Piaget v. Chomsky

No one has ever achieved quite as powerful a position in Psychology as Jean Piaget holds today. His considerable success is due partly to the strength of his massive and comprehensive theory about childhood and partly to his personal history. For many years, he fought a lonely battle against Behaviourism with the major school of Psychology at the time. The main aim of the Behaviourists was to explain behaviour in very simple reflex-like terms, and to them Piaget’s appeal to complex intellectual structures was as absurd as it was unscientific. But it was the Behaviourists’ insistence on simplicity which led to their own collapse: in the end, it became embarrassingly obvious that their theory could not account for the complexities of the behaviour of rats, let alone for human behaviour. The Behaviourists’ failure, together with the general advance of structuralism, led to a widespread assessment of Piaget’s long and Ayatollah-like isolation. To many people, it had seemed an eccentric folly: now it began to look like a courageous fight for a sane approach to human behaviour.

He became, and has remained, very popular, but his supporters still have to overcome some forbidding obstacles. One is that he writes too much, usually at the rate of several books a year. A second is that he writes very badly. A third – less reprehensible is his use of a formidable variety of sources: unlike most psychologists, he is at home in biology, philosophy and mathematics and he seems to read more deeply in these subjects than he does in psychology. Someone like this creates a demand for a middleman, a person who will boil the master down and explain his achievements clearly and concisely. Many people have tried to do this, but though they often write more clearly than Piaget, they do not usually manage to capture the breadth of his work. They tend to concentrate either on the psychological or on the philosophical side of his theory and to leave the rest well alone.

Breadth, however, is the great strength of Margaret Boden’s introduction to Piaget. In a short but extremely versatile book* she deals with the psychological, philosophical, mathematical and biological sides of his work simply and competently. She has an interesting chapter on the relationship between his ideas and the relatively new subject of artificial intelligence, and she covers many of the criticisms that have been made of Piaget’s work. Given its length, her book cannot be completely comprehensive. But it is a marvellous starting-point, easily the best there is, for anyone setting out to understand Piaget.

Boden’s virtual omission of, for example, the question of language is understandable: Piaget has never actually done any research to find out how children learn to speak with such ease between the ages of one and five. But he does have views on the subject, which he has expressed very trenchantly. These views about the child’s mastery of language that it stems from the experiences which children have very early on in life, and that it can be explained by a general theory of intellectual development brought Piaget into head-on conflict with another equally famous structuralist, the linguist Noam Chomsky. Like Piaget, Chomsky thought that the Behaviourist account of intelligent behaviour in general, and of the acquisition of language in particular, was totally inadequate, but unlike Piaget he argued that there are highly specialised mechanisms for learning language and that many of these must be innate.

This disagreement prompted a conference at Royaumont in France in 1975, organised by Massimo Piatelli-Palmarini. Piaget and Chomsky gave the opening papers, and there followed a set of talks, some by outright supporters of one or the other of the two leading participants, and others by less committed people, most of whom were trying to find some compromise between the nativist views of Chomsky and the so-called constructivist position of Piaget. The main papers and much of the ensuing discussion, in which Chomsky and Piaget took part, have now been collected in one extremely well-edited volume† which is interleaved throughout with commentaries by Piatelli-Palmarini.

The debate was a lively one, and on the whole it seems to have gone Chomsky’s way. His main argument is that there are some grammatical rules which we could not possibly learn entirely from scratch. We know, for example, that the interrogative form of the sentence ‘The man is tall’ is ‘Is the man tall?’ We also know that the sentence ‘The man who is tall will leave,’ put as a question, would be ‘Will the man who is tall leave?’ and not ‘Is the man who tall will leave?’ Chomsky’s point is that we never make the sort of mistake found in the last sentence, even as children learning the language for the first time. Yet if children were operating at the surface and trying out a rule according to which, for example, one forms a question by starting with the first verb (‘Is the man tall?’), they might very well say things like ‘Is the man who tall…?’ They do not, according to Chomsky, because they are not operating at the surface. They have a more or less, immediate grasp of the underlying structure of the sentence, and Chomsky’s question then is: how do they come by this understanding? His answer is that it could not possibly be learned, because there is nothing in the child’s experience which could teach him about the underlying grammatical relationships of the sentences which he hears. His challenge to Piaget and his colleagues, which he repeats time and again in this volume, is that they should show him how children could learn rules of this sort from their interactions with their environment. It is a challenge which is never taken up. If there is an answer to Chomsky’s question, it was not given at the Royaumont conference.

