ABSTRACT LABOR: SUBSTANCE OR FORM?
A CRITIQUE OF THE VALUE-FORM INTERPRETATION OF MARX’S THEORY
by Fred Moseley
Universidad Autonoma Metropolitana (Mexico City)
Mount Holyoke College (Massachusetts)
Our analysis has shown that the form of value, that is the expression of the value of a commodity, arises from the nature of commodity value, as opposed to value and its magnitude arising from their mode of expression as exchange-value.
Marx (C. I. 152)
It is not money that renders commodities commensurable. Quite the contrary. Because all commodities, as values, are objective human labor and therefore themselves commensurable, their values can be communally measured in one and the same commodity. Mon ey is the necessary form of appearance of the measure of value immanent in commodities - labor-time.
Marx (C.I. 188)
ABSTRACT LABOR: SUBSTANCE OR FORM?
A CRITIQUE OF THE VALUE-FORM INTERPRETATION OF MARX’S THEORY
In the Introduction to our first book, I wrote:
All the authors agree that Marx attempted in Section 1 of Chapter 1 of Capital to derive abstract labor as the "substance of value," which exists prior to exchange - although it is not observable - and which determines the exchange values of com modities. But there is significant disagreement over the validity and necessity of Marx’s derivation. Indeed this disagreement is probably the most significant one amount the authors. This controversy has a long history beginning with Boehm-Bawerk. Smith, Arthur, and Reuten reject Marx’s derivation of abstract labor as the substance of value for two main reasons (which have also been cited by many others): the lack of an adequate reason for selecting labor as the common property of commodities that determ ines their exchange values, and the unresolved problem of reducing different kinds of skilled labor to equivalent quantities of simple abstract labor. Reuten also criticizes the type of abstraction that Marx makes in this derivation - an analytical or red uctive abstraction rather than a dialectical abstraction. Furthermore, these authors also argue that the important qualitative conclusions of Marx’s theory, such as the necessity of money, can be derived without recourse to Marx’s concept of abstract labo r as the substance of value... Murray is undecided about the validity and necessity of Marx’s derivation... The remaining authors largely accept Marx’s derivation, not as a "logical proof" but rather as a plausible hypothesis, the validity of which should be evaluated on the basis of the extent to which it can explain the important phenomena of capitalism. They also argue that the main advantage of the assumption of abstract labor is that it also provides a quantitative theory of profit, which is the main question of Marx’s theory. (pp. 10-11)
The purpose of this paper is to return to this most important unresolved issue among ourselves and to attempt to clarify our disagreements and perhaps even to narrow the extent of these disagreements. The focus of the paper will be the papers from our first book by Geert and Chris, and Geert’s book with Mike Williams.
a. Criticism of Marx
I will first consider Geert’s criticism of Marx’s concept of abstract labor as the "substance" of value and then examine Geert’s alternative concept of abstract labor, which he argues does not involve the difficulties of Marx’s concept. (I use "Geert" here as a shorthand both for Reuten and Williams (1989) and Reuten (1993)).
Geert’s main criticism of Marx’s concept of abstract labor in these works has to do with the well-known "reduction" problem. Geert argues first of all that Marx’s reduces different kinds of concrete labors to abstract labor by means of a "reductive " abstraction (in which use-values are simply disregarded), rather than a "dialectical" abstraction (in which the contradiction between private labor and social labor is transcended by positing abstract labor). However, Geert acknowledges that this is a m atter of methodological preference and not necessary a logical flaw in Marx’s concept (1993, p. 103).
The main problem with Marx’s concept of abstract labor, according to Geert, is that it requires a set of reduction coefficients (or "discounting" coefficients) by means of which different kinds of concrete labor (of different skills and different i ntensities) are reduced to homogeneous units of abstract labor. This reduction of quantities of concrete labor (CLi) to quantities of abstract labor (ALi) is accomplished by means of an equation like the following:
(1) ALi = ki (CLi)
where ki stands for the reduction coefficient for each type of concrete labor. In Marx’s theory, this quantity of abstract labor is then assume to determine the value added component of the price of commodities, according to the following equati on:
(2) VAi = m (ALi)
where m is the inverse of the value of money, or the rate at which abstract labor produces money new value per hour (also called the monetary expression of value), which is taken as given by Marx.
