Althusser's 'return to Marx'

Louis Althusser and socialist strategy posted by lenin

I was going to proceed with the series on Poulantzas, but to do so properly, I need to address the influence of Louis Althusser. There is, as Ellen Meiksins Wood has pointed out, a trajectory that can broadly be sketched with Althusser, Poulantzas and Laclau/Mouffe as its three compass points: from Maoism to Eurocommunism to 'radical democracy' and the abandonment of class politics as a 'fundamentalist', 'essentialist' error.

Yet to simply read the failings of his followers back into Althusser's project would be a travesty as unfair as E P Thompson's execration of the 'Stalinist' Althusser. It is, in fact, a bitter irony that many of Althusser's followers ended up in the social-democratic camp, as this was precisely the fate that his audacious and original reconstitution of Marxism was intended to help avoid. The fact that it didn't is not necessarily a reflection on the failings of the project; rather, it shows that Althusser was perhaps over-confident in the ability of revolutionary theory to overcome the deficiencies of political practice - particularly on the part of the French communist party (PCF) of which he was a member for most of his political life. He was to acknowledge a "theoreticist deviation" among his failings when he came to review and rectify his work in later years.

For, the motivation for Althusser's 'return to Marx', and the attempt to found a 'left-wing' critique of 'theoretical Stalinism', was political and strategic. As he put it, "that there can be no tactics that do not depend on a strategy – and no strategy that does not depend on theory". So, I will recount, in a highly abbreviated form, the elements of this philosophical enterprise with an eye to its political context - while, in deference to the subject, respecting the autonomy of theory. Also, since I will make no attempt to be original, I'll append a very short, selective bibliography at the end of this post. For now, I'll just say that by far the best guide is Gregory Elliott's Althusser: The Detour of Theory, which is thankfully in print again.

The "dogmatist night"

Louis Althusser joined the PCF in 1948, just as the Cold War was commencing. The party had emerged from its participation in World War II, however belated, with considerable prestige. In the 1945 elections, it won the largest share of the vote, with 26% - 5m votes concentrated in the manual working class. Its membership was at half a million, and would increase to about 800,000 in the next few years. Its leaders joined the Cabinet, and did their best to direct popular radicalism into the reconstruction of French capitalism. Yet, it was at just this point that it was about to be thrust into a political ghetto. The marginalising forces were precisely those of the Cold War. The Socialists under Guy Mollet joined the 'Atlanticist' camp, forced the Communists out of government, and led some brutal class struggles against French workers, not to mention vicious colonial counterinsurgency campaigns. At the same time, Moscow-oriented parties were being instructed to take a 'left turn'. This meant breaking from the 'Popular Front' alliances, and polarising between the pro-peace socialist camp and the dupes of imperialism. At an ideological level, it meant dusting off the shibboleths of 'Marxism-Leninism' which had been stowed for the duration of the 'Popular Front' period, and denouncing a host of ideologies (such as psychoanalysis, for example) as 'bourgeois', 'cosmopolitan' and 'reactionary'. On a more positive level, it also meant that the party was more willing to side with workers in ongoing struggles. But amid a wave of anticommunist hysteria, the PCF found itself increasingly isolated and unable to build on its previous successes. The resulting loss of éclat saw the party retreat to an ideological fortress, with party theoreticians extirpating heresy and administering what Althusser would characterise as a "dogmatist night".

Yet, for him, the PCF was the party of the French working class, and the only party capable of posing a threat to French capitalism. Therefore it was a duty of Marxist intellectuals to be members. He was not unique in this judgment. Sartre, no ally of Stalinism, nonetheless insisted at the height of the Cold War in 1952 that a defeat for Stalinism in France would be a defeat for the working class. The PCF was the sole unifying instance capable of imparting "class-being" to the workers. So, Althusser bore the long night of dogma in relative silence in acceptance of party discipline. It was not until the événements of 1956 - in February, Khrushchev's 'secret speech' at the Twentieth Congress of the Communist Party of the Soviet Union, attacking Stalin's legacy; and in November, the invasion of Hungary - that Althusser began to break the silence. The glasnost following Khrushchev's speech allowed heterodox currents to emerge, most prominent among which were various strands of Marxist Humanism.

