The Lord, having apparently grown tired of hearing a certain sort of behaviourist boloney talked about animals, seems to have designed a most unusual missile for dealing with it. The warhead consists of a skilled, experienced, professional animal trainer. The directive system – which is the most surprising component – is an acute, trained student of contemporary philosophy, psychology and literary criticism, who has paid particular attention to the work of Wittgenstein. The propellant is a general dislike of humbug and a particular commitment to the proper treatment and understanding of nonhuman animals. While any two of these elements have at times been found in combination before, the whole set certainly has not, any more than watches have been found on uninhabited planets or monkeys have typed out the Bible. My present business, however, does not extend to the theology of the matter: it is simply to consider the job as it has been done.
The view that is up for scrutiny is the one which is still taken for granted as established and scientific by most psychologists and many others of the learned. It is the view that outlaws all application of mental concepts to non-human animals. At an everyday level, it often takes the form of a factual denial, as if animals had been experimentally proved to be unconscious or incapable of thought. Officially, though, no rash suggestion is made that such empirical tests are possible. The view is a philosophical one about the limits of our knowledge. We cannot, it holds, attribute mental predicates to animals, because these would refer to hidden, unknowable phenomena, to private sensations quite outside our ken. Words describing thoughts or emotions are simply the names of such private states. If, therefore, we say that any animal is angry or frightened, that it hates cars, or is bored or puzzled, or is looking for its kittens, we are always making an illicit ‘anthropomorphic’ projection. If we cannot help using such words, we must put them in protective shudder-quotes to show that we do not really mean them. But really we ought to get rid of them entirely, replacing them by proper scientific descriptions of outside behaviour. Otherwise we are not real academics, but crude proles, contaminated by sentimental superstition.
Now this is such an incoherent view that it tends to leave its critics gasping. For some time, however, two lines of lethal criticism have been converging on it, and it is Vicki Hearne’s achievement to have put them together. The first line is the philosophical one about privacy. If words like ‘anger’, ‘fear’ and ‘boredom’ really were just the names of hidden private sensations, we could not, it is noted, use them to talk about human beings either, and such language could never have got off the ground. Words like these actually refer to ramifying, complex patterns of life, which have interlocking inner and outer aspects. Metaphysical behaviourism radically misunderstood how language works. That is why its attempts to reform it were doomed, and also largely why it has failed as a general map of the human scene – a failure that leaves it with no obvious qualification to dictate about the animal one either. Are animals so different? Here we encounter the second line of criticism, which is a factual one. People who seriously work with animals and know them well reject the behaviourist account root and branch. They report that any attempt to use it in their work turns out, not just unilluminating, but paralysing and often actually dangerous. Traditional ‘anthropomorphic’ ways of thinking, by contrast, fit the facts and lead to successful practice. Might this just be because the trainers were not academics and did not apply the concepts properly? Vicki Hearne’s experience is interesting here.
I noticed that in obedience and riding classes, people with training in the behavioural sciences hadn’t much chance of succeeding with their animals, and that the higher the degree held by the person, the worse the job of training was likely to be. And one of the reasons I was the audience for so many lectures on the wrongness of the trainers’ way of thinking and talking was that the psychologists and philosophers had to bring their animals to me because they couldn’t house-break them, induce them to leave off chewing up the children or, in the case of horses, get them to cross the shadow of a pole laid on the ground ... To the extent that the behaviourist manages to deny any belief in the dog’s potential for believing, intending, meaning etc, there will be no flow of intention, meaning, believing, hoping etc going on. The dog may respond to the behaviourist, but the behaviourist won’t respond to the dog’s response ... The behaviourist’s dog will not only seem stupid: she will be stupid.
This testimony should not come as a surprise. There have been plenty of others like it before. Thus the psychologist Donald Hebb has described how his department once decided to rationalise its dealings with animals by moving over entirely into behaviourist terminology. After a short time the project had to be abandoned, not only because the workers could no longer understand each other’s accounts of what was happening, but because they were continually getting bitten. The terminology is not just unfamiliar: it is one which makes the practice of certain everyday arts downright impossible. And we know that the same impossibility would arise for somebody who tried to use only behaviourist terminology for conducting their relations with other human beings.
