Marx's Critique

Marx rejected both the positive conception of a homogeneous field of given economic phenomena -- and the ideological anthropology of the homo oeconomicus (etc.) which underlies it. Along with this unity he therefore rejected the very structure of the object of Political Economy.

First let us see what was the fate of classical anthropology in Marx's work. For this purpose I shall make a rapid survey of the major regions of the economic 'space': consumption, distribution and production -- in order to see what theoretical place is occupied in it by anthropological concepts.


A. C O N S U M P T I O N

We can begin with consumption, which seems a direct concern of anthropology since it involves the concept of human 'needs '. In the 1857 Introduction, Marx showed that economic needs cannot be defined unambiguously by relating them to the 'human nature' of the economic subjects. In fact, consumption is double. It does include the individual consumption of the men in a given society, but also productive consumption, which would have to be defined as the consumption which satisfies the needs of production to consecrate the universal use of the concept of need. This kind of consumption includes: the 'objects' of production (natural materials or raw materials, the result of labour transforming natural materials) and the instruments of production (tools, machines, etc.) necessary for production. A full part of consumption is therefore directly and exclusively the concern of production itself. A full part of consumption is therefore devoted not to the satisfaction of the needs of individuals, but to allowing either simple or extended reproduction of the conditions of production. From this statement Marx drew two absolutely essential distinctions, both of which are absent from classical Political Economy: the distinction between constant capital and variable capital, and the distinction between two departments of production, Department I, devoted to the reproduction of the conditions of production on a simple or extended basis, and Department II, devoted to the pro- duction of the objects of individual consumption. The proportion between these two departments is governed by the structure of production which intervenes directly to determine the nature and the quantity of a full part


page 166

of the use-values which never enter consumption for need but only production itself. This discovery plays an essential part in the theory of the realization of value, in the process of capitalist accumulation, and in all the laws that flow from it. This point is the object of an interminable polemic of Marx's against Smith, which he returns to several times in Volumes Two and Three and which is echoed in Lenin's critique of the populists and their teacher, the 'romantic' economist Sismondi.[38]

However, this distinction does not settle all the questions. It may be true that the 'needs' of production avoid any anthropological determination, but it remains true also that part of the product is consumed by individuals, who satisfy their 'needs' with it. But here, too, we find that anthropology's theoretical pretensions have been shattered by Marx's analysis. Not only does Marx define these 'needs' as 'historical' and not as absolute givens -- (The Poverty of Philosophy, pp. 41-2; Capital, T.I, pp. 174, 228; Vol. I, pp. 171, 232; Vol. III, p. 837, etc.), but also and above all he recognizes them as 'needs' in their economic function, on condition that they are 'effective' (Vol. III, pp. 178, 189). The only needs that play an economic part are those that can be satisfied economically: those needs are not defined by human nature in general but by their effectivity, i.e., by the level of the income at the disposal of the individuals concerned -- and by the nature of the products available, which are, at a given moment, the result of the technical capacities of production. The determination of the needs of individuals by the forms of production goes even further, since production produces not only definite means of consumption (use-values), but also their mode of consumption, including even the wish for these products (1857 Introduction, op. cit., p. 13). In other words, individual consumption itself, which interconnects use-values and needs in an apparently immediate fashion (and therefore seems to derive directly from an anthropology, but a historicized one), refers us to the technical capacities of production (the level of the forces of production ) on the one hand, and on the other to the social relations of production, which fix the distribution of income (the forms of the division
38 Although there is no time to do it here, I should like to note that it would be of great interest to study these long critiques of Marx's in order to find out on the one hand what distinguishes Marx from Smith in this crucial matter and on the other how and where he locates the essential difference -- in order to find out how he explains Smith's incredible 'oversight ', 'blindness ', 'misconstruction ' and 'forgetfulness ' which are at the root of the 'absurd dogma' that dominates all modern economics, and finally, in order to find out why Marx felt the need to begin this critique four or five times over, as if he had not got to the bottom of it. And we should then discover, among other epistemologically relevant conclusions, that Smith's 'enormous oversight' was directly related to his exclusive consideration of the individual capitalist, i.e., of the economic subjects considered outside the whole as the ultimate subjects of the global process. In other words, we should discover once again the determinant presence of the anthropological ideology in its directly effective form (essential references: Capital, Vol. II, pp. 189-227 and 359-436; Vol. III, pp. 811-30; Theories of Surplus-Value, Vol. I, pp. 90-100.)


page 167

into surplus-value and wages). This last point leads on to the distribution of men into social classes, which then become the 'real' 'subjects' (insofar as that term is applicable) of the production process. The direct relationship between 'needs' thus defined and an anthropological basis becomes therefore purely mythical: or rather, we must invert the order of things and say that the idea of an anthropology, if it is possible at all, must first take into consideration the economic (non-anthropological) definition of those 'needs'. Those needs are subject to a double structural, i.e., no longer anthropological, determination: the determination which divides the products between Departments I and II, and assigns to needs their content and meaning (the structure of the relation between the productive forces and the relations of production). This conception therefore rejects classical anthropology's founding role in economics.


