Many a slip between cup and lip

Noam Chomsky's sleight of word fails to convince Jacques Guy.

Despite the general appeal of its subject - how we learn to speak - this new book by Noam Chomsky seems to be intended for cognoscenti only. The Chomskyan school of linguistic theory has experienced an explosive growth since its beginnings in the 1950s. With it grew a jungle of acronyms and opaque terms: TG, UG, Government and Binding theory, c-command, X-bar theory, and so on. Readers are expected to be familiar with them as Chomsky dispenses with explanations, and there is no glossary.

In a nutshell, his thesis is that, as new-born infants our daily exposure to our mother tongue is grossly insufficient to account for our early mastery of it. (This is called "poverty of the stimulus".) How then can we learn? Because we are born with some sort of hard-wired circuit. A hard-wired circuit where only a few crucial connections, a few switches, remain undecided. Flick a switch here, another there, and you have English. Flick some more here and there, and you have Chinese. This would neatly explain how we soon become fluent in our mother tongue, despite, according to Chomsky, having so little exposure to it. In his own words: "We can think of the initial state of the faculty of language as a fixed network connected to a switch box; the network is constituted of the principles of language, while the switches are the options to be determined by experience. When the switches are set one way, we have Swahili; when they are set another way, we have Japanese."

These "principles of language" - which, it is claimed, apply to all languages - are innate; however, setting the switches is learnt. From which comes Chomsky's "minimalist program": find the switches, and figure out what you get when you flick this and that switch. The new quest for the handful of switches and their settings that determine the language we speak, supersedes and voids all his past theories and constructs, writes Chomsky: "The minimalist program seeks to show that everything that has been accounted for in terms of (levels of deep and surface structure) has been misdescribedI for those of you who know the technical literature, that means the projection principle, binding theory, Case theory, the chain condition, and so on.... There should be no indices or phrasal units and no bar levels (hence no phrase-structure rules or X-bar theory...)".

It must be difficult for anyone unfamiliar with Chomskyan linguistics to grasp the enormity of this about-turn. Imagine Einstein late in his career declaring his theories misguided all along, and that space is not curved after all. But the analogy stops there, for the twin foundational weaknesses of Chomskyan linguistics perdure: its divorce from the realities of language, and its ubiquitous confusion of the concepts symbolised by words with the words themselves.

Consider, for instance, this thought experiment: "Suppose [a] cup is filled from a tap connected to a reservoir in which tea has been dumped (say, as a new kind of purifier). What is in [this] cup, is water, not tea, even if a chemist could not distinguish it [from the contents of a cup of tea made in the usual manner]." That is what I call a linguistic illusion - magician's legerdemain or rather, if the word existed, "legerdemot". How is the reader tricked into confusing tea with water? Very simply - through Chomsky's use of a non-standard definition of water: "H²O give or take certain impurities." Stop and think for a moment. Certainly, that definition covers water, but it also covers ice, clouds, steam, rain and, with increasing amounts of impurities, beer, wine and even you, readers, who are 70 per cent H²O and the rest "impurities". Since the definition is inept, the subsequent discussion has as little to do with the reality of language as the deferents and epicycles of Ptolemaic astronomy have to do with the reality of the solar system. Look up "water" in any dictionary, and Chomsky's conundrum vanishes: "a colourless liquid essential for animal and plant life and which constitutes, in its impure form, rain, oceans, etc."

Other arguments are divorced from physical reality. Thus we are told that "most of the universe's water exists in the glassy state (in comets)" and then we are invited to a thought experiment on a twin-Earth, "where they happen to make their glasses from tails of Earthly comets. Suppose Earthly Oscar arrives on Twin-Earth and asks for water, pointing to G [a glass filled from a tap]. Is he right if he is referring to the glass and wrong if he is referring to its contents?" But on Twin-Earth where glasses are made of ice, water poured into one will either melt it, or freeze. So Earthly Oscar is asking for water, pointing either at a puddle or at a block of ice. The attempted distinction between glass and contents is just another meaningless conundrum.

Again, think about how you were tricked into seeing a meaningful question in this. Instead of being told that glasses on Twin-Earth are made of ice, you are told that most of the water in the universe occurs in comets in "the glassy state" (note how the word ice is avoided) and that Twin-Earth glasses are made "from tails of Earthly comets". ("Earthly comet" is a contradiction in terms; and it is the core of comets that contains ice, not the tail, which is fine dust, practically insubstantial.) Your attention has been diverted by this comet story long enough for you to forget that water poured into ice turns into ice or melts the ice - and to miss how a jabberwock has been pulled out of a linguistic hat.

