Contradiction and Overdetermination
Louis Althusser (1962)


I should like to stop here for a moment to examine a passage from Engels’s letter to Bloch that I deliberately ignored in the preceding article. For this passage, containing Engels’s theoretical solution to the problem of the basis for the determination ‘in the last. instance’ by the economy, is, in fact, independent of the Marxist theses that Engels was counter-posing to ‘economist’ dogmatism.

No doubt this is only a letter. But as it constitutes a decisive theoretical document for the refutation of schematism and economism, and as it has already played a historical role as such and may well do so again, we should not conceal the fact that his argument for this basis will no longer answer to our critical needs.

In his solution, Engels resorts to a single model at two different levels of analysis:

First Level

Engels has just shown that the superstructures, far from being pure phenomena of the economy, have their own effectivity: ‘The various elements of the superstructure . . . in many cases preponderate in determining their (the historical struggles) form.’. But this poses the question as to how, under these conditions, we should think the unity of this real, but relative effectivity of the superstructures – and of the determinant principle ‘in the last instance’ of the economy ? How should we think the relation between these distinct effectivities? What basis is there within this unity for the role of the economic as a ‘last instance’? Engels’s reply is that ‘There is an interaction of all these elements (the superstructures) in which, amid all the endless host of accidents (that is, of things and events, whose inner connection is so remote or so impossible of proof that we can regard it as non-existent, as negligible) the economic movement finally asserts itself as necessary’. So the explanatory model goes like this: ‘the various elements of the superstructure’ act and react on one another to produce an infinity of effects. These effects can be assimilated to an infinity of accidents (infinite in number and with an inner connection so remote and therefore so difficult to discover that it is negligible), amid which ‘the economic movement’ asserts itself. These effects are accidents, the economic movement is necessity, their necessity. For the moment I shall ignore the accidents-necessity model and its presuppositions. What is remarkable in this text is the role it attributes to the different elements of the superstructure. It is just as if, once the action-reaction system was set in motion between them, they were charged with finding a basis for the infinite diversity of effects (things and events as Engels puts it) between which, as if between so many accidents, the economy picks its sovereign way. In other words, the elements of the superstructure do have an effectivity, but this effectivity is in some way dispersed into an infinity, into the infinity of effects, of accidents, whose inner connection may, once this extremity in the infinitesimal has been reached, be regarded as non-existent. So the effect of this infinitesimal dispersion is to dissipate the effectivity granted the superstructures in their macroscopic existence into a microscopic non-existence. Of course, this non-existence is epistemological (‘we can regard’ the microscopic connection ‘as non-existent’ – it is not said to be non-existent, for knowledge). But whatever the case, within this infinitesimal microscopic diversity the macroscopic necessity ‘finally asserts itself’, that is, finally prevails.

Two comments should be made here.

First comment

This schema does not give us a true solution, but an elaboration of one part of the solution. We learn that in their mutual action-reaction the superstructures cash their effectivity as infinitesimal ‘things and events’, that is, as so many ‘accidents’. We see that the solution must be based at the level of these accidents, since their object is to introduce the counter-concept of the (economic) necessity which is determinant in the last instance. But this is only a half-solution since the relation between these accidents and this necessity is neither established nor made explicit; since (in what is really a denial of the relation and the problem posed by it) Engels presents even this necessity as completely external to these accidents (as a movement which finally asserts itself amid an infinity of accidents). But if this is so then we do not know whether this necessity is really the necessity of these accidents, and, if it is, why it is. This question is left unanswered.

