Notes on: Poulantzas, N.  (1980)  State Power Socialism, London: Verso

[This is Poulantzas's last book, and it comes at the end of a long debate with his critics about how actually to do a marxist analysis of the modern state. As some of the stages in the debate with Miliband and Laclau show, an adequate analysis must not be too general, or else it risks turning into a formalist and abstract account of invariant elements which somehow combine to produce specific cases. At the same time, it has to remain within the core elements of Marxist theory, and talk about class struggle and exploitation, or else risk sliding off into bourgeois philosophy or sociology. This book consists of a series of fairly specific essays about aspects of the modern state. Poulantzas tells us that he wishes to avoid any suspicion of offering a general formalist theory. At the same time, he wants to fight off people like Foucault and the  'new philosophers' who were proposing a break with marxism altogether. He tells us in addition that he is not going to offer a fully concrete analysis, because that would take too long, and nor is he going to refer to the classic works of Marx to justify what he's trying to say -- he urges us to consult his earlier works. What follows is a few notes on some of the essays which interested me, but the whole book is fascinating and of considerable contemporary relevance.]

Introduction chapter one. 'On The Theory of the State'
The big issue in the 20th century is how the state, power and social classes are related together. It is now common to see the state as an arena in which the classes struggle for influence. This view necessarily suggest that there is some invariant kernel or function for the state, with some more specific and peripheral aspects over which the parties contend. However, common experience if not theory shows an inherent link between the central activities of the modern state and sectional interests. It is tempting here to see the state as a simple instrument for class dictatorship instead. However, the state is a special apparatus which is not reducible to class interests as directly as this. As history shows, the modern bourgeoisie develop a state to maintain their domination, but why this particular form? There are clearly bits that do not simply express class interests. An adequate theory would argue:  (a) the state is not reducible to the political domination of a class;  (b) state apparatuses are not the same as state power -- the state has an  'institutional materiality' of its own. There is a link to the relations of production and the social division of labour, but it is not as simple as the models reviewed above suggest. Class struggles are still inherent, constitutive even.

Similar arguments apply to notions of the economic base. This is not a set of eternal economic core elements which reproduce themselves, but an aspect of class struggle. The state is not just the superstructure which controls the economy  [see below]. There is no general model of economic, political and ideological 'levels' resting on an economic base, which combine to produce specific forms  [the model developed by Althusser, and Balibar -- see diagram in file]. These views collude with the notion that there is some core essentialist economy which self reproduces. We must abandon the goal of a general theory of economic change as expressing the unfolding inner mechanisms of the economy. The same goes for analysis of the other levels, which also tended to be seen as having an invariant core [function] at the centre of more or specific aspects.

(a) The economic is a site of the relations of production and exploitation  [through the usual mechanisms of the extraction of surplus value]. It never reproduces itself. It has always required a presence of the state, although the precise forms of presence vary.

(b) The state and the economic take on different meanings in different modes of production. They cannot be given an abstract meaning which manifests itself in different combinations. The mode of production itself fixes the boundaries between the levels, and the relations and articulations  'within the unity of the mode of production' (18)  [an improvement on the old notion of determination in the last instance or not?]

(c) The relations are shown in examples like the notion of possessions or private property. For example, property rights were tied in various ways in feudalism, but  'free labour' in capitalism provides no 'traditional' possessions except labour power. The extraction of surplus value in capitalism leads to the relative separation of institutions, and new spaces for the state to occupy. This provides an inherently capitalist form of the political within the economic itself. This form can seem  'fused' [as in ';stamocap'], but  this is only in comparison to feudalism and competitive liberal capitalism.

This is an example of the point that there can be no general theory of constant objects like the state, which risks  'formalist theoreticism' (19). Similarly, there can be no abstract object for sociology or bourgeois politics to study either. We need a theory of the capitalist state and its specific formation. No general theory can be found in Marx, and the work of the new philosophers to rethink the concept of power as an abstract force  [which includes Foucault and Deleuze] is mistaken.

There are deficiencies in Marxist theory which need addressing. There is an evolutionary strand, for example, which sees the capitalist state as the realisation of some pure state  [presumably from Hegel]. There are no general theories of transition to socialism available either, only  'applied theoretical - strategic notions' (22), guides to action.  'There is always a structural distance between theory and practice, between theory and the real' (22). One good consequence of this is that it is nonsense to say that theoretical Marxism somehow produced Eastern totalitarian socialism. It is always possible to use a theory for different ends, and there is no truth in the Popperian line that totalitarian societies must arise from closed theoretical systems. Marxism can be used to illuminate the path of socialist regimes and their transitions, but studying totalitarianism always involves something more than general theory.

