The Ghost in the Machine Read online

  Watson's book Behaviourism, in which he rejected the concepts of consciousness and mind, was published in 1913. Half a century later, Professor Skinner of Harvard University, who is probably the most influential contemporary academic psychologist, proclaims the same views in even more extreme form. In his standard work Science and Human Behaviour the hopeful student of psychology is firmly told from the very outset that 'mind' and 'ideas' are non-existent entities, 'invented for the sole purpose of providing spurious explanations. . . . Since mental or psychic events are asserted to lack the dimensions of physical science, we have an additional reason for rejecting them'. [5] By the same logic, the physicist may, of course, reject the existence of radio waves, becanse they are propagated through a so-called 'field' which lacks the properties of ordinary physical media. In fact, few of the theories and concepts of modern physics would survive an ideological purge on Behaviourist principles -- for the simple reason that the scientific outlook of Behaviourism is modelled on the mechanistic physics of the nineteenth century.

  The 'cynical onlooker' might now ask: if mental events are to be excluded from the study of psychology -- what is there left for the psychologist to study? The short answer is: rats. For the last fifty years the main preoccupation of the Behaviourist school has been the study of certain measurable aspects of the behaviour of rats, and the bulk of Behaviourist literature is devoted to that study. THs development, odd as it seems, was in fact an unavoidable consequence of the Behaviourist's definition of scientific method (the 'fourth pillar' mentioned above). According to his self-imposed limitations, the Behaviourist is only permitted to study objective, measurable aspects of behaviour. However, there are few relevant aspects of human behaviour which lend themselves to quantitative measurement under laboratory conditions, and which the experimenter can investigate without relying on introspective statements about private events experienced by the subject. Thus, if he wanted to remain faithful to his principles, the Behaviourist had to choose as objects of his study animals in preference to humans, and among animals rats and pigeons in preference to monkeys or chimpanzees, because the behaviour of primates is still too complex.

  Rats and pigeons, on the other hand, can, under appropriately designed experimental conditions, be made to behave as if they were indeed conditioned reflex automata, or almost so. There is hardly a self-respecting psychological faculty in the Western world without some white albino rats disporting themselves in so-called Skinner boxes, invented by that eminent Harvard authority. The box is equipped with a food tray, an electric bulb, and a bar which can he pushed down like the lever of a slot machine, whereupon a food pellet drops into the tray. When a rat is placed into the box, it will sooner or later press the lever down with its paw, and will be automatically rewarded by a pellet; and it will soon learn that to get food it must press the bar. This experimental procedure is called 'operant conditioning' because the rat 'operates' on the environment (as distinct from Pavlovian 'classical' or 'respondant' conditioning, where it does not). Pressing the bar is called 'emitting an operant response'; the food pellet is called a 'reinforcing stimulus' or 'reinforcer'; withholding the food pellet is a 'negative reinforcer'; the alternation of the two procedures is 'intermittent reinforcement'. The rat's 'rate of response' -- i.e., the number of times it presses the bar in a given period of time -- is automatically recorded, plotted on charts, and regarded as a measure of 'operant strength'.* The purpose of the box is to enable the Behaviourist to realise his cherished ambition: the measurement of behaviour by quantitative methods, and the control of behaviour by the manipulation of stimuli.

  * Operant strength is usually measured, for technical reasons, by the 'rate of extinction' -- how long the rat will persist in pressing the lever after the supply of pallets has been stopped.

  The Skinner box did produce some technically interesting results. The most interesting was that 'intermittent reinforce -- ment' -- when pressing the bar was only sometimes rewarded by a pellet -- could be as effective, and even more effective than when it was always rewarded; the rat, which had been trained not to expect a reward after every try, is less discouraged, and goes on trying much longer after the supply of pellets has been stopped, than the rat which had previously been rewarded after every try. (The words 'expect' and 'discouraged' which I have used would, of course, be disallowed by the Behaviourist because they imply mental events.) This proudest achievement of some thirty years of bar-pressing experiments is a measure of their relevance as a contribution to psychology. As one eminent critic, Harlow, wrote already in 1953: 'a strong case can be made for the proposition that the importance of the psychological problems studied during the last fifteen years has decreased as a negatively accelerated function approaching an asymptote of complete indifference'. [6] Looking back at the further fifteen years that have passed since this was written, one would come much to the same conclusion. The attempt to reduce the complex activities of man to the hypothetical 'atoms of behaviour' found in lower mammals produced next to nothing that is relevant -- just as the chemical analysis of bricks and mortar will tell you next to nothing about the architecture of a building. Yet throughout the dark ages of psychology most of the work done in the laboratories consisted of analysing bricks and mortar in the hope that by patient effort somehow one day it would tell you what a cathedral looked like.