What would be said if acceptance of a particular linguistic theory were found to interfere with people’s use of language, or acceptance of a particular theory about muscular control were upsetting their driving? Here, and still more plainly in the engineering field, people would surely think it was the theory that had some explaining to do. It would seem plausible that the practitioners knew some facts the theorists had missed. There are other cases, of course, where it is the practice that gets discredited – necromancy, say, or beating one’s children, or human sacrifice to improve the weather. But then there are other objections to these practices. Is there any reason why training dogs to guide the blind, or to find lost property, lost children and injured mountaineers, or to guard people, should be grouped with these objectionable practices?
The veto is not usually put in these terms, and of course practices like these are not normally mentioned at all in the ‘scientific’ literature in question. What is normally held to nullify all such evidence is its source. Evidence which comes from a particular kind of people – namely, ones who like animals, attend to them and spend a lot of time with them – is treated as tainted because such people are, by definition, ignorant, sentimental, confused, superstitious and non-academic. As Vicki Hearne points out, this principle has the strange effect of excluding all well-informed evidence from outside the tribe. Reliance on the principle received a rude shock when Professor Donald Griffin’s book The Question of Animal Awareness hove across the horizon. It is getting another now, for Vicki Hearne, too, is a fully paid-up American academic (she teaches English at Yale). Griffin used his solid realism and inspired common sense chiefly on the neurological and ethological aspects of this long-standing scandal. Ms Hearne carries the war further into the philosophical hinterland from which the deeper justifications for the behaviourist approach have always been held to issue.
What are these justifications? The epistemological one just mentioned does not stand alone: what else is involved? For instance, when students are carefully trained to translate what they say into behaviourist language, this practice is treated as embodying, not just sceptical caution, but certain positive factual beliefs about people and animals. What are these? Vicki Hearne remarks:
Another habit that students had, curiously, to be cured of was the habit of supposing that one animal might hide from another animal. (I have never known a hunter to be successfully cured of this habit of mind.) ... In order to hide, it was carefully explained, one had to have a concept of self. Not only that, one had to have the concept of self given by the ability to speak academic language, or at least a standard human language – a concept of self that depends on the ability to think. And, as one philosopher informed me unequivocally, any sort of thinking requires ‘first-order quantification theory’.
Here are two very startling factual beliefs. One is that animals manage to get all the benefits of hiding without actually having the least idea what they are doing. The other is that people have to know what they are doing in a sense which only a few trained scholars can lay claim to, in order to do something which is, as it happens, among the very first deliberate actions of their lives. Hiding games, such as playing peep-bo, are among the earliest games small babies take to, long before they have the slightest interest in trying to talk. And when, later, they continue to use hiding in play, they do it in a way scarcely distinguishable from that of puppies, kittens and other young animals who take equal delight in it. (That, if one may mention so vulgar a matter, is why children and animals can play together.) The contemporary idea of human uniqueness as centring on language quite often turns out in this way to mark, not the actual species barrier, but the line between certain trained Western adults and the rest of humankind.
This, unluckily, is a more up-to-date philosophical issue than the epistemological one about privacy. A lot of contemporary philosophers have fallen for these ideas, and their way of life often does even less to correct it by bringing them in touch with animals than it does for the psychologists. For people whose skills are predominantly verbal in the first place, the rewards of treating language as a self-contained system, covering the whole domain of thought and distinct from life, are very seductive. So seductive, indeed, as to prevent Wittgenstein’s later work, which exploded this paper empire, from being taken seriously, especially in America. What Wittgenstein pointed out was that language cannot possibly be a self-contained system, that it is seamlessly woven into the way of life to which it belongs. ‘To imagine a language is to imagine a form of life.’ Vicki Hearne says that this remark became intensely important to her when she found that she was inhabiting two completely different worlds of discourse – that of the trainers and that of the psychologists and philosophers. At first, she adds, ‘I tended to think of the trainers as skilful perhaps, but philosophically naive. I hadn’t noticed that genuine mastery of anything entails sound philosophical thought of one sort or another.’ (Compare Ryle’s insight that ‘knowing how’ is a deeper matter than ‘knowing that’).
