Abstractly Human Labour and the Reduction Problem


“In Das Kapital Marx distinguishes between concrete and abstract labour. But what is abstract labour? In Section 1 I develop an explicitly normative account of this notion and argue that such a reading is essential for understanding what motivates the labour theory of value, not just of Marx himself but in general. Then, in Section 2, I use this normative reading in order to reconstruct and resolve a serious problem for Marx’s theory of value and his conception of political economy. This is the so-called reduction problem, the problem of how Marx can plausibly claim that all forms of concrete labour allow of being “reduced” to abstract labour. In conclusion I point out the implications of the normative reading of abstract labour for an elaboration of the notion of commodity fetishism and for Marx’s concept of political economy.”


In Chapter One of the first volume of Das Kapital Marx distinguishes two ways in which one can view productive labour. On the one hand, it is the creation of something with the capacity to satisfy some need or desire at least presumed to be had by some consumer of it. When described in this way, productive activity is what Marx calls concrete labour, that is, more or less efficacious poi?sis in which quite specific raw materials are worked upon, quite specific tools employed and quite specific skills brought to bear in the creation of something with utility—a use-value (Gebrauchswert). Each instance of concrete labour thus differs qualitatively from others: the tailoring of a coat differs qualitatively from the spinning of ten yards of linen.

On the other hand, productive activity is, when undertaken for the sake of exchange in some social division of labour, the creation of something with value (Wert). That is, it is the creation of something with the general property of being reliably and regularly exchangeable at market for a given amount of any other commodity similarly at market—a general property whose instantiations to specific amounts and specific commodities constitute the exchange-values (Tauschwerte) of the product.1 When described in this way, productive activity is what Marx calls abstract labour, the exercise or expenditure (Verausgabung) of human labour power, abstractly considered—abstractly human labour (abstrakt menschliche(r) Arbeit—MEW 23, S.52) or simply abstract labour.

Now Marx regarded his distinction between concrete and abstract labour as a decisive contribution to political economy2 and I believe he is right in this. For the notion of abstract labour which emerges from this distinction is essential to understanding what motivates the labour theory of value, not just of Marx himself but in general.3 And only once this motivation is clearly understood does it become possible to resolve an important problem for Marx’s theory of value and thereby for his conception of political economy and of the critique it must undertake of itself. This is the so-called reduction problem, the problem of how all forms of concrete labour can be, as Marx puts it, reduced to multiples of abstract labour—a problem which arises because the labour theory of value must somehow accommodate the possibility that different forms of concrete labour create value at different rates. Now the reduction problem is in fact deeper than the better known transformation problem—deeper not just because any solution to the latter presupposes a solution to it but also because it is a truly philosophical problem. For resolving the reduction problem requires a truly philosophical elaboration of the concept of abstract labour, one which brings out the deeply philosophical presuppositions of the labour theory of value and indeed of Marx’s whole conception of political economy. So let us turn now to see what Marx means by abstract labour and to elaborate it more extensively than Marx himself.

§ 1: What is Abstractly Human Labour?

Marx writes,

If we … disregard the use-value of those entities which constitute the material bodies of the commodity [Warenkörper], there remains only one property left to them, that of being a product of labour. Yet the product of labour has [itself] already undergone a change in our hands. In abstracting from its use-value, we abstract at the same time from the bodily components and forms which make it a use-value. It is no longer a table, a house, yarn or any other useful thing. All its sensual properties are extinguished. It is [therefore] also no longer the product of carpentry, masonry, spinning, or any other specific kind of productive labour. Along with the useful character of the products of labour disappears the useful character of the forms of labour presented in them and so the different concrete forms of these labours also disappear. They no longer differ from each other but are all of them reduced to the same human labour, abstractly human labour. (MEW 23, S.52; my translation)

One might well ask4 how Marx could possibly think that when we disregard the character of a commodity as useful, whatever that means precisely, we are left with one character only, that, namely, of being a product of labour. Surely all sorts of characters remain, for example, of being scarce relative to demand, of being subject to forces of supply and demand, or simply of being a natural object. Yet a little charity enables one to make sense of this passage: Marx is here not deductively arguing but rather phenomenologically developing, through an account of the experience of relating to something as a commodity, a working hypothesis as to the deep-structure of the act of exchange itself. For the question he is seeking to resolve is the following: Which of all the properties and features of a commodity which undeniably play some causal role in shaping the thinking and behaviour of parties to its exchange are relevant for understanding the act of exchange itself? The process of abstraction undertaken in theory is meant to track phenomenologically, thereby bringing to the surface, an abstraction undertaken in reality, by the parties to exchange themselves—an abstraction away from the utility of the commodity to those features of it to which they attend in the act of exchange itself.5

Now a commodity’s use-value obviously plays a causal role: the one party to exchange wants what the other is offering to such a degree that the commodity has sufficient utility to motivate embarking upon exchange. What if, however, there should be a certain kind of exchange in which expectations of utility and preference, once they have motivated the act of exchange, are ‘disregarded’ by the parties to exchange in order that they might bring certain other conceptual resources and capacities to bear? What if this kind of exchange should be paradigmatic in the sense that it is a condition of the possibility of all acts of exchange, hence of the whole social practice of production and exchange to which any such act belongs? Evidently, if this should be true, then, as much as expectations of utility and preference are obviously implicated in exchange, if one only considers them, one will persistently fail to capture how expectations of utility, preference and, through them, forces of supply and demand drive an ongoing, self-reproducing social practice of production and exchange.

I wish to attribute this line of thought to Marx in order to make sense, in the first instance, of his talk of abstraction but ultimately, of his concept of abstract labour itself. Marx’s talk of abstraction is precisely his effort to identify those conceptual resources and capacities which shape the paradigmatic act of exchange itself, once considerations of utility and preference have played their initiating causal role.6 And this effort is born of the conviction that concepts of expected utility and preference, and, through them, of supply and demand, are on their own unable to account for market operations. In Das Kapital Marx never argues for the conviction that these concepts are insufficient but he certainly asserts it, in a passage preceding the one just cited:

(I)t is precisely the abstraction from their use-values which evidently characterizes the exchange relation of commodities. In the exchange relation, one use-value is just as good as another, as long as it is present in the proper quantity. Or, as old Barbon says, “One sort of wares are as good as another, if the values be equal. There is no difference or distinction in things of equal value.” (MEW Bd. 23, S. 52)

Here Marx is saying precisely that in the paradigmatic case concepts appropriate for describing the dimension of utility and use-value are not relevant in the enactment of the exchange relation. Considerations of utility have already played their practical role, which is simply to cause the performance of the act of exchange. In entering the paradigmatic exchange relation the actors themselves abstract from these considerations in order to bring other conceptual capacities to bear.7 In this sense, abstraction occurs not primarily in the subject but in the object of investigation;8 in this sense, it is a real abstraction.

