C. D. Broad, Mind and Its Place in Nature , 1925


The Traditional Problem of Body and Mind

      In the last Chapter we considered organisms simply as complicated material systems which behave in certain characteristic ways. We did not consider the fact that some organisms are animated by minds, and that all the minds of whose existence we are certain animate organisms. And we did not deal with those features in the behaviour of certain organisms which are commonly supposed to be due to the mind which animates the organism. It is such facts as these, and certain problems to which they have given rise, which I mean to discuss in the present Chapter. There is a question which has been argued about for some centuries now under the name of "Interaction"; this is the question whether minds really do act on the organisms which they animate, and whether organisms really do act on the minds which animate them. (I must point out at once that I imply no particular theory of mind or body by the word "to animate". I use it as a perfectly neutral name to express the fact that a certain mind is connected in some peculiarly intimate way with a certain body, and, under normal conditions with no other body. This is a fact even on a purely behaviouristic theory of mind; on such a view to say that the mind M animates the body B would mean that the body B, in so far as it behaves in certain ways, is the mind M. A body which did not act in these ways would be said not to be animated by a mind. And a different Body B', which acted in the same general way as B, would be said to be animated by a different mind M'.)

      The problem of Interaction is generally discussed at the level of enlightened common-sense; where it is assumed that we know pretty well what we mean by "mind", by "matter" and by "causation". Obviously no solution which is reached 'at that level can claim to be ultimate. If what we call "matter" should turn out to be a collection of spirits of low intelligence, as Leibniz thought, the argument that mind and body are so unlike that their interaction is impossible would become irrelevant. Again, if causation be nothing but regular sequence and concomitance, as some philosophers have held, it is ridiculous to regard psycho-neural parallelism and interaction as mutually exclusive alternatives. For interaction will mean no more than parallelism, and parallelism will mean no less than interaction. Nevertheless I am going to discuss the arguments here at the common-sense level, because they are so incredibly bad and yet have imposed upon so many learned men.

      We start then by assuming a developed mind and a developed organism as two distinct things, and by admitting that the two are now intimately connected in some way or other which I express by saying that "this mind animates this organism". We assume that bodies are very much as enlightened common-sense believes them to be; and that, even if we cannot define "causation", we have some means of recognising when it is present and when it is absent. The question then is: "Does a mind ever act on the body which it animates, and does a body ever act on the mind which animates it?" The answer which common-sense would give to both questions is: "Yes, certainly." On the face of it my body acts on my mind whenever a pin is stuck into the former and a painful sensation thereupon arises in the latter. And, on the face of it, my mind acts on my body whenever a desire to move my arm arises in the former and is followed by this movement in the latter. Let us call this common-sense view "Two-sided Interaction". Although it seems so obvious it has been denied by probably a majority of philosophers and a majority of physiologists. So the question is: "Why should so many distinguished men, who have studied the subject, have denied the apparently obvious fact of Two-sided Interaction?"

      The arguments against Two-sided interaction fall into two sets: -- Philosophical and Scientific. We will take the philosophical arguments first; for we shall find that the professedly scientific arguments come back in the end to the principles or prejudices which are made explicit in the philosophical arguments.

Philosophical Arguments against Two-sided Interaction.

      No one can deny that there is a close correlation between certain bodily events and certain mental events, and conversely. Therefore anyone who denies that there is action of mind on body and of body on mind must presumably hold

  1. that concomitant variation is not an adequate criterion of causal connexion, and
  2. that the other feature which is essential for causal connexion is absent in the case of body and mind.
Now the common philosophical argument is that minds and mental states are so extremely unlike bodies and bodily states that it is inconceivable that the two should be causally connected. It is certainly true that, if minds and mental events are just what they seem to be to introspection and nothing more, and if bodies and bodily events are just what enlightened common-sense thinks them to be and nothing more, the two are extremely unlike. And this fact is supposed to show that, however closely correlated certain pairs of events in mind and body respectively may be, they cannot be causally connected.

      Evidently the assumption at the back of this argument is that concomitant variation, together with a high enough degree of likeness, is an adequate test for causation; but that no amount of concomitant variation can establish causation in the absence of a high enough degree of likeness. Now I am inclined to admit part of this assumption. I think it is practically certain that causation does not simply mean concomitant variation. (And, if it did, cadit quaestio.) Hence the existence of the latter is not ipso facto a proof of the presence of the former. Again, I think it is almost certain that concomitant variation between A and B is not in fact a sufficient sign of the presence of a direct causal relation between the two. (I think it may perhaps be a sufficient sign of either a direct causal relation between A and B or of several causal relations which indirectly unite A and B through the medium of other terms C, D, etc.) So far I agree with the assumptions of the argument. But I cannot see the least reason to think that the other characteristic, which must be added to concomitant variation before we can be sure that A and B are causally connected, is a high degree of likeness between the two. One would like to know just how unlike two events may be before it becomes impossible to admit the existence of a causal relation between them. No one hesitates to hold that draughts and colds in the head are causally connected, although the two are extremely unlike each other. If the unlikeness of draughts and colds in the head does not prevent one from admitting a causal connexion between the two, why should the unlikeness of volitions and voluntary movements prevent one from holding that they are causally connected? To sum up. I am willing to admit that an adequate criterion of causal connexion needs some other relation between a pair of events beside concomitant variation; but I do not believe for a moment that this other relation is that of qualitative likeness.

      This brings us to a rather more refined form of the argument against interaction. It is said that, whenever we admit the existence of a causal relation between two events, these two events (to put it crudely) must also form parts of a single substantial whole. E.g. all physical events are spatially related and form one great extended whole. And the mental events which would commonly be admitted to be causally connected are always events in a single mind. A mind is a substantial whole of a peculiar kind too. Now it is said that between bodily events and mental events there are no relations such as those which unite physical events in different parts of the same Space or mental events in the history of the same mind. In the absence of such relations, binding mind and body into a single substantial whole, we cannot admit that bodily and mental events can be causally connected with each other, no matter how closely correlated their variations may be.

      This is a much better argument than the argument about qualitative likeness and unlikeness. If we accept the premise that causal relations can subsist only between terms which form parts of a single substantial whole must we deny that mental and bodily events can be causally connected? I do not think that we need.

      (i) It is of course perfectly true that an organism and the mind which animates it do not form a physical whole, and that they do not form a mental whole; and these, no doubt, are the two kinds of substantial whole with which we are most familiar. But it does not follow that a mind and its organism do not form a substantial whole of some kind. There, plainly, is the extraordinary intimate union between the two which I have called "animation" of the one by the other. Even if the mind be just what it seems to introspection, and the body be just what it seems to perception aided by the more precise methods of science, this seems to me to be enough to make a mind and its body a substantial whole. Even so extreme a dualist about Mind and Matter as Descartes occasionally suggests that a mind and its body together form a quasi-substance; and, although we may quarrel with the language of the very numerous philosophers who have said that the mind is "the form" of its body, we must admit that such language would never have seemed plausible unless a mind and its body together had formed something very much like a single substantial whole.