It was the fact that the regime itself was incorporating this humanism into its official ideology, and the fact that the PCF could do the same without skipping a beat - the PCF witchfinder-general Roger Garaudy was converted overnight into a humanist and supporter of theoretical pluralism - aroused Althusser's suspicions. Moreover, this happened without affecting the underlying tenets of 'Marxism-Leninism' in the slightest. This de-Stalinization, he inferred, was a right-wing de-Stalinization that would lead to the social-democratization of communism. Already, the PCF had been attempting to navigate its way out of the ghetto via an alliance with the Socialists - "democratic adventurism" in Althusser's assessment - contributing to a political passivism, support for colonialism in Algeria, and eventually blank perplexity when General Charles de Gaulle established a dictatorship, the 'Fifth Republic', in May 1958. In resisting this tide, Althusser aimed to strengthen the political practice of French communism. As he put it:

"I would never have written anything were it not for the Twentieth Congress and Khrushchev’s critique of Stalinism and the subsequent liberalisation. But I would never have written these books if I had not seen this affair as a bungled destalinisation, a right-wing destalinisation which instead of analyses offered us only incantations; which instead of Marxist concepts had available only the poverty of bourgeois ideology. My target was therefore clear: these humanist ravings, these feeble dissertations on liberty, labour or alienation which were the effects of all this among French Party intellectuals. And my aim was equally clear: to make a start on the first left-wing critique of Stalinism, a critique that would make it possible to reflect not only on Khrushchev and Stalin but also on Prague and Lin Piao: that would above all help put some substance back into the revolutionary project here in the West."

Thus, he began work on his reconstitution of Marxism. The hallmark of the ensuing period was tactical quietism and strategic offensive. He forebore criticisms of PCF policy, while working to transform it in the longue durée by means of a theoretical assault on PCF dogma.

"The first left-wing critique of Stalinism"

Before going any further, I should say something about Althusser's attitude to Stalinism. First of all, his claim to be embarking on the "first left-wing critique of Stalinism" implies an ignorance of Leon Trotsky's work, among others. This meant he initiated his critique from within the radius of Stalinism, cutting himself off from extant critical resources. Inevitably, he had to make use of what was there 'in front of him'. Thus, despite attacking Stalinist dogma and lamenting its victims, he did not hesitate to endorse aspects of Stalin's thought if it was useful to him in the theoretical struggle. Despite Stalin's own theoretical dilettantism, he did not resile from attributing theoretical 'discernment' and 'perspicacity' to him at times. And he obliviously championed 'Marxism-Leninism' against the regime, as if it was not the official ideology, and not therefore a caricature of both Marxism and Leninism.

Second, his ideological pole star throughout this period was the People's Republic of China (PRC), whose attack on Khrushchevite 'humanism'* provided much of the stimulus for this "left-wing critique". (*Khrushchevism "substitutes humanism for the Marxist-Leninist theory of class struggle and substitutes the bourgeois slogan of Liberty, Equality, Fraternity for the ideals of communism. It is a revisionist programme for the preservation and restoration of capitalism".) This obviously left him to some extent hostage to the credibility of the PRC. One can also read into his theoretical work certain related political connotations. Ellen Wood rightly points out that Althusser's insistence on the relative autonomy of the 'instances' is congruent with the politics of the Cultural Revolution. Still, one must take care to avoid the tone of a health warning here: "the theoretical formulation of the 'relative autonomy' of 'instances', if taken in extreme quantities, can lead to Maoism". It is not reducible to any such political context. Less so is this theoretical innovation exclusively governed by the need to explain how a Stalinist bureaucracy can arise in a 'socialist' economy - various strands of dissident Marxism managed to explain this to their own satisfaction without recourse to Althusser.

For Althusser, the two dominant theoretical purviews, which he classed as 'humanism/historicism' and 'economism', were symmetrical in their teleological structure. Economism was the "poor man's Hegelianism" shared by both Kautsky and Stalin, a Marxist version of technological determinism in which the development of the productive forces was the sole determinant of historical change and guarantor of the inevitability of socialism. Humanism/historicism, in place of productive forces, posited 'Man' or the working class as the constitutive subject of history, whose progress would bring history to its communist terminus. In an extremely schematic way, these errors could be understood respectively as a rightist and a leftist deviation. As he explained to a New Left Review interviewer: "The rightist deviation suppresses philosophy: only science is left (positivism). The leftist deviation suppresses science: only philosophy is left (subjectivism)".

The chief problem with humanism/historicism was that it relativised Marxism and thus brought its epistemological status into doubt. By describing Marxism as an ideology of the proletariat whose validity could be verified by its contribution to the historical advance of the class, it effaced the gap between science and ideology. For Althusser, the scientificity of Marxism was what distinguished it from all speculative philosophies of history. To compromise its scientificity was to weaken its explanatory power. And there was no doubt for Althusser of its extraordinary power. This "scientific revolution" had opened up the 'continent' of History to scientific knowledge, in exactly the same way that Thales had opened the 'continent' of mathematics and Galileo the 'continent' of physical nature. Marxists were "only just beginning to explore" this new continent, but the explorations were already at risk from theoretical revisions. Althusser acknowledged the strengths of humanism as a rejection of an inhuman and dogmatic tyranny, and welcomed the intellectual thaw that it signposted. Nonetheless, as Gregory Elliot puts it, "it ultimately summed up to a romantic anti-capitalism and anti-scientism, as politically voluntarist as it was philosophically speculative".

Even Althusser's sympathetic reviewers are damning about Althusser's schema here. His engagement with Marxist humanism was unidimensional, dealing only with the doctrine as espoused by the theoreticians of Communist parties after 1956 - "Stalinism with a human face" - and ignoring the work of those who left the Communist parties but remained Marxists. The category of 'humanism/historicism' assimilates an incredibly diverse array of thinkers, from Lukacs to Korsch, Gramsci, Sartre, Merleau-Ponty, Goldmann and Della Volpe (avowedly anti-Hegelian). He does not acknowledge the Frankfurt School, and despite his hostility to revisions which dilute Marxism, his own approach would involve vitalising Marxism with the supplement of rationalist philosophy in the form of Spinoza and Bachelard. And it is a point repeatedly made, with justice, that Althusser's tendency to subsume qualitatively distinct theorists under a single category of theoretical error - the "Continent of Theoretical Error" as Norman Geras described it in his spirited joust with Laclau and Mouffe - is redolent of the tendency he discerned and derided in Hegel to characterise all the manifold phenomena of a conjuncture as mere expressions of a single moment in the development of the Idea.

Nonetheless, this version of 'humanism' formed the negative reference point for Althusser's attempt to 'return to Marx', and this much may be said for it right away: it neither concedes a single thing politically to those using the critique of Stalinism to move to the right, nor does it simply lapse into a conservative fortification of dogma. Althusser's response to what he saw as a crisis of revolutionary theory was to undertake a critical and robust re-evaluation of Marxism. And the question he posed, in For Marx, was fundamental: "What is Marxist philosophy? Has it any theoretical right to existence? And if it does exist in principle, how can its specificity be defined?"

As should be clear by now, this philosophy was to owe little to Hegel. Engels had attempted to extract the rudiments of dialectical materialism from Hegel's Logic, standing the dialectic 'on its feet' the better to "discover the rational kernel within the mystical shell", as Marx described it in an Afterword to Capital volume one. The 'three laws' of the dialectic in Engels' Anti-Duhring were lifted straight from the Logic and applied to what were then the most cutting edge ideas in science. Althusser argued that this formula of standing of the dialectic on its feet was utterly misunderstood if it taken to mean that one simply had to invert Hegel's postulates, putting matter in place of spirit: this "will merely produce a new materialist metaphysics". Marxists who took this step had misunderstood what Marx was doing Capital, volume one, part one. For here, the apparently Hegelian structure of Marx's exposition is akin to a process of peeling back the mystical layers of Hegel's approach to extract whatever is rational - though, contrary to the kernel/shell metaphor, he insisted that even the kernel had to be 'contaminated' by Hegelian idealism. The major contributions of Hegel were his attack on Kantian subjectivism which could provide the basis for a materialist defence of scientific objectivity, and the concept, implicit in the Absolute Idea, of " a process without a subject" which informed Marx's analysis of Capital. I should say that Althusser's critics justifiably point to his highly selective reading of Marx's writings, thus ignoring a great part of the 'mature' work that is still determinedly Hegelian. In fact, however, this point rebounds to Althusser's advantage: the most Hegelian passages in, for example, Capital, are also the most liable to lapse into dogmatically deterministic assertions. And Althusser, after all, never said that the 'epistemological break' separating the young from the mature Marx was a clean break - quite the contrary.

But, if Althusser otherwise rejected the Hegelian legacy, insisting that what Marx offered was more than a Hegelianised version of English political economy, from whence could this philosophy emerge? It was implicit, he said, in "the logic of Capital". It was also implicit in revolutionary practice, particularly in what he regarded as the two great successes of Marxism, the revolutions of 1917 and 1949. This is what motivated the 'return to Marx', and it is why he could brush off charges of 'scholarly fetishism' with airy contempt.

