The first of the above-mentioned difficulties has no place in science. The second, though it may find a place there, would not create the sense of deadlock that the third may do. So it is only the third of the difficulties that demands immediate attention.
But their comparative remoteness from science may not prevent the first two difficulties from worrying a scientist as they must worry everyone else, and they are likely to do so more persistently if apprehended only dimly than if fully realised and squarely faced. So a few words about both of them may help those interested in the Problem of Control to free themselves from irrelevant preoccupations and to concentrate their attention on the difficulty that most matters.
It is only too disconcertingly true today that many of the most important conclusions to which physicists have come are in conflict with the judgement of common sense, and it would be better if laymen could become more reconciled to this fact. But the vain hope of bringing all those aspects of reality about which physicists tell us within the universe of discourse of ordinary folk, the hope of establishing a "Common Sense Philosophy", persists. It has given rise to a curious ambivalence towards physics. Many laymen feel on the one hand a resentful incredulity, and on the other an awed uncomprehending admiration concerning discoveries that are so far removed from the traditional instinctive convictions of mankind. "My common sense tells me that a lot of science must be wrong" and "What an amazing world of undreamt marvels science has revealed!" express, for many, almost simultaneous reactions.
These people have not made up their minds whether they consider common sense to be a very reliable or a very limited instrument of investigation. But physicists know that it is more than limited, and that it is quite often unreliable. While a few philosophers still fruitlessly attempt to reconcile the whole of man's knowledge with the judgement of common sense, physicists discovered long ago that there is neither the need nor the possibility to do so.
The conflict has been sharpened during the present century by discoveries about the nature of space. It must seem logically absurd to a layman to speak of curved space, of limited space, even of discontinuous space, i.e., to attribute physical properties to empty space. But the notions that space can be curved, that it may be of limited extent, that it may be discontinuous, that it has physical properties, are now commonplace with physicists. Such ways of thinking about space will have prepared them for the further conclusion that the judgement of common sense about the significance of location may be faulty. If one asks whether active existence is inseparable from location, a layman will answer "yes" without hesitation. A physicist, however, is likely to doubt whether space is necessarily the container of the whole of active reality. It will occur to him that no law has so far been formulated according to which what is must be somewhere. Until proof for or against such a law has been found, he can but retain an open mind.
And yet to a physiologist, the connection between our conscious minds and the controlled movement of our limbs is anything but direct, anything but fully observable, anything but indivisible, anything but uncomplicated, anything but psychologically comprehensible. Far from being direct, the connection between decision and performance is by a path consisting of many cascaded mechanisms connected to form a complicated system of loops and junctions; indivisible and uncomplicated are the last terms that would occur to a physiologist when he thinks about the connection between a decision and the resultant action. Far from being fully observable, there are parts of this path that have not yet been observed by any scientist. Our craneman observes only the extreme ends of the path — the decision to act at the one end and the movement of his hand at the other. He does not observe what happens to cells in his cortex, to nerves, endplates, muscle fibres.
Then again, far from being psychologically comprehensible, the performance of stretching out a hand requires that the craneman shall do things to mechanisms inside his body of which, with all his self-knowledge, he considers himself quite incapable. No doubt he thinks he knows exactly what he is doing when he stretches out a hand. But he does not know that he is controlling the co-ordinated timing of hundreds of thousands of muscle fibres. No doubt he thinks he knows how quickly thought can work. He has observed the speed with which a pianist can ensure the correct sequence of notes on the keyboard and such speed seems to him just, but only just, comprehensible. But he does not know that for the sounding of a single one in that torrent of notes, the pianist is controlling the performance of a vastly more complicated drill performed by a vastly greater number of muscle fibres. That so much is achieved in so short a time is far, far beyond anything psychologically comprehensible.
