What are social classes in Marxist theory? They are groups of social agents, of men defined principally but not exclusively by their place in the production process, i.e. by their place in the economic sphere. The economic place of the social agents has a principal role in determining social classes. But from that we cannot conclude that this economic place is sufficient to determine social classes. Marxism states that the economic does indeed have the determinant role in a mode of production or a social formation; but the political and the ideological (the superstructure) also have an important role. For whenever Marx, Engels, Lenin and Mao analyse social classes, far from limiting themselves to the economic criteria alone, they make explicit reference to political and ideological criteria. We can thus say that a social class is defined by its place in the ensemble of social practices, i.e. by its place in the ensemble of the division of labour which includes political and ideological relations. This place corresponds to the structural determination of classes, i.e. the manner in which determination by the structure (relations of production, politico-ideological domination/subordination) operates on class practices—for classes have existence only in the class struggle.footnote¹ This takes the form of the effect of the structure on the social division of labour. But it should be pointed out here that this determination of classes, which has existence only in the class struggle, must be clearly distinguished from class position in the conjuncture. In stressing the importance of political and ideological relations in the determination of classes and the fact that social classes have existence only in the class struggle, we should not be led into the ‘voluntarist’ error of reducing class determination to class position. From that error flow extremely important political consequences, which will be mentioned in the sections dealing with technicians, engineers and the labour aristocracy. Yet the economic criterion remains determinant. But how are we to understand the terms ‘economic’ and ‘economic criterion’ in the Marxist conception?

The ‘economic’ sphere is determined by the production process and the place of the agents, i.e. by their distribution into social classes by the relations of production: in the unit consisting of production/consumption/ division of the social product, it is production which has the determinant role. The distinction between the classes at this level is not, e.g. a distinction based on relative sizes of income (a distinction between ‘rich’ and ‘poor’), as was believed by a long pre-Marxist tradition and as is still believed today by a whole series of sociologists. The undoubted distinction between relative levels of income is itself only a consequence of the relations of production.

What then are the production process and the relations of production which compose it? In the production process, we find first of all the labour process: it is that which in general designates man’s relation to nature. But the labour process always appears in an historically determined social form. It exists only in its unity with relations of production. In a society divided into classes, the relations of production consist of a double relation which encompasses men’s relations to nature in material production. These two relations are relations first between men and other men—class relations, and secondly between the agents of production and the object and means of labour—the productive forces. These two relations thus concern the relation of the non-worker (the owner) to the object and means of labour and the relation of the immediate producer (direct worker) to the object and means of labour. These relations involve two aspects: (a) Economic ownership: by this is meant the real economic control of the means of production, i.e. the power to assign the means of production to given uses and so to dispose of the products obtained. (b) Possession: by this is meant the capacity to put the means of production into operation.

In every society divided into classes, the first relation (owners/means of production) always coincides with the first aspect: it is the owners who have real control of the means of production and thus exploit the direct workers by extorting surplus value from them in various forms. But this ownership is to be understood as real economic ownership, control of the means of production, to be distinguished from juridical ownership, which is sanctioned by the law and belongs to the superstructure. Certainly, the law generally ratifies economic ownership, but it is possible for the forms of juridical ownership not to coincide with real economic ownership.

We can illustrate this by two examples, beginning with the case of the big farmers in the division of classes in the countryside. According to Lenin, they belong to the rich peasantry, even though they do not have formal, juridical ownership of the land, which belongs to the rentier capitalist. These big farmers belong to the rich peasantry not because of their high incomes, but because they have real control of the land and the means of labour, of which they are thus the effective economic owners. I mention this case merely as an example. Space does not permit a detailed analysis of the class divisions of the peasantry, which does not in itself constitute a single class. It should, however, be pointed out that we can divide the countryside into big landowners, rich, medium and poor peasants, such that each class encompasses groups arising from different forms of ownership and exploitation, only if we make a rigorous distinction between formal, juridical ownership and real, economic ownership.

The case of the ussr and the ‘socialist’ countries is a second example. This is highly controversial, but it should not be omitted here. In these countries, formal, juridical ownership of the means of production belongs to the state, which is held to be ‘the people’s state’; but real control (economic ownership) certainly does not belong to the workers themselves—as we can see from the extinction of the Soviets and the workers’ councils—but to the directors of enterprises and to the members of the party apparatus. It can therefore be argued that the form of collective juridical ownership conceals a new form of economic ‘private’ ownership; and hence that one should speak of a new ‘bourgeoisie’ in the ussr. In reality, the abolition of class exploitation cannot be equated simply with the abolition of juridical private ownership, but with the abolition of real economic ownership—i.e. control of the means of production by the workers themselves.

These considerations have a bearing on the question of the transition to socialism. By keeping in mind the all-important theoretical and real distinction between economic and formal, juridical, ownership, we can see that the mere nationalization of enterprises is not a solution. This is not only because nationalizations are adapted by the bourgeoisie to their own interests. It is because even when they are accompanied by a change in state power, a nationalization or a take-over of the economy by the state only changes the form of juridical ownership. In the last resort, the one thing which can fundamentally modify economic ownership and thus lead to the abolition of classes is the control of production by the workers themselves.