B. D I S T R I B U T I O N

Since distribution has been revealed as an essential factor in the determination of needs -- alongside production -- let us examine this new category. Distribution, too, has two aspects. It is not only the distribution of income (which refers to the relations of production), but also the distribution of the use-values produced by the production process. But we know that these use-values include the products of Department I, or means of production -- and the products of Department II, or means of consumption. The products of Department II are exchanged for individual's incomes, hence as a function of their incomes, hence as a function of the distribution of incomes, hence as a function of the first distribution. As for the products of Department I, the means of production, intended for the reproduction of the conditions of production, they are not exchanged for income, but directly between the owners of the means of production (this results from the realization diagrams in Volume Two): between the members of the capitalist class, who have a monopoly of the means of production. Behind the distribution of use-values, therefore, we can trace the outline of a different distribution: the distribution of men into social classes exercising functions in the production process.

In its most banal conception, distribution appears as the distribution of products, and thus as further away from and quasi-independent of production. But before distribution is distribution of the product, it is: (1) the distribution of the instruments of production, and (2) what is a further definition of the same relationship, the distribution of the members of the society into the different kinds of production (subsumption of the individuals under determinate relations of production). The distribution of the product is obviously only the result of this distribution which is included within the production process itself and determines the articulation of production (Marx: 1857 Introduction, op. cit., p. 17).


page 168

In both cases, whether by the distribution of income or by the distribution of means of consumption and means of production, the index of the distribution of the members of society into distinct classes, we are therefore referred to the relations of production and production itself.

Our examination of categories which at first sight seemed to demand the theoretical intervention of an anthropology of the homo oeconomicus and, for this reason, might have seemed to make it well-founded, has therefore produced two results: (1) the disappearance of anthropology, which has ceased to play its founding role (determination of the economic as such, determination of the 'subjects' of the economy). The 'planar space' of economic phenomena is no longer doubled by the anthropological space of the existence of human subjects; (2) A necessary reference, implied by the analysis of consumption and distribution, to the site of the true determination of the economic: production. Correlatively, we see this theoretical deepening as a transformation of the field of economic phenomena: their former 'planar space' has been replaced by a new pattern in which the economic 'phenomena' are thought within the domination of the 'relations of production ' which define them.

The reader will have recognized one of Marx's basic theses in this second result: it is production that governs consumption and distribution, not the reverse. Marx's whole discovery is often reduced to this basic theory and its consequences.

But this 'reduction' runs into one small difficulty; this discovery is as old as the Physiocrats, and Ricardo, the economist 'of production par excellence ' (Marx), gave it its systematic form. In fact, Ricardo proclaimed the primacy of production over distribution and consumption. We must go even further and admit, as Marx does in the 1857 Introduction, that Ricardo claimed that distribution constituted the peculiar object of Political Economy because he was alluding to the aspect of distribution which concerns the division of the agents of production into social classes (1857 Introduction, op. cit., p. 17). But here too we must apply to Ricardo what Marx said of him with respect to surplus-value. Ricardo gave every outward token of recognizing the reality of surplus-value -- but he always spoke of it in the forms of profit, rent and interest, i.e., within other concepts than its own. Similarly, Ricardo gives every outward token of recognizing the existence of the relations of production -- but he always speaks of them in the form of the distribution of income and products alone -- i.e., without producing their concept. When it is only a question of identifying the existence of a reality behind its disguise, it does not matter if the word or words which designate it are inadequate concepts. This is what enabled Marx to translate the language of his predecessor in an immediate substitutional reading, and to pronounce the words surplus-value where Ricardo had pronounced the word profit -- or the words relations of production where Ricardo had pronounced the words distribution of income. This is all right so long as there is no need to do more


page 169

than designate an existence: it is enough to correct a word in order to call the thing by its name. But when it is a matter of the theoretical consequences arising from this disguise, the affair becomes much more grave: since this word then plays the part of a concept whose inadequacy or absence has serious theoretical effects, whether the author in question recognizes them (as Ricardo did the contradictions he ran into) or not. Then one learns that what one had taken for a reality disguised in an inaccurate word is a disguised second disguise: the theoretical function of a concept disguised in a word. On this condition, variations in terminology may be the real index of a variation in the problematic and the object. However, it is just as if Marx had made his own division of labour. On the one hand, he was content to carry out a substitutional reading of his predecessors: this was a sign of the 'generosity' (Engels) which always made him calculate his debts unselfishly, and in practice treat 'producers' as 'discoverers'. But on the other hand, though in different places, Marx revealed that he was as pitiless towards the theoretical consequences drawn by his predecessors in this blindness as he was to the conceptual meaning of the facts which they had produced. When Marx criticized Smith or Ricardo with the utmost severity because they were unable to distinguish between surplus-value and its forms of existence, he was in fact attacking them because they did not give a concept to the fact that they had managed to 'produce'. We can clearly see that the mere 'omission' of a word is really the absence of a concept, since the presence or absence of a concept is decisive for a whole chain of theoretical consequences. And in return, this illuminates the effects of the absence of a word on the theory which 'contains' this absence: the absence of a 'word' from it is the presence in it of a different concept. In other words, anyone who thinks he only has to re-establish a 'word' which is absent from Ricardo's discourse is in danger of deceiving himself as to the conceptual effect of that absence, he is reducing Ricardo's very concepts to mere 'words'. In this cross-over of false identifications (the belief that the construction of a concept is no more than the re-establishment of a word; the belief that Ricardo's concepts are mere words) we must look for the reason why Marx could both exalt his predecessors' discoveries when they had often only 'produced' them without 'discovering' them, and criticize them just as sharply for the theoretical consequences, although these consequences have merely been drawn from the 'discoveries'. I had to go into this amount of detail in order to situate the meaning of the following judgement of Marx's:

Ricardo, who was concerned to conceive modern production in its determinate social articulation, and who is the economist of production par excellence, precisely for this reason explains not production but distribution as the basic theme of modern economics (1857 Introduction, op. cit., p. 18).


page 170

'Precisely for this reason' means:

. . . [he] instinctively conceived the forms of distribution as the most definite expression of the fixed relations between the agents of production in a given society (ibid., p. 17).


The 'fixed relation between the agents of production in a given society' are precisely the relations of production, and when Marx took them into consideration, not in the form of an 'instinctive ' feeling, i.e., in the form of the 'unknown ' -- but in the form of a concept and its consequences, it revolutionized the object of classical economics, and with the object, the science of Political Economy as such.

Marx's peculiarity, indeed, does not lie in his having claimed or even demonstrated the primacy of production (Ricardo had already done this in his own way), but in his having transformed the concept of production by assigning to it an object radically different from the object designated by the old concept.


C. P R O D U C T I O N

According to Marx, all production is characterized by two indissociable elements: the labour process, which deals with the transformations man inflicts on natural materials in order to make use-values out of them, and the social relations of production beneath whose determination this labour process is executed. We shall examine these two points in succession: the labour process (a) and the relations of production (b).

(a) The labour process

The analysis of the labour process involves the material and technical conditions of production.

The labour process, . . . the activity whose aim is the production of use-values, the appropriation of external substances for needs, is the general condition for exchanges of matter between man and nature, a physical necessity of human life, and is therefore independent of all its social forms, or rather common to all of them (Capital, T.I, p. 186; Vol. I, pp. 183-4).

This process can be reduced to the combination of simple elements, of which there are three: ' . . . (1) the personal activity of man, or labour strictly speaking; (2) the object on which that labour acts; (3) the means with which it acts' (T.I, p. 181; Vol. I, p. 178). The labour process therefore implies an expenditure of the labour-power of men who, using defined instruments of labour according to adequate (technical) rules, transform the object of labour (either a natural material or an already worked material or raw material) into a useful product.

This analysis brings out two essential features which we shall examine in succession: the material nature of the conditions of the labour process, and the dominant role of the means of production in the labour process.


page 171

First feature : Every productive expenditure of labour power presupposes material conditions for its performance, which can all be reduced to the existence of nature, either directly, or as modified by human activity. When Marx writes that 'labour is, in the first place, a process which takes place between man and nature, and in which man starts, regulates, and controls by his own activity the material exchanges between himself and nature. He opposes himself to nature as a natural force', he is stating that the transformation of material nature into products, and therefore the labour process as a material mechanism, is dominated by the physical laws of nature and technology. Labour-power, too, is included in this mechanism. This determination of the labour-process by these material conditions is at its own level a denial of every 'humanist' conception of human labour as pure creativity. As we know, this idealism has not remained in the state of a myth, but has reigned in political economy itself, and from there, in the economic utopias of vulgar socialism: e.g. in Proudhon (the people's bank project), Gray ('labour bonds'), and finally in the Gotha Programme, whose opening line proclaimed:

Labour is the source of all wealth and culture,

to which Marx replied:

Labour is not the source of all wealth. Nature is just as much the source of use-values (and it is surely of such that material wealth consists !) as labour, which itself is only the manifestation of a force of nature, human labour-power. The above phrase is to be found in all children's primers and is correct insofar as it is implied that labour is performed with the appurtenant objects and instruments. But a socialist programme cannot allow such bourgeois phrases to pass over in silence the conditions that alone give them meaning . . . . The bourgeois have very good grounds for falsely ascribing supernatural creative power to labour. ('Critique of the Gotha Programme,' Selected Works in One Volume, London 1968, p. 319).


