Monday, December 8, 2014

Why language is neither an instinct nor innate

The ideas of Noam Chomsky, popularised by Steven Pinker, come under fire in Vyvyan Evans's book The Language Myth: Why language is not an instinct

IS THE way we think about language on the cusp of a revolution? After reading The Language Myth, it certainly looks as if a major shift is in progress, one that will open people's minds to liberating new ways of thinking about language.

I came away excited. I found that words aren't so much things that can be limited by a dictionary definition but are encyclopaedic, pointing to sets of concepts. There is the intriguing notion that language will always be less rich than our ideas and there will always be things we cannot quite express. And there is the growing evidence that words are rooted in concepts built out of our bodily experience of living in the world.

Its author, Vyvyan Evans, is a professor of linguistics at Bangor University, UK, and his primary purpose is not so much to map out the revolution (that comes in a sequel) but to prepare you for it by sweeping out old ideas. The book is sure to whip up a storm, because in his sights are key ideas from some of the world's great thinkers, including philosophers Noam Chomsky and Jerry Fodor.

Ideas about language that have entered the public consciousness are more myth than reality, Evans argues. Bestsellers by Steven Pinker, the Harvard University professor who popularised Chomksy in The Language Instinct, How the Mind Works and The Stuff of Thought, come in for particular criticism. "Science has moved on," Evans writes. "And to end it all, Pinker is largely wrong, about language and about a number of other things too..."

The commonplace view of "language as instinct" is the myth Evans wants to destroy and he attempts the operation with great verve. The myth comes from the way children effortlessly learn languages just by listening to adults around them, without being aware explicitly of the governing grammatical rules.

This "miracle" of spontaneous learning led Chomsky to argue that grammar is stored in a module of the mind, a "language acquisition device", waiting to be activated, stage-by-stage, when an infant encounters the jumble of language. The rules behind language are built into our genes.

This innate grammar is not the grammar of a school textbook, but a universal grammar, capable of generating the rules of any of the 7000 or so languages that a child might be exposed to, however different they might appear. In The Language Instinct, Pinker puts it this way: "a Universal Grammar, not reducible to history or cognition, underlies the human language instinct". The search for that universal grammar has kept linguists busy for half a century.

They may have been chasing a mirage. Evans marshals impressive empirical evidence to take apart different facets of the "language instinct myth". A key criticism is that the more languages are studied, the more their diversity becomes apparent and an underlying universal grammar less probable.

In a whistle-stop tour, Evans tells stories of languages with a completely free word order, including Jiwarli and Thalanyji from Australia. Then there's the Inuit language Inuktitut, which builds sentences out of prefixes and suffixes to create giant words like tawakiqutiqarpiit, roughly meaning: "Do you have any tobacco for sale?" And there is the native Canadian language, Straits Salish, which appears not to have nouns or verbs.

An innate language module also looks shaky, says Evans, now scholars have watched languages emerge among communities of deaf people. A sign language is as rich grammatically as a spoken one, but new ones don't appear fully formed as we might expect if grammar is laid out in our genes. Instead, they gain grammatical richness over several generations.

Now, too, we have detailed studies of how children acquire language. Grammatical sentences don't start to pop out of their mouths at certain developmental stages, but rather bits and pieces emerge as children learn. At first, they use chunks of particular expressions they hear often, only gradually learning patterns and generalising to a fully fledged grammar. So grammars emerge from use, and the view of "language-as-instinct", argues Evans, should be replaced by "language-as-use".

The "innate" view also encounters a deep philosophical problem. If the rules of language are built into our genes, how is it that sentences mean something? How do they connect to our thoughts, concepts and to the outside world?

A solution from the language-as-instinct camp is that there is an internal language of thought called "mentalese". In The Language Instinct, Pinker explains: "Knowing a language, then, is knowing how to translate mentalese into strings of words." But philosophers are left arguing over the same question once removed: how does mentalese come to have meaning?

Here, Evans switches from demolishing the old world order to previewing the new. He roots language in the things we do as living, moving creatures. Take the task of hammering in a nail. Unsurprisingly, a part of our motor cortex will leap into action as we work. The surprise is that, when you use or hear an expression such as "he hammered the nail", then the same bit of the brain lights up as when we see or do hammering. Our thoughts aren't taking place in an abstract mentalese but are "embodied", arising directly in and from experience.

That is a starting point. Evans maps out how we might get from these basic concepts to much more abstract ones, like love. And he goes on to show how different a "language-as-use" grammar will be. We expect that "words" and "grammar" – that is, rules for combining words – are wholly separate, but there's a surprise in the new view: there is no systematic distinction between words and rules.

Hopefully your first reaction to that is "What!" and your second is to reach for the book, because, time and again, it will give you the particular thrill of questioning received wisdom. Whether the new view will lead to a revolution in linguistics, or whether the empire will strike back with overwhelming force we don't yet know. But we are assured that Evans has a further episode, provisionally titled The Meaning Making Engine, in production.
Anderson, Alun. 2014. “Why language is neither an instinct nor innate”. New Scientist. Posted: October 20, 2014. Available online:

No comments: