial capitalist is
both the exploiter of wage-labour but also responsible for organising the
creation of surplus-value. 19 The capitalist, like the wage-labourer, is a
functionary of the process - a functionary essential to capitalist produc-
tion. Accumulation, then, must of necessity involve class struggle. It is
not a neutral process but one necessitating a real opposition of interest.
The landowner, by contrast, is inessential, superfluous to the CMP,
although landownership was necessary for its historical development. 2o
Capital is thus a social relation - but also, and this is the fourth
point here, it takes a number of forms - a money - a commodity-
and a productive - form. Unlike the simple process of exchange of
C -M -C, when a commodity is exchanged for money which is
then used to buy another commodity, capital involves the exchange
M-c:::f;; --P--CI--MI. Money is advanced as capital (C)
38 The Anatomy of Capitalist Societies

in order to expand it. With M the capitalist buys means of production

(MP) and labour-power (LP) - these are combined and set to work by the
capitalist. This is the real abode of production where surplus-value is
produced. The consequence of such productive activity is that com-
modities are produced (Cl) - these, when sold on the market, yield
increased money to the capitalist (M 1). The process begins and ends
within the sphere of circulation. Each of the forms of capital pres upposes
the other forms and each one only functions as capital when they follow
these functions in the circuit of money-capital. Incidentally, this circuit of
capital can also be represented as a circuit of productive-capital
(beginning withP) and as a circuit of commodity-capital (beginning with
C). It is only when these circuits have been analysed that we can theorise
the other two forms of capital, forms that in fact developed historically
prior to that of industrial capital. Merchant capital consists of the circuit
M-C-Ml within the sphere of circulation; while bank capital consists
of M-Ml. In neither case is production (P) involved in the circuit.
Fifth, capital always consists of separate capital units, the relations
between them being unplanned, anarchic, antagonistic and regulated by
the laws of competition. Capital is essentially competitive - each unit
has to struggle to minimise its costs of production and to maximise its
mass of profits. If not, it will be merged with other capitals, or disappear
altogether as capital. This competitive structure applies, whether the
average capital unit is small or large, local, national or transnational.
There are three important consequences. First, it is always difficult to
establish the political unity of capital, not only because of the different
circuits (industrial, commercial, and banking) and different degrees of
centralisation (small, medium and large), but also because of this
irreducible competitive structure. Second, there is always intense
competition in the market(s) for labour-power, so that, except for that
which is in short supply (for whatever reason), there will be a tendency
not to pay above the market-rate. And third, the effects of these
competitive relations is to produce a set of immanent laws of capitalist
development, laws which result from the attempt by each capital unit to
maximise its mass of profits, from the particular competitive situation
that it faces. I shall not discuss here the law of the tendency for the rate of
profit to fall, and of the mobilisation of the counter-tendencies. I shall
simply take it that capitalist accumulation assumes certain law-like
characteristics, and that it is an irreducibly crisis-ridden process.
So far then I have made some brief points about the nature of capitalist
production. I now want to consider in more detail the relationship
 The Spheres of Production and Circulation 39

between this sphere of production and that of circulation. In particular, I

will discuss those aspects of distribution, exchange and consumption
which lie outside the sphere of production. I will consider what
relationship they bear to production and to each other. More precisely I
shall considl!r the degree to which the social relations of production
provide us with a theory of distribution and an explanation of social class.
In the following I shall not say much about the elements of exchange
and consumption. I take it that significant aspects of both of these
elements are not directly determined by production. The conditions under
which exchange takes place are in part determined by the state, as we
shall see in Chapters 6 and 7. The patterning of consumption - what
kinds of commodities different classes or other social groupings buy - is
largely determined outside production itself, within the sphere of civil
society. However, that the conditions of appropriate exchange should be
state-determined, and that consumption patterns should result from the
processes of civil society, are both consequences of the social relations of
production, in particular, ofthe fact that such production presupposes its
differentiation from the sphere of circulation. Distributional relations,
though, are more complex and raise difficult issues concerned with
Marx's theory of wages. 21
The problem is that labour-power, unlike all other commodities, is not
produced by capitalists for profit. It is produced, or reproduced, in the
sphere of civil society. So while all other commodites are produced
within capitalist production, labour-power is produced elsewhere,
outside capitalist relations of production. If so, why should the labour
theory of value apply? Why is it not possible for workers to push up their
wages perhaps to the extent that surplus-value is eliminated completely?
Why should labour-power be remunerated at a value equal to the costs of
its own (re)production? Marx's answers to these questions are problem-
atic. In general, he maintained that the existence of an industrial reserve
army of the unemployed would prevent real wages from rising above the
cost of reproducing labour-power. This is a form of unemployment not
resulting from deficiencies of effective demand - rather it is that
unemployment which exists when all capital is fully manned (or
personned). Unfortunately, though, Marx does not demonstrate that there
will always be a reserve army sufficiently large to prevent real wages
rising above the costs of reproduction. 22 This is because he presents no
general explanation of the rate of population growth, or indeed of net
migration flows, nor does he show why technical change will be
necessarily labour-saving. Indeed it would be reasonable to expect that
40 The Anatomy of Capitalist Societies

with low wages, technical change would be significantly capital-saving

rather than the opposite. Further, Marx expected that the expansion of
capitalist industry into the precapitalist sectors would lead to massive
reductions in the labour employed per unit of output. This has occurred
but is a historically limited process as economic activity has become
increasingl y capitalistic.
Two other problems may be mentioned in his theory of wages. First,
there is a need for a theory of wage differentials. What determines the fact
that a craftsman may earn two or three times what a semi-skilled operative
may earn? Marx's answer is that this depends on the time and labour
involved in the training of the skilled worker - in other words, that their
costs of (re)production are higher and that this is stored up within the
labourer and is then manifest when slhe actually begins work. 23 Qearly
this would not account for the wages of those employed where an
untrainable natural ability is involved - but these could be seen as
earning a form of rent. It is also a matter of controversy as to whether such
wage differentials are entirely independent of demand considerations. It
might be claimed that wages in certain occupations are higher because of
the disutility of the job. Workers therefore have to be especially
compensated in order to induce them to take on generally undesirable
labour. But this point seems relatively less important since it would only
apply were the industrial reserve army to be too small. In other words,
were Marx to have produced an adequate theory of the industrial reserve
army then that objection would not apply. A further difficulty rests on the
unclear role which Marx attributes to the class struggle, and especially on
the apparent ability of certain groups of workers to increase their real
income through various methods of collective bargaining, especially that
of enforcing a closed shop, or via long periods of apprenticeship. Within
certain occupations there is no nationally applicable labour market and
thus, whatever the state of the industrial reserve army overall, within
particular occupations there is no surplus oflabour to ensure that workers
are paid merely the costs of reproducing even their highly trained
labour-power.
The second problem relates to this last point. It is clear that over the last
century or so there has been a long-term increase in the level of real
wages - in the United Kingdom they are now perhaps four times as high
as in the 1890s. 24 It is clear that in the process of remunerating workers
there is an 'historical and moral element' in assessing the 'necessary
means of subsistence' involved in reproducing average labour-power. 25
Thus the laws of determination of wages are highly flexible and Marx's
 The Spheres of Production and Circulation 41

argument is imprecise. How does one determine what are subsistence

needs if one incl udes a 'historical and moral element '? Is it in fact
possible to devise such needs apart from those which are purely
physiological?
Neo-Ricardians have generally argued that trade-union pressure, the
organised power of the working class, has been responsible for shifting
real wages upwards. 26 The gains oflabour have been achieved becuse of
their bargaining effectiveness vis-a-vis capital. Such writers have seen in
the following quotation from Wages, Price and Profit support for their
contention:

although we can fix the minimum of wages, we cannot fix the

maximum .... The fixation of its actual degrees is only settled by the
continuous struggle between capital and labour, the capitalist con-
stantly tending to reduce wages to their physical minimum, and to
extend the working day to its physical maximum, while the worker
constantly presses in the opposite direction.
The matter constantly resolves it into a question of the respective
powers of the combatants. 27

Likewise Marx argues:

As to the limitation of the working day in England, as in all other

countries, it has never been settled except by legislative interference.
Without the working men's continuous pressure from without, that
interference would never have taken place. 28

It would therefore seem that at least within limits the bargaining

relationship between capital and labour does have significant effects both
at the level of real wages and of state action. The problem in Marx's
formulations, however, is the following. For all his apparent attention
lavished on classes and class struggle, he does not theorise exactly how
and why such struggles may effect distributional or state policy
outcomes. Their effect is indeterminate and that is because Marx fails to
theorise classes as opemting within civil society. Indeed for him classes
are nowhere or everywhere - they may have no effect on the outcome, or
they may determine it, but in either case this is seen as consistent with the
claims of his theory. Consider his distributional theory briefly here. On
the one hand, wages are principally determined by the subsistence needs
of the unskilled worker, although such subsistence is given an 'historical
42 The Anatomy of Capitalist Societies

and "moral" 'element. The role of the industrial reserve army is central
on this account in minimising wage increases. On the other hand, wages
are determined by the bargaining or class struggles between labour and
capital - in this case there would seem to be no upper limit to the relative
gains of labour. In the first case the class struggle plays no role; in the
second it plays the primary role.
The correct position is one in which adequate theorisation is provided
of how such class struggles may affect the patterning of distribution. As I
will argue in the next chapters, classes exist, operate and initially struggle
within civil society. For the moment we should note two determinants of
distribution derived from the social relations of production. First, there is
a strong uppermost limit on the level of real wages and that is supplied by
the movements of the rate of profit within the economy. Capitalist
production is production for profit - without profits being earned
capitalists will not advance M with the expectation of realising MI.
Second, the size and spatial allocation of the industrial reserve army
places major limitations on the effectiveness of such struggles to improve
real wages. Marx generally talked of the industrial reserve army in total
but it is clear in fact that its spatial distribution is of equal importance. In
other words, as a result of changes in production methods, resulting from
changes in concentration and centralisation, major and crucial changes
will occur in the organisation oflocallabour markets; for example, in the
relationship between primary and secondary markets. However, it should
be noted that certain determinants of the size and distribution of the
industrial reserve army lie outside production and within civil society
(e.g. attitudes to family size, the degree of immobility of labour).
In the next chapters I shall discuss classes and their interrelationships in
more detail, after I have analysed the concept of ideology. For the present
let me make some brief assertions which will serve to conclude this
chapter, but also more importantly will serve as a set of background
premises for Chapters 4 and 5.

1. Each individual seller of labour-power has to struggle in order to

sustain and to improve the conditions under which their labour-power is
sold. These forms of struggle are highly diverse, including absenteeism,
slow-downs, sabotage, collective industrial action, etc. This presump-
tion, that the sellers oflabour-power have to struggle, is the only essential
characteristic of labour. Thus, unless specific factors intrude, such as
extra payments out of revenue, or legal or customary preventions against
employment of specific means of struggle, then such struggles will ensue.
 The Spheres of Production and Circulation 43

2. Where it is possible, such struggles oflabour will take a collective

form. Again unless specific factors intrude, the seller of labour-power
will seek to struggle with other sellers, who are selling the same kind of
labour-power, or who are on approximately the same wage-level, or who
work in the same workshop, unit or factory. Such struggles will
necessarily take a sectoral form, and will be crucially structured by the
organisation of the relevant market( s) for that category oflabour-power.
Many such struggles will involve conflict, not only with capital, but with
the sellers of other categories of labour-power. Struggles within the
sphere of circulation are central to the capitalist economy.
3. Such struggles can ha ve significant effects, in increasing real wages
and or the share of wages in relation to profits. This depends on the
strength of the relative labour movement, which depends in part on the
circumstances in which capitalism was initially established.
4. Yet the possible gains to be made are limited, and depend on the
exploitation of specific conditions. Capitalists will seek to undermine
such conditions, in particular, through raising the mobility of capital. The
major division within capitalist society is that of capital and wage-labour.
It is this which structures and delimits the forms of struggle between
specific capitalists and workers.
5. The importance of conflict within the sphere of circulation means
that the characteristic form of struggle in capitalist society is an
economistic one. Further, the bases of such struggles are diverse, not
only involving categories of worker, but other categories brought
together within circulation; namely, consumers, house-purchasers,
old-age pensioners, car-owners, etc. The importance of such economistic
struggles can never be overcome. So that, although categories of worker
may know that they are exploited, it is only in quite exceptional
circumstances that the entire system will be put in jeopardy. Whatever
kinds of consciousness exist, all sellers of labour-power have to continue
to extract as much as they can, through whatever means are available,
from the sphere of circulation.
4
The Critique of Ideology

In this chapter I shall briefly and schematically consider some issues in

the Marxist theory of ideology. I shall suggest that there are three main
theories that have been developed, but that each of them suffers from
certain deficiencies. Thus, it is necessary to dispense with the concept of
ideology per se and instead to analyse ideological effects, these being
viewed in a non-functionalist manner. Much of what is conventionally
viewed as ideology is properly to be conceptualised as civil society. I
shall discuss various aspects of this notion, especially the analyses of
classes, of popular-democratic struggles and ofthe sphere of reproduc-
tion.
First, then, we can see that there is no coherent theory of ideology in
Marx's writings.! I think, however, that this is, as contrasted with the
state, no mere coincidence. It stems, I suggest, from his actual approach
to the subject. He does not adopt a coherent analysis of the concept of
ideology precisely because ideology does not exhibit such coherence. He
analyses a variety of ideological effects, because that is how ideology is
to be viewed, as effects which exhibit no particulr coherence or unity.
There is no essence to ideology, given either in some characteristic
residing in each individual human subject, or in a function which 'it'
performs in all societies, or in all examples of a particular kind of society.
Thus, the attempt by Marxists to locate ideology as one element, structure
or instance on a par with the economy and the state connot be successful.
This is as true of older notions of base and superstructure (including the
'ideological superstructure ') as it is of more modern notions of the
economic, political and ideological instances.
It is clear nevertheless that the term 'ideology' indicates a realm of
questions that need investigation. It is suggestive of a set of issues or
problems, as Althusser would say: 'dig here'. I shall suggest by the end of
 The Critique of Ideology 45

the next chapter the following distinctions. First, much of what is

characterised as ideology is simply no more than social practice, either of
a class, or of other social forces. Such practices are not necessarily unified
and bear no necessary relation to each other. The ideas embodied within
such practices may be relatively formulated, so constituting a philosophy
for that social grouping, or relatively unformulated, so remaining at the
level of common sense. Such ideas may be specific to that social force, or
more general. These are all nevertheless matters for substantive investi-
gation. Second, the state will seek to organise and mobilise these
practices into a national framework, especially through manipulating
popular sentiment into a coherent and comprehensible hegemonic
structure. The state will never fully succeed in this, except possibly and
paradigmatically at times of war mobilisation. In all other situations such
mobilisation creates a counter-mobilisation, the organisation of resis-
tance through existing or new forms of social practice. Third, it is only
certain of these practices that are to be viewed as ideological; that is,
when two conditions hold: (i) that embodied within such a practice is a
concealment of the causes, nature and consequences of that practice or of
some other practice; and (ii) that this concealment is in the interests of one
or more of the dominant social forces within that society, (for example,
capitalists, state bureaucrats, men, whites, etc.).2 However, Ido not hold
that the explanation of such an ideological effect rests with the function
that it performs either for that dominant social force, or for the overall
society. Part of the reason for this is that it is not at all clear just what
contribution is made by a particular ideological effect. Thus, if we
consider what would have been the case, were that effect not to have
occurred, then in general we would find that the society would still have
been reproduced. If this is the case then it is problematic to argue that
meeting such a functional requirement of ensuring social reproduction
actually explains the practice in question.
There are, I believe, three related theories of ideology in Marx. I shall
now summarise each, the first two briefly because they are fairly familiar.
The first can be termed the 'class-theory', or more generally the 'social
being that determines their consciousness' theory. 3 It was commonly
argued, prior to Hegel and the Hegelians, that concepts and theories, or
more generally 'ideas', are logically and temporally independent of the
societies in which they have been developed. In the Hegelian revolution it
was asserted that this was not the case, that ideas are themselves historical
and constitute the fundamental basis upon which societies depend. Ideas
are thus the motor of history. Marx, as is well known, argued rather that
46 The Anatomy of Capitalist Societies

the starting-point of historical analysis is with real individuals, their

actions and their material conditions oflife, with the interrelations of man
and man (sic) and man and nature. 4 Humans, as opposed to animals,
produce their own means of subsistence, and the variations in the forms of
such production give rise to distinctively different types of society,
different states, forms of politics and systems of belief. Consciousness,
therefore, is produced out of different material conditions - it is not some
disembodied entity which gives rise to different forms of life. It does not
determine life, but is determined by it. Thus, we may speak only of
different, living individuals as they actually exist, and of consciousness
as their consciousness. So we consider not the abstract notion of 'man',
but of real, living men, conditioned by the development ofthe forces and
relations of production , these giving rise to different forms of conscious-
ness, different concepts, ideas and beliefs.
Moreover, it is among such forms of consciousness, to which the real
conditions give rise, that there are the illusory notions of traditional and
Hegelian philosophy, those conceptions which presuppose a separate or
determining spirit. So the upsidedownness of ideology, seeing the real
world as the product of the idea, is itself a product of the real relations, of
the social nature of material life. It is the 'historical life-process ' which
causes men and their circumstances to appear upside down, as in a camera
obscura. 5 Feuerbach, as a Left Hegelian, is rejected on three main
grounds. First, he only discusses 'Man' and the relationship with
'Nature', and fails to consider the interactions between real, living
men - the ensemble of social relations. Second, Feuerbach conceives of
man's relationship to the world as one essentially involving perception
and sensation, seeing man as a 'sensuous object'. Marx argues rather for
the analysis of 'sensuous activity' ,6 for the concrete interactions with
other men and nature. And third, Feuerbach fails to see how the
transcendence of alienation does not merely involve critical speculation
and understanding, but rather practical concrete activity, of transforming
the very nature of social interactions. What thus moves history is not the
contradiction between 'man' and his alienated essence, but rather specific
social practices, a revolutionary practice with definite organisational
forms, strategy, tactics and personnel.
Gasses are obviously central to Marx's attempt, not merely to
understand the world, but to transform it. However, there are two
different metaphors he employs with respect to the relationship between
class and ideology, metaphors which are often assumed to be equivalent.
On the one hand, Marx says that it is social being that determines social
 The Critique of Ideology 47

consciousness, in other words, that each class through its relationship to

the means of the production and its general conditions of existence
develops a particular class culture or ideology. Different classes have
distinctive 'social beings', different material circumstances and hence
different interests. They thus develop separate ideologies relating to the
specific interests that they articulate. On the other hand, Marx says that
the economic base produces a specific superstructure, a system of state
and legal apparatuses, and forms of ideology that hold the society
together. The 'real foundation' gives rise to a 'legal and political
superstructure', and 'definite forms of social consciousness'.7 If we
consider these forms of social consciousness we find that the ideas of 'the
ruling class are in every epoch the ruling ideas'; thus 'the class which is
the ruling material force of society, is at the same time its ruling
intellectual force'. 8 Ideology on this scheme is viewed negatively, it
functions to limit or frustrate the mobilisation of class practices; while on
the first model it is viewed positively as the expression of class interests.
Thus, the two metaphors 'being-consciousness', 'base-superstructure'
are not equivalent to each other and give rise to differing conceptions of
ideology. But also both positions are incorrect since they presuppose that
there are unified entities of 'ideology'. Thus on the second metaphor,
'base-superstructure' , it is wrong to think that there is a single ideological
form which establishes the unity of a given social formation. There are
ideological effects of certain forms of practice within the economy, the
state and especially civil society. Many ofthese effects derive from forms
of practice which are not directly determined by capitalist relations,
which are only indirectly dependent upon such relations (e.g. the
ideology of the family being dependent upon the existence of domestic
labour, separated from productive labour). In terms of the first metaphor
'being-consciousness', so-called class ideologies are better viewed not
directly as ideologies at all. There are merely forms of class practice, of
interest, ritual, know-how, symbols, 'illusions, modes of thought and
views of life ',9 which do not necessarily form a unity. Each class practice
mayor may not overlap with that of other classes. There mayor may not
be relations of domination between different class practices. In all cases
class practices are merely one form of practice within capitalist society.
And, finally, there is no reason to assume either a unity of each class
practice, nor that ideologies are nameplates carried around on the backs
of particular social classes.
The other two theories, of 'fetishism' and of the 'subject', do not
presuppose that ideologies are class-related. In the former, the two main
48 The Anatomy of Capitalist Societies

issues are, first, just what is it that constitutes a 'fetishism', what is being
fetishised; and second, how generally applicable is this process, does it
account for all ideological notions in a capitalist society?10 Briefly, then,
Marx's theory of commodity fetishism can be stated as follows. First, in
capitalism, as a commodity-producing society, individuals are linked to
each other not directly, but through things. It is relations between things
which is the form taken by the relations between people. In particular, it is
the sphere of circulation that enables both the transfer and the equalisation
of the products of people's labour. Private labour becomes social as a
result of such exchange. This equalisation takes place through the
quantities in which the products of different kinds of labour exchange;
and this ensures the exchange of different quantities of labour-time
embodied within different commodities. Labour is thus not directly
exchanged, an equivalence between labour-times is not directly estab-
lished. The equivalance of exchange between the products of different
producers is achieved indirectly through the exchange of commodities, or
things. Second, once these relations between things are established they
are coercive. Individuals cannot avoid submitting to the social process
which is the consequence of the mass of individual transactions between
producers. The relations between individuals are determined by the
law-like relations between things. They are experienced as relations of
necessity. Third, the fetish consists of (a) failing to see that such
determining relations between things are in effect the exchange of
different quantities of labour-time; and hence (b) improperly attributing
natural properties to material objects making a fetish of their material
form. For example, it is fetishistic to see profit as arising out of the natural
productivity of the material objects that constitute capital. Profit in fact
arises out of the social relations of production, as a consequence of
which, the exchange relations between things (cotton yam, loom, cloth,
etc.) are established and dominate the behaviour of separate individuals.
They are experienced as relations of necessity.
I don't intend here to consider the various criticisms that have been
made ofthese arguments. The only issue I shall consider is whether this
could constitute a general theory of ideology. This is indeed what some
writers have recently suggested. McDonnell argues, for example, given
that all social relations have become developed forms of the capital
relation and moments in the total process of production, it is essential to
found a theory of ideology on that of capital and especially on the notion
of commodity fetishism. 11 Apart from certain individuals who will not be
taken in by such fetishistic forms, and apart from where ideology is
 The Critique of Ideology 49

deliberately manipulated for strategic reasons, then the concept of

fetishism is the basis for the theory of ideology. However, he understands
this very generally. Thus, any of the divisions within society are seen as a
fetishism, they come to be explained in terms ofthe natural characteris-
tics of the things themselves, and not in terms of the underlying social
relations. He employs this notion of ideology to consider some of the
debates over the public expenditure cuts. But fetishism progressively
disappears from his account, ideology remains, continuing to be
understood in a reductionist manner. He is unable to sustain the claim that
all of ideology, even on his account, can be reduced to fetishism.
In brief, then, I have the following objections to McDonnell's
argument. First, ideology is interpreted in a reductionist manner, it is
simply reflective of the capital relation and of the crises ofaccumulation.
How are political ideologies derived from the processes offetishisation?
Second, the place of ideology within the social formation is untheorised,
so that the degree of 'ideological legitimation' required remains
unspecified. Third, McDonnell does not demonstrate what effects follow
some particular ideological shift - how exactly does it sustain the British
social formation? Fourth, it just cannot be claimed that fetishism is the
basis of all ideology, unless (and this is unclear) ideology is defined in
terms of fetishism. But if he does not, then McDonnell is not justified in
trying to account for all kinds of different positions over the cuts in terms
of fetishistic relations. Fifth, it is not nearly so clear (in contemporary
capitalism) that commodity fetishism is particularly important. Because
of the concentration and centralisation of capital, the growth of the state
and the organisation of dominated social forces (especially trade unions),
itis much less likely, either that material objects will be seen as natural, or
that the social bases of the relations between things will be ignored. Thus,
it is less probable that payments to labour will be fetishised, as seen as
stemming purely from their natural characteristics and unrelated to the
social conditions under which they exchange for capital. 12 So even if the
theory of commodity fetishism was once generally applicable to capitalist
ideology, it is progressively less relevant to advanced capitalist society,
certainly in analysing anything as specific as different ideological
responses to public expenditure cuts.
I shall now consider an approach in which there is rejection of the basic
premises of the fetishism theory. I shall consider the 'subject' theory
through Althusser's seminal essay on 'Ideology and Ideological State
Apparatuses ',13 partly because of Althusser's explicit rejection elsewhere
of the concept 'civil society', and partly because it is one of the few recent
50 The Anatomy of Capitalist Societies

attempts to theorise ideology in a remotely rigorous fashion. He argues:

(1) There has to be reproduction of the productive forces and the

relations of production, in order for capitalist production to continue.
(2) This reproduction is secured in large measure by ideology; for
example, in the teaching in schools not just of appropriate skills but
of the submission of labour-power to the rules of the established
order.
(3) This ideology is to be seen as part of the state. Besides the repressive
state apparatus, which functions by violence, there are a number of
ideological state apparatuses (religious, educational, family, legal,
media, etc.) which function primarily by ideology. Such ISAs
(ideological state apparatuses) exhibit unity.
(4) Religion was the dominant ISA in precapitalist societies. In modem
capitalist societies it is education which plays the determinant role in
reproducing capitalist relations of production.
(5) What is represented in this and in all other ideology is the imaginary
relation of those individuals to the real relations in which they live.
Such an ideology consists not merely of sets of 'ideas' but is endowed
with a material existence, of actions, practices and rituals. Further,
there is no practice except by and in an ideology.
(6) Ideology has the function of 'constituting' concrete individuals as
subjects. It does this by 'hailing' or interpellating concrete individu-
als. Thus, ideology is to be seen as constituting subjects, not in
making them incorrectly perceive an external reality.
(7) The individuals interpellated as subjects by ideology are both free
centres of initiative, authors of and responsible for their actions, and
subjected to higher authority, to God, or to the law, or to their boss,
or whatever. There are no subjects except by and for their subjection,
as freely accepting the commands of the other.
(8) Such ideology in general takes a specific form given the particular
nature of the class struggle, of the forms of exploitation and the need
to ensure their reproduction.

Let me consider some of the deficiencies in his account. First, then, his
argument is functionalist. In effect he is saying that anything which lies
outside the forces and relations of production, and outside the repressive
aspects of the state, is 'ideological' and that this has a specific function,
namely, to reproduce those forces and relations. Thus, the explanation of
any of these apparatuses in a specific social formation (e.g. the family in
 The Critique of Ideology 51

Britain, the Church in France, etc.) lies with the functions it performs for
reproducing the social relations of capitalist production within that
formation. Althusser himself, in an appendix to this text, acknowledges
this is too general and adds to the analysis the specific conditions of the
class struggle. But this raises two problems: first, that an attempt is made
to solve 'abstract' problems with the analysis of the 'concrete' as though
these can be added together unproblematically; and second, that the class
struggle is treated merely as the conflict at the relations of production
between capitalists and workers. But the class struggles and indeed the
struggles of other popular forces are greatly more complex than this,
involving numbers of different classes, fractions, strata and popular
forces. Ideology therefore should not be reduced to merely that which
functions to reproduce the class relations of capitalist and worker.
Furthermore, as Hirst shows, Althusser equates the relations of
production with the distribution of agents to places within the social
division of labour .14 This means that no explanation is given of how the
forms of relation between agents, such as the circuits of capital, are
reproduced. The reproduction of agents is not same as the reproduction of
the social relations between agents. Moreover, certain of the occu-
pational distinctions within the labour force, of managerial/non-
managerial, manual/non-manual and skilled/unskilled, do not corre-
spond with the relations of production. Thus, it cannot be argued that the
former categories, which are clearly linked with the provision of differing
periods and forms of education, are derived from the latter, the relations
of production. It is difficult to know how and why the social relations of
production (of capital and its fractions, wage-labour, petty-commodity
production) need for their reproduction the precise differentiation of the
labour force and of educational provision into the various occupational
divisions.
Ideology cannot therefore be viewed as simply reproducing the
dominant relations of production. This is also because, underlying
Althusser's argument here, is the notion of structural causality, that each
mode of production secures its own conditions of existence and is
therefore capable of its eternal reproduction. IS However, it is possible to
reject the concept of structural causality without concluding that
dominant relations of production do not secure any oftheir conditions of
existence. Where capitalism is the dominant mode, then, as we have
seen, there are separate spheres of production and circulation. The latter,
as we shall see, is the basis for the interpellation of individual subjects.
But more generally, the existence of an individualised sphere is the basis
52 The Anatomy of Capitalist Societies

of civil society, for sets of relations between agents outside the economy
and outside the state. Such relations cannot be said to be functionally
necessary for capitalist relations, in the sense that they, and only they, are
appropriate to sustain particular capitalist relations. But given the social
relations of production, and given a possible hegemony within the state,
then they represent the particular relations which may enable capitalist
production to take place. There is, however, nothing necessary about the
relations of civil society. Even if capitalist production occurs, even if the
state is organised successfully by a hegemonic power bloc, it still does not
follow that civil society is so structured that it is most functionally
appropriate for the existing social relations of capitalist production. It
may be, but this is a contingent matter which cannot be inscribed into the
concept of civil society, as Althusser tries to inscribe this into the concept
of ideology. Civil society embraces complex sets of social relations,
aspects of which are directly structured by the dominant relations of
production (for example, the division between domestic and productive
labour), while other aspects are not (for example, different patterns of
child-rearing).
Althusser, however, rejects any distinction between the private and the
public; hence; what I take to be the private institutions of civil society (the
famil y, trade unions, political parties, etc.) are for him the ideological
apparatuses of the state. The reason why the private/public distinction has
no effectivity is because it is internal to bourgeois law and only valid in
the relatively subordinate domains in which bourgeois law exercises its
'authority'. The state escapes its effectivity because it is above the law,
indeed the state is the very condition of the distinction between the public
and private. That distinction cannot thus be the basis for the definition of
the state and of its separation from ideology. So, for Althusser, what
matters for the various apparatuses is not whether they are formally
'public' or 'private' , but how they function. The ISAs function primarily
by ideology, although secondarily by repression (e.g. punishment in
schools). The repressive state apparatuses by contrast, function princi-
pally by repression, secondarily by ideology. Now it might be wondered
why the function of ideology should make certain of these apparatuses
part of the state - indeed just what is the basis of the unity of the ISAs.
The answer is that all the ISAs function 'beneath the ruling ideology'.16
Indeed all the state apparatuses are ultimately unified by the repressive
and ideological dominance of the ruling class. There are a number of
objections to these claims.
A first set of problems in Althusser's conception of ISAs lies with his
 The Critique of Ideology 53

economlStlc conception of social class. For him class is defined

economically and he adopts the base-superstructure topography of the
1859 Preface. Given the definition of class economically, then ideologi-
cal and political relations exist merely to allow or to frustrate the full
development of that class ideologically and/or politically, but it is the full
development of relations fundamentally determined at the level of the
economic base. The unity ofthe IS As stems from the unity ofrul ing-cl ass
ideology, which in turn stems from the unity of the ruling class,
determined economically. But this is problematic because, although
certain classes are principally determined at the level of the relations of
production, they only exist within the sphere of civil society. Thus, if we
consider the means of their representation, these means are not merely
facilitators/preventors of pre-existent relations. Rather the relations of the
particular civil society are the very means by which class relations are
expressed. Such class relations can never exist in a pure form - they
always exist within the specific conditions or means of representation
given by a particular civil society; thus, by the particular sets of
distributional, familial, geographical, racial, gender and generational
relations comprising that civil society.
Second, although Althusser equivocates as to whether 'the ruling
class' is a single class or a number of classes, he has in fact to assume that
it is a single class. This is because there is no other way in which he can
demonstrate the unity of the ISAs. But even on his economistic
conception there would in fact be a plurality of dominant social classes
and fractions (of ground-rent landlords, industrial, banking, commercial,
comprador capitalists, new middle classes, traditional petty bourgeoisie).
Thus, it is impossible to conceive of the ruling class as a single class, and
to conceive of the ruling ideology as a unity. This means that we have to
reject Althusser's conception of the state as the instrument of the unified
rul ing class. Rather, the state is to be seen as that place where a ruling bloc
(what Poulantzas calls a 'power bloc ') is cemented out of the differing
and in part conflicting classes and fractions. Let us consider just one quote
where Althusser's instrumentalist conception of the state is formulated:
'To my knowledge, no class can hold State power over a long period
without at the same time exercising its hegemony over and in the State
Ideological Apparatuses. '17 Yet in point offact, 'to my knowledge', it is
possible for certain class(es) to be dominant ideologically, while not
exercising dominance in strict political terms, or in strict economic terms.
This is the point ofPoulantzas 's distinctions and his emphasis on the state
as that place where a power bloc is cemented out of the disparate 'ruling'
54 The Anatomy of Capitalist Societies

classes/fractions .18 Althusser continues from the quote above to maintain

that he needs only one example of this, namely, Lenin's attempt to
revolutionise especially the educational ISA to make it possible for the
Soviet proletariat, having seized state power, to secure the future
dictatorship of the proletariat and the transition to socialism. But that is an
extraordinary claim, because, although Althusser may only need one
example, that example may well be the exception and not the rule. After
all, it is a description not of a capitalist society but of a society in which
the proletariat has just seized state power. Thus, what may be true in that
situation may have no implications for capitalist societies, characterised
as they are both by a plurality of ruling classes and fractions, and by a
sphere of civil society lying outside the state. Althusser obviously rejects
the latter point but his grounds for doing so are unclear. He aims to extend
the Marxist concept of the state beyond the purely repressive aspects
which Lenin in particular emphasised (the 'special bodies of armed men
having prisons, etc. , at their command '19). He thus rejects the limitations
of that position and wishes to argue that the state contains a number of
non-repressive, ideological institutions (as Gramsci, of course, ar-
gued20 ). That argument is well worth making and wholly legitimate. But
what he then does is to make a further but unjustified leap into claiming
that all the private 'ideological' institutions are in fact part of the state.
But this has three deleterious effects: first, to dissolve the specificity of
the state and state power; second, to ignore the importance of ideological
effects within production itself; and third, to lead him to posit the
illegitimate unity both of ruling-class ideology and of the ruling class
itself.
A final set of problems with regard to the ISAs concern the nature and
role of class struggle. 21 He does not show how and why the reproduction
of capitalist relations might ever be broken; indeed, where do ideologies
of resistance stem from and get reproduced? What is the likely role of
'ideological work '? What is the basis of subordinate ideologies? Indeed
how could we understand ideologies which serve capitalism only
indirectly, such as sexism, racism, scientism, and so on? He tends to class
reductionism and a failure to indicate how 'ideology' could escape from
the state under the dictatorship of the proletariat.
I shall now consider more briefly the second part of his article, where
he argues that the category ofthe subject is the constitutive category of all
ideology, whatever its determination and its historical location. He
considers how individuals are 'obviously' subjects, arguing that this
'obviousness' is an ideological effect. Ideology, without ever appearing
 The Critique of Ideology 55

