If by ‘socialism’ we mean what Marx used to call the higher stage of communist society, the question of the political superstructure that would be adequate to it is raised neither in theory nor in practice.footnote In theory, because this society—to everyone according to his needs—is a classless society, hence unpolitical by definition. In practice, because not only does a society of this kind not yet exist in reality, but more importantly its very ‘problematic’, and indeed even the merest glimpse of the concrete conditions of its advent, are absent from the programmes of all revolutionary parties, whether in power or not. What has, however, constituted a crucial problem in the past, and still constitutes such a problem today, both in theory and in practice, is the political form which corresponds to the period of transition, in its two essential moments—the taking over of power by the proletariat on the one hand, and the substitution of co-operative production for the wage-labour system on the other. Since the publication of The Communist Manifesto, controversies on this theme have been at the centre of all schisms. The length and intensity of such disputes went together with a great ambiguity. The debates seem to have been polarized on two main issues. Can the proletariat come to power by peaceful means? And can it exercise power in a democratic way? Both are artificial problems.

For no political party is violence an end in itself, any more than non-violence could be an autonomous ‘categorical imperative’. These are only means, and between these means and the corresponding ends there is no direct structural correlation. There can be an indirect one, to the extent that the radicalism of the goals determines the toughness of the resistance; but then the correlation is conjunctural, and does not pertain to the doctrine. Furthermore, the extent to which we may theoretically forecast this resistance and make practical preparations to break through it does not pertain to strategy, but to tactics; it cannot, therefore, be the basis for any a priori theoretical position on the need for violence. Even the most opportunist or the most conservative of parties would not hesitate to take up arms to defend itself against a violent attempt by its opponents to usurp a legally assumed power. On the other hand, the most genuinely revolutionary party would be only too happy to avoid civil war, if the circumstances were such that they could assume and hold power peacefully. ‘Peacefully’, however, does not mean ‘legally’, in the sense of a compliance with the rules of the game as established within the existing system; any more than ‘violently’ necessarily means placing oneself outside the legal framework.

To complete this confusion, democratic ways of assuming power are frequently assimilated to the idea of a gradual, evolutionary passage from one mode of production to the other, whilst, on the other hand, open violence is assimilated to a discontinuity of production relations—just as if ‘reform’ was a synonym of non-violence, and ‘revolution’ of insurrection. Yet history is full of examples where mere rivalry between groups representing not social classes but only social strata (in other words, differentiated according to their respective positions in relation, not to the means of production, but to the state superstructure) have degenerated into conflicts more bloody and ravaging than others that have accompanied deep social transformations. A comparative analysis of the upheavals which followed the post-war decolonization in Africa would be telling in this respect. And if we go back to the revolutions that marked the passage from the feudal mode of production to capitalist relations, the lack of symmetry between the magnitude of social changes and the scale of the armed struggles that have punctuated them is still more obvious. The political compromise between the rising bourgeoisie and the great landowners in England formed the basis both for the establishment of a notably pure capitalist mode of production, and for a relatively peaceful passage from the old to the new order. By contrast, the historical necessity in which the French bourgeoisie found itself to rely on the small peasants in its struggle against the feudal lords made it possible to couple extremely bitter, bloody and protracted struggles with an incomplete social transformation and hybrid production relations.footnote1

If the period between the assumption of political power by the proletariat and the latter’s own disappearance were a true transition, such as was visualized by the founders and classics of Marxism—i.e. neither an autonomous mode of production, nor a political system organized for its own sake, but a complex of transformation processes: in other words, an aggregate of acts destroying the old social relations—there would be no break in continuity involved. This period should not be seen as having its own dialectic, but as forming an integral part of the general dialectics of the clash between capitalism and communism. It would represent not a stage in which we settle down, but a thoroughfare that has to be traversed. If these conditions were present, it would be idle to try to gauge the degree of ‘repressively’ or ‘autocracy’ of the system, analyse them in terms of democracy and dictatorship. As with the assumption of power, here too violence is a mere function of the toughness of opposition. The concept of ‘dictatorship of the proletariat’ does not refer to form, but to content. It simply means that, after having seized the government, the proletariat assumes political power openly and directly.

Thus, the relevant questions are of a different kind. Synthetically, they can be placed under three main headings. 1. Is it possible for the proletarian party to seize the bourgeois state apparatus, using for this purpose the latter’s own institutions? 2. In the affirmative case, is it possible for the proletariat to transform society, while maintaining these same institutions? If not, what specific kind of state, particularly regarding means of repression, would be appropriate for this task? 3. When this task is completed, what can assure us that this specific state will wither away—somehow disappear of its own accord—according to traditional Marxist theory?