The problem, according to Geert, is that Marx never explained how these reduction coefficients (ki) are determined, and it appears impossible in principle to explain this determination independent of the market and prices. This is a pro blem because it renders Marx’s theory "not operational," which presumably means that Marx’s concept of abstract labor cannot be observed and counted in empirical research.
I believe that this [concept of abstract labor] makes a quantitative procedure at the empirical level very dubious... [In order to get to the concrete empirical level] we need a procedure to quantify the discounting coefficients. It is hard to see how this could be done prior to the market. (97-98)
As I have indicated, an interpretation of abstract-labor embodied seems near to what Marx was getting at. But I believe that there are too many problems to stick with it. The problem is not so much that of the reductive abstraction ... (although I do have methodological objections to it). This need not prevent it from being developed into a consistent and applicable theory. The difficulty, however, is that this [abstract labor] is taken to be value, and that value is taken to be a premarket enti ty. It is not clear how we can ever make this theory operational... Thus at some stage a procedure has to be developed for getting to the discounting coefficients. It has not been shown how this can be done prior to the market. Indeed, if we need the mark et to get to the coefficients, then we can no longer hold that value is a premarket entity. (103)
My response to this criticism is that Marx simply took these reduction coefficients as given, as did Smith and Ricardo before him. Geert considers this a serious weakness because, without a knowledge of these reduction coefficients, the theory cann ot be made operational. But this criticism assumes that all theoretical concepts must be observable; in other words, it assumes a form of positivism or empiricism. This positivist requirement seems very odd coming from a Hegelian-Marxist. Marx’s ke y concepts of essence and appearance (e.g. money and prices as the "necessary form of appearance" of the unobservable essence of abstract labor), which Marx of course adopted from Hegel, seem to be entirely missing from Geert’s Hegelian interpretation of Marx’s theory.
I argue, to the contrary, that a scientific theory may involve the postulation of unobservable entities and that Marx’s concept of abstract labor is such an unobservable entity.
These unobservable entities are then used to explain the observable phenomena of interest; in the case of Marx’s theory, to explain prices, surplus-value, etc. The appropriate way to empirically test such a ("metaphysical") theory is to evaluate the ex planatory power of the theory on the basis of the range of important phenomena that can be explained on the basis of the postulated unobservable entities, or the extent to which the main conclusions of the theory are supported by the actual historical evi dence, in comparison with other theories. Richard Miller (1987) has emphasized that Marx’s theory is of this nature and has called it a "realist" interpretation. Derek Sayer (1979) has also presented a similar interpretation. Marx’s basic hypothesis, that the aggregate price of commodities is proportional to the aggregate abstract labor required to produce them, cannot be tested directly because the quantity of abstract labor is not directly observable (and also because the value of money, the factor of p roportionally between abstract labor and prices, is also unknown). But the validity of this basic hypothesis can be evaluated on the basis of the range of explanatory power of this hypothesis (which I argue is considerable and much greater than any other existing economic theory) (see Moseley 1994 and 1995). Therefore, the fact that Marx’s concept of abstract labor is "not operational" is not necessarily a defect. Rather, this is (again) a matter of methodological preference. The validity of this methodol ogical procedure cannot be decided on a priori theoretical grounds, but can only be decided on the basis of its comparative explanatory power.
b. Geert’s alternative concept of abstract labor
There seem to be subtle differences between the articulation of the concept of abstract labor in (1989) and (1993), so I will consider them separately, beginning with (1989). In (1989), the concept of abstract labor is defined, not in units of labor-time (as in equations (1) and (2) above), but instead in units of money. More specifically, the abstract labor contained in a particular commodity is defined as the product of the quantity of private labor contained in the commo dity (li) and the "monetary expression of labor" for this particular private labor (mi), the latter of which is in units of money per labor-hour. The quantity of private labor (li) appears to be equal to the average of the actual quantities of concrete labor in the different plants in the given industry. mi refers to the realized monetary expression, as opposed to the ideal monetary expression of labor (I am ignoring here "ideal prices" and the ide al monetary expression of labor.)