It was this same utopianism that led Smith and all the utopians who have followed him on this point, to leave out of their economic concepts any formal representation of the necessity for the reproduction of the material conditions of the labour process, as essential to the existence of that process -- and therefore to abstract from the current materiality of the productive forces (the object and the material instruments of labour) implied in every production process (in this respect, Smith's Political Economy lacks a theory of reproduction, an indispensable element of any theory of production). The same idealism of labour made it possible for Marx, in the 1844 Manuscripts, to call Smith the 'Luther of Political Economy' because he reduced all wealth (all use-value) to human labour alone; and to seal the theoretical union of Smith and Hegel: the first because he reduced the whole of political


page 172

economy to the subjectivity of labour, the second because he conceived 'labour as the essence of man'. In Capital, Marx breaks with this idealism of labour by thinking the concept of the material conditions of every labour process and by providing the concept of the economic forms of existence of these material conditions: in the capitalist mode of production, the decisive distinctions between constant capital and variable capital on the one hand, and between Department I and Department II on the other.

This simple example enables us to assess the theoretical and practical effects induced in the field of economic analysis itself by merely thinking the concept of its object. Once Marx thought the reality of the material conditions of production as belonging to the concept of production, economically 'operational' concepts emerged in the field of economic analysis (constant capital, variable capital, Department I, Department II) which revolutionized its arrangement and nature. The concept of its object is not a paraeconomic concept, it is the concept of the construction of the economic concepts necessary for an understanding of the nature of the economic object itself: the economic concepts of constant capital and variable capital, of Department I and Department II, are merely the economic determinations, in the field of economic analysis itself, of the concept of the material conditions of the labour process. The concept of the object exists immediately then in the form of directly 'operational' economic concepts. But without the concept of the object, these concepts would not have been produced, and we should have remained in Smith's economic idealism, exposed to all the temptations of ideology.

This is a crucial point, for it shows us that to call ourselves Marxists it is not enough for us to believe that the economy, and in the economy, production, govern all the other spheres of social existence. It is possible to proclaim these positions and yet, at the same time, develop an idealist conception of the economy and of production, by declaring that labour constitutes both the 'essence of man' and the essence of political economy, in short by developing an anthropological ideology of labour, of the 'civilization of labour', etc. Marx's materialism, on the contrary, presupposes a materialist conception of economic production, i.e., among other conditions, a demonstration of the irreducible material conditions of the labour process. This is one of the points where a sentence from one of Marx's letters to Engels which I have referred to above is directly applicable: the sentence in which Marx points out that he 'attributed much more importance to the category use-value ' than did any of his predecessors. This is a stumbling-block for all the interpretations of Marxism as a 'philosophy of labour', whether ethical, personalist or existentialist: especially Sartre's theory of the practico-inert, since it lacks any concept of the modality of the material conditions of the labour process. Smith had already related the current material conditions of the labour process to past labour: he thus dissolved the currency of the material conditions required at a given moment for the existence of


page 173

the labour process in an infinite regression, in the non-currency of earlier labours, in their memory, Hegel was to resurrect this idea in his theory of 'Erinnerung '). Similarly, Sartre dissolves the current material conditions whose structural combination governs all effective labour and every current transformation of a raw material into a useful product in the philosophical memory of an earlier praxis, itself second to another or several other earlier praxes, and so on down to the praxis of the original subject. In Smith, who was writing an economic work, this ideal dissipation had important theoretical consequences in the realm of the economy itself. In Sartre, it is immediately elevated into its explicit philosophical 'truth': the anthropology of the subject, latent in Smith, takes the open form of a philosophy of freedom in Sartre.

Second feature. The same analysis of the labour process reveals the dominant role of the 'means of labour '.

The use and fabrication of means of labour . . . is characteristic of the specifically human labour-process, and Franklin therefore defines man as a tool-making animal. Relics of by-gone means of labour possess the same importance for the investigation of extinct economic forms of society, as does the structure of fossil bones for a knowledge of the organization of extinct species of animals. It is less what is produced (macht ) than how (wie ) it is produced, and by what means of labour, that enables us to distinguish different economic epochs. Means of labour supply a standard of the degree of development of the labourer and they are also indicators (Anzeiger ) of the social relations in which he labours (Capital, T.I, pp. 182-3; Vol. I, pp. 179-80).


