Why the mind has a body

Front matter

siipips

IMm

HP

Preface

Vll ever prove to be the truth, it would be in a translated form in which our common-sense belief in the efficiency of mind would be recognized as essentially accurate. In Dr. Stout’s theory and mine this prophecy finds its fulfilment.

It only remains for me to express my obligations to the friends who have assisted me in the preparation of the book. I owe most to my father, Rev. A. H. Strong, D.D., of the Rochester Theological Seminary. My thanks are due to Professor James, to Mr. Henry Rutgers Marshall, and to my colleague Professor Cattell, who read all or parts of the book in manuscript and helped me with their criticisms and suggestions. I have also received some valuable hints from my friends Professor George Santayana and Dr. Dickinson S. Miller. Finally, I should feel that something essential had been omitted if I did not gratefully acknowdedge my immense indebtedness to the philosophical writings of Professor Paulsen and to Professor James’s Principles of Psychology, on which I have nourished myself for years and which have been the main influences in shaping my thought.

Lakewood, N. J.

Conclusion

The problem of the relation of mind and body takes, for contemporary thought, the form of the issue between interactionism and automatism. Interactionism regards the brain as an instrument employed by the mind in its dealings with the world of objects. It accordingly asserts in sensation an action of the body on the mind, in volition an action of the mind on the body. Automatism conceives the brain-process as the physical basis or condition of consciousness. It therefore holds that consciousness merely accompanies the brain-process, without exerting any influence upon it.

But the term ‘ automatism,’ when thus defined, covers two widely different theories. These are the ‘ conscious automaton theory ’ of Huxley and the 1 parallelism ’ of Clifford. Huxley regards consciousness as an effect of brain-action, an effect which does not in its turn become a cause and react on the brain, and which is therefore wholly without efficiency. He thus retains still a one-sided causal relation between body and mind. Clifford holds that consciousness and the brain-process merely flow along side by side, the brain being no more responsible for consciousness than consciousness for what happens in i l the brain. He wholly denies causal relations between body and mind.

This last theory seems alone entitled to the name of parallelism. For, although on Huxley’s theory also the stream of consciousness flows along beside the brain-process, yet the corresponding points of the two are not directly opposite, the mental states being always a little later than the brain-events which cause them. Moreover, the idea meant to be conveyed by the spatial figure of parallelism is that of the absence of causal relations, of mere concomitance. On the other hand, both theories are in strictness of speech automatist, since both hold that the brain gets along by itself, without the help of the mind. In the ordinary conception of automatism, however, there lies, in addition to this, some thought of a dependence of consciousness on the brain, some notion of the brain as the primary thing and of consciousness as “ somehow superadded ” ; and I think it therefore allowable to employ the term c automatism ’ for Huxley’s theory only, which needs a name, and for which it is difficult to find an exactly appropriate one.

There are thus three distinct theories as to causal relations between mind and body : interactionism, asserting that the causal influence runs in both directions — in sensation from the body to the mind, in volition from the mind to the body; automatism, maintaining that it runs in one direction only — always from the body to the mind ; and parallelism, denying all causal influence and holding the relation to be of a different nature. Between these theories the contest is fairly triangular : for a theory which asserts causal relations in one direction only is at least as distinct from a theory which denies causal relations as from one that asserts them in both directions.

From causal theories we must distinguish sharply what may be called ultimate theories of the relation of mind and body : theories, that is, which seek to explain and make intelligible to us why mind and body are connected at all. The 1 double aspect theory,’ with its assertion of one reality manifesting itself under two diverse forms, is an example of such a theory. Whatever we may think of the example, to some such theory we must sooner or later come, if we would solve the problem completely. We cannot rest satisfied with mere determination of the causal issue. For suppose interactionism to be proved true, we should be as far as ever from understanding why two such interacting things are juxtaposed. Or, if parallelism were demonstrated, we should have still to learn how this parallel march of nervous and mental events was brought about. On the other hand, all ultimate theories would appear to rest upon and be varieties of the three causal theories already mentioned.

Philosophers are at present pretty exclusively occupied with the causal issue ; and it certainly seems as if they were right in attacking it first. The three causal theories, when compared with each other, show peculiar points of agreement and difference.

If automatism and parallelism alike deny the action of mind upon body, automatism and interactionism agree with each other and differ from parallelism in asserting causal relations of some sort between body and mind ; and it is a question which is the deeper gulf. I believe the deeper gulf by far is that between parallelism and its rivals. The great question is not in which direction the causal influence runs, but whether the relation of mind and body is such as to permit of causal relations at all.

Let us stop for a moment, before going further, to consider the consequences to a series of sciences which would result from asserting or denying causal relations between body and mind.