As Geert express this definition of abstract labor:
... mi li is the expression for abstract labor ... (p. 95)
The product mi li is itself assumed to be equal to the value added component of the price of the commodity (yi); i.e.
(3) yi = mi li
The exact meaning of this equation is not entirely clear to me. To represent a meaningful theory, two of the three variables must be determined independently of this equation and one determined by this equation. Of the three variables in equation ( 3), which two are determined independently and which one is determined by this equation?
As mentioned above, li seems to be the average quantity of concrete labor required to produce a given commodity. This average quantity of labor seems to be determined independently of equation (3), in principle by counting the actual num ber of concrete labor hours within capitalist firms and then calculating the average for each industry.
Of the two remaining variables, yi and mi, which one is determined by equation (3) and which one is determined independently of it? Geert’s answer to this question is not entirely clear to me. I could find no clear explicit discussion of the quanti tative determination of the yi’s (value added) and the mi’s (monetary expression of labor). But it seems to me that the yi’s are implicitly determined independently of this equation and the mi’s are determined by th is equation. If the yi’s were determined by this equation and the mi’s independently of it, then the determination of the mi’s would have to be explained in some way. Otherwise, Geert’s interpretation would be subject to the same cri ticism that he makes of Marx’s reduction of concrete labor to abstract labor: it could not be "operationalised." Marx’s "reduction problem" would not be overcome, it would just be transferred from labor to the monetary expression of labor. But no explanat ion of the determination of the mi’s is presented. Therefore, it seems to me that, from Geert’s perspective, the mi’s are not determined independently of equation (3), but are instead determined by this equation.
However, if the mi’s are determined by this equation, then the yi’s cannot be determined by this equation. Instead, the yi’s appear to be taken as given, as the value added component of actual market prices in the r eal capitalist economy. Therefore, the consequence of Geert’s rejection of Marx’s concept of abstract labor as the substance of value (as determined independently of prices) seems to be that this interpretation presents no quantitative theory of th e determination of the magnitude of value added, or of the total price of commodities. The value added is simply taken as given, not explained. Instead, what we appear to have is a theory of the magnitude of the monetary expression of labor (mi ) (of the "value productivity of different kinds of concrete labor), which depends on the given value added (yi) and the observable concrete
Since there is no theory of value added, there is also no quantitative theory of surplus-value, or profit. In one brief passage, surplus-value is equated with the difference between value added and wages, as follows (p. 104):
(4) s = y - W
= ml - wl = (m-w) l
But since the magnitude of value added (y) in this equation is taken as given rather than explained, so also in effect is the magnitude of surplus-value. Value added is determined outside this theory, and therefore so is surplus-value.
Furthermore, the important conclusions of Marx’s theory of surplus-value - e.g. the conflict over the length of the working day, the conflict over the intensity of labor, the tendency toward technological change, the falling rate of profit - follow from Marx’s premise that the quantity of value added is determined by the quantity of abstract labor, as in equation (2). These conclusions do not follow from the "value-form" interpretation of Marx’s theory because this interpretation rejects Marx’s the ory of value added, based on abstract labor as the substance of value. For example, the conflict over the length of the working day follows from the assumption that the quantity of surplus-value produced in a given working day depends in part on the quant ity of value added produced in that working day, which in turn depends on the length of the working day, or on the quantity of hours in the working day. It follows from this premise that in order to increase the quantity of value added produced in a day, capitalists must endeavor to increase the quantity of labor hours in the working day (or to increase the intensity of labor). Hence the conflict over the length of the working day (and over the intensity of labor). This conclusion does not follow from the "value-form" interpretation, because in that interpretation the quantity of value added produced in a given day does not depend on the quantity of labor hours in that day. Instead, the quantity of value added is taken as given, i.e. is not explained and in particular is not explained by the quantity of labor hours.