      (ii) We must, moreover, admit the possibility that minds and mental events have properties and relations which do not reveal themselves to introspection, and that bodies and bodily events may have properties and relations which do not reveal themselves to perception or to physical and chemical experiment. In virtue of these properties and relations the two together may well form a single substantial whole of the kind which is alleged to be needed for causal interaction. Thus, if we accept the premise of the argument, we have no right to assert that mind and body cannot interact; but only the much more modest proposition that introspection and perception do not suffice to assure us that mind and body are so interrelated that they can interact.

      (iii) We must further remember that the Two-sided Interactionist is under no obligation to hold that the complete conditions of any mental event are bodily or that the complete conditions of any bodily event are mental. He needs only to assert that some mental events include certain bodily events among their necessary conditions, and that some bodily events include certain mental events among their necessary conditions. If I am paralysed my volition may not move my arm; and, if I am hypnotised or intensely interested or frightened, a wound may not produce a painful sensation. Now, if the complete cause and the complete effect in all interaction include both a bodily and a mental factor, the two wholes will be related by the fact that the mental constituents belong to a single mind, that the bodily constituents belong to a single body, and that this mind animates this body. This amount of connexion should surely be enough to allow of causal interaction.

      This will be the most appropriate place to deal with the contention that, in voluntary action, and there only, we are immediately acquainted with an instance of causal connexion. If this be true the controversy is of course settled at once in favour of the Interactionist. It is generally supposed that this view was refuted once and for all by Mr Hume in his Enquiry Concerning Human Understanding (Sect. VII, Part I). I should not care to assert that the doctrine in question is true; but I do think that it is plausible, and I am quite sure that Mr Hume's arguments do not refute it. Mr Hume uses three closely connected arguments.

  1. The connexion between a successful volition and the resulting bodily movement is as mysterious and as little self-evident as the connexion between any other event and its effect.
  2. We have to learn from experience which of our volitions will be effective and which will not. E.g., we do not know, until we have tried, that we can voluntarily move our arms and cannot voluntarily move our livers. And again, if a man were suddenly paralysed, he would still expect to be able to move his arm voluntarily, and would be surprised when he found that it kept still in spite of his volition.
  3. We have discovered that the immediate consequence of a volition is a change in our nerves and muscles, which most people know nothing about; and is not the movement of a limb, which most people believe to be its immediate and necessary consequence.

      The second and third arguments are valid only against the contention that we know immediately that a volition to make a certain movement is the sufficient condition for the happening of that movement. They are quite irrelevant to the contention that we know immediately that the volition is a necessary condition for the happening of just that movement at just that time. No doubt many other conditions are also necessary, e.g., that our nerves and muscles shall be in the right state; and these other necessary conditions can be discovered only by special investigation. Since our volitions to move our limbs are in fact followed in the vast majority of cases by the willed movement, and since the other necessary conditions are not very obvious, it is natural enough that we should think that we know immediately that our volition is the sufficient condition of the movement of our limbs. If we think so, we are certainly wrong; and Mr Hume's arguments prove that we are. But they prove nothing else. It does not follow that we are wrong in thinking that we know, without having to wait for the result, that the volition is a necessary condition of the movement.

      It remains to consider the first argument. Is the connexion between cause and effect as mysterious and as little self-evident in the case of the voluntary production of bodily movement as in all other cases? If so, we must hold that the first time a baby wills to move its hand it is just as much surprised to find its hand moving as it would be to find its leg moving or its nurse bursting into flames. I do not profess to know anything about the infant mind; but it seems to me that this is a wildly paradoxical consequence, for which there is no evidence or likelihood. But there is no need to leave the matter there. It is perfectly plain that, in the case of volition and voluntary movement, there is a connexion between the cause and the effect which is not present in other cases of causation, and which does make it plausible to hold that in this one case the nature of the effect can be foreseen by merely reflecting on the nature of the cause. The peculiarity of a volition as a cause-factor is that it involves as an essential part of it the idea of the effect. To say that a person has a volition to move his arm involves saying that he has an idea of his arm (and not of his leg or his liver) and an idea of the position in which he wants his arm to be. It is simply silly in view of this fact to say that there is no closer connexion between the desire to move my arm and the movement of my arm than there is between this desire and the movement of my leg or my liver. We cannot detect any analogous connexion between cause and effect in causal transactions which we view wholly from outside, such as the movement of a billiard-ball by a cue. It is therefore by no means unreasonable to suggest that, in the one case of our own voluntary movements, we can see without waiting for the result that such and such a volition is a necessary condition of such and such a bodily movement.

      It seems to me then that Mr Hume's arguments on this point are absolutely irrelevant, and that it may very well be true that in volition we positively know that our desire for such and such a bodily movement is a necessary (though not a sufficient) condition of the happening of just that movement at just that time. On the whole then I conclude that the philosophical arguments certainly do not disprove Two-sided Interaction, and that they do not even raise any strong presumption against it. And, while I am not prepared definitely to commit myself to the view that, in voluntary movement, we positively know that the mind acts on the body, I do think that this opinion is quite plausible when properly stated and that the arguments which have been brought against are worthless. I pass therefore to the scientific arguments.

Scientific Arguments against Two-sided Interaction.

      There are, so far as I know, two of these. One is supposed to be based on the physical principle of the Conservation of Energy, and on certain experiments which have been made on human bodies. The other is based on the close analogy which is said to exist between the structures of the physiological mechanism of reflex action and that of voluntary action. I will take them in turn.

      (1) The Argument from Energy. It will first be needful to state clearly what is asserted by the principle of the Conservation of Energy. It is found that, if we take certain material systems, e.g., a gun, a cartridge, and a bullet, there is a certain magnitude which keeps approximately constant throughout all their changes. This is called "Energy". When the gun has not been fired it and the bullet have no motion, but the explosive in the cartridge has great chemical energy. When it has been fired the bullet is moving very fast and has great energy of movement. The gun, though not moving fast in its recoil, has also great energy of movement because it is very massive. The gases produced by the explosion have some energy of movement and some heat-energy, but much less chemical energy than the unexploded charge had. These various kinds of energy can be measured in common units according certain conventions. To an innocent mind there seems to be a good deal of "cooking" at this stage, i.e., the conventions seem to be chosen and various kinds and amounts of concealed energy seem to be postulated in order to make the principle come out right at the end. I do not propose to go into this in detail, for two reasons. In the first place, I think that the conventions adopted and the postulates made, though somewhat suggestive of the fraudulent company-promoter, can be justified by their coherence with certain experimental facts, and that they are not simply made ad hoc. Secondly, I shall show that the Conservation of Energy is absolutely irrelevant to the question at issue, so that it would be waste of time to treat it too seriously in the present connexion. Now it is found that the total energy of all kinds in this system, when measured according to these conventions, is approximately the same in amount though very differently distributed after the explosion and before it. If we had confined our attention to a part of this system and its energy this would not have been true. The bullet, e.g., had no energy at all before the explosion and a great deal afterwards. A system like the bullet, the gun, and the charge, is called a "Conservative System"; the bullet alone, or the gun and the charge, would be called "Non-conservative Systems". A conservative system might therefore be defined as one whose total energy is redistributed, but not altered in amount, by changes that happen within it. Of course a given system might be conservative for some kinds of change and not for others.

