Geoffrey Sampson


Empiricism v. Nativism

Nature or Nurture?

My book The ‘Language Instinct’ Debate offers novel insight into a classic controversy. (The ‘Language Instinct’ Debate is a new and enlarged edition of a book first published under the title Educating Eve. The new edition came out simultaneously in paperback and hardback in 2005.)

There is a very long-standing debate about the relative importance of nature and nurture in the development of the human intellect. Are we creatures endowed from birth with rich structures of knowledge and understanding, which require the stimulus of experience only in order to be jolted into conscious awareness? Or do we begin life essentially as blank sheets of paper on which the outside world writes what it may, and which begin with no predisposition towards one eventual set of contents rather than another?

Put as I have just expressed it, the view which assigns the major responsibility to nature rather than nurture may sound the more attractive of the two positions. But from another aspect, maximizing the role of nurture may yield the nobler picture of mankind. We think of ourselves as a supremely creative species. Other animals behave broadly as they always have behaved, but Man is fertile in novel ideas: wasps’ nests look the same as wasps’ nests have looked for millennia, perhaps for millions of years, but human constructions change dramatically century by century, indeed decade by decade. The life of a 21st-century Englishman would be largely incomprehensible to his own ancestors of a few centuries ago, let alone to a Chinese or an Indian near the dawn of civilization. Yet, if human thought processes are essentially preprogrammed, there is little or no true creativity underlying all this apparent inventiveness. What looks like the creation of a novel idea is really just the waking-up of a cognitive item that was waiting there all along, in the minds of many people who seemed unaware of it.

If on the other hand human cognition results from an interplay between experience and minds that are equipped with no specific initial contents, then we are bound to ascribe true creativity to the human mind. Individuals must be capable of “making ideas out of nothing”, bringing into being cognitive elements whose potential existence could in no way be inferred from the structure of men’s minds before the ideas were invented.

Plato, Descartes, Locke

Clearly this issue is not an all-or-nothing question. It is about where truth lies on a spectrum of possibilities. (And it is not to be decided by the relative attraction of alternative views but in terms of the evidence for them.) Nature must have some role in human cognition, otherwise it would be inexplicable that every human has a rich cognitive life and no stone has a cognitive life at all. Conversely, nurture must play a role — if human cognition depended exclusively on innate mental structuring, we would all be born able to name the capital of New Zealand and write good C programs.

But although the two extremes of the scale are both senseless, thinkers have differed heavily on whereabouts between the extremes the truth lies. The terms nativism and empiricism are used for views which emphasize the role of nature and of nurture respectively. (Rationalism is sometimes used as a synonym for nativism — but “rationalism” is ambiguous, it often refers to reliance on reason rather than emotion, which is another matter entirely.) The word empiricism is from Greek empeiria “experience” (nothing to do with empires!).

Classically, Plato held that the role of “nurture”, or experience, was far smaller than common sense might suggest. According to Plato, a child begins life with knowledge already present within him — there is no such thing as learning new things, “what we call learning is really just recollection” (Phaedo 72e, Hugh Tredennick tr.). Similar views were argued in the 17th century by René Descartes. Both Plato and Descartes stated their point of view in an extreme form (though the specific examples they offered in its support did not suggest innate knowledge of anything as contingent as the capital of New Zealand).

Conversely, Descartes’s English contemporary John Locke argued that there are no innate ideas.

How comes [the mind] to be furnished? ... Whence has it all the materials of reason and knowledge? To this I answer, in one word, from experience ... (Essay Concerning Human Understanding, Book II, §1.2)
Locke recognized that human beings are innately equipped with minds capable of a variety of operations, but he denied that this initial endowment includes any particular ideas or knowledge.

These contrasting concepts of human nature have tended to correlate with contrasting political ideals. Plato advocated a somewhat horrifically authoritarian State; the links between Plato’s authoritarian politics and his theory of ideas were drawn out in vol. i of Sir Karl Popper’s The Open Society and Its Enemies. Locke founded not only the empiricist philosophy of mind but also the liberal theory of politics which in the following centuries became the accepted political ideal of the English-speaking world. Empiricism and liberal politics are linked by the idea that freedom to experiment is needed in situations where authoritative knowledge is not given in advance.

Noam Chomsky and Linguistic Nativism

In recent decades, it has been argued that scientific study of language gives us new evidence favouring a strikingly nativist account of human cognition. According to this view, genetics fixes the contents of our minds just as it fixes the detailed structure of our bodies.

