In lieu of more substantive matter which I'm busy working on, and because I have so much else to do, I'm just posting up these possibly interesting reflections on ideology which I threw together this morning in a fit of ennui.

Where do you get those ideas from? There is an extremely high chance that most of them were produced by others for you, that you came by them in schools, your family, from the mass media, maybe a religious institution, from a political party, a trade union, the academia, think-tanks, PR firms or any one of a network of institutions or apparatuses which make it their business to produce ideas for you. (Yes, you - this is interpellation we are talking about here.)  This is how Goran Therborn in one of his early works, The Power of Ideology and the Ideology of Power, visualises the process:

I know this isn't a shock. You didn't really think you came up with those thoughts all by yourself. You are Marxists, and you know very well that ideologies (and sciences for that matter) are produced collectively, even if not in a collectivist mode.  Or, if you prefer, they are historically, socially produced.

But what do those who aren't Marxists think? Aren't they often inclined to believe that somehow their ideas are spontaneous, natural, or arise from a process of open-ended personal reflection? I was going to say it was a particularly petty bourgeois illusion, but it seems to be implicit in ideology as such, the belief that ideas are the result of some inner spark or inspiration: in theoretical language, that agents, rather than being constituted as subjects by ideology, are the constituting subjects of ideology.  "I decide what I think," the subject of ideology says, "I mean, yes, there are all sorts of ideas being presented to me every day, but as soon as I obtain adulthood I am in a position to critically evaluate those ideas.  I choose my ideology; it doesn't choose me."  In this sense, the concept of interpellation is an affront to common sense, for it says that ideology 'chooses' its subject, that it addresses the subject with who she really is, enabling her to become the 'bearer' (Trager) of the role assigned to her by the structure.

We won't spend too long inspecting the structure of this interpellation.  Althusser sets up a primordial 'scene' of interpellation, a fable in which the agent is 'hailed' by the law ("hey you!"), stops, turns, and in the act of turning is subjectivated, accepts that she really is the "you" addressed by law.  Crucially, this submission to ideology involves misrecognition: as with Lacan's 'mirror stage', the resulting identity is a false totalization of the fragments of experience.

Following Judith Butler, we could problematize this concept along the following lines.  Althusser relies on a complex set of theological figures to set up his notion of 'hailing'.  The very act of 'turning' and thus accepting subjectivation implies an act of conscience, an appropriation of guilt.  For Althusser, ideology is centred around an imaginary 'Absolute Subject' - redolent of Lacan's 'Big Other', the guarantor of the symbolic order - which is experienced by the subject as the source of 'law', guaranteeing that (so long as you are a good subject and 'behave') everything is just as it seems and will continue to be for eternity, and will be okay.  Religious ideology is thus treated as exemplary of ideology as such, the 'naming' in religious ritual typical of subjectivation.  There is no mystery about this - Althusser practised Catholicism long after he became a marxist, and seems never to have got out of the habit of performing some of its rituals.  But this process requires an addressee already predisposed to receiving such a message, already implicated in its terms; that is, it requires a subjection that already presupposes a subject. 

At first it seems as if we're saying that the narrative of the 'scene' of interpellation cannot be 'true'.  But, of course, as Butler also notes, this concept of interpellation is allegorical.  It stages a 'hailing' that never actually literally happens in this way; it narrativises a process or set of processes that is said to be resistant to narrativization.  As Butler has it, "the 'call' arrives severally and in implicit and unspoken ways", which a too literal reading of 'interpellation' might overlook.  There is a long process in which each agent in the division of labour is equipped with, for example, the sorts of linguistic skills appropriate to her location, which will enable her to carry out the tasks assigned to her by the structure: a boss learns to 'speak properly' (ie authoritatively, persuasively, dominatively) to her employees; a worker learns to 'speak properly' (ie deferentially, respectfully, warily) back.  Over time, the agent becomes a subject by internalising the rules and attitudes, materialised in actions, inserted into practices, governed by rituals, which constitute the subject.  The agent, just as she is mastering these skills, submits to the dominant ideology: is 'hailed', 'turns' to face the law, and is subjectivated. The fabular character of 'interpellation' is one reason why the many pointed criticisms of the 'scene' depicted by Althusser - for example, that it is is too dyadic, or is phallocentric - are thoroughly well-founded, yet far from adequate to finish off the concept of interpellation.  The scene could be re-staged in multiple thoughtful and creative ways without losing what is essential.

However, Butler's case is more complex.  It is not simply that the 'scene' belies the reality of ideology in many ways.  The problem lies in the prehistory of the subject alluded to above; the preconditions necessary for the reproduction of the dominant ideology.  'Interpellation' appears to require an anxious, knowing, 'guilty' agent anxious to acquit herself by conscientiously mastering the skills that will make her a subject.  Doesn't this, as Lacanian critics approvingly suggest, require another dimension to the subject, a materia prima, or rather an immateria prima, a psychic dimension that is radically distinct from any possible incarnation in material practices?  Doesn't it require the resuscitation of the Cartesian subject?  Doesn't Althusser's repudiation of the ontological dualism of the material and the ideal simply collapse into another dualism?  Butler's rejoinder to the Lacanians (in the person of Mladen Dolar) is that: i) it mistakes a grammatical requirement for a pre-existing subject, or psychic remainder, in the context of exposition for a logical requirement; and ii) its rejection of materialist monism relies on a reduction of the material to the phenomenal, or empirically given, so that the absences that punctuate a ritual can be seen as governed by a non-material symbolic order - in counterpoint, Butler deploys the althusserian materialism of articulation, in which these absences are not 'ideal' but are bound to the phenomenon as "its constitutive and absent necessity".

Nonetheless, the terms of Althusser's formulation, the theological figures deployed and particularly the restrictive role of conscience in subject-formation, leave unresolved the question of the failure of interpellation.  In Althusser's terms it is difficult to think how a 'bad subject' could exist, since the very condition of being a subject involves mastering the skills that subject one to the dominant ideology and make one therefore a 'good' subject.  

Now, Butler doesn't do any more than signpost possible ways out of this dilemma: hinting at forms of de-subjectivation which might allow one to oppose the 'law' without denying one's complicity in it.  But it's clear that without resolving this, the danger is either that one collapses into a mechanistic, functionalist account of ideology, or a voluntarism or decisionism predicated implicitly or explicitly on this theological remainder.  Althusser didn't solve this problem.  As Jameson points out, the ISA essay was "programmatic", a manifesto of sorts, an "agenda, still incompletely fulfilled".  And I will have occasion to talk about the politics of this in a future post.  However, I would say that this psychic remainder, this kernel of interiority irreducible to the social and material world, constantly recurs in marxist accounts of ideology, either explicitly or in the form of ambiguities or silences.

The ambiguity, or unresolved tension, in historical materialist approaches to ideology is basically this: is ideology materially determined, or is it a material substance in itself?  This is an ambiguity which persists in one of the greatest marxist writers on ideology, Volosinov who, asserting that his approach is monistic, nonetheless makes this statement: "Every ideological sign is not only a reflection, a shadow of reality, but is also a material segment of that very reality."  Not only, but also.  Not just shadow, but material too.  Does this seem like a detail?  Well, I can't be the only one to have encountered this sort of ambivalence repeatedly.  Goran Therborn, arguing from a post-althusserian perspective, tended to speak of the material determination of ideas, leaving the question of ideology's ideal or material status unresolved.  

Or one encounters more directly formulations which tell us that "mind is developed upon the basis of matter", but that "the human mind cannot simply be reduced to matter". And is this not simply a straightforward exposition of Marx's approach in The German Ideology?  For on the one hand, Marx says that "neither thoughts nor language in themselves form a realm of their own, that they are only manifestations of actual life", while on the other these manifestations are "phantoms formed in the human brain ... sublimates of their material life-process ... bound to material premises".  And so, the ambiguity seems to be resolved on the side of the material determination of ideas, which are but "phantoms".

If in fact the ambiguity is to be resolved in this way, that is if we accept that there is an order of reality that is separate from matter (a subjective reality, a 'consciousness' that is linked to the material world but not directly or wholly of it), then it is surely at the expense of monism.  To say that this is a consequence of such a stance is not necessarily to disprove it.  And it is clear why it is attractive: precisely because of the need to found emancipatory politics on the self-activity of the masses, and therefore the need to explain the bases of resistance.  Somehow, the working class is capable of being an agent of revolutionary change, despite the effects of the dominant ideology.

So when Althusser argued that ideology has no "ideal or spiritual existence" but only a material existence, that it exists only "in an apparatus, and its practice, or practices" - gamesomely suggesting that readers entertain the notion sympathetically on the grounds of materialism - surely he was expressing a conception that was deeply controversial to the majority of marxists.  At least, particularly  for those of a Lukacsian persuasion, such a conception must appear to be complicit in the objectification that is characteristic of ideology as such.  For if ideas themselves are material, then its bearers might well be nothing but objects; nothing but effects of a structure, bearers (Trager) of a role assigned to them by the structure.  There would seem to be no space for will or intention ('consciousness') in such a process.  So, it is this potentially mechanistic consequence of theoretical anti-humanism that is objectionable.

But, returning to Marx's formulation in The German Ideology, if ideas are "only manifestations of actual life", in what sense are they manifest?  To whom and in what they do they manifest themselves, if not in fact in the material practices and apparatuses?  In what sense are ideas other than "actual life"?  Is it not at least arguable that in such moments, Marx was using an old conception (the ideal-material dichotomy) to express an emerging one (the different levels of determination in a complex social whole, the relative autonomy and specific effectivity of ideology)?  Elsewhere in The German Ideology Marx says, in what I think is a satirical moment, that "thoughts and ideas acquire an independent existence" only when a certain division of labour creates agents who produce thoughts and ideas independently.  In other words, the ideological conception of thoughts and ideas having an ideal existence is an illusion that arises from the social practices of petty bourgeois producers of ideology.

Approaching this from another point of view, part of the argument hinges on the base-superstructure metaphor and how it is deployed.  It is difficult to miss the way in which, when some theorists talk about the 'material', they refer to the realm of production, productive relations, commodity circulation - indeed, all that which appears as 'the economy'.  The issue that is being addressed in the language of the 'material determination of ideology' is nothing other than the determination of ideology by productive relations, class struggle and all of the various attributes of the social formation apart from its ideologies.  And a very important reason why people don't want to 'reduce' ideology to matter is precisely because of the need to recognise the specific effectivity of ideology, its peculiar forms and instantiations.  

After all, the question of 'reduction' shouldn't really arise otherwise: it is not a question of saying that ideology is 'reducible' to matter; merely that it is itself a material process.  'Reduction' only comes into it because the ontological problem of the status of ideology (as matter, or ideal substance) has been interjected into the epistemological problem posed by the base-superstructure metaphor (of the relationship between economy, politics and ideology in marxist theory).  And if I'm right in suggesting this, then surely the ISAs essay opened up a potentially fruitful terrain of investigation in solving this latter problem, drawing attention to the various mediating levels particular to ideology (actions, practices, rituals, apparatuses), but also opening up the field of the specific institutional preconditions of ideology.  Not only that, but this demarche is conducted within a sophisticated and potentially dynamic account of the determination and overdetermination of ideologies by the whole complex structure which it is articulated to; that is, it allows us to shift the terms of the problem from the 'material determination of ideology' to that of the complex and mediated determination of ideology by the sub-structures with which it is articulated.  I just suggest we consider it - in the name of materialism, say.

But that still leaves us with the other dilemma.  So, we hypothesise that ideology has only a material and no spiritual or ideal existence; that the relationship between ideology and 'politics' or 'the economy' is not that between matter and the ideal.  Doesn't it still seem that, for the masses to be capable of conducting a revolutionary self-emancipation, we need some notion of resistant interiority, some equivalent to Chomsky's (obviously deliberately simplified) notion of an 'instinct for freedom'?  That is, a conception of 'human nature' as an active constituent in the process of subject-formation?  It will not be sufficient to clarify, as Marta Harnecker exhaustively does, what theoretical anti-humanism does and does not entail for the self-emancipation of the working class.  The point is that the problem is raised about how to understand the basis for resistance, including resistance to the dominant ideology.  And, as we have said, that problem wasn't satisfactorily dealt with in the terms of Althusser's original formulations.  

I don't propose, in this post, to articulate the solution that evaded both Althusser and Butler.  I merely want to reiterate that we can't fall back on any conception of 'interiority', which risks reifying socially, ideologically produced divisions (the interior-exterior division is nothing other than the mind-body problem in another idiom anyway).  Still less can we find salvation in 'human nature'.  Insofar as it goes beyond a simple description of biological needs (for food, warmth, sociality, orgasm) or needs derivative of those (eg, the need to appropriate knowledge which meets those primary needs), which by itself would be neither 'interior' nor an adequate explanation for social resistance, and proceeds into a theology of some 'natural' state of unity with our 'species-being', alienated since the Fall, it restores idealism.  What theories of 'human nature' must invariably do is eternize contingent and historically produced relations and situations.  Such solutions are worse than the problem they address.  If a theory of subjects capable of explaining resistance is available, it must be on the terms of the materialist approach to ideology; that is, of ideology's purely material and not spiritual or ideal existence.

What follows, then, is one attempt to navigate beyond this impasse.  I hasten to say that this is intended as the beginning of an argument, since I can't claim to have resolved every tension or lacuna in my position.  I will also say upfront that it tries to solve the problem in a largely formal way with only incidental references, by way of examples, to the kinds of historical data that will stand up the argument presented here.  That is mainly to keep the argument under some sort of control.  Doing the subject historical and historiographical justice would require a book, and I don't know how to write one of those.

There are those on the Left who adopt a version the law-violence dichotomy.   E P Thompson, as I mentioned in a previous post, was one of those.  In a concluding discussion in his book, Whigs and Hunters: Origin of the Black Act, Thompson acknowledges the structural selectivity of law, asserting not merely that it has been used as an 'instrument' by the ruling class, but that its very form has been such that it tends to produce outcomes favourable to the reproduction of class domination.  Nonetheless, he reaches this striking conclusion in defence of the 'rule of law':   

In assessing this, I think we have to take into account the related contexts of Thompson's theoretical commitments and his politics of socialist humanism.  In his famed attack on Althusser, The Poverty of Theory, one of the points he made was that law did not politely keep to one level, but could be found busily constituting and mediating conflicts at every level of society.  He thus rejected the base-superstructure metaphor, and particularly the althusserian version which held that the different levels of this structure were 'relatively autonomous'.  The same argument can be found in Whigs and Hunters, which draws out what is intended to be an implication of this: law is not merely a superstructural imposition 'from above', but a medium through which the oppressed and exploited conducted their struggles.  Thus the ideology of the 'freeborn Englishman' could be estimated an excellent one, the 'notion' of the rule of law and its supposed equity an unqualified human good.  The challenge was to compel the law to live up to its promise, and the ultimate condition for its fulfilment was the abolition of the class systems which perverted and distorted the law.

