Bill is of course right in principal; you can refute a theory with empirical evidence. But empirical refutation is messy and inconclusive. For instance, the typical "refutation" of Marx's prediction of increasing social polarization was always that he didn't foresee the trade unions that would guarantee workers' stake in the current system. For obvious reasons, I haven't heard this for awhile.
In contrast, proofs of logical inconsistency (if valid or even if deemed to be valid) are neat and conclusive. "[O]n _a priori_ grounds, something which is internally inconsistent is thereby false. Nothing, no empirical evidence or theoretical argument, no matter how voluminous or persuasive, can ever make it true. There is no need to listen to it. Dialogue immediately shuts down." (I'm quoting from a recent, short and non-technical -- well, as non-technical as possible -- paper of mine, "The Significance of the 'Internal Inconsistency' Allegations," that I've copied below.)
It is because they desire neat, conclusive refutations that Marx's critics focus on what seem to be inessential technical minutiae of his value theory, rather than big things of obvious importance. Marx's value theory is the foundation of so much of _Capital_. If you can kick away the foundation, the whole house collapses.
Jim Baird led this all off by asking: has there ever been a serious attempt to refute Marxian economics by a "mainstream" economist? The only thing I have been able to find in my (admittedly scant) reading has been of this character - just assume it away, call it "discredited", without ever telling the reader what it is.
Most authors do just what Jim said, but it isn't a dismissive ploy or cowardice. They're just not reinventing the wheel. There's a century of "proofs" of internal inconsistency that are accepted nearly universally. When they say it is "discredited," they're often implicitly invoking these "proofs."
I think Jim's first question is not quite the right one. It presumes a division different from the actual one. There's no hard and fast dividing line in this realm between "mainstream" and "Marxian" economics. It is true that "mainstream" economists haven't been so busy "refuting" Marx's work, but that's because they're happy to leave this job to Marxist and other heterodox economists, and to refer to their conclusions if need be. And the latter have made quite a research program out of "refuting" Marx on grounds of logical inconsistency, particularly in the past 25 years. The actual division is between Marx and his (Marxist, non-Marxist, and anti-Marxist) critics.
The only main difference in this context between the mainstream (and Austrian) economists, and the Marxist and other heterodox critics of Marx, is that the former tend to use the "refutations" to say Marx was all wrong, whereas the latter try to say that their "corrected" or "completed" versions of Marx's theory vindicate the spirit and main thrust of it. But such claims should be taken with a grain of salt -- hell, a pillar of salt. The theoretical conclusions of their revised versions are, at key points, contrary to Marx's own theoretical conclusions. For instance, he concludes from his theory that the profit rate tends to fall as a result of the adoption of more mechanized, higher productivity technology, while they conclude from theirs' that this tends to raise the profit rate.
I agree with Michael when he writes that the "refutations" set up a straw man and knock it down (see the paper below). That's why they've been so easy to demolish -- you just need to interpret the text differently. But that doesn't mean that they are "not very effective." The "refutations" have been so effective that almost all economists -- Marxist, non-Marxist, or anti-Marxist -- have accepted and still accept them. The fact of the matter, particularly in the case of the Marxist and other left economists, is that the straw men are not mere concoctions -- they flow out of how these folks interpret the world and Marx, the questions they ask, and so forth. This is why they are so resistant to acknowledging that the "proofs" of internal inconsistency have been refuted -- the Marx that emerges as internally coherent is not one they can stomach.
As for Chris's post, I think a lot has changed in 3/4 of a century. I don't think Rubin's point is applicable to the debate today.
*******
THE SIGNIFICANCE OF THE "INTERNAL INCONSISTENCY" ALLEGATIONS
Andrew Kliman, Department of Social Sciences, Pace University, Pleasantville, NY 10570 USA. (914) 773-3951. Andrew_Kliman at msn.com
Presented February 28, 1998, in the International Working Group on Value Theory miniconference at the Eastern Economic Association Conference, New York, NY
For more than 100 years, academic economists' discussion of Marx's own value theory, including his law of the tendential fall in the profit rate, has had one single central focus. They have attempted to prove that it is internally inconsistent. Until fairly recently, the validity of the "proofs" of internally inconsistency has gone almost unquestioned. In many quarters, this remains the case.
For instance, the fact that these "proofs" have been refuted has unfortunately not yet made its way into the broader public discourse surrounding Marx. Only four months ago, The New Yorker published a piece by John Cassidy (1997) which was so appreciative of Marx that the overline declares him to be "the next thinker." When it comes to Marx's value theory, however, the myth of internal inconsistency is so ubiquitous and hegemonic that Cassidy disposes of the issue with a single comment: Marx's "model of the economy, which depended on the idea that labor is the source of all value, was riven with internal inconsistencies, and is rarely studied these days" (Cassidy 1997:252).