And, first of all, to the most general of the natural sciences, physics. Here the issue at stake is the universal validity of the principle of the conservation of energy. Is this principle, as Helmholtz declared it, “ a universal law of all natural phenomena,” 1 or only, as a recent interaction ist writer would have us believe, “ a useful and valid working-hypothesis applicable to certain classes of physical fact ” ? Interactionism stands committed to the latter view, automatism and parallelism support the former.

Next, this is a vital question for biology. Sooner or later the biologist must come to close quarters with the question of the place of consciousness in evolutionary theory. Hitherto two tendencies in biology have disputed the field: the thorough-going

1 Helmholtz, Vortrdge und Reden , p. 151.

Darwinian, presupposing the mechanical theory of life, and ignoring the possibility that the mind is a factor in bodily evolution ; and the Lamarckian, entertaining and building upon that possibility. Many biologists at the present day believe fortuitous variation and natural selection inadequate to account for the actual phenomena of evolution. They suspect the presence of an additional factor, whose nature remains to be determined. The study of instinct brings them close upon the confines of intelligence ; instinct seems to involve feeling, and feeling thus to be a factor no less efficient than bodily organs in determining survival. Perplexed to account by physical causes alone for the origin of favorable variations and for their preservation while as yet insufficient in amount to affect survival, what wonder if they have thought to find the needed additional factor in consciousness ? 1

Whether this interactionist tendency in biology be significant or ephemeral, we need not inquire. That it is a creationist hypothesis on a small scale, attributing to the animal mind an intervention in physical affairs which biologists have latterly denied to the divine, is perhaps no conclusive argument against it. It has been plausibly urged, in proof of such intervention, that if consciousness did not contribute to survival by acting upon the body it would never have been evolved. We may reserve this argument for later consideration, merely noting at

1 See Cope, Primary Factors of Organic Evolution , pp. 13, 14 and 495-517.

present that it must make the greatest difference to biology whether it is legitimate thus to appeal to consciousness as a factor in physical evolution.

Next, brain-physiology and mental pathology would be affected. The physiologist, agreeably to that materialism of his which forms so excellent a principle of method but so dubious a philosophy, is accustomed to trace all cerebral events to bodily causes, such as stimulating nerve-currents, nutrition, oxygenation. From Griesinger’s enunciation of the principle that insanity is brain-disease the alienist dates the birth of modern psychiatry. Must the physiologist give up his rule of method, and the alienist his fundamental axiom ?

When we turn from the natural to the philosophical sciences, we find that the consequences would be no less far-reaching. And, first, to general philosophy or metaphysics. It needs no argument to show that, according as we decide for causal relations or against them, our entire conception of the universe will be a different one. For to reject causal relations is to hold that in the brain none but mechanical causes are at work, that brain-events taken with other physical events form a closed circle ; and this is to set the capstone upon the mechanical conception of the universe. Whereas to assert causal relations is to contradict this conception by positing a hiatus in the physical series.

Finally, parallelism involves the denial of the physical efficiency of mind, and automatism the denial of its general efficiency. These doctrines thus apparently leave ns no longer the authors of our acts! We shall find, when we come to consider the matter, that automatism does this really, since the causal inertness of mind is one of its cardinal principles. Parallelism, on the other hand, which denies only the physical efficiency of mind, may perhaps prove consistent with its possessing some other kind of efficiency, and we must hold our judgment in suspense. But it is plain already how momentous is this issue to ethics and the philosophy of mind.

Thus a whole series of scientific and philosophical conceptions of the first order — the principle of the conservation of energy, the mechanical theory of life, the biological doctrine of evolution, the fundamental postulates of brain-physiology and mental pathology, the philosophical conceptions of mechanism, efficiency, free-will — all converge and come to a focus in the problem of the relation of mind and body. Not only so, but every one of these conceptions is vitally engaged, and will be found to stand or fall or suffer total transformation, according as we espouse interactionism, automatism, or parallelism.

Let us now consider by what means a decision between the three theories might be brought about.

It is evident, to begin with, that the question is partly at least one of fact. When we examine the three theories closely, we find that they are not merely different interpretations of the same admitted facts, but assume differences in the facts themselves. These differences are of three kinds.

(1) Interactionism assumes that for certain physical events, namely, those immediately consequent upon our volitions, no adequate cause is discoverable on the physical side. It assumes, in other words, the non-existence of certain physical facts whose existence is assumed by automatism and parallelism. And the analogous proposition might be worked out for the case of sensation : automatism and parallelism assume that the brain-event which accompanies or immediately precedes a sensation — the sensational brain-event, as we shall find it convenient to call it — goes on to produce all the physical effects which it would have produced had there been no sensation; whereas interactionism assumes (or ought in logic to assume) that the brain-event exhausts itself in producing the sensation, and therefore has no physical effects.