(d) It is necessary to separate the concepts of mode of production and social formation. The latter is the site of the state and of class struggles. We should not proceed by attempting to build an ideal type or model  [as does bourgeois sociology]. Nor are social  formations mere concretisations of modes of production  [as in orthodox marxism]: they are  'actual sites of the existence and reproduction of a mode of production' (25). These actual sites require concrete analysis not a general theory.

The relation with the mode of production  'constitutes [the state's] primary relation with social class and the class struggle' (25). That relation determines the form of separation from the relations of production and produces an institutional framework. The production process requires a unity to be established between the labour process and the relations of production (defined as  'the dual relationship of economic property and possession' (26)). The relations of production exercise a primacy over the labour process  [this language of primacy and primary still seems vague -- is it weak determinism or not?]. The productive forces in the labour process do not produce, in crystallisation, the relations of production  [so the old idea that technological change produces social change -- even hinted at in Marx --  is rejected]. Relations of production and productive forces are articulated in a process of production and reproduction, and the notion of primacy involves necessary political and ideological relations present in the productive forces, expressed in class powers. Those political and ideological relations are always present, although the forms of their presence vary according to the mode of production. It is not just the economic that is reproduced, but political and ideological domination itself must be reproduced, and this is the inherent role of the state  'as the factor which concentrates, condenses, materialises and incarnates politico - ideological relations in a form specific to the given mode of production' (27). The primacy of the productive relations also means that a social division of labour necessarily dominates any technical division of labour -- the latter is always  'incorporated' in the former  (27).

The relations of production are therefore inherently within [again, what does this  'within', this  'presence'actually mean? We know it is not a simple presence, but what actually is it?] social classes and power struggles.There are no social classes prior to class struggle [my emphasis]. Social classes are never simply objectively there in society, awaiting the development of consciousness before they are active.

Chapter two  'The ISAs: Does the State equal Repression plus Ideology?

Ideology should be understood as a set of material practices as well as ideas, which become 'cemented' in social life  (28). Ideological relations constitute the relations of possession and private property and the social division of labour. Ideology is always related to social class. Ideology represent an essential aspect of the power of the dominant class. Dominant ideology appears in state apparatuses. Those apparatuses reproduce ideology, especially the ISAs, but so do those apparatuses which specialise in violence. Repression is still important in the modern state. The state does violence to bodies, primarily, including the development of a bodily order or regulation  [so the State lies behind the disciplinary regimes of Foucault?].

However, the division between RSA and ISA is not particularly useful, despite Gramsci's tactical use of it. For Althusser, the state does nothing else but repress and do ideology -- this classically underemphasises the positive role of the state in economic activity itself  (and arises from Althusser's implicit belief that the economy is invariant and self regulating). Althusser assumes that what the state does is to repress,  persuade, and trick, involving a  'structure of concealment/inversion' (30). But the state does organize a genuine consensus, based on material activity as well as ideological trickery. It attempts an 'equilibrium of compromises' [= hegemony], but also offers genuinely significant measures for the benefit of the working class. It did this even in fascism, which underneath the strong ideology also tackled unemployment and improved purchasing power.

Does the state simply conceal its intentions? Modern politics consist of open declarations of strategy, even though the discourse is fragmented and stratified  [mediated differently to different social groups]. The state frequently openly represents the dominant classes and discusses tactics with them. It also generates knowledge of its own activities, including official statistics. Of course there are still areas which are secret, where information is restricted to membership of approved networks, but it is more common that somehow the masses fail to hear what is being openly discussed [what the heck is the evidence for this?] .

Finally, RSAs and ISAs are often connected, and the same institutions often do both -- for example the army  [which also  'wins hearts and minds'].

Chapter three States Powers and Struggles

Marxism classically neglected the role and functioning of the state, but never simply ignored it -- powers emanates from the relations of production and their primacy, as the example of private property shows. It follows that:

(1)  'Power' is not a separate and neglected issue for marxism as Foucault or Deleuze argue. It is inherent in class struggle -- the  'capacity of each class to realise its specific interests in relation opposition to that capacity in other classes' (36)  [a marxified version of Weber's definition of power here?].

(2) Power is not simply focused centrally in the state. The state is defined very narrowly, in terms of its juridical functions, in Foucault and Deleuze. Poulantzas sees the state as involving all the apparatuses of hegemony. These apparatuses can appear as external signs of power, but derive their strength from being part of the state [ a problem glossed here -- see Laclau]. Power exceeds even the state apparatuses, however, and gets extended into knowledge, and consumption [more hints of Weber here, that power is the main analytical tool to describe the more specific institutions]. Class struggles are the fundamental form of power, however.

The state does have a definite constitutive role in social relations, however. There can be no power outside the state, despite libertarian socialist movements such as Socialisme ou Barbarie [and Foucault as we shall see below]. These movements often simply assume that the state is less important, but what they need is a proper analysis of its role. Rather than deciding whether the state actually proceeds class struggle or the other way around, we should see that the state is always implicated in class struggle, and social relations.