During the last 16 years or so, however, a new interpretation of the quantitative dimension of Marx's value theory has emerged, an interpretation now known as the "temporal single-system" or "TSS" interpretation. This interpretation has refuted every single one of the alleged "proofs" of internal inconsistency. This claim cannot be substantiated in the space of 20 minutes, of course, so I will not attempt to do so. Instead, I refer you to the now-voluminous and still-growing body of TSS research (see, e.g., Freeman and Carchedi (eds.), 1996; Kliman and McGlone, 1997).
It may be in order, however, simply to list some key theoretical results of this interpretation. In each case, they replicate one of Marx's results and therefore refute a long-standing charge of internal inconsistency:
(1) Rising productivity combined with a rising organic composition of capital themselves produce a tendency for the profit rate to fall. (2) The level of the profit rate is invariant to the distribution of profit. (3) Even when value is redistributed across sectors, the profit rate is s/(c+v), surplus-value divided by the constant and variable capital advanced. (4) The level of the profit rate is influenced by production conditions in luxury industries. (5) Total price equals total value. (6) Total profit equals total surplus-value. (7) Values cannot be negative or zero, even under joint production. (8) Positive surplus-value is both necessary and sufficient for positive profit, even under joint production.
No one has successfully challenged our ability to obtain these results, and I do not believe the fact that we obtain them is in dispute any longer. So, what is our "trick"? How have we accomplished the seemingly impossible?
The answer is simple. None of Marx's critics ever tested what he actually wrote to see if it was internally consistent or not. What has passed, and continues to pass, for "proof" of internal inconsistency is in fact a type of "bait and switch." It invariably takes the following form. Marx says X and Marx says Y. Transform X into X' and transform Y into Y'. Prove that X' and Y' are incompatible. Then conclude that Marx was guilty of internal inconsistency.
This is the real "transformation problem."
What helps make this pass for proof is that X and Y are often transformed into X' and Y' by translating them into mathematical notation. Terms in natural language which might admit of different interpretations are thus given univocal meanings. For instance, "price" becomes a vector p, and "value" becomes a vector v. p is always p. v is always v. People, especially those who manipulate the symbols — and they have been the ones to allege internal inconsistency the loudest and most persistently — get accustomed to substituting the symbols for the concepts. It can take a long time to discover that they're not the same things. In fact, it took almost 80 years after Bortkiewicz's (1952 [1907]) canonical "proof" of Marx's inconsistency until a few people began to realize two things. First, although p is always p and v is always v, "price" isn't always "price" and "value" isn't always "value." Input prices differ from output prices, and input values differ from output values. Second, whether or not p = v, the price paid to acquire an input is the sum of value advanced for it.
These two points are trivially simple and, in retrospect, they are self-evident. Taken together, they are the temporal single-system interpretation. And, taken together, they make sense out of what had heretofore seemed incoherent in Marx's value theory.
Given the hostility with which the TSS findings have been greeted in some quarters, however, I want to make sure that what I'm saying here will not be misconstrued. None of our results imply that Marx was "right." They imply rather that, according to the TSS interpretation, his theoretical conclusions do follow from his own premises. None of our results prove, strictly speaking, that Marx's value theory is internally consistent. They prove, rather, that the "proofs" of internal inconsistency are invalid. (There's a difference.)
None of our results make other interpretations impossible. It is still possible to defend one's interpretation against ours, and even to convince oneself that one has done so successfully. Let me point out, however, that this line of argument cannot put the genie back in the bottle. It cannot make the proofs of internal inconsistency valid or make their validity an open issue once again. The very fact that an interpretation exists, according to which Marx's value theory is internally consistent, is sufficient to demonstrate that inconsistency has not been proven. Thus, the only way to rescue the "proofs" of internal inconsistency is to demonstrate that the TSS interpretation is impossible. In the absence of such a demonstration, intellectual honesty requires that Marx's critics, both Marxists and non-Marxists, openly renounce the failed "proofs" of internal inconsistency.
Intellectual honesty also requires that one refrain from attributing to Marx theoretical propositions and results that actually contradict what he wrote. I refer here to the practice, common among Marxists especially, of portraying their theories as "corrections" of his "errors" or "completions" of something he allegedly didn't complete, when these theories in fact contradict his propositions or results in one way or another. And I include, as part of this, the practice of calling something "the" Marxist theory when it contradicts what he wrote. It is not. It may be "a" Marxist theory, but it is not the Marxist theory of Marx himself.
