Study Suggests Primate Calls May Use Supra-Regular Recursive Grammar

Human language is intricate and adaptable. It expresses our intelligence and cultures, as well as our desire to make sense of the world and communicate with one another.

Languages generally operate according to particular patterns. Small units of grammatical meaning can be embedded within the broader framework of a sentence, creating intricate syntactic layers.

For a long time, scientists have considered these embedded linguistic patterns - called supra-regular recursive grammar - to be a trait that separates humans from every other animal on Earth.

However, mounting evidence indicates that some of our fellow terrestrial species could employ comparable grammatical approaches in their communication.

Primate calls and supra-regular recursive grammar

New research indicates that three non-human primate species may occasionally rely on supra-regular recursive grammar in their calls: the common marmoset (Callithrix jacchus), olive colobus monkey (Procolobus verus), and bonobo (Pan paniscus).

"We show for the first time that animals generate call sequences that can be captured by supra-regular recursive grammars and hence share more similarities with human language than previously thought," writes University of Zurich linguist Remo Nitschke and his colleagues.

The research is available as a preprint on the bioRxiv server and has not yet undergone peer review, though it offers an early look at the team's findings.

Because researchers do not know what individual elements of primate vocalisations mean, or how they combine, they used sophisticated mathematics to examine the sequences of calls.

They built two separate mathematical models to assess recordings of human speech and other sounds alongside primate calls. Both models predicted the likely next call sequence, but each relied on different assumptions when generating those predictions.

One model predicted the development of a call sequence on the basis that sounds occur in a linear order. This is known as probabilistic regular grammar.

The second incorporated nested hierarchies. It considered the placement of smaller sound units within a call sequence when predicting what would follow, an approach known as probabilistic context-free grammar.

Mathematical models of primate communication

Together, these models reflect important distinctions between regular grammar - an ability known to exist in some animals - and supra-regular grammar, which has been considered exclusive to humans.

The team first evaluated its method using human recordings. According to the authors, this confirmed that the models were dependable, at least for the purposes of this investigation.

Applying the models to primate calls suggested that vocalisations from all three species included features of supra-regular grammar.

"These findings challenge syntax as the major transition in hominid evolution and opens up prospects for alternative evolutionary scenarios," the authors write.

Even so, it remains too early to state with certainty that these primates genuinely use supra-regular grammar to communicate.

The findings rely on probabilities: models were used to interpret the structures of communication sequences whose meaning is not actually understood. As a result, the calls probably included supra-regular grammar, rather than demonstrably doing so.

"Yet another possibility is that the groupings reflect a purely formal structure, similar to the syntax of music in humans or song in animals," the authors note. "Disentangling these possibilities requires much further research."

The datasets were also relatively limited, meaning additional research will be required to establish whether the apparent patterns consistently persist across larger collections of calls from these primates.

"Despite these limitations, it is clear that the traditional view needs to be revised: some parts of call sequences of nonhuman primates are in fact most likely generated by supra-regular grammars," the authors add.

This may suggest that, in linguistic terms, humans are less distinct from our primate relatives than previously assumed.

"The absence of a supra-regular capacity has long been held to be the single-most important demarcation line in hominin evolution," Nitschke and team conclude.

"Our findings reset the agenda for research on the evolution of language: if not syntax, what is the critical step that enabled the evolution of language from a primate-like communication system?"

The preprint is available on bioRxiv.

This article was fact-checked and edited by Fiona MacDonald. Although we take pride in our process, we are only human. If you notice an error, please let us know.