Review: Method and the dialectic
Mike Macnair completes his review of: Guglielmo Carchedi, 'Behind the crisis: Marxs dialectics of value and knowledge'. Haymarket Books, 2012, pp303, £20
In the first part of this review, last week, I said that Carchedi’s book is an important one which should be widely read, though not uncritically; but that part of the review, which addressed issues of the theory of capitalist crisis and of the social construction of knowledge, was fairly extensively critical.1 This part of the review - on the foundations of Marx’s critique of political economy and on the dialectic - has more of the “important book which should be widely read” side of the story.
Chapter 1 of the book concerns method and the dialectic; chapter 2, ‘Debates’ - issues in the foundation of Marx’s critique of political economy - and (in section 6) the alien ‘rationality’ of the marginalists’ homo economicus. I will address chapter 2 first.
Carchedi begins chapter 2 with the assertion that Marx’s theory “has been the object of sustained attacks aimed at showing its logical inconsistency. The critique has centred upon four issues: abstract labour as the only source of value, the materiality of abstract labour, the law of the falling rate of profit, and the so-called ‘transformation-problem’ ... If the critiques are proven to be correct, there would be no sound platform on which to build a radically alternative view of capitalism and thus of its tendency towards crises and towards its own supersession” (p53).
We should pause for a moment here. Marx’s critique of political economy is addressed to the specific features of capitalism. A substantial element of Marx’s critique is the claim that political economy fails to grasp the specificity of capitalism, instead treating features of capitalism, like generalised commodity exchange, as existing from the time of the first human social interactions.
As such, the critique logically entails not only the possibility of a non-capitalist future, but also the actual existence of a non-capitalist past, out of which capitalism has emerged. This issue is by no means absent from Capital - for example, in the elaborate account of the creation of the proletariat through enclosures, etc, in part 8 of volume 1. It is not absent from the Grundrisse or the Contribution to the critique of political economy. A substantial part of Marx’s work after the publication of volume 1 of Capital consisted in studies in pre-capitalist social and land relations.2 Anti-Dühring went out formally in the name of Engels, but, contrary to a very widespread belief on the left, was effectively a joint work, with Marx seeing the drafts before publication and contributing at least one chapter.3 It, too, pays substantial attention to the pre-capitalist past.
Having regard to this, it is, in fact, perfectly possible to infer from ‘historical materialism’ - the claims of Marx and Engels about capitalism as one among a series of different historical social forms - that all social forms come into existence and will in due course pass away: there is no reason to suppose that in capitalism history has come to an end. This argument is independent of the validity of the specific form of Marx’s critique of political economy.
Of course, it does not follow that what will come after capitalism is working class rule/socialism or a path to communism. For that conclusion, in the first place more argument is needed, and the critique of political economy is part of that argument (a demonstration, among other things, that Proudhonist mutualism, which tries to attain equality, while preserving commodity production, would merely be at best a road back to a redeveloped capitalism). But this point is by no means dependent on any single interpretation of Marx’s unfinished critique. Secondly, the upshot is not determined, but will be a matter of human choices, and it is quite possible that capitalism will end in human extinction through generalised nuclear war or through passing the tipping point at which global warming becomes runaway.
Why, then, should the critiques of logical inconsistency in Capital imply that “there would be no sound platform on which to build a radically alternative view of capitalism and thus of its tendency towards crises and towards its own supersession”?
The answer is that the US state and its satellites during the cold war brought up a lot of academic heavy artillery against the theoretical claims of Marxism. The academic gunners took aims according to their fields. The economists argued the superiority of marginalism either in its Keynesian form, or Sraffa’s ‘neo-Ricardian’ version of Keynes, or in ever more esoteric mathematical micro-economics (Arrow-Debreu equilibrium, etc). Carchedi polemicises mainly against self-identified Marxist students of political economy who accepted part of these critiques.
The philosophers took aim principally at dialectic; and especially at the ‘transition from quantity to quality’, which, they argued, denied the ‘reality’ that all change is actually gradual. Hence, though few were crude enough to make the point directly, the only legitimate socialist politics is Fabian gradualism. Though the arguments have not disappeared, since 1989 pro-capitalist ideologues have themselves become revolutionists (‘colour revolutions’, etc), and the arguments for the necessary gradualness of change are an embarrassment to them.