One of the three constitutive elements of the labour process (object of labour, means of labour, labour-power) is therefore dominant: the means of labour. It is this last element which enables us to identify within the labour process common to every economic epoch the specific difference which will distinguish between its essential forms. The 'means of labour' determine the typical form of the labour process considered: by establishing the 'mode of attack' on the external nature subject to transformation in economic production, they determine the mode of production, the basic category of Marxist analysis (in economics and history); at the same time, they establish the level of productivity of productive labour. The concept of the pertinent differences observable in a variety of labour processes, the concept which makes possible not only the 'periodization' of history, but above all the construction of the concept of history: the concept of the mode of production is thus established, with respect to our present considerations, in the qualitative differences between different means of labour, i.e., in their productivities. Need I point out that there is a direct relationship between the concept of the dominant role of the means of labour and the economically 'operational' concept of productivity? Need I note the fact that classical


page 174

economics was never able to isolate and identify this concept of productivity -- a fact Marx attacked it for -- and that its misconception of history was linked to the absence in it of the concept of mode of production ?[39]

By producing his key concept of the mode of production, Marx was indeed able to express the differential degree of material attack on nature by pro- duction, the differential mode of unity existing between 'man and nature', and the degree of variation in that unity. But as well as revealing to us the theoretical significance of taking into consideration the material conditions of production, the concept of the mode of production simultaneously reveals to us another determinant reality, corresponding to the degree of variation in the 'man-nature' unity: the relations of production :

Means of labour not only supply a standard of the degree of develop- ment to which human labour power has attained, but they are also indices (Anzeiger ) of the social relations under which production is carried on . . .


Here we discover that the man-nature unity expressed in the degree of variation in that unity is at the same time both the unity of the man-nature relationship and the unity of the social relations in which production takes place. The concept of the mode of production therefore contains the concept of the unity of this double unity.

(b) The relations of production

We have thus arrived at a new condition of the production process. After studying the material conditions of the production process, which express the specific nature of the relations between men and nature, we must now turn to a study of the social conditions of the production process: the social relations of production. These new conditions involve the specific type of relations between the agents of production which exist as a function of the relations between these agents on the one hand and the material means of production on the other. This adjustment is crucial: the social relations of production are on no account reducible to mere relations between men, to relations which only involve men, and therefore to variations in a universal matrix, to inter-subjectivity (recognition, prestige, struggle, master-slave relationship, etc.). For Marx, the social relations of production do not bring men alone onto the stage, but the agents of the production process and the material conditions of the production process, in specific 'combinations'. I insist on this point, for reasons which are related to Rancière's analysis of certain of Marx's expressions,[40] where, in a terminology still inspired by his early anthropological philosophy, it is tempting to oppose, literally, relations between men and relations between things. But the relations of production necessarily imply relations between men and things, such that the
39 For all these questions, barely outlined in this chapter, see Étienne Balibar's essay -- especially his important analysis of the concept of productive forces.
40 See Lire le Capital, first edition, 1965, Vol. I, pp. 93ff.


page 175

relations between men and men are defined by the precise relations existing between men and the material elements of the production process.

How did Marx think these relations? He thought them as a 'distribution' or 'combination' (Verbindung ). Discussing distribution in the 1857 Introduction, Marx wrote (op. cit., pp. 17-18):

In its most banal conception, distribution appears as the distribution of products, and thus as further away from and quasi-independent of production. But before distribution is distribution of the product, it is: (1) the distribution of the instruments of production, and (2) what is a further definition of the same relationship, the distribution of the members of the society into the different kinds of production (subsumption of the individuals under determinate relations of production). The distribution of the product is obviously only the result of this distribution which is included within the production process itself and determines the articulation of production (Gliederung ). It is obviously an empty abstraction to consider production while ignoring this distribution which is included in it, while, on the contrary, the distribution of products is implied by this distribution, which originally forms a moment (Moment ) of production . . . Production must start from a certain distribution of the instruments of production . . .


This distribution thus consists of a certain attribution of the means of production to the agents of production, in a certain regular proportion fixed between, on the one hand, the means of production, and on the other, the agents of production. This distribution-attribution can be formally conceived as the combination (Verbindung ) of a certain number of elements which belong either to the means of production or to the agents of production, a combination which occurs according to definite modalities.

