Native speakers of Pirahã, in the Amazon lowland jungle, have no words for left or right, they use the same term for blue and green, and their definitions of red, black and white turn out to be similes, rather than dedicated words.
These once-isolated people, a tiny group, have no system of numbers; their sentences cannot accommodate subordinate clauses or other forms of recursion (embedding phrases), and they are not impressed by the Gospel of St Mark in Pirahã, not least because it is a story composed by someone they do not know, about someone they have never heard of, in a time and place that has no meaning for them. The Pirahã people tend to confine their discourse to things they know about, and their verb forms can be suffixed to distinguish between hearsay, inference and observation. They have no perfect tense.
On the other hand, they can also sing, hum, yell and whistle information to one another. So they have four additional speech forms as well as a very precise vocabulary for their environment and everything in it that matters to them. If there is some deep structure that underpins all 7,000 human languages – a universal grammar or language acquisition device or language instinct, already hard-wired in the human brain at birth – Pirahã seems to be an exception.
For Daniel Everett – linguist, anthropologist and once an evangelist missionary in the Amazon – the case settles an old argument about the nature of language. The exceptional language of the Pirahã people seems to be a unique cultural tool – like their knowledge of plant toxins, and their ability to fish with a bow and arrow – adapted for their exceptional circumstances. It is just another finely honed instrument from the human cognitive toolbox: we have large brains, we are social animals, we co-operate, we have a lucky arrangement of lungs, larynx, pharynx, palate, tongue, teeth and lips. We can speak, and so language has evolved, just as our brains and bipedal locomotion have evolved.
Language, in the Everett formula, is the sum of cognition plus culture plus communication. There is no need for a language instinct to set a three-year-old suddenly talking nineteen to the dozen. The infant's ambient culture compels the order of subject, verb and object, the potency of individual words and phrases (such as "nineteen to the dozen"), and the precise choice of phonemes.
This claim has reportedly annoyed the hell out of other linguists, among them Noam Chomsky, one of the high priesthood of the discipline, and the founder of the belief in what, for shorthand, is called a universal grammar. It also presents a challenge to the arguments of the psychologist Steven Pinker, author of The Language Instinct, a 1994 bestseller. The notion of language as an innate human talent received a colossal fillip that year with the identification of one British family, some of whose members, through three generations, were perfectly ordinary, while others had a very precise and puzzling problem with the rules of language. This was interpreted as evidence for a "grammar gene".
This, to be fair, was before the genome of even the simplest bacterial organism had been sequenced, during an era in which researchers were betting that humans inherited more than 100,000 genes, perhaps even a million. Among these might be a gene for schizophrenia, a gene for intelligence, for being good at the 100m sprint and for learning to manipulate sentences.
The picture has changed since the human genome project ended in 2003. The awesome bundle of human complexity turned out to be delivered by about 23,000 genes; many more than a fruit fly, certainly, but many fewer than the maize plant. Whatever it is that lets us relish the preposterous loquacity of Mr Micawber, condemn the hubris of footballers and compile scenarios for a Greek debt default, all on a brief bus ride, it won't be a simple genetic turn of the screw in a larger than usual primate brain.
This, however, is not an argument in which the rest of us stand much chance of judging which side is right. Everett is the western maestro of Pirahã: what we know about the language is what he tells us, and what he tells us is so strange that some linguists have even asked if he might not be the victim of some sustained Borgesian prank (it wouldn't be the first time that an isolated community with an advanced sense of humour had deadpanned a visiting anthropologist).
But beyond that lies another difficulty: arguments about what the shape and capacity of a language tells us are quite difficult to follow. Is recursion really the thing, as Chomsky claims, that makes the difference between the pragmatic social communication of great apes and the infinitely resourceful language of humans? Is Pirahã really such an exception? Does the number of pronouns in a language really correlate inversely with a culture's technological advance? Do words for colour really provide an index of the way a language has advanced?
There is another reason for confusion. Everett – once a kid in a Texas rock band, before he became a trainee missionary – is the kind of scholar often described as "flamboyant". He has not, however, written a flamboyant book. During the heaviest going, he wades into academic discourse rather as an Amazon traveller might wade through a lagoon in the rainy season, stirring up more mud than clarity. "More ordered thoughts and communication would have been facilitated in a beneficent cycle of evolutionary improvement," he says, in a passage I don't suppose he could translate into Pirahã.
It seems churlish to complain that a book about language is a bit prolix, but fewer words might have forced more pointed argument. That said, most of it is revelatory. There is nothing about humans that is quite as astonishing as language. It is a finite resource with seemingly infinite possibilities. It constrains thought, but it also delivers vision far beyond its apparent evolutionary purpose.
Persevere through the thickets of disciplinary defensiveness and the components of language become clearer: a theory of mind, intentionality, contingent judgment, a lot of cultural expectation and some biology that makes possible not just the usual vowels and consonants but also words made up of clicks, glottal stops, plosives and the voiceless dental bilabially trilled affricate (rare, even in the Amazon). As a bonus, you will also learn why you shouldn't talk with your mouth full.
• Tim Radford's The Address Book will be published by Fourth Estate in April.
Making sites popular is about more than Twitter and Facebook. Burgess Powers makes social: personal