Similar arguments apply to the tendency toward technological change and the falling rate of profit. With respect to the falling rate of profit, Marx’s main argument is that increases in the surplus-value produced per worker are less and less able t o offset the negative effects of an increase in the constant capital invested per worker because the quantity of surplus-value produced in a given day is determined by, and hence limited by, the quantity of labor hours in that day. Once most of the working day is already surplus labor, then it becomes harder and harder to increase surplus labor still further, and hence to increase the surplus-value produced per worker still further. This conclusion follows only from Marx’s quantitative theory of va lue and surplus-value, which assumes that the quantity of value added depends on the quantity of abstract labor hours, as in equation (2).
Hence the explanatory power of Marx’s theory of surplus-value is very greatly reduced if Marx’s concept of abstract labor and the assumption that the quantity of valued added is determined by the quantity of hours of abstract labor (i.e. equation 2 ) are rejected. This seems to me to be a very high cost. Perhaps, Marx’s concept of abstract labor, even with all its alleged problems and limitations, should be reconsidered.
In (1993), abstract labor appears first to be defined differently - in units of labor-hours, rather than in units of money:
The concrete labor Li and Lj cannot be added up as particulars. However, because of the dialectical contradiction of the particular simultaneously being universal labor, the labor time (I) and (j) can in principle be added up as l abor hours in the abstract (l): li + lj = l, even though we can not add them up as particulars...
At this level, li + lj is fairly empty; it is indeed an abstract statement; a statement at the abstract level of theory ... (107)
In terms of quantities, the li’s appear to be equal to the Li’s, i.e. equal to the actual quantities of labor-time. The only difference is that the Li’s consider these hours of labor as "concrete labor" and the l i’s consider these same quantities of labor-hours as "abstract labor." There seems to be no quantitative difference between the two.
Then, as a result of market exchange, abstract labor li is "concretized" as mli, as in (1989)
(in this passage, m is used instead of mi):
In the market, value actually gets shape in its expression in terms of money. If we restrict ourselves to the output, and to the value-added component, we may write m for the monetary expression of labor, as it is actually realized in the market. Thus m is the value productivity of labor. In the market then, the contradiction
Li X l is transcended at a more concrete level into what I provisionally call mLi (but should call mli, as indicated below. mLi is a sum of money ... In the market, labor actually takes the value-form. This labor is actually converted (transformed) int o an abstract entity. It is actual and capitalist abstract labor, which is capitalist value...
Perhaps a somewhat subtle differentiation (that is, for the current purposes) is that in the market the Li X l contradiction is transcended by way of positing it, more concretely, in the abstract moment (l) of the contradiction. Thus we shou ld write mli. (pp. 107-08)
Thus, abstract labor appears to be defined as both l and ml. Perhaps l is abstract labor "in the abstract" and ml is abstract labor "in the concrete."
ml is then equated with value added (y), and we are back to equation (3). But the problems associated with this equation remain. This equation does not provide a theory of the determination of value added, as one component of the price of commoditi es. It particular, it is not assumed that the quantity of value added is determined by the quantity of abstract labor.
Instead, value added is simply taken as given, and used in the determination of m (the monetary expression of labor). Hence, Geert’s elaboration of abstract labor "in the abstract" and abstract labor "in the concrete" does not change anything fundament al in the nature of equation (3). Therefore, we arrive at the same conclusion as before: Geert’s rejection of Marx’s concept of abstract labor as the substance of value results in the loss of a theory of price, and hence also the loss of a theory of surpl us-value. This rejection also loses much of the explanatory power of Marx’s theory of surplus-value, since the most important conclusions of Marx’s theory no longer follow from this interpretation, as discussed above.