to, imposes such 'obviousness' such that each individual can recognise
the other as an already constituted, concrete, distinguishable and
irreplaceable subject. The mechanism by which ideology effects this is
that of interpellation or hailing. Ideology 'recruits' or 'transforms'
individuals into subjects by interpellation 'which can be imagined along
the lines of the most commonplace everyday police (or other) hailing:
"Hey, you there!" '22 By turning round, metaphorically, the hailed
individual becomes a subject since he (sic) recognises that the hail was
really addressed to him. Althusser's conception here is based on analogy
with the law, the subject being a generalisation from the more specific
'juridic subject'. The categories of law are those of subject, object,
freedom, will and so on. The law treats each individual as a consistent
subject who, unless it can be shown otherwise, is 'fully responsible' for
his actions and hence can be punished if he transgresses the law. The law
falsely treats the individual as though he were a fully consistent subject,
there is an imaginary wholeness, an imaginary conception of himself as
the author of his own actions.
Althusser also argues that each individual is 'always-already' a
subject, even before birth. The unborn child will bear its father's name
and will therefore 'have an identity and be irreplaceable' .23 The child is
thus always-already a subject determined as such by the particular
familial - ideological configuration into whkh he will be born. The child
is presumed to fit into this subjectivity and to respond to the appropri-
ate hailing. It is assumed that each individual has a ready-made subject -
and that there is a one-to-one fit between the individual and the sub-
ject.
Briefly, the main problems are as follows. First, it is wrong to assume
that the individual and the subject are the same, since it is possible for
there to be subjects who are not individuals. 24 It is also incorrect to view
such subject-ivities as easily acquired, as though it were just a matter of
being hailed. Freud and Lacan have shown the problematic acquisition of
sexual subjectivity - of the need to traverse successfully the mirror phase
and the resolution of the castration complex in the oedipal phase.
Althusser only cryptically hints at this. He also assumes incorrectly that
the mirror phase lasts throughout one's life, that subjects continue to be
interpellated (say as economic subjects) through the imaginary relation
with the real conditions of existence of say the workplace (to be hailed as
a 'foreman '). In fact, Althusser doesn't indicate what relationship holds
between different interpellations - is it possible to be hailed as both
foreman and member of the working class - and what consequences
56 The Anatomy of Capitalist Societies

would follow. The process by which the constitution of individual

subjects occurs is not one which is always successful and always
non-contradictory. How can each individual always-already be a
subject - if the effect of interpellation is to constitute individuals as
subjects?
These difficulties arise because of the particular function which the
'subject' is supposed to bear in Althusser's argument. On the one hand,
he aims to demonstrate the precise function of ideology in reproducing
the existing relations of production, via the concept of the ISAs. On the
other hand, he attempts to develop a general theory of ideology via the
interpellation ofthe individual subject. Yet to realise the first objective,
the second has to be transformed away from its Lacanian origins. Partly it
has to be transformed because a general theory of ideology is an
impossibility. The transformation though consists in: (a) treating the
process of constitution of the subject as non-contradictory and unprob-
lematic; (b) seeing each hailing as involving a similar imaginary
relationship, whether it is of one's sex, one's nationality or one's
occupational status; (c) identifying the process of constitution with that
of hailing such that the constituted subject is a passive and determined
bearer of wider social relations; and (d) providing no theory of the
interrelations between different interpellations such that on the crudest
empirical grounds the educational ISA can be judged more central than
the familial ISA. Thus, what Althusser's apparently radical argument
amounts to, is that under capitalist relations individuals are interpellated
in such a way that they are hailed as those subjects necessary for the
reproduction of the existing relations. Hailing brooks no argument - it is
assumed that the individual who is hailed responds, is situated and freely
acts in such a way so as to reproduce his own subject-ion.
One major effect of these criticisms is to question the thesis that there is
a unity to ideology - that it is in a sense an entity which can occupy a
distinctive relationship to other entities. Hirst has recently taken this
criticism furthest maintaining that none of the relations of the political,
the economic or the ideological can be viewed as instances, as definite
sectors, of a totality governed by their place in the whole and subject to its
limits.2s He arrives at this conclusion in part through a critique of
Althusser's unitary concept of ideology, through an insistence on its
necessary heterogeneity. He also argues that the reorientation of theory
towards emphasising the non-homogeneity of social relations enables us
to avoid the 'workerism' and 'essentialism' of existing Marxist dis-
course, to avoid, for example, the treatment of the Women's Movement
 The Critique of Ideology 57

as an auxiliary of the anti-capitalist struggle. 26 He further claims that we

have to consider ideology only through its effects, and that this is a matter
of political calculation. We cannot consider whether and in what ways
ideologies 'represent' underlying social relations. We must break more
convincingly than Althusser did with the classic concept of 'represen-
tation', and this is because we must reject epistemology.
My argument here partly goes along with Hirst, but then stops and
argues in a different direction. First, I fully agree that there is nothing
unified about 'ideology '27 and that much of what is usually located under
that category is properly to be seen simply as differentiated, diverse and
heterogeneous social practices. Certain ofthese are linked fairly directly
to capitalist social relations, others most definitely are not. Thus, it is
important not to reduce various forms of popular struggle to class
struggle, to an essentialist, workerist conception of politics. Also
ideology itself is only to be viewed through its effects. However, Hirst
makes three errors. First, it does not follow that the economy and the state
are characterised by the same heterogeneity as ideology. Indeed the
heterogeneity of ideology, or what I term in the main 'civil society' , is not
something true in essence of all social practices. It just happens to be the
case in regard to certain civil society relations within capitalist societies.
Incidentally, it is paradoxical for Hirst to present such an essentialist
argument. Second, the determination of the general relations between the
economy, the state and civil society, is necessary in order to effect a
proper 'political calculation '; without which it is difficult to see how this
might be achieved. Third, the argument against epistemology does not,
as we shall see, succeed, and therefore it is possible to reconsider the
problem of 'representation'. To do so, however, is not to reintroduce
Althusser's rigid distinction between 'ideology' and 'science'.
I shall now consider and criticise the argument presented by Hindess
and Hirst as to why we should reject epistemology. Following that I shall
briefly characterise what I mean by 'ideological effect', premised as it is
on the existence of epistemology. According to Hindess and Hirst
epistemology conceives of both a distinction between, and a correlation
between, a realm of discourse and a realm of objects of discourse. 28 In
empiricism the correlation between the realm of discourse and the realm
of objects is effected through the category of experience. It is through the
sense-experience of individual human subjects that the latter is known
and comes to be systematised within the former. There is thus a privileged
level which is given by the forms of discourse which directly designate
what is given to the experience of human subjects. The objects of
58 The Anatomy of Capitalist Societies

discourse are thus knowable directly and systematically. The category of

human experience ensures a direct correlation between the objects of
discourse and discourse itself.
Rationalist epistemology likewise conceives of a distinction between
the two realms. The correlation between them is effected through the
claim that there is a privileged level of a determinate body of concepts.
The relations between the concepts within the discourse are seen as
mirroring the relations between the objects of discourse, the relations
between the objects being part of the rational order in which the world is
constituted. Thus, the investigation of the relations between concepts is
that which yields the essence of the real, a process which makes no
reference to, or use of, the category of the individual subject. The internal
relations between concepts yield knowledge of the connections in the real
between the objects of discourse.
Hindness and Hirst reject both epistemologies. They argue that it is
impossible to refer to objects existing outside of discourse as a measure of
the validity of that discourse. Indeed it is improper to refer to objects of
discourse. They do not exist. The only entities that any discourse refers to
are in fact constituted in it and by it. What is specified in a discourse are
the entities which that discourse specifies; there cannot be extra-
discursive objects outside of which the discourse somehow yields
knowledge, either in empiricist or rationalist fashion. To argue that there
are is to effect a dogmatism, to claim that there is a privileged level which
is in some way the sure basis or guarantee of knowledge.
The first difficulty is that their argument is self-negating. Their position
is one which should just as much prevent us from discussing a discourse
as an object of discourse, as it prevents us from discussing wages or a
table as objects of discourse. How can they say anything about discourse
and its objects when what they say is part of their own discourse, which
strictly speaking cannot have an object as such? If they are talking about
anything then there must be certain objects of their discourse; that is, that
lie outside the discourse itself. But if so then their argument collapses. Or
they are not talking about anything outside their discourse, in which case
we need not pay attention anyway.
Second, it is unclear whether they also avoid dogmatism. They state
that it is necessary to found concepts which start from questions posed in
the arena of 'political calculation and struggle'.29 But why, one might
ask, is this the case? Why is this level privileged? Why does this discourse
take priority over other discourses? Further Hindess and Hirst are keen to
emphasise that theirs is not an idealist position, in which ideas or concepts
 The Critique of Ideology 59

are constitutive of the external world. They maintain that there is a realm
of objects which lie outside discourse, but their point is that these are not
to be seen as objects of discourse. There are only objects internal to any
particular discourse, plus those outside, but which are not objects of
discourse. Let us consider what now happens if a particular matter
becomes a question posed in the arena of political calculation and
struggle. Pres umabl y this will occur because of the interrelations between
the objects outside discourse. However, immediately we describe this we
locate our analysis within one discourse or another. Suppose that large
wage increases take place in the British economy, something that poses
major questions of political calculation for the Left. Immediately it can be
seen that our description of that process is itself linguistic and conceptual.
It is located within a discourse, where the term 'wage' has a particular
meaning especially through its relation with other terms (for example,
'profit', 'rent', 'interest', 'salary', etc). Now, it is also clear that such an
increase of 'wages' will have major implications for one or more
'theoretical' discourses; for example, monetarist, or Keynesian, or
Marxist theories of distribution and price formation. As Hindess and
Hirst say: 'questions cannot simply be taken as they are given in political
debate, they require critical theoretical evaluation'. 30 In other words,
there are relations of interdependence and priority between discourses, in
particular between 'political discourse' where direct political questions
are posed, and 'theoretical discourse'. But it is not at all clear how this
differs from someone else who might argue that:

(1) there is an external objective world (that is, objects outside

discourse);
(2) this world poses questions and issues (what Hindess and Hirst call
'political') ;
(3) any description of such questions is necessarily conceptual (that is,
within a discourse);
(4) there are interdependent relations by which 'theory' can illumine and
explain questions posed within 'pre-theoretical' or political dis-
course.

There are still epistemological questions involved in the relationships

(1) to (4), especially between (2) and (4). How does 'critical theoretical
evaluation' operate unless it can improve upon direct political experience
and the forms of discourse present within it? Are there not certain forms
of discourse which produce a very limited basic for political action? Are
60 The Anatomy of Capitalist Societies

not particular kinds of political calculation based on notions lacking

'critical' theoretical assessment? More recently, Hirst argues that
discourses and practices (can there be a practice without a discourse, or
vice versa?) do employ criteria of appropriateness or adequacy. What he
objects to is the claim that there is one trans-discursive epistemological
criterionY So his is not a relativistic position, only one which denies the
existence of a general epistemological absolute. Thus, Hirst provides no
grounds which would prevent us from employing criteria of adequacy to
'test' particular claims relating to the concepts and beliefs entailed within
a particular social practice, such as that of 'understanding social
relations' .
We have already seen that Hirst argues for the consideration of
ideological effects, but he also claims that we cannot designate certain
effects as 'false' (as in 'false consciousness ').32 He says that it is absurd to
say that anything that has effects is false - it would be like saying that a
black pudding or a steamroller is false. However, this is a poor analogy.
A certain social practice may have effects which are 'false' , or as I argue
below produce 'concealment', because they are not material objects in
the sense that a black pudding is such an object and may (will?) produce
indigestion as an effect. This is because within any such practice will be a
discursive structure which contains propositions; such propositions may
well have as their referents the structure of that social practice or its
conditions of existence, or the consequences of that practice on others. In
other words, the system of concepts embodied within that practice may
well be such that it yields sets of proposition which must involve the
concealment of the causes, nature or consequences of that practice, or
indeed some related social practice. Thus, the first condition for the
designation of a particular social practice as ideological is that there must
be a concealment which is derived from the system of concepts embodied
within that practice. This is a restrictive condition, since any system of
concepts can yield fairly diverse sets of propositions. It is only where that
system of concepts is such that the derivable propositions must involve
one or more of the forms of concealment, that the effect is to be
designated ideological. It should be noted that although such ideological
effects are embodied within specific social practices, they are not to be
reduced to, or conflated with, such practices. Ideology does not simply
have a material existence, but exists within and through ideas, through
the relations of concepts and propositions. 33 Let me now list some forms
of concealment. 34 A particular social practice has an ideological effect
when the concepts embedded in that practice:
 The Critique of Ideology 61

( 1) do not permit explanation in terms of wider scale social and historical

changes: there is an inappropriate isolation of practices;
(2) do not enable satisfactory specification of the practice in question;
thus, there may be a conflation ofpractices which should properly be
differentiated;
(3) produce an eternalisation of the practice, and a failure to see it as
historically bounded and hence that it may be transcended;
(4) obscure the social relations which underlie existent relations between
material objects, hence, where such objects are seen to have powers
which stem not from the social but from their natural characteristics;
(5) obscure the interrelations between this practice and one or more
other practices;
(6) hide the conflicts of interest between the differently located subjects
within that practice.

I have so far dealt with the first condition under which a social practice
is to be designated as ideological. The second condition is that such a
designation wiII only be made when it is shown that this concealment is in
the interests of one or more of the dominant social forces within that
practice; for example, of capitalists, or of men, or of whites, or of
Protestants, or of state bureaucrats, etc. I am making clear here that there
are diverse social practices within capitalist societies and that the
beneficiaries of ideological forms are likewise diverse. It is not however
generally the case that the explanation of the ideological effect directly
follows from the fact that such an effect is in the interests of a dominant
social force. It may be, but this would only be so in extreme conspiratorial
situations. Furthermore, the main consequence of such an ideological
effect is on that practice itself, not necessarily on the wider society. It
may have a non-existent or a negative functional consequence for the rest
of the society. It is wrong to suppose that because it exists and persists, it
is therefore functional.
Finally, two provisos of my argument should be noted. First, I am not
considering whether individual subjects or consciousnesses correctly or
incorrectly perceive social reality. I shall show in the next chapter that it
is within civil society that individuals are constituted as subjects, in
particular as bearers of particular kinds of social relations. However, here
I have been concerned to show that embedded within certain interpel-
lations are discursive structures which systematically prevent and
obscure the understanding of the nature of either their own, or of other
social practices. I am not operating here within the problematic of the
62 The Anatomy of Capitalist Societies

subject. Second, I do not intend to argue that there is a simple dichotomy:

true/false consciousness, or non-concealment/concealment. Rather,
there are degrees of concealment. One reason for this is that a
dichotomistic formulation makes it difficult to conceive of discursive
change and development, the understanding of which seems essential to
comprehending conflict and struggle, of both classes and of the other
important social forces in civil society. To argue that there are ideological
effects of certain social practices does not mean that one is committed to a
true/false or a science/ideology dichotomy.
5
The Practices of
Civil Society

I have so far maintained that there is an important distinction to be

maintained between social practice on the one hand, and ideological
effect on the other. By categorically distinguishing between the two we
can avoid an over-integrationist conception of capitalist social forma-
tions. However, there is clearly an opposite error we could make and that
is to fall into cultural pluralism. We might find ourselves arguing that
there are simply many different social practices, each with its own
culture, that they bear no particular relationship to each other, that some
happen to have ideological effects, and that there is no systematic
relationship between them. In this chapter I intend to set out an account of
civil society which avoids these deficiencies but which, at the same time,
renders coherent the manifest diversity and variability of capitalist social
formations.
Let me begin by considering some distinctions drawn by Gramsci, l
between 'common sense', 'philosophy', and 'hegemony'. The first of
these we can understand as the more or less immediate and unreflective
culture of a particular class or other social grouping. The second,
'philosophy', refers to an organised set of conceptions which entails an
active transformation of that culture. And 'hegemony' can be seen as the
distinctive relationship between infrastructure and superstructure charac-
teristic of a particular social formation; or to express it in the terms of this
book, the distinctive articulation between the economy and the state
effected by the practices of civil society. Johnson points out how the
notion of a class's 'common sense' or culture is not only consistent with
Marx's overall writings but is actually discussed, albeit in a cryptic and
untheorised manner in CapitaP. Marx refers to the 'general level of
civilisation' of the labourer, he refers to his (s ic) habits, customs,
conventional standards of life and skill, his 'moral' notions, his forms of
64 The Anatomy of Capitalist Societies

family and sexual relations and so on. These appear as fetters or

restrictions, as bases of working-class resistance, against the onslaught of
capital. However, once class cultures are seen as important, three further
consequences follow. First, there is no reason to suppose that other
categories of subject apart from classes do not also develop specific
'common sense' understandings: such as gender, generational, ethnic
and residential social groupings. Second, within any such groupings a
distinction may be made between a relatively unformulated common
sense, and a relatively formulated 'philosophy' oriented around various
bases and dimensions of resistance. This is not incidentally to resurrect
the classic 'in-itself' - 'for-itself' distinction because any culture is
partly in- and for-itself containing a mixture of 'common sense' and
'philosophy'. I shall argue below against the notion of a class essen-
tialism, and maintain instead that only by exploring a specific civil
society can one identify the possible bases and dimensions ofresistance. 3
Third, it is difficult to demonstrate that particular class cultures are in fact
tightly bounded and discrete. Neither the discussions in Gramsci nor
Marx show that this is the case. Rather, it is a historical question, as to the
continuities and discontinuities between the cultures of different social
groupings. However, one aspect of this is important to bring out, and that
is the character of the state. Very broadly it is the state that attempts to
establish and sustain hegemony, an articulated structure that conceals and
deflects the contradictions and conflicts inherent within civil society, and
inherent between the state and civil society. This does not mean,
however, that hegemony entails cul tural homogeneity, merely an
articulation which effects concealment and deflection. It is itself a
contradictory process since the attempts to hegemonise such structures
may well either strengthen existing forms and bases of resistance (for
example, of working-class culture), or generate new bases focussed
around popular struggles within civil society against the state.
What this demonstrates is a point recently discussed by E. P.
Thompson. 4 He argues that historical materialism must be concerned, not
only with the circuits of capital, circuits in a sense within the economy,
but must also address itself to the circuits of power, and of the
reproduction of ideology. He suggest, but fails to demonstrate, that these
belong to a different logic and to other categories of analysis. Further-
more, although he refers to the interaction between these circuits ('every
circuit sparks across the other'S), he fails to indicate how to understand
such interactions. He suggests that the notion of a determinant economy
should be seen as 'setting limits' and 'exerting pressures ';6 but this
 The Practices of Civil Society 65

formulation is only of limited helpfulness. What is imperative is to

specify both the limits and the dynamic interrelations of different circuits.
For example, we shall see below that the state's overall functioning is
determined by the laws of motion ofthe capitalist economy. Yet it does
not respond directly to that economy, but in a mediated fashion, through
the social practices constitutive of civil society. And in its efforts to
transform civil society it generates and reinforces resistance within it, and
this in turn reacts back both on the state and on the economy, through, for
example, relative changes in the willingness of different groupings to
enter the labour market. It would seem that Thompson's categories of the
circuits of 'power' and the 'reproduction ofIdeology' need amplification;
hence the analysis below of the practices of civil society and of their
location between the circuits of capital and the state.
I shall begin here by consideration of the concept of social class and of
the following problem involved in its analysis? In conventional accounts
social classes are conceptualised, both as categories of economic agents
('class-in-itself'), and as political and ideological forms ('class-for-
itself'). 8 What relationship holds between them? There are two orthodox
positions. In the first, the latter are reduced to the former; politics and
ideology are merely expressions of class interests determined within the
economy. In the second, politics and ideology are not reduced directly to
the economic; rather they are thought to represent economic interests. I
take it that the reductionist position is incorrect for reasons stated at
certain points in this book. What, though, of the representational
position? This can be seen to involve three aspects: the content of what is
represented (class interests), the means of representation (political/
ideological apparatuses), and the representation itself (the practices of
specific political forces). According to Cutler et al. , ifthe representation
cannot be reduced directly to class interests, then it must be that the means
of representation are independently effective. So such means cannot be
reduced to class interests, since they in part determine the representation.
Classes cannot therefore be conceived of as economic-political-
ideological unities where the political/ideological m..::rely represent
economically determined class interests. Politics and ideology are thus
separate apparatuses. Political issues are to be evaluated, not in terms of
some alleged representation of 'true class interests', but in terms of
definite political practices and the attainment of specific political
objectives. What is wrong with this by now influential thesis?9
The difficulty which they identify in the classic treatments of social
class stems from their adherence to the conceptual distinction between
66 The Anatomy of Capitalist Societies

economics on one hand, and politics/ideology on the other, in other

words, to a base: superstructure dichotomy. Classes for them are to be
conceived of as categories of economic agents who occupy definite
positions with regard to particular forms of possession of, or separation
from, the means and conditions of production. The only question then
is whether such economic classes are directly expressed or represen-
ted within political/ideological forms. It is relatively easy to show that
they are not, but this only follows because of Cutler et al. 's adherence to
the conventional conceptual dichotomy of base and superstruc-
ture.
However, if we conceptualise social classes, not as categories of
economic agents, but as categories of subject within civil society,
different consequences follow. There are two types of such classes,
which are given by the different forms of struggle: they are, as Laclau
puts it, 'class-struggle' and 'classes-in-struggle'.10 In the former there are
capitalists and workers whose relationship is structured by the production
and appropriation of surplus-value. Neither class can exist except in a
relationship of struggle, by subjects bearing the capitalist function against
subjects bearing the labour function. The relationship is determined by
the central relationship of the capitalist mode, the struggle over the
production and appropriation of surplus-value. It is only subjects bearing
one or other function who can engage in such struggle; that is, class
struggle. The position of each subject, bearing the function of either
capital or labour, is one of struggle against the opposing class subjects.
But two points should be noted which distinguish this from the reductionist
position criticised earlier. First, classes only exist in struggle within civil
society; they do not exist first of all as economic classes which may/may
not then realise their economic interests at the political/ideological level.
Classes only exist within civil society - the form that they take is given,
first, by the current patterns of capital accumulation and relations
between the functions of capital and labour; second, by the forms of
gender, age, racial, regional and national interpellation within civil
society; and third, by the forms of political organisation and state
apparatuses. There are no pure classes determined economically-
classes exist only through the precise manner in which the dominant
capitalist relations, with a particular division of the antagonistic functions
of capital and labour, are manifest within civil society. Such social classes
are thus necessarily over-determined.
Second, this account only pertains to the 'class struggle' between
capitalists and workers; there are also 'classes-in-struggle'. Such classes
 The Practices of Civil Society 67

are those categories of subject who occupy a common position in

relationship to the means of production, but it is a relationship which does
not entail direct antagonism which another social class. Examples of such
classes would include ground-rent landlords, the so-called new middle
class, the traditional petty bourgeoisie, and the lumpenproletariat. There
are three ways within capitalism in which the struggles of these classes
are different from those between capitalists and workers. First, each
'class-in-struggle' can be conceptualised without regard to another class
with which they are directly antagonistic (this would not be true of the
landlord class within feudalism, where the capitalist class would be a
class-in-struggle). Second, as a consequence, these classes-in-struggle
are less directly structured by capitalist relations of production than are
capitalists and workers. Although this is somewhat imprecise, what I
wish to convey is the thesis that the social practices within civil society
are heterogeneous in their degree of determination by capitalist relations.
Thus, with classes-in-struggle, their form and effectivity is more directly
dependent on the nature of the specific civil society and on the particular
modes of interpellation. The nature of their struggles, the symbols, rituals
and practices within which it is conducted, derive in large measure from
those accessible in civil society. Third, because each of these classes are
not determined by a directly antagonistic relationship with another class,
the dominant contradiction will not be one of ruling class/ruled class.
Rather, it will be: the state/the people, or: the large corporations/the
people.
So far I have maintained that in capitalism there are two main forms of
class struggle: class struggle proper and classes-in-struggle. Laclau also
suggests that there is popular-democratic struggle, involving the organi-
sation of 'the people' based on non-class forms of interpellation, of
gender, generation, race, region and nation. This is very important since
it means that class struggles and classes-in-struggle take place within an
already structured civil society, structured in terms of various forms and
effectivities of popular-democratic struggle. In particular, they have to
take place within the organisation of the 'people' on the basis of
nationality, on the particular national organisation of each civil society.
As a way of clarifying my argument here let me briefly consider
Laclau's position a little further. He says that:

the level of production relations always maintains the role of

determination in the last instance in any social formation. This in itself
established the priority of the class struggle over the popular-
68 The Anatomy of Capitalist Societies

democratic struggle, since the latter takes place only at the ideological
and political level (the 'people' do not, obviously exist at the level of
production relations).l1

However, this suggests that 'class struggle' and 'popular-democratic

struggle' are located at different places within the social formation, the
latter only occurring on the top two floors (the ideological and the
political !). But this means that it is unclear how these different classes and
popular-democratic forces actually come into struggle with each other. It
is prefemble to argue that the popular-democratic struggles, classes-in-
struggle, and the class struggle all take place within civil society. It is
only the class struggle between capitalists and workers which is directly
structured by the social relations of capitalist production. However, to
say that the class struggle is structured by the social relations of capitalist
production is not to argue that it is determined by them. Capitalist
production, very crudely, can be viewed as comprising two functions, of
capital and oflabour, and of the necessary relations connecting them. In
this no particular specification is provided of which individuals happen to
occupy such functions and of the nature of struggle between them. I will
now consider further the nature of the 'economy' and 'civil society' and
see how these terms refer to very different kinds of entity: the former to
determinate social relations in which it is presupposed that there are
individuals who bear one or other specified function; and the latter to
social practices which both constitute and presuppose the existence and
intemctions of specific human subjects. I will now amplify this
distinction, beginning with the former.
It is clear that in his economic analysis Marx views the individual
merely as the 'bearer' or 'support' of specific social relations of
production. As Marx says in the 1859 Preface: 'men enter into definite
relations that are indispensable, and independent of their will, relations of
production' .12 Thus, in the analysis of economic relations: 'individuals
are dealt with here only in so far as they are personifications of economic
categories, the bearers (Trager) of particular class-relations and in-
terests ';\3 the 'characters who appear on the economic stage are merely
personifications of economic relations; it is as the bearers of these
economic relations that they come into contact with each other' .14
Individuals thus only appear to the extent that they bear particular
relations, of buyer or seller, of capitalist, landlord or wage-labourer. It is
only when and in so far as they fulfil a particular function that they appear
within the analysis. Marx talks of the 'bearers of the various functions in
 The Practices of Civil Society 69

the production process' .15 Even the capitalist is 'merely capital person-
ified and functions ... solely as the agent of capital' .16 Two further
aspects of this analysis should be noted. First, although clearly an
economic character will be stamped or impressed upon individuals, it
does not follow that all bearers will bear their particular function in
identical fashion. Indeed capitalists may not function solely as the agent
of capital; but if this is the case then these capitalists will cease after a
period to bear that function since their capital will become devalued.
Second, it is a precondition of individuals acting as economic agents, as
bearers of one function or another, that they act as subjects, as conscious
bearers endowed with a will. 17 This means that it is condition of existence
of capitalist production that there is a realm of practices where such
individual subjectivities are constituted and reproduced; that is, in civil
society.
How then are we to view the relationship between the social relations
of production and the realm of civil society? The crucial link between the
two is provided by the sphere of circulation which is part of both the
economy and of civil society, and is central to the interrelations between
the two. Capitalist relations necessitate the separation between produc-
tion and circulation, and it is the latter which produces independent
individual subjects separated from and free of natural bonds. The effect of
exchange relations is that:

the individual, each one of them, is reflected in himself as its exclusive

and dominant (determinant) subject. With that, then, the complete
freedom of the individual is posited: voluntary transaction; no force on
either side; positing ofthe self as means, or as serving, only as means,
in order to posit the self as end in itself. 18

So although Marx is not concerned to analyse the specific differences

between individuals, he is not at all indifferent to those processes which
produce independent, individual subjects as effects. However, where his
work can be faulted is in its implication that the generation of relatively
free, isolated self-interested individuals only results from the sphere of
circulation. What seems central to establish is that this sphere is the
precondition of civil society, but that within the latter there is a plethora
of social practices which in various ways are all responsible for
constituting and reproducing human subjects. Furthermore, these are
practices which vary greatly in the degree to which they are determined
by capitalist social relations. Each has to be considered in its own
70 The Anatomy of Capitalist Societies

right - having its independent effectivity and potential forms of political

practice.
The significance of this for the analysis of social classes is clearly
considerable. Let me clarify. First, it means that we cannot talk as though
there is some essential character or nature of a particular social class
which simply has to be released or expressed. There is not an essence to
working-class or capitalist practice which must merely be realised. All
we can say is that these individuals who bear one or other function will
struggle to sustain or to improve their material conditions of life. Nothing
else follows necessarily, neither the politicisation of struggle, nor the
generalisation of struggle from specific individuals to the wider social
class. Second, there is always a diversity of bases upon which individuals
may mobilise, and these depend upon the particular structuration of the
given civil society.19 The following are the main dimensions of
structuration in the generally descending order of importance:

1. Spatial organisation oflabour and of residence; that is, into nations,

regions, cities, towns, countryside, neighbourhoods.
2. Sexual division division of labour; that is, that differential
allocation of gender into spheres of production/reproduction, and the
organisation of sexual relations.
3. Religions/ethnic/racial allocation of subjects.
4. Differentiation of subjects on the basis of trade-union and profes-
sional associations, artistic and leisure organisations, political parties,
media institutions, etc.
5. Generational allocation of subjects.

Identification of the degree of homogeneity of a particular class depends

on the interaction between these dimensions of structuration, together
with analysis of the divisions within the capitalist labour process; that is,
of the division between mental labour and manual labour, and of the
degrees and forms of skilling/deskilling. Third, although such classes
exist and function within struggle, in many cases other struggles
predominate and minimise the salience of class struggle. All the sources
of structuration above may be the basis of struggle involving small or
large numbers. Such struggles will result from the attempt by individuals
to maintain or improve their material conditions of life. They thus relate
to their conditions resulting from the overall capitalist economy.
However, although capitalist relations are dominant it does not follow
that class relations (i.e. 'class-struggle ') are primary. Fourth, to a
 The Practices of Civil Society 71

significant extent class struggle is a matter of struggling to establish the

salience of class. It is not simply a means of expressing already given
class relations. Whether or not such relations are salient is a contingent
matter depending upon the particular structure of that social formation.
Gasses only exist in struggle; and there is no necessary reason why class
struggle dominates other forms of struggle. We will consider in Chapter 8
how the structure of the civil society affects different fonns of struggle.
Fifth, even when we consider classes themselves, because of the
disjuncture function/subjects, it is possible for individuals to perform
aspects of both capital and labour functions. An example would be a
production engineer. Further, it is also possible to identify ambiguously
structured social classes, classes-in-struggle, such as the new middle
class which performs the function of capital while not owning Ot
possessing the means of production. 20 In these cases of 'ambiguously
structured' individuals and classes the practices of civil society and their
relationship to the state will be more salient than in the case of individuals
unambiguously performing a capital or labour function. 21 Their attach-
ment is likely to be greater to popular democratic rather than class
politics.
We can perha ps ill ustrate the importance of the social practices of civil
society by referring briefly to two Marxist-influenced writers who have
been well aware of the relevance of these practices. Jakubowski points
out that the term 'social being' does not simply refer to economic
relations. 22 In any society it is necessary to consider the entire set of social
relations, not seeing these as undetermined, but as producing quite
distinctive effects. He says that it is not simply a question of which class
an individual belongs to, but also:

The particular layer or role which they occupy within that class, the
social position of the family (e.g. the white-collar 'proletarian' who
comes from a once rich bourgeois family), and above all, the
ideological traditions of the group (religion, received political or
philosophical ideas and prejudices, etc.).23

Jakubowski does not, however, provide a way of theorising these further

aspects of our 'social being'. By contrast Habermas has attempted to
develop the conceptual distinction between 'work' and 'interaction' and
to argue that Marx makes a cardinal error in reducing the latter to the
former. 24 To a degree this is unjustified since for Marx capitalist relations
involve both work and interaction, that capitalism is simultaneously
72 The Anatomy of Capitalist Societies

material and social. But it is not unreasonable to argue that Marx does not
so much reduce interaction to work, as ignore these interactions outside
work, or more precisely outside production. Hinted at, but ultimately
obscured in the notions of work and interaction as modalities of action, is
the categorical distinction between the social relations between agents in
the capitalist economy, and social interactions between subjects within
civil society. Let me now turn to the latter in more detail.
Civil society embraces a wide variety of social practices which involve
the constitution and actions of individual human subjects. Excluded from
this are the apparatuses of the state whose characteristics will be
discussed in Chapter 7. To elaborate on the conception of civil society it is
unnecessary to have developed a general theory of the subject, as certain
writers have recently argued. 25 All that is required here is to acknowledge
that individuals are not present 'always-already' as subjects, but that they
have to be formed as such through social experience. Further, the process
of constitution of subjectivity is achieved through language, through the
positioning it accords to individuals within particular discursive forma-
tions. It is not merely that individuals are allocated roles , since this
implies that they are already formed as subjects. It is that through social
experience, through their position within various discourses, individuals
view themselves and act as autonomous centres of creativity, conscious-
ness and initiative. They are thus determined as conscious and self-
reflexive, to act as autonomous, whole and independent subjects. The
most important interpellations ofthe subject are those ofspatio-temporal
location and of gender. The effect of the former is that individual subjects
are constituted who are aware of their presence as subjects resident with a
particular spatial location (street, town/countryside, region, nation) at a
given period of time (born of a particular generation defined by its place in
relation to others). The interpellation of gender is to produce antonomous
sex-ed subjects, each defined by its relationship of difference with the
other. Gender, like spatio-temporallocation, is socially constructed and
produced through the differences entailed within language. The other
interpellations, of ethnicity, class, religion, politics and so on, can only
be understood and related to these primary interpellations. They mayor
may not be in contradiction with them. As Laclau says of a familial
interpellation (this would embrace both spatio-temporal and gender
interpellations ):

When a familial interpellation, for example, evokes a political

interpellation, a religious interpellation, or an aesthetic interpellation,
 The Practices of Civil Society 73

and when each of these isolated interpellation operates as a symbol of

the others, we have a relatively unified ideological discourse. 26

Yet, on the other hand, in a period of 'ideological crisis' there is not this
inter-discursive connotation:

Each one of the sectors in struggle will try and reconstitute a new
ideological unity ... one of the possible ways of resolving the
crisis ... is to deny all interpellations but one, develop all the logical
implications of this one interpellation and transform it ... into a
principle of reconstruction of the whole ideological domain. 27

I shall not say any more here on the processes of constitution, of the
mechanisms within discourse which produce subjects as effects. I shall
simply assume their operation, and note in passing the inadequacy of any
theory of ideology or of civil society which simply deals with this
constitution of subjects, and not with their resulting actions. There are, I
believe, three spheres of civil society, that of circulation, that of
reproduction and that of struggle. I have dealt to some degree with the
first and the last; let me now consider the second in some detail.
The sphere of reproduction consists of these social practices within
which individuals are reproduced as subjects willing and able to bear
functions, either within capitalist production or within the state. This
involves the biological, the economic and the cultural reproduction of
individual subjects. The most important aspect of this involves the
reproduction of labour-power. The problem about this from the view-
point of capital is that labour-power, although a commodity, cannot be
produced, or reproduced, as a capitalist commodity. What happens is that
individual subjects are constituted principally within the sphere of
reproduction, and they then may sell their labour-power within the sphere
of circulation. As we saw in Chapter 3, the conditions of exchange are
significantly structured by the relations of production. However, that is
by no means to say that the sphere of reproduction is simply determined
by relations within the sphere of production. Butthe consideration ofthis
question in any detail requires an el ucidation, however sketchy, of certain
aspects of sexual ineqUality and of the differential role of male and
females subjects. I shall concentrate upon this aspect of the sphere of
reproduction and ignore other potential bases of popular-democratic
struggle.
Broadly speaking there are two main traditions in recent theoretical
74 The Anatomy of Capitalist Societies

attempts to explain the subordination of women. On the one hand, there

are 'economic' or 'materialist' accounts in which this subordination is
seen to stem from the kind of work or production in which women are
primarily involved. It is because they are principally employed as unpaid
domestic servants, reproducing, economically and biologically, the
commodity of labour-power, that is responsible for generating sexual
inequality and from which ideologies of female inferiority derive. 28 On
the other hand, 'ideological' or 'culturalist' explanations concentrate on
non-economic determinants. Thus, certain writers focus on the differen-
tial socialisation of men and women, for instance, on how women are
brought up to conceive ofthemselves as maternal, to see child-rearing as
the appropriate form of self-fulfilling activity.29 Other literature in the
psychoanalytic tradition concentrates on how the sexed subject is formed
in ideology through the operation of patriarchal family relations. 30 In the
following I shall adopt a position partly drawing on both of these
orientations, although I shall do no more than sketch out the theoretical
space for such a synthesis. 3 !
The first tradition, the 'economic', begins from an essential and highly
illuminating premise; namely, that commodity production under
capitalist social relations marginalises domestic work and hence the
family and women. Thus, the characteristic work of women takes place in
a sphere separate from, and subordinate to, that of men. Domestic work
comes under the sway of capital, is significantly affected by it, and
changes in part because of the alterations in its requirements. To the
extent that women engage in work within the spheres of production and
circulation, this is typically in circumstances which reflects their other
involvement in the sphere of reproduction. This also means that no
coherent programme for reforming sexual inequality can be envisaged
which does not involve transforming the sexual division oflabour. Purely
culturalist interpretations of female subordination are therefore at best
inadequate, and at worst highly misleading, if this primary material
subordination of women within the sexual division oflabour is ignored. A
psychoanalytic focus on the universal structure of patriarchy is also
flawed for this reason. However, it does not follow from the establish-
ment of the importance of the separation and subordination of domestic
work to capitalist relations, that we can thereby establish the location of
women within capitalist social formations. Because the primary cause of
female subordination is 'economic' it does not follow that women's work
is thereby part of the capitalist economy. Capitalism is important in the
way that it effects the separation between production/circulation, and
 The Practices of Civil Society 75

reproduction; labour, to produce use-values, takes place in both but only

production/circulation constitutes the capitalist economy. Reproduction,
I suggest, is one sphere of civil society - the biological, economic and
cultural reproduction oflabour-power, focused in and around the family
household, is no less work for being outside and marginalised from the
spheres of capitalist production and circulation. To understand it as such
demonstrates the relative independence oflabour-power from capital and
from the state, and why it necessitates exha ustive mechanisms in order to
bring it under control and domination within the capitalist labour process.
I shall now amplify my argument so far by claiming, first, that the
reproduction of labour-power within civil society, and in particular
within the family, is not something that necessarily follows within
capitalism. Second, I shall show that domestic work is not part of the
capitalist economy; it is labour that is, technically speaking, not socially
necessary, abstract and social. It has some quite distinctive features. 32
First then, why does reproduction take place within civil society, in
particular within the family? Why were Marx and Engels wrong in
believing that the family form especially within the proletariat would
collapse with the general development of capitalist relations? Much
discussion on this issue is conducted in rationalist terms; that is, given
relations between the elements of the CMP, then what does this logically
imply for non-capitalist social relations, especially the family? But this is
an incorrect approach since it ignores how the CMP only comes into
existence and ultimately into dominance on the backs of pre-existing
social forms. The family may lose its functioning as relations of
production, but it is nevertheless a distinctive social practice, of primary
importance in the constitution and maintenance of individual subjec-
tivities. It thus does not follow that because a particular social practice is
'non-capitalistic', that it will ultimately disappear because it is logically
incompatible with 'capitalist relations'. What is crucial is the establish-
ment ofthe precise effectivity of that particular practice within the social
struggles which are consequent upon such social relations. We must not
consider the comparison in abstract between two or more sets of social
relations; rather it is essential to consider the specific social groupings
produced within civil society, and to consider how in their struggles they
will undermine or strengthen particular social practices. It will be useful
here to consider briefly the working class in nineteenth-century Britain
and its relationship to the family.
Humphries considers and rejects two theses. 33 In the first it is argued
that because (a) the family is a property relation, and (b) the proletariat is
76 The Anatomy of Capitalist Societies

propertyless, therefore (c) there is no material base for the family and it
will disappear among the working class through proletarianisation. 34 In
the second, the need for the family is deduced from the needs of capital;
thus the domestic labour performed within the family enables wages to be
lower than would have been the case had working-class living standards
been solely dependent on purchased commodities. 35 So, on the first
argument, the family will disappear with the development of capitalist
relations; on the second it will be sustained because it is in the interests of
capital. Humphries argues, however, that both arguments omit a crucial
consideration; namely, the role and effectiveness of working-class
resistance in maintaining the family form against the onslaught of capital.
So, contra the first argument, Humphries argues that the family will not
disappear through proletariarisation; and, although agreeing with the
conclusion of the second, argues that this stems not from the needs of
capital but from the resistance of the working class itself. In particular,
she argues that the family form was sustained in nineteenth-century
Britain because of the following advantages:

(1) of allowing a greater measure of control over the labour market and
hence permitting the development of a 'family wage ';
(2) of providing a viable, sustained source of social support in
conditions of chronic uncertainty;
and
(3) of providing support to non-labouring members without recourse
to those alienating and demoralising forms provided within the state.