The historians and political scientists took aim mainly at ‘historical materialism’. Being more politically sophisticated than their counterparts in economics departments, they set out to lever apart Marxism using Marx and Marxists. To support Bernstein’s coalitionism, they call in aid Luxemburg’s critiques of Kautsky and the German Social Democratic Party centre, adding post-1918 communist critiques of ‘Kautskyan automatism’. To oppose the claims of Marx and Engels for the leading role of the proletariat as a class, and the idea that historical materialism implied that capitalism would naturally come to an end, they emphasised Marx’s (unpublished) draft letters to Vera Zasulich on Russia; and urged the significance of the ‘Asiatic mode of production’ in Marx and Engels’ early writings as showing the absence of any inherent directionality in history: this idea was, it was argued, ‘Eurocentric’. In all except the last of these, the academic cold warriors were indirectly supported by ‘official communist’ ideology aimed at backing up the ideas of the people’s front and national roads to socialism.
A part of the ‘new left’ of the 1950s-70s in effect fought back against cold war criticisms of the critique of political economy and of dialectics, but swallowed cold war criticisms of historical materialism. Hence, it was necessary to take radical distance from Engels (who was supposedly a ‘vulgariser’ of Marx, and the originator of Kautsky’s ‘mechanical Marxism’); and equally, hence, the idea that there could be an alternative to capitalism could only be drawn from Marx’s critique of political economy itself. Carchedi’s argument here seems to be (inexplicitly) within this framework.
The irony is that within the academy, it is ‘historical materialism’ which remains, to use Imre Lakatos’s phrase, a “progressive research programme” with profound, albeit usually dilute, influence on the profession of the study of history, and producing real results; while academic ‘Marxist economics’ has come to display the symptoms of a ‘degenerating research programme’, a marginal niche activity mainly concerned with arcane internal polemics.
To some extent, this results from the fall of the USSR, widely seen as a disproof of Marx’s economic claims; to some extent from the ferocious offensive of the US and other capitalist states and of direct capitalist donors to economics departments in favour of ‘neoliberal’ versions of marginalism. This is particularly problematic because to actually test Marxist predictive claims about the economy would require major research resources for number-crunching which are, in the current climate, not available to Marxist academics.4
It remains true, however, that it is not the right option to accept the pro-capitalist ideologues’ critiques of historical materialism and attempt to build the critique of capitalism and the idea of an alternative on the critique of political economy alone.
In chapter 2 Carchedi’s approach is to defend Marx’s logical consistency, partly by using a temporal dialectic. There is, in a sense, a prior question: why not just accept the dominant marginalism? In the concluding section of the chapter, Carchedi interrogates the logical consistency of ‘orthodox economics’ through the figure of homo economicus. This section is extremely strong.
The point is that, in order to make its basic claims about supply and demand, marginalism presupposes that an egoistic and exploitative ‘utility-maximising’ ‘rationality’ is sufficiently normal human behaviour to allow the calculation of supply and demand curves. Considered as an empirical claim, this appears to be simply false. The marginalists attempt to escape from this problem by treating altruistic behaviour as pleasurable for the actor, and therefore utility-maximising.
But on this assumption, it cannot possibly be the case that marginalism predicts any outcome, because it necessarily predicts all outcomes. Some additional examples beyond Carchedi’s: early Chicago school economist Frank Knight argued that armed robbery is utility-maximising, since the robber’s investment in weapons and risk shows he values the goods taken more highly than the victim; more recently Michele Piccione and Ariel Rubinstein have, with somewhat satirical intention, shown that a ‘jungle economy’ in which goods are allocated by the use of force displays the same equilibrium tendencies urged for the market by marginalists.5
This is, in a sense, not a new point: Geoff Hodgson makes it in How economics forgot history (London 2001). Carchedi develops some of the implications: for example, in the case of status goods and financial products, demand rises if price rises. John Weeks develops others in The irreconcilable inconsistencies of neoclasssical macroeconomics (New York 2012). Carchedi also makes the specific and very important point that the marginalist model can only be made to work at all if time is, in fact, expelled.
In substance, the ‘predict all outcomes’ problem of theories of utility-maximising makes them just an economic variant of Panglossianism or the strongest versions of the immediate determination of all events by divine providence (“Are not two sparrows sold for a farthing? And one of them shall not fall on the ground without your Father” - Matthew 10.29). The fact that this stuff is taught as ‘economics’ and absolutely dominant in schools and universities is as if flat-earthism was to be the basis of the prescribed syllabus in physics, astronomy and geography.
The upshot is that there are reasons to reject all subjective-utility theories of value (not just the orthodox marginalist/neoclassical versions), which are independent of the question whether Marx’s interpretation of the labour theory of value can be made to work.
As I quoted above, Carchedi addresses four debated issues in the problem of making Marx’s labour theory of value work: abstract labour, its materiality, the law of the falling rate of profit, and the ‘transformation problem’.
In relation to the general point that only abstract labour creates value, Carchedi takes aim at a couple of relatively soft targets: the idea that it would be possible to have a capitalist economy in which all production was carried on by machines, humans having only the function of consumers; and the ideas that profits are a reward for risks or for the special skills of capitalists. Both are dismissed with suitable expedition. Not addressed are the somewhat harder targets of (a) the Austrian school argument that profit arises from time preference and the ‘roundaboutness’ of production: ie, the need to immobilise assets in the course of production (a standard example being that Austrians argue that value arises without labour in the ageing of wine and spirits); and (b) Keynes’ argument that interest is an incentive not to hoard money. The answers to both these arguments depend at the end of the day on arguments from the specificity of capitalist production and, in particular, the role of money.
Carchedi offers a more elaborate, and a fundamentally important, argument on the materiality of abstract labour. In Marx, concrete labour creates use-values; abstract labour, “the expenditure of human labour in the abstract”, creates exchange-value. He begins with a critique of Chris Arthur’s argument in The new dialectic and Marx’s Capital (Boston 2004) that abstract labour is immaterial and merely an aspect of the value-form. Against this view, Carchedi makes a fundamental point: the expenditure of human labour-power in the most general or abstract sense is a physiological fact. This argument can then be deployed against other value-form theorists, such as John Milios and Michael Heinrich.
The point can be looked at another way. For the marginalists, unemployment arises because wages are ‘sticky downwards’: that is, if only workers would accept lower wages, more employment would be offered. Trade unions and state welfare are then the obstacles to full employment. But - to take only one example - Robert C Allen in his Enclosure and the yeoman (Oxford 1992), though arguing using marginalist methods, is unable to avoid the point that, in spite of growing unemployment in the region he studies, in some ‘privileged sectors’ (agriculture, weaving and building) the wages of the employed did not fall. The explanation he offers is that the wage was so close to subsistence level that any reduction would imply that the worker simply could not work: more calories (and access to clothes, shelter, etc) are necessary to the ability to work (Marx’s ‘labour-power’) than merely to avoid immediate starvation. For Allen these are wages “above the market clearing level”; but the actual logic is that there is no such thing as a “market clearing level” of wages in the marginalists’ sense. The cost of reproduction of labour-power is a real floor on its price, and one which is inconsistent with the idea that markets tend to clear.
Once we accept that this is the case, we are, in fact, driven towards a labour theory of value: not necessarily in the sense that abstract labour is the yardstick which ‘measures’ different use-values to allow exchange, as Marx argued at the beginning of Capital volume 1, but in the sense that the abstract labour content of commodities constrains or limits possible prices if socially necessary activities are to be carried on under generalised commodity production; and, the larger the number and variety of commodities involved, the tighter these limits become. Though the money form, on the other hand, allows very considerable divergence of prices from these limits, this is a contradiction, and the limits ‘return’ in various forms, notably crisis.
I have discussed briefly in the first part of this review Carchedi’s account of crisis in terms of the tendency of the rate of profit to fall (FRP). As I said there, I am not convinced that simple and direct use of FRP in production is a satisfactory explanation of the tendency to cyclical crises in capitalism. Nonetheless, Carchedi’s arguments in section 4 of chapter 2 that the dominant tendency is towards FRP and the counter-tendencies identified are just that seem to me strong. The issue is whether this secular tendency to FRP leads to a tendency to increasing severity of crises - or to a tendency to monopolisation/oligopolisation and statisation.