Why not? It may be that Chomsky is right and that there is no answer. But there is another possibility which is never very clearly discussed in this book, and it centres on the difficulty of establishing any kind of causal hypothesis about children’s development. By and large, child psychologists are concerned with two quite different questions. They try to establish exactly what changes happen during childhood. Children, they show, typically crawl, then walk, master this grammatical Construction before that one, think about three-dimensional space in one way and later in a completely different way. These developmental sequences are easy enough to establish. All you have to do is look at children when they are, say, three years old, and again, six months later, and again when they are four, and plot the changes in their behaviour. These are often dramatic, which means, of course, that they can be demonstrated without much difficulty. That is why child psychologists have no great empirical problem with the first question (Question 1) about the ways in which children change as they grow older.

But Question 2 is more difficult. This is the causal question. Given that children change, what makes these changes happen? The problem is that the changes happen outside the psychologist’s laboratory. If you cannot see a thing happening it is very difficult to say what makes it happen. This, no doubt, is why Piaget’s causal arguments are quite generally unsatisfactory. They do not really tell us what makes children become more logical as they grow older and better at understanding physical and social rules, any more than they answer Chomsky’s question about language. They are too vague. Piaget talks of changes happening because the child is thrown into intellectual disequilibrium. The child’s existing intellectual structures cannot cope with things that happen to him, and so he reorganises them and thus reaches a new stage of intellectual development which has its own temporary equilibrium. Later, the whole process happens again, and then again and again – in exactly the same way from infancy through to adolescence. Such a theory avoids brass tacks and it is easy to see how it could not answer Chomsky’s pointedly specific question, or indeed any other specific causal question about children’s development. The few attempts made by Piaget’s colleagues to provide some empirical support for his causal hypotheses about intellectual development were all miserable failures. But this does not mean that all the changes are innately determined. Perhaps with a more incisive hypothesis, and with better methods (not yet devised) for testing causal hypotheses, it might be possible to show how a child’s experiences teach him logical principles, or even Chomsky’s rules.

To Question 1, Piaget has more definite answers. He thinks that children start their lives in a state of almost total ignorance and ineptitude, and that it takes a very long time before they master even the simplest logical rules. The principle of invariance, class inclusions, transitive inferences, cardinal and ordinal number, the Euclidean framework, moral imperatives these are all completely out of the young child’s reach. Ostensibly, the Royaumont volume is not about this side of Piaget’s work. Chomsky and he disagree, not on the way children’s language develops, but only on what makes it develop. But towards the end of this volume two papers throw considerable doubt on Piaget’s extravagant claims about young children’s inabilities. Mehler shows (as Margaret Boden’s book clearly confirms) that Piaget’s empirical evidence that children are illogical is usually very weak because it depends on tests which seem deliberately designed to mislead. In other experiments which make the same logical demands as Piaget’s, children manage pretty well. Another argument, convincingly developed here by Fodor, against Piaget’s theory about logical development is that children who start life as ineptly as Piaget claims would be quite unable to build up all the intellectual structures which form the centre of Piaget’s theory of development.

Piaget’s strengths and weaknesses are clear. Almost single-handed, he rescued the study of intellectual development from sterility. He posed a long set of extraordinarily interesting and erudite questions about children’s development. He produced a coherent theory about the nature of this development, though a far less compelling account of its causes. But his empirical evidence is often unconvincing, and he betrays a certain unprofitable impatience with explanations other than his own. The beauty of his work, and its blemishes, are accurately captured in Margaret Boden’s account and in the volume edited by Piatelli-Palmarini.