Politically, this commitment was related to Thompson's rejection of Stalinism.  It requires no great stretch of the imagination to see why a certain radical English liberalism would be a comfort in the face of 'totalitarian' Eastern tyranny: the gap between 'arbitrary power' and 'the rule of law' seems quite obvious when separated by an iron curtain.  (I don't suggest that Stalinism was actually characterised by an absence of the 'rule of law', but I think that's implicit in Thompson's view, just as it is explicit in many marxist accounts of Nazi rule).  Of at least equal import, though, is his political disappointment, his retirement from activism by the late 1960s, and his complete befuddlement and hostility in the face of the second New Left.  Thompson's weary impatience with young marxists bemoaning 'the fuzz' is of a piece with his disdain for May 1968 as "a rich kid's revolutionary farce", his wariness about Third World revolution, trendy Parisian philosophers and all of the concerns of the New Left which struck him as either modish or politically immature.  To this extent, 'Stalinism' could function as a polemical convenience, enfolding within its compass a range of political and theoretical positions which were disagreeable to the English idiom and romantic moralism that he unapologetically deployed.  In this sense, I think Thompson bears out Althusser's claim that the post-1956 turn to humanism among formerly Stalinized marxists represented a 'right-wing' de-Stalinization; the introduction of feeble bourgeois ideologies of 'liberty' and 'humanity' into marxism was but the mood music for a rapprochement with liberalism.  Still, there are stronger grounds on which one might criticise Thompson's account here, chief among which is that his anti-theoreticist and empiricist tendencies, leave him without the theoretical means to fully break with the 'instrumentalist' account of the law that he is criticising.  I'll come back to this.

Another example of a leftist defence of the law, from a left-reformist perspective, would be Ian Taylor's arguments which were influential on the Labour Left in the early 1980s.  In his article 'Against Crime and For Socialism', Taylor sought to define a leftist defence of law in the context of growing social turmoil and dislocation arising from capitalist crisis and Thatcherite austerity.  The idea was that the left had failed to produce a convincing answer to popular anxieties about crime, largely because it was stuck in two irrelevant analytical modes: either revolutionary anti-statism, or social democratic authoritarianism.  Any analysis which began with a monolithic view of the state, as an instrument for class domination or an uncontested terrain dominated by a single class interest, was intellectually simplistic and strategically barren.  Taylor's argument suggested that law and the state were contested terrains, and he championed the 'community' response to crime favoured by the Bennite Left, organizing local campaigns while imposing democratic and accountable structures on the police so far as possible.  The obverse of this 'community' based politics was a reproach to criminality as anti-social behaviour.  Crime, he argued, was an ideological category which worked on real social material to produce the phenomenon of 'underclass' or 'lower class' criminality.  Thus, the left had to challenge the ideology by displacing the real content from its rightist-populist articulations.  The argument that criminality derived from despair, worklessness, inequality and poverty would not do.  Rather, criminal behaviour was really a kind of self-interested enterprise, distributed across classes in different ways, and a product largely of capitalist individualism.  Among the 'lower class' elements, crime arose from the breakdown of solidaristic communities, or from layers that had never been part of a cohesive, 'respectable' working class.  Social order was worth defending against this sort of predation; policing was an absolute necessity; but it had to be conducted on a popular socialist basis which simultaneously deployed policies to rebuild and stabilise afflicted communities.

This was a very common argument on sections of the Left, especially in the context of E P Thompson's intervention.  Stripped of its more radical prescriptions, the basic analysis of crime as anti-social individualism became the basis for a new authoritarianism within the Labour Party, eventually grounding Tony Blair's 'tough on crime, tough on the causes of crime' campaign and New Labour's punitive campaigns against anti-social behaviour.  Of course, this idea of crime as a form of self-interested enterprise was also offered as an explanation for last summer's riots.  It is vapid: capitalist individualism is everywhere, but crime is not.  Nor is crime necessarily conducted on the basis of 'rational self-interest'.  For all the frankly idiotic commentary freaking out about the theft of consumer goods during the riots, one thing that stood out was the frequency with which those involved acted without reference to material self-aggrandisement, often placing themselves in harm's way in order to conduct sometimes successful but ultimately futile fights with the law.  Altruism, group solidarity and a desire for collective adventure and liberation were salient features of the riots.  Part of the problem here is polysemy: that is, it can mean so many things to say, as Taylor does, that law is ideology.  Quite what that means in practice depends on your conception of ideology and its scope.  I would argue that Taylor, influenced by certain Gramscian motifs, treated 'crime' as an ideological category only inasmuch as it was manipulated by the Right, not inasmuch as it was produced by the ideological action of the law itself.  But again, this I will return to.

What I want to stress is that these kinds of arguments, correctly attempting to oppose a simple theory of law as an expression of class domination, miss what is central to law.  It isn't adequate to say that law mediates the actions of the powerful, and that such mediation is an improvement on unmediated power.  There is no such thing as unmediated power: no way that power could not be materialized in a particular medium.  And what is the medium here?  What accounts for the 'lawness' of law?  Wherein resides the specific aspect that makes legality what it is?  I will develop an argument about this, but the first clue is the constant presence of violence.  If law is a scene of contestation, what decides is violence.  If law has an imperative character, what gives it compulsory force is violence.  To the extent that law is operative at all, it is nothing other than "congealed violence".  Law is, in this aspect, nothing other than a particular organization of violence.

An example from history.  In a previous post, I discussed the 'Massive Resistance' mounted by Southern states in the wake of Brown vs. the Board of Education.  A series of Supreme Court rulings, from Smith vs. Allwright to Brown vs. the Board of Education, and then subsequent fast-track endorsement of the federal government's right to impose Voting Rights, challenged the white supremacist system in various of its aspects.  There was no claim here that the Court was interested in immediate reform.  The approach was to phrase such decisions in such a way as to grant maximum leverage to Southern states in the timing, scale and manner in which they de-segregated.

Importantly, much of the relevant scholarship is explicit in stating that the outcome of these arguments was determined less by the juridical arguments, deduction-from-precedent and so on, than by politics. It was the more centralized, 'interventionist' state arising from the 'New Deal', and a judiciary reconfigured by FDR, that made the difference in a very short space of time between a pro-segregation Supreme Court and one that was imposing reform. Later, the Truman administration's amicus curiae on the international ramifications of segregation made all the difference in Brown vs. the Board of Education. Putting a marxist gloss on this, I argued that underlying this development was i) the rise of monopoly capital and inter-regional firms requiring a further centralization and rationalization of the state of a kind that had begun in the 1910s, and having a different relationship to racial formations than plantation or textile capital, ii) the strategic shift in the Democratic Party's base in response to the emergence of an immigrant working class vote in the North, union insurgency and the growth of the African American vote outside of the South, and iii) the delayed emergence of a civil rights movement effectively deploying disruptive power to force the question and lever open divisions within the ruling class.

At any rate, this process - a Second Reconstruction by many accounts - necessitated the re-organization of state violence in the constitution of class relations. This was manifested quite literally in the federalization of national guards, the use of 101st Airborne, and so on, to physically coerce local state authorities into complying and dropping the various city ordinances, state laws, policing and parapolitical operations comprising 'Massive Resistance'. Between the state of Kentucky and the central state, force (not 'sovereignty', 'states' rights', the constitution, etc.) decided.  Law revealed itself to be, in a sense, nothing but congealed violence.