      So far we have merely defined a "Conservative System", and admitted that there are systems which, for some kinds of change at any rate, answer approximately to our definition. We can now state the Principle of the Conservation of Energy in terms of the conceptions just defined. The principle asserts that every material system is either itself conservative, or, if not, is part of a larger material system which is conservative. We may take it that there is good inductive evidence for this proposition.

      The next thing to consider is the experiments on the human body. These tend to prove that a living body, with the air that it breathes and the food that it eats, forms a conservative system to a high degree of approximation. We can measure the chemical energy of the food given to a man, and that which enters his body in the form of Oxygen breathed in. We can also, with suitable apparatus, collect, measure and analyse the air breathed out, and thus find its chemical energy. Similarly, we can find the energy given out in bodily movement, in heat, and in excretion. It is alleged that, on the average, whatever the man may do, the energy of his bodily movements is exactly accounted for by the energy given to him in the form of food and of Oxygen. If you take the energy put in food and Oxygen, and subtract the energy given out in waste-products, the balance is almost exactly equal to the energy put out in bodily movements. Such slight differences as are found are as often on one side as on the other, and are therefore probably due to unavoidable experimental errors. I do not propose to criticise the interpretation of these experiments in detail, because, as I shall show soon, they are completely irrelevant to the problem of whether mind and body interact. But there is just one point that I will make before passing on. It is perfectly clear that such experiments can tell us only what happens on the average over a long time. To know whether the balance was accurately kept at every moment we should have to kill the patient at each moment and analyse his body so as to find out the energy present then in the form of stored-up products. Obviously we cannot keep on killing the patient in order to analyze him, and then reviving him in order to go on with the experiment. Thus it would seem that the results of the experiment are perfectly compatible with the presence of quite large excesses or defects in the total bodily energy at certain moments, provided that these average out over longer periods. However, I do not want to press this criticism; I am quite ready to accept for our present purpose the traditional interpretation which has been put on the experiments.

      We now understand the physical principle and the experimental facts. The two together are generally supposed to prove that mind and body cannot interact. What precisely is the argument, and is it valid? I imagine that the argument, when fully stated, would run somewhat as follows: "I will to move my arm, and it moves. If the volition has anything to do with causing the movement we might expect energy to flow from my mind to my body. Thus the energy of my body ought to receive a measurable increase, not accounted for by the food that I eat and the Oxygen that I breathe. But no such physically unaccountable increases of bodily energy are found. Again, I tread on a tin-tack, and a painful sensation arises in my mind. If treading on the tack has anything to do with causing the sensation we might expect energy to flow from my body to my mind. Such energy would cease to be measurable. Thus there ought to be a noticeable decrease in my bodily energy, not balanced by increases anywhere in the physical system. But such unbalanced decreases of bodily energy are not found." So it is concluded that the volition has nothing to do with causing my arm to move, and that treading on the tack has nothing to do with causing the painful sensation.

      Is this argument valid? In the first place it is important to notice that the conclusion does not follow from the Conservation of Energy and the experimental facts alone. The real premise is a tacitly assumed proposition about causation; viz., that, if a change in A has anything to do with causing a change in B, energy must leave A and flow into B. This is neither asserted nor entailed by the Conservation of Energy. What it says is that, if energy leaves A, it must appear in something else, say B; so that A and B together form a conservative system. Since the Conservation of Energy is not itself the premise for the argument against interaction, and since it does not entail that premise, the evidence for the Conservation of Energy is not evidence against interaction. Is there any independent evidence for the premise? We may admit that it is true of many, though not of all, transactions within the physical realm. But there are cases where it is not true even of purely physical transactions; and, even if it were always true in the physical realm, it would not follow that it must also be true of transphysical causation. Take the case of a weight swinging at the end of a string hung from a fixed point. The total energy of the weight is the same at all positions in its course. It is thus a conservative system. But at every moment the direction and velocity of the weight's motion are different, and the proportion between its kinetic and its potential energy is constantly changing. These changes are caused by the pull of the string, which acts in a different direction at each different moment. The string makes no difference to the total energy of the weight; but it makes all the difference in the world to the particular way in which the weight moves and the particular way in which the energy is distributed between the potential and the kinetic forms. This is evident when we remember that the weight would begin to move in an utterly different course if at any moment the string were cut.

      Here, then, we have a clear case even in the physical realm where a system is conservative but is continually acted on by something which affects its movement and the distribution of its total energy. Why should not the mind act on the body in this way? If you say that you can see how a string can affect the movement of a weight, but cannot see how a volition could affect the movement of a material particle, you have deserted the scientific argument and have gone back to one of the philosophical arguments. Your real difficulty is either that volitions are so very unlike movements, or that the volition is in your mind whilst the movement belongs to the physical realm. And we have seen how little weight can be attached to these objections.

      The fact is that, even in purely physical systems, the Conservation of Energy does not explain what changes will happen or when they will happen. It merely imposes a very general limiting condition on the changes that are possible. The fact that the system composed of bullet, charge, and gun, in our earlier example, is conservative does not tell us that the gun ever will be fired, or when it will be fired if at all, or what will cause it to go off, or what forms of energy will appear if and when it does go off. The change in this case is determined by pulling the trigger. Likewise the mere fact that the human body and its neighbourhood form a conservative system does not explain any particular bodily movement; it does not explain why I ever move at all, or why I sometimes write, sometimes walk, and sometimes swim. To explain the happening of these particular movements at certain times it seems to be essential to take into account the volitions which happen from time to time in my mind; just as it is essential to take the string into account to explain the particular behaviour of the weight, and to take the trigger into account to explain the going off of the gun at a certain moment. The difference between the gun-system and the body-system is that a little energy does flow into the former when the trigger is pulled, whilst it is alleged that none does so when a volition starts a bodily movement. But there is not even this amount of difference between the body-system and the swinging weight.

      Thus the argument from energy has no tendency to disprove Two-sided interaction. It has gained a spurious authority from the august name of the Conservation of Energy. But this impressive principle proves to have nothing to do with the case. And the real premise of the argument is not self-evident, and is not universally true even in purely intra-physical transactions. In the end this scientific argument has to lean on the old philosophic arguments; and we have seen that these are but bruised reeds. Nevertheless, the facts brought forward by the argument from energy do throw some light on the nature of the interaction between mind and body, assuming this to happen. They do suggest that all the energy of our bodily actions comes out of and goes back into the physical world, and that minds neither add energy to nor abstract it from the latter. What they do, if they do anything, is to determine that at a given moment so much energy shall change from the chemical form to the form of bodily movement; and they determine this, so far as we can see, without altering the total amount of energy in the physical world.