The primacy of practice

Marxist philosophy, in Althusser's schema, is one of two distinct levels of Marxist theory. These are: historical materialism, being "the Marxist science of the development of social formations"; and dialectical materialism, being the as yet unconstituted philosophy that is implicit in Marx's work. This philosophy would, if elaborated, form the epistemological bedrock of historical materialism, point out its flaws, suggest solutions, and supply concepts adequate to its problems. Above all, it would safeguard the theory as a scientific practice against debilitating ideological dilutions. If historical materialism is the science of history, then, dialectical materialism is the "theory of science and of the history of science", or more broadly "a theory of theoretical practice". One upshot of this conception of theory as a practice, is to render the problem of the unity of theory and practice unintelligible.

Among Althusser's first tasks is to discern what the "theoretical practice" is, of which dialectical materialism is the theory. The concept of practice itself, the "primacy of practice" in Marx, is central: "We can assert the primacy of practice theoretically by showing that all the levels of social existence are the sites of distinct practices: economic practice, political practice, ideological practice, technical practice and scientific (or theoretical) practice." But to speak of an undifferentiated "practice" is to succumb to ideology. Marx's breakthrough was to provide a "theory of the different specific levels of human practice". These different levels - economic, political, and ideological - are the 'relatively autonomous' 'instances' of the social totality. Practice can nonetheless be defined as "any process of transformation of a determinate given raw material into a determinate product, a transformation effected by a determinate human labour, using determinate means (of 'production')". The determinant moment in this process is the labour of transformation itself. Theoretical practice, then, is a form of production, divisible into three 'Generalities': Generalities I, a set of raw materials (concepts, abstractions) into relation with one another; Generalities II, the "means of theoretical production" which are brought to bear on the raw materials; and Generalities III, the end product, knowledge or the thought-concrete.

Althusser proceeded to refine this understanding of knowledge-as-production against the empiricist conception of knowledge as vision. For empiricism, knowledge begins with a relation between subject and object: the 'object' of knowledge is objective reality, on which the subject performs a labour of abstraction through which it separates the essence from the inessence of the object. In so doing, the subject comes to possess the essence and thus has knowledge of it. For Althusser, this was flawed on two accounts. First, it was wrong to assume that the object of knowledge was objective reality. The object of knowledge is what results from the labour of abstraction, which forms the raw material for the process of theoretical production. Second, its two-ply model of reality actually inscribed the knowledge of reality (essence) within that reality itself, thus collapsing the distinction between thought and the real. For Althusser, it was crucial for materialism that the independence of reality from thought should be maintained.

It follows that Althusser must reject the empiricist conception of knowledge as a 'model' composed of abstractions, a model which is necessarily a poor cousin of the reality which it tries to emulate. If, in the empiricist view, the abstraction is always a hugely simplified 'essence' of the real object, and therefore theory can never be adequate to the complex reality it describes. For Althusser, however, the object of knowledge is not a given (the 'real object') but rather a raw material consisting of previously worked on concepts and abstractions, and is therefore susceptible to continual refinement by the means of theoretical production. There is no inherent limit to a theory fully adequate to its subject. This is extremely important. As Norman Geras put it, "if the object of knowledge in the strict sense is not the real object, the object which is known finally, via the object of knowledge, is the real object."

Following both Bachelard and Spinoza, Althusser insisted that science and ideology were wholly distinct kinds of knowledge. Though the process of knowledge production in each case is formally equivalent, science is marked by a foundational rupture, an 'epistemological break' with common sense. The opacity of the everyday is such that only such a profound break could found a genuinely objective knowledge. But how is it possible to tell the difference, to verify that what one is doing is science and not ideology? Althusser turned down the pragmatist option of judging a project by its successes: theory is successful because it is true, he argued; it is not true because it is successful. The proper way to judge a science was by criteria immanent to it. A research programme could only devise its own criteria of verification.