The ‘transformation problem’ arises because Marx postulated in Capital volume 3 the formation of a uniform ‘general rate of profit’ through competition, leading to the movement of capital into the most profitable spheres. The ‘general rate of profit’ is an inheritance from the classical economists (and one maintained by the marginalists), but it is not without explanatory significance. Its function in Marx is to explain the market dynamics of rent, interest and ‘merchant profit’ (commodity futures speculation) under capitalism, as distinct from pre-capitalist societies, where very different social dynamics affect all three phenomena (rent is customary and ‘grows out of the harvest’; interest is either illegal or heavily regulated by law; merchant profit is so unpredictable that merchants are dependent on the formation of specialist social groups involving mutual assistance, commonly identified by nationality or religion, such as Jews, Parsees, etc).
Marx worked up ‘prices of production’ distinct from direct labour values, which are not actual market prices, but prices formed on the assumption that each capital receives the general rate of profit in proportion to capital advanced. In this way surplus value created in particular industries at very varying levels is redistributed among capitalists.
The Austrian-school author, Eugen Böhm-Bawerk, argued that this procedure involved Marx smuggling into his theory of labour value an implicit contribution of value from capital. Ladislaus Von Bortkiewicz introduced a more specific objection: namely that for consistency both inputs and outputs needed to be valued at ‘prices of production’. Mathematical solutions to this ‘problem’ have attained an impressive degree of obscurity. Carchedi argues for a variant of the ‘temporal single systems interpretation’ (TSSI), also supported by Andrew Kliman, Alan Freeman and others, with the difference of the emphasis on dialectic. The essence of the TSSI, as Carchedi presents it, is that inputs do not have to be revalued at the time of output, since what is sold is a different commodity from what is purchased; where there is a change in market values of inputs during production which affects output prices, this is also a form of redistribution of value between different commodities. Von Bortkiewicz, and hence also Marxist ‘simultaneist’ interpreters of the ‘transformation problem, have tacitly introduced into Marx Leon Walras’s general equilibrium and Say’s Law (that markets inherently clear); a point made by Alan Freeman in several articles.
There is an alternative approach to this problem, not addressed or critiqued by Carchedi, which is that argued by Immanuel Farjoun and Moshé Machover in Laws of chaos (London 1983) and followed by (for example) Paul Cockshott and others’ Classical econophysics (Oxford 2009). This is to deny the reality of the general rate of profit. If so, the ‘transformation problem’ disappears because the transformation procedure is redundant: surplus value is not, in fact, redistributed between capitals in proportion to capital advanced.
Kliman, interviewed by Nick Rogers in this paper, argued that Farjoun’s and Machover’s point is true but does not dispose of the argument that Marx was logically inconsistent, because the absence of a general rate of profit is merely contingent.6 Carchedi does not cite Farjoun and Machover, but in effect gestures towards their point by writing in his argument mainly in terms of the average rate of profit rather than the general rate of profit.
I am not convinced that the general rate of profit is actually a wholly redundant argument. On the one hand, the case for it on the basis of movement of capital into the most profitable businesses is, I think, untenable. This is not only empirically because if there were an equilibrium outcome it would be a probabilistic distribution rather than an arithmetical mean (Farjoun’s and Machover’s point). It is also because assuming that, since capital tends to move into the most profitable businesses, theory should be constructed on the basis that it has done so sufficiently to produce a ‘general rate of profit’ is like assuming because entropy tends to a maximum that thermodynamics should be constructed on the initial assumption that it has already done so; or that since, in the long run, all organisms die, biology should be constructed on the initial assumption that all organisms are dead.
The general rate of profit is not fully redundant, however, because of the role it plays - as I indicated above - in accounting for capitalist market dynamics in relation to rent, interest (and other financial profits) and ‘merchant profit’.
This point relates back to Heinrich’s objection to FRP theories discussed in the first part of this review. Heinrich’s basic point was that Engels chose to base Capital volume 3 on an 1865 manuscript which contained arguments Marx had in effect rejected in subsequent work; hence Marx’s researches in the middle and later 1870s and his non-publication of volumes 2 and 3. Part of the meaning of these researches is that Marx had in effect abandoned the scheme of the four-volume Capital and returned to the scheme of the Critique of political economy, as necessarily including at least land and consideration of the specificity of capitalism in relation to rent by comparison with pre-capitalist forms; that is, he recognised that the scheme of explanation of market rents, etc in the manuscript which became Capital volume 3, which is dependent on the general rate of profit and hence the transformation procedure, did not work.