Fortunately, we need not often be worried by this discrepancy between our psychological understanding and our physiological knowledge. When we think about the decisions that we take and the actions that result from those decisions, we have occasion to dwell on our controlling minds, but we rarely have occasion to dwell at the same time on synapses, neurons, muscle fibres. Hence we can afford to speak as though control were direct from mind to limb and to forget about all the cascaded mechanisms that lie along the connecting path. When, on the other hand, we think about these mechanisms, we have occasion to dwell on physical facts, but we rarely have occasion to dwell at the same time on the controlling mind. It is remote from the material things in which we are interested at the moment, and we can afford to forget its very existence. In short, if we approach the path from the one end we think in terms of psychology and if we approach it from the other we think in terms of physiology; but we do not often have any need to think in terms of both disciplines at one and the same time. In consequence, we are not worried by the disconcerting fact that physiological facts are psychologically incomprehensible.
Only when, as in the present enquiry, one is obliged to give thought to the whole of the path and to remember that the unperceived elaborate drill performed by the muscle fibres is as real as the perceived control the mind exercises over it, does the problem of integrating the two disciplines present itself with any insistence. It is then that human frailty may succumb to the temptation of evading the problem by ignoring one or the other of the two disciplines. A layman who is told that an apparently simple and indivisible act like clasping a switch handle can be analysed into a stupendous number of controlled drill movements performed by countless muscle fibres may well react as Goethe did when he learnt that Newton had analysed white light into its component prime colours. Goethe testily declared that light is indivisible; he produced his own strange colour theory in justification. And to this day, literary men have been known to express their sympathy with his repudiation of that analytical approach to truth that, in their opinion, is a defect of scientific method. They sympathise because it is clear to them that Goethe's views on the nature of light were psychologically comprehensible and that Newton's views certainly were not. But Goethe did not realise, perhaps no poet could realise, that not all truth needs to be psychologically comprehensible.
According to the master controller theory which I mentioned in Chapter XVI, there is a co-ordinating mechanism in the brain, something analogous to the automatic gear change mechanism in a motor-car. This is said to ensure by purely physical means that the correct sequence of contractions shall occur among the muscle fibres. For any particular movement of a limb, there is supposed to be a particular setting of this master controller — just as there is a particular setting of the gear-changing device for each required position of the gears in a motor-car.
If it did exist in the brain, such a device would make it seem easier for the mind to control the limbs, just as an automatic gear change makes it easier for the driver to control the car. There would be fewer things for the mind to do and so there would be more time for each performance. The controlling mind would have to take action only when a new setting of the master controller was required; and if one complete gesture could be determined by a single setting, that would not be so very often. If the mind did not have to co-ordinate things in greater number and at greater speed than things can be coordinated by our hands and feet, the process by which our limbs are controlled might be a little more psychologically comprehensible than the picture presented by the knowledge that physiologists have gained. But the master controller theory does not solve one single real problem. Its authors have never even thought out what the real problems are.
What these authors would like to prove can best be made clear by reference to Chapter XIV. They would like to prove that the path between stimulus and response contains only causation without control. But in order to do this, they do not attempt to find reasons why there should be no control at all — far from it. They invent a theory according to which the control consists in the selection of the setting of a master controller. They seem to believe that an impersonal stimulus, without the activity of any controlling mind, could perform the necessary setting or adjustment of the master controller, while it could not select the moment of time at which each muscle fibre was to contract. None of the biologist-philosophers to whom we owe such theories has been able to tell me why it constitutes causation without control to select a suitable setting of a master controller, and causation with control to select the moment of time when a primary relay shall operate.
If one explores the same path from the opposite end, one has to employ the method of introspection. One observes one's own thoughts and feelings, one's motives; one is aware of the decisions that one takes. But one does not observe what each muscle fibre is doing, or what each synapse is doing; one does not observe what the controlled elements of the primary relays are doing. One does not even observe what the mind is doing as it controls the moments of time at which these elements change from one state to the other. What happens at the part of the path where it first has location is buried deeply beneath consciousness. Though by use of the Freudian technique, one may penetrate far below the surface of the conscious mind, one cannot penetrate as deeply as that.