But what are these other conceptual resources and capacities brought to bear in the paradigmatic exchange relation? Here we need to understand why much classical political economy identified labour as the source of value. It did so because it believed that this move permitted development of a notion of ‘objective’ or ‘true’ value. And a notion of ‘objective’ or ‘true’ value was thought to be needed because the stability and self-reproduction of the social practice of exchange, hence of production for exchange, required it to be operative in at least sufficiently many acts. For this reason, neither Adam Smith nor David Ricardo nor the English communist John Francis Bray, whose views Marx expounds at length in The Poverty of Philosophy, ever give a proper argument as to why the labour theory of value is true.9 Rather, they simply posit it as a claim with a certain intuitive plausibility and crucially also the capacity to serve as a fundamental postulate10 with which to make economic reality theoretically tractable—this by providing the basis upon which to develop a description of this reality as a system of rule-governed self-reproducing social interaction.

Evidently, the underlying thought here is quasi-Kantian: exchange is essentially a move in a social practice of production and exchange, hence in a stable and regular, rule-governed division of labour. But for a stable and regular, rule-governed division of labour to be possible, sufficiently many acts of exchange must be governed by a conception of, and indeed a concern for, a rate of exchange grounded in the nature of the commodities exchanged, hence not simply the result of agreement induced by bidding. This idea of a rate of exchange grounded in the nature of the commodities themselves is what is meant by admittedly windy talk of ‘objective’ or ‘true’ value. And when the rate is thus grounded, the act itself is paradigmatic. Of course, a given act of exchange may diverge from the paradigm. Yet even in its divergence it remains related to the paradigm as an exception to the rule—as, for example, a lie relates to truth-telling. In this ontological rather than ethical sense, the divergent act is a sub-optimal realisation of the concept of exchange.

Now I am not concerned to argue here either for this quasi-Kantian thought or for the claim that it is implicit in advocacy of notions of ‘objective’ or ‘true’ value. Rather, I wish simply to intimate how this quasi-Kantian thought leads very naturally to the idea that precisely labour is the source of such value. Consider Smith’s example of trade in deer and beavers.11 The trade depicted in this example need not be understood as an original act from which the social practice of exchange, and ultimately of production for exchange, with its complex division of labour, either actually or hypothetically arose.12 Smith himself appears to have understood it in this fashion but even for him this is not its primary function. For the primary function of this example is to reveal the basis for the notion of a ‘natural’ price, that is, an ‘objective’ or ‘true’ rate of exchange which is not simply the result of arbitrary agreement between the parties to exchange. We are to imagine a deer hunter who wants beavers, a beaver catcher who wants deer. Deer thus have utility for the one, beavers utility for the other. These utilities or use-values cause them to enter into exchange. At the same time, we are to assume that both parties seek not simply to maximise utility but do so at a ‘true’ rate of exchange grounded in the nature of the goods traded. At first blush, this assumption barely makes sense—until, that is, we make a further one: we assume that by a ‘true’ rate of exchange, by a ‘natural’ price, both we ourselves and our parties to exchange mean a fair rate of exchange.13 Only this shift to the normative gives substance to talk of a ‘true’ rate of exchange.14

Clearly, a concern to achieve a fair rate of exchange rules out bidding, that is, a procedure whereby one party to exchange suggests a rate as advantageous as they think they can get away with, then is haggled down to rate upon which both parties agree.15 For bidding evinces no concern for fairness: it has not been excluded either that an exaggerated level of self-love, in other words, selfishness, is motivating the one to take advantage of the other; or that an exaggerated level of other-love, that is, self abnegation, is motivating the one to gift to the other. So how should our deer hunter and beaver catcher proceed? Clearly, they themselves, and not just we political economists reflecting on them, must abstract from their ‘natural’ differences to their ‘non-natural’ sameness in respect of rights and entitlements. In the light of this normative sameness the task becomes one of identifying some perfectly natural property which provides a criterion for determining what it would take for each to receive fair and proper due. And only one such natural property offers itself as such a criterion: the trouble each has gone to in obtaining the goods offered for trade. Each must get what they need or desire at a rate which equally compensates for this expenditure of labour time. In effect, deer hunter and beaver catcher, each acting out of a concern to treat the other not just as a means but also as an end, agree on the length of their respective toil as what in the circumstances counts as treating the other fairly.16

This immediately suggests the following account of what Marx means by abstractly human labour. Abstract labour is concrete labour viewed in a certain way or respect and Smith’s example suggests what this way or respect is: the property of being the average toil and trouble of a morally considerable being, that is, a being which deserves never to be treated just as a means but always also as an end. For only a morally considerable being deserves to be treated fairly.17 That Marx means precisely this is intimated by the following passage in his discussion of the commodity:

Aristotle could not infer, from inspecting the form of value itself, that in the form of commodity-values all labour is expressed as equal human labour, and therefore as labour of equal validity—because Greek society was founded on the labour of slaves, hence had as its natural basis the inequality of men and of their labour-powers. The secret of the expression of value, namely the equality and equal validity of all kinds of labour because and insofar as they are human labour in general, could not be deciphered until the concept of human equality had already acquired the stability of a popular conviction. This, however, first becomes possible in a society in which the commodity form is the universal form of the product of labour, in which, therefore, the dominant social relation is the relation between men as possessors of commodities. (MEW Bd. 23, S.74)

Clearly, the equality at issue here cannot be equality with respect to kind, competence or diligence since in these concrete respects human labour is manifestly unequal. Marx can only mean equality with respect to the normative significance human labour has simply in virtue of being human labour, i.e., the labour of morally considerable beings.18

But Smith’s example is useful in a further respect, for it also hints at what Marx himself wishes to do with the concept of abstract labour. As with Smith, so, too with Marx, there is no suggestion that a concern for fairness governs the behaviour of literally all individuals engaged in trade. So the concept of socially average labour time need not govern the behaviour of all such individuals. It immediately follows from this that Marx cannot regard the so-called law of value (Wertgesetz) as a natural law in the sense in which Newton’s three laws of mechanics are. It can at best be a “natural law” in the sense in which according to Locke equality is the first law of nature.19 Locke speaks of a law because in the state of nature equality is a moral principle not just true but also governing of behaviour. Crucially, it governs behaviour in a weaker and a stronger sense: in the weaker sense it governs just in case sufficiently many individuals adhere to it for the long-term stability and resilience of the social interactions for which it is a law; and in the stronger sense it governs just in case sufficiently many individuals adhere to it for this social order to count as realising what the law is meant to achieve, namely, equity or fairness.

