The Poulantzas project 1
Nicos Poulantzas in his final book, 1978:
The constructivist image of ‘base’ and ‘superstructure’, which is supposed to allow the determining role of the economic sphere to be visualised after a fashion, cannot in fact provide a correct representation of the articulation of social reality, nor therefore of that determining role itself. It has proved to be disastrous in more ways than one, and there is everything to be gained from not relying on it. For my own part, I have long ceased to use it in analysis of the State. [Poulantzas, 1978: State, Power, Socialism, Verso, p. 16]
Nicos Poulantzas in 1964, in the earliest paper that has been translated into English:
When they involve the superstructures, as in the case of art, law and the state, or morality, the most universal-concrete, general-particular – in short, simple-complex – concepts cannot be directly referred to the base: they can only be captured by preliminary research into their historical relations with the base. What analysis can set out from is, on the one hand, the specificity of the superstructure in general and its fundamental dialectical division – for it refers to a historically determinant division – from the base; and, on the other, the specificity of a certain law or state, a certain art, a certain morality, situated in time and space. [Poulantzas, 1964: "Marxist examination of the contemporary state and law and the question of the alternative", in James Martin (ed.), 2008, The Poulantzas Reader: Marxism, law and the state, translated by Gregory Elliot, Verso, p. 25]
Raymond Williams once wrote that “any Marxian theory of culture must begin by considering the proposition of a determining base and a determined superstructure”, even though “from a strictly theoretical point of view this is not, in fact, where we might choose to begin.” [“Base and superstructure in Marxist cultural theory”, New Left Review, I/82, 1973: p. 3] Poulantzas seems to have felt similarly obliged in his early work, and a similar discomfort is evident in the shifty, hedging, ill-defined terminology that make it unclear in places (“universal-concrete, general-particular – in short, simple-complex”?!).
The first two of his papers I will deal with – “Marxist examination of the contemporary state and law and the question of the ‘alternative’” (1964) and “Sartre’s Critique of Dialectical Reason and law” (1965, also in the Martin ed. Verso 2008 collection) – are pervaded by the problem of honouring the letter of the base-determines-superstructure metaphor while subverting its spirit. In my view, the centrality of the metaphor within some strands of Marxism is the blowing up of a single paragraph far beyond all proportion in Marx’s work. But, as Raymond Williams argued, because of its importance to the tradition, if you ditch everything that employs it, a lot of baby will go out with the bathwater. In any case, it has been deployed with completely different meanings and emphases: the determination of the superstructure by the base may highlight the division between the two spheres, i.e., their relative autonomy from one another, or it may emphasise the determination, reducing the ‘superstructure’ to a reflection of the base. The metaphor, in itself, does not say very much.
In Poulantzas’s case, from the beginning he uses it to emphasise the autonomy of superstructure – and the law in particular – from the economic ‘base’. The early work can be read, then, in light of his later escape from and rejection of the metaphor, and I will trace this thread through his work as I progress.
In the 1964 essay he leaves the terms undefined, working instead by criticising earlier Marxian accounts of law, and thereby building a negative picture of his own view from which we are left to infer the meaning of his concepts. He begins with Reisner and Vyshinsky (the latter best known as Prosecutor General during Stalin’s purges of the late 1930s and his related jurisprudential principle that ‘confession of the accused is the queen of evidence’). He agrees with their definition of the law as a “conceptual ensemble of behavioural norms and rules” – that is, it expresses ‘what should be’ socially to reinforce the practices that reproduce the base (alongside other superstructural domains such as morality, religion, the state and “even art (albeit in a different sense)” [p. 26]). But he charges them with reading the ‘what should be’ directly from the interests of the capitalist class. Their picture of capitalist law is (at least in Poulantzas’s telling) an uncomplicated instrument by which capital maintains its domination.
I have no problem with his rejection of such a view, but the precise argument is unclear because his own conception of the structures is not spelled out. He argues that Reisner and Vyshinsky misconceive the relationship between “the juridical-state sphere” and the economic base:
Establishing an external, direct relationship, in the indicated sense, between law and the state and class struggle, which (let us not forget) is situated on the level of social relations of production – not on the economic level of forces and modes of production – this conception shuts off access to the economic level. [p. 27]
The criticism of an ‘external’ relationship between law/state and class struggle is well and good, but how does this sit with the apparent drawing of a firm distinction between ‘social relations of production’ and ‘the economic level of forces and modes of production’? The ‘let us not forget’ implies a confidence that readers within his milieu (this was originally published, by the way, in Sartre’s Les Temps Modernes) would know exactly what was meant by the distinction – but it is not clear to me. I am sympathetic to Derek Sayer’s interpretation that ‘forces’ and ‘relations’ of production are different ‘cuts’ from the whole economic structure rather than ontologically separate spheres. [The Violence of Abstraction, 1987] My best guess is that Poulantzas means to distinguish between ownership relations on the one hand, and the market system of production on the other.
