Marxism and Mediation - Richard Gunn


Richard Gunn writes about the concept of mediations: for Marx, mediations of the contradictions inherent in the commodity-form (the central contradiction is between use-value and exchange-value, and its mediation is money) 'does not abolish these contradictions, but rather provides the form [read: the mode of existence] within which they have room to move. Published 1987 in Common Sense, no. 2, pages 57-66.

Werner Bonefeld also wrote on Marxism and the Concept of Mediation for the same issue. That can be found here.

The full archive of Common Sense can be found at:

Submitted by UseValueNotExc… on November 15, 2023

In both Hegelian and Marxist thought, the concept of mediation figures as a central dialectical category. That the category does theoretical, and revolutionary, work is clear. What is less clear, to myself at any rate, is what might be termed the conceptual geography of the category itself. It is this conceptual geography which, as a preliminary to further discussion, the present paper attempts to clarify. A more pretentious title for what follows might be 'Prolegomena to a Reading of Marx'.

To mediate is to bring about a relation by means of a relating (an “intermediate”) term. A mediation is the relating term itself. To count as a mediation, a relating term must be more than a mere catalyst or external condition (however necessary) of the relation: rather, it must itself be the relation. It must constitute it, in the way that for example – and the example is offered merely heuristically – a rope linking two climbers is constitutive of the relation in which they stand.

If a mediation is, thus, the relation which it establishes, it does not follow that just any relation counts as a mediating term. A mediated relation is distinct from a relation for which, to render it intelligible or accurately describe it, no reference to a relating term need be made – for example, a relation of juxtaposition. A relation of this kind is an immediate relation (which, for its part, may be catalysed or necessitated in this or that way).

Within the conceptual field of mediation, as so far outlined, various possibilities exist. Two (or more) terms may be related (mediated) by means of a third, or further, term; or a single term may be related (mediated) to itself by a second term. Where a single term is mediated to itself, the relation between it and its mediation may or may not be reciprocal. Where it is reciprocal, there exist two terms each of which is the other's mediation, and each of which is mediated by the other to itself. This gives an idea of the internal richness of mediation's conceptual field: either there may exist two (or more) terms plus their mediation; or there may exist a single term plus its mediation; or there may exist two terms each mediating, and mediated by, each other. The first of these three possibilities is, perhaps, the one with which the notion of mediation is most commonly associated. (It is closest, for example, to dictionary definitions of 'to mediate'.) However, the third possibility is quite explicitly invoked by Hegel when he envisages a situation in which each of the two terms 'is for the other the middle [the mediating] term, through which each mediates itself with itself and unites with itself' (Hegel 1977 p. 112). The example he gives is that of a mutually recognitive relation between individual self-conscious subjects.

A further, and all-important, step is taken in exploring the concept of mediation when it is noticed that the process of mediation may be such as to bring about not merely a relation, but an internal relation: it is exclusively such instances of mediation which concern Hegel and Marx. (In the case of a single term which is mediated to itself, the corresponding possibility is that the process of mediation “totalises” discrete aspects into an internally related whole.) Prior to the mediation, that which is mediated may or may not have been internally related (or self-related). But, even supposing that it was, the mediation may establish a fresh internal-relatedness (or a fresh totalisation). If a (fresh) internal-relatedness or totalisation is established by the process of mediation, then the following is the consequence. Since (a) an internal relation is constitutive of the terms which it relates, and since (b) a mediation is itself – as already indicated – the relation of the term(s) concerned, we can say: in such cases, the mediation is the mode of existence of the related term(s). This can also be expressed – Marx and Hegel so express it – by saying that in such cases the mediation is the form or appearance of the term(s) which it internally relates.