The law of value has a similar status. In the first instance, it is valid for exchange as it is paradigmatically. That is, for there to be exchange, hence production for exchange, at all, the law of value must hold sway in the first and weaker sense: sufficiently many transactions must adhere to it for the social practice of production for exchange to possess stability and resilience. But this does not entail that it holds sway in the second and stronger sense. It does not follow that sufficiently many transactions conform to it for exchange to count as generally fair or just.20 In fact, this practice as a whole may be structurally unjust—in which case the law of value holds sway contradictorily.

This has a crucial implication for the critical significance of the law of value: because it is not a law in the natural scientific sense, it can permit exceptions in a wider sense than can a natural scientific law. Exceptions to natural scientific law are possible only insofar as they are cases in which the ceteris paribus clause inherent to such law is not satisfied.21 But in the case of the law of value22 there can be whole swathes of exchange-value which are simply not determined by labour time. Yet even in these cases the law of value still governs. It governs in the sense that the social practice of production and exchange to which anomalies belong must adhere to some sufficient degree to the law of value—in the first instance, for its stability, in the second instance, for the fairness of its exchange.

Finally, let us note two more important features of Smith’s example, the one relevant for the transformation, the other for the reduction problem. With regard to the transformation problem: note that in Smith’s example there is no mention of any fixed or constant capital. Presumably, our deer hunter and beaver catcher did not catch their respective quarries with their bare hands. Arrows were shot, traps were set and presumably this took its toll on the capital equipment employed. Yet there is no mention of this. Why? Because there need not be. Insofar as each has made their own capital equipment, this can be factored into the labour power expended on catching their quarry. Insofar as each has obtained their capital equipment through some kind of prior trade, we may factor into the labour power expended on catching the current goods the labour power expended on the goods traded in order to secure the capital equipment. Either way, the expenditure of fixed or constant capital can be construed as so much expenditure more of labour power. But now a crucial question arises: why should one thus construe it? There is really only one answer: fixed or constant capital, even when it exactly replicates the output of human labour, is not morally considerable. It does not toil in the requisite, that is to say, normatively significant sense. This is what it means to say that fixed or constant capital does not create value. The stage is now set for the need to explain how value is transformed into price.

With regard to the reduction problem: note that in Smith’s example there is no mention of reduction, either of deer hunting to beaver catching and vice versa, or of both either to abstract labour or to what we shall later see Marx to call simple labour. Why? Because this, too, is not needed.23 In this kind of case no common denominator need be identified in terms of which two very different kinds of concrete labour might be compared with respect to the rate at which they create value, hence contribute to determining a fair rate at which use-values are traded. After all, if Smith’s deer hunter and beaver catcher are equally masters of their craft, then, all else being equal, one day’s labour is no more or less toil and trouble for the one than it is for the other. So all that needs to be assumed in this case is that both parties to the trade have exercised their respective skills, whatever these skills might be, to the degree of competence and diligence normal or average for them, given the individuals they are. This labour time can then be factored into the determination of a fair rate. The ad hoc, occasional character of the case ensures that the one form of concrete labour counts for as much or as little as the other; the coefficient relating the one to the other is necessarily one.24

Now both features arise directly from the normative significance of the notion of abstract labour as interpreted here. So the question arises as to how this interpretation might help with problems of transformation and reduction. Since there is insufficient time to explore implications for the transformation problem, let us explore the ramifications of this interpretation for the in any case conceptually prior problem of reduction. Smith’s example reveals that the need to map different forms of concrete labour on to one another, such that one can meaningfully speak of the one form’s creating more value than the other, only arises when the trade at issue is a move in a social practice of production for exchange. This result is crucial to understanding what the reduction problem is and why Marx’s somewhat cavalier treatment of it is not as absurd as his critic Eugen von Böhm-Bawerk25 maintains.

§ 2: Understanding the Reduction Problem—Complex and Simple versus Abstract and Concrete Labour

Marx’s labour theory of value must pass a crucial test: it must permit one to show how it is possible for different forms of concrete labour to create value at different rates. In the first instance, the need for this arises from a simple empirical fact. As Böhm-Bawerk points out, “it is obvious even to the casual observer that … (t)he day’s product of a sculptor, of a cabinetmaker, of a violin-maker, of an engineer, etc., … does not contain an equal value but a much higher value than the day’s product of a common workman or factory hand, although in both the same amount of working time is “embodied.”” (Böhm-Bawerk 1949/1896, p.80 (S.103)) But in the second and more important instance, the need arises for a conceptual reason: Since commodities obviously need not exchange one to one, then, if the labour theory of value is true, neither can the concrete labours which produce them create value one to one. If the labour theory of value is to account even for the merely possible behaviour of exchange value, then concrete labour must be scalable across its different forms.

Now the scalarity required must have something to do with the character of labour as concrete. That is, differences in rate of value creation must be due at least in part to some difference in the labours themselves, hence a difference in them qua concrete since there is no difference between them qua abstract: from sculptor to engineer there is, we may assume, a day’s toil at the same level of mastery of different skills, with the same diligence in the application of these skills. At the same time, as Smith’s example shows, scale in contributions to value across different concrete labours is conceptually extrinsic to the character of labour as either abstract or concrete. The proportion between different forms of concrete labour thus cannot be determined solely by any property inherent to concrete labour itself,26 irrespective of its social setting. So Marx must regard the determination of the coefficients presupposed by different rates at which different forms of concrete labour create value as the result of some relation of concrete labour to its social context. What is this relation?

In order to answer this question, let us turn to Marx’s account of the reduction problem. How does he conceive that relation to social context which determines potentially different coefficients for different forms of concrete labour, such that it becomes possible for one day’s labour sculpting statuettes to create as much value as five days’ labour breaking stones? Indeed, does he clearly see that specifying such a relation is the task? Commodities, says Marx, such as coats and yards of linen, “(i)nsofar as they are values,27 … are things of like substance,” even though the activities which produced them

… are qualitatively different kinds of labour. … [For] (i)f we disregard the determinate character of productive activity, and therefore the useful character of labour, there remains only its character as the expenditure of human labour power. (MEW Bd. 23, S. 58)

This like substance is clearly the character of these commodities as products of human labour in its character as abstract in the normative sense indicated above. The expenditure of human labour power also has, however, a material or natural side: “Tailoring and weaving,” for example, “although qualitatively different productive activities, are both the productive expenditure of human brains, nerves, and muscles” and in this physiological sense “they are both human labour.” Crucially, productive activity also has an anthropological, hence social, cultural and historical side: the human labour of which tailoring and weaving are “ … but different forms … must be more or less developed in order to be expended in this or that [concrete] form.” In general, productive activity for the sake of exchange is the socially embedded expenditure of concrete labour power and only as such does it create value. Only in a social, cultural and historical setting can it produce something with the property of, as Marx puts it (MEW Bd. 23, S. 59), presenting (darstellen) abstract human labour—presumably presenting it to others.28 So only in such a setting can it produce something which counts as having value, hence counts as a commodity.29