The relation between the state in class struggle has been recently misunderstood by the so-called new philosophers, including Foucault  [an interesting aspect of the discussion is that these new philosophers apparently rely on Lacan and his account of the Symbolic as always involving power -- power and intervention therefore becomes the key to all social order, the reverse of the libertarian position]. The issue of precedence is really one of presuppositions rather than (positivist) chronology, however. It must be admitted that Engels is partly responsible for this  'myth of origins' (42).

We return to the point that the relations of power exceed even class relations, although there are still highly  'pertinent' to social class. The relations between men and women are one example. These relations of power still need  institutions to reproduce them and are still the focus of attention by the state, which ensures a class pertinency -- the domination of women is connected to the domination of the working class [very controversial and much debated by feminists, of course -- see file].

So: (a) class power is the cornerstone of power;  (b) political power is primordial, although it is grounded in economic power  [weasel to avoid Weber?];  (c) in capitalism, political power occupies a specific fields, although it is linked throughout the social formation;  (d) political power is  'concentrated and materialized by the state' (44).

Foucault and Deleuze suggests that power operates at a much more diffused and micro-level, but this simply ignores the role of the state and classes. The idea probably originates in American bourgeois sociology, but appears new in French thinking.

The position is therefore more complex than either the libertarians or the  'statists' suggest. Of particular interest is those who accuse Marx of being statist -- that is replacing the private sphere with a centralised state. There is no evidence for this in Marx, but critics follow a  'well-known Stalinist logic: the best proof of people's guilt is the lack of all proof. For if there is no proof, it must be because they have hidden it; and if they have hidden it, then they must be guilty' (46, quoting Ranciere).

Part II, chapter 3  'Towards a Relational Theory of Power?'

[This is a critique of Foucault, but I also think it fits really well those gramscians who wanted to 'elaborate' Gramsci with Foucault and others, including discourse theorists as in Laclau and Mouffe, and abandon the notion of class struggle].

The marxism that Foucault criticises is largely Stalinism. Power is indeed relational, but marxist like Poulantzas have long said so [as in the arguments above about power as a necessary aspect of struggle]. But the power of class struggle varies according to its place in the system of relations between the economic, the political, and the ideological. That depends on the conjuncture and on the position and strategy used towards other classes, the 'opponent strategy' (147). The state is not a thing but an expression of power, a site of organisation of the dominant class. Therefore Foucault gets it wrong:

(1) Power has a precise basis, for example in exploitation, and finds a place in all apparatuses of power, a place in the state. It is  'fundamentally but not exclusively determined by exploitation' (148). Class struggle contaminates all the other struggles. For Foucault, power is somehow  'pure', and  'resistance' must arise, but only in principle, as an abstraction. What is the source of resistance? Other 'new philosophers' have stressed the rights of human beings. Deleuze stresses the mechanism of desire. Foucault simply naturalises power and resistance. One consequence of this is that every struggle only confirms the master role of power -- hence his pessimism.

(2) Struggle has a primacy over state apparatuses, but does not merely try to grab state power as something external. Foucault seems to agree with this, because he argues that power is a relation. However there is an ambiguity -- power refers to the whole process  [especially its positive aspects, where it is linked to knowledge] and/or just one pole of a whole system of power/resistance. This is why domination can never be escaped for Foucault, because it would require something radically outside the entire system of power. He finds this outside force in the  'plebs' who manage somehow to escape [again this is an  'anthropological'concept, a quality inherent in groups or individuals, not a concrete social group]. Plebs are rapidly integrated as soon as they try to get strategic  [the old Weber argument], which leaves only  'emptiness' (151). For Poulantzas, there always are limits to the exercise of power, arising from class struggle. These limits are already inside apparatuses of power, they are integral. There can never be anything outside -- for example, people do not escape the political apparatus by simply not participating in it.

(3) The state is not simple but a  'material condensation of a relationship of forces' (153). The masses are inherently involved in some state apparatuses, such as the school, a conscript army and mass political parties, but are not involved in state apparatuses such as the police, judiciary, and civil service. However, they are still present, at a distance [sic]. This distance or gap between those state apparatuses and the masses constantly has to be bridged or filled. Excessive distance leads to legitimation problems for the state, and an accumulation of internal contradictions  [which sounds like the fatal effects of apathy in Baudrillard -- the masses simply refuse to supply any kind of information, which completely screws up efficient administration].

(4) Must all strategies lead to integration? Not working-class ones -- working-class organisations can manage to remain autonomous even within the relationship of forces. Indeed, effective political action requires this. They can resist incorporation  [I would need to be persuaded of this]. In practice, working-class organisations must constantly manage incorporation in one direction and openness to their members in the other. The most open socialist alternatives, such as self management, or rank and file organisations, still must locate themselves in the conventional relations of power.

back to reading guides