'Equality before the law' is a cardinal principle of liberal-democratic ideology, and it is one that historians like E P Thompson, and sympathetic theorists, have taken very seriously.  Thompson's critique of the inequities of law, of unjust law, took as its standard the principle of legal equity itself: of the civil and political rights of the 'freeborn Englishman'.  A certain type of marxist critique of this notion holds that it is a purely formal equality which conceals or conveys a content of inequality.  Such, as we will see, is the position of Evgeny Pashukanis.  This is not to say that such a critique must deny the efficacy of formal equality - on the contrary, Pashukanis' work is an attempt to comprehend this formal equality as part of the fundamental, cellular structure of law.  But it involves a different emphasis.  It is not that the law poses a standard of equality which it doesn't realise in practice, but rather that the legal relationship between abstract, formally equal individuals is just the form that class domination takes in a capitalist society characterised by generalised commodity exchange.

Against both interpretations, which actually converge in some of their fundamental assumptions, I want to suggest that the law does not give us formal equality.  Agents are assigned a bundle of rights and obligations depending on their location within the relations of production, such that their formal legal position with respect to one another, irrespective of how these rights and obligations are realised or elaborated in the process of jurisprudence, is not one of equality.  At the most simple level, a capitalist is not equal with a worker, even at the 'formal' legal level.  It will surely be objected that this refers to the substantive content of law, not its form.  Such a reproach is related to schools which treat the legal form as a foundational grammar of abstract legal subjects derived from certain aspects of the capitalist mode of production - most notably, as mentioned, the Pashukanis school which treats the legal form as a derivative of or homologous with the commodity form.  But I don't think the 'form' of law is restricted to its most abstract expression, even if one accepts the grammar of the commodity-law homology.  There are different levels of abstraction at which 'form' makes itself known.  Putting it like this entails that, if the form-content dichotomy is to be useful, it must be revised.  In its current articulation, it seems as if form is something settled at the highest level of abstraction, and content is settled at the highest level of concreteness.  Arguably, however, one finds 'form' and a corresponding 'content' from the first abstraction to the last concrete-in-thought.

As importantly, I am inclined to doubt that the commodity form theory explains what it is supposed to.  Let me reiterate the problem briefly.  The question of the 'legalness' of law comes up for a number of reasons.  First, because in order to truly understand the iniquitous 'content' of law it must be stated why that 'content' - class power, racism, patriarchy, etc - takes the 'form' that it does.  Second, because it is not possible to understand what law does in capitalism specifically, and in particular social formations, without understanding what makes it law in the first place.  That is, what is definitive of law across historical epochs and modes of production, and what is specific to this mode of production.  The Pashukanisian argument is that the legal form is the commodity form; that legal relations first appear in the exchange of commodities between formally free and equal subjects. The commodity relation depends on the items for sale being exchanged between formally equal subjects, who each enter the transaction as voluntary participants, as bearers of rights, particularly property rights.  And it is through this relationship that the legal form is "congealed violence".  For in such transactions, inescapably, there is the potential for one party to alienate the property of the other without consent, which must not happen if it is to be a true commodity relationship.  In the real process of such transactions, of course, coercion almost always intervenes in some way and at some level.  Either it structures the contract of exchange itself, or it calls into question the interpretation of the contract, thus producing a process of legal argument that can ultimately only be terminated through superior violence.  So it is that the grammar of abstract, formally equal legal subjects 'congeals' violence and domination, and the main forms of violence are of course class violence.  It is not that law's promise of equality simply can't be realised in the capitalist mode of production, but rather that this formal equality is the language that class domination under capitalism must speak in.

This seems to answer a number of problems.  First of all, it offers an apparently smooth transition between the determinant structure, the commodity form, and the resulting superstructure, law.  Second, it seems to allow us to pinpoint both the historical continuity of law and its discontinuity: the commodity form may explain the legal form in general, but it is only with the generalization of commodity circulation under capitalism that law expands from localised instances of contract to a universal structure.  Third, it seems to show the precise moment at which violence and consent are mutually articulated in the legal form.  It is so elegant a solution, and absolutely consistent with certain lines of Marx's thought.  But still, I don't think it does quite what it seems to.  First, the smoothness of the transition here is only apparent - unless the homology is reduced to sheer coincidence, unless the legal form just is the commodity form.  The latter is, I think, China Mieville's position.  In what appears to be a deft theoretical innovation, he argues that the law is split between the legal form, which is the commodity form and is proper to the base, and the practice of law, which is a superstructural phenomenon.  Thus, the legal form is but a particular juridical iteration of the commodity form, which is activated and materialized at a different (superstructural, political-ideological) level. 

Again, this is an attractive solution, but I will stick to my position.  I think this conception is inseparable from a set of positions on the base-superstructure controversy which are perfectly arguable but which I suggest we reject. First, China, following Pashukanis, rejects the analysis of law as ideology on the grounds that it reduces law to a set of ideas in people's heads.  Yes, there are all sorts of ideological aspects to, and ramifications of, the law; but, per Pashukanis, we must seek the "objective reality" of law "in the outside world", as opposed to merely the "subjective reality".  This desire to find the specific materiality of the law is part of the reason why Pashukanis tries to trace the legal form to the base, into which he subsumes commodity circulation.  As I will argue in a moment, however, this rests on an unnecessarily reductive conception of ideology.  In fact, it is an idealist conception of ideology which is at fault here, not the suggestion that law is ideology.  Althusser's argument that ideology consists of  "material actions inserted into material practices governed by material rituals" strikes me as the beginning of a superior approach.

Second, relatedly, the argument seems to me to rest on the idea that the superstructure stands in a purely external relationship to the base.  Thus, if the legal form just is the commodity form in a juridical iteration, it must belong to the base, forming part of the foundational structure that will then give rise to a superstructure in which the class-determined 'content' of law will be elaborated.  If, however, you don't assume this relationship of exteriority, it makes sense to see the different levels of the structure as being imbricated, mutually articulated.  Ideology, politics, law, etc., are constantly iterating different aspects of productive relations in their own idiom, constantly intervening, constantly constituting and being constituted by those relations.  And I would say that the fact that there is a juridical iteration indicates that it is already a superstructural phenomenon.  In itself, accepting this way of looking at it doesn't have to pose a problem for the Pashukanis school.  It is quite possible for this juridical iteration of the commodity form to be a superstructural form and still comprise the cellular structure of law.  However, this does allow us to displace the problem.  With Pashukanis, we're trying to locate the materiality of the law in a property of the infrastructure.  Instead, we can now think about ways in which the mutual articulation of infrastructural and superstructural levels determines both the form and content of law.  And that means, we can allow that the form of law is determined by more than one property of the infrastructure, by more than commodity circulation - we can introduce, at different levels of abstraction, productive relations, the social and technical division of labour, the geo-economic unity of a social formation, class struggle and so on.

Moreover, it enables us to think the continuity and discontinuity of law in a far more specific way.  In the Pashukanisian scheme, the emergence and persistence of law is coterminous with the emergence and persistence of markets and commodity trade.  What persists is the basic legal form. Similarly, the change taking place between pre-capitalist and capitalist legality is seen mainly at a quantitative level: ie, in a shift from localised instantiations of the legal form to its universalization.  I think something more than a quantitative shift takes place.  Even at the level of the commodity form itself, there is a qualitative transformation.  In precapitalist, 'simple' circulation, the relationship of commodities to the social division of labour, and thus the commodity relation itself, is completely different.  Many of the inputs of existing commodities have not themselves been commodified; labour has not been generally waged.  The transaction is much more directly determined by political force, or rather by traditional rules, by notions of a 'fair price' arising from religious and political intervention appropriate to feudal social relations, and so it is not necessarily an exchange of equivalents.  In fact, unequal exchange in buying and selling must be considered the norm in pre-capitalist commodity circulation.  The relation of equivalency only seems to operate where the law of value operates; that is, where exchange value understood as the amount of socially necessary labour time embodied in a commodity, is operative.  That is, the commodity form only assumes the form of the exchange of equivalents, requiring the abstract legal subjects which we are invited to see as the universal grammar of law, under capitalism.  It is not that certain formal aspects of the exchange - the act of production for exchange, the use of an exchange medium, the juridical iteration of the exchange relationship - don't stay the same.  It is that the formal aspect which is supposed to ground the legal form changes entirely.

