23. Does the Identity Theory Escape the Fate of Epiphenomenalism?

Before starting to criticize the identity theory, let me make it quite clear that I regard it as a perfectly consistent theory of the relation of mind and body. In my opinion the theory may therefore be true.

What I regard as inconsistent is a wider and stronger theory: the materialist view of the world, which involves Darwinism, and which together with the identity theory leads to a contradiction — the same as in the case of epiphenomenalism. My thesis is that the identity theory combined with Darwinism faces the same difficulties as epiphenomenalism.

Admittedly, the identity theory is somewhat different from epiphenomenalism; especially from an intuitive point of view. From this point of

view it seems not so much a form of psychophysical parallelism,²⁴ but, rather, close to dualistic interactionism. For in view of

(3) World 1_m = World 2 we have, by (4), that World 1 interacts with World 2. Moreover, the reality of World 1_m (= World 2) and its efficacy could not be stressed more strongly. All this takes World l_m (= World 2) far away from epiphenomenalism.

Moreover, the identity theory has the great advantage over epiphenomenalism that it gives a kind of explanation — and an intuitively satisfying explanation — of the nature of the link between World 1_m and World 2. In parallelistic epiphenomenalism, this link has just to be accepted as one of the ultimate inexplicables of the world — as a Leibnizian pre-established harmony. In the identity theory (whether we take or do not take the term "identity" quite seriously) the link is satisfying. (It is at least as satisfying as that between the inside view and the outside view of reality in Spinozism.)

All this seems to distinguish the identity theory sharply from epiphenomenalism. Nevertheless, the identity theory is as unsatisfactory as epiphenomenalism from a Darwinian point of view. But we (and especially the materialists among us) need Darwinism as the only known explanation of the emergence of purposeful behaviour in a purely material or physical world, or even in a world which at some stage of its evolution was confined to World 1_p (so that at this stage World 1_m = World 2 = 0).

Thus it is my thesis that my critical remarks about epiphenomenalism apply here too, mutatis mutandis, though admittedly with less intuitive force.

For the identity theory is, by intention, a purely physicalistic theory. Its fundamental principle is still the principle of the closedness of World 1; which leads to the lemma that (causal) explanation, so far as it is knowledge by description, must be in terms of strictly physical theory. This allows us (perhaps) to accept the emergence of a new World 1_m; but it does not allow us to explain that the characterizing feature of this World 1_m is that it consists of mental processes, or that it is closely linked with mental processes.

On the other hand, we must demand that all major novelties emerging under the pressure of natural selection must be explained entirely within World 1.

To put it in another way, the World 2 of the identity theory remains with respect to the Darwinian point of view logically in exactly the same situation as the epiphenomenal World 2. For although it is causally effective, this fact is irrelevant when it comes to explaining any causal action of World 2 upon World 1. This has to be done entirely in terms of the closed World 1.

The real thing, the thing in itself, and causality known by acquaintance — all this remains, from the point of view of the physicalistic principle and of knowledge by description, outside of the physical explanation, and, indeed, of what is physically explicable.

The principle of the closedness of World 1 demands that we still explain, truthfully, my going to the dentist in purely physical terms. But if so, the fact that World 1_m is identical with World 2 — the world of my pains, my aim to get rid of them, and my knowledge about the dentist — remains causally redundant. And this is not changed by the assertion that another causal explanation, a World 2 one, is also true: it is not needed; the world works without it. But Darwinism explains the emergence of things or processes only if they make a difference. The identity theory adds a new aspect to the closed physical world, but it cannot explain that this aspect is of advantage in the struggles and pressures of World 1.² For it can explain this only if the purely physical World 1 contains these advantages. But if so, then World 2 is redundant.

Thus the identity theory, contrary to its intuitive character, is logically in the same boat as a parallelist theory that employs the physicalist principle of the closedness of World 1.

^{24 2} Jeremy Shearmur has drawn my attention to a very similar argument in Beloff [1965],