If this seems circular, that is because it is: as Gregory Elliot puts it, "there is something inherently dogmatic and viciously circular in employing Marxist philosophy to guarantee the status of Marxist science". Unable to avoid this problem, Althusser attempted to pose the question in another way: by what mechanism did theoretical practice, which took place exclusively at the level of thought, effect the cognitive appropriation of reality? But that did not yield a satisfactory answer either. As such, and given the manifest deficiencies of Marxism in the face of the last century's challenges, the claim for Marxism's scientificity looks shaky. Althusser would abandon the idea that dialectical materialism was a science of sciences, arguing in Lenin and Philosophy that philosophy was not itself a science but the class struggle in theory. Where that leaves the epistemological status of Marxism is something that is still open to debate.

Befrore proceeding, it's worth drawing out a few implications from this account of theoretical practice and its autonomy. Althusser was defending "the right of Marxist theory not to be treated as a slave to tactical decisions". But a quite different political consequence of this was a certain kind of elitism, in which Marxist theory was the work of specialists who should be left alone. To be clear, Althusser understood that Marxist theory could not have emerged without the workers' movement (in conjunction with certain "theoretical elements". His critics who ascribe to him a form of idealism in which the 'epistemological break' of Marxism is a purely theoretical phenomenon, estranged from class struggle, are wrong. The autonomy of theory is only relative, after all. Yet, for Althusser, the workers' movement did not, and by implication could not, produce Marxist theory "by its own devices". He takes the Lenin of What Is To Be Done to argue that Marxist theory must be imported into the working class movement from outside it, a hypothesis with which he concurs. Workers have a 'class instinct', which enables them to arrive at objectively correct proletarian 'class positions', but they are incapable of being spontaneously revolutionary. So much the worse with theorists, who are instinctively petty bourgeois, and must constantly militate against these prejudices. As it comes down to it, only the vanguard party provides the link between the class and revolutionary theory. Without the party, revolutionary theory cannot be imported into the class; without revolutionary theory, the class cannot make a revolution. And Althusser is insistent that the empirical data confirms this - revolutions have only been made where Marxist theory has been accepted by the workers' movement (Russia and China), whereas those situations where revolution is most distant also happen to be those where Marxist theory was never widely accepted in the workers' movements (Britain), or was adulterated in a social democratic fashion (Germany).

Althusser was thus committed to a version of Leninism which has these days, and for good reason, fallen out of favour. As Valentino Gerrattana argues, despite Althusser's intention to break with Stalin, he constantly evaded the question of the latter's relationship to Lenin and Leninism, never clarifying the issue of how much "theoretical Stalinism" was a break from Bolshevism. As a result, he committed himself to certain 'Leninist' positions that more closely resemble what Hal Draper called 'Leninology' than historical Leninism. Even so, I would again caution against using this political error to dismiss the theory. There is in fact no good reason why one should infer from the autonomy of theoretical practice that the working class movement can produce no revolutionary theory by itself. The fact that theoretical and political practices are distinct, does not mean that workers are incapable of performing both.

Signposts in the "continent" of history

In addition to these epistemological concepts, and the extremely interesting account of knowledge that they sum up to, Althusser elaborated a set of historical concepts. We have already introduced the idea of the relative autonomy of 'instances', that is of different levels of practice within a social totality. But what does such a 'totality' consist of, and what constitutes its different levels? Each level, precisely, is a structure, so that the social whole is a structure of structures.

Something of the nature of the structure is imparted in Althusser's discussion of 'Generalities II', where he introduces the notion of the 'problematic'. The problematic is a theoretical structure which brings the concepts and abstractions into a certain relationship with one another. But it is rarely explicit in the theory that it regulates: it is submerged. This is as true of Marxism as other doctrines, which is why it is necessary for Althusser and his followers to embark on close, symptomatic readings of Marx's texts to bring the problematic to light. Despite being submerged, the problematic plays an active role, and this not in any metaphorical sense. It is not the subject that 'sees' the objects that become known through theoretical practice; it is the problematic. The subject does not play the determinant role in knowledge production; the structure does. And this is typical. This structure is not an audible, visible, tangible object, but rather a "process without a subject" discernible solely through its effects. The only subjects pertinent to it are those constituted and governed by the structure. This is Althusserian anti-humanism. It is a commonplace that this total demotion of the subject was deeply problematic for any revolutionary praxis that might follow from it - after all, if subjects are constituted by the structure, wherein lies the possibility of their emancipation from it? Althusser attempted to deal with this by stating that this was a necessary explanatory step; the theoretical reduction of subjects accounted for the practical reduction of subjects under capitalism. Yet, this by itself didn't solve the problem of how subjects could escape this practical reduction. Ultimately, to avoid a structural-functionalist account in which the structures merely perform in ways that are functional for their reproduction, Althusser relies on his concepts of overdetermination and contradiction, which we'll come back to.