My point here is not that the general rate of profit is not logically conceivable, nor to accept von Bortkiewicz’s ‘correction’ to Marx. It is that the argument of which the ‘transformation’ is part depends on a hypothetical general redistribution of surplus value, flowing from a ‘completion’ of the working out of competition, which would, if it were ever to occur, bring capitalist dynamics to an end. And this cannot be an explanation of the actual transition to ‘market rates of interest’ and ‘market rents’, etc, which occurs fairly early in the development of capitalism (in fact, shortly following from the formation of organised financial markets, originating in state deficit financing).
The TSSI authors, Carchedi here included, are nonetheless surely right to insist that Marx’s Capital is not a theory of capitalism as an equilibrium system, but as a disequilibrium system, staggering forwards through time from crisis to crisis. Carchedi concludes his section on the ‘transformation problem’ by linking this issue to that of his interpretation of dialectic. The ‘transformation’ involves transitions from potential value created in production to realised value in goods sold, and of the same goods back to potential value, as they are incorporated as inputs in the following production period. To dialectic, therefore, we now turn.
Chapter 1, ‘Method’, argues for a temporal interpretation of Marx’s dialectic. The argument is buttressed in appendix 3, which argues that this concern with temporality is manifested in Marx’s Mathematical manuscripts.
Carchedi begins by limiting his approach. He will not follow Engels as to ‘dialectics of nature’, but makes claims limited to social reality. Further, he does not claim that his approach is applicable to all modes of production, but only that it is applicable to capitalism. And, finally, he is not engaged in Marx exegesis (though he quotes Marx at various points), but argues that his approach is to be supported for reasons of consistency and explanatory power (p2).
The starting point is empirical observation, though this is filtered through interpretive frameworks. The observed phenomena are the starting point, consisting of social relations and processes of change in unity-in-contradiction (p3). Social phenomena are always both realised (past) and potential (future possibilities). For example, in the case of the commodity, it is a realised product of human labour, but contains potentially social value which is realised only when it is sold (pp4-5). The realised and the potential have to be held in a ‘unity of identity and difference’: hence a unity of opposites in the temporal dimension (pp6-8).
Related to this, social phenomena are always both determinant and determined (pp8-18). At pp8-11 Carchedi argues that humans attempt to develop their potentialities, but within the framework given by the class order, which is defined by the “ownership-relation”: to be understood not as juridical ownership, but as who has the power of decision as to what is to be produced, for whom and how; which, itself, is an outcome of the class struggle. At pp11-12 he argues for the ultimately determinant character of production relations, using quotations from the Grundrisse, but at the end of the day on the rather simple basis that only what has already been produced can be consumed; though the distribution may well determine what is produced in the next round of production. More generally, “only previously existing phenomena can determine the actualisation of other phenomena” (p16). Thirdly, social phenomena are subject to constant movement and change (pp18-22). Because they are inherently contradictory, this movement is tendential rather than mechanical.
In section 3 of the chapter, Carchedi addresses dialectics of individuals and social phenomena. His key move here is to distinguish between concrete and abstract individuals - abstract individuals being individuals so far as they are members of some realised social group and as such are replaceable (pp22-25). The social phenomena are potentially present in the concrete individuals (and vice versa, concrete individual phenomena are potentially present in their social relations and processes); with an extensive set of corollaries (pp25-31).
Section 4, ‘Class analysis and the sociology of non-equilibrium’, draws out some implications: for example, that “social structure is not static but dynamic” and that “reproduction is not equilibrium” (both at p32). Carchedi also recommends the work of Resnick and Wolff in New departures in Marxian theory (Oxford 2006).
Section 5 addresses the relation of his arguments to Engels’s temporal dialectic in Dialectics of nature. While recognising similarities in treatment, Carchedi insists that he makes claims only about social reality, and objects that Engels’ attempt to ground dialectics in nature implicitly assumes the neutrality of the natural sciences, which Carchedi rejects (for reasons discussed in the first part of this review).