The conclusion that there is such a thing as a primary diathesis must, therefore, result in a readjustment of those traditional philosophical views about the nature and activity of the mind. Part of the readjustment must be to attribute to the unconscious parts of the mind much more than common sense would attribute to it, much more than most philosophers would attribute to it, much more than even Freud has attributed to it. Part of the readjustment must be to regard consciousness as a very superficial phenomenon; perhaps even to view the whole of mind as but an occasional manifestation of the more important and more general diathete, life. But that is a very big theme, and I shall have to reserve its discussion for another occasion.
Another name for them would be eudiathetous mechanisms. The word "mechanism" implies that they are material systems. It also implies that they conform to the definition of a mechanism that I have given in Chapter XVI. They serve to convert diathesis from one form into another. And the word eudiathetous implies that, unlike all other mechanisms, they are directly susceptible to the action of a diathete. What happens to them does not depend only on the surrounding matter; it also partly depends on the direct action of an influence without location. The diathesis that is converted into another form by a eudiathetous mechanism reaches that mechanism literally from nowhere.
At the material beginning of every path for diathesis, there must be a eudiathetous mechanism. These mechanisms are, therefore, the most universal and the most indispensable of all instruments of control. For other instruments, one may find alternatives; if an electric motor is not available an internal combustion engine can be made to serve the same purpose. But no system of mechanisms, no matter how they are disposed, can serve the purpose of control unless a eudiathetous mechanism is placed at the material beginning of the path through which the control passes. The most automatic machine that man can devise could, conceivably, respond to a word of command. But the controlling word must be spoken before the machine acts; and in control of the speech muscles of the person who controls the machine is a set of cascaded relays with, at the beginning of the set, a eudiathetous mechanism. These small devices can be looked on as the prototypes of all mechanisms.
It has already been pointed out that the material beginning of every path for diathesis is in living substance. So eudiathetous mechanisms occur only in such substances. And for reasons mentioned briefly in Chapter XVII they are probably not only associated with mind but also with life: so one must postulate their presence in all living substance. Without them, there can be no control of metabolism, no control of growth, no control of any vital process. It is by the presence of eudiathetous mechanisms that living substance can be distinguished from lifeless substance. As they lack these devices, man-made machines can occur only at the end of a path for diathesis, never at the beginning.
Hence eudiathetous mechanisms are of the utmost importance in biology. To ignore them on the grounds that their study belongs to metaphysics and not to science, as I fear many biologists will be inclined to do, would be to turn one's back on a field of study that is likely to prove of exceptional fertility. That elusive property called vitality may very well depend largely on the state of health of these tiny mechanisms. If only more could be discovered about how they are constructed and how they work, many problems that are now puzzling biologists would, I have no doubt, be nearer to their solution. But it is not only biologists who should be concerned with such mechanisms.
Questions about the construction and operation of eudiathetous mechanisms fall also within the field of study of physicists and, more particularly, physical chemists. The most immediate and the most puzzling problems are problems in mechanics. How may it be possible, one is here and now led to ask, for any sort of mechanisms to be susceptible to the action of an influence without location? The third of the difficulties mentioned at the beginning of this chapter, and the only one that scientists ought to recognise, is the difficulty of finding a satisfactory answer to this question. Let me attempt to show, as cogently as possible, both why the difficulty is considerable and why it cannot be lightly ignored.
Clearly the performance of each one of these bodily mechanisms on which the performance of the muscle fibres depends is also ordered and not random. This applies to the endplates, the neurons, the synapses. It must apply to all the mechanisms in the cascaded series. It must apply to the first of these mechanisms, to the ones that I have called primary relays. It must apply to the controlled elements of these relays, to the devices that I have called eudiathetous mechanisms. In the sense in which the performance of the muscle fibres is ordered and not random the performance of the eudiathetous relays is ordered and not random.