What has allowed these practices to gain acceptance is precisely the myth that Marx's own value theory, as he himself stated it, has been proven to be internally inconsistent. Were that true, one would have only two choices; either reject it outright, or scavenge among its remains, picking and choosing the parts one likes and discarding the rest. But it isn't true. Hence, to attribute something to Marx that contradicts what he in fact held, or to pretend that it has been proven internally inconsistent, is to suppress his own theory, to exclude it from consideration.
To understand this, and to understand why some of us are so disturbed by the way in which the debate over Marx's value theory is conducted, one must be sensitive to a crucial distinction. The distinction is between statements that something is internally inconsistent and statements that one disagrees with it, or thinks it is untrue.
When one expresses disagreement or says that something isn't true, one recognizes the right of others to have a different view of the matter, a different theory, say. It is possible that a fruitful dialogue may then ensue. When, however, one declares something to be internally inconsistent, one denies others' right to think it. That is because, purely on a priori grounds, something which is internally inconsistent is thereby false. Nothing, no empirical evidence or theoretical argument, no matter how voluminous or persuasive, can ever make it true. There is no need to listen to it. Dialogue immediately shuts down.
>From the very beginning of the internal inconsistency debate,
more than 100 years ago, up to the present, Marx's critics
(Marxists as well as non-Marxists) have been quite conscious that
the function of their allegations of internal inconsistency is to
suppress debate. Rather than trying to express their
disagreement with his value theory, they have intentionally
sought to exclude it from further consideration.
To take just one of far too many possible examples, Roemer (1981) recognized clearly that if Marx's law of the tendential fall in the profit rate is logically incoherent, then any empirical investigation into whether it holds true is unnecessary as well as pointless. The Okishio theorem does not hold that, empirically, the profit rate has failed to fall for the reasons Marx claimed, but that, logically, it is impossible for the profit rate to fall for these reasons: "if one believes Okishio's model then there is no increase possible in the organic composition of capital so great as to reduce the rate of profit. What, then, is the point of tracking the organic composition …?" (Roemer, 1981:113, emphasis added).
To see the suppressive function of the internal inconsistency myth even more clearly, let us examine a refreshingly different critique of Marx's law of the tendential fall in the profit rate, one which opens up dialogue rather than suppressing it. I refer to the critique made by Rosser (1997), in his review of the recent collection Marx and Non-equilibrium Economics. Rosser's argument is an empirical one; he points to several phenomena which, to him, suggest that Marx's law is not currently operative (he was writing before the East Asian economic crisis erupted).
Now, the difference is not merely that Roemer argues on the ground of logic, whereas Rosser argues on the ground of fact. Rather, the reason that Rosser is willing and able to discuss the empirical relevance of Marx's law at all is that he first accepts that "Kliman sets out to reestablish the 'Law' in light of the [TSS] interpretation contained within this book and does so" (Rosser, 1997, emphasis added). It is only after one accepts that the law as formulated by Marx is at least logically coherent that the question of its explanatory power can then be re-opened. For more than 100 years, and especially in the last 25, the allegations of internal inconsistency have all but prevented us from advancing to this stage. Isn't it high time that we do so?
The International Working Group on Value Theory (IWGVT) is therefore committed to re-opening "the discussion on value … without the presupposition of any established standard, tradition or source of authority regarding either value or Marx." This does not mean that "anything goes." We do "believe it is possible to assess the merits of contesting theories in debate. [The IWGVT] seeks to create an atmosphere for this debate — which does not at present exist — such that all value theories, and all readings of value theorists, may discuss on an equal footing, referring in their support neither to the evidence of authority nor the conviction of doctrine but to reasoned and logical discussion based on textual evidence for readings and factual evidence for theories" (International Working Group on Value Theory, 1997).
REFERENCES Bortkiewicz, Ladislaus von. 1952 [1907]. "Value and Price in the Marxian System," International Economic Papers 2.
Cassidy, John. 1997. "The Return of Karl Marx," The New Yorker, October 20 & 27, pp. 248ff.
Freeman, Alan and Guglielmo Carchedi, eds. 1996. Marx and Non-Equilibrium Economics (Cheltenham, UK: Edward Elgar).
International Working Group on Value Theory. 1997. "Who We Are," available from www.gre.ac.uk/~fa03.
Kliman, Andrew and Ted McGlone. 1997. "A Temporal Single-System Interpretation of Marx's Value Theory," Review of Political Economy; forthcoming.
Roemer, John E. 1981. Analytical Foundations of Marxian Economic Theory (Cambridge: Cambridge University Press).
Rosser, J. Barkley, Jr. 1997. Review of Marx and Non-Equilibrium Economics, Journal of Economic Behavior and Organization; forthcoming.