The argument was first constructed in the 1960s by the American Noam Chomsky, in books such as Cartesian Linguistics. In this and a series of subsequent writings Chomsky identified a large range of considerations (for instance, all human languages share certain universal structural features, and young children acquire their first language surprisingly fast), each of which, he urged, forces us to accept that “we do not really learn language; rather, grammar grows in the mind” (Chomsky, Rules and Representations, p. 134). According to Chomsky, genetics controls not only the nature of human language but human knowledge and belief — “our systems of belief are those that the mind, as a biological structure, is designed to construct” (Reflections on Language, p. 7). If there are topics for which mankind fails to come up with satisfactory scientific explanations after trying for a long time, this may be (Chomsky suggests) because the true theories of those topics are not included in the class of theories that human minds are biologically capable of formulating.

Chomsky’s nativism applies to cognition very generally. He envisages biological limitations to the development of the arts as well as science. But the evidence for it is almost all drawn from language.

This novel, biologically-based version of Platonism was extremely successful in winning converts in the 1960s and 1970s. Chomsky is a highly skilled rhetorician and polemicist; and his theories were given a helping hand by external circumstances. At the time when he was propounding these ideas about language and human nature, Chomsky was also the leading intellectual opponent of American participation in the Vietnam War. This made him an attractive figure to a generation of young Americans who (very understandably) did not want to be sent to fight in Vietnam, and were keen to listen to someone giving reasons why they should not have to. I lived in the U.S.A. myself at this period, and it was noticeable how enthusiasm for the anti-war message created audiences for the linguistic-nativism message.

Despite having won many distinguished converts, however, Chomsky’s linguistically-based version of nativism seems entirely misguided. My book The ‘Language Instinct’ Debate systematically lists all the separate strands of argument that Chomsky uses in his various writings to infer innate knowledge/ideas from observable facts about language. All of the arguments collapse on examination. In some cases the factual premisses about human language are factually false. In other cases the conclusions don’t follow. Sometimes the argument assumes what it purports to prove.

Indeed, The ‘Language Instinct’ Debate shows that some of the facts which Chomsky takes as evidence for innate knowledge of language actually tend to prove the reverse.

Alongside his nativist philosophy, Chomsky has in the past advocated a distinctive political ideal. Most commentators seem to have taken Chomsky’s linguistics and his political writings to be quite separate and unrelated, but my earlier book Liberty and Language showed that the two sides of Chomsky’s thought cohered as a single interrelated structure of ideas. Chomsky’s politics were in my view as unreasonable as his linguistic nativism.

In the 21st century, probably few people take Chomsky’s specific political ideas seriously any more. But the theory of language and mind which Chomsky initiated is now acquiring a sinister political relevance of another kind. My paper “Minds in Uniform” shows how nativist linguistics as developed by Chomsky’s successors is creating an intellectual foundation for the 21st-century version of imperialism – an obnoxious new political movement whose initial success stemmed from simple ignorance among present-day Westerners about the diversity of human cultures, but which can now quote linguistic nativism as allegedly demonstrating that cultures are really not – indeed, cannot be – diverse.

The New Wave of Nativists

During the 1980s, Chomsky’s nativist discourse moved out of the public spotlight. The world turned its attention to other political issues, and linguistics fissioned into separate subdisciplines, some of which had no relationship with the nature/nurture issue while others became too technical and inward-looking to make much impact on readers interested in broader philosophical and psychological considerations.

But, beginning in the 1990s, there has been a new wave of writing — often addressed to a more popular audience — which has revived essentially the same idea about language and knowledge being innate in human beings. These books refer to a broader range of considerations, including issues high in human interest such as case studies of pidgin languages, young children’s speech, experiments in teaching language to apes (whereas Chomsky’s arguments were rather dryly formal and mathematical).

Several different writers have contributed to this “new wave” of present-day arguments for nativism. By far the most influential, however, has been Steven Pinker’s 1994 book The Language Instinct, which must by now be one of the best-selling books on language for many decades. As my title suggests, it is chiefly Pinker’s book which I am directly addressing in The ‘Language Instinct’ Debate.

Pinker’s book is extremely well written. It has been praised in strong terms by a wide range of distinguished commentators. However, the fact that a case is argued in a way that is fascinating and enjoyable to read doesn’t make it true. The reasons for believing in a common-sense empiricist account of human language and cognition remain as undamaged by Pinker as they were by Chomsky.

Indeed Pinker and the other “new wave” linguistic nativists operate partly by taking Chomsky’s arguments for granted. And this strategy is rhetorically quite effective. When Chomsky originally spelled out an argument, the reader would assess it and might detect its fallacies; but when recent writers refer to something as having been established back in the 1960s–70s, most readers are likely to take this on trust, for lack of time and energy to check the sources.

If Chomsky’s arguments were fallacious, new arguments that rely on them can be no better founded. True, the present-day nativists do also claim to identify some extra strands of evidence, never mentioned by Chomsky. But The ‘Language Instinct’ Debate looks at these points too, and finds that they are just as misguided as the earlier claims. It is not just that the nativists’ arguments are illogical. Too often, Pinker and other modern nativists simply get the basic facts about language and languages wrong.