Marx now makes a crucial move. For he now introduces the notion of simple, indeed of simple average labour, which he treats as identical with the notion of abstract labour. He writes that human labour as such, in other words, abstract labour, “… is the expenditure of simple labour power, i.e., of the labour power which, apart from any special development, exists on an average in the organism of every ordinary individual.” (MEW Bd. 23, S. 59; italics added) Why does Marx introduce the concept of simple or simple average labour? What is it? Let us follow Marx a little further. “More complex30 labour,” he says,

… counts only as simple labour intensified, or rather, as multiplied simple labour, so that a given quantity of complex labour is equal to a greater quantity of simple labour. … A commodity may be the product of the most complex labour, but its value, by equating it to the product of simple labour, represents a definite quantity of the latter labour alone. (MEW Bd. 23, S. 59)

An initial distinction between concrete and abstract labour has now been complemented by a new one between complex and simple labour. And it seems that this new distinction partially overlaps the former because simple labour is said to be identical with abstract labour.

Clearly, in his effort to accommodate the possibility that different forms of concrete labour create value at different rates, Marx is seeking some kind of scalarity. That is, he is seeking some way of construing the different forms of concrete labour as located on a scale on which all such forms are multiples of the others. For if he can find some appropriate scalarity, he will have taken a necessary but obviously not sufficient first step towards mapping the different forms of concrete labour, via this scale, onto a scale of greater and lesser value creation. For this reason, Marx writes here of complexity and simplicity since these implicate, or at least can implicate, the scalarity required: the more complex is, or at least can be, a multiple of the simpler by a coefficient greater than one, the simpler a multiple of the more complex by a coefficient less than one. Importantly, a scale of complexity and simplicity does not require anything maximally simple or, for that matter, maximally complex, on it. That is, there need be no absolute or atomic simples or complexes. Yet such a scale does entail that for any distribution of elements on it at least one possesses an average degree of simplicity or, viewed from the other end of the scale, of complexity. This appears to be why in the passage in which Marx introduces the concept of simple labour, he identifies simple with an average labour power. The scale Marx is envisaging is thus a scale of simpler and more complex on which there is no simplest (or most complex) element; the only distinctive elements on the scale are those which count as averagely simple.31

The crucial issue is, however, what assigns individual forms of concrete labour their position on the scale. In respect of what are the different forms of concrete labour more or less simple, more or less complex, relative to some average form?32 Notice that Marx never unambiguously answers this question! True, in a footnote in Zur Kritik der politischen Ökonomie he remarks that simple labour is what the English economists call unskilled labour.33 This suggests that the term “complex” just means “skilled,” the term “simple” unskilled. And much subsequent commentary has just assumed this to be so. But this cannot be right! Firstly, in the equivalent pages of the later work Das Kapital Marx does not speak of unskilled or skilled but only of simple and complex labour. Secondly and more importantly, any unmediated identification of “complex” with “skilled”, and of “simple” with “unskilled,” is wrong in principle.

For consider what such identification is trying to achieve! It seeks to map a scale of greater or lesser complexity onto a scale of greater or lesser rates of value creation. Yet no scale of greater or lesser complexity intrinsically associates itself with a scale of greater or lesser rates of value creation. The concept of value, grounded as it is in the notion of abstract labour, is an evaluative notion, as indeed the term “value” itself implies. Consequently, the concept of value creation is also evaluative. The concept of skill, hence of greater or lesser skill, is, however, completely non-evaluative. So the claim that a gradient of items scalable in respect of greater or lesser skill, or indeed in any other non-evaluative respect, is of itself identical with, or even merely of itself tracks, a gradient of items scalable evaluatively is false. It is to assume that from an ‘is’ one can derive an ‘ought’.34

Given this, there is no need to consider here the attempts made by Hilferding, Devine and Roncaglia to ‘solve’ the reduction problem.35 These writers all assume that greater or lesser simplicity just means greater or lesser simplicity in respect of skill. So at best they fail to see what the real issue is. And one must acknowledge that Marx himself does not see the real issue clearly, for his talk of simpler and more complex is confusing and unnecessary. At best, it is simply a way of articulating the formal properties which level of skill must have in order that it be linkable with greater or lesser value creation. Obviously, such a link is possible and nothing said thus far denies that it is. The point is rather that levels of skill and rates of value creation do not link up automatically, hence must not be naively identified with one another. In particular, one must acknowledge that insofar as such a link exists, it is instituted.36 The link can only be something conventional, perhaps a custom arising only behind the backs of producers37 according to which one form of concrete labour counts as creating more value than another in virtue of its involving greater skill.38 The relation to social context that determines the coefficients fixing the rates at which different forms of concrete labour create value is normative institution. And indeed ultimately Marx sees this: “The different proportions in which different kinds of labour are reduced to simple [average] labour as their standard are established [fest__gesetzt__!] by a social process behind the backs of producers, hence seems to them to be given by custom [Herkommen].” (MEW Bd. 23, S. 59; italics added)

Now level of skill is certainly a non-evaluative, more or less scalable property which could be elevated behind the backs of producers to a standard for determining what concrete labours have higher rates of value creation. It is surely right to say that the daily product of sculpting statuettes is more valuable than that of breaking stones at least in part because the former labour is more skilled than the latter. But we now see that what is actually important in the reduction of concrete to abstract labour—more accurately, in the establishment of a scale setting diverse forms of more or less skilled concrete labour in numerical relation to cases of averagely skilled concrete labour and thereby to one another—is the aspect of normative institution, not the non-evaluative scalable property instituted. The heart of the reduction is in fact the creation of a convention with regard to some non-evaluative scalable property or other. So concrete labours need not be reduced to one another in respect of skill, they could be reduced to one another, naturally only via some concept of average labour, in some other respect. For example, a social division of labour is conceivable in which concrete labours are reduced with respect to how their products realise societal conceptions of justice (Recht) and welfare (Wohl),39 hence realise its ethical life (Sittlichkeit). This would be a social division of labour in which assembling solar panels or caring for the elderly generated value at a higher rate than turning gun barrels on a lathe or flipping burgers.