In contrast, he argues that the legal superstructure crystallises certain normative values – i.e. of ‘what should be’ – from the practices which make the economic base function in the way that it does. The owners of the means of production benefit from the functioning of that system, simply because of their position within it; further state/legal intervention in their interests is unnecessary to make them the ‘ruling class’, and in fact the crystallisation of the values of the base in laws and state activity may appear to conflict with their interests in particular cases.
Poulantzas then argues that Stuchka and Pashukanis make a different error. By focusing on the shaping of capitalist law by the practices of commodity exchange, they relate law to the base in a more general way, without a presumption that it is an emanation of the interests of the capitalist class. But according to Poulantzas, “they reduce law and the state to the base. They disregard their specific character as a coherent system of norms and thus completely ignore their relative autonomy.” [p. 28] This is tempting, he says, because class relations of production are defined in terms of property ownership, so that law might seem to be immediately part of the base. But he objects to the idea of one-way causation from the base to the legal/state superstructure.
This is the main thrust of what I take to be Poulantzas’s positive argument: that the legal form by which the norms guiding and restricting some aspects of social practice cannot but affect the norms themselves, and thereby practice. This is suggested in the quote at the beginning of this post: “on the one hand, the specificity of the superstructure in general and its fundamental dialectical division – for it refers to a historically determinant division – from the base; and, on the other, the specificity of a certain law or state, a certain art, a certain morality, situated in time and space”. Values arise from social practice, and social practice is in turn guided by norms deriving from values. The norms guiding practice, and thus underpinning a given social structure, may be moral or religious norms, or they may be legal norms – and it is interesting, and makes a difference, that the rise of capitalism was associated with a great expansion and sharpening of the legal sphere – ethical and religious norms alone were incapable on their own to sustain. It makes a difference because the legal form has a logic all its own, which cannot be reduced to determination by other social spheres: e.g., it entails consistency within the system of laws, involves concepts of legal personhood, political soveriegnty, posits judicial institutions etc. (Incidentally, Pashukanis’s General Theory of Law and Marxism  is brilliant on the internal logic of capitalist law and its partial determination by commodity exchange, and I think Poulantzas’s criticism is a little unfair.) The law transforms values simply by giving them juridical form.
The legal sphere is shaped partially ‘externally’ by the values (needs?) arising from the base, and partially ‘internally’ by the legal form. Specifically, Poulantzas sees the structure of contemporary capitalist economic practices as generating the values of liberty, equality, calculability and predictability, while the legal form itself shapes these values to generate norms that are general, abstract, formal and strictly codified.
How exactly the base generates these norms is unclear. Functionalism is implied, but not of a simple kind. In the argument against Vyshinsky and Reisner, he accepts that the law and other parts of the superstructural ‘normative ensemble’ express “what should be socially” – but that this is not necessarily “what the bourgeoisie wants”. Why the values of ‘liberty’ and ‘equality’ rather than ‘obedience’ and ‘stratification’, which are certainly functional for the capitalist economy? It is not set out explicitly, but it seems that class conflict plays a role here in generating compromise values, and Poulantzas argues that the values are capable of being turned against the capitalist social structure – that ‘liberty’ and ‘equality’ can be footholds for socialist strategy in the political sphere:
…in the process of production, the democratic values of real, concrete liberty and equality – this particular expression of the generic socialisation of man – are increasingly experienced as a pressing lack, because they are a lack that can be made good. And if the hegemonic organisation is also a strategic channeling of this ‘material force‘; and if this channelling consists in proposing models to this already existing positivity at the global political level – i.e. historically objectifying it by suggesting objectives that suit it… It is in the concrete positivity of working-class praxis that its positive political realisation must be sought, while taking account of the mediations requried to transpose a democratic economy into a revolutionary political democracy. (p. 45)
(Yes, the style is somewhat tortuous.)
Elsewhere, in a footnote, Poulantzas suggests that it is not so much that the base directly generates the value-content of the ‘normative ensemble’ that guides its practices. Rather, it determines its limits, because dysfunctional norms will be unsustainable:
For Marx… a false ideology or a superstructural system that distorts the base… remains wholly real (real-ideal) can exist ontologically as such, and possesses a historical effectivity, while not being ‘adequate’ - monism of contradiction – to the real-material. Precisely to the extent that they are not adequate, they are axiologically invalid. In effect, although genetically ‘grounded’, as historical values, in the base… they do not, or they do no longer, conform to the true meaning immanent in the real-material. Thus they are not – or are no longer – legitimated and validated by those material realities of the base that structure, at this particular moment, its historical meaning. [p. 415n]
It should be clear that Poulantzas is coming from a highly abstract philosophical position, and it makes for uneasy social science. His academic background up to this point is in jurisprudence. This paper is unsatisfactory in many ways. But it is Poulantzas’s starting point, not the end-point – and the direction in which he is straining against the traditional ‘base-determines-superstructure’ metaphor is the right one.