(2) Interactionism would seem to require the assumption that the sensational brain-event happens just previously to the sensation, since the cause must necessarily precede the effect in time, and the volitional brain-event for the same reason just subsequently to the volition. Automatism agrees with interactionism in its account of sensation, but in its account of volition reverses the temporal relation, making the volition come after the volitional brain-event instead of before. Parallelism, finally, assumes that sensational brain-event and sensation, volitional brain-event and volition happen simultaneously.

(3) Automatism and parallelism alike involve the assumption that all mental states — not only sensations and images, but also judgments, pleasures and pains, desires, volitional decisions, nay, even the consciousness of self — have events corresponding to them in the brain. Interactionism commonly disputes this, holding that the ‘ higher ’ mental states are without cerebral correlates.

These differing versions of the psychophysical facts yield a series of empirical problems, investigable presumably by the methods of natural science, quite independently of the views one may hold as to the ultimate nature of physical facts and mental states. Whether mental states are attributes of a Soul, or appearances to a Subject, or contents of a unitary consciousness, or discrete ideas, they exist in some sense ; and whether physical events are externally real, just as we see and feel them, or only modifications of the mind, they too exist in some sense : in some sense, therefore, the temporal and other relations between these two kinds of fact must exist, and have validity on any philosophical theory. This being so, no harm can result from ignoring metaphysical difficulties at the outset and seeking by empirical methods to determine these questions of fact. Indeed, it stands to reason that an exact determination of the facts should precede all discussions of theory.

I shall therefore begin with a study of the facts. I shall take up first those primary and indubitable ones which constitute the data of the problem, dividing them into classes and discussing their consistency with each of the three causal theories. I shall then deal briefly with the question, whether consciousness is immediately correlated with the total bodily process, or only with certain parts of it, and, if the latter, with what parts. This done, we may consider the questions referred to above : whether the correlation extends to all mental states, and whether it involves simultaneity or succession. But, without abandoning empirical ground, we may go still further. Having determined the facts, we may proceed to try the issue between interactionism, automatism, and parallelism so far as it is ad judicable by empirical considerations alone ; so far, that is to say, as it is a question for physiological psychology as a natural science.

It may seem to the reader that, if this empirical inquiry were satisfactorily concluded, our task would be completed and the problem solved; that physiological psychology, in other words, is in this question the tribunal of last resort. And it must be admitted that the establishment of certain facts would negate or tend to negate each of the three theories. Thus proof that the c higher ’ mental states are without cerebral correlates would tend to disprove automatism and parallelism, while evidence of the universality and simultaneity of the correlation would tell strongly against interactionism. If, then, facts could overthrow each theory, could not facts establish each ? Cannot physiological psychology, without overstepping the limits of her province, settle the whole question ?

To show that this assignment of jurisdiction, even if theoretically correct, is not in practice sufficient for the question’s settlement, I need only cite the views actually held by leading writers. There are psychologists who admit the universality of the correlation, yet uphold interactionism. There are physiologists and even philosophers who insist on the universality of the correlation, but see in it a proof of the automatist view that the cerebral correlates produce the mental states. There are others, equally convinced of such universality, who interpret it in the sense of parallelism, and hold that the relation excludes causality. Finally, there are those exceptional parallelists who deny the universality of the correlation. These conflicting interpretations of the same facts sufficiently prove that we cannot be satisfied with mere fact-determination, but must discuss the nature of the causal relation, so as to judge of its applicability between body and mind. Now, the nature of the causal relation has not usually been deemed an empirical question, but a metaphysical one.

But is there not another question, even more clearly metaphysical, which is logically prior? Would it not be well, before debating whether mind can act upon matter, to decide just what we mean by matter, and in what sense we may reasonably conceive it to exist? For if at the present day there is a point on which philosophers show some approach to agreement, it is that matter does not exist, in any such sense as the plain man supposes ; that it has no existence independently of mind. But, if matter does not exist independently of mind, it cannot be there for the mind to act upon (or, for that matter, for mental states to run parallel to). One would accordingly think some investigation of the separateness of matter and mind an indispensable preliminary to discussion of their interaction. Yet most interactionists go straight to the causal issue, assuming that we know perfectly well what we mean by matter, that it exists in the universe alongside of mind, and that the only question is whether we will permit mind to act upon it (as a hungry dog might be permitted to attack a bone), or will wickedly forbid such action. In other moods these writers are commonly idealists, but when they discuss the relation of mind and body they forget their idealism and talk like the naivest of realists.