This makes it, I maintain, the wrong axis of continuity.  I maintain instead that we must have an account of law that can incorporate pre-capitalist legal forms: theological jurisprudence such as Shari'a, law that is derivative of religious ritual such as Roman law, different forms of customary law, the Hammurabi Code, etc etc.  That is, an account of law that can comprehend forms which are not predicated on an abstract formally equal legal subjects, whose subjects are explicitly hierarchically differentiated, and which are associated with imposition (thesmos) as much as contract (nomos).  It must allow us to encompass the dense networks of legal relations which in precapitalist as in capitalist societies extend well beyond the sphere of commodity circulation and are decidely not equitable relations.  Such an approach will of necessity not permit the conclusion that bourgeois standards of equality are embedded in the legal form itself at its most abstract level, in its 'timeless structures'.  The legal relation is only infrequently in history one between equal subjects or between equal rights, and then only in a certain abstract manner relating to certain social classifications (citizenship).  Far more pervasively, it is a relation between unequal subjects - not just unequal individuals but corporate subjects, states, and so on - unequally endowed with rights and obligations. 

The solution as I see it is to return to the concept of law as ideology; that is, as a particular element in a mode of production that is proper to the ideological superstructure.  Since the different levels and elements are always-already mutually articulated, their specific form and content (including the appearance of the 'relative independence', as per Franz Neumann, of the legal order) will depend on their articulation in a mode of production and the site of its realization, the social formation.  But that still leaves wide open the question of what kind of ideology it is, and why it takes form it does.  I have cited Althusser's claim that ideology comprises "material actions inserted into material practices governed by material rituals".  If one thinks of the actions of the police officer, the practices of the police bureaucracy, the rituals of the justice system (naturally, my mind gravitates toward HUAC, SISS, and other instances of ritualised repression, but arguably ritual governs the whole field), the action of the law does seem to conform to this characterization.  Of course, for Althusser, the materiality of ideology was inseparably coextensive with the materiality of the state: hence, ideological-state apparatuses.  So, it would seem appropriate to talk about the state.  To be frank, we cannot talk about law any longer without somehow referring it back to the state's constitutive action.

There is a misplaced concreteness that is invested in the state (including by Pashukanis), in which its materiality inheres in a machinery or instrument of domination.  But this spurious institutional determinacy, based on the 'public' kernel of prisons, armed men, etc., will not hold. The state spreads itself over a much wider field than this, and constitutes areas of life that are supposedly 'private' (in part through the very public-private binary that it sustains). The materiality of the state resides in the relations of forces that it condenses in its institutional sites of power and, through its practices, its action at various levels of the social formation.  Part of this action takes the form of symbolic production, moral regulation, the production of social classifications and, as a corollary, the punishment of symbolic and moral trespass, the disruption of its classifications, etc.  This means that violence is immanent to its action, whether or not violence is immediately deployed.  And it is in this sense that I argue that law occupies the position of an ideological relation within the state's strategic field.  It is in this sense that legality appears as both at one and the same time the dominant mode of legitimacy, of consensus, and the mode in which terror and repression is constantly brought to bear in shaping the social body.   

This brings me to my first attempt to say something definitive about law: Law is the dominant form of the dominant ideology concretized in practices which are permanently inscribed with political violence.  In saying this, I do not mean that the dominant ideology or the legitimacy of the law is simply uncontested.  Nor do I mean that the dominant ideology is just the ideology of a single class, and that therefore law is nothing but the execution of the 'will' or interests of that class.  Rather, it is through contestation and class struggle that the dominant ideology reproduces itself, not as a homogenous ideal substance but as a necessarily heterogenous formation comprising elements of popular ideology which have been absorbed and their oppositional character neutralised.  Poulantzas, strongly influenced by Gramsci, incorporated this into his theoretical architecture more successfully than Althusser.  To this extent, following Poulantzas, we can say that law is a juridical condensation of the relations of forces in the social formation that it rules.

But there is a peculiar characteristic of law as ideology, which is its imperative, axiological-deductive character.  Whether the law's application is 'deduced' from a civil code, a constitution, or from a body of 'common law', it takes the form of a chain of juridical logic, the unfolding of a set of determinations from a body of axioms and norms.  This is nothing other than a rationalized, predictable structure of domination and decision-making within the field of the state.  This rationalization is something that emerges from the double freeing of workers from the means of production, the social division of labour arising from this and the operation of the law of value as a homogenizing dynamic.  Its imperative aspect is accounted for by the immanence of political violence to its actualization.  Law in this sense is a mode of disciplinary power, an articulation of the dominant ideology (within a capitalist mode of production) with the political violence which secures its dominance. 

This brings me back to the question of formal equality.  I have said that law is usually not a relation between equal rights.  But the appearance of formal equality is not a bagatelle, not an optical illusion.  Rather, it does appear at different times in the sphere of legality, in certain forms of legal subjectivity.  Citizens are formally equal, even if capitalists and workers are not.  But it seems to me that this arises from the action of the political first and the legal second.  In terms of the formal equality prevalent in bourgeois democracy, it arises from a type of individualization created by the capitalist state.  The Pashukanis school argues that formal equality between empirically different individuals is strictly homologous with the formal equivalence between qualitatively different commodities.  In effect, (meaning, tendentiously put), the formal equality of individuals in law is a kind of exchange value of the subject.  But I think this takes for granted the individualization that it describes, and which forms the basis for the legal relation.

Either one must assume that the individuals thus mentioned are understood as 'biological' individuals (which I certainly think is implicit in Pashukanis, but is problematic because the existence of such individuals can in no way be just assumed), or the process of individualization must be assumed to have taken place through the commodity relationship itself (which I think is problematic because the relation between traders doesn't have to be a relation between individuals), or it must be something that emerges from the freeing of productive agents from territorial-personal bonds and the rupturing of the chains of signification cementing agents within that bondage. If it is the latter, as Poulantzas argues (I think convincingly), then this crucial link in the grammar of law is established by means of an effect of isolation which has roots in the atomization of productive agents in the capitalist division of labour, and is iterated at a juridico-political level by the capitalist state.  Thus, a capitalist type of individuality is produced.  And in the sense that individual subjects thus produced, juridico-political citizens, are all equally subject to the rule of law, formal equality prevails.  This is also simultaneously a totalising mechanism, inasmuch as these are subjects to the extent that they are incorporated within a national body, within the rule of law as such.

With this capitalist division of labour, the double freeing of labour from means of production, the replacement of tribute with the homogenisation of labour processes into abstract labour-time, comes a simultaneous granulation of social space into an imperfectly homogenised grid of locations (sites of production and consumption, as well as extended reproduction) in contrast with land-as-sovereign-body. The organization of production for surplus value entails rationalization, the bureaucratic standardization of units of space-time, the individualization and isolation of agents as competitive producers, the administration of production relations and thus of class antagonisms along scientific-rational and predictable lines. It entails the organization of political violence according to a system of rationalized axioms.  I think this accounts for the main chracteristics of law in its capitalist form.