Second comment

It is astonishing to find Engels in this text presenting the forms of the superstructure as the source of a microscopic infinity of events whose inner connection is unintelligible (and therefore negligible). For, on the one hand, we could say exactly the same of the forms of the infrastructure (and it is quite true that the detail of microscopic economic events might be said to be unintelligible and negligible!). But, more important, these forms as such are certainly forms as principles of reality, but they are also forms as principles of the intelligibility of their effects. For their part they are perfectly knowable, and in this respect they are the transparent reason of the events that derive from them. How could Engels pass so rapidly over these forms, their essence and their role, and only consider the negligible and unintelligible microscopic dust of their effects? More precisely, is this reduction to a dust of accidents not absolutely opposed to the real and epistemological function of these forms? And since Engels invokes it, what else did Marx do in the Eighteenth Brumaire than give an analysis of the action and reaction of these ‘different factors’? a perfectly intelligible analysis of their effects? But Marx was only able to perform this ‘proof’ because he did not confuse the historical effects of these factors with their microscopic effects. The forms of the superstructures are indeed the cause of an infinity of events, but not all these events are historical (cf. Voltaire’s remark that all children have fathers, but not all ‘fathers’ have children), only those of them that the said ‘factors’ retain, select from among the others, in short, produce as such (to take just one case: every politician in government makes a choice among events according to his policies and also his means, and promotes the chosen ones to the rank of historical events, even if it is only, for example, by suppressing a demonstration!). So, to sum up, on this first level: (1) we have not yet been given a true solution; (2) ‘cashing’ the effectivity of the forms of the superstructure (which is all that is in question here) as an infinity of microscopic effects (unintelligible accidents) does not correspond to the Marxist conception of the nature of the superstructures

Second Level

And, in fact, at the second level of his analysis we find Engels abandoning the case of the superstructures and applying his model to another object which does this time correspond to it: the combination of individual wills. We also find that he answers the question by giving us the relation between the accidents and the necessity, that is by finding a basis for it. ‘History is made in such a way that the final result always arises from conflicts between many individual wills, of which each again has been made what it is by a host of particular conditions of life. Thus there are innumerable intersecting forces, an infinite series of parallelograms of forces which give rise to one resultant – the historical event This may again itself be viewed as the product of a power which works as a whole, unconsciously and without volition. For what each individual wills is obstructed by everyone else, and what emerges is something that no-one willed. Thus past history proceeds in the manner of a natural process and is essentially subject to the same laws of motion. But from the fact that individual wills – of which each desires what he is impelled to by his physical constitution and external, in the last resort economic, circumstances (either his own personal circumstances or those of society in general) – do not attain what they want, but are merged into a collective mean, a common resultant, it must not be concluded that their value is equal to zero. On the contrary, each. contributes to the resultant and is to this degree involved in it.’

I apologise for this long quotation, but I had to give it in full as it provides the answer to our question. Here indeed, the necessity is established at the level of the accidents themselves, on the accidents themselves, as their global resultant: so it really is their necessity. The answer missing from the first analysis we really are given here. But on what condition do we get it?. On condition that we change objects, that our starting-point is no longer the superstructures, their interaction and ultimately their microscopic effects – but individual wills, confronted and combined in relations of forces. So it is as if the model applied to the effectivity of the superstructures had really been borrowed from its true object, the object we are now dealing with: the play of individual wills. It is now clear why it failed with its first object, for that was not its real object, and why it should go on to a second, which is its real object.

How, then, does this proof work? It relies on the model of a parallelogram of forces derived from physics: the wills are so many forces; if they confront one another by twos in a simple situation their resultant is a third force, different from either but none the less common to both, and such that though neither can recognise itself in the third, each is none the less a party to it, that is, its co-author. So one basic phenomenon appears straightaway, the transcendence of the resultant with respect to the component forces: a double transcendence, in relation to the respective size of the component forces – and in relation to the reflection of these forces on themselves (that is, to their consciousness, since we are dealing with wills). Which implies: (1) that the resultant is quite different in size from each force by itself (higher if they add together, lower if they oppose each other); (2) that the resultant is, in essence, unconscious (it does not correspond to the consciousness of each will – and at the same time, it is a force without a subject, an objective force, but, from the outset nobody’s force. That is why it immediately becomes the global resultant that may be ‘viewed as the product of a power which works as a whole, unconsciously and without volition’. It is clear that we have now found a basis and an origin for this force that triumphs in the last instance: determination by the economy is no longer external to the accidents amid which it asserts itself, it is the internal essence of these accidents.