Crucially, the social totality, the structure of structures, is quite unlike Hegel's 'expressive' or 'spiritual' totality, in that the structure has no centre, and no original essence of which its parts are mere expressions. It is the sum of irreducible, multiple instances, each with its own history, each with its own complex structure and each governed by its own temporality. Importantly, each also has its own specific efficacy, so that the political and ideological are by no means merely reducible to the economic. In this sense, Althusser's account of the social has remarkably similar starting points to those of post-structuralists, albeit he draws radically different conclusions about the intelligibility of the structure, and his emphasis on "unity in difference" or "complex unity" distinguishes his difference from the free play of signifiers that is Derrida's différance. And he rescues his analysis from simple pluralism by asserting the determining role of the economic level 'in the last instance'. That is, though the social formation has no centre or essence, within it there is a "structure in dominance" in which one or other of the instances is dominant. And, it is the economic level that determines which of these instances will be dominant in any social formation. (I find it helps to think of the example of feudalism, in which mode of production the instance of politics was dominant, due to the specific mode of surplus extraction in the economic instance.)

It's also worth examining the difference between a social formation and a mode of production in Althusser's analysis. The difference between the two is partially a difference in the level of abstraction. The mode of production is the combination of the elements of a social whole at its most abstract; the social formation is the site in which those elements are concretely present in all of their specificity. The 'elements' of the mode of production can be assigned to an elementary table, listing the labourer, the non-labourer, some means of production, etc. Seemingly, these 'elements' are constant through all modes of production, with only their specific relationship in a given combination varying. Yet, this conclusion has to be qualified by Althusser's stress on 'overdetermination'.

The concept of overdetermination is taken over from Freudian psychoanalysis, in which it refers to the condensation of potent dream-thoughts, wishes, etc., in a single image. The 'overdetermination' of a point in the structure means, analogously, the condensation of all of the relations and 'contradictions' within the totality in that single node. 'Contradictions' in Althusser's useage are purely 'historical', referring to a number of different kinds of complexity, such as paradox, antagonism, ambiguity, combined and uneven development, etc. He does not subscribe to any of the materialist metaphysics of Anti-Duhring. When he refers to an accumulation of contradictions in a given social formation, he is referring to just those tensions, antagonisms and ambiguities that undermine the unity and cohesion of the formation and make possible its disintegration and overthrow. A consequence, at any rate, of overdetermination is that the elements in a mode of production cannot be the same regardless of how they are articulated. Since each point is overdetermined by every other point in the matrix, the specific content of these elements must vary depending on their articulation.

The concepts of overdetermination, determination in the last instance, and the structure in domination, were cited by Althusser to repudiate critics who described his work as structuralist. It is another matter entirely whether he succeeded in avoiding certain structural-functionalist temptations; certainly, his followers didn't always do so. (I mentioned Poulantzas' tendency to collapse into such explanations, at least in his earlier work). However, as Stuart Hall points out, by thinking systematically about the different levels and kinds of determination in a social formation, by locating this complexity and plurality in Marx, he enabled the thinking of concrete historical situations and ideological formations within the Marxist 'problematic', as well as the analysis of political situations in their specificity and lines of antagonism. Indeed, as we learn from Machiavelli and Us, one of the most important historical concepts for Althusser is the 'conjuncture' as an "aleatory, single case", comprising not merely a sum of elements but their unity in a 'contradictory' system. In this respect, Althusser is quite close to Gramsci, whom he otherwise lumped in with the 'historicists'.

We have seen that he avouched a strict distinction between ideology and science which seems to be in some doubt. Yet, one of the bases for this distinction is a suggestive analysis of the "lived relation" between "men" and their world. To be more precise, ideology expresses, not the relationship between "men" and their conditions of existence, but the way they live, or imagine that relationship. The idea of inhabiting the world without ideology is utopian and futile: "there is no practice except by and in ideology".

By the same token, however, "there is no ideology except by the subject and for subjects". The term 'imagine' alludes to Lacan's concept of the pre-symbolic 'imaginary order', and the associated doctrine of the 'mirror phase'. In the 'mirror phase', subjects come to recognise themselves in reality, and to imagine their relationship to that reality as if they were fully autonomous, constitutive subjects This, pace Althusser, is what ideology does: it constitutes subjects so that they imagine themselves to be constitutive subjects; it "interpellates" them. And in giving them an imaginary relationship to their actual relation with the world, it distorts their real situation and binds them to the social structure. This concept of ideology as, not merely erroneous beliefs but a lived relationship to one's situation, has proven enormously fruitful. Much of Hall's splendid repertoire is unthinkable without it, as he is the first to acknowledge.