This brings me back to the arguments from Thompson and Taylor which I was disputing at the start.  For they both correctly argue that the law is not just an instrument, but is also a field of contestation.  Yet they conclude from this that the law has a legitimate function which is the mediation of power, the suppression of violence, the containment of anti-social actions and so on.  What this misses is that law has this role only inasmuch as it is a mode of class domination: it only mediates class power inasmuch as it reproduces class power through its action; it only suppresses violence inasmuch as it is the organization of violence monopolized by the capitalist state; it only contains anti-social action in as much as the state defines, through the law, what is anti-social.  I said Thompson and Taylor don't fully break with the instrumentalist account of law and the state that they seem to reject, and this is true to the extent that their evaluations imply that the same apparatus can be inhabited and put to use by one or other class in support of any political objective; that any seeming structural selectivity is due to a misuse of the instrument which can be put right by proper use.  They also both greatly over-estimate and misunderstand the role of formal equity in the law.  It is not a promise that law makes, but which goes unrealised due to the distorting role of class power.  Rather, it is a juridico-political form which really does exist and is realised within the capitalist mode of production precisely as a mode in which the political and ideological dominance of the ruling class is secured.  That it is also the form in which this domination is provisionally and partially contested does not change this fundamental relationship.  To do away with class power is not to free law from its shackles and distortions, but to do away with law as such.

This is part II of the long delayed Poulantzas series, this time on the problem of the capitalist state.  Poulantzas made several distinctive, ground-breaking contributions to state theory.  Or, I should say, to capitalist state theory since in his view a generic theory of the state was impossible.  One can derive some "general theoretical propositions" about the state from the study of its types, but they "can never be anything other than applied theoretical-strategic notions".   The two major works of his dealing with the capitalist state are Political Power and Social Classes (1968 - hereafter PPSC), written within an althusserian problematic, and State, Power, Socialism (1978 - SPS), which advances a relational view of the state and dispenses with some of the earlier althusserian themes.

I will disappoint some people by not immediately treating in detail the politics of each theoretical phase, but I do intend to return to this in a later post.  Suffice to say for now that the two major works cover a shift from 'Marxism-Leninism' of a more or less critical variety (PPSC, finished days before the occupation of the Sorbonne and the beginning of the May 1968 uprising in France) to a left variant of Eurocommunism (SPS, published during a crisis of marxism, especially of althusserian marxism, and containing passages aimed at the nouveau philosophes).  Strategically, and with regard to the capitalist state, this involved a shift from a nominally revolutionary approach to a 'centrist' approach - centrism, in the terminology of the Third International, being a position suspended between reform and revolution.

In each case, Poulantzas was arguing for a strategy commensurate with the politics of the communist formation (the Greek Communist Party of the Interior - KKE-I) that he was a member of.  This breakaway from the Greek Communist Party (KKE) was active in the resistance to the colonels from 1968-1974, and represented the non-Stalinist wing of the party.   It was initially one of the more left-wing communist parties, but moved to the right throughout the 1970s.  Confounding expectations, this did not improve its standing among Greek voters.  In the 1977 elections, it was the party advocating hardline Stalinism (the KKE) that reaped the lion's share of the communist vote, while the KKE-I's modernising Eurocommunist position received a derisory vote.  The KKE-I was famed among the intelligentsia, but never broke out of its ghetto of less than 3% of the vote, with membership in the region of 12-14,000 in contrast to the KKE's votes of 9-11%, and membership of between 100-120,000. The fact that Poulantzas' major Eurocommunist text followed the 1977 result suggests that even if he had been aware of the historic failure awaiting the Eurocommunist project, he would have continued in the same direction as he saw no future in orthodox alignments, and expected Stalinism be superseded by some form of 'democratic socialism'.

Before delving into Poulantzas' theoretical innovations, I must make a note on his method.  As he said in his critique of Miliband, any historical materialist approach to the capitalist state must clearly state its epistemological criteria in order to properly situate the concrete historical data it works with.  Absent this, it becomes an exercise in empiricism.  His own works, particularly PPSC, are to a very large extent concerned with outlining these protocols.   His approach, as such, has been taxed with the stigma of 'formalism' and (pace Miliband) 'hyper-abstractionism'.  The burden of this criticism is that Poulantzas spent more time parsing texts from the marxist canon and arguing through their implications, than examining concrete state formations.  This is not entirely unfair, and to the extent that it is true, Poulantzas was being typically althusserian: a close, symptomatic scrutiny of texts being the modus operandi of the Althusser Circle.  But the point is overstated.  The survey of the typologies of the capitalist state in PPSC, for instance, largely draws on current sociological and historical research.  The argument about the ambiguous role of state personnel in SPS draws from the immediate experience of May 1968 in France.  Moreover, there is something praiseworthy in Poulantzas' re-evaluation of first principles, the painstaking clarification of concepts.  Though this responded to concrete political problems, usually crises - of Greek communism, of democracy, of marxism, etc - his response was far from intellectually defensive.  He took theoretical risks in order to make marxism adequate to the present.  Only by doing so is it possible to make any sort of progress.

In PPSC, Poulantzas' approach to the capitalist state was, as I have suggested, conducted within the problematic of althusserian marxism.  That is, he sought to understand the state in terms of the specific role of the political 'instance' or level within the capitalist mode of production.  Recall that for Althusser, the mode of production is a 'structure of structures', an articulation of political, ideological and economic levels in which the economic level indirectly determines the content of the political and ideological levels 'in the last instance'.  At the same time, the political and ideological levels intervene in the economic in an 'overdetermining' fashion - that is, the effects and 'contradictions' that accumulate at one level of the structure are condensed in each point of the whole.  (I hope this explanation makes some sense - a lot is being omitted here.)  For Poulantzas, therefore, to understand the capitalist state was to understand: i) the role of the political instance in the capitalist mode of production; ii) the specific way in which the political intervenes in the economic, and is determined by the economic in the last instance, and iii) the relationship of the state to the field of class relations, and thus class practices.

Under capitalism, the political has a certain 'relative autonomy' from the economic and ideological levels.  (Please bear in mind in what follows that the term 'relative' is as important as the term 'autonomy'.)  One way of arguing this might be to claim that capitalism is distinguished by an extrusion of politics from direct relations of production and surplus extraction.  Whereas under feudalism, the levels are 'mixed', with those appropriating surplus labour also wielding direct political power, they are separated out under capitalism.  But Poulantzas rejects this.  Rather, his analysis rests on the so-called 'effect of isolation'.  That is, under capitalism the labour process is subject to both collectivization and separation.  On the one hand, labour processes are carried out in a more dependent, cooperative manner than ever before; on the other hand, they are within certain limits carried out independently of one another, in a competitive fashion, "without the producers having to organize their cooperation to begin with".  At the level of politics, this results in the setting up of agents in the productive process as 'individuals/subjects'.  This is not merely an ideology but a real juridical relation, which intervenes in and structures the productive process so that agents actually experience socio-economic relations as fragmented and atomised processes.  The 'effect of isolation' is thus "terrifyingly real".  It "has a name: competition", and it affects not just direct productive relations but "the whole ensemble of socio-economic relations".

The capitalist state in this sense appears as the "strictly political unity" of these relations.  "It presents itself as the representative of the 'general interest' of competing and divergent economic interests", whose class character is concealed precisely by the isolation effect.  The state thus systematically conceals its own political class character, representing itself as a popular-national state, with "the people-nation" "institutionally fixed as the ensemble of 'citizens' or 'individuals' whose unity is represented by the capitalist state".  The effect of isolation is the "real substratum" of this state.  But it is precisely in "putting itself forward as the representative of the unity of the people-nation" that the state assumes this relative autonomy with respect to class relations.