      (2) The Argument from the Structure of the Nervous System. There are purely reflex actions, like sneezing and blinking, in which there is reason to suppose that the mind plays any essential part. Now we know the nervous structure which is used in such acts as these. A stimulus is given to the outer end of an efferent nerve; some change or other runs up this nerve, crosses a synapsis between this and an afferent nerve, travels down the latter to a muscle, causes the muscle to contract, and so produces a bodily movement. There seems no reason to believe that the mind plays any essential part in this process. The process may be irreducibly vital, and not merely physico-chemical; but there seems no need to assume anything more than this. Now it is said that the whole nervous system is simply an immense complication of interconnected nervous arcs. The result is that a change which travels inwards has an immense number of alternative paths by which it may travel outwards. Thus the reaction to a given stimulus is no longer one definite movement, as in the simple reflex. Almost any movement may follow any stimulus according to the path which the afferent disturbance happens to take. This path will depend on the relative resistance of the various synapses at the time. Now a variable response to the same stimulus is characteristic of deliberate as opposed to reflex action.

      These are the facts. The argument based on them runs as follows. It is admitted that the mind has nothing to do with the causation of purely reflex actions. But the nervous structure and the nervous processes involved in deliberate action do not differ in kind from those involved in reflex action; they differ only in degree of complexity. The variability which characterises deliberate action is fully explained by the variety of alternative paths and the variable resistances of the synapses. So it is unreasonable to suppose that the mind has any more to do with causing deliberate actions than it has to do with causing reflex actions.

      I think that this argument is invalid. In the first place I am pretty sure that the persons who use it have before their imagination a kind of picture of how mind and body must interact if they interact at all. They find that the facts do not answer to this picture, and so they conclude that there is no interaction. The picture is of the following kind. They think of the mind as sitting somewhere in a hole in the brain surrounded by telephones. And they think of the efferent disturbance as coming to an end at one of these telephones and there affecting the mind. The mind is then supposed to respond by sending an efferent impulse down another of these telephones. As no such hole, with efferent nerves stopping at its walls and afferent nerves starting from them, can be found, they conclude that the mind can play no part in the transaction. But another alternative is that this picture of how the mind must act if it acts at all is wrong. To put it shortly, the mistake is to confuse a gap in an explanation with a spatio-temporal gap, and to argue from the absence of the latter to the absence of the former.

      The Interactionist's contention is simply that there is a gap in any purely physiological explanation of deliberate action; i.e., that all such explanations fail to account completely for the facts because they leave out one necessary condition. It does not follow in the least that there must be a spatio-temporal breach of continuity in the physiological conditions, and that the missing condition must fill this gap in the way in which the movement of a wire fills the spatio-temporal interval between the pulling of a bell-handle and the ringing of a distant bell. To assume this is to make the mind a kind of physical object, and to make its action a kind of mechanical action. Really, the mind and its actions are not literally in Space at all, and the time which is occupied by the mental event is no doubt also occupied by some part of the physiological process Thus I am inclined to think that much of the force which this argument actually exercises on many people is simply due to the presupposition about the modus operandi of interaction, and that it is greatly weakened when this presupposition is shown to be a mere prejudice due to our limited power of envisaging unfamiliar alternative possibilities.

      We can, however, make more detailed objections to the argument than this. There is a clear introspective difference between the mental accompaniment of voluntary action and that of reflex action. What goes on in our minds when we decide with difficulty to get out of a hot bath on a cold morning is obviously extremely different from what goes on in our minds when we sniff pepper and sneeze. And the difference is qualitative; it is not a mere difference of complexity. This difference has to be explained somehow; and the theory under discussion gives no plausible explanation of it. The ordinary view that, in the latter case, the mind is not acting on the body at all; whilst, in the former, it is acting on the body in a specific way, does at least make the introspective difference between the two intelligible.

      Again, whilst it is true that deliberate action differs from reflex action in its greater variability of response to the same stimulus, this is certainly not the whole or the most important part of the difference between them. The really important difference is that, in deliberate action, the response is varied appropriately to meet the special circumstances which are supposed to exist at the time or are expected to arise later; whilst reflex action is not varied in this way, but is blind and almost mechanical. The complexity of the nervous system explains the possibility of variation; it does not in the least explain why the alternative which actually takes place should as a rule be appropriate and not merely haphazard. And so again it seems as if some factor were in operation in deliberate action which is not present in reflex action; and it is reasonable to suppose that this factor is the volition in the mind.

      It seems to me that this second scientific argument has no tendency to disprove interaction; but that the facts which it brings forward do tend to suggest the particular form which interaction probably takes if it happens at all. They suggest that what the mind does to the body in voluntary action, if it does anything, is to lower the resistance of certain synapses and to raise that of others. The result is that the nervous current follows such a course as to produce the particular movement which the mind judges to be appropriate at the time. On such a view the difference between reflex habitual, and deliberate actions for the present purpose becomes fairly plain. In pure reflexes the mind cannot voluntarily affect the resistance of the synapses concerned, and so the action takes place in spite of it. In habitual action it deliberately refrains from interfering with the resistance of the synapses, and so the action goes on like a complicated reflex. But it can affect these resistances if it wishes, though often only with difficulty; and it is ready to do so if it judges this to be expedient. Finally, it may lose the power altogether. This would be what happens when a person becomes a slave to some habit, such as drug-taking.

      I conclude that, at the level of enlightened common-sense at which the ordinary discussion of Interaction moves, no good reason has been produced for doubting that the mind acts on the body in volition, and that the body acts on the mind in sensation. The philosophic arguments are quite inconclusive; and the scientific arguments, when properly understood, are quite compatible with Two-sided Interaction. At most they suggest certain conclusions as to the form which interaction probably takes if it happens at all.

Difficulties in the Denial of Interaction.

      I propose now to consider some of the difficulties which would attend the denial of Interaction, still keeping the discussion at the same common-sense level. If a man denies the action of body on mind he is at once in trouble over the causation of new sensations. Suppose that I suddenly tread on an unsuspected tin-tack. A new sensation suddenly comes into my mind. This is an event, and it presumably has some cause. Now, however carefully I introspect and retrospect, I can find no other mental event which is adequate to account for the fact that just that sensation has arisen at just that moment if I reject the common-sense view that treading on the tack is an essential part of the cause of the sensation, I must suppose either that it is uncaused, or that it is caused by other events in my mind which I cannot discover by introspection or retrospection, or that it is caused telepathically by other finite minds or by God. Now enquiry of my neighbours would show that it is not caused telepathically by any event in their minds which they can introspect or remember. Thus anyone who denies the action of body on mind, and admits that sensations have causes, must postulate either

  1. immense numbers of unobservable states in his own mind; or
  2. as many unobservable states in his neighbours' minds, together with telepathic action; or
  3. some non-human spirit together with telepathic action.
I must confess that the difficulties which have been alleged against the action of body on mind seem to be mild compared with those of the alternative hypotheses which are involved in the denial of such action.

