mainly on the strength of an interpretation of the principle known as the conservation of energy—an interpretation of it, however, which many of the ablest physicists disallow. The energy of the physical world, it is maintained, is a strictly invariable amount; matter, therefore, cannot act on mind, for such action would entail a decrease, nor can mind act on matter, since that would entail an increase, of this energy. In other words, the material world is held to be a “closed system”; and as all the changes within it are mass-motions, there can be none which are not the effect and equivalent of antecedent mass-motions. But now this statement must be established on physical grounds: to assume it otherwise would be openly to beg the very question at issue. For if mind does act on matter, the physical mechanism is subject to changes from without, and so often its motions are not due to antecedent motions; and this—the common-sense view—cannot, of course, be summarily dismissed as impossible or absurd. Now, energy is essentially a metrical notion, and its conservation in finite and isolated material systems has been ascertained by careful quantitative experiments. To say that the energy of the material universe is constant is only a way of expressing the generalization of this result—is tantamount, in other words, to saying that it holds of all finite isolated systems. The whole universe may perhaps be called isolated, but we do not know that it is finite. We cannot, therefore, apply metrical concepts to it; and consequently we cannot interpret the conservation of energy as meaning that the physical part of it is a closed system. But if not a closed system, then the energy of a given group of bodies may be increased or decreased without interaction between that group and other bodies—may be increased or decreased by psychophysical interaction, that is to say. And, moreover, such psychophysical interaction would not invalidate the conservation of energy, rightly understood; for that merely means that the energy of a group of bodies can be altered only from without, and this might happen whenever such interaction occurred.[1] We seem, therefore, justified for the present in rejecting psychophysical parallelism as one of the three possible modes of relating mind and matter regarded as attributes of the real. Not only are there psychological as well as biological objections which it has not yet overcome, but there are so far no physical grounds in its favour.
At this point we may again for a moment turn aside to consider a modified form of the doctrine—the so-called Conscious “Conscious Automaton” Theory. Automaton Theory, an attempt to blend the old Cartesian views concerning the minds of man and brute. According to Huxley,[2] the best known modern exponent of this theory, “our mental conditions are simply the symbols in consciousness of the changes that take place automatically in the organism.” This consciousness is supposed “to be related to the mechanism of the body simply as a collateral product of its working, and to be as completely without any power of modifying that working as the steam-whistle . . . is without influence upon the locomotive's machinery”: thus “the feeling we call volition is not the cause of a voluntary act, but the symbol of that state of the brain which is the immediate cause of that act.” In other words, physical changes are held to be independent of psychical, whereas psychical changes are declared to be their “collateral products.” They are called collateral products, or “epiphenomena,” to obviate the charge of materialism, and to conform to the interpretation of the conservation of energy that we have just discussed. Such a theory is, strictly speaking, one of parallelism no longer: rather it adopts, instead, the first of the two possibilities we have noted above as opposed to parallelism. According to it, matter is the initiating or independent variant, on whose changes mind simply follows suit. It is open to two fatal objections. First, it is methodologically unsound: its psychology is physiological in the bad sense. It regards all states of consciousness as passive, i.e. as ultimately either “feelings” or “reflexes.” Volitional activity is declared illusory; and if this be true, intellectual activity must be illusory too. But to detect illusion requires experience of reality—we only know the sham by knowing the genuine first; and even passive states could not be experienced as such save by contrast with states that are active. To the physical side, then, we naturally turn for this knowledge which we are told is not to be found on the psychical; and we do so the more readily as, according to the present theory, the physical holds the primary place. But we turn in vain; for matter is inert, and its energy only “works” by taking the line of least resistance, like water running down hill. Moreover, such activity as we are in search of could only be found here in case the physical mechanism showed signs of being intelligently directed, and that would also be evidence that psychical activity is not illusory. Is, then, the physical side after all primary? No, we reply: the assumption is epistemologically unsound. This is our second objection. The order implied in the distinction of physical phenomena and psychical epiphenomena is contrary to all experience and indefensible. A physical phenomenon is either actually perceived or possibly perceptible; otherwise it is devoid of empirical reality altogether. But objects of perception are so far psychical; that is, they belong to immediate or individual experience. Therefore we cannot regard them as independent of this experience, nor this as their collateral product, i.e. as epiphenomena. Again, the phenomenality supposed to be common to both involves, as we have already seen, a fundamental identity in the standpoint of each: they belong to the same continuous experience at different levels. And lastly, their abstract, merely quantitative, character shows that it is the concepts of physics, and not the facts of immediate experience, that are symbolic, and so to say epithetic. The attempt—either empirically or speculatively—to outflank mind by way of matter is an absurdity on a par with getting into a basket in the hope of being able to carry oneself.
These epistemological considerations may help us to deal with the prime and ultimate argument for strict parallelism. When all is said and done, it is urged, still the interaction of mind and matter remains inconceivable. But this is hardly a sufficient reason for denying what is prima facie a fact. Occasionalists, from Geulincx to Lotze, have acknowledged the same obscurity in all cases of transeunt action. Yet they did not venture to deny that sensations were interruptions in the psychical series, the “occasions” for which were only to be found in the physical; nor that purposive movements were interruptions in the physical series, the “occasions” for which were only to be found in the psychical. And surely such a position is more in harmony with experience than that of the parallelisms, who maintain that each series “goes along of itself”—a statement which, as we have repeatedly urged, contradicts psychology and assumes the physical “explanation” of life. Whereas occasionalism leaves the question of ultimate means to be dealt with by a metaphysics which will respect the facts,[3] parallelism forecloses it on the basis of a ready-made metaphysics—modern naturalism, that is to say—in which psychology as an independent science is entirely ignored. Starting with a dualism as absolute as that of Descartes—but replacing his two substances by one, enjoying the otium cum dignitate of the Unknowable—starting, too, from the physical side, no wonder such a philosophy finds that what is for us the most familiar and of the supremest interest, the concrete world of sense and striving, is for it the altogether inconceivable, the supreme “world riddle.” And yet if the naturalist could deign to listen to the plainest teachings of psychology and of epistemology, the riddle would seem no longer insoluble, for his phenomenal dualism and his agnostic monism would alike disappear. The material mechanism which he calls Nature would rank not as the profoundest reality there is to know: it would rather become—what indeed “machine” primarily, connotes—an instrumentality subservient to the “occasions” of the living world of ends; and so regarded, it would cease to be merely calculable, and
- ↑ The possibility is enough: we cannot tell what actually happens, and do not, therefore, know how far the direction of matter by mind calls for a modification or limitation of physical hypotheses. Cf. Ward, Naturalism and Agnosticism (3rd ed., 1906), ii. 73-86.
- ↑ Essay on “Animal Automatism,” Collected Essays, vol. i.
- ↑ Cf. Lotze, Metaphysik, § 61 fin.