Likitkijsomboon’s (1995) main criticism of Reuten and Williams (1989), and of the "value-form" interpretation in general, is the essentially same criticism that I have made above: that, by abandoning Marx’s concept of abstract labor as the substanc e of value, the "value-form" interpretation provides no quantitative theory of price.
The modern attempts to "reconstruct" Marx’s value-form theory render themselves one-sides and logically impoverished by denouncing the quantitative aspect, and by collapsing Marx’s multi-dimensional categories of value-form and price into a mystified c oncept of pure form without content and abstract money imposed from without. (p. 74)
All this, from Rubin to Reuten, is in effect a process of conceptual collapse of abstract labor into social labor, the magnitude of labor-value into its realized value in exchange (price), and value into exchange-value, which finally degenerates into t he rejection of abstract labor, labor-value and labor-exchange ratios, leaving in place only exchange-value, the value-form, price-form and "money," all without quantitative grounds. (p. 93)
However, Likitkijsomboon’s criticism is not developed thoroughly enough. In particular, he does not examine in detail the logic of the relation between labor, the monetary expression of labor, and value added, as in equation (3) above.
Geert (1995) responds to Likitkijsomboon’s criticism as follows:
Likitkijsomboon correctly feels that value-form theory’s first concern is a qualitative rather than a quantitative one...
This is not to say that, subsequently, quantification has to be outlawed from such as approach. In Reuten and Williams (1989: ch 1-2) we have set out a value-form theoretic development from the abstract-labor theory of value. In a short-hand notation, (see p. 95 and 104) it states Y = ml (where Y is value added, m is the monetary expression of labor, and l social aggregate labor), thus m is the value productivity of labor; surplus value (s) [is given by] s = (m-w)l (where w is the wage rate).
However, we have seen above that these equations, although they are quantitative relations, do not provide a quantitative theory of value added and surplus-value. Value added is taken as given rather than explained, and hence so also is surplus-val ue. Likitkijsomboon’s criticism is essentially correct, just not detailed and thorough enough. (Likitkijsomboon’s critique also does not consider Reuten (1993) - which is partly my fault since I was a RRPE referee for this paper!)
a. Criticism of Marx
Chris’ main criticism of Marx’s theory of value in (1993) is that Marx did not prove the necessity of value as the common property of commodities that determines their exchange-values. Another criticism is that, even if one accepts the e xistence of value (as defined above), Marx did not prove the necessity that labor must be the substance of value. In particular, Chris argues (following a long line from Boehm-Bawerk on) that Marx arbitrarily excluded commodities which are not prod ucts of labor (such as land) from his logical deduction.
To begin with, Chris argues that Marx did not prove the necessity of value:
This is probably the most important turning point in the whole theory of value, and is by far the most controversial... Marx skips over this transition astonishingly quickly. He simply declares that in the exchange relation of iron and corn there is a quantitative identity... Critics have been vociferous in denying the necessity the necessity of any such inference [Bailey, Boehm-Bawerk, Cutler, et al., Moore, etc.]...
It seems to me that this argument [i.e. Bailey, etc.’s argument] has much more force than most Marxists allow. At this point we only have the postulate of identity in essence and of common measure. (1993, pp. 75-76)
Chris’s criticism here seems to be that Marx did not provide a sufficient argument for assuming that commodities must contain a common property that determines their exchange-values. But Marx did provide and argument, which Chris here ignores: that the exchange of commodities is the exchange of equivalents. From the exchange of equivalents, it follows of necessity that commodities must contain a common property in terms of which their equivalence is determined. As Marx quoted Aristotle: "The re can be no exchange without equality and no equality without commensurability." (C.I. 151) To recall the key paragraph in which this argument is made:
Let us now take two commodities, for example corn and iron. Whatever their exchange relation might be, it can always be expressed as equation in which a given quantity of corn is equated to some quantity of iron, for instance 1 quarter of corn = x cwt of iron. What does this equation signify? It signifies that a common element of identical magnitude exists in two different things, in the 1 quarter of corn and similarly in x cwt of iron. (C.I. 127; emphases added)
This point is Marx’s main critique of Bailey in the Theories of Surplus-Value - that Bailey did not recognize that the exchange of equivalents requires a common property in terms of which their equivalence is measured and determined ( TSV.III. 124-47).