  The De-Humanisation of Man

  However, if the futility of these experiments would be the only reason for criticism, then one would indeed be flogging indignantly a dead horse. But, incredible as it may seem, the Skinnerians claim that the bar-pressing experiments with rats, and the training of pigeons (about which more presently), provide all the necessary elements to describe, predict and control human behaviour -- including language ('verbal behaviour'), science and art. Skinner's two best-known books are called The Behaviour of Organisms and Science and Human Behaviour. Nothing in their resounding titles indicates that the data in them are almost exclusively derived from conditioning experiments on rats and pigeons -- and then converted by crude analogies into confident assertions about the political, religious and ethical problems of man. The motivational drive of the rat is measured by the number of hours it has been deprived of food before being put into the box; human behaviour, according to Skinner, can be described in the same terms:

  Behaviour which has been strengthened by a conditioned reinforcer varies with the deprivation appropriate to the primary reinforcer. The behaviour of going to a restaurant is composed of a sequence of responses, early members of which (for example, going along a certain street) are reinforced by the appearance of discriminative stimuli which control later responses (the appearance of the restaurant, which we then enter). The whole sequence is ultimately reinforced by food, and the probability varies with food deprivation. We increase the chances that someone will go to a restaurant, or even walk along a particular street, by making him hungry. [7]

  Next in importance to Skinner of Harvard in shaping academic psychology was the late Clark Hull of Yale; his pupils still occupy key positions in the academic world. His system differed on technical points from Skinner's, but his basic outlook was the same: he, too, expressly postulated that the differences between the processes of learning in man and rat are merely of a quantitative, not of a qualitative, order:

  The natural-science theory of behaviour being developed by the present author and his associates assumes that all behaviour of the individuals of a given species and that of all species of mammals, including man, occurs according to the same set of primary laws. [8]

  The unique attributes of man, verbal communication and written records, science, art, and so forth, are considered to differ only in degree, not in kind, from the learning achievements of the lower animals -- once more epitomised, for Hull as for Skinner, in the bar-pressing activities of the rat. Pavlov counted the number of drops which his dogs salivated through their artificial fistulae, and distilled them into a philosophy of man; Professors Skinner, Hull and their follo
wers took an equally heroic short cut from the rat in the box to the human condition.

  Skinner's most impressive experiment in the 'prediction and control ofbehaviour' is to train pigeons, by operant conditioning, to strut about with their heads held unnaturally high. He turns on a light; then food appears in a place where the pigeon can only reach it by stretching its neck; after a while, each time the light is turned on, the pigeon stretches its neck, expecting the food. How does one extrapolate from this to the prediction and control of human behaviour? Skinner explains (his italics):

  We describe the contingency by saying that a stimulus (the light) is the occasion upon which a response (stretching the neck) is followed by reinforcement (with food). We must specify all three terms. The effect upon the pigeon is that eventually the response is more likely to occur when the light is on. The process through which this comes about is called discrimination. Its importance in a theoretical analysis, as well as in the practical control of behaviour, is obvious. . . . For example, in an orchard in which red apples are sweet and all others sour, the behaviour of picking and eating comes to be controlled by the redness of the stimulus. . . . The social environment contains vast numbers of such contingencies. A smile is an occasion upon which social approach will meet with approval. A frown is an occasion upon which the same approach will not meet with approval. Insofar as this is generally true, approach comes to depend to some extent upon the facial expression of the person approached. We use this fact when by smiling or frowning we control to some extent the behaviour of those approaching us. . . . The verbal stimulus 'Come to dinner' is an occasion upon which going to a table and sitting down is usually reinforced by food. The stimulus comes to be effective in increasing the probability of that behaviour and is produced by the speaker because it does so. [9]

  How to Manipulate Tautologies

  Skinner did not intend to write a parody.* He means it seriously. Less obvious, however, than the monumental triviality of its pronouncements is the fact that the pedantic jargon of Behaviourism is based on ill-defined verbal concepts which willingly lend themselves to circular arguments and tautological statements. A 'response', the layman would imagine, is an answer to a stimulus; but 'operant responses' are 'emitted' to produce a stimulus which occurs after the response; the response 'acts upon the environment in such a way that a reinforcing stimulus is produced'. [10] In other words, the response responds to a stimulus which is still in the future -- which, if taken literally, is nonsensical. An 'operant response' is not in fact a response, but an act initiated by the animal; but, as organisms are supposed to be controlled by the environment, the passive term 'response' is mandatory in the whole literature. Behaviourism is based on S-R theory (stimulus-response theory) as first defined by Watson: 'The rule or measuring rod, which the Behaviourist puts in front of him always is: can I describe this bit of behaviour I see in terms of "stimulus and response"?' [11] These S-R bits are regarded as the 'elements' or 'atoms' of the chain of behaviour; if the R for 'response' were eliminated from the terminology, the chain would fall to pieces, and the whole theory collapse.

  * In a memorable essay, 'Pavlov and his Bad Dog' (Encounter, London, Sept. 1964), attacking the English brand of Behaviourism, Kathleen Nott pointed out three main characteristics of this kind of jargon: '(1) Grandiose-inflationary or "Bullfrog" (B.f.); (2) Disguise by obviousness or "Poe" (E.A.P.) and (3) Pejorative reference to unacceptable concepts or other psychological theories, or Giving a Name a Bad Dog (B.D.).'

