I’m sure you’ve heard that humans share about 99% of our DNA with bonobos—well, now it looks like we might share a few building blocks of grammar too.
A recent study published in Science (Berthet et al., 2024) caught my attention because it offers a rare glimpse into what our ancient ancestors might have sounded like – not with words, exactly, but with whistles, grunts & surprisingly sophisticated combinations of sound. The study focused on bonobos – one of our closest living relatives – & found that their calls may exhibit nontrivial compositionality, a key ingredient of human language.
What is compositionality?
At its core, compositionality is the idea that the meaning of a whole is built from the meanings of its parts. Linguists often divide this into trivial (basic combinations like “eat banana”) & nontrivial forms – where one element modifies or transforms another (e.g., “banana bread” isn’t just a banana & some bread—it’s something new). The latter is crucial in human syntax & semantics.
Until now, evidence of nontrivial compositionality in the animal kingdom was vanishingly rare. But this study, led by Mélissa Berthet & Simon Townsend, offers compelling evidence that bonobos may be using structured combinations of calls in ways that go beyond mere add-ons. That’s big news for both linguistics & evolutionary biology.
How was the study done?
Berthet spent five months in the forests of the Democratic Republic of the Congo, recording over 700 bonobo vocalisations in natural settings. Each call was coded with 300 contextual variables (what the bonobo was doing, how others reacted, etc.). From this, the researchers built a kind of semantic map of the bonobos’ seven most frequent calls using techniques borrowed from human linguistics.
The breakthrough came when they analysed combinations of calls. For instance, a “peep-whistle” often occurred in delicate social interactions. The team found that several of these combinations behaved compositionally—not just stacking meaning but modifying it—something we’d previously thought only humans could do.
Why does it matter?
Nontrivial compositionality is a linguistic holy grail. It’s what allows us to say things like “the cat that sat on the mat that was under the table that collapsed…” & have it all make (some) sense. If bonobos can even partially do this, it chips away at the idea of a uniquely human linguistic divide.
Of course, these bonobo utterances aren’t full sentences. They don’t map neatly onto human grammar or vocabulary. But they suggest that the cognitive building blocks of syntax may have deeper evolutionary roots than we imagined.
As Townsend put it, “It’s another layer chipped away from the apparent wall of human uniqueness.”
What about the sceptics?
Some scientists, like Shane Steinert-Threlkeld, questions whether the compositional structure observed in bonobo calls really reflects productivity like human grammar. He suggests they may be idiomatic combinations—more fixed expressions than truly generative syntax. Like “break a leg” to mean “good luck”, the meaning might not be transparently derived from its parts. Nevertheless, even critics acknowledge the value of this dataset—one of the most extensive ever collected in wild primate communication.
Teacher Takeaways?
Meaning isn’t always compositional – Idioms, collocations & phrasal verbs often don’t follow neat rules. This mirrors the tension in the bonobo study—are their call combinations rule-based, or more like “chunks”? Great prompt for discussing lexical chunks in class.
Evolution as engagement – People love a good story. Use this as a springboard to explore other animal communication systems—or even to discuss what makes our language unique.
If nothing else, studies like this give us a thrilling glimpse into the cognitive roots of our own linguistic abilities. Before grammar rules, before dictionaries, there were calls in the forest—layered, purposeful, maybe even a little bit like us.
What’s the most unexpected way you’ve explored language structure in the classroom?
Leave a Reply