We may also add that struggling for a 'family wage' and hence for the
non-employment of women and children weakened the ability of capital
to deskill and hence to lower wage-levels. Overall, then, the concentra-
tion of non-capitalist and non-state social relations within the family both
emerge from working-class struggle, and help to strengthen that struggle.
What is the role of capitalist social relations in producing this?
Capitalism effects a division between production and reproduction, but
this does not mean that the latter has to be either privatised or socialised,
non-commodified or commodified. It is though reasonable to expect a
shift towards commodification of reproduction as capital seeks out newly
profitable opportunities. But it is not correct to assume, as Zaretsky does
for example, that reproduction is necessarily privatised, that capitalism
per se creates the division between the socialized production of
commodities and the privatised reproduction oflabour-power. 36 What we
 The Practices of Civil Society 77

have further to consider is how the division between productive

consumption and the consumption to reproduce labour-power affects
particular classes and their form of struggle. We have seen that the
working-class family, and hence the privatised reproduction of labour-
power, is an achievement of struggle, and strengthens that struggle.
What, then, are the social conditions that permit this to occur? There are
three points. First, capitalist societies are based upon the exchange of
commodities rather than the exchange ofwomen. 37 As a consequence this
permits some development of a civil society but the form it takes and its
relationship to both the economy and the state is a matter for struggle.
Second, given that capitalism develops within a society where there is an
already existent family form, then the latter will be generally sustained if
possible, since it provides immediate support for peoplefrom capitalism.
It is misleading to refer to the 'bourgeois family'. The family is in part a
protection against capitalist relations. Third, the reason why this can and
often does occur is because of the spatial organisation of the sphere of
reproduction. It can be differentiated from capital because of the
organisation in space of separate households, within which the relatively
'free' interactions of subjects can be achieved. What occurs within
families (of all classes) is thus outside capitalist relations and the state,
and is relatively independent of them both.
I shall now briefly consider the nature of domestic work, showing that
it is not a direct or indirect part of the capitalist economy. I shall concl ude
by making a few points about the state and reproduction which will
prepare the ground for the next chapters on the state itself.
Domestic work is unproductive. It is outside the capitalist economy
since it is not socially necessary, abstract, and social .38 This can be seen
in a number of respects. First, domestic work is not labour that is
allocated by the law of value. It is performed independent of such
allocation. Hence, fluctuations in the price oflabour-power do not affect
the performance of domestic work. It will even be performed when its
product - that is, labour-power - cannot be sold at all. Second, this also
means that the relations of commodity exchange do not affect the
domestic labour process. There may be overproduction of the commod-
ity, or there may be 'inefficient' production. In either case, the domestic
worker does not produce the product, labour-power, as a commodity for
exchange. Changes in the domestic labour process do not result from
changes in the conditions of exchange of this commodity for others.
Third, domestic labour is not abstract because there is no equalisation and
interchangeability of it with other concrete labours. As Marx says: 'the
78 The Anatomy of Capitalist Societies

maintenance and reproduction of the working class' is left 'to the

worker's drives for self-preservation and propagation'. 39 Ignoring the
resort to drives, what this means is that it is left to women, outside and
marginalised, who reproduce a commodity that does not interchange and
establish equalised relations with other commodities. And this is in part
because there is no competition between households to minimise the
labour-time embodied in their particular products. As we have noted,
'inefficient' households do not fail to sell their products. There is no
relationship between the labour-time involved in reproducing labour-
power and the wages that the sellers of that product can achieve within
exchange. It is impossible to establish the socially necessary tasks of
domestic work, or that there is any socially necessary labour-time
towards which individual labour-times will necessarily converge. There
is no mechanism by which domestic work becomes reduced to socially
necessary and abstract labour. The duration of domestic work is
unregulated. It might take thirty hours a week, or a hundred or more. 40
There is no mechanism which equates the exchange between the
wage-labourer and the domestic labourer.
There are important consequences of this for women's struggles. I
have earlier characterised these as popular, or popular-democratic, and in
general they do not coincide with class struggles. They are usually based
on the dichotomy, women: men, or, the people: male capitalists and state
bureaucrats. These struggles are highly problematic to sustain because:

(a) Women primarily function as domestic workers.

(b) Male workers have sought to develop and protect a 'family wage'.
(c) Women cannot determine whether or not they are exploited as
domestic workers since there is no socially necessary labour-time
establ ished.
(d) To the extent they also function as wage-workers, they mainly
have the status of an industrial reserve army.
(e) As domestic workers women are individualized, as wage-workers
they are in the weakest parts of the labour-market.41

One crucial effect of their location within the sphere of reproduction is

that a major form of struggle, both individual and collective, is to escape
from that sphere, to enter the spheres of production/circulation, or the
state. So this is a strange kind of struggle , ofthe subordinated grouping,
not trying directly to change the relationship with the dominant grouping
(Le. males) but attempting to opt out (at least for periods) and to join
 The Practices of Civil Society 79

another subordinate social grouping, of wage-labourers. So struggle

consists of attempting to escape from legal-economic dependence on the
wage-labourer, a dependence produced by the fact that while one can
change an employer, it is much more difficult to change a husband. 42
I have thus attempted to demonstrate the manner in which reproduction
is realised. It occurs within a set of social practices, relatively
independent of capitalist relations, and, as we shall see, of the state.
However, the establishment of that relative independence is a matter of
struggle, revolving around existing forms of practice present with the
development of capitalist relations. The form of work within the sphere of
reproduction is distinct from that within the capitalist economy. It is as a
result difficult to organise and regulate by capital or by the state; as we
shall see the state is not necessarily able to ensure appropriate subjects
willing and able to bear functions in the spheres of production/
circulation. This shows the importance of avoiding the kind of Marxist
functionalism in which the state is seen as perfectly adept at resolving the
problems generated within the rest of a particular capitalist social
formation. In particular, analysis of civil society is essential in order to
avoid such functionalism. Consider the following points made by
McIntosh on the state and family households. 43 First, the state in Britain
does not particularly intrude in internal family relations in order to protect
women; second, it does in part sustain the family household and hence
strengthens the industrial reserve army status of women; yet third, on
occasions by maintaining the family house hold through aiding women as
reproducers of labour-power, it weakens their status as an industrial
reserve army. These three points illustrate the importance of a civil
society analysis, of seeing the state, both as responding to the demands of
various social groups (especially male workers and the notion of a family
wage), and in only being able to intervene in the economy through
devising policies which affect specific individual subjects who will be
engaged in various forms of social struggle, struggles which will often
preclude the successful realisation of the objectives of the state in that
conjuncture. To consider this further we must now tum to explore the
state in detail, and of the circuits through which it is interdependent with
civil society and with its constitutive spheres of circulation, reproduction
and struggle.
6
The State: A Critique

It is now commonplace to note that Marx did not provide a general,

well-articulated theory of the capitalist state. There are rather a number of
different texts in which he (and/or Engels) makes reference to the state in
capitalist societies. We can, at the cost of considerable simplication,
classify these into the following approaches: 1 the state as a parasite, as
the private property of state officials;2 the state as mystification, while
apparently representing general interests of the society, in fact represent-
ing specific interests;3 the state as the reflection of the economic base
responding to, and hence facilitating, the developing forces of prod uc-
tion;4 the state, as a set of essentially repressive institutions, which
function as the instrument of class rule;5 the state as a social regulator
moderating and channelling the struggles between classes, in cases
through suppressing the interests of specific capitalists;6 and the state as
an ideal collective capitalist standing alongside capital and sustaining its
pattern of accumulation. 7
Two points need to be made about these various approaches. First,
each of them suffers from well-known deficiencies. The first three are
highly partial. Thus, the state is not simply parasitical, nor does it
function merely to mystify. Nor is it correct to reduce the state to the
economic base, most obviously because this conceals the manifest
diversity of existing capitalist states. The other three approaches are more
defensible, but each suffer from limitations which will be explored
below. Second, even if we were to combine two or more approaches (as,
of course, Marx/Engels did themselves on occasions) there is still nothing
resembling a theory of the state in capitalist societies.
To rectify this we can begin by trying to identify the criteria for such an
adequate theory. 8 Such a theory would need to indicate, both the general
form of the state in capitalist societies - that is, as a system of political
 The State: A Critique 81

domination separate from both the economy and civil society; and also
why this would appear to be best expressed as representative democracy,
why democracy seems to be the best possible political shell for
capitalism. A theory of the capitalist state would be adequate to the extent
that:

(1) The theory is based on the specific characteristics of capitalist social

formations; that is, both on the nature of the eMP, and on its
relationship of dominance over other forms of social and private
labour. Such societies, founded on the dominance of the commodity
form, entails a state different from that found in pre-capitalist
societies. There is no reason for thinking that it is possible to develop
a theory of the state in general, only theories of states in particular
kinds of society.
(2) The state is not reduced to the economic, or to be seen as the
instrument of a dominant economic class. This is so partly because
the dominant economic class does not always control the state, and
indeed may be said to be exercising its domination most effectively
when it is not. It is also because this 'instrumentalist' conception of
the state contravenes criterion (1), namely, in implying the neutrality
of the state which can thus become the instrument of whichever class
happens to assume economic dominance. In general, this position
ignores how the state takes a particular form and pursues certain
policies because of its structural interrelationship with other elements
of capitalist societies, not because it happens to be the direct or
mediated instrument of the economically dominant class.
(3) The state is not viewed functionally, as automatically developing its
form or changing its policy in response to the needs of the capitalist
system. It must not be assumed that each state always has the most
functional consequences for the overarching structure. This is
because it operates in the context of individuals, groups, and classes,
struggling to sustain their material conditions of life. The state does
not develop automatically but in the context of such struggle. But this
does not mean (pace criterion (2) ) that the state is the instrument of
anyone class, or indeed of an alliance of classes. But it is to be
viewed as operating in the context of such struggles, and if it changes
its form or its policy, this will be through its responses to the nature
and effectiveness of social struggle. Major changes in the state, such
as the progressive expansion of the franchise, can only be understood
in part through the successful waging of concerted action by the
82 The Anatomy of Capitalist Societies

popular forces in that society at the time. It also follows from this that
the state does not necessarily pursue the 'best' policy from the
viewpoint of the maximum accumulation of capital. Capitalist states,
as elite theorists are keen to argue, operate within the context of
different groups and classes struggling for power. This also means
that such social struggles are reproduced within the state itself.
(4) Successful explanation can be given of the specific institutional
characteristics of the capitalist state. This means that purely
functional theories ofthe state are incorrect for two reasons. First,
because if we define the state in terms of functions, as Poulantzas
defines the state as the general factor for social cohesion, then we
cannot exclude certain aspects of capitalist society which are to be
seen as outside the state proper. We are forced to maintain that
anything which functions as the general factor of social cohesion is
part of the state, as Althusser has to argue about the institutions of
civil society which he identifies as ISAs. Second, such a functionalist
viewpoint fails to enable us to account for the distinctive character-
istics of the state as a set of institutional apparatuses. If we consider
Pashukanis's well-known question as to why the subordination of
one class by another takes the form of an impersonal mechanism of
public authority,9 then we must give an account of the very nature
of that 'public authority'.
(5) It is recognised that there is no single capitalist state, merely a
multiplicity of conflicting nation-states. Much discussion seems to be
premised on these being the capitalist state; and it being necessary to
devise a theory of that state. to Thus, when Marx claims that the
modern state is merely the executive committee for managing the
common affairs of the whole bourgeoisie, this cannot be strictly
correct. ll The whole bourgeoisie consists of capitalists located
within different nation-states, within a world system of conflicting
states. No such state can possibly manage all the 'common affairs' of
the whole bourgeoisie; merely of those sectors of capital located,
through either ownership or operation, within its territory. Further-
more, states do not simply coexist one alongside the other; any
particular nation-state exists within the context of other nation-states,
and in protecting the interests of 'its' capital. Nation-states repro-
duce, but in a different form, the essentially competitive nature of the
capitalist world economy.
(6) Finally, full recognition is given to the diversity of capitalist states,
and to the changes in state form and state policy. Thus, it is wrong to
 The State: A Critique 83

assume, as Poulantzas once seemed to, that the capitalist state is

paradigmatically a 'popular-democratic' state. It is also wrong to
argue that all capitalist state will ultimately become fascistic, that
this higher stage of capitalism will necessarily produce the same
state form.12 It is through the interposing of civil society that we
can see that similar capitalist economies do not necessarily produce
similar state forms. Indeed, any such state depends on a parti-
cular constellation of class forces, which mayor may not be sus-
tained.

I have thus suggested six criteria for an adequate theory of the capitalist
state: that it is based on the specific characteristics of capitalist societies;
that the state is not reduced to the economic, or seen as the instrument of
the dominant economic class; that the state is not conceptualised in
functional terms; that explanation is provided of its specific institutional
form; that theorisation is provided not of the capitalist state, but of
capitalist states and of their interrelations; and that it is recognised that
capitalist states are diverse and bear no straightforward one-to-one
relationship with the development of the capitalist economy.
The effect of employing these criteria is to render as partly otiose
certain of the contemporary debates on the nature of the capitalist state.
For example, Miliband claims that: 'the "ruling class" of capitalist
society is that class which owns and controls the means of production and
which is able, by virtue of the economic power thus conferred on it, to use
the state as its instrument for the domination of society'. 13 This clearly
contradicts criterion (2); his approach also contradicts criteria (5) and (6)
since he assumes that the capitalist state can always be defined
'instrumentally', that there is little difference between the differing
examples of the 'western system of power'. Poulantzas's position by
contrast is contradictory. On the one hand he claims that 'the state has the
particular function of constituting the factor of cohesion between the
levels ofa social formation' (italic in original). 14 This contradicts criteria
(3) and (4), and probably (1) as well. It is functionalist, in which cohesion
is effected only by the state, where anything which effects cohesion must
constitute a part of the state, and where it is presumed that the state will
always perpetuate the unity ofa given social formation. Yet, on the other
hand, he argues that a necessary feature of any social formation is a power
bloc, a contradictory unity of the politically dominant classes and
fractions under the protection of the hegemonic class, a protection that
never eliminates the struggle between these different constituent social
84 The Anatomy of Capitalist Societies

forces. The political class struggle is thus one in which the dominant
classes are organised into a power bloc, and where the dominated classes
are disorganised out of politics. This view of the state, although class
rather than capital-theoretical and therefore contradicting criterion (1),
has important implications explored in the following chapters. In
particular, it implies that the state cannot simply function so as to realise
the interests of capital. It is often unable to act, or it acts in ways
detrimental to capital. Poulantzas thus attempts to sustain functionalist
and a neo-Gramscian conceptions of the state simultaneously. 15
I will now discuss what I take to be the most significant set of literature
on the subject, that produced within the German state debate. This has
been particularly addressed to criterion (1), to the necessity of founding a
theory of the state on the specific characteristics of the CMP. More
specifically, they have considered how to derive the state from the nature
of capital. What is it about capital, of commodity-producing capital units
based on 'exploitation', which necessitates an official public
authority - that is, the state - to maintain such relations. I am going to
discuss the German literature, partly because of the range and sophisti-
cation of material they present. But it is also worth discussing because
this has come to be an influential tendency in Britain, significantly
because of the work of Holloway and Picciotto (hereafter HP).16 Initially,
I will consider their generally informative and helpful overview of the
literature in the Staatsableitung (state-derivation) debate in the Federal
Republic. The limitations identified in their 'Introduction' can then be
seen in more detail in the substantive contributions to the debate.
They argue, not unreasonably, that discussion on the state in Britain
had got stuck in the infertile rut of the Miliband/Poulantzas debate. 17 This
debate concealed major similarities, since both authors, according to HP,
focused on the political as an autonomous object of study. It is dubious
how true this is, but it is certainly the case that both tend to provide a
class-theoretical rather than a capital-theoretical orientation to the state.
HP go on to say that: 'neither Poulantzas nor Miliband denies the validity
of Marx's famous dictum that political forms can be understood only
when related to the "anatomy of civil society", but neither of them
considers it important to analyse this relation with greater precision' .18
However, it is unfortunate that HP do not examine this apparently
self-evident dictum with greater precision. Marx does not say that
political forms can only be understood when related to the anatomy of
civil society; that is, to political economy. He says that political forms can
only be understood when grasped 'neither from themselves nor from the
 The State: A Critique 85

so-called general development of the human mind, but rather have their
roots in the material conditions of life', that is, in 'civil society', the
anatomy of which is to be sought in political economy .19 Thus, although
the anatomy of civil society is to be sought in political economy, the state
and legal forms are firstly to be related to civil society, not directly and
straightforwardly to the nature of capital. Indeed HP's occasional
employment of the term 'civil society'is mysterious20 since as the text
proceeds no conceptual space is allocated to such relations. What
happens is they understand political forms entirely in the light of civil
society's 'anatomy' and not at all in terms of civil society itself. HP, in
their criticism of neo-Ricardians, or Sraffians, maintain that one can only
understand the social and political development of capitalist societies on
the basis of the specific form of capitalist class exploitation, which is
based on the extraction of surplus-value. 21 But that is a pretty empty
assertion on which many Marxists would agree. The crucial question is
exactly theform of the class struggle within a given civil society, how the
essential class relation of capital and labour, based on the production and
appropriation of surplus-value, is expressed within a given capitalist
society. To reduce all class struggle of the relation of form between
capital and labour is as erroneous as failing to relate civil society to its
anatomy at all.
Their argument proceeds as follows:

1. Marx sought to penetrate behind the categories of bourgeois political

economy to discover the social relations which they concealed.
2. In particular, he sought to show that the categories of exchange-value,
price, etc., are not 'objective external reality' but merely represent
historically determined forms assumed by social relations in
bourgeois society.
3. It is therefore necessary to show how 'historically and logically' these
various social relations are derived from the 'basic form' of capital
and value and the social relations they express. In particular, separate
realms of the economic and the political do not exist - they are mere
forms taken by the social relations of capitalist production. Political
study should not therefore be an attempt to develop an autonomous
political science, but should rather involve the deciphering of the
political categories as forms of capitalist social relations.
4. Thus, we are not involved in a question of causal determination; that
is, the way in which the economic base may determine the political
superstructure. We are involved in examining why the social relations
86 The Anatomy of Capitalist Societies

of capitalist society appear in separate forms as economic relations

and political relations.

They summarise their argument as follows:

the economic and the political are both forms of social relations, forms
assumed by the basic relation of capitalist society, the capital relation,
forms whose separate existence springs, both logically and histori-
cally, from the nature of that relation. 22

What then are some of the deficiencies of HP's 'German' position?

First, we might note that the criticism they make of Miliband and
Poulantzas is somewhat overstated. These two authors are castigated for
postulating the relative autonomy of politics and the state. Clearly
Poulantzas 's suggestion that the political is always relatively autonomous
is incorrect - but in the analysis of capitalist societies there does not seem
to be much difference between this concern with relative autonomy, and
the German attempt to derive the autonomisation of the political. Second,
HP seem to believe that because the 'economic' , which is presumably the
sphere of circulation, can be shown to be underlain by the social relations
of capitalist production, therefore all other forms in bourgeois society are
similarly underlain by such capitalist relations of production, that they are
all similar fetishised appearances. 23 But this does not follow. It may be
that the economic phenomenal forms (of exchange- value, price, etc.) are
specific and privileged appearances of capitalist relations. Simply
because such relations are themselves social does not mean that they give
rise to all the other social relations of a capitalist society. HP provide no
further reasons why the political forms are forms of appearance of
capitalist social relations. Indeed, would HP argue that all relations
within capitalist society are in fact forms of the capital relation. If not,
why not and what then makes the state particular? Third, HP do not show
that the relations between the state form and production are the same as,
or analogous to, that between circulation and production. In the latter,
there is separation and integration of the two spheres - economic forms
are not outside production but are integral to it. HP do not show that the
state is integrated into production in anything like the same manner.
Fourth, HP do not evade the problem of economic reductionism. This is
because, once we try to use their formulation in specific conjunctural
analysis, we cannot avoid seeing the state's policy and structure as
determined by capital. Their attempt to Hegelianise the relationship
 The State: A Critique 87

cannot be sustained once we leave the realm of abstract theoretical

formulations. Fifth, HP argue that the process of capital accumulation is
to be seen as one of class struggle. This takes place between capitalists as
a class and workers as a class and impresses itself on all the other social
relations of a cpitalist society. However, their approach is unjustifiably
essentialist - they view such class struggle as the unfolding of an
essence, of peeling away the layers, and of the emergence of a pure,
unvarnished revolutionary essence. To the extent that this has not yet
emerged this is because of the fetishisms of bourgeois society. I will
spend some further time in Chapter 9 criticising such essentialist notions
in relationship to the state and representative democracy.
Finally, we should also note that they confuse two different senses of
form and content. On the first version, the content gives rise to or
expresses different forms. The content is the essence, the real, which
yields or from which different forms are derived. On the latter
neo-Kantian version, form refers to the formal or structural properties,
while the content refers to the particular historical configuration given to
the form. In social theory, it is the difference between early Marx and
Simmel, the former as a quasi-Hegelian, the latter as a sociological
Kantian. In the middle sections of HP's article, the former sense
predominates, mainly under the influence of passages from Capital
reminiscent of the early Marx. In the later sections the latter sense is
dominant; HP say that 'form analysis' may well at times be 'too formal
and too abstract' and that what is needed for a proper Wlderstanding of the
state is analysis of 'the dialectic of the form and content of class
struggle'.24 The latter, the 'content', consists of 'concrete' analyses of
history, of conjunctures, to fill out the unfortunately 'formal' character of
the analysis of form. They say elsewhere that:

What is required for an anal ysis of restructuring is thus not just a formal
analysis of capital, nor an empirical analysis of the course of class
struggles, but an analysis which embraces both movements, which
tries to understand social development in the dialectical interaction of
the form and content of class struggle. 25

Two points should be made. First, the sense of content is here different
from that employed earlier where it is the content that gives rise to the
forms, the forms being an expression of content. Second, content here
refers to the real, the empirical, as though the facts of history can be
grafted on to the analysis of form, which is now taken to refer to the
88 The Anatomy of Capitalist Societies

structure. So form and content have here come to mean structure and
history, or synchrony and diachrony, or even theory and the real. But,
unfortunately, the understanding of the dialectic of form and content is
rendered impossible because HP do not theorise their relation, of how
form 'determines' or 'structures' content, indeed whether form refers to
the 'theoretical', and content to the 'observational'. Are form and content
anyway ontological categories, or epistemological categories and what
relationship do they bear one to another?
I have taken HP's article to suggest that the German state literature has
not succeeded in solving many ofthe questions which it had posed. I will
now consider some of these writings. However, I cannot do justice to
them; and generally I will only make reference to a few points that they
contain. My main purpose is not exegetical. It is rather to identify certain
central issues relevant to my own discussion of the state below.
I will begin by considering the seminal article by Millier and
Neustiss,26 itself intended as a critique of some of the earlier writings of
Offe and Habermas. 27 They make a number of relevant points. First,
Miiller and Neusiiss consider the presuppositions of revisionist theories
of the state; in particular, the conception of the state as an institution
relatively independent of societal contradictions. Revisionism is pre-
mised on the assumption that the state can comprehensively and
consciously regulate the laws of motion of capitalism and of their
consequences. Underlying this is the assumption that the sphere of
distribution is absolutely autonomous from the sphere of production. It is
seen not as a necessary moment of the process of production but as an area
of politically determined state activity. State action is seen as purely
political, and it is possible to found a 'state socialism'. Millier and
Neusiiss criticise this on a number of grounds: that the sphere of
distribution cannot be isolated in this manner since it is a necessary
moment in the capitalist production process; that it suggests that the state
can itself somehow introduce socialism rather than relying on the
'working class '; and that it implies that the state is neutral and not in fact
riddled with social contradictions structured by the capital relation.
Second, they argue that the state must not been seen purely as a
'distributor', since this is one of the illusions of state socialism. For
Miiller and Neusiiss the state is not independent of society, it is from
society and its contradictory social relations that the form of the state is
derived. In relation to Marx's analysis of the Factory Acts, they criticise
the revisionist interpretation of them as representing the gradual
advancement of socialist principles in a capitalist society. 28 They point
 The State: A Critique 89

out rather how such Acts had two highly functional consequences for
British capital. First, the development ofthe Acts prevented individual
capitalists from destroying the bases of their own activity; namely,
refreshed and energetic labour-power. Second, the restrictions placed on
the length of the working day meant that production of relative rather than
absolute surplus-value assumed dominance. How was it that these
consequences resulted? They followed from two features of capitalist
social relations. First, capital consists of individual capitalists competing
with each other, so that each is encouraged to work their workers as long
and as intensively as possible. Second, capitalist relations produce
similarly placed wage-labourers who are literally forced to defend
themselves as a class. Factory legislation then is dragged out of the state
through a more or less open class struggle between the capitalist and the
working classes. As Marx says:

They curb the passions of capital for a limitless draining of labour-

power, by forcibly limiting the working-day by state regulations, made
by a state that is ruled by capitalist and landlord. 29

The maintenance of the CMP hence requires the organisation of workers

as a class, since they would not otherwise be able to safeguard their
continued existence as individualised sellers of the commodity labour-
power. Thus, the development of the state and the realisation of its social
and political functions can only be achieved through the effects of either
actual or threatened working-class struggle. But, also, such struggles
have the effect of reinforcing the organisation of workers as a class with a
'tendency to overcome capitalism and its state'. 30 Gass struggle.
therefore, both 'maintains the system', and strengthens the organised
power of the working class.
Third, the state in capitalism functions to ensure the reproduction of
labour-power. This is necessary because of the nature of the capital
relation, which is to exploit to the limit the labour-power that each capital
has purchased. Thus, the state, under pressure from the organised
resistance of the working class, produces compulsory legislation, powers
of surveillance and a growing system of bureaucratic enforcement. Such
restructuring of the state is necessary so that capital can be effectively
controlled to prevent individual labourers being exhausted through an
excess oflabour. Thus, the development ofthe capitalist state, at least in
its legislative and bureaucratic aspects, results neither from some general
requirements associated with the development of a complex economy,
90 The Anatomy of Capitalist Societies

nor solely from the need to control the working class more effectively. It
results from the need for the state to intervene in the immanent laws of
motion of capitalist production, to contradict the immediate movement of
capital, in order to safeguard capital's general interest.
Finally, to return to the first point, the effect of these struggles is to
create an illusion among the workers and their allies; namely, that the
state is separate from capital, is neutral with respect to all social classes,
and hence can be employed against capital by labour. Thus, the bases for
reformist politics are systematically reproduced within the day-to-day
experience of working-class struggle and the form in which the state is
confronted. 34
Assessment of this long, wide-ranging and contentious article is
difficult. I want to make one main point; namely, that their position is
contradictory and that this means that much of their vituperation at Offe
and others is misplaced. What their interesting account of the Factory
Acts shows is that the 'illusions' of 'state socialism' (neither term is
appropriate, incidentally) follow from working-class struggle and ex-
perience. As a result of defending itself the working class does see the
state as neutral because that is how it acts. The state does act against
individual capitals, and this does benefit each individual possessor of
labour-power. They do consider how this simultaneously strengthens the
organisation and effectiveness of the working class but it does not follow
that this increases its 'tendency to overcome capitalism and the state'. It
may simply be that this increases its effectiveness in further struggles to
use the state against individual capitals. Indeed they imply that the further
this process occurs, the more the working class used the state against
capital, the more this will produce as an effect, further extension and
autonomisation of the state. And hence it is less likely that it will be
possible to transcend the illusion of state socialism, and to overcome
capitalism and the capitalist state. MUller and Neususs are admirably
clear on the need to ground political organisation within the 'necessary
struggles of the proletariat'.32 Unfortunately for their haranguing of
'reformists' and 'revisionists', they provide little reason for seeing why
those necessary struggles will ever be non-reformist. Of course they are
sometimes, but MUller and Neususs fail to indicate why this is the case,
and indeed their argument that the Factory Acts were highly functional
for capital in effecting the dominance of relative surplus-value produc-
tion, suggests that most, or all, class struggle will somehow be
incorporated, albeit in a contradictory manner, into the functioning of the
capital relation.
 The State: A Critique 91

One of the very strong virtues of their argument is their claim that the
autonomisation of the state does not result from some automatic
mechanism in the capitalist system. For them it follows from specific
class practice. However, their formulation of autonomisation is some-
what underdeveloped. I will now discuss certain of the later contributions
to the German debate considering only their work in relationship to this
specific issue. Do they present an adequate theory by which the state is
derived from capital? Is this related in coherent fashion to the struggles of
different classes in the manner that MUller and Neususs suggest?
There are four different approaches that we can identify.33 In the first,
the state form is derived from the sphere of circulation; in the second,
from the crisis-ridden character oflater capitalism, in the third, from the
nature of capital as individual capital-units; and in the fourth, from the
capital relation as one of class domination. The first two approaches are
fairly obviously limited; the latter two have more to recommend them.
Flatow and Huisken have most famously elaborated the thesis that we
can derive the state form from the surface of capitalist society, from the
realm of 'freedom, equality, property and Bentham' .34 It is within this
sphere that there is no contradiction between particular and general
interests. All the subjects engaged in exchange have a common interest,
an 'identity of interest' , in maintaining their source of revenue in as high
and as continuous a form as possible. The common interest is shared by
all property-owners, ofland, of capital and oflabour-power. Given that
this common interest cannot be sustained by anyone separate property-
owner, it is the state which is necessitated by such exchange relations, so
as to ensure the common interest shared by each property-owner in
guaranteeing the continued conditions for exchange. Having reached a
'general derivation' of the state form, they then consider why particular
demands arising from society acquire the status of a general interest and
are implemented through the state. It is these which make up the specific
functions of the capitalist state.
It should be clear that this is related to the conception of the state which
I labelled earlier as that of mystification. Marx and Engels, in The
German Ideology, maintain that 'out of this very contradiction between
the interest of the individual and that ofthe community the latter takes an
independent form as the state'. 35 However, Flatow and Huisken's
conception fails to meet many of the criteria of an adequate theory I
outlined earlier. The very existence of a sphere of circulation, in which
relations of equality, etc., are produced, depends in capitalism, on
another sphere, that of production, where relations are of course anything
92 The Anatomy of Capitalist Societies

but equal. The 'general interest' established on the surface is highly

limited; indeed there is no such entity as the general interest, merely
specific interests claimed to be general. Hence, it becomes impossible to
deduce the state as such from the general interest since the general interest
only exists as ideology or mystification, concealing the specific interests,
'the very contradiction between the interest ofthe individual and that of
the community' (my italic) which lie behind. One might note also their
tendency to reify, to treat the three separate property-owners as
equivalent. This disregards their highly differing relations of different
classes to the underlying system of production, and the manner in which
their distributional struggles are thereby structured. Their theory there-
fore entirely abstracts from the specific struggles of differing social class,
and the effects so produced.
A second approach is to derive the state form from the crisis-ridden
nature of the capitalist economy. 36 Because the structurally uncontrolled
operation ofthe law of value has the effect of producing larger and more
persistent crises, it is necessary that there is development of the state. On
the one hand, it functions to administer needs which are not satisfied
through the existing pattern of production. On the other hand, it functions
as a means of political domination to ensure that potential class struggle is
suppressed. Thus, the nature of the state in both its bureaucratic-
administrative and repressive - class dominant aspects is accounted for
in this interpretation. However, it is a highly partial derivation: it fails to
explain why the normal processes of competition and crisis itself come to
be inadequate; it treats the state mainly in terms of political domination
and ignores its less specific functions in maintaining the general
conditions of production: it cannot explain aspects of the state such as the
nature of bourgeois law or representative democracy; and it cannot
account for the form of state development prior to that period in which the
pattern of capital accumulation became markedly crisis-ridden. As a
general principle for a theory of the state it clearly will not do; although
later we shall come back to this issue in trying to understand recent forms
of state interventionism. .
Both these first two approaches to the derivation of the state suffer a
major disadvantage. They do not derive the form of the state from the
social relations of capitalist commodity production itself, but rather from
certain appearances, in the first, from the 'equal' relations of exchange, in
the second, from the exceptional relations of crisis. I will now consider
two approaches which do attempt to derive the state form from the social
relations of commodity production.
 The State: A Critique 93

First of all, let us consider Altvater. 37 Capital, he points out, is

constituted of many different capital units, the relations between them
being unplanned, anarchic, antagonistic and regulated by the laws of
competition. As an effect ofthese competitive relations, the immanent,
inexorable laws of capitalist development realise themselves. However,
the production of certain use-values cannot be rendered profitable, and so
capital cannot be reproduced solely through the actions of many different
capital units each competing one with another. Certain social 'functions'
will not be produced by these individual capital units. Hence, there has to
be an alternative structure which provides certain necessary conditions of
existence of capital; a structure which is not subject to the limitations of
capital in that surplus-value must be produced and realised. The state is
the institution which meets the general interests of capital because it is not
itself a 'real-capitalist'. It is not itself subject to the law of value, and it
can ensure certain conditions necessary for capital accumulation, some of
which would be destroyed through competition between individual
capital units. Altvater says that the state performs these functions
necessary for the maintenance of capitalist society. And it can achieve
this because the state is a 'special institution, outside and above bourgeois
society'.38
The four functions which capital units cannot perform are the creation
of the general material conditions of production, the determination and
safeguarding of the legal system, the regulation of conflict between
capital and wage-labour, and the assurance and expansion of total
national capital on the world market. To illustrate these let us consider the
first. Altvater argues that the state takes over these material processes
which cannot be operated on a capitalist basis. He thus employs a 'deficit'
theory, the state taking over and running in a non-capitalist fashion the
production of necessary use-values. The state necessarily moves into the
'vacuum' remaining when capital units no longer find production
profitable. There has been a long-term increase in the size of the state
sector, partly because of the historical tendency for the rate of profit to
fall, and partly because of the growing level of the productive forces.
There are many difficulties with this account. As an explanation of the
pattern of nationalised industries it suffers from the following errors. It is
simply not the case that the state takes over all there material processes
which cannot be operated on a capitalist basis - many are allowed to go
to the wall and the use-values are imported. Furthermore, it is not only
unprofitable industries that become nationalised. Also, the very notion of
unprofitability is itself problematic since the state itself is engaged partly
94 The Anatomy of Capitalist Societies

in determining wage- and price-levels, and hence in the formation of the

average rate of profit. It is too simple to suggest, simply as an effect of the
tendency for the rate of profit to fall, that the consequently 'unprofitable'
units will get nationalised. It all depends, significantly, on the strength,
organisation and cohesiveness of the various social classes struggling in
that particular society.
Indeed, there is a more general criticism here: Altvater's derivation of
the state form is highly functionalist. He argues:

(1) There are these four functional prerequisites which must be met for
any capitalist society to survive.
(2) They will not be met by individual surplus-value making capital
units.
hence
(3) There develops the state which is that special institution which
meets these four functional prerequisites,
and
(4) The state is defined and explained in functional terms; it is that
institution which meets these four functional problems.