Section 6 offers an account of the relation between formal logic and dialectical logic. He argues that formal logic belongs only to the realm of the realised and excludes that of the potential. Dialectical logic admits contradiction, being a “contradiction between what has become and what can become, as contradictory to what has become” (p41, emphasis removed). Formal logic, insofar as it rules out processes of change of this sort is ideological in the defence of the existing order; but, so far as we are dealing with the realm of the realised rather than the potential, dialectical logic must employ formal logic as an auxiliary method (pp43-44).
Section 7, ‘Induction, deduction and verification’, argues that reasoning begins with the real concrete, moving from it inductively in a process of abstraction, and from the abstract, forward to working up the concrete (quoting a celebrated passage from the Grundrisse). But, unlike induction in formal logic, the premises can be contradictory because reality is contradictory, and the result will therefore explain “the movement in its characteristic features rather than in all its aspects” (p48). Verification is inextricably linked with class content: there are no neutral quantitative data or methods, though data can be employed against the purposes for which they were created; as far as logical consistency is concerned, “whenever the rules of formal logic cannot decide among contradictory (elements of) theories (all of them internally logically consistent) it is the social, class-content that decides” (p50, emphasis removed).
The idea of dialectic as addressing temporal problems of qualitative change, which is addressed by Carchedi’s approach in this chapter, is surely sound and of fundamental value. So, too, is the approach drawn from the difference between the realised and the potential. The fact that the argument is presented without Hegelianism, not as Marx-exegesis, but in terms of the explanatory value of the approach, is also an important strength of Carchedi’s treatment. This approach is a fundamental reason why the book should be widely read.
At the same time, I have already argued in the first part of this review that treating the “class content” of rival theories as dispositive of their truth-value is problematic in relation to debates within the workers’ movement. This movement is not an organic whole, but one which is necessarily constituted - even at the level of trade unions! - by conscious agreement to cooperate for common ends among people who disagree with each other widely.
In addition, it seems to me that Carchedi’s effort to amputate the objectionable ‘Engelsisms’ of temporal dialectic both necessarily fails and is unhelpful to the case for this approach.
It necessarily fails because, however much Carchedi says that he is only making claims about the social world of capitalism, the fundamental claim about time, which is made by the idea of the realised and the potential, inherently applies at least to biology and to human history in general: so that if he were to claim that it only applies to social order under capitalism, he is obliged to show how it does not apply to these other domains.
It is unhelpful to the case for the reasons already given. The belief that capitalism tends towards its own end is at the very least corroborated by the evidence of historical materialism. The attempt to give a Marxist account which strips out this aspect of Marxism weakens the argument - and makes it more dependent on the unhelpful “class content” argument.
It should be apparent, however, that I have found the book immensely stimulating to read. Though I am saying that it should be read critically, it should certainly be read.
1. ‘What drives capital’s global crises?’ Weekly Worker May 23.
2. Eg, L Krader (ed) The ethnological notebooks of Karl Marx Assen 1972; H-P Harstick (ed) Karl Marx über Formen vorkapitalistischer Produktion: vergleichende Studien zur Geschichte des Grundeigentums, 1879-80, Frankfurt 1977.
3. Eg, S Timpanaro On materialism London 1975, chapter 3, especially p77; H Draper Karl Marx’s theory of revolution Vol 1: The state and bureaucracy part 1, New York 1977, pp23-26.
4. I put on one side here the important fact that the Marxist critique of political economy predicts the regular return of crisis, which orthodox marginalist accounts do not, for two reasons. First, because Schumpeter predicted the regular return of crisis, and the Austrian version of marginalism predicts the regular return of crisis as long as the state monopoly of money-issuance continues, which is for all practical purposes the same thing (in reality, as the early 19th century US shows, abolishing state-monopoly money and the central bank produces an increased frequency and severity of crises). Second, because Marxists (meaning here not merely Marxist left groups, but also both Marx himself and professional Marxist economists) have shown a marked tendency to over-predict crisis.
5. ‘Equilibrium in the jungle’ Economic Journal July 2007, pp883-96.
6. ‘Crisis, theory and politics’ Weekly Worker September 27 2012.