Of course, if reduction is not grounded solely in the nature of concrete labour, if it is rather a conventional imposition from without, then it becomes critical to explain how and why level of skill might become the respect in which reduction takes place. After all, as Marx points out, reduction in respect of skill actually does take place.40 Surely this is only because the actual social division of labour is organised as commodity production. For in commodity production “products confront one another as the results of autonomous private labours acting independently of one another.”41 (MEW Bd. 23, S. 57) In other words, the flows of information between producers are primarily market-mediated, through competition, hence the forces of supply and demand.42 If, however, this is so, then there cannot be the kind of genuinely collective communication between producers which would be required if reduction were to proceed with respect to social good. Because in commodity production the social process of reduction must work behind the backs of individual producers, it surely can occur only in respect of skill.

Conclusion—Further Implications

The account just given of reduction, and therefore also the normative account of abstract labour that it presupposes, are significant in at least two important respects. Firstly, the explanation just given as to why in commodity production reduction should take place according to skill gestures towards a way of elaborating something left under-elaborated by Marx himself: commodity fetishism, which according to Marx arises in any form of commodity production yet is particularly extreme under capitalism. The normative reading of abstract labour suggests that commodity fetishism arises in a situation in which the information channels between producers are such that the diverse forms of concrete labour cannot be reduced rationally to one another, either in a substantively rational sense, that is, in terms of social good; or in a procedurally rational sense, that is, self-consciously and collectively.43 Some kind of reduction must occur since this is required for that meshing of concrete labours in which a coherent social division of labour consists.44 But the value of products and the value-creating character of concrete labours have a tenuous and potentially diminishing correspondence to what realises social good.45 In this sense, I suggest, the commodity form “… reflects back to human beings the social character of their own labour as an objective character of the products of labour themselves, as a natural property of these things; and so it also reflects back to producers their social relation to the totality of their labour as a social relation external to them, existing not between them but between the products of their labour.” (MEW Bd. 23, S. 86)

Secondly, the normative reading of abstract labour has implications for the concept of political economy itself. Since it entails rejection of the idea that economic reality could be captured in a mathematical model in analogy to physical reality, the law of value cannot purport to capture the behaviour of all prices across the board. In some forms of economy, exchange as it is paradigmatically could be implemented only to a degree which, although sufficient for relative stability, also contained many systematic divergences from it. But this would not refute the law of value since its role is critical rather than strongly explanatory. It is the keystone in a demonstration that the actual social division of labour does not conform to how it could and should be ordered—that indeed it systematically thwarts realisation of social good even as it generates vast quantities of readily accessible goods and services. Yet in order for the law of value to fulfil this critical role, it must nonetheless play a weakly explanatory role. Specifically, it must enable a theoretical account of the capitalist mode of production and exchange which shows how this mode presupposes for its stability, and therefore for its very possibility, a conception of exchange implicating ethical notions whose realisation is systematically undermined by the wider context within which exchange occurs. So the law of value only truly does what it is supposed to do when embedded within an account of why at least sufficiently many exchanges must conform to the ideal form, hence to the labour theory of value. Only a political economy conceived in this quasi-Kantian spirit could show that at least sufficiently many prices must oscillate around value as determined by labour time.46


Böhm-Bawerk, E. von 1949/1896 Karl Marx and the Close of His System, edited and introduced by P. Sweezy, New York: Augustus M. Kelly

Original text: “Zum Abschluß des Marxschen Systems, in Staatswissenschaftliche Arbeiten—Festgaben für Karl Knies, edited by Otto von Boenigk, Berlin 1896

Also reprinted in Aspekte der Marxschen Theorie 1. Zur methodologischen Bedeutung des 3. Bandes des Kapital, edited by Friedrich Eberle, Frankfurt: Suhrkamp Verlag, 1973. p. 25 ff.

Christensen, C. 2015 “Two Kinds of Economy, Two Kinds of Self—Toward More Manageable,Hence More Sustainable and JustSupply Chains,” in Human Ecology Review, Vol. 21, pp.3-21

Devine, J. 1989 „What is ‘simple labour’?—A Re-Examination of the Value-Creating Capacity of Skilled Labour,” in Capital and Class, Vol. 13, pp.113-131

Harvey, P. 1985 „The Value-Creating Capacity of Skill Labor in Marxian Economics,“ in Review of Radical Political Economics, Vol. 17, pp. 83-102

Hilferding, R. 1904 „Böhm-Bawerks Marx Kritik,“ in Marx-Studien, Bd. 1, pp.1-61.

Kant, I. 1787/1976 Kritik der reinen Vernunft, Immanuel Kant—Werkausgabe Bd. 3, zweite Ausgabe, Frankfurt am Main: Suhrkamp Verlag

Lewis, D. 1974 Convention—A Philosophical Study, Cambridge, MA: Harvard University Press

Marx, K. 1846/1977 Das Elend der Philosophie, in Marx-Engels Werke, Bd. 4, Berlin: Dietz Verlag, S. 63-182

Marx, K. 1859/1961 Zur Kritik der politischen Ökonomie, in Marx-Engels Werke, Bd. 13, Berlin: Dietz Verlag, S. 3-160

Marx, K. 1862/1965 Theorien des Mehrwerts, Teil 1, in Marx-Engels Werke, Bd. 26.1, Berlin: Dietz Verlag

Marx, K. 1867/1979 Das Kapital—Kritik der politischen Ökonomie, Bd. 1, in Marx-Engels Werke, Bd. 23, Berlin: Dietz Verlag

Smith, A. 1759/2002 The Theory of Moral Sentiments, edited by Knud Haakonsen, Cambridge: Cambridge University Press

Smith, A. 1776 An Inquiry into the Nature and Causes of the Wealth of Nations, available online at http://geolib.com/smith.adam/woncont.html

Tortajada, R. 1977 “A Note on the Reduction of Complex Labour to Simple Labour,” Capital and Class, Vol. 1, No. 1, pp.106-116.


  1. According to Marx the set of such properties is projectible: it has a genuine unity to it and is no mere list or enumeration. Thus Marx writes, “Ganz gleichgültig also gegen ihre natürliche Existenzweise, und ohne Rücksicht auf die spezifische Natur des Bedürfnisses, wofür sie Gebrauchswerte, decken sich Waren in bestimmten Quantitäten, ersetzen einander im Austausch, gelten als Äquivalente, und stellen so trotz ihres buntscheckigen Scheins dieselbe Einheit dar.” (MEW 13, S. 18) Marx explicitly distinguishes between value and exchange-value because the former is precisely the unity which underlies the set of exchange-values, making it projectible. Ultimately, of course, he will maintain that value supervenes in some way on the socially average amount of labour time required for the production of the commodity.