Combining this notion of mediation as the mode of existence (form, appearance) of what is mediated with the third possible shape of mediation indicated earlier, a further possibility emerges: two terms may be the mode of existence of one another. And such is indeed the case, for Hegel, with two mutually recognitive self-consciousnesses: in Hegelian usage, the expression 'recognition' carries with it a specifically constitutive force. This being so, it follows that a recognitive relation between individuals in no way requires mediation through a discrete “third term” –
for example social institutions (or in Hegel's term 'spiritual masses') such as state and civil society (Hegel 1977 pp. 300-1) – separate from, and standing over against, the individuals concerned. The Hegel of the Phenomenology is in fact emphatic that the existence of 'spiritual masses' entails alienation, and that mutually recognitive (or non-alienated) social existence is possible only when no spiritual masses or social institutions exist: mutually recognitive self-consciousness 'no longer places its [social] world and its ground outside itself' (Hegel 1977 p. 265). Thus it is that being alive to the various possible shapes of mediation – in particular, the refusal to equate mediation as such with the first of the three possibilities above indicated – allows us to discern what is in effect an anarchist stratum in Hegel's thought. And the emergence of Left Hegelianism out of Hegel becomes intelligible at the same stroke: for example, Marx's 'On the Jewish Question' appears as a restatement of the critique of 'spiritual masses' which the Phenomenology contains. In the Philosophy of Right, by contrast, Hegel reinstates spiritual masses: individuals are seen as mediated to one another via the discrete “third term” of social institutions, most notably and most famously the institutions of civil society and the state. In virtue of this reinstatement, Hegel opens himself to the criticisms delivered by Marx in his Hegel-critique of 1843. The same point may be stated differently: the Hegel of the Phenomenology emerges as the Philosophy of Right's most trenchant critic.

The expressions “form” and “appearance”, introduced earlier, require further elaboration. I should like what I have said to be taken as (in the sense which is relevant here) defining them: the form or appearance of something is its mode of existence. This definitional sense is not, of course, the sense which “form” and “appearance” receive in ordinary language: there, form is understood as opposed to content and appearance is understood as opposed to reality or essence – as though something's form or appearance might be removed or altered without thereby effecting an essential change in the nature (the content or reality) of the “something” itself. In other words, the ordinary-language usage of “form” and of “appearance” is dualistic.

By contrast, their definitional sense (the sense which is relevant so far as mediation is concerned) is non-dualistic. What this involves is made clear by Hegel in his treatment of the relation between appearance and essence. According to Hegel 'essence must appear', i.e, the appearance is the essence's mode of existence: ' not something beyond or behind appearance, but, just because it is the essence which exists, the existence is appearance' (Hegel 1892 papa. 131). The relation between appearance and essence here envisaged is non-dualistic inasmuch as it is in and through its appearance that the essence is. Essence stands out ahead of itself as appearance, and it is as thus standing ahead of itself that it exists: “appearance”, in other words, is to be understood not as a passive noun (an inert veil or cover) but as an “appearing”, i.e., in a sense which alludes to the activity of the verb. This thought is one which Hegel derives from Ancient philosophy. For Anaxagoras, similarly to Hegel and in contradistinction to Parmenides' dualistic counterposing of appearance to reality, 'Appearances are a glimpse of the obscure' (Kirk and Raven 1963 p. 394). Anaxagoras's saying is not to be understood as affirming that appearances comprise, so to say, a thin rather than a thick veil. Rather, his thought is that it is in the nature of what is not appearance – namely, being – to reveal/conceal itself or, in other words, to appear in the sense of standing (obscurely) forth. And, in fact, Marx's concepts of fetishism and of mystification register, so far as social being is concerned, an exactly parallel point.

I have dwelt on the non-dualistic meaning of the term “appearance” because this meaning is decisive for how Marx's Grundrisse and Capital are to be read. Famously, Marx speaks of penetrating through appearances to reality and urges that capitalist society appears to those who live in it in systematically misleading ways (e.g. Marx 1973 pp. 247, 674; 1976 p. 421; 1966 p. 817). Such passages are misunderstood if they are read – and of course they have been so read – as counterposing appearance to reality in a dualistic fashion, or as affirming that appearance is less real than the reality it fetishistically reveals/conceals. From the Grundrisse, it is clear enough that capitalism's appearance in terms of freedom, equality, property, etc. is a real moment in capitalist production relations taken as a whole. Marx drives this point home when he contends that social relations which appear as 'material relations between persons and social relations between things' appear 'as what they are' (Marx 1976 p. 166): this passage is unintelligible – it must seem as though Marx is endorsing a fetishised perspective – unless appearance is understood as the mediation (the mode of existence) of the relation in which the producers of commodities stand.1
If, despite all this, a dualistic understanding of the appearance/reality relation is forced upon Marx then the consequence is either determinism (reality is seen as causally conditioning an appearance which is distinct from it) or reductionism (not the appearance, but only the reality, is supposed finally to exist). Once appearances are understood as mediations no such consequences are entailed. Regarding fetishism and mystification, Marx's point is not that we can be mystified about reality, or even that we can be misled by reality, but that mystification – or “enchantment” – is the mode in which capitalist reality exists. So to say, capitalism exists as its own self-denial.