      The difficulties which are involved in the denial of the action of mind on body are at first sight equally great; but I do not think that they turn out to be so serious as those which are involved in denying the action of body on mind. The prima facie difficulty is this. The world contains many obviously artificial objects, such as books, bridges, clothes, etc. We know that, if we go far enough back in the history of their production, we always do in fact come on the actions of some human body. And the minds connected with these bodies did design the objects in question, did will to produce them, and did believe that they were initiating and guiding the physical process by means of these designs and volitions. If it be true that the mind does not act on the body, it follows that the designs and volitions in the agents' minds did not in fact play any part in the production of books, bridges, clothes, etc. This appears highly paradoxical. And it is an easy step from it to say that anyone who denies the action of mind on body must admit that books, bridges, and other such objects could have been produced even though there had been no minds, no thought of these objects and no desire for them. This consequence seems manifestly absurd to common-sense, and it might be argued that it reflects its absurdity back on the theory which entails it.

      The man who denies that mind can act on body might deal with this difficulty in two ways:

      (1) He might deny that the conclusion is intrinsically absurd. He might say that human bodies are extraordinarily complex physical objects, which probably obey irreducible laws of their own, and that we really do not know enough about them to set limits to what their unaided powers could accomplish. This is the line which Spinoza took. The conclusion, it would be argued, seems absurd only because the state of affairs which it contemplates is so very unfamiliar. We find it difficult to imagine a body like ours without a mind like ours; but, if we could get over this defect in our powers of imagination, we might have no difficulty in admitting that such a body could do all the things which our bodies do. I think it must be admitted that the difficulty is not so great as that which is involved in denying the action of body on mind. There we had to postulate ad hoc utterly unfamiliar entities and modes of action; here it is not certain that we should have to do this.

      (2) The other line of argument would be to say that the alleged consequence does not necessarily follow from denying the action of mind on body. I assume that both parties admit that causation is something more than mere de facto regularity of sequence and concomitance. If they do not, of course the whole controversy between them becomes futile; for there will certainly be causation between mind and body and between body and mind, in the only sense in which there is causation anywhere. This being presupposed, the following kind of answer is logically possible. When I say that B could not have happened unless A had happened, there are two alternative possibilities.

  1. A may itself be an indispensable link in any chain of causes which ends up with B.
  2. A may not itself be a link in any chain of causation which ends up with B. But there may be an indispensable link a in any such chain of causation, and A may be a necessary accompaniment or sequent of a.
These two possibilities may be illustrated by diagrams. (a) is represented by the figure below: --


The two forms of (b) are represented by the two figures below: --



A0   /A1A2B

      Evidently, if B cannot happen unless a precedes, and if a cannot happen without A accompanying or immediately following it, B will not be able to happen unless A precedes it. And yet A will have had no part in causing B. It will be noticed that, on this view, a has a complex effect AA1, of which a certain part, viz., A1 is sufficient by itself to produce A2 and ultimately B. Let us apply this abstract possibility to our present problem. Suppose that B is some artificial object, like a book or a bridge. If we admit that this could not have come into existence unless a certain design and volition had existed in a certain mind, we could interpret the facts in two ways.

  1. We could hold that the design and volition are themselves an indispensable link in the chain of causation which ends in the production of a bridge or a book. This is the common view, and it requires us to admit the action of mind on body.
  2. We might hold that the design and the volition are not themselves a link in the chain of causation which ends in the production of the artificial object; but that they are a necessary accompaniment or sequent of something which is an indispensable link in this chain of causation. On this view the chain consists wholly of physical events; but one of these physical events (viz., some event in the brain) has a complex consequent. One part of this consequent is purely physical, and leads by purely physical causation to the ultimate production of a bridge or a book. The other is purely mental, and consists of a certain design and volition in the mind which animates the human body concerned. If this has any consequences they are purely mental. Each part of this complex consequent follows with equal necessity; this particular brain-state could no more have existed without such and such a mental state accompanying or following it than it could have existed without such and such a bodily movement following it. If we are willing to take some such view as this, we can admit that certain objects could not have existed unless there had been designs of them and desires for them; and yet we could consistently deny that these desires and designs have any effect on the movements of our bodies.

      It seems to me then that the doctrine which I will call "One-sided Action of Body on Mind" is logically possible; i.e., a theory which accepts the action of body on mind but denies the action of mind on body. But I do not see the least reason to accept it, since I see no reason to deny that mind acts on body in volition. One-sided Action has, I think, generally been held in the special form called "Epiphenomenalism." I take this doctrine to consist of the following four propositions:

  1. Certain bodily events cause certain mental events.
  2. No mental event plays any part in the causation of any bodily event.
  3. No mental event plays any part in the causation of any other mental event. Consequently
  4. all mental events are caused by bodily events and by them only.
Thus Epiphenomenalism is just One-sided Action of Body on Mind, together with a special theory about the nature and structure of mind. This special theory does not call for discussion here, where I am dealing only with the relations between minds and bodies, and am not concerned with a detailed analysis of mind. In a later chapter we shall have to consider the special features of Epiphenomenalism.

Arguments in Favour of Interaction.

      The only arguments for One-sided Action of Body on Mind or for Parallelism are the arguments against Two-sided interaction; and these, as we have seen, are worthless. Are there any arguments in favour of Two-sided interaction? I have incidentally given two which seem to me to have considerable weight. In favour of the action of mind on body is the fact that we seem to be immediately aware of a causal relation when we voluntarily try to produce a bodily movement, and that the arguments to show that this cannot be true are invalid. In favour of the action of body on mind are the insuperable difficulties which I have pointed out in accounting for the happening of new sensations on any other hypothesis. There are, however, two other arguments which have often been thought to prove the action of mind on body. These are

  1. an evolutionary argument, first used, I believe, by William James; and
  2. the famous "telegram argument." They both seem to me to be quite obviously invalid.