This is Marx's own expression:

Whatever the social form of production, labourers and means of production always remain factors of it. But in a state of separation from each other either of these factors can be such only potentially. For production to go on at all they must combine. The specific manner (die besondere Art und Weise ) in which thus combination is accomplished distinguishes the different economic epochs of the structure of society (Gesellschaftsstruktur ) from one another (Capital, Vol II, p. 34 -- modified).


In another and probably more important text (Capital, Vol. III, pp. 770-04), on the feudal mode of production, Marx writes:

The specific economic form, in which unpaid surplus-labour is pumped out of direct producers, determines the relationship of rulers and ruled, as it grows directly out of production itself and, in turn reacts upon it as a determining element. Upon this, however, is founded the entire formation


page 176

(Gestaltung ) of the economic community which grows out of the production relations themselves, thereby simultaneously its specific political form (Gestalt). It is always the direct relationship of the owners of the conditions of production to the direct producers -- a relation always naturally corresponding to a definite stage in the development of the methods (Art und Weise ) of labour and thereby its social productivity -- which reveals the innermost secret (innerste Geheimnis ), the hidden basis (Grundlage ) of the entire social structure (Konstruktion ), and with it the political form of the relation of sovereignty and dependence, in short, the corresponding specific form of the State.


This text's developments reveal behind the two elements hitherto considered (agents of production, means of production) distinctions of quite crucial importance. On the side of the means of production we find the already familiar distinction between the object of production, e.g., the land (which played a determinant part directly in all the modes of production before capitalism), and the instruments of production. On the side of the agents of production we find, besides the distinction between labourer and labour power, an essential distinction between the direct agents (Marx's own expression) whose labour power is set to work in production, and other men whose role in the general process of production is that of owners of the means of production, but who do not feature in it as labourers or direct agents, since their labour power is not used in the production process. By combining or inter-relating these different elements -- labour power, direct labourers, masters who are not direct labourers, object of production, instruments of production, etc. -- we shall reach a definition of the different modes of production which have existed and can exist in human history. This operation inter-relating determinate pre-existing elements might make us think of a combinatory, if the very special specific nature of the relations brought into play in these different combinations did not strictly define and limit its field. To obtain the different modes of production these different elements do have to be combined, but by using specific modes of combination or 'Verbindungen ' which are only meaningful in the peculiar nature of the result of the combinatory (this result being real production) -- and which are: property, possession, disposition, enjoyment, community, etc. The application of specific relations to the different distributions of the elements present produces a limited number of formations which constitute the relations of production of the defined modes of production. These relations of production determine the connexions between the different groups of agents of production and the objects and instruments of production, and thereby they simultaneously divide the agents of production into functional groups, each occupying a definite place in the production process. The relations between the agents of production are then the result of the typical relations they maintain with the means of production (object, instruments) and of their distribution into


page 177

groups defined and localized functionally in their relations with the means of production by the structure of production.

I cannot give a theoretical analysis of this concept of 'combination' and of its different forms here: on this point I refer the reader to Balibar's paper. But it is clear that the theoretical nature of this concept of 'combination' may provide a foundation for the thesis I have already suggested in a critical form, the thesis that Marxism is not a historicism : since the Marxist concept of history depends on the principle of the variation of the forms of this 'combination'. I should just like to insist on the special nature of these relations of production, which are remarkable in two respects.

In the text I have just quoted we have seen Marx prove that a certain form of combination of the elements present necessarily implied a certain form of domination and servitude indispensable to the survival of this combination, i.e., a certain political configuration (Gestaltung ) of society. We can see precisely where the necessity and form of the political 'formation' is founded: at the level of the Verbindungen which constitute the modes of liaison between the agents of production and the means of production, at the level of the relations of property, possession, disposition, etc.[41] These types of connexion, according to the diversification or non-diversification of the agents of production into direct labourers and masters, make the existence of a political organization intended to impose and maintain the defined types of connexions by means of material force (that of the State) and of moral power (that of ideologies) either necessary (class societies) or superfluous (classless societies). This shows that certain relations of production presuppose the existence of a legal-political and ideological superstructure as a condition of their peculiar existence, and why this superstructure is necessarily specific (since it is a function of the specific relations of production that call for it). It also shows that certain other relations of production do not call for a political superstructure, but only for an ideological superstructure (classless societies). Finally, it shows that the nature of the relations of production considered not only calls or does not call for a certain form of superstructure, but also establishes the degree of effectivity delegated to a certain level of the social totality. Irrespective of all these consequences, we can draw one conclusion at any rate where the relations of production are concerned: they relate to the superstructural forms they call for as so many conditions of their own existence. The relations of production cannot therefore be thought in their concept while abstracting from their; specific superstructural conditions of existence. To take only one example, it is quite clear that the analysis of the buying and selling of labour power in which capitalist relations of production exist (the separation between the
41 One important specification. The term 'property' used by Marx can lead to the belief that the relations of production are identical with legal relations. But law is not the relations of production The latter belong to the infrastructure, the former to the superstructure.