  2. See MEW 23, S.56.

  3. According to at least Marx’s labour theory of value, the character of a commodity as having value is in some sense grounded in the amount of labour power expended in the production of the commodity, averaged for competence and diligence across the whole social practice of production for exchange in which the act of production occurs. Averaging for competence and diligence across the whole social practice is required because if the labour power upon which exchange-value supervenes were what each individual producer individually expends in the production of a particular community, then there could be no general patterns of, or regularities in, exchange. Consequently, there could be no exchange value since exchange-value implicates a pattern of or regularity in exchange. A commodity is essentially something that is produced with a certain utility, is subject to certain standards of production, and is reliably or regularly exchangeable for other commodities in the social practice of production for exchange and ultimately consumption in which it occurs.

  4. As indeed Böhm-Bawerk does ask—see Böhm-Bawerk 1949/1896, p.75 (S.99).

  5. So the abstraction undertaken at the level of theory repeat in quasi-Heideggerian fashion (Wieder-Holen) an abstraction which occurs, or at least can occur, pre-theoretically, in everyday life.

  6. Note that these concepts are either explicity mentioned or at least implicit in the properties Böhm-Bawerk adduces as ones to which one is just as led much by abstraction from use-value as the property of being a product of labour. This intimates just how much Böhm-Bawerk fails to see what Marx is really doing. In abstracting from use-value, Marx is precisely rejecting, or perhaps rather, presupposing the rejection of, the properties to which Böhm-Bawerk, as a good marginalist, would appeal in order to capture the essence of exchange and the ground of exchange-value.

  7. Naturally, they do not necessarily bring these conceptual capacities self-consciously to bear.

  8. Note how Hegelian this thought is!

  9. Smith famously writes, “The real price of everything, what everything really costs to the man who wants to acquire it, is the toil and trouble of acquiring it. What everything is really worth to the man who has acquired it, and who wants to dispose of it or exchange it for something else, is the toil and trouble which it can save to himself, and which it can impose upon other people. What is bought with money or with goods is purchased by labour as much as what we acquire by the toil of our own body. That money or those goods indeed save us this toil. They contain the value of a certain quantity of labour which we exchange for what is supposed at the time to contain the value of an equal quantity. Labour was the first price, the original purchase-money that was paid for all things. It was not by gold or by silver, but by labour, that all the wealth of the world was originally purchased; and its value, to those who possess it, and who want to exchange it for some new productions, is precisely equal to the quantity of labour which it can enable them to purchase or command.” (Smith 1776, Book I, Ch. 5)

  10. As what Kant might call an Entwurf—see Kant 1787/1976, B xiii.

  11. Smith writes, “In that early and rude state of society which precedes both the accumulation of stock and the appropriation of land, the proportion between the quantities of labour necessary for acquiring different objects seems to be the only circumstance which can afford any rule for exchanging them for one another. If among a nation of hunters, for example, it usually costs twice the labour to kill a beaver which it does to kill a deer, one beaver should naturally exchange for or be worth two deer. It is natural that what is usually the produce of two days’ or two hours’ labour, should be worth double of what is usually the produce of one day’s or one hour’s labour.” (Smith 1776, Book I, Ch.6)

  12. In other words, the example need not be understood as what Marx calls a Robinsonade. Note that if one works with Marx’s understanding of the commodity and its exchange, then the trade depicted in this example is not, strictly speaking, an act of exchange at all. Nor are the deer and beavers traded commodities. According to Marx a commodity is by definition something with value and this is its possession of the general property of being exchangeable for x amount of A, where A is any other commodity at market. This clearly entails that the act of trading a commodity (as a commodity) is essentially a move in a social practice which sustains the web of exchange values—a social practice which also includes the production-for-exchange of the commodity, with all its potentially very complex division of labour.

  13. Smith speaks of our natural sense of justice in The Theory of Moral Sentiments—see Smith 1759/2002, Part II, § ii, Chapters 2 and 3.

  14. Note that to make this move is not in any way to moralise in the way in which certain writers within the tradition of classical political economy no doubt did—certainly Bray, perhaps Smith, not, however, Ricardo—and crucially also, not Marx. The moral concepts or rather conceptual capacities to which reference is made here are ones understood and endorsed by the object of politically economic investigation, not necessarily or exclusively by its subject. To think that by making or endorsing this move, one is indulging in utopian rather than scientific thinking is comparable to thinking that one is indulging in utopian rather than scientific thinking when one argues that, however much the desire to get other people to believe things typically motivates assertion, in assertion itself speaker and hearer abstract from this perlocutionary desire and the concrete situation to the abstract illocutionary roles of asserter and assertee, which are guided by considerations of comprehensibility, sincerity and epistemic reliability.

  15. That is, the rate which is, or at least believed by the participants to be, pareto-optimal.

  16. It cannot be stressed enough that nothing in Smith’s example entails that all acts of exchange conform to the model it presents. Nor indeed does anything in the example suggest that Smith ever believed they did conform. Consequently, nothing in the example suggests that the example either actually evinces or is believed by Smith to evince any exceptionless law of actual exchange. Indeed, given his objectives in The Wealth of Nations, he believes precisely the opposite. For he is seeking to portray exchange as it is ideally, that is, how exchange-value is ideally determined, in order then to be able to show that much actual exchange and much actual exchange-value falls significantly short of this, in other words, fails to correspond to its concept—as indeed is pre-theoretically and empirically evident in the ethically and prudentially deficient behaviour of the East India Company.

    Incidentally, I would like to think that the reason why, as many including Marx have noted, Smith has two conceptions of the source of exchange-value—on the one hand, the amount of labour time invested in the production of a commodity, on the other the amount of labour which the commodity can command at market—is that he regards the latter as operative in developed and therefore potentially degenerate forms of exchange and production for exchange. After all, the second conception is really just another way of saying that value is determined by utility. When utility is the only consideration operative, a developed form of exchange is degenerate.

  17. Note how the reciprocity of the property of exchangeability mirrors the reciprocity of the property of fairness.

  18. That this is so is intimated by the repeated reference to all human labour as equally valid.

  19. Clearly, capitalist society is (relatively) stabile and resilent, so sufficiently many acts of exchange must adhere to the law of value for this (relative) stability and resilience to be possible. But capitalist society is manifestly not just and therefore does not realise that even stronger desideratum which Aristotle called the good life, Kant the perfect good.

  20. Marx must be committed to maintaining that even in a radically deficient capitalist reality the law of value holds good in the sense that even those cases of what we would pre-theoretically call exchange are only possible if at least sufficiently many of them are cases of exchange in the full and proper sense. Note that this does not entail that all cases of exchange in the full and proper sense are cases in which socially average labour time is the only determinant of exchange-value. This touches upon how to understand correctly what Marx means, or at least must when properly reconstructed mean, by the reduction of so-called complex to simple labour.