      (1) The evolutionary argument runs as follows: it is a fact, which is admitted by persons who deny Two sided Interaction, that minds increase in complexity and power with the growth in complexity of the brain and nervous system. Now, if the mind makes no difference to the actions of the body, this development on the mental side is quite unintelligible from the point of view of natural selection. Let us imagine two animals whose brains and nervous systems were of the same degree of complexity; and suppose, if possible, that one had a mind and the other had none. If the mind makes no difference to the behaviour of the body the chance of survival and of leaving descendants will clearly be the same for the two animals. Therefore natural selection will have no tendency to favour the evolution of mind which has actually taken place. I do not think that there is anything in this argument. Natural selection is a purely negative process; it simply tends to eliminate individuals and species which have variations unfavourable to survival. Now, by hypothesis, the possession of a mind is not unfavourable to survival; it simply makes no difference. Now it may be that the existence of a mind of such and such a kind is an inevitable consequence of the existence of a brain and nervous system of such and such a degree of complexity. Indeed we have seen that some such view is essential if the opponent of Two-sided interaction is to answer the common-sense objection that artificial objects could not have existed unless there had been a mind which designed and desired them. On this hypothesis there is no need to invoke natural selection twice over, once to explain the evolution of the brain and nervous system, and once to explain the evolution of the mind. If natural selection will account for the evolution of the brain and nervous system, the evolution of the mind will follow inevitably, even though it adds nothing to the survival-value of the organism. The plain fact is that natural selection does not account for the origin or for the growth in complexity of anything whatever; and therefore it is no objection to any particular theory of the relations of mind and body that, if it were true, natural selection would not explain the origin and development of mind.

      (2) The "telegram argument" is as follows: Suppose there were two telegrams, one saying "Our son has been killed", and the other saying: "Your son has been killed". And suppose that one or other of them was delivered to a parent whose son was away from home. As physical stimuli they are obviously extremely alike, since they differ only in the fact that the letter "Y" is present in one and absent in the other. Yet we know that the reaction of the person who received the telegram might be very different according to which one he received. This is supposed to show that the reactions of the body cannot be wholly accounted for by bodily causes, and that the mind must intervene causally in some cases. Now I have very little doubt that the mind does play a part in determining the action of the recipient of the telegram; but I do not see why this argument should prove it to a person who doubted or denied it. If two very similar stimuli are followed by two very different results, we are no doubt justified in concluding that these stimuli are not the complete causes of the reactions which follow them. But of course it would be admitted by everyone that the receipt of the telegram is not the complete cause of the recipient's reaction. We all know that his brain and nervous system play an essential part in any reaction that he may make to the stimulus. The question then is whether the minute structure of his brain and nervous system, including in this the supposed traces left by past stimuli and past reactions, is not enough to account for the great difference in his behaviour on receiving two very similar stimuli. Two keys may be very much alike, but one may fit a certain lock and the other may not. And, if the lock be connected with the trigger of a loaded gun, the results of "stimulating" the system with one or other of the two keys will be extremely different. We know that the brain and nervous system are very complex, and we commonly suppose that they contain more or less permanent traces and linkages due to past stimuli and reactions. If this be granted, it is obvious that two very similar stimuli may produce very different results, simply because one fits in with the internal structure of the brain and nervous system whilst the other does not. And I do not see how we can be sure that anything more is needed to account for the mere difference of reaction adduced by the "telegram argument."

The Positive Theory Of Parallelism.

      The doctrine of Psycho-physical Parallelism, or, as I prefer to call it, "Psycho-neural Parallelism", has two sides to it. One is negative; it is the denial that mind acts on body and the denial that body acts on mind. With this side of it I have now dealt to the best of my ability, and have argued that there is no reason to believe it and tolerably good reason to disbelieve it. But Psychoneural Parallelism has also a positive side, which might be accepted by one who rejected its negative side. The positive assertion of Parallelism is that there is a one-one correlation between events in a mind and events in the brain and nervous system of the body which it animates. Is there any reason to believe this on empirical grounds?

      I think we must say that it may be true, but that it is a perfectly enormous assumption unless there be some general metaphysical ground for it; and that the empirical evidence for it is, and will always remain, quite inadequate. The assertion is that to every particular change in the mind there corresponds a certain change in the brain which this mind animates, and that to every change in the brain there corresponds a certain change in the mind which animates this brain. What kind of empirical evidence could there be for such an assertion? At best the evidence would be of the following kind: "I have observed a number of brains and the minds which animate them; and I have never found a change in either which was not correlated with a specific change in the other. And all other people who have made similar observations have found the same thing." If we had evidence of this sort the positive side of Parallelism would be a straightforward inductive generalization of it; i.e., an argument from "A has never been observed to happen without B" to "A never does happen without B". But actually we have no evidence whatever of this kind. No one person in the world ever has observed, or probably ever will observe, a brain and its mind. The only mind that he can observe is his own and the only brains that he can observe are those of others. Nor is this the worst. We can very rarely observe other men's brains at all, and never when they are alive and in a state of normal consciousness. Thus the actual empirical data for the positive side of Parallelism consist of observations on brains which are no longer animated by minds at all or whose animating minds are in abeyance. And these minds could not be directly observed by us even if they were present and functioning normally.

      It will therefore be worth while to consider carefully what account of parallelism we really are justified on empirical grounds in assuming.

      (1) We have fairly good reasons for thinking that the existence and general integrity of a brain and nervous system is a necessary condition for the manifestation of a mind to itself and to other minds. We do not positively know that it is a sufficient condition; and the question whether it be so or not will have to be discussed later in this book. Our evidence is all of the following kind:

  1. In the absence of a brain and nervous system we see none of those external actions which we know in our own case to be accompanied by consciousness.
  2. The brain and nervous system are known to increase in complexity up to a certain age, and we have observed in ourselves and can infer from the behaviour of others a corresponding growth in mental complexity.
  3. Soon after men have ceased to show signs of consciousness by their external behaviour their brains and nervous systems break up. It must be admitted that it might be maintained with almost equal plausibility that these last facts show that the integrity of the brain and nervous system is dependent on the presence of the mind. We might just as well argue that the brain begins to break up because the mind has ceased to animate it, as that the mind has ceased to manifest itself because the brain has begun to break up. In fact, seeing the order in which we actually get our knowledge of the two facts, the former is prima facie the more plausible interpretation.
  4. In many cases where men's behaviour has been so odd as to suggest that their minds are abnormal, it is known that their brains have been injured or it has been found after their death that their brains were in an abnormal state. On the other hand, it must be admitted that the brains of some lunatics on dissection show no systematic differences from those of normal people.
It would obviously be absurd to talk of "Parallelism" in reference to this very general relation between the integrity and complexity of the brain and nervous system, on the one hand, and the manifestation of a human mind, on the other.

      (2) There is, however, empirical evidence which goes rather further than this. It is found that wounds in certain parts of the brain make specific differences to the mind. e.g., a wound in one part may be followed by a loss of memory for spoken words, and so on. Unfortunately, similar results can often be produced by causes like hypnotism or like those which psycho-analysts discuss. And here there is no positive empirical evidence that these specific areas of the brain are affected. Again, there seems to be some evidence that, after a time and within certain limits, another part of the brain can take over the functions of a part that has been injured. Thus the most that we can say is that the general integrity of certain parts of the brain seems to be at least a temporarily necessary condition for the manifestation of certain specific kinds of mental activity. It remains doubtful how far any given area is indispensable for a given kind of mental activity, and whether there may not be some kinds of activity which, though dependent like all others on the general integrity of the brain, are not specially correlated with any particular area. We might sum up these facts by saying that there is good evidence for a considerable amount of "Departmental Parallelism" between mind and brain.