In order that two quantities of different use-values can be equated as equivalents, it is already presumed that they are equal to a third thing, that they are qualitatively equal and only constitute different quantitive expressions of this qualitative equality. (TSV. 134; first emphasis added)
[Bailey] even forgets the simply consideration that if y yards of linen equal x lbs. of straw, this [implies] a parity between the two unequal things ... making them equal magnitudes... It is not was straw and linen that they are equated, but as equivalents. The one side of the equation must therefore express the same value as the other. The value of straw and linen must therefore be neither straw nor linen, but something common to both ... (TSV. III. 139; emphases added)
But although the commodity has a thousand different kinds of value, or a thousand different prices, as many kinds of value as there are commodities in existence, all these thousand expressions always express the same value. The best proof of thi s is that all these different expressions are equivalents which not only replace one another in this expression, but do replace one another in exchange itself. (TSV.III. 147)
Chris’s other criticism of Marx is that, even if the postulate of the value of commodities is accepted, Marx did not prove that labor must be the substance of value, and in particular Marx arbitrarily excluded commodities which are not produ cts of labor (like land) from his deduction.
I differ here from Marx in that I refuse to find it necessary to come to labor until after conceptualizing capital as a form-determination. Bringing in labor too early runs the risk of model building and committing the exposition to a stage of simple c ommodity production...
To begin with, we analyze the commodity-form itself, and only at the end give grounds for picking out as systematically important those commodities that are products of labor... [W]e avoid an appearance of arbitrariness in concentrating from the outset only on products of labor. (1993, pp. 69-70)
My response to this second criticism it ignores key elements of Marx’s overall logical method. According to Marx’s method, the price of land is explained by Marx on the basis of the rent of land, i.e. on basis of the future expectations of rent or "capitalized" rent. Rent itself is explained in Part 6 of Volume 3 as one part of the total amount of surplus-value (along with other parts of surplus-value such as interest, commercial profit, etc.). According to Marx’s logical method (as I have emphasiz ed in several recent papers), the total amount of surplus-value is determined prior to its division into the individual parts of rent, interest, etc. The determination of the total amount of surplus-value is the main subject of Volume 1 of Capital (the analysis of capital in general). The individual parts of surplus-value (or the distribution of surplus-value) are then explained in Volume 3 (an abstract analysis of competition) on the basis of the assumption that the total amount of surplus-value h as already been determined. Therefore, according to Marx’s method, the price of land cannot be explained in Chapter 1 of Volume 1. Instead, land and the price of land are "abstracted from" in Volume 1 (just as are the other individual parts of the total s urplus-value), and then explained at a more concrete level of analysis in Volume 3. These important aspects of Marx’s logical method were overlooked by Boehm-Bawerk, and by almost everybody else since, apparently including Chris.
Now, one could still argue that, even if Marx’s treatment of the price of land is accepted as logically legitimate, Marx’s argument does not prove the necessity that labor must be the substance of value. As Boehm-Bawerk argued, the substance of value could also be utility or scarcity. My response to this argument is that, yes perhaps it could be argued that e.g. utility could be the common property of commodities that determines their exchange-values, although one would have to explain how t hese concepts could be made homogeneous across different commodities. The neoclassical subjective concept of utility could not serve as a common objective property of commodities, because one would have to explain how subjective utility could be made homo genous across individuals - which neoclassical economics tried to do in its early days and failed.