I hope to be able to show: (1) that we are now dealing with the true object of Engels’s model; (2) that thanks to this readjustment, Engels has really provided the solution to the problem he raised; (3) that problem and solution only exist as a function of the adequacy of the model to its object; (4) that, as this object does not exist, neither the problem nor the solution exist; (5) that we must find some reason for this whole futile construction.

I am quite prepared to ignore Engels’s reference to nature. As the model he has used is itself physical (the earliest example of the type can be found in Hobbes and then in innumerable later versions, of which Holbach’s is noteworthy as a particularly pure case), there is nothing surprising in the fact that it refers us from history to nature. This is not a proof, it is a tautology. (Note that this is merely a matter of the model used and that the dialectic of nature is obviously not at stake in this exposition, for the very good reason that it arises in a quite different context.) Epistemologically, a tautology is null and void; but it may nevertheless have a heuristic role. It is reassuring to be able to refer directly to nature, to be sure. (Hobbes said it long ago: men tear out their hair or their lives over politics, but they are as thick as thieves over the hypotenuse or falling bodies.)

It is Engels’s argument itself that I should like to examine very closely, the argument which seems at first sight to achieve such a perfect harmony between the model and its object. But what do we find? A harmony between model and object at the immediate level. But beneath (en deca) this level and beyond (au dela), this harmony is postulated, not proved, and in its place we find an indeterminancy, that is, from the point of view of knowledge, a void.

Beneath the level of individual wills. The transparency of content which strikes us when we imagine the parallelogram of forces (of individual wills) disappears once we ask (as Engels does himself!) about the origin (and therefore about the cause) of the determinations of these individual wills. For we are referred to infinity. ‘Each ... has been made what it is by a host of particular conditions of life.’ Each individual will may be simple when it is considered as an absolute beginning, but it becomes the product of an infinity of microscopic circumstances arising from its ‘physical constitution’ and ‘external circumstances’, its ‘own personal circumstances’ ‘or’ ‘those of society in general ‘, the external circumstances which are ‘in the last resort economic’, and all these thrown together so that side by side with purely contingent and peculiar determinations we also find general determinations fin particular, the one we are discussing: the economic circumstances which are determinant in the last resort). It is clear that here Engels is mixing up two types of explanation.

The First Type

a non-Marxist type, but one adapted to its present object and to its hypotheses, viz., explanation by the infinity of circumstances or accidents (this form can be found in Helvetius and Holbach); such explanation may have a critical value (to the extent that it is destined, as was already the case in the eighteenth century, to refute divine intervention, among other things), but from the point of view of knowledge it is empty. It puts forward an infinity without content, an abstract and hardly even programmatic generalisation.

The Second Type

However, at the same time, Engels introduces a Marxist type of explanation, when he ranks among the infinite circumstances (which are in essence microscopic) those determinations which are at once both general and concrete, viz., social circumstances and economic circumstances (which are determinant in the last resort). But this type of explanation does not answer to its object, since it represents at the beginning the very solution which it is supposed to be producing and establishing (the generation of this determination in the last resort). To sum up: either we stay with the object and the problem which Engels has posed, in which case we come face to face with the infinite, the indeterminate (and therefore with an epistemological void); or from this moment we take as the beginning itself the (content-ful) solution which is precisely what is in question. But then we are no longer either in the object or in the problem.

Beyond the level of individual wills. We find ourselves confronted by the same alternatives. For, once the first parallelogram is given, we only have a formal resultant, which is not equivalent to the definitive resultant. The definitive resultant will be the resultant of an infinity of resultants, that is, the product of an infinite proliferation of parallelograms. Once again, either we trust to the infinite (that is, the indeterminate, epistemological void) for the production in the final resultant of the resultant we are hoping to deduce: the one that will coincide with economic determination in the last instance, etc., that is, we trust a void to produce a fullness (for example, within the limits of the purely formal model of the composition of forces, it does not escape Engels that the said forces present might cancel one another out, or oppose one another . . . under such conditions, what is there to prove that the global resultant might not be nothing, for example, or at any rate, what is there to prove that it will be what we want, the economic, and not something else, politics, or religion? At this formal level there is no assurance of any kind as to the content of the resultant, of any resultant). Or we surreptitiously substitute the result we expect for the final resultant, and duly rediscover in it, along with other, microscopic determinations, the macroscopic determinations which were secreted in the conditioning of the individual at the outset; this expected result, these macroscopic determinations will be the economy. I am obliged to repeat what I have just said of what was beneath the immediate level: either we stay within the problem Engels poses for his object (individual wills), in which case we fall into the epistemological void of the infinity of parallelograms and their resultants; or else, quite simply, we accept the Marxist solution, but then we have found no basis for it, and it was not worth the trouble of looking for it.