      (3) The orthodox Parallelist, however, goes much further than this, and much beyond the most rigid departmental parallelism. He would hold, not merely that there is a strict correlation between each distinguishable department of mental life and some specific area of the brain, but also that there is a strict parallelism of events. E.g., he holds, not merely that I could not remember at all unless a certain area of my brain were intact, but also that if I now remember eating my breakfast there is a certain event in this area uniquely correlated with this particular mental event. And by "unique correlation" he means that if some other mental state had happened now instead of this particular memory there would necessarily have been a different brain-event, and conversely. So far as I know there is not, and could not possibly be, any empirical evidence for this "Parallelism of Events", as I will call it. For

  1. , while a man is conscious and can observe events in his own mind, his brain is not open to inspection by himself or by anyone else. And, when his brain is open to inspection, he is not likely to be in a position to introspect or to tell others what is going on in his mind, even if something is happening there at the time.
  2. In any case the events in the brain which are supposed to correspond to particular events in the mind would be admitted to be too minute to be observable even under the most favourable circumstances. They are as purely hypothetical as the motions of electrons, without the advantage that the assumption of them enables us to predict better than we could otherwise do what states of mind a man will probably have under given circumstances.

      It seems to me then that there is no empirical evidence at all for a Parallelism of Events between mind and brain. If this doctrine is to be held, the grounds for it must be general. E.g. psycho-neural parallelism might be plausible if, on other grounds, we saw reason to accept psycho-physical parallelism; i.e., the doctrine that every physical event is correlated with a specific mental event, and conversely. And the wider doctrine might be defended as helping to explain the apparent origin of life and mind from apparently non-living and non-conscious matter. This is a question which we shall have to discuss later; all that I am concerned to argue at present is that, at the level of enlightened common-sense and apart from some general metaphysical theory of the nature of matter and mind, there is no adequate evidence for a psycho-neural parallelism of events. And, as parallelism has commonly been defended on the ground that it is established by empirical scientific investigation of the brain and nervous system, this fact is worth pointing out.

      If there is no reason for psycho-neural parallelism of events, is there any positive reason against it? Some philosophers have held that there is. They have held that, while it is possible and even probable that some mental events are correlated with specific neural events, it is impossible that this should be true of all mental events. Those who take this view generally hold that there probably is psycho-neural parallelism of events for sensations, but that there certainly cannot be such parallelism for comparison, introspection, attentive inspection, and so on. This view is taken by Mr Johnson in his Logic (Part III); and it will be worth while to consider his arguments. They are contained in Chapter VII, § 6 of that work. Mr Johnson's argument, if I rightly understand it, comes to this: We must distinguish, e.g. between the fact that I am having two sensations, one of which is light red and the other dark red, and my recognition that both are red and that one is darker than the other. We must likewise distinguish, e.g., between the fact that the dark one started before the light one, and my recognition of this fact; and between the fact that the dark one is to the left of the light one in my visual field and my recognition of this fact, Finally, we must notice that we have to distinguish different degrees of clearness and determinateness with which a perfectly determinate fact may be recognized. We may merely judge that one sensum is separate from another, or we may judge that one is to the left of the other, or we may judge that the first is as much to the left of the second as the second is to the right of a third, and so on. Now Mr Johnson contends that the sensations themselves have neural correlates, and that the determinate qualities and relations which the sensations actually have are determined by the qualities and relations of these neural correlates. But he holds that there is then nothing left on the neural side for the recognition of these qualities and relations to be correlated with. Still less is there anything left on the neural side to be correlated with the infinitely numerous different degrees of determinateness with which the qualities and relations of the sensations may be apprehended. Hence he concludes that mental events above the level of sensations cannot be correlated one to one with specific neural events. He does not explicitly draw the distinction which I have done between Departmental Parallelism and Parallelism of Events; but I think it is plain that his argument is meant only to deny the latter. He would probably admit that, if certain specific areas of our brains were injured, we should lose altogether the power of making judgments of comparison and of recognising spatio-temporal relations; but he would hold that, given the general integrity of those areas, there is not some one specific event within them corresponding to each particular judgment of comparison or of spatio-temporal relation.

      Before criticising this argument we must notice that Mr Johnson does not explicitly distinguish sensations and sensa. By a "sensation" I think he means what I should call a "sensed sensum". And he thinks that, from the nature of the case, there can be no unsensed sensa. Thus a sensation for him is a sensum, regarded as existentially mind-dependent; and, in virtue of its supposed existential mind-dependence, it counts as a mental event belonging to the mind on which its existence depends. If we like to distinguish between mental states and mental acts we can say that a sensation, for Mr Johnson, is apparently a mental state having certain sensible qualities, such as colour, position in the visual field, and so on. To recognise that one is having a sensation, that it is of such and such a kind, and that it stands in such and such spatio-temporal relations to other sensations, would be to perform a cognitive mental act. And his contention is that, whilst there is a parallelism of events for mental states, there cannot for this very reason be also a parallelism of events for mental acts. This at least is how I understand him.

      Now I must confess that Mr Johnson's argument seems to me to be so extremely weak that (knowing Mr Johnson) I hesitate to believe that I can have properly understood it. Let us suppose that the actual relative position of two sensa s1 and s2 in a visual sense-field is determined by the relative position of two excited areas in the brain, b1 and b2. Let us suppose that the actual relative date of the two sensa in the sense-history of the experient is determined by the relative date of the excitement of these areas. And let us suppose that the determinate sensible qualities of the two sensa (e.g., the particular shade of the particular colour possessed by each) is determined by the particular kind of movement which is going on in the microscopic particles within these two areas. Mr Johnson's contention seems to be that, when we have mentioned the positions of the excited areas, the dates at which they begin to be excited, and the particular kind of movement which is going on within them, we have said all that can be said about the neural events. There is nothing left on the neural side to be correlated with our acts of recognition, of qualitative comparison, and of spatio-temporal judgment; and therefore these events can have no special neural correlate. To this there are two answers which seem so obvious that I am almost ashamed to make them.

      (1) At the very utmost the argument would show only that there is nothing left within the two areas b1 and b2 to be correlated with any judgments which we happen to make about the sensations s1 and s2. But these two areas do not exhaust the whole of the brain and nervous system. Why our acts of judgment about these two sensations should not have neural correlates in some other part of the brain I cannot imagine. The situation on the mental side is that we may, but need not, make these judgments if we do have the sensations; and that we cannot make them unless we have the sensations. This is exactly what we might expect if the neural correlates of the acts of judgment were in a different part of the brain from the neural correlates of the sensations themselves: and if a certain kind of disturbance in the latter were a necessary but insufficient condition of a certain kind of disturbance in the former.