Another way to put this is that if the effect of isolation on economic struggles tends to impede class unity, resulting in sectional struggles, it is at the level of political practice that this unity must be created.  Thus, the political class struggle operates in a relatively autonomous fashion with respect to economic class struggles.  The capitalist state has to be seen in light of the political practice of the dominant classes, whose purpose is to produce class unity out of the isolation of their economic struggles, and at the same time constitute their political interests as the "general interest of the people/nation".  The relative autonomy of the capitalist state enables it to better organise the unity of the dominant classes, and to represent their interests as those of the society as a whole: this, the organisation of the dominant classes, and the disorganisation of the dominated classes, is the primary political function of the state.  It is the indispensable 'factor in unity', without which the bourgeoisie's political dominance is unthinkable.  This leads us to the question of hegemony, and the 'power bloc'.

Alongside Althusser, Gramsci is one of the major influences in Poulantzas' thought.  Even where Poulantzas felt compelled to upbraid Gramsci's 'historicism' in his earlier work, a tendency which Peter Thomas notes is "essentially discontinuous with or rhetorically external to his concrete analyses of Gramsci's theses", the trend is toward a growing articulation of his research project with that of Gramsci.  Concessions to althusserian fashion obscured this.  (In fairness to Althusser, his own later writing on Gramsci was far less schematic, and far less driven by the dismissive typologies of his earlier work.)  In PPSC, he takes over the concept of 'hegemony' and seeks to develop it with specific reference to its role in the political dominance of the ruling classes.

In Gramsci, hegemony has several senses.  In one sense, it refers to the hegemony of the proletariat within a wider anticapitalist class alliance incorporating peasants.  To this extent, the concept is continuous with its useage in the Russian context.  In another sense, it refers to a particular state of ruling class dominance.  In this perspective, hegemony is a brief historical moment, which has to be constantly worked on and constructed, in which the ruling class does not merely rule, but actually leads politically and ideologically.  In such moments, the bourgeoisie, or a fraction thereof, sets itself up as the leading class/fraction in a world-historic mission, and uses a combination of repressive, ideological and material means to incorporate subordinate classes and fractions into a system of class alliances supporting this mission.  But aside from these exceptional moments, one can also speak of hegemonic political practices - practices through which a dominant class or fraction aspires to hegemony.  This is the sense in which Stuart Hall argues that the coalition government is pursuing a hegemonic project, attempting to fundamentally alter the popular 'common sense' in a reactionary direction.

Poulantzas is at this stage solely interested in developing the concept of hegemony in so far as it accounts for "the political practices of dominant classes in developed capitalist formations".  The concept of hegemony is thus used in two senses.  First, it indicates the relation of the dominant classes of a capitalist formation to the state, and the constitution of their interests as the 'general interest'.  This reinforces the concept of the state as the factor in unity, transposing struggles from a corporate to a universal plane.  Second, it specifies the specific form in which the dominant classes unity is secured: through an alliance of classes and fractions, in which one class or fraction (usually a fraction) is dominant, or hegemonic.  This alliance, Poulantzas calls the 'power bloc'.

The need for a power bloc derives from the nature of capitalist production relations, which ensures that the ruling class is "constitutively divided into fractions" (financial, commercial, industrial, rentier, and so on).  The isolation effect, moreover, is not compensated for by any other factor - such as the factor of 'collective labour' in the working class.  This means that the dominant fractions and classes are incapable of raising themselves to the hegemonic level through their own parties: they need some other basis for unity. The power bloc comprises a "contradictory unity of dominant classes or fractions" under the leadership of a hegemonic class or fraction.  But the relation of this bloc to the state is not one of a 'sharing out' of power among the fractions.  "In the last analysis," says Poulantzas, "it is always the hegemonic class or fraction which appears to hold state power in its unity". As such, it is the hegemonic class or fraction that assures the unity of the power bloc and acts as its protector.

Returning to the argument about the relative autonomy of the state ‘machine’ from class relations, it seems here that the 'Caesarist' tendencies discussed in Marx's Eighteenth Brumaire are immanent to the capitalist type of state. Far from the state being dependent on any one of the fractions of the dominant classes, far from it securing its unity from an already unified hegemonic class or fraction, it is "the factor of the political unity of the power bloc under the protection of the hegemonic class or fraction. In other words, it is the factor of hegemonic organisation of this class or fraction". The state does not arbitrate between already constituted social forces. Rather: "Everything happens precisely as if the state permanently played the role of political organizer of the power bloc".

Poulantzas goes on to argue that the play of institutions within the state apparatus is directly related to the relations of power within the power bloc.  Though it functions as a "centralised unity", it has a set of formal separations - between legislative, judicial and executive power.  Setting aside the judicial branch, the distinction between legislative and executive power is here treated as a power relation and not merely a juridical separation: "it corresponds both to the precise relations of political forces and to real differences in the functioning of state institutions". Depending on the state in question, one of the branches always dominates, usually either the executive or legislative branch, and thus constitutes the nodal point where unitary institutionalized power is concentrated within the state organization.  The formal separation of powers reflects an internal index of subordination, inasmuch as the hegemonic class or fraction controls the dominant branch of the state.  Here, Poulantzas is drawing on Althusser's reading of Montesquieu, who coined the doctrine of the separation of powers.  In this reading, the relations between executive and legislative branch (separated into lower and upper chambers) of the French state immediately following the revolution, relates to a certain conception of the relations between social forces. The royalty controlled the executive, the nobility the upper legislature, and the ‘people’/bourgeoisie the lower legislature.  The interplay between these institutions reflected a struggle for power among these dominant classes, with the less powerful branches playing the role of allowing certain resistances on the part of subordinate fractions within the power bloc: but the centralised unity of the state remains, and power, far from being actually separated out or distributed, continues to be concentrated in the dominant branch.* 

Thus far we have encountered the capitalist state as a relatively autonomous force; a class state in a 'popular-national' form, organising the hegemonic struggles of the dominant classes; and a centralised unity acting as the factor in the unity of the power bloc, and by extension the disunity of those excluded from power.  This approach has been taxed with functionalism, and this is not the only place where a functionalist problematic can be detected in the formulations used by Poulantzas.  To describe the state as the 'factor in unity' of the dominant classes implies a degree of internal unity and consistency that would make destabilisation and disintegration hard to imagine.  But we don't have to read it in that way.  It's possible to see this 'function' of the state as, if you like, a necessary condition for bourgeois rule, which may or may not be adequately fulfilled at any given moment.  The only way to redeem the insight, though, would be to separate from the functionalist problematic and incorporate it into a new epistemological framework.

In his later work, SPS, Poulantzas made several adjustments along these lines.  In place of the focus on the regional autonomy of the political, Poulantzas came to argue that "political-ideological relations are already present in the actual constitution of the relations of production".  Therefore, the position of the capitalist state vis-a-vis the economy was not to be resolved by declaring its 'relative autonomy', but rather by showing that this position was just "the modality of the State's presence in the constitution and reproduction of the relations of production".  Poulantzas did not deny the relative separation of economic and political regions, but rather laid a different emphasis, stressing their "mutual relation and articulation - a process that is effected in each mode of production through the determining role of the relations of production".  This mutual relation and articulation, incidentally, explains why there can be no general theory of the state.

So, rather than start from a 'regional' analysis of different 'instances' in the capitalist mode of production, he re-energised his whole approach with a 'relational' analysis of the state as a strategic field brought into existence by the intersection of ruling class power networks.  In breaking with Althusser's "legalist image" of the state as a sovereign legal subject guarding the perimeters of economic sphere that otherwise reproduced itself independently, he held that the state was a set of relations that actively constituted and reproduced the economic sphere.  Far from being a juridico-political organisation standing over the economy, it concentrated within itself the political and ideological relations already present in the relations of production; it incarnated those relations, inscribing them (thus, the political and ideological dominance of the ruling class) in the "institutional materiality" of the state itself.

