THERE IS no word, in the language of the Amazonian Piraha tribe, to describe the number of columnists on this page. To the fascination of anthropologists, the Piraha cannot count: they have a word for “one” (which usually means “roughly one”); and a word for two (which tends to mean “not many”, and sounds exactly like the word for “one”, only with a rising inflection); and they have a word for “many”. A Piraha version of
Ten Green Bottles would not take long, then: one, two, a lot.
Linguists have long pondered the question of what evolutionary hiccup left the Piraha less numerate than ducks (apparently they can count to four), but now Peter Gordon, a behavioural scientist at Columbia University in New York, has come up with a simple but intriguing answer: words for numbers over three simply do not exist in the Piraha lexicon, and never have; in other words, the numerical universe of the Piraha is precisely circumscribed by the tribe’s vocabulary.
The research on the Piraha, published in Science this week, seems to reinforce the famous Sapir-Whorf thesis, which holds that every individual’s perception of the world is directly moulded by language. We are what we speak or, as Wittgenstein put it: “The limits of my language are the limits of my world.”
But does language determine thought, or the other way round? Do Koreans avoid the first person (seldom saying: “I want a biscuit”; more usually: “a biscuit would be nice”) because they are naturally modest, or have they become unnaturally modest because arbitrary patterns of language have imposed that trait? Do Chinese eschew a direct “yes” or “no” out of an innate indecisiveness and unwillingness to give offence, or is the language doing the shaping? Inuits can describe snowfall in ways that we cannot imagine, because their language provides multiple words for snow; but on subjects other than frozen water, Inuit conversation can be rather restricted.
In a recent essay, the writer Amy Tan, who is of Chinese origin, wrote: “It’s a dangerous business this sorting out of language and behaviour.” For her own part, she was offended by the notion that the lack of definitive “yes” and “no” in the language made Chinese people more irresolute; yet she also wondered whether linguistic programming explained why she can’t just tell those infuriating telephone salesmen to get stuffed.
Language does undoubtedly mould, and perhaps restrict, both national character and style of writing. English, for example, has many of the virtues associated with Britishness: expansive, flexible, eccentric. Yet compared with other languages it can seem inelegant and unromantic. The Czech novelist, Josef Skvorecky, complains that for all its versatility, English lacks “the sex appeal of feminine endings, the lure of verbal aspects, the capricious scherzo of prefixes.” Ours is a can-do language, but uptight with it.
It is this blend of national character and linguistic peculiarity that gives every language its central, untranslatable essence. The Piraha may have no word for “three”, but these hunter-gatherers have numerous other words and concepts that simply would not translate into our language. Take the Bantu word identified recently in The Times as the most untranslatable in the world: ilunga, in the Tshiluba tongue, means “a person ready to forgive any abuse the first time, to tolerate it a second time, but never a third time”. Of course, the word means something far richer than that, but only an individual immersed in the world it emerges from, where offence, honour, family, tribe, feuds and memory are paramount, could appreciate its full meaning. Conversely, the word “googly” has no meaning in the lowland Amazon, the Piraha not being keen cricket-players, probably on account of the scoring.
Trying to impose cross-cultural meaning can be disastrous, and hilarious. When Captain Cook arrived on the coast of northwest America in 1778, a group of Indians paddled out in canoes and shouted “nootka”, a form of greeting which roughly translates as: “Why don’t you take your big floating house around the other side of the bay where it’s less windy”. Cook assumed they were referring to themselves, and the name stuck; even their language is known as Nootkan, based on a mistranslation from itself.
Some linguists fear that with the spread of English and the internet, the idiosyncratic essence of individual languages will gradually vanish, boiled into a soupy and debased universal tongue: to turn the Sapir-Whorf thesis on its head, our perceptions of the world will not vary much, since we will all speak the same language, becoming mutually comprehensible, and very dull.
But language, the greatest of all human inventions, is more resilient than that, constantly evolving and adapting. The internet has forged numerous, sometimes rather wonderful mutant forms of English, just as the collision between languages in the past has produced some of the most remarkable literature. The great strength of English is its multiple borrowings from other languages and influences, layer on layer. Joseph Conrad’s richly evocative, finely-wrought prose, for example, owes much to his first language being Polish. Writing in another tongue forced him to think even harder about meaning: “If I had not written in English, I would not have written at all,” he once said.
Language does condition thought, but also changes with behaviour and thinking. Future generations of the Piraha will doubtless learn to count in the conventional way, but culturally they will probably continue to think: one, two, a lot. The numerically limited language of the Piraha may seem incomprehensible to us, but it has its own, inner coherence.
Which inevitably brings us to George W. Bush, who has brilliantly invented a language only he can truly understand, the perfect example of the fraught relationship between language, meaning and thought. As the President observed to an audience in Oregon last week: “I hope you will leave here and walk out and say: ‘What did he say?’ ” And that is exactly what they did.
Join the Debate at comment@thetimes.co.uk