It is no less indispensable to have a clear idea of what we mean by mind. Despite the apparent testimony of consciousness to its own efficiency, writers of mark are found to declare it exempt from or superior to the causal category, and thus too lofty a thing to soil its fingers with common work. Some, like Kant, resolve it into a series of phenomena no more deserving of the predicate of' reality than those of matter. It is therefore as necessary to examine whether consciousness is real and capable of acting as whether matter is real and capable of being acted upon.

For these reasons it is clear that our inquiry must have a metaphysical part, devoted to a critical examination of the conceptions of matter, mind, and causation. But this is equally necessary for a further reason : because without it we should be unable to cope with our problem in its full extent. Of the two questions which together constitute this problem, only the first, as to the causal or non-causal character of the relation, could by any possibility be decided by empirical facts and arguments alone. The second, how two such disparate things come to be associated at all — “ultimate of ultimate problems,” in the words of Professor James1 — is none the less an integral part of our task. The likelihood of any light being thrown upon it in the course of the present inquiry may appear to the reader small. He will at least admit that any treatment of it presupposes definite conclusions regarding the nature of matter and mind. From which it would follow that our problem belongs, not to physiological psychology as a natural science, but to the branch of metaphysics known as the philosophy of mind, whose task it is to deal with the philosophical questions growing out of psychology.

I shall accordingly supplement the empirical inquiry already announced with a metaphysical investigation of the nature of matter and mind, and shall follow this up with renewed discussion of the problem from the vantage-ground of the critical conceptions thus attained. Possibly after this metaphysical clearing of the atmosphere the causal question may appear in a somewhat different light, and even the second and more ultimate of the two questions prove less intractable than it now seems.

I am aware that readers whose metaphysical investments have hitherto brought small returns will take a less hopeful view of the issue of our undertaking. Metaphysics is in discredit at the present day, not altogether undeservedly. But metaphysics in this question is inevitable. Is it, or is it not, a real question whether material things exist independently of the mind, or only for the mind ? Is it, or is it not, a real question whether the mind itself is real, active, and efficient ? Is it, or is it not, indispensable to answer these questions before debating whether the mind can act on the body? To these queries there can be but one answer. We must study the facts, and we must examine the empirical arguments, but when this is done, whatever our misgivings, we must bid farewell to the terra firma of experience and embark upon the ocean of metaphysics.

But is this ocean navigable ? Can we hope, as the result of our inquiry, to reach any assured and binding conclusions ? A widespread notion of metaphysics makes it synonymous with ‘ speculation ’ — a discipline whose processes escape rational control, whose results are therefore essentially doubtful. In the natural sciences, we are told, a new theory can be put to the test of experiment and its truth or falsity quickly determined, but in metaphysics this is impossible. I have no desire to undertake the defence of all that passes under the name of metaphysics. But I would point out that our problem, perhaps alone among those with which metaphysics deals, affords an opportunity for something that may be quite accurately described as a metaphysical experiment. In the psychophysical relation mind and matter are somehow associated. We are to investigate the nature of these two things quite irrespectively of their association. We are then to make trial, whether the conceptions at which we have arrived enable us to explain the association. If perchance they should, would not our procedure be like that of the chemist who, suspecting that oxygen and hydrogen together made water, isolated these gases and brought them together and found that in truth they do ? A similar success would give to our theory, it seems to me, something like experimental confirmation.

i Sill

Part I

EMPIRICAL

Chapter I

The facts which exemplify the relation of mind and body may be divided into three classes, according as they seem to show action of mind on body, or action of body on mind, or dependence of mind on body. I say “ seem to show,” for if the facts really showed these things, it is evident that the first and second classes would prove interactionism, and the third class automatism. It is therefore an important question for us to discuss, whether the facts necessitate one theory or another, or can be interpreted in terms of all three.

Apparent Action of Mind on Body

The most conspicuous instance of this is of course volition. When I will a bodily movement, and the movement takes place, I may be pardoned for supposing that the mind has acted on the body. But volition is only one case among many. Recent psychologists

20 WHY THE MIND HAS A BODY [cnjj

tell its that all mental states are followed by bodily changes — that “all consciousness leads to action.” This is true of desires, of emotions, of pleasures and pains, and even of such seemingly non-impulsive states as sensations and ideas. It is true, in a word, of the entire range of our mental life. The bodily effects in question are of course not limited to the voluntary muscles, but consist in large part of less patent changes in the action of heart, lungs, stomach, and other viscera, in the calibre of blood-vessels and the secretion of glands. A few examples will help to make our ideas definite.

