In response to:

On Your Marx from the December 20, 1979 issue

To the Editors:

I am grateful for Peter Singer’s generous review of my Karl Marx’s Theory of History (NYR, December 20), but I reject his charge that I participate in a “popular misconception” about Darwin’s theory of evolution. My chapters on functional explanation, in which I exploit Darwin on Marx’s behalf, defend controversial claims, and Peter Singer is entitled to be unpersuaded. But his criticisms of what I say about Darwin are off the mark.

Of course I know that the proposition that “long necks help giraffes survive” was not original to Darwin. And, as can be gathered from pages 271 and 285-286 of my book, I substantially agree that “Darwin’s distinctive insight was into the mechanism by which the forces of evolution operate.” But my argument, bypassed by Singer, and presented at some length, was that Darwin’s insight does not upset, but rather illuminates, the fact that species have the valuable features they do because they are valuable: and that is a functional explanation, since an effect of a feature, its value to the species, is used to explain the feature.

So I do not agree that I misconstrue Darwin. But I must also point out that Singer exaggerates when he says that, according to me, “explanation in Marx takes a form similar to explanation in Darwin.” As just indicated, I hold that Darwin discovered the mechanism required by functional explanations which were provided before he discovered it, and which were not refuted, but vindicated, by his discovery. I invoke Darwin to show the respectability of functional explanation as part of my defense of historical materialism, since it posits functional explanations. I do not, however, claim that Marx, like Darwin, described the mechanism(s) involved in the functional explanations he hypothesized, and my own discussion of mechanisms in Chapter X is an attempt to fill this gap in classical historical materialism. But the incompleteness of Marx’s explanations does not mean, as Singer would have it, that Marx’s theory was teleological in a nonscientific sense. Darwin thought organisms have the useful features they do because they are useful before he discerned the relevant mechanism, but he was not therefore, at that earlier stage, a party to unscientific belief in design or purpose.

It would therefore not follow from his failure to specify mechanisms that Marx believed that history was subject to a transcendent controlling purpose. He might, nevertheless, have believed just that. This brings me to Singer’s contention that in the body of my book I “blithely ignore” what he takes to be the “import” of its first chapter: that Hegel and Marx had similarly “teleological” conceptions of history. But I said at the end of Chapter I that Marx’s theory, though rooted in Hegel’s, was an attempt to replace a speculative conception by a scientific one. I agree that there are traces of the theory’s Hegelian origin in Marx’s mature writings, particularly in those he did not publish. But my stated aim was to present historical materialism as scientifically as possible, and I therefore, legitimately, ignored the Hegelian residue.

Singer-like doubts about my treatment of functional explanation occupy a good part of a review of my book by Jon Elster, forthcoming in Political Studies in 1980. I reply to Elster in the same number, and those who think my use of functional explanation on behalf of historical materialism questionable will find an extended defense of it there.

G. A. Cohen

Reader in Philosophy

University College, London

Peter Singer replies:

On rereading the pages to which Cohen refers, I withdraw the charge that he participates in a popular misconception about Darwin’s theory. His error is more sophisticated than the simple belief that giraffes have long necks because it enables them to survive, though it is related to this popular error. He thinks that—as his letter puts it—“value to the species” performs an essential role in a Darwinian explanation of a feature like the length of giraffes’ necks. I maintained in my review that a Darwinian explanation eliminates the need to talk of the function of long necks. I could have added that it eliminates the need to talk of “value to the species” as well. All it requires is that we talk about the effect that variations in the lengths of the necks of earlier generations of giraffes had on their ability to survive and reproduce.

To get back to Marx, I did not regard my review as establishing that Marx’s theory of history is teleological in a nonscientific sense. I merely suggested this as a possible reason why Cohen’s extraordinarily able attempt to present historical materialism as scientifically as possible seemed, in the end, not entirely convincing. Obviously the issue needs further discussion and I am glad to see that Cohen will be replying to doubts like mine at greater length in the journal he cites.

This Issue

February 21, 1980