The "strategic field" of the state, in Poulantzas' terms, is defined quite broadly.  While Foucault and Deleuze charged marxists with ignoring the political power relations in institutions beyond the state, such as asylums, hospitals, sporting apparatuses etc., for Poulantzas these were "included within the strategic field of the state".  This is not to say that these were constituted as sites of power by the state: power in the marxist sense goes well beyond the state even in the broad sense understood here.  It is to say that these sites of power "do not stand in an external relationship to the state", which increasingly penetrates every sphere of social reality, "dissolving thereby the traditionally 'private' texture".  This understanding of the "strategic field" brings into focus one of the problems for the analysis of state forms, that of 'parapolitics'.  Take, for example, the Ku Klux Klan organisations of the 1950s.  These were not bodies with an explicit, codified relationship to any public authority.  Yet, their illicit hierarchies and relations (with governors, police commissioners etc), their protection of explicit hierarchies through the administration of racial violence, and their relation to the political class struggles of the Southern ruling classes, all place them firmly in the "strategic field" of Southern state forms.  They were partially, but not wholly, constituted as political powers by the state.  They did not occupy privileged sites of political power, but power was delegated to them by those who did occupy them.  This ambiguous position does not only manifest itself in the case of covert political violence.  One of the ways in which neoliberal statecraft manifests itself, for example, is the proliferation of so-called 'quangos' which perform state-like functions within a remit defined by the state.  There is a whole ensemble of institutions, stretching out well beyond the public kernel of policemen, bureaucrats, armies etc which are not understood to be part of the state but which nonetheless fall into its strategic field.  And beyond that, there is no social reality that does not in some way constitute itself in relation to the state.

Without further spelling out what forces are at play in this "strategic field", however, the phrase risks becoming a mere incantation.  The major forces at work in any society are class forces.  The positioning of these forces in the strategic field of political power depends on the relations of production, and the social division of labour that emerges from it.  For Poulantzas, the latter mainly manifests itself in the form of a division between mental and manual labour.  The state constantly re-constitutes this division, through the education system and by other means, and is itself the distinctive embodiment of intellectual labour.  By reproducing this division, moreover, it deprives the popular classes of the intellectual skills necessary to penetrate its bureaucratic discourses.  This case is simply unconvincing in its original form, and leads to unsustainable conclusions about the formation of classes.  (See the previous post on Poulantzas' thinking about the division of labour and classes in contemporary capitalism).  Given the proletarianisation of occupations that involve intellectual labour, I would suggest that we might better think of the division as one between executive/managerial and menial/subordinate labour.  With that adjustment, we can then return to the relationship between the state and the dominant classes.

In SPS, Poulantzas held to his previous argument regarding the primary political role of the state, viz. the organization of the dominant classes, and disorganization of the dominated classes.  It does this by unifying a power bloc politically, while linking fractions of the dominated classes to the power bloc in various ways so that they are unavailable for counter-hegemonic struggles.  But his new methodological approach required a different understanding of this role.  For, if the power relations that were condensed in the state were primarily class relations, it followed that the strategic field of the state must be traversed by class struggles.  Rather than merely allowing for resistances by fractions within the power bloc, he laid a great deal more emphasis on strategies pursued by dominated classes either within the state, or impacting on the state.  He allowed for beach-heads of resistance on the part of popular classes within different layers of the state.  These were by no means equivalent to the centres of power within the state occupied by the dominant classes: this would imply a permanent state of dual power within the capitalist state itself.  But the strategic calculations of the latter could be modified by the struggle of popular classes.  We might add that the divisions mentioned earlier, between menial and executive labour, are reproduced within the state apparatus.  (It would be difficult to understand the public sector strikes otherwise - unless, like Zizek, you maintain that they represent the revolt of a salaried bourgeoisie struggling for privileges and a share of rent extracted from the proletariat.  In which case, you're easily gulled.)  So the state is riven with class struggles.  But it is also exceeded by them.  For though it attempts to incorporate class relations on terms favourable to the power bloc, because these relations are characterised by struggle they always exceed the capacity of apparatuses to incarnate them.  This is certainly some distance from a 'functionalist' treatment of the state, and it helps us to understand more precisely certain aspects of our own situation.  After all, one of the weaknesses (far from the major one) of the anti-cuts movement in the UK is the weakness of political representation of the working class, the absence of footholds in the state at most levels, including the lowest council chamber.  We cannot be indifferent to the fact that only Caroline Lucas and a few Labour lefts even try to conduct such representation in the commons.  We need only look at the Linke to see the difference that such representation, such footholds, can make to class struggles outside of parliament.  At the same time, such an understanding does not, to my mind, lend itself to the substitution of parliamentary struggles for all others.

Poulantzas also refined his thinking on the state's role in the production of hegemony, arguing that the distinction between repressive and ideological state apparatuses in Althusser was misleading (we know from our own experience that repressive institutions such as the police and courts have a strong ideological role).  In addition, I have said that the linking of different subaltern fractions to the power bloc is a role of the state, but in SPS Poulantzas clarifies that this is not only a political or ideological operation: the state must constantly produce a material substratum for mass consent: "even fascism was obliged to undertake a series of positive measures, such as absorption of unemployment, protection and sometimes improvement of the real purchasing power of certain sections of the popular masses, and the introduction of so-called social legislation. (Of course, this did not exclude increased exploitation through a rise in relative surplus-value - quite the contrary.)"  Again, this is an advance on PPSC, and in those accounts which overestimate the power of the 'ideological state apparatuses'.

One of the problems that remains was hinted at by Goran Therborn.  Therborn, writing before the publication of SPS, contended that for all of Poulantzas' innovations in state theory, he paid remarkably little heed to the internal organisation of the state, and specifically the state apparatus.  But with SPS and its reflections on the "institutional materiality" of the state, its argument that the institutions-apparatuses of the state concretise the relations of political-ideological dominance in the wider society, we can no longer substantiate that claim.  Yet there remains an aporia, as far as I can tell: that is, the implications of the state's internal organisation for political strategy are drawn out incorrectly.  So, for example, when Poulantzas writes on the possibility of a transition to democratic socialism, he focuses on the differing class locations of state personnel.  In normal situations, the state is so organised that a general 'line' will emerge from the interplay of strategies and tactics of the dominant classes with the institutions of the state itself, and that 'line' will successfully be imposed on dissident and antagonistic elements within the state.  But during a crisis, he argues, of a scale like that which shook France in May 1968, the diverging class positions will result in a fracturing of the state personnel which, if sensitively handled, can help effect the transition.  I think this places far too much weight on the strategic significance of such divisions, and doesn't follow through on the correct (to my mind) understanding that he has earlier developed, the implication of which is that the dominant classes would continue to command the most strategically important positions within the centralised unity that is the state.  

Now, this doesn't result in a straightforwardly reformist position on Poulantzas' part. He still maintains that the working class must build structures of rank-and-file self-government to challenge liberal democratic forms of representation.  But this is as much to apply pressure to the capitalist state as to develop alternative, socialist forms of democracy.  The strategic perspective that follows from this mediates between reform and revolution.  Perhaps it says something that the only place where something like this strategy has been implemented and yielded some gains - not socialism, of course -  is the highly exceptional case of Venezuela where the struggle of the popular classes really has traversed the state right to the top with no serious reversal as yet in sight.  (Poulantzas as a co-author of "21st Century socialism" - anyone?)  But I think that if Poulantzas' superior insights are taken seriously, their logic is revolutionary.