So the problem we face now is this: why is everything so clear and harmonious at the level of individual wills, whereas beneath this level or beyond it, it all becomes either empty or tautological? How is it that this problem, so well posed, corresponding so well to the object in which it is posed, should become incapable of solution as soon as we move away from its initial object? A question which must remain the riddle of riddles until we realize that it is this initial object which commands both the transparency of the problem and the impossibility of its solution.

Indeed, Engels’s whole proof hangs by that very particular object, individual wills interrelated according to the physical model of the parallelogram of forces. This is his real presupposition, both in method and in theory. In this respect the model does have meaning: it can be given a content, it can be manipulated. It ‘describes’ apparently ‘elementary’ bilateral human relations of rivalry, competition or co-operation. At this level what was previously the infinite diversity of microscopic causes might seem to be organised in real, and discrete, and visible unities. At this level accident becomes man, what was movement above becomes conscious will. This is where everything really begins, and it is from this point that deduction must begin. But unfortunately this so secure basis establishes nothing at all, this so clear principle merely leads to darkness – unless it withdraws into itself, reiterating its own transparency as a fixed proof of all that is expected of it. Precisely what is this transparency? We must recognise that this transparency is nothing but the transparency of the presuppositions of classical bourgeois ideology and bourgeois political economy. What is the starting-point for this classical ideology, whether it is Hobbes on the composition of the conatus, Locke and Rousseau on the generation of the general will, Helvetius and Holbach on the production of the general interest, Smith and Ricardo (the texts abound) on atomistic behaviour, what is the starting-point if not precisely the confrontation of these famous individual wills which are by no means the starting-point for reality, but for a representation of reality, for a myth intended to provide a basis (for all eternity) for the objectives of the bourgeoisie? When Marx had so thoroughly criticised the myth of the homoœconomicus in this explicit presupposition, how could Engels return to it so naively on his own account? How, if not by a fiction quite as optimistic as the fiction of bourgeois economics, a fiction closer to Locke and Rousseau than to Marx, could he suggest to us that the resultant of all the individual wills and the resultant of these resultants, actually has a general content, really embodies determination by the economy in the last instance. (I am thinking of Rousseau, whose dearest wish was that the particular wills, cut off from one another, might come together in a fair vote, producing that miraculous Minerva, the general will !). The ideologues of the eighteenth century (with the exception of Rousseau) did not demand that their presupposition should produce anything but itself. They just asked that it should provide a basis for the values already embodied in the presupposition, and that is why the tautology did have a meaning for them, but one obviously denied Engels, who, for his part, hoped to discover the exact opposite of the presupposition.

This is why, in his own text, Engels ultimately reduces his own claims almost to nothing. What is there left of his schema, his ‘proof’? Just the one expression, that given the whole system of resultants, the final resultant does contain something of the original individual wills: ‘each contributes to the resultant and is to this degree involved in it’. This is a thought which in a quite different context might reassure minds uncertain of their grasp on history, or, given that God is dead, uncertain of the recognition of their historical personality. I would go so far as to say that it is a desperate, honest thought which nourishes despair, that is, hopes. (It is no accident that Sartre, basing himself on Engels’s ‘question’, on the question of the ‘basis’ for and genesis of the necessity of history, should pursue the same object, with arguments which are equally philosophical, while of quite different inspiration.)