Oearly, then, Altvater's approach fails to meet criterion (3) above,

that a theory of the state should not be functionalist. There are a number of
other related difficulties. First, Altvater provides no explanation of why
and how the state develops. It simply developes as a capitalist state in an
apparently automatic fashion, responding to each functional demand
emerging from within 'society'. It is also incorrect to suggest that the
state maintains all functions necessary to maintain capitalist society. This
view leads to an over-extension of the concept of the state to embrace
many relations which are not distinctive to the state itself, for example,
the family. Furthermore, Altvater takes class conflict to occur only within
capitalist society, and that the state comes along to regulate such conflict
from outside. But this ignores how such conflict occurs also within the
state, it is itself a major arena of class struggle. As a result, the state is
often unable to carry out its functions successfully. The contradictions of
capitalist accumulation are not only reproduced within society but also
within the state itself.
In the next chapter I will refer to another article which adopts a position
partly similar to Altvater, that by Blanke, Jurgens and Kastendiek. 39 For
the moment let us consider certain of Hirsch's writings, which provide
examples of the derivation of the state form from the class relation of
 The State: A Critique 95

capital and wage-Iabour. 40 The state form is not derived from the nature
of capital as so many different capital units, but from the historically
changing nature of the relationship between capital and wage-labour.
The appropriation of surplus-value in capitalism does not depend on
direct relations of force or coercion. It depends rather on the hidden laws
of reproduction which operate behind the backs of the bearers of such
laws. It is essential for capitalist relations that all barriers which prevents
the unimpeded circulation of commodities based on the principle of equal
exchange are abolished. Hence, it is necessary that the direct producers
are deprived of control of the means of force; and further, that this is
localised in a social instance raised above the economic process of
reproduction. Thus, Hirsch maintains that both the creation of formal
bourgeois freedom and equality, and a state monopoly of force, follow
from the very nature of commodity production. The nature of capitalist
rule means that the 'ruling class' must grant to the body which secures its
domination an existence formally separate from it. As Pashukanis says:
'the principle of competition prevailing in the bourgeois-capitalist world
does not allow any possibility of linking political power to the individual
economic enterprise'. 41 Two consequences follow from the emergence of
such a centralised state apparatus: first, the suppression of the various
'feudal' constraints and relations of dependence which penneate society;
and second, the separation of the political apparatus of bourgeois society
from real individual and common interests. The state moreover expresses
a particular historical form of class rule, and what is crucial is to analyse
precisely the changing nature of that rule. The developments in the form
of the state result from the changing character ofthe capitalist production
process, that process being seen as one of essentially class struggle. The
specific activities and measures of the state develop, not because of the
abstract logical development of capitalist relations, but from the specific
state of class struggle and of the effectivity of political movements. The
state is therefore a system of centralised coercive domination which seeks
to guarantee the accumulation process, to maintain existing relations of
production and to force through developments in the labour process. The
process by which the state reorganises, or restructures the process of ac-
cumulation, to mobilise the counter-tendencies to the fall in the rate of
profit, is a process of class struggle. Its nature, and the resulting form of
the state, historically develops and changes.
There are four other important aspects to note. First, Hirsch empha-
sises the 'imperfection, incompleteness and inconsistency of state
activity'42 as a result of the general conflict and collision of opposing
96 The Anatomy of Capitalist Societies

political forces. The state does not represent the institutionalisation of the
general interest. Second, the state cannot overcome the contradictions
inherent in the accumulation process, its continued existence as a
particular form of social relations depends on the reproduction of the
capital relation, and its very policies may well exacerbate such
contradictions. Thus, the growing pattern of state intervention to regulate
the reproduction process of capital will itself intensify social conflict,
between wage-labour and capital, and between fractions of capital.
Third, the state does not simply orchestrate contradictions from
outside - it is itself ridden with such contradictions which are reproduced
in and between the different apparatuses which constitute the state.
Finally, although one can establish that the state will provide certain of
the general material conditions of production , one cannot determine what
must be the nature of such infrastructural provision at any particular
moment, nor whether that need will in fact be provided. It depends on a
complex range of determinants, which Hirsch summarises as the
penetration and development of capital at the time. Much more in his
work could be referred to, particularly his analysis of the changes in the
form of the state in modem, monopolistic forms of capitalism and the
heightened nature of the class struggle.
I will now make some critical comments. His argument proceeds:

(1) Capitalism is a system in which surplus-value is produced and

appropriated, within the spheres of production and circulation,
without direct relations of force, dependence or restraint.
(2) The process of surplus-value production and appropriation must be
sustained by relations of force.
Thus
(3) Such relations of force must be situated within an institution separate
from the relations of commodity production; that is, within the state.

This brings out clearly a number of difficulties. First, his argument does
not demonstrate why the relations of force cannot be exercised within
each capital unit. Although clearly relations within the sphere of
circulation should be characterised by an absence of direct force,
dependence and restraint, it does not follow 'logically' or 'historically'
that relations within production should be so characterised. And indeed
we know that they are not 'historically'. Moreover, it does not follow
from the fact that direct control should be absent from capitalist relations,
that there will in fact be such an absence. Hirsch does not provide an
 The State: A Critique 97

account of the mechanism which produces as an effect such an absence,

as well as the corresponding concentration of force within the state. His
argument is teleological - that capitalism is to be characterised in a
certain manner, as having a particular end-state. One can then see any
development - for example, that ofthe bourgeois state - as brought into
being by, and contributing to, that process which yields the particular
end-state.
A further problem concerns the nature of the state itself. Hirsch
emphasises its essentially coercive nature, by contrast with say Altvater's
functionalist position. Yet clearly the state is not purely coercive, and
indeed much of Hirsch's substantive argument demonstrates exactly the
forms of, and limits to, various kinds of economic intervention. So we
might consider what relationship holds between these different aspects.
Does the derivation of the form of the state, based on its monopoly of
force, lead to an understanding of the other aspects of state policy? How
do they relate together? Furthermore, the concentration on the monopoly
of force suggests that the state is relatively unified, that it has a unity
based on this monopoly. (I assume that he means a monopoly of the
'legitimate' means offorce.) Yet, at the end of The State Apparatus and
Social Reproduction' Hirsch maintains that the state is heterogeneous and
increasingly chaotic, and indeed that this is functional for its ability to
maintain complex relations with various class, strata and fractions. The
state, he says, falls apart into a conglomerate of 'relatively unconnected
part-bureaucracies'. Two points need making here. First, if this is the
case, then it is doubtful whether it can be claimed that there is a unity of
the state let alone one based on the monopoly of force. One wonders how
Hirsch can know that these relatively unconnected part-bureaucracies are
in fact part of the state. Second, Hirsch maintains that this development
occurs so that the state can manage complex relations with various
classes, strata and fractions 'which are the conditions of its ability to
function as guarantor ofthe domination of the bourgeoisie'.43 Indeed he
suggests that the state must relate to and support itself on competing
individual capitals having, under the conditions on the world market,
extraordinarily different valorisation interests. Yet the rest of his
argument had tried to show the overwhelming centrality of the relation-
ship between wage-labour and capital, in which the struggles of
individual capitals, class fractions and other classes and strata had been
almost irrelevant. Regrettably, Hirsch does not theorise all these sets of
relations; he, like his British followers, tending to employ the capital/
state couple.
98 The Anatomy of Capitalist Societies

Thus he does not theorise the relationship between, on one hand, the
exploitati ve relations between capital and wage-labour, and, on the other,
the multiplicity of relations between those individuals, classes, strata and
fractions that constitute civil society. As I will hope to demonstrate it is
only by considering that relationship systematically that we begin to make
full sense of the capitalist state. Many of the extremely perceptive
contributions made within the state-derivation debate can then be located,
not in relationship simply to 'capital' but in the more complex relations in
which civil society stands between the economy and the state. This I will
now attempt to show.
7
The State: Some Proposals

In 1844 Marx summarised the relationship between state and civil

society:

[The state] is based on the contradiction between public and private

life, between universal and particular interests. For this reason, the
state must confine itself toformal, negative activities, since the scope
of its own power comes to an end at the very point where civil life and
work begin. Indeed, when we consider the consequences arising from
the asocial nature of civil life, of private property, of trade , of industry,
of the mutual plundering that goes on between the various groups in
social life, it becomes clear that the law of nature governing the
administration is impotence. For, the fragmentation, the depravity and
the slavery of civil society is the natural foundation of the modern
state .... If the modem state desired to abolish the impotence of its
administration it would have to abolish contemporary private life. And
to abolish private life it would have to abolish itself, since it exists only
as the antithesis of private life.!

Although there are a number of deficiencies in this argument, it does have

one important strength. Marx shows that it is essential to relate the state to
the overall character of civil society, and not just to the relations of
production. It is within civil society that various limitations are placed on
the state, that it is substantially impotent t(' act in various ways and that
this stems from 'civil life' and the 'mutual plundering' that occurs
between the various social groups. It is clear from this that civil society
embraces both the relations between individual subjects, its 'fragmen-
tation' and the relations between contending 'groups in social life '. The
capitalist state is to be related to these social relations of civil society in
100 The Anatomy of Capitalist Societies

two ways. First, they substantially determine the limits, boundaries and
content of the state, of changes in its form, functioning and policy. The
state thus acts within the context provided by the contending relations
between individual subjects and wider social groupings. It is never
simply omnipotent, nor does it simply react to the demands and needs of
the economy. Second, the state is to be viewed not merely in terms of its
reaction to the social relations constitutive of civil society. Rather the
state is itself to be seen as formed actively, in seeking to establish and
sustain a particular constellation of social forces. However, the estab-
lishment and maintenance of such a power bloc is never achieved once
and for all; and where such a bloc is not sustained the state may often be
unable to identify or to pursue the policy that is most appropriate from the
viewpoint of the 'economy'. In the rest of this chapter I shall attempt to
elucidate these two aspects of the relations between the state and civil
society.
Let me begin by recapitulating the criteria for an adequate theory of the
capitalist state: it should be based on the specific characteristics of
capitalist social formations; it should not be reductionist or instrumen-
talist or functionalist; it should explain the state's specific institutional
form; and it should not be assumed that there is merely one capitalist state
or that there will be no substantial transformations in the form of any such
state. It is, however, clear that the first two criteria have posed major
problems for theorists of the state. It has proved to be difficult to relate the
state to the specific characteristics of capitalist social formations without
that leading to reductionist or instrumentalist interpretations of that state.
In the last chapter I have discussed some of the essentially reductionist
formulations produced by the German theorists who have attempted to
derive the form of the state from the nature of capital. I want now to
suggest further reasons why their arguments are generally deficient.
First, they contradict the first criterion because they relate the
state - or more precisely derive the state form - from capital. What they
ignore, theoretically at least, is the fact that capitalism comprises both the
relations embodied in the nature of capital and the relations of civil
society. It is impermissible to derive the state simply from capital,
because that is to neglect the determination of the state by the relations of
civil society. Second, they present a form of rationalist argument in
which it is presumed that the logical relations between certain concepts
within discourse yield us knowledge ofthe real relations between objects
outside discourse. In Chapter 1 we saw certain objections to this kind of
argument, and that it is necessary to formulate concepts which are not
 The State: Some Proposals 101

merely logically related; for example, that the concept of the state form
can be logically derived from the concept of capital. It is necessary to
formulate concepts which refer to various kinds of social grouping and to
their characteristic forms of struggle. It is only then that we can identify
the mechanisms which may ensure that the logical relations between
certain concepts will yield us knowledge of the relations between the real
objects outside that discourse. It is identification of the mechanism of
working-class struggle which makes Millier and Neusiiss's paper so
interesting. They show that the development of capitalist relations has the
effect of producing similarly located wage-labourers who are literally
forced to defend themselves as a class. Factory legislation then is dragged
out of the state through working-class pressure. But the effect of such
pressure is that one essential precondition of capitalist production,
labour-power, is thereby reproduced. What is crucially important in their
argument is that they do not merely derive the role of the state in
reproducing labour-power from the nature of capitalist relations. They
seek to show that there is a particular mechanism, of working-class
struggle and its bases, and that this produces through state action both the
reproduction oflabour-power and a strengthening ofthe organised power
of popular social forces. They thus demonstrate the deficiency of the
rationalist arguments of many of the other German theorists.
I will now try to elaborate on their approach by considering the
relationship between capitalist relations of production, civil society and
the state. I shall try to show that the specific nature of the capitalist state
cannot be directly derived from the capitalist relations of production; but
also that such relations do determine the general nature of the state. Let
me try to resolve this apparent paradox.
On the one hand, the capitalist state (bearing in mind the multiplicity of
such states) possesses a form which is given by its attempt to sustain the
overall conditions under which profitable accumulation can take place
within its national territory. This demand on each state results from the
structure within which it is situated. It cannot avoid attempting to sustain
such conditions, although each state will vary in both its internal structure
and in the policies it happens to pursue. So minimally each state is located
within a structure in which it attempts to provide and to sustain those
conditions appropriate for profitable capital accumulation. In this sense,
but only in this sense, is the state to be related to the nature of capital and
the changing requirements of the accumulation process.
Yet, on the other hand, the state embodies a set of social relations
which are distinct from the process of production itself. And interposed
102 The Anatomy of Capitalist Societies

between production and the state are the multifarious relations of civil
society. Thus, each capitalist state, in its attempt to sustain the changing
conditions for profitable accumulation, in fact operates within the
heterogeneous relations which comprise civil society. It is this contradic-
tory location which is central to our understanding of the state. In its
attempt to guarantee accumulation it seeks to organise, legislate and
orchestrate the diverse relations of civil society. This means that it is
incorrect to derive the state directly from capital, but it is also incorrect to
relate the state only to civil society, since, as I have suggested, its overall
form is given by its attempt to provide and sustain the overall conditions
appropriate for profitable capital accumulation.
Let us now consider those aspects of civil society which were identified
in Chapter 4: first, the sphere of circulation and the activities of subjects in
exchange; second, the social relations within which labour-power is
reproduced economically, biologically and culturally; and third, social
classes and popular democratic forces. It was also noted that these
practices do not form a unity, and each bears different relations both to the
state and to the relations of production. I will argue that the form of each
capitalist state is given by its relationship to these three aspects of civil
society. In its attempt to sustain profitable capital accumulation the state
is forced into particular relations with the agents of change and the sphere
of circulation, with the reproduction of labour-power and with social
classes and popular democratic forces. It is its relationship to these three
aspects of civil society simultaneously that account for the form taken by
the state, that it is a unified set of institutions based on the centralisation of
the means of violence within a specified territory, it passes and
implements generally applicable laws, it administers bureaucratically the
interests of those citizens resident within that territory, and it devises and
pursues policy through forming and sustaining a power bloc out of the
contending social classes and other social forces. 2 I will discuss each of
these briefly before proceeding to discuss why this form of state is to be
seen as res ul ting from the three aspects of civil society.
First, then, state power is predicated by the monopoly of centralised
physical coercion within the national territory. This is not to argue that
state power is purely coercive. Indeed it is most effective when it is not
simply physical; when, for example, citizens willingly and self-
consciously obey the generally agreed on laws and regulations of that
state. However, a system of laws has always to be guaranteed by bodies
of armed men able and willing to enforce the law, to catch and punish
transgressors. So it is this monopoly of physical coercion which lies
 The State: Some Proposals 103

behind the system of laws and which sustains its effectiveness. Two
further points should be made. First, the state itself transgresses the law
on many occasions, at which point higher-level justifications are brought
into play. However, in such situations the state will be put under closer
scrutiny and challenge, and may need to employ further the 'repressive
state apparatus' to support its transgression. Second, much Marxist
literature assumes that Lenin's 'special bodies of armed men'3 mainly
function to sustain legally sanctioned relations within the national
territory. This is misleading since much of the means of physical coercion
held by a state in fact sustain the external relations of that nation-state,
especially through deterring other nation-states from attack. It is clear
that the role of the state in the twentieth century has been significantly
enhanced because of its centrality in the organisation of warfare. The
relationship of nation-states to the organisation of warfare is clearly of
great importance and needs to be explored in much more detail.
The second feature of the state has already been referred to, namely its
unity. In theorising the state it is necessary to argue both that it has a unity
and to indicate its basis. If we do not do this then we cannot distinguish
between it and civil society. And once we cannot distinguish between the
two, albeit problematically, then the argument collapses into a sociologi-
cal institutionalism. 4 So if it is important to distinguish between the state
and civil society, and, if the latter has a unity, what is its basis? As should
be clear by now, the unity of the state lies in its monopoly of physical
coercion, both internally, and externally vis-a-vis other nation-states.
Thus, the state comprises those institutions within a capitalist society
which directly and interdependently are based upon the centralised
organisation of physical coercion. This may be a direct dependence, as in
the case of the legislature, whose laws are implemented through the
application of, or threat of application of, sanctions of the police and
judiciary. Or it may be an indirect dependence, where the state activities
rest upon the forces of centralised coercion in regulating in the
background, entry to and discharge from those institutions; and in
ensuring sufficient taxation revenue to finance them. What in fact ensures
that a particular apparatus is part of the state is that it is located in a
particular structured relation with the other state apparatuses, of the
national and local executive, legislature and judiciary;5 of the police,
army and security services; of the educational, welfare and health
services; and of the various apparatuses of economic administration and
organisation. The overall effect ofthese relations is that the state 'rules',
it passes laws and ensures a degree of compliance, it raises taxes, it
104 The Anatomy of Capitalist Societies

implements policy, it fights wars and so on. And it is able to do all this
because underlying its complex and disparate activities and apparatuses is
the centralised concentration of coercive power. It is not of course that
individuals and indeed social groups may not possess some means of
physical coercion (the ownership of guns in the United States, wide-
spread patterns of domestic violence 6 ), but merely that in contemporary
capitalist societies (as well as those of Eastern Europe, etc.) there is an
exceptional concentration of organised coercion within the state, how-
ever disguised this may be. Finally, it is important to note how this
coercive power is not diffused throughout the state, as in Absolutism, but
is concentrated within specific apparatuses.
The third feature of the state is that is passes and implements generally
applicable laws. Very little of the activities of the state are free from
lawful direction - and those themselves must be specified in legal terms.
Thus, the state is centrally involved in passing and implementing laws
which relate both to itself and to the citizens born and/or living within its
territory. These are general laws, and, with certain exceptions, they do
not apply to specific classes or ethnic or religious groups. They regulate
and direct the form of the interactions between individuals and various
other legally constituted bodies. The construction of such a system oflaw
has the effect of depersonalising power, of making the exercise of power
flow from the implementation of a set of general laws . Thus citizens obey
the law rather than obeying specific individuals or groups or classes.
Further, the state produces through the law individual juridic subjects
possessing certain justiciable rights; and also certain general categories,
for example, of landowners or minors or consumers who each possess
different kinds of juridical right. One effect 6fthe law is to fragment class
relations. The state constitutes and interacts with individual juridic
subjects, as well as with the general categories of legal subject.
Fourth, the activities of the state are organised bureaucratically. State
policy is implemented through the application of predetermined rules by
hierarchically organised bureaucrats. The tasks that each performs are
closely specified and there is little possibility of innovative action.
Bureaucratic officials are supposedly 'neutral' having little or no personal
or class involvement. They neutrally decide on the issues and administer
the policies appropriate within their specified sphere of competence.
Bureaucratic officials are paid by salaries, they are appointed and not
elected, their work within the bureaucracy constitutes a career, and this
work is divorced from their private life. Many of these features are of
course well known. There are, however, some contentious issues. First of
 The State: Some Proposals 105

all, we might consider the degree to which such bureaucratic regulation is

technically the most efficient, especially in the later monopoly forms of
capitalism. 7 Further, it is unclear the degree to which bureaucrats can and
will act as a relatively autonomous social category.s Also, in modern
capitalism it appears that the bureacuracy increasingly dominates the
legislature; we might consider both the causes of this and its consequ-
ences for the class struggle. 9 And there is the issue as to the degree to
which the 'administration of things' in advanced societies is necessarily
bureaucratic or whether this is tied to specific social relations of
production. \0 Certain of these issues are addressed below.
Fifth, the state operates within the context of the contending relations
between classes, class fractions and popular democratic forces. The
actions of the state can neither eliminate the contradictions of capitalism,
nor can it act independently of them. The effecting of a major change in
the state form or policy always involves the establishment of at least a
temporary power bloc out of the politically dominant social forces. The
establishment of such a bloc is problematic, involving balancing out,
manipulating, coercing, compromising and bargaining between the
different classes, fractions and popular forces. This means that it is never a
question simply of the capitalist class running the state, that it is simply
the ruling class, or a capitalists' state. Nor is it correct to consider that the
state is always able to implement the 'correct' policy from the viewpoint
of capital accumulation; indeed the state may often implement the
'wrong' policy, or the 'right' policy but atthe wrong time. The state must
not be viewed as automatically reacting to the demands of capital
accumulation. Indeed, for substantial periods there may be no power bloc
established at all; merely a number of politically dominant classes,
fractions and social forces with no particular organisation, unity or
policy .11 The establishment of a power bloc is always problematic, and it
is a necessary condition for the securing of major transformations in state
form and policy.
This concentration upon the formation and reproduction of the power
bloc does not mean that I am advocating a purely 'fractionalist'
position.1 2 It is now argued that there is in fact a major divergence
between 'fundamentalists', and 'Poulantzians' or 'fractionalists'. The
former emphasise that it is capitalists as a class that exploit workers as a
class, and that this relationship impresses itself upon all specific
capitalists and workers. For the 'fundamentalist' Marxist the struggles
between different fractions of capital, or the problems of cementing a
power bloc out of various classes, fractions and forces, are subordinate to
106 The Anatomy of Capitalist Societies

the overall detennining struggle between capital and wage-labour to

produce and realise surplus-value, and the role of the state in maintaining
this dominance of capital.
It is clear that 'fundamentalism' is in part correct; namely, that the
dominant relations in a capitalist society are those within the economy,
between capital and wage-labour. As we have seen, this has the effect of
producing within civil society, the class struggle between capitalists as a
class and workers as a class. It also produces the essential form of the
state, which is given by its functioning to maintain the overall profitable
accumulation of capital. The objective of politics is thus given by the
struggle over the dominant mode of production, between capitalism and
socialism. However, two points of reservation have to be made. First,
capitalists and workers are categories of subject produced within civil
society and such classes are fissured by the bases of structuration present
within each civil society. There are no pure classes of capitalists and
workers existing outside the forms of differentiation effective within civil
society. And further within civil society there are other 'classes-in-
struggle' and popular democratic forces and struggles.
Second, unlike the pre-Absolutist feudal state which locally sustained
direct payment of feudal rent to the lord, the capitalist state does not
function directly in the relations between capital and wage-labour. The
state operates within civil society and especially within those sites where
individual subjects, popular forces and social classes struggle to control
and to transfonn their conditions of reproduction within the sphere of
circulation, within the relations of distribution, exchange and consump-
tion. Thus, the state in capitalism is to be viewed as structurally
detennined to ensure the overall conditions for profitable capital
accumulation, but it does so, and can only do so, within the conflicting
and multifarious relations constituted by civil society. In particular, it has
to, but may fail to, orchestrate a power bloc out of the diverse social
elements that makes up any particular civil society. I will now try to show
that the form of the state is given by the distinctive features of civil
society, and indeed that the different form taken by capitalist states is
related to differing relations between these particular features.
There are, it will be remembered, three different aspects of civil
society: the sphere of circulation, the sphere of reproduction and the
sphere of struggle. It is the relative significance of the first two that affects
the sphere of struggle, and this in turn has immense implications for the
detennination of the state. Where the sphere of circulation dominates
the sphere of reproduction, then social struggles revolve around the
 The State: Some Proposals 107

establishment and maintenance of relations of exchange equality between

the owners of the different commodities, including labour-power. In that
case, state policy is more directly circumscribed by the nature of the
economy and the struggles of different social groupings to establish
conditions of formal exchange-equality. By contrast, where the dominant
sphere is that reproduction, then state policy is more indirectly mediated.
There are highly complex and interdependent social struggles involving
numbers of different classes and popular forces, each concerned to
establish and sustain the conditions of reproduction of their labour-
power. The state then has to pursue a much wider range of policies,
located within a broadened institutional base in order to guarantee the
conditions of profitable accumulation. I shall try to support this
argument, first, by considering Pashukanis's alternative conception
of the state. Second, especially in the next chapter, I shall argue it di-
rectly.
To put the argument very briefly, Pashukanis fails to relate law and the
state to the other aspects of civil society apart from that of the sphere of
circulation. 13 As is well known, Pashukanis tries to show how the form of
law, individual juridic subjects, is to be related to the characteristic form
ofthe capitalist economy, namely the commodity. And further, he shows
why the mechanism of constraint consists of an impersonal public
authority isolated from the specific relations of each capitalist enterprise.
Pashukanis believes that both ofthese forms, oflaw and the state, are to
be derived from the nature of the commodity, from the distinctive
cell-form of the capitalist economy. I shall suggest 'that although this is
correct in those social formations in which the dominant sphere of civil
society is that of circulation, this is not so when the dominant sphere is
that of reproduction. I will now try to show this by briefly summarising
certain aspects of Pashukanis's materialist theory of the law.
First, then, Pashukanis rejects a common Marxist critique of the law in
which it is argued that the content of laws and of legal institutions are
consistent with, and determined by, the material interests of the dominant
class or classes. 14 For him it is essential to analyse the form of law itself
which can be abstracted from any specific content. However, this does
not mean that we can identify a universally true or appropriate conception
of the law. Rather the law itself only develops in specific contexts;
namely, where the isolation and opposition of specific interests occur.
And this results from the existence of commodity exchange, as a result of
which people come to be seen as legal subjects, or having a legal
personality. It is when the product of labour becomes a commodity and a
108 The Anatomy of Capitalist Societies

bearer of value that people acquire this capacity to be legal subjects and
bearers of subjective private rights. The social relations of production
thus present themselves both as the value of commodities in exchange and
as people's capacity to be the subject of such rights. Each person has thus
a juridically constituted will which makes him or her free and equal to
other commodity owners. The idea of a subject as the abstract bearer of
rights is thus produced by the relations of commodity exchange. And the
general development of bourgeois relations leads to law itself becoming
abstract in character. The legal norm takes on the logicall y perfected form
of abstract universal law . The legal subject is, as Pashukanis puts it, the
abstract owner of commodities raised to the heavens. IS And central to
these relations is that of the contract, an agreement which is concluded
between autonomous wills, between autonomous juridical subjects.
Contract law is thus the paradigm form of law for Pashukanis.
And by contrast with Stuchka, for example, Pashukanis claims that the
form of all law - private, criminal and public - is to be explained in
terms of the nature of commodity exchange. In modem punishment - for
example, in criminal law - the essential notion is that the blameworthy
individual is punished by serving an appropriate time in prison. There is
thus the idea of equivalent exchange, the deprivation of freedom to sell
labour-power for a time commensurate with the scale of the crime, not
say until he/she dies or can be bought out. There is dominance of
'equivalent retribution'.
Unfortunately Pashukanis devotes himself only briefly to the further
question as to why the state or public law 'detach[es] itself from the ruling
class and take[s] on the form of an impersonal apparatus of public power'
separate from each commodity producer. 16 He argues that a state is
needed which guarantees the general conditions of exploitation rather
than each separate exploitative relation. The reason for this is that each
wage-labourer is not compelled to work for a specific capitalist but only to
alienate his/her labour-power to one capitalist or another. Pashukanis
argues that since this alienation consists of a formal relation between two
autonomous commodity owners, it is necessary that there is an
impersonal public authority to guarantee such contractual relations
between each capitalist and each worker. More generally, the form of the
state is given by its functioning to maintain the overall system of legal
regulation in a capitalist society, and in so doing to preserve the
dominance of the capitalist class. Arthur says: 'There is, therefore, the
coexistence of a legal form relating' 'independent and equal persons" on
the one hand, and, on the other, the material reality of the rul e of one class
 The State: Some Proposals 109

over another in the bourgeois state - but mediated, as we have seen,

through the rule of law. '17
There are many issues raised by Pashukanis's provocative thesis,
incl uding his rejection of the notion of a proletarian law. The main debate
has focused around whether it is possible, and legitimate, to derive the
form of law from the sphere of circulation, rather than from either the
sphere of production, or the state. IS It is clear that there are many
difficulties here: Is it really possible to reduce all oflaw to that of contract
and property? How can we distinguish between crirninallaw and simple
coercion if the former is only seen as an exchange of individual
equivalents? Is it correct to see all legal subjects as human subjects? Does
law not have partial effect itself on the relations of production? Does
wage-labour really need juridical equality? What happens when as in
British labour-law such equality was not established for many years? Is
not law full of contradictions and therefore not a coherent unity? What
role does struggle play in developing the legal form, and what effect do
changes in the latter have on forms of struggle? In relation to his argument
on the state there are some further specific difficulties in his attempt to
derive it from private law and hence from the sphere of commodity
exchange.
First, he minimises the coercive character of the state, merely noting
that it has to appear to emanate from an abstract collective person and to
appear to be in the interests of all parties to legal transactions. Coercion
seems to be conceived of as something external to the law and the state
rather than as constitutive of it. Further, he fails to consider the
importance of coercion in the interventions made by the state in the
struggles both between various social forces and between such forces and
the state. Second, he relates the state only to the sphere of circulation and
to its functioning in constituting separate juridic subject within civil
society. Yet this ignores the two other spheres of civil society. We have
just seen the importance of the sphere of struggle in relationship to the
coercive character of the state. The sphere of reproduction is also of
importance and means that the form of the state is not given purely by its
role in sustaining legal relations based on the sphere of circulation.
Incidentally, Pashukanis seems to be aware of the problem, since he says
that juridical theory of the state which attempts to encompass all state
functions would be inadequate nowadays. 19 However, he fails to provide
anything apart from a juridical theory, and he does not attempt to account
for any other functions of the state. Third, he provides no account ofthe
mechanism by which this form of the law and the state develop. He
110 The Anatomy of Capitalist Societies

simply tries to deduce them from the nature of commodity exchange, and
his argument thus suffers from the previously noted deficiencies of
rationalist forms of demonstration and proof. What should be provided in
each case is an exposition of how and why the different contending social
classes and other social forces produce, as a consequence of their
struggles to sustain their material conditions of life, the development of
legal regulation guaranteed by the centralised coercion of the state. There
is a premature leap to considering the relations supposedly characteristic
of a fully developed capitalist social formation, where these directly
mirror the logical relations between the concepts of the commodity, law
and the state. 20 The crucial political question which is posed by the
consideration of civil society is to determine the effects that working-
class and popular struggles have on the development of the law and the
state. I will argue below that such actions are of major imortance, but this
does not mean that socialism can be established through a reformist
political strategy. What has to be explained is the process of transmuta-
tion. Why is it that the organised power, organisation and strength of
particular social classes has the effect of producing specific forms of the
law and state-forms which may well be more 'suitable' from the
long-term viewpoint of capital accumulation? How is that transmutation
to be explained? Indeed, how and why in specific circumstances do the
political 'victories' of the working class result in a more developed
version of the capitalist state?
The significance of this point can be seen if we consider the argument
of Blanke et at., which is in part an attempt to expand Pashukanis's
analysis more specifically into the area of the state. They argue, like
Pashukanis, that out of the commodity relation as the specific, reified
form of the cohesion of social labour , there arises the form of the law and
of legal regulation between isolated legal subjects. Such law must be
enforced and its appropriate enforcement must be guaranteed by coercive
force. Thus they do not derive the state from the commodity, only the
form of coercion. They then say that it is necessary to develop certain
principles of form which this coercive force must observe if it is to
conform adequately to the form of the commodity. Such principles are to
be found in the concept ofthe general law , which is impersonal, abstract
and public. Just as the nature of commodity production separates human
social relations into the material relations of legal persons, so the
cohesion of society constitutes itself in a dual manner as abstract, and
'supra-personal'. They then proceed to consider decisive functional
changes in the extra-economic coercive force that follows from the
 The State: Some Proposals III

development of money into capital and of labour into wage-labour. In so

doing they connect the state to the changing relations between the spheres
of circulation and reproduction within civil society. But what they do not
do is systematically to link these developments with the sphere of
struggle. Let us consider two examples of such lacunae in their text.
Thus, they say that the first typifying feature of politics consists of
struggles to establish or to interpret legal rightS. 21 This occurs at that stage
in which there is a dominance of the sphere of circulation (over
reproduction) and it is this which structures the sphere of struggle.
Struggles thus resolve around the effort of the popular forces to extend
and consolidate legal rights - to generalise such rights so that each
person is constituted as an equivalent legal subject. However, Blanke et
al do not consider the effects ofthese struggles, merely pointing out the
nature of the first typifying feature of politics. This failure to consider the
effects of the sphere of struggle can also be seen in the second example
here, where they point out the value of other commodities in the
reproduction of labour-power is 'the life process of the concrete being
with his concrete needs' .22 Such reproduction they say will always be a
matter of struggle involving contestation over 'the quantity of indispens-
able necessaries required to maintain life'. 23 At which point they quote
Marx's chapter on 'The Working Day' from volume 1 of Capital. But
again they fail to develop this further. What will be the effects of such
struggles to work limited hours for particular amounts of income? How
are the struggles by the different sellers of labour-power related to the
growth and development of the state?
Broadly speaking a major deficiency of the Blanke et al. argument is
their failure to develop a proper concept of civil society, one that
embraces not only the sphere of circulation, but also the spheres of
reproduction and struggle. It is these which are singularly combined,
such that it is only through struggle that the labour-power of wage-
labourers is reproduced; and, in particular, that such struggles increas-
ingly extend beyond the realm of civil society and embrace the state. And
it is the state itself whose principal function is to sustain, guarantee and
orchestrate suitable quantities of appropriate labour-power. Having
therefore seen certain limitations of the arguments of both Pashukanis and
of Blanke et al. , I will now put forward more positively some important
propositions on the state, law and labour-power.
One characteristic error in much debate on the state is to adopt a
position approximating to the following. It is assumed that capital
develops some general interest in maintaining capitalism and that the
112 The Anatomy of Capitalist Societies

development and elaboration of the state occurs in order to secure this,

especially to provide these conditions ofreproduction ofthe commodity,
labour-power. This may be seen to occur either in an automatic systemic
fashion, or from the active intrusion of the interests and struggles of the
specific capitalist class. By contrast with either ofthese interpretations of
the capitalist class, it is held that the working class has an essentially
non-reformist revolutionary interest to destroy capitalist relations. A
reformist strategy is then inherently a demonstration of false conscious-
ness. Of course it is not denied that most workers are reformist, but it is
argued that this is for especial reasons. Reformism does not follow from
the characteristic social relations of capitalist production; but rather
from particular features, such as the existence of a labour aristocracy, or
from divisions within the working class, or from a collaborationist labour
movement. Thus the development of the state is seen on this view as a
class state, a system of political domination sustaining and guaranteeing
the general interests of capital, or of the capitalist class.
There are a number of deficiencies of this view. First, consider the
following quotation from the Grundrisse:

Every capitalist knows this about his worker, that he does not relate to
him as producer to consumer, and [he therefore] wishes to restrict his
consumption, i.e. his ability to exchange, his wage, as much as
possible. Of course he would like the workers to other capitalists to be
the greatest consumers possible of his own commodity. But the relation
of every capitalist to his own workers is the relation as such of capital
and labour, the essential relation. 24

The reason for this is the competitive nature of the capitalist economy, of
the fact that capital is comprised of very many capital units. The relations
between them in toto are unplanned, anarchic and competitive, although
of course specific capitalists will try to minimise the deterious effects of
competition. Each capital will seek to reduce to a minimum its costs of
production, and especially the cost of the labour-power which it employs.
This is achieved in various ways, which include directly cutting wages
below the value of its labour-power, lengthening the working day and
increasing the intensity of labour, either by getting each worker to work
harder, or by raising the organic composition of capital. It is of course
clear that if the hours and intensity of labour are increased so much, a
natural limitation is reached when labour-power is destroyed faster than it
is biologically reproduced. However, the reason why this generally does
 The State: Some Proposals 113

not occur is not simply because of the development of a general interest of

capital not to destroy the commodity labour-power. The competitive
nature of capitalist relations could not by itself ensure this. The principal
reason why labour-power is not destroyed is because of the organised and
sustained resistance of wage-labourers. It is only their resistance which
prevents both the continuous extensions of the working day and increases
in the intensity oflabour. Wage-labourers of all sorts are literally forced
to defend their interests, to struggle to sustain their material conditions of
life. The development of the state, its institutional and policy-making
elaboration, is thus significantly the outcome not directly of the capitalist
interest but of that of wage-labourers. To say therefore that the state in
capitalist societies is a 'capitalist' state is in one sense incorrect. It is only,
I would argue by conceptualising the such societies into the economy,
civil society and the state that one can see how the class struggle in civil
society does significantly determine state form and policy. However, and
this is crucial, working-class and popular struggles do not effectively
transform such states into a popular or workers' state able to eradicate
capitalist relations. There is nothing inconsistent about arguing both that
state structure and policy significantly reflect working-class and popular
struggles, and that such structures and policies do not amount to a
substantial transformation in the direction of a 'workers' 'state willing
and able to abolish capitalism from within. In much Marxist discourse, it
is held that these are inconsistent, so to believe in the possibility that the
state does change under worker resistance is simultaneously to believe
that capitalist societies can be reformed and become socialist.
One implication of my argument so far is to highlight an interesting
paradox. Whatever may be the beliefs of the capitalist class, they do
function in such a way that undermines one of the essential conditions of
capitalist production, namely refreshed and energetic labour-power.
Wage-labourers by contrast, whether or not they happen to be re-
volutionaries, function in part to sustain capitalist relations through their
struggle to reproduce their own labour-power. This may be put
over-simply. The effect of capitalism is to produce within civil society
one class, the working class, who are literally forced to struggle in order
to survive, to struggle, often involving state action, over employment,
hours, wages, conditions of work, safety rules, old-age, sickness,
holiday and unemployment benefits. Yet, it is this struggle which is
essential for the reproduction of labour-power without which capitalist
production cannot continue.
One rider should be put here. I am not claiming that, in specific
114 The Anatomy of Capitalist Societies

capitalist societies, groups of progressive capitalists do not come to

develop some general conception of capitalist interest, in particular of
reproducing the labour-power of the workers. However, in general, these
represent responses on the part of sections of capital to the struggles of the
working class to defend themselves. Advances that are made are
therefore to be seen as concessions which are forced out of the capitalist
class as a result of the forms of industrial and political resistance that the
working class develop. Where such forms of resistance do not occur, no
such general interest can be seen to have developed.
If we view the history of capitalism in this way there are some
advantages. First, it means that there is no likelihood of emasculating this
history, of implying that the countless struggles of various working
classes have been to no avail since, even without them, the self-
equilibriating capitalist system would have ensured the reproduction of
their labour-power. Second, it demonstrates materially why working
classes have been literally forced by capitalist relations to struggle to
reform the conditions of their existence, and thus, as we shall see in the
next chapter, to be reformist. They literally have had no choice, just as
they have no choice in working for one capitalist employer or another.
They have to struggle against capital, because without that struggle they
would not survive. Finally, this viewpoint highlights some important
points about the state itself. This can be seen by considering briefly two
possible explanations of the growth in the size and the significance ofthe
state, and especially of the welfare state, in the United Kingdom. In the
first, it is argued that the growth in the welfare state results from class
struggle, from the struggles of the working class, as a result of which the
size and significance of the state has been greatly increased. 25 The
expansion of welfare and educational services and benefits, of the
nationalisation of the infrastructure, and the development of national
insurance, have all been at the expense of capital. In the second, the
growth of the welfare state has been seen as functional for capital
accumulation. 26 Thus the development of the state has meant the
socialisation of certain of the costs of reproducing labour-power, and of
the basic infrastructure (energy, transport), and it has brought off the
working class by incorporating it more fully within the interstices of
capitalist society.
The first explanation, while highlighting the importance of the class
struggle, suffers from two problems. First, the form of the state is
unrelated to the laws of motion of capital and the changing requirements
of capital accumulation. Second, it is premised upon a reformist
 The State: Some Proposals 115

conception of the state which is seen as more or less autonomous of all

other social relations. The second is problematic because it omits the
class struggle altogether, as though this has no effect at all on the degree,
the timing and the forms of state policy, expenditure and structure.
It is only the state that can limit the power of each capital unit to destroy
the basis of its own reproduction and accumulation. The state imposes on
each capital various laws, rules, conditions, procedures and taxations
which govern the operation of these separate units. And these result
significantly from the force and organisation of working-class and
popular politics, from the determined resistance of the working class and
others to reproduce the conditions of their own reproduction. Such
working-class resistance is never straightforward since it will entail
alliances and coalitions with other classes, fractions and social forces.
And it will always meet the constraints of action imposed by the existing
degree and forms of capital accumulation, the relationship of national to
international capital, the movements of money capital and the constraints
of the balance of payments, etc. Furthermore, the representatives of the
working class increasingly seek power within the state, so that the
separate capital units can be more effectively determined, thus attempting
to effect those policies which more successfully ensure the reproduction
of their conditions of existence. What is the relationship of law to these
processes? I shall make a couple of extremely brief comments.
I have already referred to the idea that there are circuits of 'power and
ideology' , which are analogous to the circuits of capital. In the latter the
commodity money is crucial in effecting the division between, and
integration of, production and circulation. In the circuits of power and
ideology there is a corresponding medium, and that is the law. I have
attempted to represent this in Fig. 7.1. Law is the intermediary between
civil society and the state. The demands made within the former consist of
changes in the form of law and/or lawful direction by the latter; yet at the
same time one major form of interpellation within the spheres of civil
society is that of legal subjects by the state.
Law is therefore not simply part of the capitalist base or of the
superstructure; nor is it to be directly derived from capital or to be seen as
conterminous with the state. 27 Nor, on this account, is it to be viewed as a
unity - different parts of the law are differently integrated into the
circuits connecting the state and civil society. It also follows that
developments in the law do not occur automatically. For example, if all
adult individuals in a society enjoy equal legal rights this only occurs
because of three interconnecting determinants: (a) dominance of the
116 The Anatomy of Capitalist Societies

,--------
Sphere of capitalist

l~7
Sphere of circulation ~
1 Sphere of struggle }
Civil
society
Sphere of reproduction /

\ law
L-;-S_p_e_C-:-ify_ _ _ State
function

Figure 7.1 The Basic Structure of Capitalist Social Forma-

tions

sphert:: of circulation; (b) popular democratic struggles under the

hegemony of a major class or other popular force articulated around this
theme; and (c) a power bloc able to neutralise opposition and to sustain
effective support for may be a longish period of time.
In the next chapter I shall consider these points in more detail. In
particular, I will consider the central paradox of the capitalist state: that
much of its form and policy results from the struggles of the sellers of the
commodity labour-power, yet at the same time it attempts to reproduce
that labour-power in a form which enhances its use-value to capital. In the
following chapter I shall consider the form of a civil society within which
social struggles result in representative democracy being established. In
both chapters I shall try to show that the circuits interconnecti~g civil
society and the state through the intermediation of law are of central
importance.
8
The State, Labour-power
and Class Struggle

Let me begin here by summarising some of the main implications of the

preceding chapters. Having thus set these out, I shall then consider
certain further aspects of labour-power, politics and the state.
First, then, it is wrong to view class relations as purely or essentially
economic. Gasses, I argue, only exist at the level of civil society, subject
to various forms of structuration. It is wholly wrong to conceive of classes
at the level of the abstract, or pure, mode of production. Classes then exist
in a structured or differentiated form, and their existence presupposes the
other classes with which they interrelate. Two forms of class struggle can
be identified: 'class struggle' between capitalists and workers; and
'classes-in-struggle'. For any capitalist society it is necessary to
determine whether one or other of these struggles is dominant, or indeed
whether that which is dominant is a form of popular democratic struggle.
Second, although capitalism is to be characterised as a system of
generalised commodity exchange it is clear that the commodity labour-
power has certain unique characteristics. In particular, it cannot be
produced like other commodities. It cannot be produced as a capitalist
commodity. The conditions for its reproduction lie outside the process in
which capitalist commodities are themselves produced. Furthermore, the
sellers of this commodity, of labour-power, cannot simply alienate it, in
the manner in which all other commodities can be alienated. The seller of
labour-power is him or herself the bearer of that labour-power. Such
sellers of the commodity have thus an interest in its price quite unlike the
interest that the sellers of other commodities happen to have. Moreover,
the purchasers of this labour-power have a peculiar interest in it since this
commodity, unlike all others, possesses a particular use-value, namely
that it is itself a source of value, and hence of surplus-value. Thus, while
the purchasers of labour-power, the capitalist, will seek to purchase it at
118 The Anatomy of Capitalist Societies

the minimum price, they will not want, as a class, to destroy that
commodity since it is the very basis of value. So, although the
relationship between capitalists and workers is antagonistic, the form that
this antagonism takes is profoundly structured by the centrality to the
relationship of the commodity of labour-power. There are two crucial
implications. First, the sellers oflabour-power have to struggle in order to
sustain the conditions under which their labour-power can be reproduced.
This emphasis on struggle is central, and ensures the continued and
profound importance of so-called economistic struggles. Thus, whatever
may be the long-term ambitions of wage-labourers , in the short term they
can do nothing but attempt to wrest sufficient material conditions in order
to reproduce their labour-power. Without such struggles they would not
ensure that reproduction. They thus have a partial interest in capitalist
social relations to the extent that their labour-power is reproduced within
such relations. Second, both capital and labour have an interest in this
reproduction of labour-power, labour obviously because they are
themselves the bearers of that labour-power. But it is also one of the
necessary conditions of capital accumulation and so capital-in-general, if
not individual capitalists, seek to ensure that there are sufficient quantities
of various categories of the commodity labour-power. Thus, while
individual capitalists are in a directly antagonistic relationship with
labour this is not the case with capital-in-general. The manner in which
capital and labour are inextricably bound up with each other, at the same
time as being antagonistic, stems from the crucial importance of the
unique commodity labour-power.
Third, it is incorrect to view the state simply as the bourgeois state, the
state of the bourgeoisie. So, given that there is a domain of civil society
lying between the economy and the state, it does not follow that the state
is the instrument of that class or classes dominant within that civil society.
Thus, it is not necessary that a neo-Gramscian position entails the view
that the state is the instrument of the dominant class; so neo-
Gramscianism does not entail instrumentalism, as Oarke suggests.!
However, it obviously does not follow from this that state form and policy
instead simply reflect the demands ofworking-class and popular struggle,
that the capitalist state is the instrument of the working class. The
problems of this view can be seen from Navarro's recent book on the
development ofthe health service in Britain; his view is well-summarised
in the following:

there emerges a popular demand for assuring the availability of

 The State, Labour-power and Class Struggle 119

services .... And that demand eventually determines a response from

the dominant class, a response based on the necessity for that class to
legitimate the social order in which it holds dominance. And that
dominance further reflects itself in the nature of the response, i.e., it
primarily mirrors and reproduces the class interests and ideology of
that dominant class. Indeed ... the specific interest groups, incl uding,
among others, professional interests, creep in and substantially shape
its final form.2

If we consider, say, the background to the National Health Insurance

Act of 1911 Navarro argues that there was 'popular demand' for such
legislation. He says that this was shown by increasing industrial
militancy, the shift away from craft and textile unions and the growth of
socialist and Marxist views at home and abroad. But this is problematic:
for example, he points to the peaking of strike activity during and after the
depression of 1890-3. But what is the precise connection between
industrial militancy in that period and health-insurance legislation fifteen
to twenty years later?3 Indeed, what would count as a demonstration of
such a connection? Is it the case that the 'dominant class' always
responds? If not, what determines whether or not it does respond? And
what determines the success of its response if any?
Navarro uses this argument to demonstrate that the social legislation of
the 1906 Liberal Government did not have anything to do with the
ideology of noblesse oblige of the aristocracy or of the nascent
bourgeoisie. But he does not show this. He merely reiterates that such
legislation stemmed from (a) the social demands by labour for better
wages and conditions, and (b) the social needs of the capitalist class for a
fitter labour force. But why and how did these come together through the
Liberal Government? How did the social needs of the capitalist class
manifest themselves as such, given that specific capitalists themselves
did not favour such legislation? What is the connection, if any, between
the need to legitimate the social order and the ensuring of a healthy labour
force?
There are thus a number of crucial difficulties in Navarro's analysis,
difficulties which stem from the failure to theorise adequately the
relations between the economy, class struggle and the state. His account
improves when he considers the actual nature of legislation passed and
implemented; how the Liberals in Parliament rejected a nationalised
health service funded out of general revenues. He also shows, in other
cases, how the professional interests, especially of the doctors in the
120 The Anatomy of Capitalist Societies

hospital section (the 'upper class'), have consistently structured and

transmuted the content of the popular demands being articulated at the
time. Let me then summarise three of my main objections to his account.
First, he treats the state mainly in terms of its function to legitimise the
capitalist economy and its attendant class relations, not in terms of its
functioning to organise the conditions appropriate for that economy to
continue. 4 Second, he equates the dominance of the hospital doctors with
that of the dominance of the capitalist class, both in terms of their class
background and in terms of their present functioning. 5 However, this is
only so on a sociological, non-Marxist conception of Class; and he never
shows why medicine bears a particular relation to capital and hence what
role it plays in relation to the certification of fit/unfit labour-power. 6
Third, and related to this, Navarro operates in terms of a rather crude
'conflict' model of capitalist society, of a struggle between capitalists and
workers, with various middle-class professionals in between. He fails to
consider the nature of possible power blocs, such as the one the Liberals
achieved in the first decade of this century, and of class alliances. He does
not explain the connection between this particular class action and the
nature of the resulting legislation. Why legislation to reproduce labour-
power? What are the developments in the nature of civil society which
change the sphere of struggle in this way?
Navarro's text illustrates a more general point; namely, the necessity to
integrate into the analysis other forms of struggle apart from that between
the major social classes. These other social groupings, of race, gender,
generation, and so on, engage in popular democratic struggles, and not
essentially in struggles to change the underlying class relations. Thus
there are social practices which may have important political effects,
which have their own specificity and autonomy, and which cannot be
seen as the 'property' of a particular class. It is where the sphere of
reproduction is dominant in civil society that there are a number of
different bases on which politics may develop, and these are in part
separate from class struggle and class-in-struggle. Furthermore it is
important to acknowledge the crucial importance of popular struggles
which are oriented around the axis of the people - the power bloc, or the
people - the state. This, too, occurs as a consequence of the dominance
of the sphere of reproduction. There are three crucial consequences.
First, in contemporary capitalism working-class or socialist politics are
never the exclusive form of politics; there are always present popular
democratic struggles of various sorts. Second, no working-class or
socialist movement can exist and mobilise without important popular
 The State, Labour-power and Class Struggle 121

democratic elements, the nature and importance of which will obviously

vary. Third, such elements may be mobilised by forces other than
working-class or socialist movements, as Fascism stood for 'the people'
in the German and Italian Fascist movements between the wars.7
These points further imply that there is no necessary ideological
consensus in capitalist societies. Indeed the very existence of civil society
signifies the opposite, that there are a diversity of different practices, with
no necessary ideological conformity. I have thus argued against the
notion of a dominant ideology, and hence against the claim that all (or
most) practices within a particular capitalist society serve to reproduce
that society. There are numbers of practices within each civil society
which have no necessary relationship with the 'class-struggle', which is
structured by, but not determined by, relations within the economy.
However, three provisos on this general argument need to be made and
emphasised. First, it is not, of course, the case that the practices of civil
society are uninfluenced by the economy, but the degree to which this is
so varies considerably. Second, capitalist societies are at times charac-
terised by a high degree of ideological conformity, by a particular
structured relationship between the state and civil society. But even at
such times, for example, in the welfare consensus of post-Second World
War Britain, there will be a substantial amount of socially structured
ideological dissent. And third, it is clear that contemporary capitalist
societies, even if they are not characterised by a simple hegemony, are
characterised by a continuing effort by the state to organise this popular
democratic sentiment. This stems from the fact that capitalism is
organised into nation-states and that the state consistently seeks to
identify and reinforce common nationality. There is thus a national
framework which is reproduced and in terms of which people interpret
and evaluate events and experiences. The mass media is particularly
important in emphasising that there is one perspective on events, that
there is a national consensus and there are hence no irreconcilable
conflicts of interest. 8
This means that the establishment of a degree of ideological consensus
is a matter of struggle, as the state attempts to sustain and reproduce a
particular relationship with civil society, in particular to organise the
popular democratic forces into a given national structure. And in civil
society, social classes and other forces in seeking to reproduce their
labour-power may actively seek out such organisation through the state,
or they may resist it in various ways. Thus the relative autonomy ofthe
state stems from the existence of civil society and the fact that there are
122 The Anatomy of Capitalist Societies

patterns of interrelationship between the two. Although certain practices

in civil society are determined by the economy, and the form of the state is
given by its need to attempt to preserve the overall conditions for
profitable accumulation, these interrelationships are partially auton-
omous of the capitalist economy. It is these interrelationships between the
state and civil society which guarantee the relative autonomy of the
former and the partial independence of certain practices within the latter.
This also means that different forms of the capitalist state, such as
representative democracy, fascism, corporatism and authoritarianism, in
fact imply different relations between civil society and the state. For
example, corporatism is not a theory of the economy, or of the
articulation of particular interest groups, or of the state - but it is in a
sense all ofthese things, a theory of the direct representation of corporate
groups of civil society within the state rather than the indirect represen-
tation of individual citizens who cast their vote for one representative or
another.
Finally, I have emphasised a number of features of the state. Let me
emphasise four here. First, any capitalist state faces many difficulties
both because of the pressures from civil society and because of its
inability to work out and implement correct policy. Indeed it may often
implement a counter-productive policy. It simply cannot be expected to
implement that policy most appropriate from the viewpoint of maximum
capital accumulation, although its form is given by that need. Second, I
have argued that it cannot be assumed that a power bloc will be formed
out ofthe politically dominant classes; this mayor may not occur. Third,
state form and policy should always be related to specific struggles and
not seen as the embodiment of abstract principles, such as bureaucracy
being the realisation of the principle of rationality. Fourth, the actual
form or policy implemented rarely if ever fully reflects the demands of the
social forces articulating them. So the popular demand for fairness in
treatment, and not for a class-bias, helped to develop the bureaucratic
form of administration and hence to restructure the state into a less
controllable, more oppressive structure. Because of the particular
articulation of the state and civil society, no one class or social force is
able to implement exactly what it demands.
A possibly helpful way in which my argument can be summarised is to
say that it is neither a class nor a capital-theoretical approach. 9 The state is
not to be derived either from the dominant class(es) whose supposed
instrument it is, nor from the nature and movements of capital. It is rather
to be seen as resulting from the interdependent relations between the
 The State, Labour-power and Class Struggle 123

economy and civil society: the fonner sets its demands, the latter provides
the context within which it struggles to resolve them. I shall now expand
on this by considering a number of related issues: labour power,
periodisation and nationalisation in this chapter; democracy and par-
liamentarism in the next.
In a recent article Aumeeruddy, Lautier and Tortajada argue that the
essential problem in the analysis of the state is its relation to the re-
production of labour-power. \0 They argue that the foundation of the
state must be sought in the wage-relation, and hence in the relation
between the state and labour-power. They suggest that what it is that
establishes the state as a separate entity beside and outside civil society is
the fact that the production and circulation of commodities does not
automatically reproduce that which is both external and essential;
namely, labour-power which cannot be produced as a capitalist com-
modity. State intervention is thus necessary, first, to produce this
particular commodity, to constrain the 'bearers' oflabour-power to enter
into the wage-relation; and second, to manage the fragmentation of the
collective labourer, to divide and separate workers who are in fact united
through their collective production of value. II Furthennore, the state,
like labour-power, is 'external' to the production of commodities, its
externality being dependent on that of labour-power.
It should be clear that there are similarities between my argument and
that of Aumeeruddy et al. However, there are some difficulties in their
formulation which are important to consider. First, they provide a
peculiarly biased interpretation of recent debates on the state, arguing that
these have been mainly oriented around the nature of economic policy
and not around the very nature of the state. They further suggest that of
those authors who have opened up the latter question there is agreement
that: 'The state is an instrument of domination at the service of the
dominant class. '12 They disagree with this because such a view seems to
obscure the 'fundamental, open, violence that is the attribute of the
state'.13 However, they fail to theorise such violence beyond implying
that it stems in some way from the foundation of the state upon the
regulation of the wage-relation. One wonders what to make of the claim
that the existence of the wage-relation 'demands overt violence every
time such violence is thought necessary'.14 What is the relationship
between violence and the other aspects and features of the state? What
happens in those situations when violence is not applied?
Second, as we have seen, they assert that the foundation of the state is
to be found in the wage-relation. This is never argued, merely asserted,
124 The Anatomy of Capitalist Societies

and glosses serious difficulties. First, they sometimes imply that

everything that the state does can be reduced to its function in reproducing
labour-power, in minimising its costs of reproduction and in reproducing
its conditions of existence. At other times this function is merely seen as
the foundation of the state, which then presumably engages in other
activities, although these are not specified. Second, they themselves
acknowledge that at times the regulation of the wage-relation has not been
undertaken by the state they mention how in nineteenth-century
Europe reproduction was effected, in a sense, on the basis of the simple
destruction of the bearers oflabour- power. Yet if this was the case then
how can the foundation of the state be as A umeeruddy et al. argue. This
would imply that if the state was not constraining the bearers of
labour-power to enter into the wage-relation, then there was no state. So
in nineteenth-century Europe, for example, since labour-power was not
adequately reproduced it would follow absurdly that there was no state.
Third, they never consider whether the state is in fact able to reproduce
appropriate labour-power. They assume that the functional requirements
will be met in an adequate form. They ignore the social conditions which
determine state policy, and indeed the singular involvement that the
'bearers' oflabour-power have in ensuring that such labour-power will be
reproduced. Thus, to the extent that 'the costs of existence of capitalist
production '\5 are born by capitalist and shared out by the state, this is not
something inscribed in the nature of the state but results from the
struggles of wage-labourers to ensure their existence as labour-power,
healthy, literate and numerate, and effectively transported from home to
work.
Finally, there is an interesting section in which Aumeeruddy et al.
maintain that, outside commodity production, there are various 'labour
processes', 'control processes', and 'other activities', including affect-
ive, sexual and authority relations .16 Although it is important to recognise
the existence and significance of these, they remain in an untheorised
form. Indeed they appear to ambrace both relations within the state and
within civil society, and as such they clearly bear differing relationships
with the economy. To what degree then, and in what forms, is the
reproduction of labour-power the consequence of state intervention
(through 'overt' or 'masked violence '); and to what degree is it the result
of these other processes outside the state as well as outside the production
and circulation of commodities?
One of the deficiencies of their argument is the failure to provide a
periodisation of the tendencies which they discuss. All they say is that in
 The State, Labour-power and Class Struggle 125

the early stages of capitalist development, with the dominance of absolute

surplus-value production, there was no state intervention in the wage-
relation; and of course in later capitalism highly complex forms of such
intervention have occurred. However, I wish to argue that it is also
necessary to categorise the different forms of social struggle that may take
place and to relate these to different periods of capitalist development.
Thus, with regard to the development of the class struggle and the state
there are two main periods in the development of capitalist societies: that
in which the sphere of circulation is dominant within civil society; and
that in which the sphere of reproduction is dominant. In the former, the
relations of exchange dominate the relations of distribution and consump-
tion. It is essential for the process of accumulation that commodity-
capital is quickly realised as money - capital and returns to the sphere
of production. Thus, there is a strong basis for the establishment of
'Freedom, Equality, Property, and Bentham', 17 for the establishment of
those conditions which permit the most efficient and fastest circulation
of commodities. But it is erroneous to suggest that such conditions fol-
low automatically from the dominance of the sphere of circulation. The
establishment of such conditions results from particular social struggles,
without which such conditions will not be established. It is therefore
illegitimate to claim, as Holloway and Picciotto do, that a particular
pattern of accumulation, of absolute surplus-value production, produces
as an effect the dominance of the sphere of circulation, and that this
produces a particular form of the state, its 'liberal movement'.18 It is
rather that while the establishment of the sphere of circulation does follow
from the production of absolute surplus-value, no particular develop-
ments of the state need follow. They depend on specific social struggles
which I will ill uminate below. As Blanke et al. say: 'The owner oflabour
power as a free wage-labourer with the full and equal rights of a citizen
was able to develop only through long class struggles. In no way does he
[sic] arise from the surface forms of competition. '19 Furthermore, the
shift from absolute to relative surplus-value does not mean there is a shift
away from the dominance of the sphere of circulation within civil society.
That again depends on the forms of social struggle and the effects that
they may have.
Within this sphere there are four main forms of struggle. First, there is
the attempt by the different fractions of the nascent capitalist class to
establish monetary, measurement and transportation systems within the
national territory appropriate to the developing exchange economy. They
also seek to establish legal procedures backed up by the state so that
126 The Anatomy of Capitalist Societies

market transactions can be contractually established, sustained and

enforced, and if necessary to establish procedures for sueing and being
sued. Second, there is that struggle, mainly pursued by popular and
working-class forces, to establish conditions in which all individuals
exchange commodities in a position of approximate legal equality. Thus
the effect of such struggle, if successful, is to produce a set of social
relations within which formally free and equal juridic subjects are
constituted, subjects who take themselves to be the authors of their own
actions, who sue and are sued, who buy and sell commodities, who
alienate their own labour-power or purchase that of others, and so on.
Third, there is struggle around the conditions which affect exchange,
which determine the price at which different commodities are bought and
sold. In particular, with relation to labour-power, there is struggle over
the categories of worker (male/female/child), the time they are available,
their capacity to organise, and so on. In situations in which the conditions
are highly unfavourable, then extensive mobilisation will be probable,
generally along the lines in which the systematic denial of rights happens
to exist. And fourth, there will be struggle by different classes and social
forces to extend the principle of exchange equality into the political
sphere, to ensure that they possess the right to vote, to join a political
party, to stand for election, and so on. The four most important bases for
the establishment for such politics are class, gender, race and religious
groupings. I will discuss below some of the conditions which determine
the effectiveness of such struggles in the establishment of representative
democracy. At this stage I wish to emphasise three points.
First, the development and extension of the 'liberal' state, safeguard-
ing and extending the circulation of commodities, does not follow
directly from the nature of an economy characterised by the production of
absolute surplus-value. It only follows through the effects, often
unintended, of the forms of struggle in which the various classes and
social forces in that civil society happen to engage in. Second, to the
extent that such struggles do have the effect of ensuring the extension and
elaboration of the system of generalised commodity exchange, then this
has two consequences: (i) to accelerate the rate of accumulation; and (ii)
to shift the dominant sphere of civil society from that of circulation to that
of reproduction. To the extent that such struggles do not have these
effects then the rate of accumulation will be lower and struggles will
continue to be structured within the sphere of circulation. Although the
production of relative surplus- value may be dominant, in cases through
the 'importation' of such relations of production from other social
 The State, Labour-power and Class Struggle 127

formations, it does not follow that the sphere of reproduction will be

dominant within civil society. There will thus be uneven development
between advanced economic development and 'reactionary' political
forms oriented around the attainment of relations of formal exchange-
equality. In such situations there will be at times intense struggles focused
upon the bases along which such formal exchange-equality has not been
established, along that of religion in Ulster for example, or ethnicity in
the United States. Third, this indicates that although one may conceive
how in general formal exchange-equality is established in society, there
will always be exceptions to this, the most noticeable being, of course,
that of women. Partly, the political significance of this is obscured
through their central involvement in the sphere of reproduction, but this is
not to deny the continuing importance of the inequality they experience as
they bring their labour-power to the market. 20 Fourth, the significance of
establishing formal exchange-equality for the majority of the sellers of
labour-power is considerable. It changes the parameters of politics in the
advanced capitalist societies and represents a major gain for the popular
classes. But there are two provisos in this. On the one hand, such an
achievement has to be protected since reversals may well occur, as has
happened in Latin American over formal political rights. 21 And on the
other hand, it aids the development of relative surplus-value with certain
deleterious consequences, among which is the shifting in the sphere of
struggle to that of the dominant sphere of reproduction.
Let me now consider the situation where the sphere of reproduction is
dominant within civil society. Unlike the former case where the central
characteristic is the establishment of exchange-equality in the present
between buyers and sellers of different commodities, in this case the
central feature is the emphasis placed on the distribution of wages and of
surplus-value, and on the consumption of commodities; that is, distri-
bution and consumption are dominant. This can be seen from the
viewpoint of both capital and labour. In the case of the former, the
tendency for the rate of profit to fall is experienced in part as a series of
crises of realisation, so that the later stages of capitalism are characterised
by a tremendous emphasis on sustaining and enlarging consumption. The
growth in the concentration and centralisation of capital increasingly
means that the general conditions of production cannot be provided by
individual capital units - reproduction cannot be therefore assured
simply through exchange processes. The growth in the scientific and
technological character of capitalist relations means that such skills have
to be themselves produced, or reproduced, among the bearers of certain
128 The Anatomy of Capitalist Societies

kinds oflabour-power. From the viewpoint oflabour the gains achieved

through the achievement of formal exchange-equality in the sphere of
circulation, have a number of consequences in the period in which the
sphere of reproduction dominates: (1) a commitment to the conservation
of gains already achieved, (2) a demand to extend them to other areas,
especially to the political, (3) an emphasis upon extending struggle to
increase consumption, (4) a willingness to engage in distributional
struggles. Three particular features are important. First, the very
establishment of formal exchange-equality in the period in which the
sphere of circulation dominates, means that it is possible and likely that
different occupational groupings will compare and evaluate their incomes
with each other. The establishment of such formal legal equality is
important in leading each group in society to evaluate how well they are
doing with many other groups. Second, the fact that the relations within
exchange are less directly influential means that patterns of distribution
are a matter for struggle. They are not simply determined and in part
legitimated through the market. They are much more the consequence of
social struggles. Third, a crucial lesson learnt in the sphere of circulation
period is that changes are achieved only through collective forms of
struggle and not through individual forms, in particualr, through trade
unions, political parties and pressure groups of various sorts.
As pointed out earlier, it is a contingent question as to whether the
sphere of reproduction assumes dominance. It depends on the effectivity
of struggle in the previous period - struggles which mayor may not yield
relations of formal exchange - equality. If they do, then the following
seem to be important features of the struggles characteristic of the
dominance of the sphere of reproduction: (a) all of the social forces in
civil society may engaged in struggle so minimising the political effect of
specifically class struggle; (b) class struggle, even if far more organised,
is less intense because of the achievement of the gains of formal legal
equality; (c) the reproduction of labour-power in particular necessarily
involve state action and policy; (d) the development of the state reflects
the growing and conflicting demands for state action from these differing
social groupings; (e) there is a general shift of demand from commodified
to non-commodified forms of state action; and (f) the shift from exchange
to consumption/distributional dominance entails greater control over
what is produced and the distribution of rewards that follows from its
consumption. 22 Incidentally, I am not suggesting that state form and
policy follow simply from these characteristics; to do so would be to
reduce the state to the relations of civil society. There are two further
 The State, Labour-power and Class Struggle 129

critical determinants: first, the nature of the economy, the changing

patterns and forms of accumulation, the laws of motion of the capitalist
economy, the existence of crises, and so on; and second, the nature of the
state itself, especially the balance of class and other forces present within
it.
Let me now consider these characteristics of struggle within the sphere
of reproduction. First, there is a proportionate decline in the importance
of class as opposed to other kinds of social struggle. 23 This is in part
because certain groupings, such as women or ethnic groups, will seek to
establish conditions of total exchange- and political-equality. But it is
also because the more such general formal equality is achieved, the
greater the variety of bases on which groupings will seek to establish a
degree of substantive equality. There is no reason why mobilisation will
only take place along class lines. This is especially so when a particular
kind of subject, such as women, youth, blacks, and so on, undergo
similar experiences in relation to the labour market and their organisation
by capital. The importance of the reproduction of labour-power also
means that there are substantial sectors oflabour-power either engaged in
such reproduction (teachers and lecturers, hospital workers, etc.) or in
being themselves reproduced (student, patients, etc.).
Second, once the sphere of circulation is no longer dominant within
civil society, then the likely forms of class struggle generally become less
intense. 24 Of course, there are some exceptions to this, in part depending
on the rate and form in which formal exchange-equality and later formal
political equality is achieved. However, the attainment of basic private
and public rights dramatically weakens the effectivity of class struggle.
This is because a major determinant of that struggle is that of the
systematic denial of rights, say to all workers; once they are achieved or
even partially achieved then a major basis of class action disappears and
other bases of social struggle became as important. Wage-labourers may
seek to reproduce their material conditions of life in a variety of different
kinds of social struggle.
Third, the reproduction oflabour-power necessarily involves the state.
This is because the nature of capital is such that it is rare for it to be in the
interests of any particular capital-unit to bear the costs of reproducing its
own labour-power. This is at least the case when there is a need for
specialised forms of training, medical care, sickness and welfare
benefits, and so on. As the sphere ofreproduction assumes dominance it
is the state which ensures that all capital-units bear some of the costs of
their provision.
130 The Anatomy of Capitalist Societies