It may seem as though such a view inscribes mystification so deeply in capitalist social reality that the emergence, from capitalism, of revolutionary theory and practice becomes all but impossible. But precisely the opposite is the consequence if, as we shall see, capitalist appearances are modes of existence of relationships which are antagonistic through and through. It is the non-dualism of the appearance/reality relation which allows antagonisms to be matters of experience – to be 'glimpsed', in Anaxagoras's meaning – in however self-contradictory and distorted a way. Once appearance is dualistically severed from reality, by contrast, antagonism is placed outwith the domain of experience and the basis for a politics of revolutionary self-emancipation is undermined.

As with “appearance”, so with “form”. Marx's characteristic mode of questioning is always to ask “Why do these things take these forms?” (e.g. Marx 1976 pp. 173-4). The “things” concerned are production relations which are always, except in communist society, class relations, i.e., relations of struggle: in existing society it is the capital-labour which is “formed” – which gives form and is re-formed – in varying ways. Marx's project is 'to develop from the actual, given relations of life the forms in which these have been apotheosized' (Marx 1976 p. 494). The “forms” concerned are the commodity-form, the value-form, the money-form, the wage-form, the state-form, etc. If “form” is understood dualistically, i.e. as opposed to content which is distinct from it, then once again (for reasons parallel to those given in connection with “appearance”) either determinism or reductionism results. In the event, however, forms are to be understood as mediations (as modes of existence, or appearances) of the class relation – under capitalism, the capital-labour relation – and hence of the struggle in which that relation consists. (On the centrality of class struggle to all the categories of Capital, see Cleaver 1977: every single category in Marx's critique of political economy is designed to contribute to the description of the mediations – the modes of existence – of class struggle, and this is one reason why Capital is to be seen as presenting a critique of political economy rather than a rival political economy on its own behalf.)

It is worth noticing that all of the mediations set forth by Marx stand to be mediated in their turn: for example, exchange-value is the mediation (the mode of existence or appearance) of value, and is for its part mediated by the money-form. For Marx, as for Hegel, no process of mediation is definitive: mediated terms may themselves call for remediation, and far from being static or merely “structural” the process of mediation and remediation is one in which the praxis of class struggle – and therefore capitalism's response to labour's insurgency – is inscribed. Better: mediation and remediation are at issue in class struggle, inasmuch as mediations are forms of class struggle. As usual, it is the categories which thematize activity – here, the activity of struggle – which are given primacy by Marx. Understood thus, the concept of mediation explodes all deterministic readings and establishes revolutionary subjectivity at the very centre of Marx's work.

This being so, there can be no question of revolutionaries having to intervene from outside (like Leninist vanguardists) in inert social structures in order to conjure struggle into existence or to generate praxis from process, since it is as mediations of struggle and as at issue in struggle that social “structures” and social “processes” exist. In this sense, for Marx as for Hegel (and in opposition to every variety of bourgeois or pseudo-Marxist sociology), a social world 'is not a dead essence, but is actual and alive' (Hegel 1977 p. 264). It follows that the politics entailed by a reading of Marx in the light of the category of mediation is, with Luxemburg, a politics of spontaneism: but in the Marxist tradition Luxemburg's category of spontaneism has been understood no less confusedly than the category of mediation itself. At the close of the present paper, I shall offer brief comment on what I take the category of spontaneism to involve.

