The Evolution of Language

Animals that know a thousand words exist; animals that join two of them do not

There was a border collie called Chaser whose owner trained her four or five hours a day for three years. In the 2011 paper she knew the names of 1,022 toys — and the researchers randomly re-paired names with commands to show she knew the name pointed at the object itself. Training stopped because her owner ran out of time, not because she stopped learning.

The bonobo Kanzi went further: he used printed symbols on a board as words, and researchers reported he understood sentences he had never heard, at roughly the level of a two-and-a-half-year-old — though there was one subject, the scorers knew what to expect, and no other lab reproduced it, so this one is still argued about.

But all of these stop in the same place. Around age two a child starts combining words, and combines them by rule — "car went" is not "went car". More to the point, once the combining starts there is no ceiling: finitely many words, infinitely many sentences. No animal has been seen doing this in the wild.

So "is language uniquely human" is the wrong shape of question. Break it into parts and it gets clear: can you change your own vocalisations by listening to others; can you use a sign for something not in front of you; can you guess what another individual knows; can you slot one whole thing into another whole thing. The first three turn up piecemeal in other animals. The fourth does not.

The first one is especially interesting. Learning to reshape your own voice is found in songbirds, parrots, hummingbirds, whales, bats and elephants — branches far apart on the tree, each hitting on it independently. And chimpanzees, our closest relatives, barely have it. That is exactly the pattern from the convergent evolution issue: the same thing hit upon independently many times over, and absent in your nearest relative. Human control of the voice was not inherited up the ape line; our branch added it later.

One thing said plainly: there is no such thing here as "language evolved in order to communicate". The ledger only reads one way — among variants that happened to appear, those that let individuals get meaning across and thereby leave more descendants grew as a share of each generation.

It may have started with the hands, or with the tune, and each side holds one hard piece of evidence

If it was not inherited up the line of ape calls, where did it start? Two roads are mainly on the table. One says hands came first. Wild chimpanzees have a recorded repertoire of more than thirty distinguishable gestures, a good number of which map reliably onto a specific request — follow me, give me that, climb on my back. These are more flexible than their calls: a call is welded to the animal's current state (hurt, frightened, found food), while a gesture is aimed at one particular individual, and if it goes unseen it gets repeated.

There is a lovely experiment on this road. In 1998 researchers compared the hands of congenitally blind children and sighted children while they talked. Blind speakers, who had never seen anyone gesture, gestured anyway; and they kept gesturing even when told their listener was blind too. Gesture needs neither a model to copy nor an audience to watch. Hand movement and speech look like they come out of the same machinery, not like decoration added to speech.

The other road says the tune came first, which was Darwin's own guess: whole emotional songs with no separable parts, later cut up and fitted with meanings. What supports it is that the exaggerated pitch swings adults use with infants show up across many cultures, and that babies catch prosody long before they catch words. The two roads are not necessarily rivals; plausibly both channels ran and only later converged on one.

But what this section is really for is not which side wins; it is why the question cannot be settled. This layer leaves no trace. Throat, tongue and brain tissue do not fossilise, and gestures and grammar are not even objects; what can be dug up is bone, stone tools and ancient genomes. A Neanderthal hyoid bone found in an Israeli cave in 1989 is close in shape to ours — which rules out "their throats simply could not produce complex sound", yet tells you nothing about what they said, let alone whether hands or tune came first.

The gene called "the language gene" for twenty years has a better real story than the legend

In the 1990s a three-generation London family came to researchers' attention. About half its members had severe speech difficulty — not simple mispronunciation but an inability to organise the movements of mouth and face — and they also scored clearly low on grammar tests. Half affected, half not, which points at a single genetic change. It was found in 2001: one altered site in a gene called FOXP2 on chromosome 7. The press took it from there and made it "the language gene", and that label is thoroughly wrong.

FOXP2 is a transcription factor — in plain terms a switch governing when hundreds of other genes turn on and off. Nor is it ours alone: mice have it, birds have it, fish have it, and it has barely changed over enormous stretches of time. In 2007 researchers worked on zebra finches: while a young bird is learning its song, expression of this gene rises in one brain region, and suppressing it leaves the learned song inaccurate and incomplete. In 2009 another group put the two human-version changes into mice; the mice's calls changed, and so did the shape of neurons in one brain region — not a mouse starting to talk, but the switch really does sit on circuitry to do with vocalisation.

The part worth telling comes next. A famous 2002 paper reported that the human version of FOXP2 had been through strong selection within the last 200,000 years — the trace a useful variant leaves when it appears and sweeps through a population. It went into countless popular accounts, and the date lined up neatly with the story of modern human behaviour suddenly getting complex.

In 2018 a team redid the analysis on far larger and far more diverse genomes. The signal was gone. The original rested on a few stretches of sequence from about twenty people, with African populations barely represented; look at African populations on their own and the sweep disappears. What had been read as selection looks more like an artefact of who carried which version out of Africa. Add the 2007 finding that Neanderthals carry the same two changes we do, and it is not even specific to our species.

So today's version reads: FOXP2 really is involved in speech, because speech is the first thing to fail when it breaks. But "the most conspicuous failure" and "the only thing it governs" are different claims — for a node with hundreds of things downstream, whichever downstream thing is least robust is the one that lends the node its name. Hunting for the gene was the wrong search; the thing to find is the network. And being overturned made the field better: a pretty story that has survived a test is worth far more than one never tested.