      (2) But we could answer the argument without needing even to assume that the neural correlates of judgments about sensations are in a different area of the brain from the neural correlates of the sensations themselves. We have to remember that the same area may contain at the same time microscopic events of different scales of magnitude. Let us take a purely physical analogy. The same piece of metal may be at once hot and glowing. We have extremely good reasons to believe that both these apparent characteristics are correlated with microscopic motions which are going on throughout the whole volume occupied by the bit of metal. The heat is supposed to be correlated with the random movements of molecules and the light with the jumps of electrons from one stable orbit to another. The large scale events can go on without the small-scale events (a body may be hot without glowing); but the more violent the large-scale events the more frequent will be the small-scale events (a body begins to glow if it be heated enough). Now I cannot imagine why the same thing might not be true of the neural correlates of sensations and the neural correlates of our judgments about our sensations. Suppose that the neural correlates of sensations were large-scale events in a certain area of the brain; and suppose that the neural correlates of our judgments about these sensations were small scale events in the same area. Then I should expect to find that sensations could happen without our making judgments about them; that we could not make the judgments unless we had the sensations; and that it would be more difficult not to make the judgments as the sensations became more intense, other things being equal. And this is exactly what I do find. It seems to me then, either that I have altogether misunderstood Mr Johnson's argument, or that there is nothing whatever in it.

      There remains one other point to be discussed before leaving the subject. It is true, as Mr Johnson points out, that we make judgments of various degrees of determinateness about the same perfectly determinate fact Does this raise any particular difficulty against the view that every act of judgment has a specific neural correlate? I do not think that it does, if we avoid certain confusions into which it is very easy to fall. I suppose that the difficulty that is felt is this: "Every neural event is perfectly determinate; how can an indeterminate judgment have a determinate neural correlate; and how can there be different determinate neural correlates for all the different degrees of determinateness in judgments?" To this I should answer (i) that of course the differences on the neural side which would correspond to different degrees of determinateness in the judgment are not themselves differences of determinateness. But why should they be? The differences on the neural side which correspond to differences of shade in sensations of colour are not themselves differences of shade. If, e.g., the area which is correlated with judgments about our sensations be different from the area which is correlated with the sensations themselves, we might suppose that differences in the determinateness of the judgment were correlated with differences in the extent or the intensity of the disturbance within this area. If, on the other hand, we supposed that our sensations were correlated with large-scale events, and our judgments about these sensations with small-scale events in the same region of the brain, we might suppose that differences in the determinateness of the judgment are correlated with differences in the frequency of these small-scale events. There is thus no difficulty, so far as I can see, in providing neural correlates to every different degree of determinateness in our judgments. (ii) It is perhaps necessary to point out that what is called an "indeterminate judgment" is not an indeterminate event; every event, whether mental or physical, is no doubt perfectly determinate of its kind. E.g., whether I merely judge that some one has been in the room or make the more determinate judgment that John Smith has been in the room, either judgment as a psychical event has perfectly determinate forms of all the psychical determinables under which it falls. The indefiniteness is in what is asserted, not in the act of asserting as such. Hence the problem is not, as it might seem to a careless observer, to find a determinate neural correlate to an indeterminate psychical event; the problem is merely to find a determinate neural correlate to a determinate psychical event which consists in the asserting of a relatively indeterminate characteristic.

      I conclude then that no adequate reason has been produced by Mr Johnson to prove that there cannot be specific neural correlates to mental acts as well as to mental states. I have also tried to show that there neither is nor is likely to be any empirical evidence for the doctrine that all mental events have specific neural events as their correlates. Hence the positive doctrine of Psycho-neural Parallelism of Events seems to me to be a perfectly open question. This is not perhaps a wildly exciting result. But it is not altogether to be despised, since it leaves us with a perfectly free hand when we try to construct a speculative theory of the relations of matter and mind which shall do justice to all the known facts. For the known facts neither require nor preclude complete Psycho-neural Parallelism of Events.

Summary and Conclusions.

      I wish to make quite clear what I do and what I do not claim to have done in this chapter. I have definitely assumed that the body and the mind are two distinct entities, which are now in a very intimate union, which I express by saying that the former is "animated by" the latter. I have raised no question about the exact nature or origin of this relation of "animation"; and I have not considered the apparent growth of mind in the individual or the apparent development of consciousness from the non-conscious in the course of the earth's history. Again, I have taken the body to be very much as common-sense, enlightened by physical science, but not by philosophical criticism, takes it to be; I have supposed that we know pretty well what a mind is; and I have assumed that causation is not simply regular sequence and concomitant variation, though these are more or less trustworthy signs of the presence of a causal relation. These are the assumptions on which the question of interaction has commonly been discussed by philosophers and by scientists; and it would be idle for me to conceal my opinion that it has been discussed extraordinarily badly. The problem seems to have exercised a most unfortunate effect on those who have treated it; for I have rarely met with a collection of worse arguments on all sides. I can only hope that I have not provided yet another instance in support of this generalization

      My conclusion is that, subject to the assumptions just mentioned, no argument has been produced which should make any reasonable person doubt that mind acts on body in volition and that body acts on mind in sensation. I have tried to show the extreme difficulties which are involved in attempting to deny that body acts on mind. And I have tried to show that the apparently equal difficulties which seem to be involved in attempting to deny that mind acts on body could be evaded with a little ingenuity. Thus One-sided Action of Body on Mind is a possible theory. But there seems to me to be no positive reason for accepting it, and at least one reason for doubting it) viz., the conviction which many men have (and which Mr Hume's arguments fail altogether to refute) that we know directly that our volitions are necessary conditions for the occurrence of our voluntary movements.

      If these conclusions be sound, Parallelism, considered as an alternative which excludes interaction, has no leg left to stand upon. But Parallelism has a positive side to it which is perfectly compatible with Interaction, and is therefore worth discussing for its own sake. I distinguished between the metaphysical doctrine of Psycho-physical Parallelism and the more restricted doctrine of Psycho-neural Parallelism. And I divided the latter into Departmental Parallelism and Parallelism of Events. It seemed to me that there was good empirical evidence for a considerable amount of Departmental Parallelism, but that there was not and is not likely to be adequate empirical evidence for Parallelism of Events. On the other hand, I came to the conclusion that Mr Johnson's arguments to prove that complete parallelism between mental and neural events is impossible were quite unsound.

      This, I think, is as far as the discussion can be carried at this level. One thing seems to me to emerge clearly even at this point. If interaction has to be denied at a later stage it can only be because the relation between mind and body turns out to be so intimate that "interaction" is an unsuitable expression for the connection between a particular mental event and its correlated bodily event. This would be so if e.g., Materialism were true, so that the mind was just some part of the body. It might be so on a Double-aspect Theory, or on a theory of Neutral Monism. But we cannot decide between such general theories until we know more about the true nature of Mind and of Matter, and have taken into consideration questions about origin and development of minds which we have hitherto explicitly left out of account. Thus the final discussion of the question can come only near the end of the book.

Contents -- Go to Chapter IV