  Another omnipresent term in contemporary psychological jargon -- which has even found its way into political jargon -- is the ugly word 'reinforcement'. What exactly does it mean? According to Skinner's 'law of conditioning': 'if the occurrence of an operant is followed by presentation of a reinforcing stimulus, the strength [of that operant] is increased'. [12] And how is a 'reinforcing stimulus' defined? 'A reinforcing stimulus is defined as such by its power to produce the resulting change [in strength] .' [13] Translated into human language, we arrive at the tautology: the probability of repeating an action is increased by reinforcement, where 'reinforcement' means something which increases that probability.* As one of Skinner's critics wrote: 'Examining the instances of what Skinner calls reinforcement, we find that not even the requirement that a reinforcer be an identifiable stimulus is taken seriously' (Chomsky). [15] According to Skinner, 'a man talks to himself . . . because of the reinforcement he receives' [16]; thinking is 'behaving which automatically affects the behaviour and is reinforcing because it does so' [17]; 'just as the musician plays or composes what he is reinforced by hearing, or as the artist paints what reinforces him visually, so the speaker engaged in verbal fantasy says what he is reinforced by hearing or writes what he is reinforced by reading' [18]; and the creative artist is 'controlled entirely by the contingencies of reinforcement'. [18a] Fortunately, in Skinnerian parlance, the word 'control' is as empty as 'reinforcement'. Originally, in talking of pigeons and rats, 'prediction and control of behaviour' had a concrete meaning: by giving and withholding rewards, the animal's behaviour could be drastically shaped by the experimenter. But in the case of the writer who is controlled by the 'contingencies of reinforcement', the word 'control' refers to the fact that his 'verbal behaviour may reach over centuries or to thousands of listeners or readers at the same time. The writer may not be reinforced often or immediately, but his net reinforcement may be great' [19] (which accounts for the great 'strength' of his behaviour, whatever that means). Thus the environment which 'controls entirely' the writer's verbal behaviour includes stimuli centuries ahead; and determines whether he should hammer out on his typewriter a tragedy or a limerick.

  * The 'strength' of an operant is measured by the probability of it being repeated in similar conditions. [14] The tautological nature of the so-called law of conditioning has been repeatedly pointed out before.

  This brings us to the Behaviourist's attitude to human creativity. How can scientific discovery and artistic originality be explained or described without reference to mind and imagination? The following two quotations will indicate the answer. The first is again from Watson's Behaviourism, published in 1925; the second from Skinner's Science and Human Behaviour, published thirty years later; thus they enable us to judge whether there is any substantial difference between the paleo-Behaviourist and neo-Behaviourist attitudes. (Some readers will perhaps notice that I have already used the same passage from Watson in The Act of Creation, for it happens to be the only passage in his fundamental book in which creative activities are discussed):

  One natural question often raised is, how do we ever get new verbal creations such as a poem or a brilliant essay? The answer is that we get them by manipulating words, shifting them about until a new pattern is hit upon. . . . How do you suppose Patou builds a new gown? Has he any 'picture in his mind' of what the gown is to look like when it is finished? He has not. . . . He calls his model in, picks up a new piece of silk, throws it around her, he pulls it in here, he pulls it out there. . . . He manipulates the material until it takes on the semblance of a dress. . . . Not until the new creation aroused admiration and commendation, both his own and others, would manipulation be complete -- the equivalent of the rat's finding food. . . . The painter plies his trade in the same way, nor can the poet boast of any other method. [19a]

  In the article on 'Behaviourism' in the 1955 edition of the Encyclopaedia Britannica you will find five columns of eulogy for Watson. His books, we are told, 'demomtrate the possibility of writing an adequate, comprehensive account of human and animal behaviour without the use of the philosophical concept of mind or consciousness'. One wonders whether the author of the Encyclopaedia Britannica article (Professor Hunter of Brown College) would really regard the above quotation as 'an adequate and comprehensive account' of how Hamlet or the Sistine Chapel came into being.

  Thirty years after Watson, Skinner summed up the Behaviourist's views on how original discoveries are made in Science and Human Behaviour: 'The result o
f solving a problem is the appearance of a solution in the form of a response. . . . The relation between the preliminary behaviour and the appearance of the solution is simply the relation between the manipulation of variables and the emission of a response. . . . The appearance of the response in the individual's behaviour is no more surprising than the appearance of any response in the behaviour of any organism. The question of originality can be disposed of. . . .' [20]

  Needless to say, the 'organisms' referred to are once more his rats and pigeons. Compared with Watson's, the language of the Skinnerians has become more dehydrated and esoteric. Watson talks of manipulating words until a new pattern is 'hit upon', Skinner of manipulating 'variables' until 'a response is emitted'. Both are engaged in question-begging on a heroic scale, apparently driven by an almost fanatical urge to deny, at all costs, the existence of properties which account for the humanity of man and the rattiness of the rat.