What have we left now? One sentence in which the final resultant is no longer long-term economic determination, but ‘the historical event’. So individual wills produce historical events! But a closer look shows that strictly speaking we can only admit that the schema gives us the possibility of an event (some men confront one another: something must happen, or nothing, which is also an event waiting for Godot), but absolutely not the possibility of a historical event, absolutely not the reason that will distinguish a historical event as such from the infinity of things which happen to men day and night, things which are as anonymous as they are unique. The problem must be put the other way round (for once !), or rather, in a different way. It is never possible to explain a historical event – not even by invoking the law which makes quantity pass over into quality – if one proposes to derive it from the (indefinite) possibility of non-historical events. What makes such and such an event historical is not the fact that it is an event. but precisely its insertion into forms which are themselves historical, into forms which have nothing to do with the bad infinity which Engels retains even when he has left the vicinity of his original model, forms which, on the contrary, are perfectly definable and knowable (knowable, Marx insisted, and Lenin after him, through empirical, that is, non-philosophical, scientific disciplines). An event falling within one of these forms, which has the wherewithal to fall within one of these forms, which is a possible content for one of these forms, which affects them, concerns them, reinforces or disturbs them, which provokes them or which they provoke, or even choose or select, that is a historical event. So it is these forms which command the whole, which already contained the answer to Engels’s false problem – whose solution, it must be admitted, he could never have reached, since there was never any other problem than the one he posed on the basis of purely ideological presuppositions since it never was a problem!

Of course, once again, there had certainly been the semblance of a problem for bourgeois ideology: to rediscover the world of history on the basis of principles (the homoœconomicus and his political and philosophical avatars) which, far from being principles of scientific explanation, were, on the contrary, merely a projection of its own image of the world, its own aspirations, its own ideal programme (a world which would be reducible to its essence: the conscious will of individuals, their actions and their private undertakings . . .). But once this ideology without which this particular problem could not have been Dosed, had been swept aside by Marx, how could this problem still remain the problem posed by this ideology, that is, how could it still remain a problem?

To close this too lengthy commentary, allow me two more remarks: one epistemological; the other historical.

It should be noted, vis-a-vis Engels’s model, that every scientific discipline is based at a certain level, precisely that level at which its concepts find a content (without which they are the concepts of nothing, that is, they are not concepts). Such is the level of Marx’s historical theory: the level of the concepts of structure, superstructure and all their specifications. But if the same scientific discipline should set out from another level than its own, from a level which is not the object of any scientific knowledge (such as, in our case, the genesis of individual wills from the infinity of circumstances and the genesis of the final resultant from the infinity of parallelograms . . .), to produce the possibility of its own object and of the concepts corresponding to it, then it will fall into an epistemological void, or, and this is what gives it its vertigo, into a philosophical fullness. This is the fate of the search for a basis Engels undertook in his letter to Bloch: and we find it impossible to distinguish in it between the epistemological void and the philosophical vertigo, since they are nothing but one and the same thing. Precisely this passage, with its arguments borrowed from the kinds of model used in the natural sciences (and ultimately this is their only precaution, a purely moral one), Engels is merely a philosopher. His use of a reference ‘model’ is philosophical. I insist on this point deliberately, for there is a more recent example of the same kind, that of Sartre, who also tries to find a philosophical basis for the epistemological concepts of historical materialism (he has one advantage over Engels in this respect, he knows it and says so). It is enough to refer to certain pages from the Critique de la raison dialectique (for example, pp. 689) to see that if Sartre rejects Engels’s answer and his arguments, he basically approves of the attempt itself. All that separates them is a quarrel over- the means employed, but on this point they are united in the same philosophical task. It is only possible to bar Sartre from his path by closing the one Engels opened for him.

But this makes it necessary to pose the problem of the philosophical leanings which can be detected in certain of Engels’s writings. Why beside genial theoretical intuitions do we find in Engels examples of this step backwards from the Marxist critique of all ‘philosophy’? This question could only be answered by a history of the relations between Marxist thought and ‘philosophy’, and of the new philosophical theory (in the non-ideological sense) which Marx’s discovery brought with it. Obviously, I cannot enter into this problem here. But perhaps we have to be convinced of the existence of the problem before we will find either the will or the way to pose it correctly and then resolve it.

Further reading: Dialectical and Historical Materialism Stalin | Georg Lukacs | Jean-Paul Sartre | Concepts of Capital, Geoff Pilling