It is not remotely plausible that sentimentality could be used to underpin all this complex, difficult and often dangerous work. In fact, as she points out, sentimentality wrecks it at once. Nor does it make much sense to talk of the trainers’ attitude as some sort of ‘necessary illusion’. What is needed is, first, to look realistically at the facts about high-grade training of dogs and horses – to see what the process actually involves, both for trainers and trained. Second, to see how well or badly these facts fit into our current notions about relationship, communication and meaning, about speech and understanding, and about what sort of a community such language presupposes. Third, to alter the notions to fit the facts, not try to reverse the process.
The carefully reasoned upshot of Vicki Hearne’s survey is that the training relationship is not only one of true mutual understanding, but is also a moral one, involving mutual trust and responsibility. She begins by contrasting the condition of a properly trained dog or horse with that of Washoe and the other talking apes. She records her shock on visiting Washoe and finding her locked in a cage. Adult chimps are very dangerous: to take one for a walk requires extraordinary precautions. Whatever their education, they may always attack. Vicki Hearne points out the strangeness of an academic approach to animal communication which concentrates wholly on a case not involving the basic trust on which real communities are built.
A good police-dog has not only a large vocabulary but also extraordinary social skills. He understands many forms of human culture and has his being within them. He can be taken to the scene of a liquor-store robbery and asked to search, with the handler trusting that he won’t molest the customers or other police officers or the clerk behind the counter. He knows what belongs and what doesn’t ... He can take down a criminal who is attacking his handler on Monday and on Tuesday play with the patients at the children’s hospital. These dogs then are glorious, but for anyone familiar with working dogs they are not surprising, any more than your pet dog is surprising in his or her ability to distinguish between your friends and strangers.
All this, however, is taken for granted as humdrum stuff, irrelevant for scientific purposes – no doubt because human dignity, which finds comparison with Washoe quite worrying enough, would not tolerate being put into the same-frame of reference as a police-dog. There is another difficulty, though, which occupies Vickie Hearne a good deal in the survey of training procedures and their meaning which takes up most of her book. This is the libertarian objection which many people today feel to the whole business of training – an idea of it as essentially oppressive and brutal. She gives a twofold answer.
First, this idea is factually mistaken, because training only succeeds if it is conducted as a co-operative enterprise. Domineering, power-seeking people are as sure to fail in it as sentimental ones. Secondly, what is very interesting philosophically: the use of imperatives is not a mark of brute force, an instrument for making agents act without consent and without understanding. Far from this, imperatives are central to language. ‘The possibility of giving a command is prior to (both historically and in terms of the unfolding structure of the syntax) such things as naming, referring, advising and full trust ... If I can’t say to you, should the occasion arise, “Duck” or “Put the broom in the closet when you’re done,” or “No mayonnaise on mine,” or “Lower down, Ah, that’s it,” then it is not clear what sort of relationship we can have.’ Imperatives, in fact, may simply be answers to the question ‘What shall I do now?’ and there are many situations where this is a perfectly proper question, for people as well as dogs. That both dogs and people often have a right to ask such a question and have it answered is a part of this book’s message. This part answers the moral case which often underlies the ‘scientific’ one for ignoring animal training: the belief that trainers manipulate animals by treating them as mindless cogs. I do not see what is left of the behaviourist position when Vicki Hearne has finished, though it will be interesting to watch how long academic inertia manages to delay recognising this. Five years?
Send Letters To:
London Review of Books,
28 Little Russell Street
London, WC1A 2HN
Please include name, address, and a telephone number.