An additional virtue of the concept of mediation is that it makes possible a theorising of the relation between class struggle and struggles of other kinds. For example, the relation of class oppression to sexual oppression has been a topic of notorious difficulty in both feminist and Marxist thought: sexual and class oppression are intertwined, but of course sexual oppression is older than the capital-labour relation. The necessary insight here is to the effect that capitalist valorisation is not a closed dynamic. i.e., not merely one which destroys, externally, all 'patriarchal and idyllic' pre-capitalist forms (although just such a view seems to be implied in, for example, the Communist Manifesto's opening pages). Rather, it is to be seen as an open process of totalisation which is always ready to incorporate – viciously, voraciously – whatever in pre-capitalism can serve its purposes and lies ready to hand. It incorporates such elements as its own mediations, and in so doing re-“forms” them (understanding “form”, here, in the definitional sense specified above). In this way, capitalism re-forms the family and transforms sexual relations within the family into a “form” of the capital-labour relation itself: the nuclear family comes into being cotemporally with industrial capitalism (Shorter 1976). The sexual relation becomes a mediation of the class relation and vice versa. Women's unpaid labour in the nuclear family serves as a free subsidy to capital so far as the reproduction of labour-power is concerned.2
Thus, sexual emancipation presupposes, but is not reducible to, class emancipation (and vice versa). This analysis is the opposite of reductionist because it construes the process whereby capital re-forms sexual relations as one of struggle and implies neither that all of existing sexual oppression is a consequence of this re-formation – although it is all affected by it – nor that sexual oppression will be automatically terminated once the capital-labour relation is destroyed.

In passing, it can be noted that capitalism's continuing employment of pre-capitalist relations is crucial not merely for any concrete understanding of the capital- labour relation's mediations but also for an understanding of the sources of legitimacy upon which capital may draw. (An example is racist legitimacy, bound up with a heritage of anti-semitism and slavery, so far as the capitalist state is concerned. Amongst other things, this heritage makes it possible for capital to organise a flow of “immigrant” labour-power to and fro across the state's boundaries and in accordance with valorisation's needs.) To see capitalist valorisation as a closed and sheerly self-sustaining dynamic, and capitalist legitimacy as stemming solely from the exchange relation, is to downplay its capacity for incorporating, as its own mediations, that which is or was non-capitalist – and so to underestimate the strength (deriving from flexibility) of that to which revolutionary struggle is opposed. The sheer 'formalism' of the exchange relation would supply capitalism only with a weak legitimation; the 'substantial' sources from which it can derive strong – but nonetheless always problematic – legitimation are both older and more “irrational” and mythic than a Marxism impressed with the hegemony of liberal values would suppose (Horkheimer and Adorno 1969). Sometimes, fascism is analysed as an archaic throwback to times before capitalist rationality prevailed; if this is accepted, however, the conclusion must be that all capitalist states are fascist on precisely this score. The advantage of the category of mediation, here, is that it allows us to break away from the image of a “pure” capitalism overlain and sullied by what Stalinist Marxism terms 'survivals' from a pre-capitalist past. On the contrary to such views: the strength of capital is its capacity to re-form pre-capitalist relations as its own mediations and thereby to translate them into modes of existence of itself.

What I have said about “form” sheds light on yet another contentious area of Marxist theorising, this time an area of a methodological kind. One of the central topics addressed in Marx's 1857 Introduction to the Grundrisse manuscripts is that of the relation between categories which are (in their designation) abstract and categories which are concrete, and we learn that, instead of starting from the concrete and abstracting from it, we must start from the abstract and show how the concrete is composed out of it – the concrete, here, being viewed as 'the concentration of many determinations, hence unity of the diverse' (Marx 1973 p. 101). I shall not attempt to explain all the issues raised in this complex passage, but only to draw out a distinction between two ways in which the abstract/concrete relation can be understood.