But it is not just this aspect of the interpretation of ideology that is relevant here. Again, the 'relative autonomy' of the instances has some bearing here. For, while it was a staple of certain crude, mechanistic Stalinism that certain ideologies had a necessary class belonging (this played a useful polemical role if one wanted to debate intellectual opponents), it has equally become a staple of poststructuralist ideologies that there is no correspondence whatever between ideological struggles, which are purely contingent, and the 'economic base'. Thus, Laclau and Mouffe insisted that Gramsci's theory of hegemonic struggle was burdened by an essentialist remnant, which tried to anchor these struggles in a unified economic level - but only if contingency was expelled from the economy could this anchoring be effective. So, Laclau and Mouffe denied any correspondence between classes and ideologies. I think what you find in Althusser's conception, however, is quite different from both. The 'relative autonomy' of instances allows that there is no necessary class connotation to a given idea, (Althusser was witness to all those futile attempts by the Stalinists to bissect scientific theories according to some imputed class position), but does not permit that there is necessarily no such connotation. This allows for a certain open-endedness of political struggles and class practices, for the aligning of heterogenous interests and perhaps even antagonistic subject-positions. Moreover, Althusser's stress on the practical aspect of ideology underlines that any successful articulation between different elements has to be constructed.

No strategy without theory... theory without strategy?

If there is no strategy without theory, it does not follow that a given theory is suited only to one strategic purpose. The relation between political and theoretical practice is more complex, less determinate, than such a view would allow. So, if Althusser did conceive of his theoretical innovations as political interventions, it doesn't mean that he followed what might be the expected path that his theoretical work would imply. Politically, he was neither a Stalinist nor a reformist; theoretically, he was waging an attack on the practice of the French communists. Yet, membership of the party was his sole route, as he perceived it, to an organic connection with the working class and thus to meaningful theoretical work. And because he did not break fully with Stalinism, he could not break fully with reformism, and remained bound to a party moving perpetually in a social democratic direction. Nor did he help renew PCF political practice along revolutionary lines. Toward the end of his career, the PCF was embarking on a Eurocommunist route to the margins.

Althusser's career ended in a squalid tragedy when, under the influence of a mental illness that had grown more intense over the years, he strangled his wife, Hélène. K S Karol's article, 'The Tragedy of the Althussers', listed below, is one of the few that remembers Hélène Althusser for something other than this horrible end - recounting her role in the French Resistance, as a Communist, and as a social scientist. Louis Althusser was consigned to an asylum, before living more or less hermetically until his death. Did Althusser's project fail? By the scale of its own extraordinary ambition, it did fail in many respects. It particularly failed to establish the absolute scientificity of Marxist theory, and thus its safeguard against party bosses, revisions and rivals. Yet, given the real deficiencies of Marxism exposed by the twentieth century experience, there is perhaps only a Beckettian choice between failing and failing better. Althusser failed better.

***

Select bibliography: Louis Althusser & Étienne Balibar, Reading Capital, NLB, 1970; Louis Althusser, For Marx, Allen Lane, 1969; Louis Althusser, Lenin and Philosophy and Other Essays, NLB, 1971; Louis Althusser, 'On the Twenty-Second Congress of the French Communist Party', New Left Review 104, July-August 1977; Louis Althusser, The Humanist Controversy and Other Writings, Verso, 2003; Gregory Elliott, Althusser: The Detour of Theory, Brill, Leiden & Boston, 2006; Norman Geras, 'Althusser’s Marxism: An Account and Assessment', New Left Review 71, January-February 1972; K S Karol, 'The Tragedy of the Althussers', New Left Review 124, November-December 1980; Alex Callinicos, Althusser's Marxism, Pluto Press, 1976; Alex Callinicos, Is There A Future for Marxism?, Macmillan, 1982; Valentino Gerratana, 'Althusser and Stalinism', New Left Review 101-102, January-April 1977; Stuart Hall, 'Signification, Representation, Ideology: Althusser and the Post-Structuralist Debates', Critical Studies in Mass Communication, Vol 2, No 2, June 1985