Fourth, this socialisation of the costs of provision clearly reflects the

growing demands from the wide variety of social groupings within civil
society. All such groupings seek the state to provide for their
reproduction - there are none of the limitations characteristic of the
period in which the sphere of circulation is dominant. Obviously this will
also incl ude the economically dominant classes and fractions, the
professional and new middle classes, as well as wage-labourers of
various kinds. The effect is that there is little disincentive on any
particular grouping to moderate the demands that it places on the state,
the regulation of such demands having to take place through fiscal crises
and dramatic cuts in public expenditure. 25 Thus the widely noted
extension of the state in the later stages of capitalism results in part from
the large number of demands placed upon it, demands which are difficult
to resist stemming as they do from the wide variety of social groupings in
civil society. Politics in the period of dominance of the sphere of
reproduction increasingly involves the state. Thus there is a changed
relationship with civil society. On the one hand, the state is far more
subject to determination by the characteristic forces of civil society; and,
on the other hand, the state is enhanced in the degree and kinds of activity
in which it can engage. Yet to the degree to which the former occurs, the
effectivity of policies that are enacted is increasingly problematic. There
will at times be strong demands from different groupings for a functional
representation within the state in a manner which transcends the limited
effectiveness of the parliamentary representation. Any trends towards a
corporatist relationship between state and civil society must be in part
understood in terms of the demands articulated by the relevant classes
themsel ves. In certain societies this will incl ude the capitalist class,
which will benefit from the expansion of the state, especially where there
is a strong and active working class. Hence the paradox that the latter may
result in an increased centralisation of the state, which is then less
responsive and which cannot be so easily shifted. 26
Fifth, there will be considerable conflict within civil society and within
the state over the form taken by the extensions of the state. There will be
conflict and struggle between the adherents of commodified and of
non-commodified state policies. 27 The former invol ve those interventions
by the state which work through, and reinforce mechanisms within, the
sphere of circulation; these include taxation changes, unemployment
payments, cash forms of welfare, profit, price and wage controls, health
and educational provision through payment, regional and other subsidies
to private industry, etc. Non-commodified state policies involve the state
 The State, Labour-power and Class Struggle 131

providing a use-val ue and not a commodity, thereby a voiding the sphere

of circulation into which all commodities enter and are evaluated against
each other. In the case of activities organised by the state we can
distinguish between a relatively commodified form, Renault, say, in
France, from the relatively non-commodified form, such as public utility
provision in Britain. Very broadly speaking the shift of dominance from
the sphere of circulation to that of reproduction means that political
demands are increasingly oriented to non-commodified rather than
commodified policies. Why is this so? There are two main reasons. First,
with a commodified and/or selective policy there is difficulty in ensuring
that all members of a particular category acquire the particular use-value
in question. It is easier to ensure this if the use-value does not have to be
purchased on the market. Second, many different social groupings will be
likely to support such non-commodified policy; in particular, it will
effectively mobilise popular democratic sentiment oriented around the
contradiction, the people:the state. However, although it is likely that
political demands from the popular classes will take this form, it is also
progressively the case that the state will attempt to resist such demands,
or at least to moderate them often in a substantially significant manner.
This is because such a non-commodified programme will be more
expensive, and since it will be financed out of general taxation it will
thereby divert surplus-value that would otherwise be appropriated by
capital. 28 Therefore much of capital, and probably also the traditional
petty-bourgeoisie, will struggle against non-commodified policy. Such
opposition will often be sufficient to prevent the implementation of such a
policy. It is only if such classes have in part been politically neutralised
that a non-commodified policy may be implemented. 29
Finally, the dominant elements in the sphere of circulation are those of
consumption and distribution. This has two consequences. First, all
groupings are implicated in a competitive struggle for maximising their
share of distribution; there are few restrictions which prevent such a
competitive struggle, which will obviously be heightened during crises.
This means that the state cannot but be involved in mediating the various
claims of the differing social groupings. Second, the importance of
reproduction means that groupings are progressively involved, not
merely in sustaining an income adequate for their reproduction, but in
influencing the pattern of production itself. There is thus a shift in the
focus of political demand, from (1) achieving formal equality of
exchange conditions, which many groupings have only partially attained
even in later capitalism; (2) struggling for an improving distribution of
132 The Anatomy of Capitalist Societies

income and wealth; and to (3) ensuring that their reproduction can be
assured through the consumption of particular use-values, and for a
specific balance between the commodified and non-commodified form;
that is, that the processes of exchange do not yield an appropriate balance
of use-val ues .
I am now going to approach this last point in more detail. I shall analyse
the development of nationalised industries in the United Kingdom, in
other words, with the forms of struggle necessary to produce as an effect
the taking into state ownership of a major industry previously organised
under capitalist relations of production. My comments will be very brief
and selective. For reasons that I have already indicated this is not the
characteristic form of struggle in the later stages of capitalist develop-
ment. Thus the paradox that, to the degree to which nationalisation is
necessary in order to restructure particular industries, the interests of
capital are increasingly problematic to ful fill, since there is less likelihood
of establishing an effective power bloc willing and able to carry through
such nationalisation. I shall initially consider the limitations of an
economic reductionist analysis of this issue, as in Holloway and
Picciotto's (HP) theory of the state and class struggle. They operate with a
three-stage model of the development of the form and functions of the
state. 30 In the final stage, 'The socialisation of production and the
tendency of the rate of profit to fall', HP argue the following:

As capital is forced, in the struggle for accumulation, to strive to

overcome the limitations ofthe state form, it tendentially undermines
that particularisation of the state which is a preconditon of its own
existence. Increasingly the state intervenes directly in the production
process, taking over particular industries and reorganising the actual
process of value creation and exploitation.

And in the next paragraph:

The tendential undermining of the separation of state from society does

not mean that that separation is overcome. The undermining does,
however, pose a threat to the mystification of the political, to the
fetishised appearance of the neutrality of the state. The state is
increasingly identified with capital, reformism - the strategy of using
the state against capital - is increasingly abandoned, even as a
justificatory ideology used by the traditional reformist parties. 3 !
 The State, Labour-power and Class Struggle 133

In the former quote then they argue, first, that there is a progressive
increase in the direct intervention of the state in the production process,
and second, that this follows from the need for restructuring implied in
the process of accumulation. There are a number of difficulties in this
argument. Partly, it is empirically dubious - there is certainly no simple
unilinear increase in such interventions, and there are examples of
reversal - for example, of the reduction in state ownership in Germany in
the post-war period, or even of the selling back of some state industry to
private ownership in the United Kingdom. It is also difficult to know how
to interpret the considerable variation in state ownership in different
capitalist societies. What are we to make of the fact that the most
advanced capitalist society, the United States, has not been subjected to
the same pressures - why isn't the degree of state intervention higher and
not lower than elsewhere? How important in that case is the fact that the
United States is the leading capitalist economy and that we have to
consider the location of each economy within the overall chain of
capitalist economies, within the world economy?32 However, the central
problem in the first quotation is its reductionism, that the actions of the
state follow from the need for capital to be restructured in certain ways.
This ignores both that the form that this restructuring takes is partly a
political question, so that the effects of nationalisation rather than, say,
the encouragement of centralisation under private ownership are diffe-
rent. But also the very existence of nationalisation will only occur if there
is a power bloc willing and able to carry through such a measure. This
involves analysis of the leading political parties and their articulation with
different classes, fractions and social forces, not only in terms of the
orchestration of support but also of the neutralisation of opposition. The
importance of this issue can be seen if we consider the second quotation
above.
In this HP suggest that reformism, which they define as the strategy of
using the state against capital, has been progressively abandoned in the
advanced capitalist economies. Yet if this is so one wonders how the state
can increasingly intervene in the production process, interventions which
may include nationalising major industries. In Britain, for example, if the
Labour Party has abandoned reformism, then how can it be that there will
be increasing direct interventions in the production process, given the
manner in which such interventions by the state are in fact taken to be part
of a strategy for the achievement of socialism in Britain? HP fail to
consider this issue, namely, the form and determinants of class and other
struggles within civil society, and of their articulation with the state.
134 The Anatomy of Capitalist Societies

Before I try to explain certain aspects of the process of nationalisation, let

me indicate one crucial reason why their analysis fails.
Throughout much of their article they refer to 'class struggle' , to the
struggle between capital as a class and wage-labour as a class. They argue
that the process of capital accumulation is essentially one of class
struggle, and that it is a deficiency of some of the German literature that
the capital relation is not seen as one of class struggle. 33 However, their
use of 'class struggle' as a shorthand for the capital relation has the effect
of making their approach seem simultaneously theoretical or logical, and
empirical or historical. As a result, actual struggles of specific groups in
particular capitalist societies can be viewed as direct expressions of the
capital relation, of the 'class struggle' , one term among others of Marx's
theory ofthe capitalist mode of production. The employment of the term
'class struggle' conflates the treatment of the real (class struggle implying
voting, strikes, lockouts, speeches, organisations, etc.) with the theoreti-
cal (class struggle as a shorthand for the capital relation). This is both to
commit certain errors of epistemology, but also more importantly of
substance. Thus, it is not the case that the development of the state form is
only to be related to the 'class struggle' in the capital relation sense. It is
also to be related to class struggle in the sense of the specific class and
other social groupings of influence in particular capitalist societies.
Oearly the policies demanded and implemented relate to the nature of
capital accumulation, to the tendency and counter-tendencies for the rate
of profit to fall and to the nature of crisis. However, the precise effectivity
of the accumulation process depends on the manner in which such
features are mediated through specific classes, social forces, and indeed
through the state itself. Holloway and Picciotto point out how the state
'must remain essentially external to the process of accumulation ';34 they
do not, and indeed cannot, explore how class and other struggles are also
partially external to that process of accumulation.
I will now finish this chapter by discussing the process of nationalis-
ation in more detail. Two kinds of explanation of the pattern of
nationalisation can intially be seen as incorrect. On the one hand, it might
be said that it will be those industries which are unprofitable which get
nationalised, and as the processes of capital accumulation work them-
selves out this will lead to a series of nationalisations as one industrial
sector after another becomes unprofitable and falls into the arms of the
state. 35 On the other hand, it might be held that the industries which get
nationalised reflect the ambitions of the labour movement and of its
political representatives. Thus the industries nationalised are the so-
 The State. Labour-power and Class Struggle 135

called commanding heights of the ecnomy. and such a policy may reflect
the attempt to effect a major shift in the balance between the state sector
and private capital. The former is thus an economistic account. the latter a
politicist interpretation. Both are inadequately one-sided.
Let me begin with the economistic account. It is clear that one of the
features of capitalist development is that capital is mobile. it moves in and
out of sectors in pursuance of maximum profit. This means that capital
will not be attracted to those sectors in whichever branch of production
where profits below the average are being made. Now this process would
not be problematic if capitalism were a well-equilibrated world-wide
system in which labour was as mobile as capital. There are, however, two
features of capitalist accumulation which are essential to recognise. First,
it takes places within nations and is fundamentally structured by national
territorial division; second, it involves the bearers of labour-power who
are relatively geographically immobile. Nationalisation must be related
to these two aspects of the pattern of accumulation.
First, let me consider the implications of the fact that accumulation
takes place within nation-states. In a relatively advanced economy many
commodities will be imported and will be paid for through the export of
other commodities. However, not all such commodities will in fact be
imported - some commodities which would have been will in fact be
produced at home. This may occur through the state's encouragement of
private capital to produce in conditions in which it would not otherwise
ha ve done, or by the direct provision by the state in either a commodified
or non-commodified form. Thus one crucial aspect of state intervention in
the economy is to ensure that a balance of production is sustained, so that
adequate supplies of different commodities are available, commodities
being in highly complex interdependencies with each other. When
therefore I referred earlier to the form of the state as given by its role in
sustaining the conditions of continuing capital accumulation this can be
seen to have two interrelated aspects: first, restructuring of specific
industrial sectors; and second, the ensuring of a balanced structure of
output of the different branches and secotrs of production. Why then on
fairly rare occasions does the state intervene, not in the sphere of
circulation, but directly in the sphere of production, taking over the
ownership and control of a particular private enterprise? We can list a
number of factors that may lead a particular enterprise to be nationalised:
first, that an industry is strategically important, such as energy or
aerospace; second, that, because its interdependence with other industrial
sectors, it is especially important for that country's balance of payments,
136 The Anatomy of Capitalist Societies

either in providing a market for many others (such as motor vehicles), or

in providing a major cost of production (such as transport or fuel); third,
that an industry is a major employer oflabour-power and there are few if
any alternative employers (such as coal in Britain); and fourth, that the
industry is failing to provide adequate supplies of appropriately priced
commodities needed by other sectors (such as steel in Britain).
Now although one or more of these conditions apply it is obviously the
case than not all such industries actually get nationalised. Nationalisation
represents a particular kind of industrial restructuring. It forces through
the laws of motion of the capitalist economy faster and more effectively
then would be achieved under private ownership. The centralisation of
capital is effected in often spectacular fashion. 36 Likewise nationalisation
invariably entails the massive rationalisation of separate plants, and the
release of capital for more profitable uses. 3? There will tend to be major
reductions in the labour-power employed, and consequential increases in
labour productivity.38 In Britain, nationalisation has involved the
massive restructuring of a fairly small part of British industry (10-15 per
cent assets, 8 per cent of employment) but one that makes 30 per cent of
all new manufacturing investment. 39 It is a form of restructuring based
upon pre-existing devisions between different industrial sectors. It has
the effect of preserving ownership and control within the nation-state.
Given these effects it might be thought surprising that there is not more
rather than less nationalised industry. It is clear, however, that it is a fairly
rare occurrence; of the various industries that might be nationalised, only
a small number are. The main reason for this is that nationalisation, which
is a particularly dramatic form of restructuring, necessitates a power bloc
able to implement such a programme. I see this as singularly problematic.
In Britain the Labour Party has been best able to orchestrate such a power
bloc. But this raises a curious paradox. One of the most notable effects of
nationalisation has been for the quantity oflabour-power employed to be
cut dramatically .40 Yet historically in much of Western Europe it has been
the trade unions and the political representatives of labour who have
struggled most strongly for nationalisation. Why is this?
The first reason was hinted at earlier; it stems from the structural
asymmetry of capital and labour. To a very significant extent labour-
power is relatively immobile geographically. Capital is not. At least in its
money and commodity forms it is mobile, it can be moved around the
world relatively freely, save for the various exchange controls, quotas,
tariffs, etc. The bearers of labour-power are relatively fixed. They live in
and are bound to a given national territory. Labour therefore, in order to
 The State, Labour-power and Class Struggle 137

reproduce its material conditions of life, is forced to seek the more

efficient operation ofthat capital on which it depends. So as to reproduce
its own conditions of existence it must seek to sustain that capital situated
within the state-determined national territory. As a result we have one
major reason why the representatives of labour-power will tend to seek
the nationalisation of private capital. However, as we have seen, such
nationalisation will generally involve a massive restructuring of that
industrial sector, a restructuring which will certainly reduce the labour-
power employed.
Second, nationalisation will be a policy of the labour movement in
advanced capitalist societies to the extent that a statist conception of
socialist politics is held. It must be generally believed that state relations
are at least partially antagonistic to capitalist relations, and that the former
can be directed in such a way that capitalist relations can be modified or
controlled. This form of politics is thus premised upon the dichotomy
between capital and the state. To the extent to which that dichotomy is
perceived to be less clear cut, then there is a decreased possibility that
nationalisation will be taken to be a central component of class struggle.
Of course, as we have argued, there is always in fact a dichotomy
between capital and the state since the latter is external to the processes of
surplus-value production. However, to the extent to which there seems to
be a changed relationship between civil society and the state, then it may
appear that there is little or no dichotomy between the state and capital,
and hence little chance and effectivity of using the state against capital.
It does not though follow from this that there will be more
nationalisation the greater the dichotomy between the state and capital.
This is because we also have to consider the effect of this dichotomy on
other classes, apart from that of the working and popular classes. In
particular, the greater the dichotomy between the relations of production
and the state then the more determined will be the opposition of the
capitalist class and of its various agents to nationalisation. This is because
nationalisation does involve the transfer of the ownership of the means of
production from private hands to the state, and is therefore directly
antagonistic to the interests of capital in one sense. The opposition of
capital to nationalisation will be less, the weaker the dichotomy between
capital and the state. 41 Clearly nationalisation will only be possible where
capital is in general neutralised - two circumstances which facilitate this
being opposition and conflict between the different fractions of capital,
and the existence of an extremely strong and vigorous working-class and
popular movement. 42
138 The Anatomy of Capitalist Societies

I do not intend to consider here all the determinants of an appropriate

power bloc which could effect a major nationalisation in the United
Kingdom, or in other of the advanced capitalist economies. It seems clear
that the Conservative Party is usually unable to nationalise major
industries (not forgetting Rolls - Royce) and this is because of two
connected reasons. First, it is generally responsive to the particular
section of capital likely to be nationalised, as they were, for example, to
the steel industry.43 Second, the Conservatives are less likely to be able to
orchestrate a power bloc effective enough to push through national-
isation, both because ofthe difficulty in neutralising much of capital, and
of organising on their side the demands of certain working-class and
popular forces. The Labour Party, by contrast, has been the party of
nationalisation, because it has been willing to act against particular
sections of capital, and because it represents wage-labourers, who have a
singular interest in the conditions which ensure their reproduction.
Hence, in relationship to certain issues, the Labour Party is able to act
more effectively in the interests of capital to ensure its expanded
accumulation. It is less subject to the demands of that capital about to be
nationalised, and it has a strong relationship to the labour movement
which can, on occasions provide the foundation for the orchestration of a
successful, reforming power bloc. This may work in one of two ways:
either negatively, to prevent the pursuance of a particular policy; or
positively, to implement policy.
I will now make a few further brief comments on the forms of class and
other struggles in contemporary capitalism. First, the process by which
capitalist relations are established will affect the effectiveness of, and
relations between, class and other struggles. There are a number of
dimensions: pre-existing social and spatial division; speed of develop-
ment; degrees of violence, dependence, and spatial and industrial
unevenness; the role of the state; the strength of civil society; and the
characteristic forms of hegemony . The interplay between these mean that
there are 'critical moments' in the history of struggle, in which a
particular structured relationship is effected, and which is often very
difficult to change. 44 Second, and connected with this, it is important to
consider the relationship between popular-democratic struggles articu-
lated around the people: the power bloc contradiction, and class
struggles, and how this may, especially at times of 'national crisis' (after
1945, for example), provide the basis for successful orchestration of an
alternative power bloc willing and able to carry through, say, the
nationalisation of particular industries. Third, we should also consider
 The State, Labour-power and Class Struggle 139

the nature of the state and the degree to which the social forces activated
in the sphere of reproduction operate within the state itself. As a result the
dichotomy between the state and civil society breaks down, as the state
both attempts to moderate struggle and is simultaneously riven by
struggle. And the greater this is the case, the progressively more difficult
it becomes to use the state to reform civil society, and hence to restructure
the capitalist relations of production. There is thus a major contradiction
within British society (as well as in certain other societies). Social classes
and other social forces have been forced to 'enter' the state in order to
ensure their own reproduction; but to the degree to which this has been
successful, reformism becomes increasingly unrealisable, especially to
the degree to which the representatives of capital devise mechanisms to
neutralise in part the popular forces. 4S Fourth, it does not follow from this
that there is a 'legitimation crisis' in contemporary capitalist societies,
rather that the particular structured relationship between the state and
civil society is one which progressively prevents the resolution of the
problems of capital accumulation within the economy. It seems neither
appropriate nor sensible to characterise this in terms of a lack of
'legitimation' , as though there is some quite specific level oflegitimation
sufficient to ensure social continuity. There may well be a 'crisis of he-
gemony';46 but this is less important than the inability of the sellers of
labour-power to devise a coherent basis for reformist politics (let alone of
a revol utionary politics). It is an implication of this chapter that reformist
politics are often one precondition for the successful restructuring of the
economies of late capitalist societies. 47
9
Capitalism and
Representative Democracy

It has now become fairly commonplace to maintain that there is a

'functional fit' between capitalist social relations and representative (or
liberal or bourgeois) democracy. This has been argued by conservatives
such as Friedman, who claim that capitalist economic organisation
promotes political as well as economic freedom;1 by radicals, such as
Macpherson, who maintains both that liberal-democracy is found only in
countries with capitalist enterprises and that, with a few exceptions, all
countries with such enterprises develop liberal-democratic institutions;2
and by Marxists such as Lenin, who argue that a democratic republic is
the best possible political shell for capitalism. 3 Yet there are obviously a
number of difficulties in these arguments. On the one hand, if we consider
all these societies in which the CMP has been dominant, then probably
only a minority have in fact been organised in a representative democratic
manner. Political democracy, by which I mean universal adult suffrage, a
representative parliament to some extent controlling the executive and
judiciary, and institutionalised freedoms of election, speech, association,
and so on is a twentieth-century phenomenon mainly of Western
European and North American capitalism. It did not occur in nineteenth-
century capitalism, nor has it developed in large numbers of peripheral
capitalist societies. And of course even in the core capitalist countries
there have been some distinctive departures from such representative
democracy. On the other hand, what Macpherson calls this 'close
correspondence of liberal-democracy and capitalism'4 has not been
adequatel y explained. Just why is it that capitalist social relations both
necessitate and produce specific political practices which have particular
functional consequences for these dominant capitalist relations? What is
the mechanism which produces liberal or representative democracy? And
is it true that it is always the best possible shell for capitalist relations?
 Capitalism and Representative Democracy 141

In the following I shall argue, first, that there is nothing inevitable or

necessary about the growth of representative democracy; its development
therefore depends on quite specific determinants. Second, it is only in
capitalist societies with (a) a civil society relatively autonomous from the
state; (b) a relatively large well-mobilised working class; and (c) other
mobilised popular classes and forces, that the minimal conditions exist
for organising a popular-democratic struggle to extend the rights of
citizenship to all adults. Third, the establishment and the maintenance of
democratic forms depends on a specific balance of class forces which can
be broadly specified. It is not possible to argue that representative
democracy is or is not the best possible shell for capitalism. This is a
contingent matter. In particular, I shall argue against the thesis that it is
the best possible shell, in that situation in which the capitalist class is
politically and ideologically dominant. 5
Let me begin by considering Macpherson's still illuminating discus-
sion. He points out that the 'liberal democracies' as we know them were
'liberal first' and 'democratic later'.6 In other words, before they were
democratic in the sense indicated above, they were liberal. By this he
means that the society in general and the state in particular were organised
on the basis of freedom of choice, individuals were free to make the best
arrangements within the market-place. Instead of a society based on
custom, status and authority, there was a liberal market society based on
individual mobility, contracts and choice. The liberal state which
functioned to sustain these relations in the market-place developed as a
system of alternating or multiple parties, whereby government could be
held responsible to one or other section of the dominant class or classes.
This was not particularly democratic. The task of the competitive party
system was to uphold the competitive market society by ensuring that the
government was responsive to the changing majority interests of those
who were running the market society.
However, Macpherson says, this society did after a time generate 'a
pressure for democracy which became irresistible '.7 Those without a vote
saw that they had no weight in the political market. They had no political
purchasing power and hence their interests were not consulted. They thus
demanded the vote themselves and there were no good grounds for
refusing this since the liberal society was founded upon equal individual
rights. However, such attainments was only realised after many decades
of agitation and organisation, democracy thus coming as a late addition to
the competitive market society and the liberal state. Macpherson
emphasises two points. First, the attainment of democracy was some-
142 The Anatomy of Capitalist Societies

thing extra, it had to be struggled for, often over a long period. Second,
democracy was both demanded and admitted on the grounds that it was
unfair not to have it in a competitive society. It was something that the
competitive society needed.
However, this argument is problematic. Although it is clear that the
extension of the franchise did result from struggle, it is necessary both to
indicate the form that this struggle took, and the determinants. It has not
been able to occur in all capitalist societies. Furthermore, it is not the case
that the extension of the franchise was fought for simply to effect the
logical completion ofthe 'competitive society'. It was fought for so that
specific further struggles could be sustained, struggles which increas-
ingly appeared to demand a political rather than simply an economic
settlement. It is thus abstract to suggest that it was simply the logic ofthe
competitive society which produced a democratisation of the liberal
society. There is thus a tension in Macpherson's argument. On the one
hand, the move towards representative democracy results from struggle;
on the other hand, it results from the logic of the competitive market
system. To overcome this contradiction in his argument it is necessary to
show how the nature of particular capitalist societies (their 'logic ')
produces class and other movements which have as their effect (if not
necessarily their only or main aim) the establishment of representative
democracy. I will discuss this issue by firstly considering Barrington
Moore's impressive if flawed attempt to analyse the social origins of
democracy and dictatorship. 8
As is well known, Moore mainly discusses two forms of class struggle,
that between the landlords and peasants and that between the landed
upper class and the urban capitalist class. Given that the particular society
is not governed in absolutist fashion by a monarch; that is, that there is
some balance between the nobility and the monarchy, then there are two
central preconditions for democratic development. First, it is that there is
the commercialisation of agriculture, that agriculture becomes capital-
istically organised and hence that the peasantry is transformed into
capitalist tenant-farmers and landless wage-labourers. This means that no
highly repressive state is required in order to maintain the exploitation of
the peasantry by the landlord class (through whatever system of
rent-payments happens to be in force). Hence, the state can be
democratised. Second, the commercialisation of agriculture also means
there is a basis for a progressive fusion of the agricultural and industrial
capitalist classes, with the general dominance of the latter. Where this is
the case it ensures an overall diffusion of democratic practice, especially
 Capitalism and Representative Democracy 143

where there is strong urban organisation. Moore says that where there is
no urban bourgeoisie there will be no democracy.9 It is where agriculture
is not commercialised and the peasantry is not eliminated that a
reactionary alliance of landlords and a more weakly organised capitalist
class develops, an alliance which weakens and in cases prevents the move
towards a democratic state form. In these cases 'dictatorship' of one form
or another results.
This is not the place to consider this text in any detail. Besides noting
the weak and very vague notion of democracy that is used,IO the main
deficiency lies in Moore's failure to specify precisely just what are the
social conditions of representative democracy. He seems to think that
simply if the capitalist class is dominant and establishes a peaceful fusion
with the increasingly commercialised landed class, then democracy will
ensure. This is of course not true. As Macpherson suggests, what this
may well do is to establish a liberal state, but a liberal state is not
necessarily representatively democratic. That requires two further fea-
tures: first, a realm of civil society separate from the state; and second,
organised working-class, feminist and, in cases, racial movements whose
actions produce, as effects, universal suffrage and related freedoms.
Moore is therefore incorrect to assume that simply because the capitalist
class has acquired power this will in itself produce representative
democratic forms. It may tend to lead to certain developments which are
conventionally related to universal suffrage, such as a party system and a
legislature, constitutionalism, a judicial system guaranteeing property
and contracts, and so on. But it is only if universal adult suffrage is
struggled for that, in certain circumstances, it will be realised. Moore,
like many other writers, suffers from the delusion that ther~ is something
about capitalists and capitalism in their essence that will lead to
representative democracy, unless some other factors actively prevent this
from occurring. Moore therefore employs what is in effect a class-version
of modernization theory. Thus, if the reactionary peasantry has been
destroyed and the landed upper class is subordinated to the capitalist
class, then the latter will ensure the progressive modernisation of that
society, in particular of ensuring a relatively free and democratic political
system. In Britain, he maintains, repression was unnecessary - the
peasantry has been destroyed, the working class could be disciplined
within the workplace - and hence freedom and democracy, a modem
political system, could be slowly but irreversibly introduced. There was
no need for a repressive, unfree, un-modern political system.
Now there are a number of curious aspects to these arguments. In
144 The Anatomy of Capitalist Societies

particular, Moore equivocates over the relationship which exists between

the structure of any society and those classes present and active within
them. Clearly, in analysing peasant revolution as in China, this cannot be
understood without seeing the peasantry as itself making history, because
after all it makes the revolution. But in his treatment of western history
the popular forces tend to disappear, except as something simply to be
controlled. And in his account of English development he minimises the
importance of working-class struggle in helping to produce England's
'relative freedom' and political democracy. He sees the existing
politicians, under the influence of the dominant bourgeoisie, handing out
reforms, legislating to protect the poor and committed to ideals of liberty
and democracy. A corollary of Moore's modernisation thesis is a
commitment to the Whig presumption of the far-sighted, benevolent,
liberal and democratic Parliament reforming the society and itself and
leading gradually but inexorably towards the modernized, humane and
essentially just society of the present. 11 So while Moore criticises
modernisation theory for ignoring the role of violence in the relations
between lord and peasant, his elevation of the importance of that class
struggle leads him in tum to neglect the concrete manifestations of class
struggle between worker and capitalist, and of the partial effectiveness of
the former within these struggles.
In order to demonstrate one of the important arguments of this book let
me now set out those reasons as to why class and popular-democratic
struggles are of significance in capitalist societies, partly in producing
'ruling class' containment, but also in forcing through important but
transmuted political and social reforms. I am not arguing that all reforms
simply result from such struggles, merely that, in many cases, but not all,
a necessary condition for such reforms is the existence of a social
grouping engaging in particular forms of struggle. I am also not arguing
that it is only the struggles of obviously subordinate groupings which are
effective; nor that the effect of struggle is to realise what that class or
social force actually seeks (we have noted above the problem of
'transmutation ').
(1) The first reason is that without such a hypothesis it is difficult to
understand the enduring commitment in Britain to reformist politics, or to
representatively democratic institutions. Supposed we assume what
many Marxist theories imply; namely, that working-class and popular
struggle has no effect in producing either representative democracy or
social reforms. Then, of course, the explanation of why there is
widespread commitment to both these practices is problematic, and
 Capitalism and Representative Democracy 145

recourse has to be made to a legitimating ideology.12 But as is well

known, these explanations are problematic, suffering as they do from
some of the same deficiencies as normative functionalist theories. 13
Prima facie , then, it seems more plausible to argue that the major reason
why there is popular commitment to representative democracy and social
reformism is because they do in part stem from popular struggle. They are
not simply illusions foisted on the masses, but they are in some sense
developments the masses themselves have brought about. They are not
mere bourgeois institutions. It is rather that the western state is a 'popular'
state in this respect; namely, that it stems from popular struggles to
enlarge, strengthen and democratise the state against capital.
This further means that the issue that has obsessed Marxist
commentators - namely, why the working classes in contemporary
capitalism are non-revolutionary - is wrongly posed. It is wrongly posed
because it is presumed that such societies have remained unchanged, that
because they are still capitalistic then the working class should still be
revolutionary. But this argument is problematic, partly because it is not at
all clear that the early working classes were revolutionary as opposed to
merely militant, but mainly because it ignores the degree to which such
societies have in fact changed, significantly because of working-class and
popular pressure. It may be that the fundamental patterns of exploitation
remain unaltered, but there have been highly significant transformations
in civil society, in the state and in their relations with each other and the
economy. The 'illusions of state socialism '14 result from the considerable
changes that have occurred, many of which, especially those relating to
the reproduction of labour-power, stem from individual and collective
actions by the sellers of such labour-powers.
(2) The second basis for our argument is similarly indirect. If we are
faced with choosing between two explanations of the same phenomenon,
one that is functionalist and one that is non-functionalist, then we should
choose the latter, other considerations being relatively equal. It is now
clear that (a) functionalist theories are in general deficient, (b) many
Marxist theories of the state are functionalist, thus (c) such Marxist
theories of the state are similarl y deficient and are al ways to be considered
inferior to plausible class theories. 15 In particular, it seems highly
implausible to argue that the progressive extensions of the franchise in
Britain represented the most functional means by which the system
ensured that the working class - and, later, women - were to be
incorporated into nineteenth- and early-twentieth-century British
capitalism. Nor incidentally does it seem plausible to argue that these
146 The Anntomy of Capitalist Societies

extensions of the franchise represented the most functional adaptation of

the system as conceived by the 'ru1ing class', that the latter in some sense
so understood the dynamics of British capitalist development that they
could correctly estimate just when the suffrage shou1d be enlarged. If
both these theories are to be rejected then we again have prima facie
grounds for believing that working-class and popu1ar struggles are of
some import.
(3) Against this, however, it might be argued by, say, Barrington
Moore that the crucial determinant, given a capitalistic agricu1ture, is in
fact a modernising and strong bourgeois class. But we have already seen
deficiencies in this account - in particu1ar because it becomes very
difficult to see why such a capitalist class would seek to transform a
liberal state into a democratic state. 16 Why shou1d a capitalist class aim to
end the system of propertied - voting unless they were forced to do so?
Furthermore, it is incorrect to argue that the capitalist class is always the
modernizer, seeking to reform there conditions under which the mass of
the population live. As Engels pointed out: 'The Ten Hours Bill provided
an excellent meeting ground for these reactionary classes and factions to
join forces with the proletariat against the industrial bourgeoisie. '17 This
also shows that, although the dominant classes may often be opposed to
each other with respect to certain crucial areas of policy, it is normally the
case that progressive changes can be vetoed unless there is strong
organised popu1ar support. This indicates the essential importance of
analysing the domain of civil society and of the potential alliances and
blocs which can be sustained there. In relation to nineteenth-century
England, Johnson discusses the political effects of the alliance of
capitalist agrarian and industrial classes. It has meant, among other
things, that the commonest form of English radicalism has been, not
socialism, but an anti-aristocratic popu1ism. 18 Such a ru1ing alliance both
fuelled popu1ar opposition to their ru1e, but also deflected it into first
liberal and then social democratic opposition.
(4) There is another negative reason we must deal with here. One
reason why many Marxists dispute the argument I am trying to sustain is
that to allow that class struggle might have significant effects is to imply
that reformism could be a viable political strategy. But this does not in
fact follow. To argue that working-class and other popu1ar struggles have
highly significant effects within capitalist societies is not to imply that
such societies can be transformed by reforms into socialist societies. This
is because of the structured organisation of such societies. As we saw in
Navarro's discussion of the health service, what is important is the
 Capitalism and Representative Democracy 147