Then we would have two rival hypotheses concerning the value of commodities. I would suggest that the appropriate way to determine the relative validity of these two hypotheses is to evaluate their relative empirical explanatory power (as I have discus sed in the first section of this paper). On the basis of which hypothesis can one explain a greater range of important phenomena of the capitalist economy? I would argue that there can be no doubt about the conclusion of such an empirical appraisal. One c an explain a much wider range of important phenomena of capitalism (especially related to the dynamic development of capitalism over time) on the basis of Marx’s assumption that the substance of the value of commodities is abstract labor than one could ex plain on the basis of the alternative assumption that the substance of value is utility (or scarcity).
I think that this is what Marx had in mind is his well-known letter to Kugelmann in which Marx responded to a similar criticism by a reviewer in the Centralblatt, that in Chapter 1 Marx had failed to prove the logical necessity of labor as t he substance of value, by saying that this criticism is based on ignorance of the scientific method:
As for the Centralblatt, the man is making the greatest possible concession in admitting that, if one means anything at all by value, the conclusions I draw must be accepted. The unfortunate fellow does not see that, even if there were not chapt er on "value" in my book, the analysis of the real relations which I have would contain the proof and demonstration of the real value relations. All that palaver about the necessity of proving the concept of value comes from complete ignorance of t he subject with and of scientific method. (SC. 196)
In other words, even if no argument whatsoever were given in Chapter 1 to support the hypothesis that abstract labor is the substance of value, and this hypothesis were simply assumed as the starting point, the explanatory power demonstrated in the res t of Capital provides the best proof of this hypothesis. (If anyone knows how to get a copy of this review in the Centralblatt, please let me know. I have made a few efforts to obtain a copy, but without success. I did learn somewhere that t he full name is Literarisches Centralblatt fur Deutscheland. The issue is No. 28,4 for July 1868. It is signed "U".)
b. Chris’ alternative interpretation
We will now examine Chris’ alternative derivation of the necessity of value as the common property of commodities that determines their exchange values and of the necessity of labor as the substance of value.
Chris’s alternative derivation of value is not entirely clear to me, but it seems to me that Chris’s alternative derivation is actually very similar to Marx’s derivation. Both rely on the crucial point that the exchange of commodities is the exchange of equivalents. From this key point, the necessity of value as a common property (i.e. commensurability) follows in both derivations. The only difference is that the exchange of equivalents is itself established in different ways: in Chri s’ derivation through the actual exchange process and in Marx’s derivation through an abstract analysis of the nature of commodities as general equivalents. According to Chris:
It is the very process of commensuration that posits commodities as value masses in the first place. The actuality of value and its expression or measure develop together at the same time.
The exchange relation has to be grasped as simultaneously constitutive of value and serving as its expression. (1993, p. 78)
But I do not see how Chris’s derivation is superior to Marx’s. Chris’ criticism of Marx’s derivation is that Marx did not provide sufficient argument to justify the assumption of a common property. But Marx did provide such an argument - the exchan ge of equivalents. This is the same argument that Chris provides, except the exchange of equivalents itself is derived in a different way.
Chris’ alternative derivation of labor as the substance of value (as I understand it) can be summarized briefly as follows: The necessity of labor is derived from the capital-form as self-expanding value. In order to expand in value, capital must produce use-values. And in order to produce use-values, capital requires labor. Therefore, labor must be the substance of value. In Chris’ words:
What then is the condition next required to grant necessity to the existence of capital as self-valorization? ... It is here that we remember that at the outset we stated that a primary condition of exchange is the world of use-values. With capi tal we reach of form of circulation of commodities that is its own end, but the self-valorization process still rests for its possibility on the emergence into being of the goods themselves from some external source...
The problem is solved if the goods are themselves produced by capital and reduced to moments in its own circuit...
The activity of production is an activity of labor. Hence, capital must make that activity its own activity. Only now does the presentation find it necessary to address labor. The limitlessness of accumulation inherent in the form of capital is
given a solid ground in productive labor. (1993, pp. 84-85; emphasis added)
Furthermore, the uniqueness of labor does not does follow from a consideration of use-values. Labor is indeed necessary to produce use-values, but so are means of production and, in some cases, land. In the production of use-values, there is nothin g unique about labor. Therefore, Chris’ conclusion, that labor must be the sole substance of value, does not follow. One could argue, in exactly the same way, that because means of production (or land) are necessary to produce use-values, means of product ion (or land) are also a substance of value.