  21. Someone like Böhm-Bawerk is happy to permit exceptions in this sense.

  22. As in the case of the rules of language or of a game, which are also honourable in the breach.

  23. That no such reduction is needed becomes all the more evident when one considers an example in which the misleading and superfluous trappings of a Robinsonade are not present. Imagine that a garden maintenance man needs an eye test and an optometrist needs some rubbish removed from the garden. Given their individual levels of normal or average skill, the optometrist must expend two hours of labour on the eye test while the maintenance guy requires a whole morning to remove the rubbish. Fairness dictates that the optometrist undertake two eye tests, for the maintenance guy himself, the other perhaps for his kid, in return for removal of the rubbish. Here, too, we have a case of ad hoc, occasional trade. And here, too, there is no role for any reduction of the one kind of concrete labour to the other.

  24. A third and a fourth feature are also worth noting: firstly, in Smith’s example no kind of averaging across the productive roles of deer hunter or beaver catcher is needed. This is because no productive roles are involved. Perhaps our deer hunter and beaver catcher do play a productive role in some social division of labour, perhaps indeed these roles are those of deer hunter and beaver catcher respectively. Nonetheless, on this occasion they are not acting these or any other roles out. (Note that this does not entail any rejection of the thesis that participation in such a social practice is a condition of the possibility of such one-off, occasional trades.) There is no move here in any social practice of producing for exchange and of exchanging. As far as determining a fair rate at which the deer and the beavers on offer change hands is concerned, all that matters is the time and trouble of each individual qua the morally considerable individuals they are. So all that needs to be assumed is that each has exercised their skill, whatever the level of this skill might be, to a degree of competence and diligence normal or average for this individual, not for the productive role.

    Secondly, as the third feature intimates, some kind of averaging is nonetheless involved. If when hunting the deer actually traded, the deer hunter had had a hangover which caused hunting to proceed sub-optimally at a rate of one and a half deer per day, fairness would have required this to be taken into consideration. So even in this ad hoc, occasional case, the relevant property of the two labour times is not their actual but the normal or average length of such labour times for this individual to which these actual labour times are subject. Although there is no social average, no average of performances of diverse actors standing in productive and commercial relation to one another, there is an average across the performance of an individual.

    Note that a rate of production determined by excessive application to the task, no matter what the causes of excessive diligence were, would also yield an unfair outcome. For it would mean that the excessively diligent would be disadvantaged should the other not be equally in excess. If being fair to the other requires, even in such an ad hoc and occasional case as this, that each party take some kind of responsibility for the competence and diligence of their productive activity, such that they acknowledge any degree of substandard performance as having to be taken into consideration, then being fair to the other also requires that extreme application to the task, whatever its causes, also be discounted, since it disadvantages the over-performer. This intimates a crucial point about what it means to have that concern for fairness which Smith imputes to his deer hunter and beaver catcher from the outset: such concern is quite distinct from an altruistic concern for the other. It essentially involves a sense of one’s own legitimate claims and to this extent stands so to speak between what Hutcheson and others of his time called sympathy and what they understood by self-love.

  25. https://en.wikipedia.org/wiki/Eugen_B%C3%B6hm_von_Bawerk.

  26. As Böhm-Bawerk points out—see Böhm-Bawerk 1949/1896, p.83 (S. 105).

  27. That is, genuine, that is to say, projectible unities of diverse exchange-values such as the exchange-value possessed by one coat of being equivalent to twenty yards of linen, the exchange-value possessed by twenty yards of linen of being equivalent to one coat, the exchange-value possessed by one yard of linen of being equivalent to one-twentieth of a coat, and perhaps also the exchange-value possessed by one yard of linen of being equivalent to two bushels of wheat, hence the property possessed by one coat of being equivalent to forty bushels of wheat, and so on, ad indefinitum.

  28. So perhaps Marx means that only in such a setting can concrete labour make its character as abstract labour recognisable by others, both in the sense of Erkennen and in the sense of Anerkennen. The notion of recognition at issue here is at once cognitive and performative.

  29. See MEW Bd. 23, S. 59.

  30. The translation at the Marxists Internet Archive (https://www.marxists.org/) renders Marx’s noun phrase “Kompliziertere Arbeit” as “Skilled labour.” The basis for this is a footnote in the earlier work Zur Kritik der politischen Ökonomie (MEW Bd. 13, S.16), in which Marx claims that simple labour (einfache Arbeit) is what the English economists call unskilled labour. Clearly, if this is right, then more complex labour is more skilled labour, just as the translation assumes. (Note, incidentally, that the translation disregards Marx’s use of a comparative.) But the fact that no such footnote occurs in the later work Das Kapital indicates that Marx had second thoughts about this—and rightly so. As is pointed out below, it is incorrect to identify simple average labour with unskilled labour and indeed to claim, as Marx still does even in the later work, that simple average labour just is abstract labour. In fact, the distinction between abstract and concrete labour is not the distinction between simple and complex labour. And the latter distinction is in fact merely a continuum or gradient of more or less skilled which possesses according to Marx a certain average: this average is what Marx in fact means by simple labour.

  31. That Marx does mean by simple labour something averagely, not ultimately simple is shown by the following passage: “Wir nehmen an, daß die Spinnarbeit einfache Arbeit, gesellschaftliche Durchschnittsarbeit ist.“ (MEW Bd. 23, S. 204)

  32. Or more accurately, forms; the ordering here is obviously only partial.

  33. See MEW Bd. 13, S. 16.

  34. Note that this is arguably the true and deeper point underlying Böhm-Bawerk’s objection to any argument that the reason why more skilled forms of concrete labour have higher rates of value creation is because they tend to produce, in the same amount of time, commodities with higher prices. Böhm-Bawerk points out (Böhm-Bawerk 1949/1896, pp. 83-86 (S. 105-107)) that on the labour theory of value, higher prices are supposed to be reflective of higher value and on this basis he concludes that Marx’s whole idea of associating different levels of value creation with different levels of skill is circular. Now in one way this is not correct; in one way the procedure to which Böhm-Bawerk is objecting is not disastrously circular. If one builds into one’s labour theory of value the idea that different rates of value creation are associated with different levels of skill, because one empirically observes that different product prices are associated with different levels of production skill, then obviously one cannot treat the theory as explaining this difference between prices. One could, however, still maintain that prices are proportional to value and this is, pace Böhm-Bawerk, what really concerns Marx. There is, however, a deeper point to Böhm-Bawerk’s objection. What is disastrously circular, i.e., question-begging, is the suggestion that the association of different levels of value creation can be motivated by, or derive from, nothing more than the character of skill itself and differences between skills. Insofar as this is what Böhm-Bawerk inchoately recognises Marx to be doing, he is right. It is precisely the fallacy of attempting to derive an ‘ought’ from an ‘is’.