process of transmutation, of how changes produced by popular pressure

are transmuted into something less acceptable to those popular forces.
This has the following paradoxical consequence. Many of the changes
produced in this manner within capitalist societies satisfy none of the
major classes in such societies. The end result has been unintended by any
major social element. It would seem that an essential feature of further
explanation here would lie in developing a basis on which to account
systematically for such transmutations.
(5) It seems that the most effective condition under which representa-
tive democracy comes to be established is that of warfare. 19 In particular,
in only three 'western' countries has the attainment of universal suffrage
been unrelated to foreign war (these being Australia, New Zealand and
Switzerland). The importance of warfare works in one of two ways.
Either, the existing non-democratic regime is defeated in war, and as in
the cases of Germany (after both World Wars), Italy and Japan, this led to
broadly democratic regimes being established. Alternatively there is
national mobilisation against a foreign country and this produces
significant extensions of the franchise, in Norway, for example, as a
result of its unresolved dispute with Sweden, or in Belgium and Britain
during the First World War. However, in every case this condition is
associated with that of a large-scale popular movement, of labour, or of
the agricultural petty-bourgeoisie, or of women, or of an ethmcally
disadvantaged category, or by some combination of these. Therborn
concludes from his survey that: 'Bourgeois democracy has always
succeeded mass struggles of varying degrees of violence and protracted-
ness. The first inherent tendency will be found in the conditions favouring
popular struggle. '20 Thus, to the extent that capitalist development has
generated conditions favouring such struggles, such as literacy, com-
munications, large cities and work-places, etc., then capitalist relations
have helped to generate representative democracy. However, generally
the growth of suffrage has needed other factors as well, a defeated or a
divided ruling class, or one trying to generate loyalty in a time of national
crisis, or one without a dominant reactionary landed element.
(6) The final reason why it is correct to argue that popular struggle is
essential to franchise and other reforms is because of the obvious strength
of opposition to such reforms in societies where they are threatened.
Again, this demonstrates the weakness of Barrington Moore's argument
in which, in nineteenth-century Britain, he minimises the strength of both
popular agitation and of reaction and control. Johnson maintains that the
new working class, not only made itself ala E. P. Thompson, but also
148 The Anatomy of Capitalist Societies

substantially made nineteenth-century society. As opposed to Moore,

Johnson argues that 'it is not so much middle class agitation nor the
amateur culture of aristocracy that made England' 'relatively free" , but a
plebian agitation and culture'. 21 This can be seen in a number of areas: the
achievement of representati ve democracy, the liberty of the press and of
public meeting, and the preventing of a fully bureaucratised state as
envisaged by the Chadwickian programme of Poor Law Reform.
I have not attempted to provide an account of all the determinants of
whether capitalist society will generate a powerful enough popular
movement sufficient to realise a representative democratic state. Two
considerations so far ignored are: first, the degree of differentiation
between the state and the economy, and second, the degree of
differentiation between the state and civil society. On the first, where the
state is separate from the economy, then it can be seen as representing a
different kind of organisation, which can be employed to control that
economy more effectively. In other words, to the degree to which the
economy is separate, then this strengthens the reformist basis within that
society. This will be much less marked where the state controls the
economy more directly. And, on the second, the greater the differentia-
tion between the state and society, the more that classes and other social
groupings will seek representation within the state, rather than to control
it directly . And that consists essentially ofthe representation of individual
citizens, not of classes or other social groupings. This again serves to
support reformism; namely, that the franchise should be extended to that
particular category, so reforming the relationship between civil society
and the state. Clearly, this condition consists in the existence of a viable
sphere of civil society which in turn depends upon an institutionalised
sphere of circulation separate from production. The most important
condition of this lies in the existence of free wage-labour system rather
than a labour-repressive system.
It might be argued against my position that, even if it were the true that
a fundamental condition ofthe development of bourgeois democracy has
been proletarian agitation, it does not follow that such democratic forms
are not highly functional for capitalist social relations. Indeed it could be
that such functions enable us to explain, not the origin of bourgeois or
representative democracy, but its persistence. 22 I will not consider the
general dimensions of these arguments here, merely, the particular form
put forward by Poulantzas. He argues that representative democracy,
first, permits the different fractions of capital to bargain with each
other, so enabling a compromise to emerge which is in the long-term
 Capitalism and Representative Democracy 149

interests of capital; second, it enables the institutionalisation of popular

struggles and hence of social reformism; and third, it atomises classes and
other social groupings into individualised citizens subject to a national
rather than a class interest.
The first point seems reasonable except for the fact that there may be
contradictory interests between, say, industrial and financial capital
which cannot be so resolved. According to Longstreth, British capitalism
has long been characterised by a dominance of finance capital, and this
has been achieved, not through Parliament, but through the Bank of
England and the Treasury.23 Is there in fact anything which can be
designated as the 'long-term interests of capital '? This relates to the more
general presumption in this whole style of argument; namely, its class
essentialism. It is assumed that each of the major classes has an essential
interest, in the working class to destroy capitalist relations, in the
capitalist class to promote and further the long-term interests of capital.
But once we recognise theoretically the importance of civil society then
these essential interests cannot be sustained. We find individual agents
struggling to further their material interests, atomistically or collectively.
This mayor may not yield common interests, but, on the worker side, this
cannot be expressed simply as the destruction of capitalist relations. And
on the capitalist side no simple identificatIon is possible of what would be
in the long-term interests of 'capital'. Thus, if we consider the argument
that representative democracy is crucially important in disorganising the
dominated classes, we can see that there are erroneous assumptions built
into this. First, there is the assumption that class struggle, without the
workings of representative democracy, will effect the political unification
ofthe working class. Yet this is simply not correct. Indeed we have seen
that in that period in which the sphere of reproduction is dominant in civil
society, then this will make such unification less not more likely, for
reasons that have little to do with representative democracy itself.
Second, it is presumed that the main relation of the state to the sellers of
labour-power is political and not economic; in other words, to disorganise
the dominated class or classes. This claim is doubtful because it fails to
explain why the state might function in this way, given the general
rejection of instrumentalist theories of the state. Just what is the
mechanism which ensures that the state functions politically to atomise
and disorganise the dominated classes?
Finally, in this discussion of representative democracy let us consider
Jessop's discussion. He correctly states that whether democracy is the
best political shell for capitalism is a contingent matter and one not
150 The Anatomy of Capitalist Societies

dependent on the innate qualities or effects of such institutions. 24 It

depends on the political conjuncture within which such democratic
institutions are located. In particular, the bourgeois democratic republic
is the best possible political shell for capital to the extent that the bour-
geoisie is politically and ideologically dominant. Yet, when the bour-
geoisie is not dominant at the levels of both political forces and of the
articulation of ideologies. then the rule of capital is seriously threatened,
and this may result in either an exceptional form of the state or the
instituting of a socialist republic. What are the difficul ties in this account?
The main problem is again that of class essentialism. Thus, it is
presumed that the working class, if politically dominant, will seek to
institute a socialist repUblic. This is never explicated; no reasons are thus
given as to why such a dominance would not result in a social democratic
state aimed at running capitalism more efficiently but also more fairly. Of
course, 'dominance' might be defined by Jessop in a way such that this
reformist solution would not be seen as one in which the working class
was in fact dominant. Indeed the very notion of dominance here is highly
problematic. Political dominance of the bourgeoisie is perhaps a
relatively clear notion provided we ignore the problem of class fractions,
although whether the bourgeoisie can be said to be politically dominant
during a Labour government in Britain is contentious. But ideological
dominance is much more problematic. Has the bourgeoisie been
ideologically dominant in post-war Britain? Was it indeed dominant in
early-nineteenth-century capitalism? Are the subordinate classes always
dominated ideologically by the superordinate classes? Abercrombie et al.
argue that no such dominant ideology thesis can be plausibly sustained. 25
My own argument on the importance of the realm of civil society would
similarly support a more pluralist view of competing ideologies. There
are further difficulties in Jessop's approach which stem from its
over-politicised conception of the state and from its functionalism.
In general, Jessop does not demonstrate that where the bourgeoisie is
dominant (politically and/or ideologically), then representative democ-
racy is the best political shell for capital. It may well be that the dominant
bourgeoisie is incapable of restructuring capital in appropriate manner,
perhaps because of its inability to form an effective power bloc. In such a
case the best political shell for capital would in fact be one which
permitted social democratic parties (representing the working class and
the people) dominance within the relevant state apparatuses in order to
effect the appropriate restructuring. In this case then representative
democracy would be the best possible political shell for capitalism,
 Capitalism and Representative Democracy 151

precisely because it enabled political domination to be in part wrested

from the bourgeoisie. It would not be the best political shell for capital
were the bourgeoisie to continue its pattern of political domination. In
this case the best political shell would probably be a military-
authoritarian one to force through restructuring against the political
opposition of the established bourgeoisie. Jessop, like so many commen-
tators, forgets that the working class and other popular forces have an
irredeemable interest in the structuring of 'its' capital. Also, in the epoch
in which capital has become internationally centralised, it is impossible to
state what is the best political shell for 'capital', does this mean national
capital or international capital? In such circumstances it may be that it is
progressively the political representatives of the bourgeois class who
have to be politically excluded from the state, in order that the interests of
national capital can be so sustained.
This discussion is inconclusive; I have referred to various consider-
ations which Jessop has ignored. Thus, while he is correct to maintain
that the conditions under which democracy is the best political shell for
capital depend on the political conjuncture, this needs to be much more
systematically elucidated. Indeed, it involves analysis of capital, of the
form of civil society, and of its interrelations with the state. Analysis must
also not presume the kind of class essentialism which pervades his
argument. This means that we should question the conventional Marxist
notion of class interest in a much more thoroughgoing manner than in
Crouch's recent paper.26 He maintains that Marxist theory is particularly
good as deriving the interests of the capitalist class because capitalism is
the structure in dominance and its interests are expressed as a continuous
series of incremental decisions aimed at maintaining that dominance.
But, this is to state the problem far too simply. It ignores the point most
emphasised by so-called Althusserians; namely, the division of capital
into different fractions, commercial, industrial and financial. But more
importantly it implies that there is one way, or series of ways, in which
'capitalism' can be maintained, and that if these policies and measures are
implemented then this is to pursue capitalist interests. Yet capitalism is a
worldwide system which happens to be organised into particular
nation-states. What is the interest of capitalism in Britain, to strengthen
British capitalism, or American capitalism; is it to concede demands to
the British working class, so increasing the rate of relative surplus-value
production, or not to concede so strengthening the hands of capitalists in
other competitors countries who also wish not to concede? Is it to
encourage social democratic parties who will more effectively restructure
152 The Anatomy of Capitalist Societies

national capital, or to encourage more 'capitalistic' parties who will

favour the interests of worldwide capitalism? In each case the answer is
unclear.
And, on the side of the working-class interests, equally severe problem
arise. Crouch argues that working-class interests can be stated as follows:
within capitalism labour is subject to constant subordination. Thus, it has
an interest in the development of any mode of production that will reduce
or eliminate that subordination. Hence, it is in its interests to support
revolutionary strategies provided two very stringent criteria apply: first,
that there is a reasonable expectation that the result of the change will be a
system which will reduce or eliminate domination; and second, that the
material gains already achieved will not be put in jeopardy. But even if we
accept this formulation, it still begs many questions. In particular, what
actually is a 'revolutionary strategy'; and more especially how do we
assess whether changes ('reforms') in a particular capitalist society are in
working-class interests? Crouch seems to see such changes as mainly
material, hence they should not be put in jeopardy and when they are, by a
conservative,labourist or Marxist party, then such interests are infringed.
But in this book I have argued in a variety of ways that by considering the
realm of civil society, how classes and other forces interrelate, then the
straightforward identification of interests is far more problematic than
this. It is, for example, quite often the case that both labour and capital
have a common interest, at least in the short run in the reproduction of the
commodity labour-power. More specifically, we often have to consider
not simply the working class but the conflict of interest within it, between
men and women, say. Furthermore, the notion of a 'power bloc'
sensitises us to the kinds of conflicting alliances and relations which are
necessary between different groupings in order for particular policies to
be pursued and hence for apparently contradictory interests to be realised.
Civil society is, moreover, the locus of popular democratic politics which
is expressed within and supported by a set of distinctive social practices.
This has to be fought for and won over to socialist politics - it is neither
necessarily capitalist or anti-capitalist.
I have engaged in this brief discussion of interest in order to provide
some defence for my apparently heretical argument, that because ofthe
changing form of capital accumulation, it is increasingly the case in many
capitalist societies that democracy is the best political shell. But it is the
best political shell, albeit one already tending to corporatism27 because it
penn its labour and its representatives, as well as the representatives of
other popular social forces, a vastly greater degree of determination than
 Capitalism and Representative Democracy 153

would otherwise be the case. Contemporary capitalism is one in which

the representatives of labour are in general part of the state, rather than
simply of civil society. Furthermore, the effects of such determination are
generally of benefit in the necessary restructuring of the each national
capitalist economy; in enlarging the role of the state; in reorganising
private capitals; and in ensuring the necessary reproduction of labour-
power. But the forms in which these policies are manifest are not such
that working-class and popular interests are necessarily satisfied thereby.
Structures develop, unintended by my single class, fraction or force,
which contradict those interests. So while it is incorrect to characterise
this as a bourgeois or capitalist state, in the sense that that class controls
the state, we have the following paradoxical situation. In contemporary
capitalism, with a general dominance of the sphere of reproduction, a
variety of working-class and popular forces significantly affect the form
and nature of the state and of state politics, often from within. The state is
thus subject to a distinctively changed relationship with civil society - a
civil society that is strengthened and increasingly resilient, that is less
directly dominated through the state. Yet the effect ofthis increasingly
diversified and effective civil society is that the class and popular
elements within it have no way of controlling and moderating capitalist
relations except through the state, since this is the alternative to capital.
So over time, and with reversals, the state comes to be strengthened,
enlarged, bureaucratised, internationalised. However, it cannot satisfy
all the demands being made ofit,28 nor avoid functioning in part to harm
the interests of these classes and forces whose partial effectivity within
the civil society in part helps to generate such developments. Modern
capitalist societies are thus characterised both by stronger states and by
stronger civil societies, both of which are responses to crises of
accumulation.
In this book I have therefore attempted to deal in some detail with what
Thompson calls the circuits of power and ideology, of what I consider to
be the circuits interlinking civil society and the state. I hope to have
shown that these are complex and contradictory, and therefore that
notions of base/superstructure, or ofthe economic/political/ideological,
should be placed once-and-for-all in the dustbin of history. It has been
argued that there is a realm of civil society which mediates between the
economy and the state. I have suggested that it is comprised of three
spheres and that struggles within it are diverse and complex in their
effects. Although such struggles cannot in normal circumstances re-
volutionise capitalist society, they do have various effects on that society,
154 The Anatomy of Capitalist Societies

in particular on the state and its attempt to sustain the overall conditions
for capital accumulation. It is necessary to consider how different forms
of struggle produce as effects different state forms and policies which,
although sustaining general capitalist relations, do differentially affect
the various classes and social forces engaged in struggle. The realms of
civil society, which in part is structured by capitalist relations of
production, delimits the potential forms of struggle - their often unin-
tended effects being in many cases functional for capital accumulation. I
have thus tried to provide the barest outlines of an approach which
enables us to theorise the role of class and popular struggle within
capitalist relations.
Notes and References

Chapter 1

1. See Poggi (1978) for an interesting non-Marxist discussion of the increasing

interdependence between state and society; see Winkler (1977), for example, on
corporatism. On base and superstructure see Hall (1977). I have been considera-
bly influenced by many of Stuart Hall's recent writings.
2. This was very marked at the 1978 British Sociological Association
Conference, for example.
3. My account of these positions is of course, highly schematic; no single
author would fall unambiguously into one or other position. Yaffe (1975),
however, is a good example of a reductionist, Hirst (1976) of an autonomist.
4. For a very clear example of this see Domhorst and Newman (1977).
5. See Althusser (1969) and Poulantzas (1973).
6. See Hirst (1976) and (1979); and see the discussion in Hall (1978).
7. See Cutler et al. (1978) pp. 264 ff.
8. See ibid. pp. 233-93.
9. See Althusser (1969) pp. 166 ff. I ignore the problem of whether there is a
fourth practice within social formations, that of scientific practice.
10. See Chapter 4 for some discussion of this concept.
11. See Althusser (1971) and Poulantzas (1973).
12. See Oarke (1977) for a characteristic example.
13. See Poulantzas (1973). For criticisms of Marxist functionalism see Crouch
(1979) and Maravall (1979); for a very recent and sustained critique of
functionalism see Giddens (1979).
14. Thompson (1978) p. 346. Also see Oarke (1977) on the links between
Althusserianism and structural-fuctionalism, although his own position seems
similarly functionalist.
15. See Dawe (1970).
16. See Maravall (1979), and Przeworski (1977), who takes a position
somewhat similar to my own (see pp. 65 ff.).
17. See Crouch (1979) p. 27 on these two points.
18. See (1980); as well as Abercrombie and Turner (1978).
19. See Hindess and Hirst (1977).
156 Notes and References to pages 7-20

20. This was one of the main claims in Althusser's and Poulantzas's relatively
early writings; see the introduction to Poulantzas (1973), for example.
21. See the very helpful discussion in Foster-Carter (1978). Also see Martin
(1979) on the difficulty of theorising the articulation of the feudal and capitalist
modes.
22. See Delphy (1977), for example, on the former.
23. See Balibar (1970) for a well-known attempt to argue the contrary position.
24. See Bhaskar (1975), Harre (1970), Harre and Madden (1975). among
others.
25. On rationalist epistemology and its critique see Hindess and Hirst ( 1977).

Chapter 2
1. See Marx (1959) p. 791 for an example ofthis position. Miliband (1969) is
the best-known contemporary example.
2. See Mandel (1975) chap. 15.
3. See the collection of readings edited by Holloway and Picciotto (1978),
where every element of capitalist society seems to be either 'capital' or the 'state'.
4. See the discussions of this interpretation in Geras (1972) and Mepham
(1972).
5. Marx (1959) p. 791. This is one consequence ofa realist interpretation of
Marx; see Keat and Urry (1975) chap. four.
6. Marx (1976) p. 176.
7. Marx (1973a) p. 107.
8. See Hindess (1978) p. 97.
9. See, for example, Althusser (1976).
10. See Hall, Lumley and McLennan (1977).
11. Poulantzas (1973) p. 124. Useful discussions of Poulantzas can now be
found in Clarke (1977), Oarke et al. (1977), Wright (1978) chap. 2, A general
account of my own views can be found in Abercrombie et al. (1976). Poulantzas,
auto-critique can be found in (1976). For Poulantzas's recent writing on the state
see (1978). There is incidentally some similarity between my views and that of
Jakubowski (1976), who talks of the economic base, the legal and political order,
and the ideological superstructure which 'crowns' the model.
12. Poulantzas (1973) p. 128
13. Ibid. p. 65; see Oarke (1977) on this argument.
14. Poulantzas (1973) p. 63, n. 8.
15. This seems notto be true oflater work; see, for example, Poulantzas (1975)
p. 18.
16. Gramsci (1971) p. 208. There is a further question highlighted by Bobbio
(1979) and Texier (1979): as to whether Marx's and Gramsci's concepts of civil
society are different. See p. 24.
17. See Clarke (1977) pp. 3-4.
18. Ibid. p. 3; see Miliband (1969) and Anderson (1977).
19. Marx (1933) p. 29.
20. See ibid. pp. 28-9.
21. Clarke (1977) p. 10.
 Notes and References to pages 20-34 157

22. Ibid. p. 10.

23. Ibid. p. 10. Lovell (1978) embodies this in even more forthright fashion.
24. I am indebted here to Anderson's analysis (1977) of the Prison Notebooks.
Other recent helpful articles are Hall et al. (1977), Mouffe and Sasson (1977) and
Mercer (1978), as well as the collections edited by Mouffe (1979a) and Davis
(1979).
25. Gramsci (1971) p. 208.
26. Ibid. p. 263.
27. Ibid. p. 160.
28. Ibid. pp. 208, 209, 235.
29. Ibid. p. 235.
30. Ibid. p. 238.
31. Ibid. pp. 161, 181, 182.
32. Ibid. p. 59.
33. Ibid. p. 12.
34. pp. 239, 246.
35. See Althusser (1971). On this interpretation of civil society see Buci-
Gl ucksmann (1974), as well as the brief discussion in Mouffe and Sassoon (1977)
p. 47 ff.
36. See Gramsci (1971) pp. 234-5 and 242-3.
37. See Mouffe (1979b) pp. 188-98, especially the critique of Poulantzas
(1973), who treats Gramsci as a fairly orthodox historicist.
38. Gramsci (1971) p. 241.
39. See Bobbio (1979) and the discussion in Mouffe and Sassoon (1977)
pp. 43 ff.
40. See Mercer (1978) on this.
41. Hall et al. (1977) p. 62.

Chapter 3.
1. In Althusser (1969). I am not going to say anything about Hegel's
conception of civil society - nor about Marx's discussion of the concept that
occurs elsewhere. My purpose here is not primarily exegetical - merely to
provide some pointers for my own argument.
2. Ibid. p. 110.
3. Marx and Engels (1968) pp. 48-9.
4. Ibid. p. 48.
5. Ibid. p. 79.
6. Marx (1976) p. 1007.
7. See Oarke (1977) in the discussion above; as well as the recent book on the
state and capital edited by Holloway and Picciotto (1978).
8. Marx (1976) p. 280; see also Marx (1973) pp. 239-47.
9. Marx (1976) p. 268.
10. Ibid. p. 269.
11. Ibid. p. 279.
12. Ibid. p. 279.
13. Ibid. p. 280.
158 Notes and References to pages 35-52

14. Marx (1959) pp. 882r-3.

15. See Marx (1973a) pp. 88-100; and Carver (1975) part l.
16. Marx (1973a) p. 100.
17. Ibid. p. 99.
18. Marx (1959) p. 814; see the discussion of this in Sayer (1979) chap. 3.
19. See Marx (1959) pp. 879-80, Marx (1956) p. 57, as well as Rosdolsky
(1977) pp. 31-5 for a more general discussion.
20. See Marx (1969) part 2, p. 152.
21. See Howard and King (1975) pp. 129-43, Rosdolsky (1977) pp. 282-313.
22. See Howard and King (1975) pp. 199-203.
23. See Meek (1973) p. 172.
24. Brown and Browne (1968) p. 344.
25. Marx (1976) p. 275.
26. See Glyn and Sutcliffe (1972).
27. Marx (1965) p. 74.
28. Ibid. p. 74.

Chapter 4

1. See Abercrombie (1980), Barrett (1979), Hirst (1979), Larrain (1979) and
Sumner (1979) for some recent contributions.
2. See Larrain (1979) pp. 50-Ion the distinction between 'ideology' and
'idealistic superstructure', which roughly mirrors mine between 'ideology' and
'civil society'.
3. Marx and Engels (1958) p. 363 and Marx and Engels (1968) more
generally. See Abercrombie and Turner (1978) for a fuller discussion.
4. Marx and Engels (1968) p. 31.
5. Ibid. p. 37.
6. Ibid. pp. 659-67.
7. Marx and Engels (1958) p. 363.
8. Marx and Engels (1968) p. 61.
9. Marx (1973b) p. 173.
10. See Marx (1976) pp. 163-77. The best discussion of this is in Rubin
(1972), which is interestingly criticised in Cutleretal. (1977) chap. 3. Otherwise
see the articles by Geras (1971), Mepham (1972) and Rose (1977).
11. See McDonnell (1978). My understanding of this article has been helped by
seeing and discussing an unpublished critique by Brian Longhurst.
12. This is discussed in a different way by Haberrnas (1976), when he analyses
how the market no longer provides a successful base for legitimation in advanced
capitalism.
13. See Althusser (1971). There are by now many discussions, the best is
undoubtedly Hirst (1976) reprinted in Hirst (1979). Otherwise, see McLennan et
al. (1977), Ranciere (1974), Sumner (1979), as well as Althusser (1976) for his
self-criticism.
14. See Hirst (1976) pp. 390 ff.
15. See Hindess (1978), among a number of sources here.
16. Althusser (1971) p. 139.
 Notes and References to pages 53-68 159

17. Ibid. p. 139.

18. See Poulantzas (1973).
19. Lenin (1963) p. 11.
20. See Gramsci (1971), as well as Althusser (1971) p. 136 and David (1978),
on the family-education couple.
21. See Sumner (1979) chap. 2 for a discussion of a number of these
points.
22. Althusser (1971) p. 163.
23. Ibid. p. 164.
24. See Hirst (1976) p. 400.
25. See Hirst (1979) p. 18.
26. Ibid. p. 12.
27. This is argued in a number of different places in this book, see especially
Chapters 1 and 5.
28. See Hindess and Hirst (1977), especially p. 10. See the reviews by Collier
(1978), McLennan (1978) and Skillen (1978).
29. Hindess and Hirst (1977) p. 73.
30. Ibid. p. 73.
31. Hirst (1979) p. 21.
32. See ibid. p. 38.
33. See Johnson (1979b) p. 59, who also makes this old-fashioned point. Also
see Johnson (1979a).
34. The importance of 'concealment' is much stressed in Larrain (1979).
Otherwise see Miller (1972) pp. 443-4 and Keat and Urry (1975) chaps 4 and 8,
on some of these points. My discussion of 'ideological effect' is only intended to
be suggestive. For example, the concept of 'interest' opens up a Pandora's box of
troublesome issues.

Chapter 5
1. See Gramsci (1971); as well as the very useful discussion in Johnson
(1979b) p. 72.
2. See Johnson (1979b) p. 72; Marx (1976) p. 275.
3. There is some similarity here with E. P. Thompson's extraordinary critique
of Althusser; see Thompson (1978) p. 356 on the category of experience.
4. Ibid. pp. 260-1.
5. Ibid. p. 260.
6. Ibid. p. 351.
7. See Cutler et al. (1977) chap. 13.
8. It is worth noting how Poulantzas (1975), for all his structural determination
by three instances, tends to employ something of the same distinction.
9. See Coward (1977) and Coward and Ellis (1977) for texts influenced by the
Cutler et al. (1977) and (1978) position.
10. See Laclau (1979) pp. 104-5 on this very useful distinction. More
generally see Przeworski (1977) on the deficiencies of the 'in-itself'/'for-itself'
distinction.
11. Ibid. p. 108.
160 Notes and References to pages 68-78

12. See Marx and Engels (1958) pp. 362-3, Generally on this issue see the very
helpful article, Molina (1971) esp. pp. 236-46.
13. Marx (1976) p. 92.
14. Ibid. p. 179.
15. Marx (1959) p. 822.
16. Ibid. pp. 8, 9.
17. Marx (1976) p. 254.
18. Marx (1973a) p. 244; and see p. 165.
19. On the concept of structuration see Giddens (1973) and Scott (1979) pp.
109-10.
20. This is a very contentious and discussed issue. See my brief account on this,
Urry (1973). For more sophisticated formulations, see Carchedi (1977), as well
as the discussion of many of these issues in Przeworski (1977).
21. See Wright (1978) when he suggests that individuals in contradictory class
locations will be more affected by political/ideological factors; as well as
Jakubowski (1976) pp. 59-60.
22. Jakubowski (1976) p. 59.
23. Ibid. p. 59.
24. See Habermas (1974) pp. 43-64, and the very useful discussion in Sensat
(1979) chap. 4.
25. See Coward and Ellis (1977), for example.
26. Laclau (1979) p. 102.
27. Ibid. p. 103.
28. This is best seen in the domestic labour debate; see note 38 below.
29. See Comer (1974).
30. See Coward and Ellis (1977), Mitchell (1975).
31. See Kuhn (1978) for an alternative and interesting attempted synthesis.
32. I have not the space here to discuss whether there is a distinctive patriarchal
mode of production. See Delphy (1977) for the most interesting discussion of
this. Since she herself also calls it a 'system' of production, I doubt that much is
lost by viewing it as such and not as a mode.
33. See Humphries (1977).
34. Ibid. p. 242.
35. Ibid. pp. 244-5. See below for some literature on the domestic labour
debate.
36. See Zaretsky (1976), as well as Foreman (1977) and Smith (1978).
37. See Mitchell (1975) on Levi-Strauss's discussion of this.
38. See Smith (1978) for a paper I have drawn on heavily in the following. I am
not implying here that domestic work could not be socialised by the state; indeed I
have just raised that possibility. But even if it was, it would still not be fully
subsumed under commodity production. Also see Himmelweit and Mohun
(1977) for a very useful summary. Kuhn and Wolpe (eds) (1978) contain a helpful
collection of relevant papers.
39. Marx (1976) p. 718.
40. See Oakley (1974) for one of the few empirical studies of domestic work.
41. See Liddington and Norris (1978) for an interesting account of the 'radical
suffragists'. i.e. women textile workers, who are to be distinguished from the
'militant suffragettes', who mainly concentrated on gaining the vote. On
 Notes and References to pages 79-86 161

women's work see Beechey (1978), especially the discussion of dual labour-
market theory.
42. See Delphy (1977). She points out that until 1965 a husband in France could
legally prevent his wife from working; see p. 12.
43. See McIntosh (1978) pp. 282-3 on functionalism, and throughout for these
points made here. Also see Ginsburg (1979) chap. 4.

Chapter 6
1. See Jessop (1977) pp. 354-6, Frankel (1979), McMurtry (1978) chap. 4,
and Jessop (l978b) are also helpful recent discussions.
2. See Marx (1970).
3. See Marx and Engels (1968) p. 45; and Marx (1975) pp. 230 ff.
4. See Marx and Engels (1958) pp. 362-4.
5. See Marx (1 73c) p. 69.
6. See Marx and Engels (1962) p. 319.
7. See Engels (1934) p. 306.
8. I am indebted here to the helpful suggestions in Jessop (1977) pp. 353-4.
9. See Pashukanis (1978).
10. For an example of this see Holloway and Picciotto (1977); for a critique of
this and a general discussion, see Barker (1978), von Braunmuhl (1978), as well
as Jakubowski (1976) p. 43.
11. See Marx (1973c) p. 69.
12. On Poulantzas, see his (1973) p. 277 for example. Also see Poulantzas
(1974) for a critique of the Comintern interpretation of fascism. The general
relationship between the state and democracy is very interestingly discussed in
Jessop (1978a); see Chapter 9.
13. Miliband (1969) p. 25.
14. Poulantzas (1973) p. 44.
15. See Jessop (1977) pp. 267-9 on neo-Gramscian theories of the state. See
also Poulantzas (1974, 1975) for somewhat revised interpretations of the state.
16. See especially Holloway and Picciotto (1977), (eds) (1978).
17. See in the following order: Poulantzas (1969), Miliband (1970), Miliband
(1973), Laclau (1975), and Poulantzas (1976).
18. Holloway and Picciotto (eds) (1978), 'Introduction', p. 3.
19. See Marx and Engels (1958) p. 503.
20. See Holloway and Picciotto (eds)( 1978), 'Introduction', p. 5, for example.
21. Ibid. p. 12. See Gough (1975) for an interestingly clear statement of a
neo-Ricardian view of the state; for a critique see Yaffe and Bullock (1975), as
well as Yaffe (1975).
22. Holloway and Picciotto (1977) p. 84.
23. There is a general tendency in the Conference of Socialist Economists to see
almost everything in capitalism as a fetishism, or fetishised appearance. One
wonders if fetishism means anything more than 'objective structure'. For an
example of this all-embracing conception of fetishism see McDonnell (1978), and
in Chapter 4. For a critique of the related conception ofreification see Keat and
Urry (1975) chap. 8.
162 Notes and References to pages 87-105

24. Holloway and Picciotto (eds) (1978), 'Introduction', p. 30.

25. Holloway and Picciotto (1977) p. 94; see also Barker (1978). I am grateful
to Scott Lash for some elucidation on form and content.
26. See Miiller and Neusiiss (1970), and the critiques by Habermas (1975) and
affe (1975). In general I will only refer to literature in the German debate
available in translation; where items are not readily available I will refer to
original sources. See Gerstenberger (1976) for a very helpful overview. A good
collection of articles is in Gesellschaft Beitriige zur Marxschen Theorie, 1
(Frankfurt - Main, 1974).
27. See Habermas (1974) and affe (1972), for example.
28. See Miiller and Neusiiss (1970) pp. 60. ff.
29. Quoted in ibid. p. 63.
30. Ibid. p. 72.
31. Although there are now tendencies operative which reduce the strength of
such illusion, see ibid. pp. 87-90.
32. Ibid. p. 84.
33. See Holloway and Picciotto (eds) (1978) pp. 19-31, and Blanke et al.
(1978) pp. 120-1.
34. See Aatow and Huisken (1973), the critique by Reichelt (1978) and Chapter
3.
35. Marx and Engels (1968) p. 45.
36. See Boccara (1971), as well as Blanke et at. (1978); pp. 120 ff. It should be
noted that I don't consider the Stamocap tradition any further in this book. See
Wirth (1977).
37. See Altvater (1973a, 1973b); and Barker (1977) for a critical discussion.
38. Altvater (1973a) pp. 98-9.
39. See Blanke et al. (1978).
40. See Hirsch (1974), (1976), (1977), (1978) for fairly succinct accounts of his
position, as well as Holloway and Picciotto (1977) (eds) (1978).
41. Pashukanis (1978) p. 141.
42. Hirsch (1978) pp. 65-6.
43. Ibid. p. 100.

Chapter 7

I. Marx (1975) p. 412.

2. On these various points there is of course a vast literature: for some that has
become recently available, see Pashukanis (1978), Poggi (1978), Poulantzas
(1978) and Wright (1978).
3. Lenin (1963) p. II.
4. This would seem to be the likely conclusion of Hindess and Hirst's recent
work: see Hindess (1977) for a clear example of this.
5. See Cockburn (1977) on the analysis of the local state.
6. On male violence see Hanmer ( 1978). I shall not consider here whether it is
therefore appropriate to consider that there is a 'patriarchal state'.
7. See Offe (1972) on this issue.
8. See Poulantzas (1973) part v.
 Notes and References to pages 105-120 163

9. See Mandel (1975) pp. 474-99 as well as Jessop (1978a), who discuss the
thesis of alternating periods of 'parliamentary' and 'executive' rule, depending on
the intensity and nature of the class struggle.
10. See Wright (1978) pp. 181-225 on the contrasting if partially related views
of Lenin and Weber.
11. Poulantzas thus seems mistaken in arguing that the establishment of a power
bloc is a necessary feature of capitalist states. However, his discussion of this
concept is nevertheless important; see (1973) pp. 229-52.
12. This is presented and criticised in a somewhat antagonistic piece by Simon
Oarke (1978).
13. See Pashukanis (1978); recent helpful discussions are Arthur (1977)
(1978), Collins (1979), Kinsey (1979), and Redhead (1978). Also see Lash
(1979) more generally on law, class and the state. Balbus (1977) is an
interestingly similar although independently worked out analysis of the legal
form.
14. See Arthur (1978) p. 11; an example of this would be the nevertheless
interesting Griffiths (1977).
15. Pashukanis (1978) p. 121.
16. See ibid. p. 139; as well as Arthur (1978) pp. 16 ff., and Lash (1979) pp.
19-20.
17. Arthur (1978) p. 17.
18. The fomer is argued by Picciotto (1979); the latter in part by Redhead
( 1978).
19. Pashukanis (1978) p. 137.
20. See Blanke et al. (1978) p. 109.
21. Ibid. p. 124.
22. Ibid. p. 127.
23. Ibid. p. 127.
24. Marx (1973a) p. 420.
25. See Navarro (1978) on the development of the health service in the United
Kingdom.
26. See Yaffe and Bullock (1975). Generally see Gouph (1979) on these
various arguments, and Ginsburg (1979) for an attempt to transcend the
dichotomy.
27. It would seem that Balbus (1977) p. 572 is wrong to argue that simply
because a ruling class-theory of the law is invalid (as in Stuchka), so a theory
which relates law to the nature of the capitalist economy is thereby validated.