Still other arguments are devoid of sense. Thus: "we do not expectI the theory of language to deal with the fact that Chinese is the language of Beijing and Hong Kong, though Romance is not the language of Bucharest and Rio de Janeiro". Since Romance is not a language but a language family, and since the Chinese of Beijing and that of Hong Kong are two very different languages, that makes as much sense as: "we do not expect number theory to deal with the fact that six is prime, although pi is even."

There are even arguments that manage to contradict both logic and intuition. Thus: "If London is reduced to dust, it - that is, London - can be rebuilt elsewhere and be the same city , London." Never mind that Londoners might beg to differ there, Chomsky continues: "If my house is reduced to dust, it (my house) can be rebuilt elsewhere, but it won't be the same house." So when London is destroyed and rebuilt in Antarctica, it is the same London but its houses are not the same houses?

Immediately after destroying and rebuilding London and his house, Chomsky adds: "If the motor of my car is reduced to dust, it cannot be rebuilt, though if only partially damaged, it can be", and comments on this:

"Pronouns involve dependency of reference, but not necessarily to the same thing; and both referential dependence and the narrower notion of sameness involve roles in a highly intricate space of human interests and concerns. Judgments can be rather delicate, involving factors that have barely been explored." Readers furrowing their brows to figure out what that could possibly mean are likely to overlook that "rebuilt" has just been used in two different meanings: to build from scratch, and to do up or restore. More "legerdemot".

The trick would not work in French, though, because French has two separate words ( reconstruire and refaire ) that force French speakers to distinguish the intended meaning. Likewise: "we can paint the door white and walk through it." This time, the trick works in French but not in Sakao, a language of Vanuatu. What you paint is the aghorhal (literally, path blocker) and what you walk through is the anathhal (literally, path opening).

Chomsky's above reference to the "notion of sameness" and the "highly intricate space of human interests and concerns" aims at reinforcing his argument of the "poverty of the stimulus": infants have too little experience to resolve such intricacies. It then leads him to claim "that there is something like an array of innate concepts", including such notions as carburettor and bureaucrat . Here, I am afraid most readers will suspect me of completely misunderstanding Chomsky. Yet, he writes: "Some, for example [the philosopher] Hilary Putnam, have argued that it is entirely implausible to suppose that we have 'an innate stock of notions' including carburetor and bureaucratI If he were correct about this..." - which clearly implies that Putnam is incorrect and that the supposition is at least not entirely implausible. But Chomsky goes much further, claiming that far from being not implausible, "the argument is at least in substantial measure correct even for such words as carburetor and bureaucrat, which, in fact, pose the familiar problem of poverty of stimulus if we attend carefully to the enormous gap between what we know and the evidence on the basis of which we know it." So, it must be the Chomskyan thesis ("poverty of the stimulus") that generates the conclusion that we are born with the notions of bureaucrat, carburettor, and, presumably too, gearbox, chihuahua, unicorn, carnival, pterodactyl and deuterium, and - why stop there? - Klingons and the face on Mars. Even if you accept that, you should still ask what explanatory value such a theory has, and whether it is not the linguistic equivalent of a theory that the universe was created on November 23 1999, about tea-time, complete with its past, right back to the big bang. Perhaps, on the other hand, you may entertain some doubts about the soundness of a theory of language that can claim that the concept of bureaucrat is innate.

A final question remains to be asked. In what way does the minimalist program, the search for "language as a fixed network connected to a switch box", contribute to the advancement of our knowledge - as distinct from, say, the neural nets of computer science? or even Petri nets? And if, for the sake of argument, we grant the innateness of linguistic concepts, in what way does that theory contribute anything at all to the question of how we learn to relate words and concepts? Is not explaining language by Chomskyan innateness like explaining the diversity of species by divine creation?

In this book, Chomsky razes to the ground the churches and cloisters of his early brainchildren. A good thing. But he does not offer us any sound foundations upon which to rebuild.

Jacques B. M. Guy, a student of Chinese, Japanese and Tahitian, used transformational grammar to teach French and Latin before receiving a PhD for his thesis on Sakao, a language of Vanuatu, from the Australian National University, and later turning to natural-language understanding.