Extended Book Review: Freedom and republicanism in Roberts’ Marx

Impersonal domination

Chapter 3 of Marx’s Inferno contains the basic ingredients of Roberts’ account of capitalist domination. The chapter associates Marx’s critique of political economy with its radical republican background, especially the work of William Thompson, John Francis Bray, Robert Owen in England, and Charles Fourier and Pierre Joseph Proudhon in France. It then argues for three main claims: (1) Marx kept the emphasis of republicanism on domination as subjection to arbitrary power, but (2) complemented it with an ‘impersonal’ account of domination of labour by capital that is at once anti-moralistic and structural, and (3) Marx’s account of fetishism is, fundamentally, about ‘a form of domination, rather than a form of false consciousness’ (p. 85). The rest of this section discusses claims (2) and (3). I shall take up (1) in the next section.

According to Roberts’ Marx, the opening chapters of Capital ‘contain, not a theory of price, but an account of the structure of human relations in commercial society. The concept of value they elaborate is, first and foremost, a tool for understanding the dominating form that these relations take’ (p. 102). Because decisions about what to consume and produce are fundamentally dependent on prices, ‘deliberate judgments about the place of commodities in a decent life’ become ‘extremely fragile’ (pp. 84–85). Moreover, because capitalist social relations are price-mediated, that is, mediated by properties of things, ‘there is no way to ensure that one’s labor power, or one’s commodity in general, is marketable’ (p. 88). It follows that both capitalists and workers are dominated under capitalism:

The capitalist, dominated by market imperatives, is compelled thereby to exploit labor, and to do so within the confines of competitive market transactions that secure for the laborer the fair value of his … labor power. (p. 102)

Therefore, Roberts thinks, ‘the dominant class in modernity, the class of capitalists, is as subject to … impersonal domination as are the laboring classes’ (p. 102). The worker is, in this sense, ‘a slave of a slave’ (p. 103).

Roberts infers – correctly, I think – that capitalist domination is not equivalent to class domination. There is, however, an ambiguity in his claim that the capitalist is ‘dominated by market imperatives’. Are market imperatives the ground of domination or its subject? The former idea is coherent: markets are the structure in virtue of which capitalists dominate workers. The latter idea, however, is not coherent. And if it is incoherent, then capitalists cannot be ‘slaves of slaves’.

Consider the idea that markets dominate (capitalists or workers). This idea issues from a fallacy of composition: the coherent thought that constituents of market structure (capitalists) dominate other constituents (workers) is illicitly transformed into the incoherent thought that market structure itself dominates capitalists and workers. In other words, statements such as ‘in the context of capitalist structure, capital dominates labour’ are logically equivalent to statements such as ‘in the context of cathedral structure, buttresses support walls’. It is a fallacy of composition to infer, from the latter statement, that cathedral structure supports cathedral walls. Rather, cathedral structure confers on buttresses a power of support over walls. By the same token, capitalist structure – Althusser’s subjectless process – confers on capitalists a power of dominion over workers. Impersonal domination therefore entails that capitalist structures themselves do not dominate. Roberts roundly criticizes Moishe Postone and Michael Heinrich (pp. 91–93) on this score, but then inexplicably asserts that ‘modernity … is a world without masters’ (p. 133, n. 141).

Now, Marx (1976) does say that ‘[t]he capitalist is just as enslaved by the relationship of capitalist as is his opposite pole, the worker, albeit in a quite different manner’ (p. 990). It does not follow that markets or market imperatives dominate capitalists. Capitalists might, for example, dominate each other by jointly constituting the ‘external coercive necessities confronting the individual capitalist’ (Marx 1976: 381). I propose to further clarify this possibility by means of the following example.

P is a landowner and Q is a landless worker. P hires Q to work on P’s land for a subsistence wage. P dominates Q. Is P’s domination of Q impersonal, in Roberts’ sense? No: the explanation of the agency that dominates Q is not reified; no ‘coercive law’ determines the wage rate that P offers. There is, by implication, no fetishism here since P’s social relation with Q neither is nor appears as a social relation between P’s commodities (money) and Q’s commodities (labour power). Call this the noncapitalist case. Now R comes into the scene. Like P, R is a landlord. P and R produce commodities, which they sell to each other and to Q. Now P and R begin to compete with each other. Labour is scarce, so competition between P and R drives up Q’s wages. This competition causes long-term prices to fluctuate, not in order to meet demand – goods markets clear – but to maintain profit. Now, this is beginning to look more like impersonal domination: an impersonal institution, the market, confers on P, Q, and R certain definite roles. If markets are competitive, as Marx assumes, then P and R have no alternative but to choose a price that maximizes profit. The circle is therefore complete: the capital circuit is internally reproducible. P and R individually dominate each other, and jointly dominate Q. Moreover, the social relation between P and Q has been transformed into a social relation between P’s money and Q’s labour power. It is therefore fetishistic. Call this the capitalist case. I will now argue that Roberts’ republicanism cannot make sense of the capitalist case.