page 178

owners of the means of production on the one hand and the wage-workers on the other), directly presupposes, for an understanding of its object, a, consideration of the formal legal relations which establish the buyer (the capitalist) as much as the seller (the wage-labourer) as legal subjects -- as well as a whole political and ideological superstructure which maintains and contains the economic agents in the distribution of roles, which makes a minority of exploiters the owners of the means of production, and the majority of the population producers of surplus-value. The whole superstructure of the society considered is thus implicit and present in a specific way in the relations of production, i.e., in the fixed structure of the distribution of means of production and economic functions between determinate categories of production agents. Or in other words, if the structure of the relations of production defines the economic as such, a definition of the concept of the relations of production in a determinate mode of production is necessarily reached via the definition of the concept of the totality of the distinct levels of society and their peculiar type of articulation (i.e. effectivity).

In no sense is this a formal demand; it is the absolute theoretical condition governing the definition of the economic itself. It is enough to refer to the innumerable problems raised by this definition where modes of production other than the capitalist one are concerned to realize the decisive importance of this recourse: Marx often says that what is hidden in capitalist society is clearly visible in feudal society or in the primitive community, but precisely in the latter societies we can clearly see that the economic is not directly and clearly visible ! -- just as in these same societies we can also clearly see that the degree of effectivity of the different levels of the social structure is not clearly visible ! Anthropologists and ethnologists 'know' what to confine themselves to when, seeking the economic, they come upon kinship relations, religious institutions, etc.; specialists in mediaeval history 'know' what to confine themselves to when, seeking for the dominant determination of history in the 'economy', they find it in politics or religion.[42] In all these cases, there is no immediate grasp of the economic, there is no raw economic 'given', any more than there is any immediately 'given' effectivity in any of the levels. In all these cases, the identification of the economic is achieved by the construction of its concept, which presupposes a definition of the specific existence and articulation of the different levels of the structure of the whole, as they are necessarily implied by the structure of the mode of production considered. To construct the concept of the economic is to define it rigorously as a level, instance or region of the structure of a mode of production: it is therefore to define its peculiar site, its extension, and its limits within that structure; if we like to return to the old Platonic image, it is to 'divide up' the region of the economic correctly in the whole, according
42 Cf. Godelier's article 'Objet et méthode de l'anthropologie économique' (L'Homme, October 1965 and in Rationalité et irrationalité en économie, Paris 1966).


page 179

to its peculiar 'articulation', without mistaking this articulation. The 'division' of the 'given', or empiricist division, always mistakes the articulation, precisely because it projects on to the 'real' the arbitrary articulations and divisions of its underlying ideology. There is no correct division and therefore no correct articulation, except on condition of possessing and therefore constructing its concept. In other words, in primitive societies it is not possible to regard any fact, any practice apparently unrelated to the 'economy') (such as the practices which are produced by kinship rites or religious rites, or by the relations between groups in 'potlatch' competition), as rigorously economic, without first having constructed the concept of the differentiation of the structure of the social whole into these different practices or levels, without having discovered their peculiar meaning in the structure of the whole, without having identified in the disconcerting diversity of these practices the region of economic practice, its configuration and its modalities. It is probable that the majority of the difficulties of contemporary ethnology and anthropology arise from their approaching the 'facts', the 'givens' of (descriptive) ethnography, without taking the theoretical precaution of constructing the concept of their object: this omission commits them to projecting on to reality the categories which define the economic for them in practice, i.e., the categories of the economics of contemporary society, which to make matters worse, are often themselves empiricist. This is enough to multiply aporia. If we follow Marx here, too, this detour via primitive societies, etc., will only have been necessary in order to see clearly in them what our own society hides from us: i.e., in order to see clearly in them that the economic is never clearly visible, does not coincide with the 'given' in them any more than in any other reality (political, ideological, etc.). This is all the more 'obvious' for the capitalist mode of production in that we know that the latter is the mode of production in which fetishism affects the economic region par excellence. Despite the massive 'obviousness' of the economic 'given' in the capitalist mode of production, and precisely because of the 'massive' character of this fetishised 'obviousness', the only way to the essence of the economic is to construct its concept, i.e., to reveal the site occupied in the structure of the whole by the region of the economic, therefore to reveal the articulation of this region with other regions (legal-political and ideological superstructure), and the degree of presence (or effectivity) of the other regions in the economic region itself. Here, too, this requirement can be faced directly as a positive theoretical requirement: it can also be omitted, and it then reveals itself in peculiar effects, either theoretical (contradictions and thresholds in the explanation) or practical (e.g., difficulties in planning techniques, whether socialist or capitalist). That, very schematically, is the first conclusion we can draw from Marx's determination of the economic by the relations of production.

