power from actual participation in or possession of ‘parcels’ of institutionalized state power, but rather from the ‘relative autonomy’ which permits the state to provide them with the political unity they otherwise lack.2
The question underlying these theoretical arguments is fundamentally a strategic one: ‘can the state have such an autonomy vis-à-vis the dominant classes that it can accomplish the passage to socialism without the state apparatus being broken by conquest of a class power?’3 Poulantzas’s answer is aimed at specific targets. He attacks ‘instrumentalist’ arguments which treat the state as a mere tool of the dominant classes. He also rejects the other side of the ‘instrumentalist’ coin, the view that the instrument can easily change hands and that, as an inert and neutral tool, it can be wielded as easily in the interests of socialism as it was formerly wielded in the interests of capital.4 In short, Poulantzas is explicitly attacking the theoretical foundations of ‘reformism’ and the political strategy of social democracy. This strategy in effect shares the bourgeois pluralist view that the state can belong to various countervailing interests, and proceeds from there to the conviction that, once representatives of the working class predominate, revolution can be achieved ‘from above’, quietly and gradually with no transformation of the state itself. Indeed, to social democrats, today’s state monopoly capitalism may appear as already a transitional phase between capitalism and socialism. Political and juridical forms, which are in advance of the economy, will simply pull the latter behind them, allowing a piece-meal transition to socialism without class struggle.
At this stage, Poulantzas’s own political prescriptions remain largely implicit, apart from this very general attack on social democracy. Although his theory of the state could be adapted to an assault on Stalinism, as he was later to do explicitly by treating Stalinism as more or less the obverse of social-democratic ‘statism’, in this early work such criticism is muted; and, indeed, there is as yet little that might be construed as ‘anti-statism’. To the extent that the book can be understood as containing implied criticisms of the PCF, they are entirely coded, as they are in Althusser’s work of the time. What can be said about Political Power and Social Classes is that it is generally intended to convey a fidelity to the Leninist tradition, in its Althusserian mediations.
There are, however, important theoretical manoeuvres in this work that have far-reaching political implications. It is here that Poulantzas begins to establish the dominance of the political, going further than his mentor Althusser, and Balibar, in distancing himself from Marxist ‘economism’. He is here perhaps signalling the Maoist leanings which will become more explicit in his next major work, Fascism and Dictatorship’, but he is also fashioning a theoretical instrument that will, as it turns out, continue to be serviceable in his subsequent shift to Eurocommunism.
Poulantzas begins by explaining the circumstances in which the political is ‘dominant’:
… in the global role of the state, the dominance of its economic function indicates that, as a general rule, the dominant role in the articulation of a formation’s instances reverts to the political; and this is so not simply in the strict sense of the state’s direct function in the strictly political class struggle, but rather in the sense indicated here. In this case, the dominance of the state’s economic function over its other functions is coupled with its dominant role, in that its function of being the cohesive factor necessitates its specific intervention in that instance which maintains the determinant role of a formation, namely, the economic. This is clearly the case, for example, in the despotic state in the Asiatic mode of production, where the dominance of the political is reflected in a dominance of the economic function of the state; or again, in capitalist formations, in the case of monopoly state capitalism and of the ‘interventionist’ form of the capitalist state. Whereas in the case of such a form of the capitalist state as the ‘liberal’ state of private capitalism, the dominant role held by the economic is reflected by the dominance of the strictly political function of the state – the state as ‘policeman’ [l’état gendarme] – and by a specific non-intervention of the state in the economic.5
In his later work, this same idea was to be stated more succinctly:
… monopoly capitalism is characterized by the displacement of dominance within the CMP from the economic to the political, i.e. to the state, while the competitive stage is marked by the fact that the economic played the dominant role in addition to being determinant.6
In other words, despite the separation of the economic and the political which is uniquely characteristic of capitalism and which survives in the monopoly phase, because of the expansion of the domain of state intervention the political sphere acquires a position analogous to the ‘dominance’ of the political sphere in pre-capitalist modes of production. Poulantzas even draws an analogy between state monopoly capitalism and the ‘Asiatic mode of production’ in this respect.
This analogy and Poulantzas’s conception of the ‘dominance’ of the political in state monopoly capitalism reveals a great deal about his point of view. His argument is based on the Althusserian principle that, while the economic always ‘determines in the last instance’, other ‘instances’ of the social structure may occupy a ‘determinant’ or ‘dominant’ place. In fact, the economic ‘determines’ simply by determining which instance will be determinant or dominant. This is at best an awkward and problematic idea; but it makes some kind of sense insofar as it is intended to convey that in some modes of production – indeed typically, in precapitalist societies – the relations of production and exploitation may themselves be organized in ‘extra-economic’ ways. So, for example, in feudalism surplus-extraction occurs by extra-economic means since the exploitative powers of the lord are inextricably bound up with his political powers, his possession of a ‘parcel’ of the state. Similarly, in the ‘Asiatic mode of production’ the ‘political’ may be said to be dominant, not in the sense that political relations take precedence over relations of exploitation, but rather in the sense that exploitative relations themselves assume a political form to the extent that the state itself is the principal direct appropriator of surplus labour. It is precisely this fusion of ‘political’ and ‘economic’ that distinguishes these cases from capitalism where exploitation, based on the complete expropriation of direct producers and not on their juridical or political dependence or subjection, takes a purely ‘economic’ form. This is more or less the sense in which Althusser and Balibar elaborate the principle of ‘determination in the last instance’. In Poulantzas’s hands, however, the idea undergoes a subtle but highly significant transformation.7
In the original formula, the relations of exploitation are always central, though they may take ‘extra-economic’ forms. In Poulantzas’s formulation, relations of exploitation cease to be decisive. For him, relations of exploitation belong to the economic sphere; and the ‘economic’ in pre-capitalist societies, and apparently also in monopoly capitalism, may be subordinated to a separate political sphere, with its own distinct structure of domination. It would, of course, be perfectly reasonable for Poulantzas to point out that the role and the centrality of the ‘political’ vary according to whether it plays a direct or indirect role in surplus extraction and whether it is differentiated from the ‘economic’. It would also be reasonable to suggest that the expansion of the state’s role in contemporary capitalism is likely to make it increasingly a target of class struggle. But Poulantzas goes considerably beyond these propositions. He suggests not only that the nature of exploitative relations can vary in different modes of production according to whether they assume ‘economic’ or ‘extra-economic’ forms, but also that modes of production – or even phases of modes of production – may vary according to whether the relations of exploitation are themselves ‘dominant’ at all. When he argues, therefore, that the ‘political’ and not the ‘economic’ is ‘dominant’ in monopoly capitalism, he is in effect arguing that the relations of exploitation (though no doubt ‘determinant in the last instance’) no longer ‘reign supreme’.
III
In 1970, Poulantzas published Fascism and Dictatorship, which represents his most overtly Maoist work. Written in the wake of May 1968, when