Domination and republicanism

According to Roberts, Marxian impersonal domination is a form of republicanism. More precisely, Marx’s theory of freedom appropriates the radical republican theories of Owen and Proudhon (pp. 75–78, 80–81, 97–100, 178–179), removes their moralistic halo (pp. 67–70, 94–95), and gives them a structuralist twist (pp. 94–101). But Marx does nothing of the sort. That is, there is much theoretical and textual evidence that the critique of political economy constitutes a break with radical republicanism. The reason is simple: republicans of all stripes, including Owen and Proudhon, conceive of domination in terms of subjection to an arbitrary will. But the ‘market imperatives’ criticized by Marx do not involve such arbitrariness. Therefore, impersonal domination and republicanism pull in different directions; they are incompatible.

The theoretical case for this incompatibility is supported by the contrast between the noncapitalist and the capitalist cases presented above. In the noncapitalist case, P’s power over Q is relevantly like the power of an absolute monarch over her subjects; it is, in that sense, arbitrary. No such arbitrariness obtains in the capitalist case: here P is constrained, on pain of competitive disadvantage, to maximize profit, which in turn requires paying Q a market-clearing wage. Crucially, none of P, Q, and R possess arbitrary power over others in virtue of the aforementioned economic facts. To be sure, P and R possess power over Q. But the explanation of that power goes from things – P’s and R’s commodities – to wills – capitalists P and R – and not vice versa. Therefore, capitalist domination is impersonal not in the sense that its subject is the market, but rather in that its subject is a will, a will that is – in a uniquely relevant sense – nonarbitrary. Modernity engenders domination without arbitrariness, something that Marx observed more keenly than his contemporaries. Individual bosses might, of course, have arbitrary power over workers in the workplace. But that form of domination, as Roberts recognizes (pp. 235–236), is qualitatively distinct from, indeed derivative of, the impersonal domination of capital over labour.

Consider a structural analogy: Q finds herself in a pit. P and R are standing above it. P makes an offer to Q: ‘either you work for me for $999 or I leave you there’. R also issues an offer to Q: ‘either you work for me for $1,000 or I leave you there’. If P does not make the offer, then R will force P to do so. If R does not make the offer, then P will force R to do so. In this example, there is nothing arbitrary about the power of P and R over Q, however one defines arbitrariness. According to Philip Pettit, for example, P’s power over Q is arbitrary just when P’s power is not forced to track the interests of Q (Pettit 2012: 58–59). The pit example shows that this definition is too narrow: P is forced to make an offer that significantly promotes Q’s interests. Yet Q is dominated. Crucially, Pettit’s account does not make room for those ‘market offers’ that Roberts thinks constitute domination (p. 97). Once again, the problem with republicans is not that their account is too focussed on will-to-will transactions, as Roberts suggests (p. 98). Rather, republicans necessarily misconstrue the nature of such transactions (I discuss this in an article entitled ‘Reclaiming Domination’, which is available from me upon request). Marx and Engels (1956) underscore the anachronism of the republican categories along similar lines:

Saint Just, on the day of his execution, points to the large table of the Rights of Man hanging in the hall of the Conciergerie and says with proud dignity: ‘C’est pourtant moi qui ai fait cela’. It was just this table that proclaimed the right of a man who cannot be the man of the ancient republic any more than his economic and industrial conditions are those of the ancient times. (p. 165)