The second conclusion is not less important. If the relations of production now appear to us as a regional structure, itself inscribed in the structure of the


page 180

social totality, we are interested in this because of its structural nature. Here both the mirage of a theoretical anthropology and the mirage of a homogeneous space of given economic phenomena dissolve simultaneously. Not only is the economic a structured region occupying its peculiar place in the global structure of the social whole, but even in its own site, in its (relative) regional autonomy, it functions as a regional structure and as such determines its elements. Here once again we find the results of the other papers in this book: i.e., the fact that the structure of the relations of production determines the places and functions occupied and adopted by the agents of production, who are never anything more than the occupants of these places, insofar as they are the 'supports' (Träger ) of these functions. The true 'subjects' (in the sense of constitutive subjects of the process) are therefore not these occupants or functionaries, are not, despite all appearances, the 'obviousnesses' of the 'given' of naïve anthropology, 'concrete individuals', 'real men' -- but the definition and distribution of these places and functions. The true 'subjects ' are these definers and distributors : the relations of production (and political and ideological social relations). But since these are 'relations', they cannot be thought within the category subject. And if by chance anyone proposes to reduce these relations of production to relations between men, i.e., 'human relations ', he is violating Marx's thought, for so long as we apply a truly critical reading to some of his rare ambiguous formulations, Marx shows in the greatest depth that the relations of production (and political and ideological social relations) are irreducible to any anthropological inter-subjectivity -- since they only combine agents and objects in a specific structure of the distribution of relations, places and functions, occupied and 'supported' by objects and agents of production.

It is clear once again, then, how the concept of his object distinguishes Marx radically from his predecessors and why criticisms of him have run wide of the mark. To think the concept of production is to think the concept of the unity of its conditions: the mode of production. To think the mode of production is to think not only the material conditions but also the social conditions of production. In each case, it is to produce the concept which governs the definition of the economically 'operational' concepts (I use the word 'operational' deliberately, since it is often used by economists) out of the concept of their object. We know which concept in the capitalist mode of production expressed the fact of capitalist relations of production in economic reality itself: the concept of surplus-value. The unity of the material and social conditions of capitalist production is expressed by the direct relationship between variable capital and the production of surplus-value. The fact that surplus-value is not a measurable reality arises from the fact that it is not a thing, but the concept of a relationship, the concept of an existing social structure of production, of an existence visible and measurable only in its 'effects ', in the sense we shall soon define. The fact that it only exists in its effects does not mean that it can be grasped completely in any


page 181

one of its determinate effects: for that it would have to be completely present in that effect, whereas it is only present there, as a structure, in its determinate absence. It is only present in the totality, in the total movement of its effects, in what Marx calls the 'developed totality of its form of existence', for reasons bound up with its very nature. It is a relation of production between the agents of the production process and the means of production, i.e., the very structure that dominates the process in the totality of its development and of its existence. The object of production, the land, min- erals, coal, cotton, the instruments of production, tools, machines, etc., are 'things ' or visible, assignable, measurable realities: they are not structures. The relations of production are structures -- and the ordinary economist may scrutinize economic 'facts': prices, exchanges, wages, profits, rents, etc., all those 'measurable' facts, as much as he likes; he will no more 'see' any structure at that level than the pre-Newtonian 'physicist' could 'see' the law of attraction in falling bodies, or the pre-Lavoisierian chemist could 'see' oxygen in 'dephlogisticated' air. Naturally, just as bodies were 'seen' to fall before Newton, the 'exploitation' of the majority of men by a minority was 'seen' before Marx. But the concept of the economic 'forms' of that exploitation, the concept of the economic existence of the relations of production, of the domination and determination of the whole sphere of political economy by that structure did not then have any theoretical existence. Even if Smith and Ricardo did 'produce', in the 'fact' of rent and profit, the 'fact' of surplus-value, they remained in the dark, not realizing what they had 'produced', since they could not think it in its concept, nor draw from it its theoretical consequences. They were a hundred miles away from being able to think it, since neither they nor the culture of their time had ever imagined that a 'fact' might be the existence of a relation of 'combination', a relation of complexity, consubstantial with the entire mode of production, dominating its present, its crisis, its future, determining as the law of its structure the entire economic reality, down to the visible detail of the empirical phenoena -- while remaining invisible even in their blinding obviousness.