Chris also provides an argument for excluding land (which has a price, but no value) from Marx’s analysis in Volume 1.
To be self-grounded, value must be produced by value. This means that only those goods produced by capital itself counts as value, as true commodities in both form and content...
Our presentation has reached the point at which nonproduced commodities are seen to retain the value-form, but only the semblance of value; they are lacking in the substance of value because they do not originate within the value circuit itself as it i s driven by valorization. They play no essential role in the dynamic of capitalist development (although two such nonproduced commodities - labor power and land - are materially essential inputs, but cannot be treated in this chapter).
(1993, p. 85)
There is an appeal here to a philosophical principle ("to be self-grounded ...") that I do not understand (which is my own limitation). And there is an additional argument ("plays no essential role ...") that is hedged in parentheses and this hedge is excluded from further discussion. But at the very least, we can see that Chris’s rationale for excluding land in Volume 1 is completely different from Marx’s rationale (the determination of the total amount of surplus-value prior to the determination of the individual parts of surplus-value, such as the rent of land), and also that Chris’s rationale provides no basis for the later determination of the price of land, as does Marx’s rationale.
Finally, I would like to raise the following further questions about Chris’ alternative derivations of value as the common property of commodities and labor as the substance of value. What are the implications of this interpretation for the rest of Marx’s theory, and in particular for the theory of value added and surplus-value, as discussed in the first section of this paper in relation to Geert’s interpretation? Is the theory of value added implied by Chris’ interpretation essentially the same as equation (2) above, i.e. the same as Marx’s theory, according to which abstract labor (L) and the monetary expression of value (m) are assumed to exist independently of value added and to determine value added? If not, what does it mean to say tha t labor is the "substance of value"? And how is value added determined?
None of these questions are explicitly addressed by Chris in this paper. This is probably an indication of less interest in these quantitative questions. But a theory of value and surplus-value must address these quantitative questions. My guess is that Chris’ answer to these questions would be more or less the same as Geert’s answers. But then, in my view, Chris’ interpretation would be subject to the same objections as Geert’s interpretation, which I have discussed above. The main objection is th at it provides no theory of value added, but rather takes value as given, which in turn implies that it provides no theory of surplus-value, the most important conclusion of Marx’s theory.
I look forward to discussing these issues in a few weeks in Tepoztlan.
Arthur, Chris (1993). "Hegel’s Logic and Marx’s Capital," in Moseley (1993).
Likitkijsomboon, Pichit (1995). "Marxian Theories of Value-Form". Review of Radical Political
Economics. 27(2): 73-105.
Miller, Richard (1987). Fact and Method: Explanation, Confirmation and Reality in the Natural and Social Sciences. Princeton, NJ: Princeton University Press.
Moseley, Fred (ed.) (1993). Marx’s Method in Capital: A Reexamination. Atlantic Highlands
NJ: Humanities Press.
(1994). "Explanatory Progress or Retrogression in Economics" A Marxian Response to Hausman’s Appraisal," paper presented at the 1994 ISMT conference.
(1995). "Marx's Economic Theory: True or False? A Marxian Response to Blaug's Appraisal," in Moseley (ed.), Heterodox Economic Theories: True or False?, 1994.
Brookfield VT: Edward Elgar Publishing.
Reuten, Geert (1993). "The Difficult Labor of a Theory of Social Value: Metaphors and
Systematic Dialectics at the Beginning of Marx’s Capital," in Moseley (1993).
Reuten, Geert and Michael Williams (1989). Value-Form and the State: The Tendencies of
Accumulation and the Determination of Economic Policy in Capitalist Society. London:
Sayer, Derek (1979). Marx’s Method. Sussex: Harvester’s Press.