Abstracting from the concrete involves abstraction in what may be termed an empiricist sense: the more I abstract, the further I move away from (concrete) reality and the less real – the more purely conceptual – my abstractions become. Marx is for his part willing to employ abstraction in this sense, as when he remarks that 'all epochs of production have certain common traits, common characteristics. Production in general is an abstraction, but a rational abstraction...' (Marx 1973 p. 86). But he adds at once that 'there is no production in general' (Marx 1973 p. 86), in the sense that production is always historically specific, and one of his objections to vulgar political economy is that it confuses abstraction in its empiricist meaning with abstraction in a sense which the notion of mediation brings to light. In this latter sense, that which is abstract can be a mode of existence (a form) of that which is historically specific and no less real than any other aspect of the concrete totality in which it inheres. Mediations, in short, may be either abstract or concrete or a (contradictory) unity of the two. The example which Marx gives of abstraction as mediation is labour, which 'achieves practical truth as an abstraction only as a category of the most modern society' (Marx 1973 p. 105) wherein value-production obtains. The 'dual character' of labour (Marx 1976 pp. 131ff.) as abstract and concrete – as productive of value and of use-value – is one among the mediations of the capital-labour relation itself. Confusing abstraction in the empiricist sense with abstraction in the sense of mediation allows the political economists to construe that which is specific to capitalism (in the present example: abstract labour) as intrinsic to production under all social formations whatever. This confusion is one aspect of the fetishism of categories to which Marx is constantly opposed.

Once again, we arrive at a point which is decisive for how Marx's own critique of political economy is to be read. To be sure, the first volume of Capital discusses “capital in general” in abstraction (more or less) from the questions posed by the existence of “many capitals”, and even at the end of Volume Three we still have to 'leave aside' the conjunctures of the world market, credit and so forth (Marx 1966 p. 831). But this in no way entails that the value-form, abstract labour, surplus-value, etc. – in short, all the central topics of Volume One – are less real than the topics approached as the arguments of Volume Three unfold. Value and labour precisely as abstractions, in the sense of mediations or modes of existence of the capital-labour relation, do real (and murderous) political and exploitative work. The mediations which Volumes Two and Three of Capital add to those of Volume One – the remediations, in other words, of these former mediations – in no way subtract from the importance of the story (the story of the capital-labour relation as a relation of class struggle) which Volume One tells. Nor is it a matter of a “pure model” of capitalism – which exists no more than does 'production in general' or a Weberian 'ideal type' – being moved “closer to reality” by successive stages. To be sure, Volume Three approaches 'step by step the form which they [the 'various forms of capital'] assume on the surface of society...and in the ordinary consciousness of the agents of production themselves' (Marx 1966 p. 25): but just here it is important to keep the sense of “forms” and of “appearances” as mediations clearly in mind. For example, “many capitals” is the mode of existence of “capital in general” and, minus the 'practical truth' – the real social existence – of “capital in general”, the intelligibility of “many capitals” disappears. The point here is more than a textual one. If Volume One is treated as presenting a “pure model” of capitalism (an abstraction in the empiricist sense), then Marxism's emphasis on class struggle – the struggle inscribed in the capital-labour relation – evaporates; both in theory and in practice, one finishes up endorsing the mystified 'ordinary consciousness' of capitalist social relations and the fetishism in which (as Volume Three demonstrates) that consciousness is steeped. To read Marx as an empiricist – as employing only an empiricist concept of abstraction – is to read him as a reformist, and both his political and his theoretical challenges are evaded at a single stroke.

One way of summing up what has been said concerning Marx is to see it as articulating further the possible shapes of mediation discussed above. For it will be apparent that, for Marxism, one application of the concept of mediation as mode of existence (as form or appearance) is of key importance, namely, the application of this species of mediation to a situation wherein, prior to mediation, an antagonistic – or self-antagonistic – relation characterises the to-be-mediated terms. Indeed the antagonism may be one strong enough to destroy the terms, as in the Communist Manifesto's scenario of 'the common ruin of the contending classes'. Hegel tells us what a mediation of antagonistic terms can mean: it can mean that each antagonistically (or self-antagonistically) related term achieves the 'power to maintain itself in contradiction' (Hegel 1971 para. 382), or in other words in its antagonism (which is not at all to say that the antagonism is removed outright or abolished). Suppose, now, that a mediation of this kind brings about an internal relation between, or within, the antagonistic term(s): in such a case, the mediation is the mode of existence not merely of the term(s) themselves but of their antagonism. The antagonism concerned is not removed, but on the contrary is sustained and set on a new footing, inasmuch as (qua mediated) it no longer consumes and destroys or undermines itself. Thus, for Marx, mediations of the contradictions inherent in the commodity-form (the central contradiction is between use-value and exchange-value, and its mediation is money) 'does not abolish these contradictions, but rather provides the form [read: the mode of existence] within which they have room to move' (Marx 1976 p, 198).

In this example, mediation allows not only the antagonistic terms but their antagonism to remain in being. Money, as the mediation of the commodity, is not just superadded to the commodity but is the mode of existence of the commodity itself: 'The riddle of the money fetish is...the riddle of the commodity fetish, now become visible and dazzling to our eyes' (Marx 1976 p. 187). In the absence of this mediation, use-value and value would remain merely juxtaposed, in the sense that use-value production, as a condition of all social existence, is by no means merely value- production and indeed points beyond it. Not the least aspect of the fetishism of commodities is the circumstance that use-value production, as a universally imposed condition of human existence, is established, through mediation, as related internally to value. Hereby, fetishistically, the existence of capitalism becomes inscribed in the ineluctibly given order of things.

Antagonism, of course, returns us once again to class struggle: if the various moments of capital are mediations (forms, modes of existence) of class struggle, then they are mediations which sustain this struggle not merely within the (broad) limits of the avoidance of 'common ruin' but within the (narrow) limits of a capital-imposed order of things. If this is so, then it seems that neither set of limits can become an issue for class struggle – social existence can involve risks neither per se nor for the powers that be – as long as these mediations are in play. And yet, since it is as an antagonistic relation – the capital-labour relation – which they mediate, it is as forms of struggle that capital's mediations always-already exist. The “play” of mediation is thus the play (the risk-taking praxis) of struggle itself. Risk, that is, is intrinsic to social existence and remains so even when it exists in the mode of being denied.

And this returns us to the topic of spontaneism, touched on above. The presence of antagonism in capital (and as capital) allows us to say that, in capitalist society, mediation always and only exists as the possibility of, so to say, going into reverse gear. Mediation exists as the possibility of demediation. Putting matters in this way allows us to avoid what would be a new form of reductionism, namely, a discovery (an uncovering) of class struggle as a level of pristine and authentic immediacy which lies under mediation's shell. Reductionism would be involved here inasmuch as immediacy would be counterposed to mediation, in a dualistic fashion, as the latter's essence and truth. In fact, what lies under mediation's shell is nothing: or, rather, the whole metaphor of a “shell” (together with its famous “kernel”) is inapplicable since the mode of existence of class struggle is the process of mediation and the possibility of demediation itself. This means that the antagonistic contradiction of mediation/demediation is intrinsic to class struggle, as Luxemburg lucidly sees: 'On the one hand, we have the mass; on the other, its historic goal, located outside of existing society. On the one hand, we have the day-to-day struggle; on the other, the social revolution. Such are the terms of the dialectical movement through which the socialist revolution makes its way' (Luxemburg 1970 pp. 128-9). This 'dialectical', or in other words contradictory and self-contradictory, movement is the movement which the term “spontaneism” connotes. In no way does spontaneism conjure, magically and romantically, a surging groundswell of immediacy which will eventually carry before it the web of mediations whose putative truth it is and to which it is externally juxtaposed. On the contrary, the contradiction inscribed in mediation is inscribed in the challenge to mediation as well, and there is no space of immediacy located outside of mediation which might supply a foothold or point of departure from which revolutionary challenge could spring. Spontaniesm connotes demediation and not the conjuring of immediacy, as Luxemburg (unlike her critics) already so sharply sees.

These two things are true: mediation exists as the possibility of demediation; and there is no immediacy, not even in revolution's camp.