  35. Hilferding attempts to link greater and lesser degrees of skill with greater and lesser rates of value creation via the greater labour involved in creating the difference in skill—see Hilferding 1904, pp.19-21. (In this regard, Roncaglia’s attempt is not essentially different from Hilferding’s.) But insofar as one assumes that greater or lesser Ausbildungsarbeiten make this link of themselves, one is trying to derive an ‘ought’ from an ‘is’. Hilferding is looking around for something which links greater and lesser skill to greater and lesser rates of value creation and he seizes upon the relation of different skills to the labours which have produced them to accomplish this. No doubt this move is grounded as follows: We all have the pre-theoretical intuition that an exercise of more skilled labour power deserves more reward at market than an exercise of less skilled labour power because of the greater trouble gone to in the acquisition of the more skilled labour power. So this intuition is shared by workers and it leads them to struggle for some recognition in the reward they receive for their labour in proportion to the value they create. The base line for their reward is, of course, what they require for the creation and maintenance of their labour power but this does not preclude their exacting from their employers recognition of the value they create. So little, indeed, does it preclude this that it is or can be the driver for an elevation in what they require for the creation and maintenance of their labour power. (As Adam Smith pointed out, these days a linen shirt is a necessity even for the simple labourer.) If this is so, then there will indeed emerge a tendency to link—or perhaps rather precisely a custom of linking!—different levels of skill with different levels of value creation as a causal consequence of the belief that higher levels of skill deserve more reward because of the toil invested in acquiring them. But this only emphasises the crucial point: the link must be instituted.

    In addition, against Hilferding one might object that he just assumes Ausbildungsarbeiten to be, at least ultimately, forms of einfache(r) Arbeit. Why this should be so is not clear and one must suspect that it is question-begging. Of all the authors who have discussed this issue, Devine comes closest to seeing what the real issue is. He claims that we do not need to choose between different concepts of skill coefficient—see Devine 1989, p.126. Indeed we do not need to choose and the reason for this is the choice is ultimately ‘arbitrary’ in the sense that it is a matter of resolving upon a convention. Consequently, so, too, is the determination of skill itself as the respect in which reduction takes place. Devine remains confused in one important respect, however: he still wants to claim that “simple labour is abstract labour” (p.128) whereas in fact one should drop this kind of talk altogether. As just pointed out, Marx’s confused and confusing talk of simpler and greater is doing no more than intimating the formal properties of scale which some feature F of concrete labour must have if a conventional link between F and rates of value creation is to be instituted.

  36. That Marx sees this is perhaps intimated by the following passage: “Das Materialisieren etc. der Arbeit ist jedoch nicht so schottisch zu nehmen, wie A.Smith es faßt. Sprechen wir von der Ware als Materiatur der Arbeit - in dem Sinne ihres Tauschwerts - , so ist dies selbst nur eine eingebildete, d. h. bloß soziale Existenzweise der Ware, die mit ihrer körperlichen Realität nichts zu schaffen hat; sie wird vorgestellt als bestimmtes Quantum gesellschaftlicher Arbeit oder Geld. Es ist möglich, daß die konkrete Arbeit, deren Resultat sie ist, keine Spur an ihr zurückläßt. Bei der Manufakturware bleibt diese Spur in der Form, die dem Rohmaterial äußerlich bleibt. In dem Ackerbau etc., wenn die Form, die die Ware, z.B. Weizen, Ochs usw., erhalten haben, auch Produkt menschlicher Arbeit, und zwar von Generation zu Generation vererbter und sich ergänzender Arbeit ist, so ist das dem Produkt nicht anzusehn. Bei andrer industrieller Arbeit liegt es gar nicht im Zweck der Arbeit, die Form des Dings zu ändern, sondern nur seine Ortsbestimmung. Z.B., wenn eine Ware von China nach England gebracht wird etc., so ist die Spur der Arbeit an dem Ding selbst nicht zu erkennen (außer bei denen, die sich erinnern, daß das Ding kein englisches Produkt ist). Also in der Art wäre das Materialisieren der Arbeit in der Ware nicht zu verstehn. (Hier kommt die Täuschung daher, daß sich ein gesellschaftliches Verhältnis in der Form eines Dings darstellt.)“ (MEW Bd. 26.1, S. 141-142)

  37. Lewis 1974 provides an account, derived ultimately from Hume, of how a convention or custom might arise behind the backs of the individuals for whom it holds good.

  38. Note Marx’s use of the word ‘gelten’: “Komplizierte Arbeit gilt nur als potenzierte oder vielmehr multiplizierte einfache Arbeit, so dass ein kleineres Quantum komplizierter Arbeit gleich einem größeren Quantum einfacher Arbeit.“ (MEW Bd. 23, S. 59) Böhm-Bawerk objects to this use—see Böhm-Bawerk 1949/1896, p.82 (S. 104)—but the upshot of the argument here is that Böhm-Bawerk misunderstands Marx.

  39. Note that this is a genuinely non-evaluative property; concrete labours are being scaled according to how their products realise socially shared belief about the right and the good. Note, too, that skill could still play a role, albeit a subordinate one; one could distinguish between concrete labours which make equal contributions to social good according to differences in skill.

  40. See MEW Bd. 23, S. 59.

  41. „Nur Produkte selbständiger und voneinander unabhängiger Privatarbeiten treten einander als Waren gegenüber.“ (MEW Bd. 23, S. 57)

  42. “Gilt deine Arbeitsstunde soviel wie die meinige? Diese Frage wird durch die Konkurrenz entschieden.” (MEW Bd. 4, S. 85) Of course, given that my hour of work can only ever count as some multiple of yours, it must be possible to determine what multiple of your hour of work my hour counts as in ways other than just competition. Marx’s claim here articulates what decides the issue under conditions of commodity production.

  43. Clearly, a substantively rational mapping requires a procedurally rational one.

  44. At one point Hilferding suggests that the point of Marx’s labour theory of value, which of course implicates the concept of ‘reduction’, is not to explain the behaviour of prices in all possible cases (as Böhm-Bawerk assumes) but to identify what holds the articulated, differentiated totality of social production, exchange and consumption together. This seems to me to be quite correct. See Hilferding 1904, S.11-12.

  45. Note, however, that any such mapping must involve at least some minimal degree of correspondence.

  46. And since these prices would surely be the ones involved in the most basic and everyday of exchanges, the ones characteristic of the ‘real’ economy, such a political economy provides a means of spelling out what one means when one thus naively speaks of the ‘real’ economy.