Chapter 8
1. See Oarke (1977) pp. 3-4; as well as pp. 00-0 above,
2. Navarro (1978) p. 11; also see Ginsburg (1979) on social security provision
and council housing.
3. Dangerfield (1966) argues, for example, that in 1911: the workers were
interested in wages, not health; and whatever may be said for or against the
Insurance Act, one cannot maintain that it raised anybody's pay'. (p. 245)
4. See Navarro (1978) pp. 109-16 on the capitalist state.
164 Notes and References to pages 120-134

5. See ibid. pp. 86-7. One might note incidentally that many doctors are in
fact the sons (and occasionally the daughters) of other doctors, not of the capitalist
class proper.
6. See Johnson (1977) p. 106 on the eMP and the medical profession.
7. See Laclau (1979) p. III in particular. Ignoring this issue is one deficiency
oflan Gough's otherwise very interesting book on the welfare state, see Gough
(1979).
8. See Hall et al. (1978) pp. 55 ff.
9. See Holloway and Picciotto (eds) (1978) pp. 9-10 on this classification of
approaches to the state.
10. See Aumeeruddy et al. (1978).
11. The second part of Aumeeruddy et al. 's article, pp. 52-60, deals with 'The
State and the "Management" of the Collective Labourer'. I will not discuss this
here because they almost entirely ignore the state and what they have to say
amounts to very little.
12. This is in Aumeeruddy et al. , p. 43. This quotation does not actually come
from anyone, the closest quote in the relevant footnote comes from Altvater, who
would presumably be most upset to discover that he had adopted such an
instrumentalist position. See my discussion of Altvater in Chapter 6.
13. Aumeeruddy et al., p. 43.
14. Ibid. p. 50.
15. Ibid. p. 51.
16. Ibid. p. 49.
17. Marx (1976) p. 280.
18. See Holloway and Picciotto (1977) pp. 86-91.
19. Blanke et at. (1978) p. 142.
20. See Beechey (1978).
21. See Therborn (1977); more generally on these issues see the classic
discussion of citizenship in Marshall (1950).
22. There is some similarity here with the arguments advanced in Esping-
Anderson et at. (1976).
23. This is discussed in the recent book by Parkin (1979).
24. I have not discussed this in the normal terms of an institutionalisation of
class conflict. My argument is consistent with that developed in Trotsky (1934);
and see Giddens (1973) and Mann (1973).
25. See O'Connor (1973) on the fiscal crisis of the state.
26. On many of these points see the neo-Gramscian, Perez-Diaz (1978),
especially pp. 44, 69-72.
27. See Esping-Anderson et al. (1976) pp. 199 ff.
28. This is extensively discussed by Yaffe and Bullock (1975).
29. See Cutler et at. (1978) p. 263, for an interesting discussion of the
importance of struggling for non-commodified policy.
30. See Holloway and Picciotto (1977). Also see the discussion in Chapter 6.
31. Ibid. p. 97.
32. See Barker (1978), Poulantzas (1975) and von Braummiihl (1978) among
recent contributors to the analysis of this issue.
33. Holloway and Picciotto (1977) p. 85.
34. Ibid. p. 96.
 Notes and References to pages 134-141 165

35. See the discussion of Altvater (1973a) at pp. 93-4.

36. In the United Kingdom, for example, nationalisation by the 1945-50
Labour Government invol ved the centralisation of 550 electricity companies, 800
mining companies, 1000 gas companies, and 3800 road haulage companies, into
four nationalised industries. See Pryke (1971) p. 21.
37. In Britain the closures of mines and railways are notorious examples of this
tendency; see ibid. p. 21.
38. In Britain, for example, between 1958 and 1968 (a period of generally
rising employment) the total labour force fell from 1,920,000 to 1,310,000, a fall
from 8.3% of all civil employment to 5.4%. Partly as a consequence, the
productivity of the nationalised industries increased between 1958 and 1967 by
5.3% per annum, compared with 3.4% p.a. for private capital; see ibid. pp.
64-77.
39. See Open University (1972) pp. 1-5.
40. This does not necessarily entail massive redundancies, although it usually
does. In the 1960s in the United Kingdom the reduction in labour-power
employed occurred in coalmining simply through decreasing the number of
bearers of labour-power; in electricity it occurred through reducing the hours
worked by each bearer. It should also be noted that I am assuming that the Labour
Party in the United Kingdom is the main form of political representation of the
British working class. See Chapter 9 for my objections to the kinds of class
essentialist arguments which would imply the rejection of this claim.
41. As in Bismarckian Germany and the nationalisation of the railways.
42. This was the case in post-1945 France.
43. See McEachern ( 1979) p. 141 in particular. This is a very useful discussion
of the role of the Labour Party in the nationalisation of the steel industry.
44. See Giddens (1977) p. 19 for a brief discussion of 'critical phases'. See
Warde (1980) on the nature of the post-1945 structure of struggle and conflict
embodied in the 'welfare state'.
45. See the discussion in Offe (1972) on the contradictions of state strategies, as
well as Saunders (1979) pp. 173 ff.
46. See Habermas (1976) on the notion ofa 'legitimation crisis'; on the 'crisis
of hegemony' see Hall et al. (1978) chap. 7 especially p. 215.
47. I shall not consider whether the fact that the British state seems consistently
to damage the long-term interests of domestic capital means that such economic
and political forms are necessarily in their later ('deathbed ') stages. I think it
should be clear that no such political/social changes necessarily follow such
economic/political contradictions.

Chapter 9
1. Friedman (1962).
2. Macpherson (1966) p. 4.
3. Lenin (1963) p. 296.
4. Macpherson (1966) p. 4.
5. This is the thesis that Jessop develops in (1978a); this is easily the most
useful article on this topic. See also Therborn (1977).
166 Notes and References to pages 141-153

6. Macpherson (1966) pp. 6 ff.

7. Ibid. p. 9.
8. See Moore (1968); for useful discussions see Johnson (1976) and Wiener
(1975).
9. See ibid. p. 418.
10. See ibid. p. 414.
11. See Johnson (1976), especially p. 16, for a very helpful discussion ofthis
point, and for some of the following argument.
12. See Miliband (1969) for a classic example of this.
13. See Mann (1970) and (1973) for the best-known expositions of this critique.
14. See the discussion in Chapter 6.
15. See the recent discussion of this in Crouch (1979) pp. 24 ff.
16. Therbom (1977) pp. 17 ff. convincingly shows that none of the great
bourgeois revolutions led directly to the establishment of representative democ-
racy. I assume universal franchise to be the most important aspect of
representative democracy.
17. Engels (1962) pp. 96-7.
18. See Johnson (1976) pp. 24-6.
19. See Therbom (1977) on this point.
20. Ibid. p. 29.
21. Johnson (1976) p. 17.
22. See Isajiw (1968) on this distinction. On these various arguments see
Poulantzas (1973) and (1976).
23. See Longstreth (1979); as well as Grant and March (1977) on the relative
weakness of the CBI, that is, the form of representation of industrial capital.
24. See Jessop (l978a) pp. 39-40.
25. See Abercrombie et al. (1980).
26. Crouch (1979) pp. 33 ff.
27. On corporatism, see the contributions to Comparative Political Studies
(April 1977); and Panitch (1978).
28. See Buchanan and Wagner (1977) and Rose and Peters (1977) among many
others on the notion of the 'overloaded' state. On this argument more generally
see Swingewood (1977).
Bibliography

ABBREVIATIONS

Am. Soc. Rev. American Sociological Review

Brit. J. Soc. British Journal of Sociology
BSA British Sociological Association
Camb. J. Econ. Cambridge Journal of Economics
German Pol. Stud. German Political Studies
J. for the Th. of Soc. Behaviour Journal for the Theory of Social Behaviour
NLR New Left Review
Pol. Stud. Political Studies

Abercrombie, N. (1980) Class, Structure, and Knowledge (Oxford: Black-

well).
- Hill. S., and Turner, B. (1980) The Dominant Ideology Thesis (London:
Allen & Unwin)
- and Turner, B. (1978) 'The Dominant Ideology Thesis', Brit. J. Soc. 29,
149-70.
- - and Urry, J. (1976) 'Oass, State and Fascism: The Work of Nicos
Poulantzas', Pol. Stud. 24, 510-19.
A1thusser, L. (1969) For Marx (Harmondsworth: Penguin Books)
- (1971) 'Ideology and Ideological State Apparatuses', in Lenin and
Philosophy and Other Essays (London: New Left Books) pp. 121-73.
- (1976) Essays in Self-Criticism (London: New Left Books).
- and Balibar, E. (1970) Reading 'Capital' (London: New Left Books).
Altvater, E. (1973a) 'Some Problems of State Interventionism: I',Kapitalistate,
1,96-116.
- (1973b) 'Some Problems of State Interventionism: II', Kapitalistate, n,
76-83.
Anderson, P. (1977) 'The Antinomies of Antonio Gramsci', NLR, 100,
5-78.
Arthur, C. (1977) 'Towards a Materialist Theory of Law', Critique, 7, 31-46.
- (1978) 'Introduction' to Pashukanis (1978) pp. 9-32.
168 Bibliography

Aumeeruddy, A. T., Lautier, B., and Tortajada, R. (1978) 'Labour-Power and

the State', Capftal and Class, 6, 52-66.
Balbus, I. (1977) 'Commodity Form and Legal Form: an essay on the 'relative
autonomy' of the law', Law and Society Review, 11, 571-88.
Balibar, E. (1970) 'The Basic Concepts of Historical Materialism' , in Althusser
and Balibar (1970) pp. 199-308.
Barker, C. (1977) 'On Altvater', unpublished, CSE Manchester Group.
- (1978) 'A Note on the Theory ofthe Capitalist State', Capital and Class, 4,
118-26.
Barrett, M., Corrigan, P., Kuhn, A., and Wolff, J. (eds) (1979) Ideology and
Cultural Production (London: Croom Helm).
Beechey, V. (1978) 'Women and Production: a critical analysis of some
sociological theories of women's work', in Feminism and Marxism, ed. A.
Kuhn and A. M. Wolpe (London: Routledge & Kegan Paul) pp. 155-97.
Benton, T. (1977) Philosophical Foundations of the Three Sociologies (London:
Routledge & Kegan Paul).
Bhaskar, R. (1975) A Realist Theory of Science (Leeds: Alma).
- (1978) 'On the Possibility of Social Scientific Knowledge and the Limits of
Naturalism', J. for the Th. of Soc. Behaviour, 8, 1-28.
Blanke, B., Jurgens, U., and Kastendiek, H. (1978) 'On the Current Marxist
Discussion of the Analysis of Form and Function' , in Holloway and Picciotto
(eds) (1978) pp. 108-47
Bobbio, N. (1979) 'Gramsci and the Conception of Civil Society', in Mouffe
(ed.) (1979a).
Boccara, P. (1971) 'Zum Staatsmonopolistischen Kapitalismus', Sozialistiche
Politik, 11.
Braunmiihl, C. von (1978) 'On the Analysis ofthe Bourgeois Nation State within
the World Market Context', in Holloway and Picciotto (eds) (1978) pp.
160-77.
Brown, E. H. Phelps, and Browne, M. (1968) A Century of Pay (London:
Macmillan).
Buchanan, 1. M., and Wagner, R. E. (1977) Democracy in Deficit: the political
legacy of Lord Keynes (New York: Academic Press).
Buci-Glucksmann, C. (1974) Gramsci et l' Etat (Paris: Fayard).
Carchedi, G. (1977) On the Economic Identification of Social Classes (London:
Routledge & Kegan Paul).
Carver, T. (ed.) (1975) Karl Marx. Texts on Method (Oxford: Blackwell).
Clarke, J., Connell, I., and McDonough, R. (1977) 'Misrecognising Ideology:
ideology in political power and social classes', in Working Papers in Cultural
Studies, 10, On Ideology, pp. 106-22.
Clarke, S. (1977) 'Marxism, Sociology and Poulantzas' Theory of the State',
Capital and Class, 2, 1-31.
- (1978) 'Capital, Fractions of Capital, and the State: "neo-Marxist" analyses
of the South African state', Capital and Class, 5, 32-77.
Cockburn, C. (1977) The Local State (London: Pluto Press).
Collier, A. (1978) 'In Defence of Epistemology ',Radical Philosophy, 20, 8-21.
Collins, H. (1979) The Marxist Theory of the Form of Law , British Sociological
Association Conference, Warwick.
 Bibliography 169

Comer, L. (1974) Wedlocked Women (Leeds: Feminist Books).

Coward, R. (1977) 'Class, "Culture" and the Social Formation', Screen, 18,
75-105.
Coward, R., and Ellis, J. (1977)Language and Materialism (London: Routledge
& Kegan Paul).
Crouch, C. (ed.) (1979) The State, Capital and Liberal Democracy', inState and
Economy in Contemporary Capitalism, ed. C. Crouch (London: Croom Helm)
pp. 13-54.
Cutler, A., Hindess, B., Hirst, P., and Hussain, A. (1977)( 1978) Marx' s Capital
and Capitalism Today (London: Routledge & Kegan Paul) vols 1 and 2.
Dangerfield, G. (1966) The Strange Death of Liberal England (London: Paladin
Books).
David, M. E. (1978) 'The Family - Education Couple: towards an analysis of the
William Tyndale dispute', in Power and the State, ed. G. Littlejohn et al.
(London: Croom Helm) pp. 158-95.
Davis, J. (ed.) (1979) Gramsci and Italy's Passive Revolution (London: Croom
Helm).
Dawe, A. (1970) The Two Sociologies', Brit. J. Soc. 21, 207-18.
Delphy, C. (1977) The Main Enemy: a materialist analysis of women's
oppression (London: Women's Research and Resources Centre).
Domhorst, R., and Newman, P. (1977) 'Which Way Forward for Communists?
Critique of the British road to socialism' ,Revolutionary Communist, 6, 4-21.
Engels, F. (1934) Anti-Duhring (Moscow: Foreign Languages).
- (1962) On Britain (Moscow: Foreign Languages).
Esping-Anderson, G., Friedland, R., and Wright, E. o. (1976) 'Modes of Class
Struggle and the Capitalist State', Kapitalistate, 4-5, 186-220.
F1atow, S. von, and Huisken, F. (1973) 'Zum Problem der Ableitung des
biirgerlichen Staates', Probleme des Klassenkampfs, 7, 83-129.
Foreman, A. (1977) Femininity as Alienation (London: Pluto).
Foster-Carter, A. (1978) The Modes of Production Controversy', NLR, 107,
47-77.
Frankel, B. (1979) 'On the State of the State', Theory and Society, 7, 199-242.
Friedman, M. (1962) Capitalism and Freedom (London: Chicago V.P.).
Geras, N. (1971) 'Essence and Appearance: aspects of fetishism in Marx's
Capital', NLR, 65 69-85.
Gerstenberger, H. (1976) Theory ofthe State: special features ofthe discussion
in the Federal Republic of Germany', in German Political Studies, ed. K. von
Beyme, vols 1 and 2 (London: Sage) pp. 69-92.
Giddens, A. (1973) The Class Structure of the Advanced Societies (London:
Hutchinson).
- (1977) Studies in Social and Political Theory London: Hutchinson).
- (1979) Central Problems in Social Theory: action, structure and contradic-
tion in social analysis (London: Macmillan).
Ginsburg, N. (1979) Class, Capital and Social Policy (London: Macmillan).
Glyn, A., and Sutcliffe, B. (1972) British Capitalism: workers and the profits
squeeze (Harmondsworth: Penguin Books).
Gough, I. (1975) 'State Expenditure in Advanced Capitalism' ,NLR, 92, 53-92.
- (1979) The Political Economy of the Welfare State (London: Macmillan).
170 Bibliography

Gramsci, A. (1971) Selections from the Prison Notebooks, ed. Q. Hoare and
P. Nowell Smith (London: Lawrence & Wishart).
Grant, W., and Marsh, D. (1977) The CBI (London: Hodder & Stoughton).
Griffiths, J. A. G. (1977) The Politics of the Judiciary (London: Fontana).
Habennas, J. (1974) Theory and Practice (London: Heinemann).
- (1975) 'A Reply to Miiller and Neusiiss', Telos, 25,91-8.
- (1976) Legitimation Crisis (London: heinemann).
Hall, S. (1977) 'Rethinking the' 'Base-and-superstructure" Metaphor' , in Class,
Hegemony and Party, ed. J. Bloomfeld (London: Lawrence & Wishart) pp.
43-72.
- (1978) 'Some Problems with the Ideology/Subject Couplet', Ideology and
Consciousness, 3,113-21.
- Critcher, C., Jefferson, T., Oarke, J.;and Roberts, B. (1978) Policing the
Crisis: mugging, the state, and law and order (London: Macmillan).
- Lumley, B., and Mclennan, G. (1977) 'Politics and Ideology in Gramsci', in
Working Papers in Cultural Studies, 10, On Ideology, pp. 45-76.
Hanmer, J. (1978) 'Violence and the Social Control of Women " in Power and the
State, ed. G. Littlejohn et al. (London: Croom-Helm) pp. 217-38.
Harre, R. (1970) The Principles of Scientific Thinking (London: Macmillan).
- and Madden, E. H. (1975) Causal Powers (Oxford: Blackwell).
- and Secord, P. (1972) The Explanation of Social Behaviour (Oxford:
Blackwell).
Himmelweit, S., and Mohun, S. (1977) 'Domestic Labour and Capitai', Camb. J.
Econ. I, 15-31.
Hindess, B. (1977) 'Marxism and Parliamentary Democracy', unpublished.
- (1978) 'Oasses and Politics in Marxist Theory', In Power and the State, ed.
G. Littlejohn et al. (London: Croom Helm) pp. 72-97.
- and Hirst, P. (1977) Mode of Production and Social Formation (London:
Macmillan).
Hirsch, J. (1974) 'Scientific-Technical Progress and the Political System',
German Pol. Stud. 1, 107-39.
- (1976) 'Bemerkungen zum theoretischen Ansatz einer Analyse des biirger-
lichen Staates' , Gesellschaft Beitrijge zur Marxschen Theorie, 8/9, 99-149.
- (1977) 'What is the Fiscal Crisis of the State? On the political function of
fiscal crisis, Conference of Socialist Economists, Bradford.
- (1978) 'The State Apparatus and Social Reproduction: elements of a theory of
the bourgeois state', in Holloway and Picciotto (eds) (1978) pp. 57-107.
Hirst, P. (1976) 'Althusserand the Theory ofIdeology', Economy and Society, 5,
385-412.
- (1979) On Law and Ideology (London: Macmillan).
Holloway, J., and Picciotto, S. (1977) 'Capital, Crisis and the State' , in Capital
and Class, 2, 76-101.
- - (eds) (1978) 'Introduction', and editors of State and Capital (London:
Arnold).
Howard, M., and King, J. (1975) The Political Economy of Marx (London:
Longman).
Humphries, J. (1977) 'Oass Struggle and the Persistence of the Working-class
Family', Camb. J. Econ. 1,241-58.
 Bibliography 171

Isajiw, W. W. (1968) Causation and Functionalism in Sociology (London:

Routledge & Kegan Paul).
Jakubowski, F. (1976) Ideology and Superstructure in Historical Materialism
(London: Allison & Busby).
Jessop, B. (1977) 'Recent Theories of the Capitalist State' , Camb. 1. Econ. I,
353-73.
- (1978a) 'Capitalism and Democracy: the best possible shell', in Power and
the State, ed. G. Littlejohn et al. (London: Croom Helm) pp. 10-51.
- (1978b) 'Marx and Engels on the State', in Politics, Ideology and the Statt.
ed. S. Hibbin (London: Lawrence & Wishart) pp. 40-68.
Johnson, R. (1976) 'Perry Anderson, Barrington Moore and English Social
Development', in Working Papers in Cultural Studies, 9, Culture and
Domination, pp. 7-28.
- (1979a) 'Culture and the Historians', in J. Oarke, C. Critcher and R.
Johnson, Working Class Culture: studies in history and theory (London:
Hutchinson) pp. 41-71.
- (1979b) 'Histories of Cul ture I Theories of Ideology: notes on an impasse', in
Barrett et al. (1979) pp. 49-77.
Johnson, T. (1977) 'Professions in the Oass Structure', in Industrial Society,
Class, Cleavage and Control, ed. R. Scase (London: Allen & Unwin) pp.
93-110.
Keat, R., and Urry, J. (1975) Social Theory as Science (London: Routledge &
Kegan Paul).
Kinsey, R. (1979) 'Pashukanis - Law and Marxism', Head and Hand, pp.
16-20, Spring.
Kuhn, A. (1978) 'Structures of Patriarchy and Capital in the Family', in Kuhn and
Wolpe (eds) (1978) pp. 42-67.
- and Wolpe, A. M. (eds) (1978) Feminism and Marxism (London: Routledge
& Kegan Paul).
Laclau, E. (1975) 'The Specificity of the Political', Economy and Society, 4,
87-110.
- (1979) Politics and Ideology in Marxist Theory (London: New Left Books).
Larrain, J. (1979) The Concept of Ideology (London: Hutchinson).
Lash, S. (1979) Law, Class and the Order of Labour, Conference of Socialist
Economists Law and State Group.
Lenin, V.1. (1963) The State and Revolution, in Selected Works, vol. 1 (Moscow:
Foreign Languages).
Liddington, J., and Norris, J. (1978) One Hand Tied Behind Us (London:
Virago).
Longstreth, F. (1979) 'The City, Industry and the State', inState and Economy in
Contemporary Capitalism, ed. C. Crouch (London: Croom Helm) pp.
157-90.
Lovell, T. (1978) The Social Relations of Ideological Production: the absent
centre of a new discourse, Conference of Socialist Economists, Bradford.
McDonnell, K. (1978) 'Ideology, the Crisis and the Cuts' , Capital and Class, 4,
34-69.
McEachern, D. (1979) 'Party Government and the Oass Interest of Capital:
conflict over the steel industry', Capital and Class, 8, 125-43.
172 Bibliography

McIntosh, M. (1978) 'The State and the Oppression of Women', in Kuhn and
Wolpe (eds) (1978) pp. 254-89.
McLellan, G. (1978) 'Correspondence', Economy and Society, 7, 193-205.
- Molina, V., and Peters, R. (1977) 'Althusser's Theory of Ideology', in
Working Papers in Cultural Studies, 10, On Ideology, pp. 77-105.
McMurtry, J. (1978) The Structure of Marx's World-View (Princeton, N. J.:
Princeton V.P.).
Macpherson, C. B. (1966) The Real World ofDemocracy (Oxford: Blackwell).
Mandel, E. (1975) Late Capitalism (London: New Left Books).
Mann, M. (1970) 'The Social Cohesion of Liberal Democracies', Am. Soc. Rev.
35,423-39.
- (1973) Consciousness and Action among the Western Working Class
(London: Macmillan).
Maravall, J. M. (1979) 'The Limits of Reformism: parliamentary socialism and
the Marxist theory of the state', Brit. 1. Soc. 30,267-90.
Marshall, T. H. (1950) Citizenship and Social Class and Other Essays (London:
Cambridge V.P.).
Martin, J. (1979) 'Peasants and Landlord in the Development of Feudalism and
the Transition to Capitalism in England' (Ph. D., thesis Lancaster V ni versity).
Marx, K. (1933) Wage-Labour and Capital (London: Lawrence & Wishart).
- (1956) Capital, vol. 2 (London: Lawrence & Wishart).
- (1959) Capital, vol. 3 (London: Lawrence & Wishart).
- (1965) Wages, Price and Profit (Peking: Foreign Languages).
- (1969) Theories of Surplus Value, parts I, 2, 3 (London: Lawrence &
Wishart).
- (1970) Critique of Hegel's Philosophy ofRight (London: Cambridge V.P.).
- (1973a) Grundrisse (Harmondsworth: Penguin Books).
- (1973b) Surveys from Exile, ed. D. Fembach (Harmondsworth: Penguin
Books).
- (1973c) The Revolutions of 1848, ed. D. Fembach (London: Allen Lane).
- (1975) Early Writings, ed. L. Colletti (Harmondsworth: Penguin Books).
- (1976) Capital, vol. 1 (Harmondsworth: Penguin Books).
- and Engels, F. (1958) Selected Works, vol. 1 (Moscow: Progress).
- - (1962) Selected Works, vol. 2 (Moscow: Foreign Languages).
- - (1968) The German Ideology (Moscow: Progress).
Meek, R. (1973) Studies in the Labour Theory of Value (London: Lawrence &
Wishart).
Mepham, J. (1972) 'The Theory ofldeology in Capital' ,Radical Philosophy, 2,
12-19.
Mercer, C. (1978) 'Culture and Ideology in Gramsci', Red Letters, 8, 19-40.
Miliband, R. (1969) The State in Capitalist Society (London: Weidenfeld &
Nicolson).
- (1970) 'The Capitalist State: a reply to Nicos Poulantzas', NLR, 59,53-60.
- (1973) 'Poulantzas and the Capitalist State', NLR, 82, 83-92.
Miller, D. (1972) 'Ideology and the Problem of False Consciousness', Pol. Stud,
20,432-47.
Mitchell, J. (1975) Psychoanalysis and Feminism (Harmondsworth: Penguin
Books).
 Bibliography 173

Molina, V. (1977) 'Notes on Marx and the Problemoffudividuality', in Working

Papers in Cultural Studies, 10, On Ideology, pp. 236-64.
Moore, B. (1968) The Social Origins of Dictatorship and Democracy (London:
Allen Lane).
Mouffe, C. (ed.) (1979a) Gramsci and Marxist Theory (London: Routledge &
Kegan Paul).
Mouffe, C. (1979b) 'Hegemony and Ideology in Gramsci', in Mouffe (ed.)
(1979a) pp. 168-204.
Mouffe, c., and Sassoon, A. S. (1977) 'Gramsci in France and Italy - a review
of the literature', Economy and Society, 6, 31-68.
MUller, W., and Neususs, C. (1970) 'The Illusions of State Socialism and the
Contradictions Between Wage-Labour and Capital', Telos, 25, 13-90.
Navarro, V. (1978) Class Struggle, the State and Medicine (Oxford: Martin
Robertson).
Oakley, A. (1974) The Sociology of Housework (London: Martin Robertson).
O'Connor, J. (1973) The Fiscal Crisis of the State (New York: St. Martin's
Press).
Offe, C. (1972) 'Political Authority and Class Structures: an analysis of late
capitalist societies', Internat. J. of Soc. n, 73-108.
- (1975) 'Further Comments on MUller and Neususs', Telos, 25,99-111.
Open University (1972) Nationalised Industries (The Open U.P.) D203, IV.
Panitch, L. (1978)Recent Theorizations of Corporatism: reflections on a growth
industry, Ninth World Congress of Sociology, Uppsala, Sweden.
Parkin, F. (1979) Marxism and Class Theory. A bourgeois critique (London:
Tavistock).
Pashukanis, E. (1978) Law and Marxism, a General theory, ed. C. Arthur
(London: Ink Links).
Perez-Diaz, V. M. (1978) State, Bureaucracy and Civil Society (London:
Macmillan).
Picciotto, S. (1979) The Theory ofthe State, Class Struggle ,and the Rule ofLaw ,
BSA Conference, Warwick.
Poggi, G. (1978) The Development of the Modern State (London: Hutchinson).
Poulantzas, N. (1969) 'The Problem of the Capitalist State' ,NLR, 58, 67-78.
- (1973) Political Power and Social Class (London: New Left Books).
- (1974) Fascism and Dictatorship (London: New Left Books).
- (1975) Classes in Contemporary Capitalism (London: New Left Books).
- (1976) 'The Capitalist State: a reply to Miliband and Laclau', NLR, 95,
68-83.
- (1978) State, Power, Socialism (London: New Left Books).
Pryke, R. (1971) Public Enterprise in Practice (London: MacGibbon & Kee).
Przeworski, A. (1977) 'Proletariat into a Class: the process of class formation
from Karl Kautsky's The Class Struggle to recent controversies' , Politics and
Society, 4, 343-401.
Ranciere, J. (1974) 'On the Theory of Ideology (the Politics of Althusser)',
Radical Philosophy, 7, 2-15.
Redhead, S. (1978) 'The Discrete Charm of Bourgeois Law: a note on
Pashukanis', Critique, 9, 113-20.
Reichelt, H. (1978) 'Some Comments of Sybille von Flatow and Freerk
174 Bibliography

Huisken's Essay "On the Problem of the Derivation of the Bourgeois State" "
in Holloway and Picciotto (eds) (1978) pp. 43-56.
Rosdolsky, R. (1977) The Making of Marx's Capital (London: Pluto).
Rose, N. (1977) 'Fetishism and Ideology: a review of theoretical problems',
Ideology and Consciousness, 2, 27-56.
- and Peters, G. (1977) The Political Consequences of Economic Overload
(University of Strathclyde, Centre for the Study of Public Policy).
Rubin, 1.1. (1972) Essays on Marx's Theory ofValue (Chicago: Black and Red).
Sassoon, A. S. (1978) 'Hegemony and Political Intervention', in Politics,
Ideology, and the State, ed. S. Hibbin (London: Lawrence & Wishart) pp.
9-39.
Saunders, P. (1979) Urban Politics (London: Hutchinson).
Sayer, D. (1979) Marx's Method: Ideology, science and critique in 'Capital'
(Hassocks, Brighton: Harvester).
Scott, J. (1979) Corporations, Classes and Capitalism (London: Hutchinson).
Sensat, J. (1979) Habermas and Marxism (London: Sage).
Skillen, T. (1978) 'Post-Marxist Modes of Production ',Radical Philosophy, 20,
3-8.
Smith, P. (1978) 'Domestic Labour and Marx's Theory of Value', in Kuhn and
Wolpe (eds) (1978) pp. 198-219.
Sumner, C. (1979) Reading Ideologies (London: Academic Press).
Swingewood, A. (1977) The Myth of Mass Culture (London: Macmillan).
Texier, J. (1979) 'Gramsci, Theoretician of the Superstructures', in Mouffe (ed.)
(1979a) pp. 48-79.
Therborn, G. (1977) 'The Rule of Capital and the Rise of Democracy , ,NLR, 103,
3-41.
Thompson, E. P. (1978) The Poverty of Theory (London: Merlin).
Trotsky, L. (1934) History of the Russian Revolution (London: Gollancz).
Urry, J. (1973) 'Towards a Structural Theory of the New Middle Gass',Acta
Sociologica, 16, 175-87.
Warde, A. (1980) Consensus and Beyond: the transformation of Labour Party
strategy since the Second World War (Manchester: Manchester U.P.).
Wiener, J. M. (1975) 'The Barrington Moore Thesis and Its Critics', Theory and
Society, 2, 301-30.
Winkler, J. (1977) 'The Corporatist Economy: Theory and Administration', in
Industrial Society: class, cleavage and control, ed. R. Scase (London: Allen &
Unwin) pp. 43-58.
Wirth, M. (1977) 'Towards a Critique of the Theory of State Monopoly
Capitalism', Economy and Society, 6, 284-313.
Wright, E. O. (1978) Class, Crisis and the State (London: NLB).
Yaffe, D. (1975) The State and the Capitalist Crisis·, VESV Conference,
Amsterdam.
- and Bullock, P. (1975) 'Inflation, the Crisis and the Post-war Boom',
Revolutionary Communist, 3-4, 5-45.
Zaretsky, E. (1976) Capitalism, the Family and Personal Life (London: Pluto).
Index of Names

Abercrombie, N. 7 Hirst, P. 7, 51, 56-60

Althusser, L. 2-4,13,16,27-8,44, Holloway, J. 84-8, 132-4
49-57,82 Huisken, F. 91-2
Altvater, E. 93-4, 97 Humphries, J. 75- 6
Anderson, P. 18,21-2 Hussain, A. 65-6
Arthur, C. 108-9
Aumeeruddy, A. 123-4 Jakubowski, F. 71
Jessop, B. 149-51
Bentham,1. 32-3, 125 Johnson, R. 63, 146-8
Blanke, V. 94, 110-11, 125 Jurgens, U. 94
Bobbio, N. 24
Kastendiek, H. 94
Clarke, S. 18-20, 118
Conference of Socialist Lacan, J. 55-6
Economists 4 Lac1au, E. 66- 8, 72-3
Crouch, C. 151-2 Lautier, B. 123-4
Cutler, A. 65-6 Lenin, V. I. 54, 103, 140
Longstreth, F. 149
Engels, F. 75, 80, 146 Lumley, B. 13

Feuerbach, L. 46 McDonnell, K. 48-9

Flatow, S. von 91-2 McIntosh, M. 79
Freud, S. 55 McLennan, G. 13
Friedman, M. 140 Macpherson, C. B. 140-2
Marx, K. 8, 10-20, 26- 8, 30- 42,
Gramsci, A. 10, 12-13, 18-25,54, 44-9,63-4,68-9,71-2,75,
63 77-8,80,82,84-91,99,111-12,
134
Habermas, J. 71-2,88 Miliband, R. 18-19, 83-4, 86
Hall, S. 13 Moore, B. 142-4, 146- 8
Hegel, G. 35,45-6 MUlier, W. 88-91, 101
Hill, S. 7
Hindess, B. 7, 57- 60 Navarro, V. 118-20, 146
Hirsch, J. 94-8 NeusUss, C. 88-91, 101
176 Index of Names

Offe, C. 88, 90 Therbom, T. 147

Thompson, E. P. 64-5, 147, 153
Pashukanis, E. 82, 95, 107-11 Tortajada, R. 123-4
Picciotto, S. 84-8, 132-4 Turner, B. 7
Poulantzas, N. 2, 4, 13-18, 26-7,
53, 82-4, 86, 148-9 Weber, M. 26
Simme1, G. 87
Stuchka, P. 108 Zaretsky, E. 76
Index of Subjects

action theory 5 Factory Acts 89-90, 101, 111, 146

autonomism 1-3, 65-6 fractionalism 105- 6
functionalism 4-6, 31, 45, 79,
capital 12, 36- 8; and the state 81-3, 93-5, 124, 145-6, 149-50
84-98
castration complex 55 gender inequality 74-9
circulation 17-18,26-43,69,88,
91-2, 106, 125-32 Hegelianism 35,45-6, 87-8
civil society 10-25, 27-32,42, hegemony 19, 21-5,63-4, 139
50-4,63-79,99-116,120-1;
gender interpellation 72-3; ideology 3-4, 14, 23, 32, 44-63,
121; dominant ideology thesis 7,
spatial-temporal interpellation
150; 'class' theory 45-7;
72-3;sphereofcirculation 26-43,
'fetishism'theory 47-9;
106, 125-32; sphere of reproduction
ideological effects 60-2; 'subject'
73-9, 106, 125-32; sphere of strug-
theory 49-56
gle 106, 125-32, 144-8; and the
ideological state apparatus 4,49-56,
state, see state and civil society
classes 15-17,26-7,41-3,45-7, 82
individual subjects 13-15, 18,25,
52-4, 66-71, 117, 124-39, 142-
54; class essentialism 70-1, 143, 34, 49-56, 68-9, 107-9
industrial reserve army 39- 43
150-4; 'class-struggle' 66, 106,
134; 'classes-in-struggle' 66-7,
labour-power 32-43, 88-91,
106
111-15, 117-39
collective bargaining 40-3
land 37, 142-3, 146
consumption 35-9, 131-2
law 104, 107-11, 115-16
corporatism 121, 130, 152
legitimation crisis 139
distribution 35-43, 131-2
domestic labour 77- 8 National Health Insurance Act,
1911 119-20
epistemology 57-60, 100-1 nationalised industries 93-4, 132-9
exchange 32-9,90--1, 106-9,
125-32 popular democratic struggle 67,
expressive totality 8 124-39, 144-9
178 Index of Subjects

power bloc 83-4,105-6,116,120, society 21-5,27-9, fO-2, 84-5,

132-9, 150-4 99-116,117-39; 141-54; and
production 17-18,26--43, see also capital 92-8, 10 1; and
capital; mode of 7-8, 29-31 crisis 92; and coercion 95-9,
profit, tendency for rate to fall 38, 102-4, 109-10, 123-4; and
127 exchange 84-6,91-2;and
labour-power 88-91, 111-15;
realist philosophy of science 8- 9 andlaw 104,107-11,115-16,
reductionism 1-3,8, 10-12,20, 122-39; and media 121; and
86-7 power bloc 83-4, 105-6, 116,
reformism 2-3, 88-9, 110, 114, 120, 132-9, 150-4; and represen-
136-9, 146-9 tative democracy 3,21-2,81,
140-54; and women 79
state: criteria for a theory of 81- 3, state socialism, illusions of 88-90
100; derivation from capital 84- structuration 70
98; fascism 83, 121; form
'trinity formula' 37
analysis 84- 8; German state
debate 84-98; instrumentalist wages 39-43
theory 18-19, 118; Marxist warfare and the franchise 147
theory of 80-1, 84- 99; rationalist Welfare State 114-15,118-21,
theory of 100-1; and 146-7
bureaucracy 104-5; and civil women 74-9, 117-39, 145, 152