Babblers speak to the origin of language

“Holy shit, man!”

Andy Russell had entered the lecture hall late and stood at the back, listening to the close of a talk by Marta Manser, an evolutionary biologist at the University of Zurich who works on animal communication.

Manser was explaining some basic concepts in linguistics to her audience, how humans use meaningless sounds or “phonemes” to generate a vast dictionary of meaningful words. In English, for instance, just 40 different phonemes can be resampled into a rich vocabulary of some 200,000 words. But, explained Manser, this linguistic trick of reorganising the meaningless to create new meaning had not been demonstrated in any non-human animal. This was back in 2012.

Russell’s “Holy shit, man” excitement was because he was pretty sure he had evidence for phoneme structuring in the chestnut-crowned babbler, a bird he’s been studying in the semi-arid deserts of south-east Australia for almost a decade. After the talk, Russell (a behavioural ecologist at the University of Exeter) travelled to Zurich to present his evidence to Manser’s colleague Simon Townsend, whose research explores the links between animal communication systems and human language. The fruits of their collaboration are published today in PLoS Biology.

One of Russell’s students Jodie Crane had been recording the calls of the chestnut-crowned babbler for her PhD. The PLoS Biology paper focuses on two of these calls, which appear to be made up of two identical elements, just arranged in a different way.

When the babblers are flying through their scrubby habitat, they commonly make a two-note flight call (AB) that sounds a lot like a squeaky toy.

When the babblers arrive at the nest to feed their chicks they make a three-note “prompt call” (BAB), which sounds like the AB flight call but prefixed by an additional B. Like the squeaky toy that’s been given an extra squeeze.

In order to demonstrate that A and B were really meaningless phonemes that the babblers were arranging to create calls with different meaning, it was time for some playback experiments [carried out by lead author Sabrina Engesser]. With babblers housed in custom-built aviaries, the researchers watched how birds responded to the A and B motifs arranged in different ways. “We played them the flight call. The prompt call. The flight call made of prompt call elements. The prompt call made of flight call elements. The one syllable [B] by itself that seems to differentiate between the calls,” says Russell.

When the captive babblers heard a flight call, they became restless and looked out of the aviary as if expecting an incoming bird. When given a prompt call, they turned to a nest lodged in their enclosures. “Together, these results confirm the two calls are distinct and encode perceptible, context-specific information,” the researchers write.

A flight call made solely from the components of a prompt call is interpreted by the birds as a flight call, and vice versa. When the B element is played in isolation, it does not elicit any increased attention in the nest. “It confuses them,” says Russell. It only triggers nest-directed behaviour when followed by an A and another B. “It therefore fits the broad definition of a phoneme,” he says.

“This is the first time that the capacity to generate new meaning from rearranging meaningless elements has been shown to exist outside of humans,” says Townsend. “Although this so-called phoneme structuring is of a very simple kind, it might help us understand how the ability to generate new meaning initially evolved in humans,” he says. “It could be that when phoneme structuring first got off the ground in our hominid ancestors, this is the form it initially took”.

This is unlikely to be a one-off. Russell and his colleagues are currently analysing two other pairs of calls where the babblers may be doing something similar. More broadly, he says, there is no way that mixing and matching phonemes is a skill limited to chestnut-crowned babblers and humans. “People have been looking in the wrong place,” says Russell. Nobody’s ever found evidence for phoneme structuring in long-range communiation like birdsong, he says.

“We propose that such systems will be most operant in the short-range communication of vocally constrained, social animals,” the authors suggest.

Engesser S, Crane JMS, Savage JL, Russell AF, Townsend SW (2015) Experimental Evidence for Phonemic Contrasts in a Nonhuman Vocal System. PLoS Biol 13(6): e1002171. doi:10.1371/journal.pbio.1002171