Roberts might, however, suggest another sense in which the market subjects Q to arbitrary power. Impersonal domination, he avers, is partly constituted by ‘the desires and preferences of multitudes, and those multitudes, being anonymous and dispersed, cannot be challenged to defend their desires and preferences, no matter how much these might spell the ruin of a whole class of people’ (p. 99). It is questionable whether this gambit is compatible with Roberts’ purported anti-moralism (pp. 56–58); it also leads him to misdiagnose the remedies to impersonal domination (see below). For now, it suffices to note that the subject of impersonal domination cannot be the mere resultant of the preferences of market agents. Pace Roberts, capitalist markets are not preference-aggregating catallaxies: Marx is not a neo-Austrian. What is missing here is an account of the way markets mediate production and money-backed desires, transforming individual whim into ‘laws working themselves out with iron necessity’ (Marx 1976: 90). Marx does not get very far into the constitution of such laws until Volume 3 of Capital. I hasten to add that these laws need not presuppose the labour theory of value, or similar Marxological baggage that Roberts rightly wants to discard (pp. 78–81).

The shape of things to come?

I turn to the political ramifications of Roberts’ analysis. These are summed up in Chapter 7, which has two parts. The first part criticizes G.A. Cohen’s argument that Marx’s optimism about the inevitability of socialist revolution issues from an ‘obstetric motif’ he inherits from Hegel. The second part complements this critique, by drawing tentative conclusions for socialism. These two parts are less convincing than the rest of the book. I discuss each of them in turn.

Cohen (2000) famously held that Marx inherits from Hegel the idea that the solution of a problem ‘is the consummation of the full development of the problem’ (p. 63). Seen in this light, the problem of socialism is solved by the full development of contradictions immanent to capitalism. Therefore, the task of the working class is to deliver the socialist new-born by minimizing the birth pangs to mother history. Cohen charges that this engenders unwarranted revolutionary optimism on Marx’s part. A necessary concomitant to this optimism is an aversion to writing ‘recipes for the cookshops of the future’ (Marx 1976: 99). Roberts wants to vindicate Marx against Cohen. He writes,

The material conditions of socialism are … the proletariat’s felt need for large-scale cooperative production, coordinated on a national and global scale. This can only be a felt need when capitalist development has broken down the laborers’ reserve of individual skills, so as to make their material interdependence obvious and robust, and when the power of the capitalist state has developed to the point where the futility of worker separatism has become equally obvious. (p. 241)

So we have concentration and centralization of capitals (p. 217), deskilling and collectivization of labour, relative immiseration, state repression, obviousness of interdependence, felt need (p. 241) … Roberts’ list grows longer if transition costs are counted. And the longer the list, the less plausible it is to deny that Marx was overoptimistic about the prospects of socialist revolution. One explanation for this optimism – a debunking explanation – is Marx’s antecedent commitment to the obstetric motif (cf. Leopold 2007: 294; Stedman-Jones 2002: 170–176).

The book concludes with a tentative sketch of what ‘republicanism in the realm of production’ (p. 253) might look like. Roberts suggests that Marxian socialism resembles Owen’s ‘notion of cooperative production … based in a form of social mediation that excludes all forms of domination’. Impersonal domination, Roberts claims, is excluded by the fact that members of cooperatives do not interact with one another as ‘independent producers exchanging goods at market prices’. He adds, ‘[t]he relations among cooperatives are … matters for deliberation and decision by assemblies of delegates’ (p. 254). What does that mean? Answer: ‘Cooperative production compels individual producers and consumers to justify their preferences and to make decisions in common about what and how to produce’ (p. 255). But how might individual consumers and producers ‘justify’ their preferences? To whom? And why? Here Roberts’ exposition has been hijacked by the republican notion that market domination entails arbitrary power. In this quixotic search for arbitrariness, Roberts locates domination in the whims of market agents. But, as I argued above, arbitrariness is not the ground of impersonal domination. And even if it were, capitalist markets are not mere aggregators of individual preferences. Roberts’ republicanism therefore saddles Marx with a dubious anarchist notion that has no basis in Marx’s texts (see Schweickart 2006 for trenchant critique of that notion).

As Roberts points out early in the book, Marx’s critique of political economy is about the way the means of production come to rule over the direct producer. It scarcely follows that the emancipation of the producer requires that market agents justify their preferences to each other. Rather, emancipation requires democratic control over the allocation and disposal of the means of production – the main determinant of the social division of labour – through a federated system of democratic workers’ councils and cooperatives. Marx excludes questions about the design of this system from the purview of his critique. Pace Roberts, this exclusion is not only methodologically suspect, but it is also a major hindrance to theoretical innovation in the ranks of the revolutionary Left.