Whether the emotions are to be regarded as causes or as reflex manifestations of the bodily changes which constitute their expression, in either case a mental state — on the traditional theory the emotion itself, on the c back-stroke theory ’ the idea that arouses it — has been followed by physical changes. And the parallel statement applies to the theories of desire and volitional effort formed on the ‘ back-stroke ’ analogy, which explain effort as a feeling of muscular strain, and desire for food, for example, as partly composed of the sensation of the watering of the mouth. On any theory, emotion is accompanied by an overflow of cerebral currents entailing profound and widespread bodily effects. Thus passion breaks forth in physical violence, joy quickens and grief depresses the vital functions, fear leads to a paralysis of movement, anger in a wet-nurse poisons the milk, etc.

The bodily consequences of ideas and states wre should describe as cognitive are no less clearly

I]

marked. The thought of blushing makes one blush, the sound of others coughing makes one cough. Among these states none is more physically potent than belief, as we see by the phenomena of hypnotic suggestion. Thus the subject’s belief that he cannot open his eyes makes opening them impossible, his conviction that a postage-stamp is a mustard-plaster enables it to raise a blister ; the hysterical patient’s belief that she cannot move her leg or open her hand creates a virtual paralysis or contracture of those parts ; and faith or confident expectation may aid in the cure of functional ailments. Indeed, the latest theory of will makes its essence consist in the conviction that the bodily movement is about to take place — a prevision which speedily brings about its own fulfilment.

But why multiply instances, when every instinctive movement, every voluntary act, perhaps even every turn of the attention, is a case in point? The apparent action of mind upon body is one of the most obvious characteristics of our mental life. Let us rather inquire, without further delay, whether the facts necessitate interactionism, or can be formulated in terms of the other two theories.

A moment’s reflection shows that the latter is the case. Though the facts seem at first sight to prove interactionism, automatism and parallelism offer alternative interpretations of them which, for anything we can now see, stand an equal chance of being correct.

Let us consider the parallelist interpretation first. The type of this class is the case of volition; and this

[CIIAP.

is the exact reverse of the case of sensation. Now, when a sensory nerve-current seems to cause a sensation, parallelism tells us that it really causes a sensational brain-event, which is merely accompanied by the sensation : and there is nothing in the facts which would permit us to gainsay the account. The sensational brain-event being wholly removed from our knowledge, the assertion of its concomitance with the sensation is as likely to be true as that of their sequence. In the same way, when a volition appears to cause a bodily movement, parallelism tells us that it is accompanied by a volitional brain-event which is the real cause of the movement ; and again, there is nothing in the bare facts to contradict such a view.

The analogy of the case of volition holds good for all other cases in which the mind appears to act on the body. Let us take what would be the most impressive, if we could be certain of its truth. We hear of men sick unto death who survive through an effort of will, or recover because of a fixed determination to do so. We read of persons in the last stages of some lingering disease, who continue alive through the strength of some affection. Waiving all doubts as to the facts, let us consider the interpretation which parallelism would put upon them if true. According to the parallelist, the determination to recover, this persistent volitional state, is accompanied by an equally persistent brain-event, which is the real agent in effecting recovery. Similarly, it is not the nurse’s anger that poisons the milk, but the agitated brainstate underlying the anger. It is not the hypnotic subject’s belief that prevents his opening his eyes, but the inhibitory brain-event corresponding to the belief. What the interactionist, in all these cases, takes for an action of mind on body is in reality, on the purely physical side, an action of the brain upon the rest of the body. Nor can we, at this stage, deny the legitimacy of the parallelist’s interpretation, or demonstrate interactionism simply by insisting on the facts. The facts are wholly neutral, and would be exactly what they are on either the interactionist or the parallelist hypothesis.

With a few slight changes the above argument may be made to show that the facts are equally consistent with automatism. This theory, like parallelism, ascribes the physical result — the recovery from illness, the poisoning of the milk, the inability to open the eyes — not to the preceding mental state, but to its attendant brain-event. Automatism differs from parallelism only in holding that the brain-event causes the mental state as well as the physical result. And this, again, is a view which cannot be refuted merely by pointing to the facts, for there is nothing in the facts that contradicts it. I conclude, therefore, that the facts which seem to prove action of mind on body do not really prove it, but can be formulated in terms of each of the three theories.

Apparent Action of Body on Mind

The facts which appear to show such action may be classified according to the mode in which the physical influence takes effect : through the sensory nerves, through the cranial walls and brain-substance, through the blood-supply, or through the working of the psychophysical mechanism itself. They pass over, as we shall see, without a break into the facts which seem to show dependence.