A new chapter in the 2005 edition of The ‘Language Instinct’ Debate looks in particular at how a cross-section of the population use the English language in everyday conversation. I draw on a large database of recordings of everyday modern spoken English to show just how unrealistic the nativists’ beliefs about language behaviour are. Ordinary speakers regularly do say things which the nativists claim nobody ever says, and systematically fail to say things which the nativists claim people often say. In speaking the way they find natural, the population at large blows the nativist case to pieces.

(If you want to read the gist of the case against linguistic nativism without wading through the full detail in my book, the bare bones were set out systematically in a talk I gave at one of the hautes écoles in Paris, now a web page.)

Silent Now the Oracles

A natural response to the foregoing would be “All right, that’s what you think – how do the nativists answer it?”

They don’t. I have been exposing the fallacies of linguistic nativism in print for forty-odd years now, but Chomsky, Pinker, and their followers have scarcely ever publicly alluded to my critique – Pinker never has, so far as I know. (This is a point identified as remarkable by Paul Postal of New York University, in the foreword he contributed to my book.)

Suren Naicker quotes Pinker as writing in 2004 that he had never found time to read my book in the seven years since its first edition appeared, because he was too busy with new books of his own! (See footnote 12, pp. 44–5, of Naicker’s Witwatersrand MA thesis, 2007.) I wonder how one squares this statement by Pinker with the fact that, in a book he published in 1999, he included his own take on a very specific argument original to my book, though without mentioning me or my book by name. (I discuss this episode on p. 128 of my 2005 edition.)

The nativists are not fools. If they drew the public’s attention to the existence of alternatives to their point of view, they know people would not go on accepting their implausible theories for long. They want the high-prestige invitations, the academic prizes, and so forth to carry on flowing in their direction. For that, they have to seem to the general public to tower above the fray, so that contrary views are not worth dignifying with an answer.

But, to people who understand how science functions, this silence amounts to an admission of intellectual defeat. Reviewing my book in Language (the world’s premier linguistics journal), John McWhorter comments that it

makes a powerful case that linguistic nativism … risks looking to scientists in a hundred years like the search for phlogiston does to us now.
He adds:
claims that linguistic nativism is less a theory than a cult start looking plausible.

— Incidentally, it is not clear that Chomsky himself any longer believes in linguistic nativism. He is an old man now; in 2002 he put his name to a paper co-authored with Marc Hauser and W. Tecumseh Fitch (Science, 22 Nov 2002) which, taken as a development of Chomsky’s theories, would reduce them to vacuity. But even if Chomsky is no longer a Chomskyan, many other influential thinkers still are Chomskyans. The Hauser/Fitch/Chomsky piece led to a rebuttal (followed by counter-rebuttal, counter-counter-rebuttal, …) by Steven Pinker and Ray Jackendoff, another leading nativist linguist, in the journal Cognition. Chomsky’s linguistics books are still being read; and, whatever he thinks or does not think nowadays, Chomsky has never said in so many words “I was wrong”. So it needs to be said for him.

Culture, not Instinct

The idea that people begin knowing nothing but capable of learning anything may be boringly traditional; it remains the most adequate account of the facts. The ‘Language Instinct’ Debate looks at the complete spectrum of linguistic arguments for nativism — from Chomsky’s works of fifty or sixty years ago to the new wave — and shows that they are all built on sand.

The message of my book is that there is no “language instinct”.

When I wrote The ‘Language Instinct’ Debate, I supposed that belief in a “language instinct” was just a mistaken idea that happened by chance to have appealed to a lot of academics who overlooked the abundant evidence showing that it’s wrong. More recently I have come to see it as a consequence of the way in which universities in the 21st century have been changed into businesses which prioritize the pursuit of revenue over the disinterested pursuit of truth via painstaking exact scholarship.

It isn’t an accident that so many academics have embraced this particular idea about human cognition: there are more academic jobs in linguistic nativism than there could ever be in the contrary belief. (I develop this point in a newer book, The Linguistics Delusion.) Similarly, it might take a lot of astrologers to work out all the various ramifications for different areas of life of Saturn being in the ascendant, Mercury being in opposition to Mars, and so forth – but it does not need many people to expound the simple truth that the planets have no influence on human affairs. I have even found one commentator, Stephen Cowley, who explicitly wrote (in Language Sciences vol. 23, 2001, p. 85) as if the employment issue were a valid reason for embracing nativism! But universities cannot just be self-perpetuating job creation schemes. Unless academics are honest about seeking truth, whether it be convenient or inconvenient, we truly are a worthless bunch.

The ‘Language Instinct’ Debate shows that anyone who is willing to look at the evidence in good faith must agree that language, and indeed human intellectual achievements more generally, are in all the ways that matter purely cultural developments. Only to a trivial extent are they predetermined by our biology.

Geoffrey Sampson

last changed 14 Sep 2022