Thus runs the first line of reasoning. Now for the second.
No further mechanism, moreover, is in control of the controlled element of a primary relay. For if it were this further mechanism would be the primary relay. And this is inconsistent with the conclusions reached in Chapter XIII.
Are we going to be told that there is nothing wrong with either conclusion, that the forces are both random and ordered? I am much afraid that that will be said, because it would provide such an easy way out for lazy people. We have all met the type of philosopher who sets out to explain away any difference between irreconcilables. He is quite prepared to prove that there is no intrinsic difference between something and nothing, between yes and no, between black and white, between pleasure and pain, between a cow and a table. His technique is to prove that the two things have something in common. On that foundation he argues that they have everything of any significance in common. Applying such a technique, he would be only too pleased to prove that there is no intrinsic difference between a random and an ordered performance. But his place is not among scientists.
Those who hold out the hope that, with the help of sham logic and metaphysical speculation, one may succeed in dodging the difficulty, will do nothing to advance knowledge. But those who seek honestly and methodically to discover the error in the above two lines of reasoning cannot fail to make a valuable contribution. So let us examine both lines of reasoning in detail.
Is either of the two sets of facts wrong? Is either of the two assumptions untenable? Is the logic faulty by means of which one of the conclusions was reached? I can think of no other possible error.
Some of the facts contained in the sets A (i) and B (i) have often enough been dodged by those who discuss the relation between mind, life and body. But none of the facts has ever been refuted. There are those who would like to ignore the existence of mechanisms in the body; there are those who would like to deny the validity of the principle of conservation of energy; there are those who would like to prove that nothing is ever done with a purpose; there are those who would like to prove that the performance of the muscle fibres is neither timed nor co-ordinated; there are those who would like to find reasons why causation with control is exactly the same thing as causation without control: but none of them has yet succeeded. So I can see no alternative to accepting the sets of facts mentioned both in A (i) and B (i).
The steps by which the conclusions A (iii) and B (iii) have been reached are quite short. The reasoning is very simple. There hardly seems to be room for an error there. I can see no means of denying that in each case the conclusion is inevitable, provided the facts and the assumptions be accepted.
So the most hopeful approach to the difficulty is in examining very carefully the two assumptions A (ii) and B (ii). Can either of them be shaken?
A mystic might be inclined to doubt B (ii). He would see no reason why a controlling mechanism should be needed to prevent forces from being random. He might well be prepared for the assumption that forces could be ordered without any material ordering instrument. But then the mystic is rarely a physicist. He places more faith in metaphysics than in physics. He can attach no meaning to the principle of conservation of energy. He uses the word "force" in a different sense to the one that it has in science. The statement that force is rate of change of momentum would convey nothing to his mind. The problem that is being discussed here does not really belong to his universe of discourse.
I do not think that any scientist or engineer would have much hope of proving that, in the absence of a controlling mechanism, forces can be controlled. And so it is assumption A (ii) that should be scrutinised first. Is it true that an ordered performance can result only from the application of ordered forces?
Here, be it remembered from Chapter XVI, order consists in the kind of order that can also be called co-ordinated timing. To say that the performance of the muscle fibres is ordered and not random is to say that the fibres contract at specific and not at random moments of time. So the question now reached can be put in this form:
Can random forces be caused to produce a specific event at specific moments of time?
If so, assumption A (ii) and conclusion A (iii) are both wrong, and there is no reason why the forces that act on a eudiathetous mechanism should not be random in spite of the ordered performance of the mechanism. The forces and the diathete, it must then be concluded, act simultaneously on the mechanism and they do so independently of each other. The forces are necessary for the mechanism to operate at all, and the diathete is necessary for the mechanism to operate at the specific moments of time required for co-ordination of the muscle fibres.
The Problem of Control will have been solved when it has become known how such a mechanism is constructed and how it works.