If this is so, then the project of revolution (the project of demediation) always contains something paradoxical and, as it were, ironic and playful (it is demediation “making its play”). What Adorno says of the dialectic of identity and non-identity applies to the dialectic of mediation and demediation as well: 'I have no way but to break imminently, and in its own measure, through the appearance [read once again: the mode of existence] of total identity' if non-identity is to come to light (Adorno 1973 p. 5), since it is as modes of existence of one another that identity and non-identity obtain. Indeed, more than analogy relates Adorno's defence of non-identity to the theme of mediation/demediation, since a good part of revolutionary struggle turns on the articulation of that which is particular and nonidentical and hence marginalized with respect to the conformism that any social order entails. This is most evidently the case with sexual politics but holds equally for class politics as well. In Georges Bataille's terms: heterogeneity is to be rescued from the homogeneity which, for example, the bourgeois exchange relation presupposes and enshrines (cf. Bataille 1985 and, on identity and the exchange relation, Adorno 1973). But rescuing particularity and heterogeneity and nonidentity must involve paradox since universality, homogeneity and identity are inscribed in the very conceptual ordering whereby any rescue-attempt must be thought through (to say nothing of the organisational forms which revolutionary practice may find itself driven to adopt). 'The concept of the particular is always its negation at the same time; it cuts short what the particular is and what nonetheless cannot be directly named, and replaces it with identity' (Adorno 1973 p. 173). As with the concept of particularity, so with the concept of demediation: in order to remain in play, it is called upon always to think against itself. And if there remains something opaque about the category of demediation, so be it. Transparency would announce it merely as a fresh mediation, and so close the conceptual space within which the figure of 'revolutionary subjectivity' finds itself able to appear.


Kosmas Psychopedis convinced me of the importance of mediation and my poor understanding of it. In the course of many discussions, Werner Bonefeld helped me to see issues at stake.


Adorno, T. W. 1973: Negative Dialectics (Routledge and Kegan Paul)

Bataille, G. 1985: Visions of Excess: Selected Writings 1927-1939 (Manchester University Press)

Cleaver, H. 1977: Reading 'Capital' Politically (Harvester Press)

Dalla Costa, M., and James, S. 1976: The Power of Women and the Subversion of the Community (Falling Wall Press)

Hegel, G. W. F. 1892: Logic [Encyclopaedia of the Philosophical Sciences Vol. I] (Clarendon Press)

Hegel G. W. F. 1971: Philosophy of Spirit [Encyclopaedia of the Philosophical Sciences Vol. III] (Oxford University Press)

Hegel G. W. F. 1977: Phenomenology of Spirit (Oxford University Press)

Horkheimer, M., and Adorno, T. W. 1969: Dialectic of Enlightenment (Allen Lane)

Kirk, G. S., and Raven, J. E. 1963: The Presocratic Philosophers (Cambridge University Press)

Luxemburg, R. 1970: Rosa Luxemburg Speaks (Pathfinder Press)

Marx, K. 1966: Capital Vol. III (Progress Publishers)

Marx, K. 1973: Grundrisse (Penguin Books)

Marx, K. 1976: Capital Vol. I (Penguin Books)

Negri, A. 1984: Marx beyond Marx (Bergin and Garvey)

Shorter, E. 1976: The Making of the Modern Family (Collins)

Taken from

  • 1Here, I speak of the mediation of a relation whereas, previously, I have spoken of the mediation of terms. What may seem like a confusion is only a verbal difficulty, which may be resolved in one of two ways. Either the expression 'term' may be understood in a broad fashion, so as to include relations as one species of term. (In this case, the relation between commodity producers is mediated to itself through the commodity form.) Or the commodity producers themselves may be understood as the 'terms' which the commodity form mediates or relates. Nothing turns on which of these alternative resolutions is adopted, and the expression 'mediation of a relation' can be understood as shorthand for this either/or.
  • 2See Dalla Costa and Jones 1976. In her notes to the 1976 edition of this work, Dalla Costa mistakenly says that women's housework is productive not merely of use-value (the use-value of labour-power) but of value and surplus value as well. If this were so, it would destroy her own argument: women's housework would increase (instead of holding down) the value of labour-power, and capital would have an interest in decreasing the amount of housework which women do.