Reinforcing my strongly held belief that what fundamentally sets humans apart isn't spoken language, or tools, or any of that, but rather the fact we write down what we know, then make those writings available to future generations to build on. We're a species distinguished from all others by our information-archival and -dissemination practices. We're an archivist species, a librarian species. Homo archivum. In my opinion.
> Now we have evidence that both chimps and bonobos have syntax, it is inevitable that this capacity for compositionality was inherited from our last common ancestor, says Leroux. “They just showed, unambiguously, that this core building block is evolutionary ancient and at least 7 million years old, and maybe even older.”<p>This is not true at all. It's like saying that tool construction must be >300 million years old because both chimpanzees and New Caledonian crows make tools. Things can be invented or discovered multiple times by different species. It <i>might</i> be inherited from a common source, but it might not.
It is exciting times in animal communication.<p>This is not just distributional information analysis in the sense that ‘tokens’ are grounded in other ‘tokens’. They’ve grounded these calls in naturalistic situational context. This is hard won data.<p>If I understand, the finding here is that bonobo calls are “non-trivially” compositional, e.g., the semantic embeddings of pairs of vocalizations point in different directions surrounding the base vocalization. But it seems there is no “trivial” compositionality in the sense that constructions like [good __] might point in a similar direction. I would expect this latter result. This seems like a conspicuous absence? Is this really compositionality? Not sure what to make of it.<p>Some interesting context: bonobos and other (non-human) great apes are believed to have more intentional and flexible control over their gestural repertoire than their vocal repertoire and that these gestural repertoires are larger. Human language likely evolved from gesture (or so some believe). So, if their vocalizations are in fact compositional, it may be a separate evolutionary prong.
It is funny how at least the press written about this sort of research seems to imply only humans have language and some new evidence might challenge that notion.<p>Really if you ever own a pet, probably any pet I bet, you find that communication in a way that is arguably a language is pretty low level stuff in the animal kingdom. And it makes sense as it is quite useful for a species to communicate things about the world. You turn your community into a meta organism: rather than continuous appendages and nerve endings you might have a meerkat a couple hundred yards observing for predators for you sharing their own senses on their own body with you through their long distance communication abilities in the form of their vocalizations or body language. Now you can solely be a meerkat and get all this information about the area without having to evolve into some lovecraftian horror with a set of eyes and ears every 100 yards.
> One core block is syntax, where meaningful units are combined into longer sequences, like words into sentences.<p>I would think that syntax is a <i>structure</i> to the sequence of symbols, not just a sequence in any order. For example:<p><pre><code> Thag ate Fish
Fish ate Thag
</code></pre>
have different meanings.
> They recorded over 300 of these observations, including what the caller was doing at the time, what was happening in the environment and the behaviour of the caller and audience after the vocalisation.<p>> To reveal the meaning of each call, they used a technique from linguistics to create a cloud of utterance types, placing vocalisations that occurred in similar circumstances closer together. “We kind of established this dictionary,” says Berthlet. “We have one vocalisation and one meaning.”<p>This is lots of manual effort, could the recent advancement in language models help decode animal languages more easily? I guess it will need lots 24/7 capture of physical movement/action and sound data and train a model (that already understands vocal English too) perhaps.
> This finding doesn’t mean that bonobos have language, though, because language is the human communication system<p>I hate this attitude.<p>Also, I'm curious how advances in AI will shape our empathy towards animals.
> For example, the phrase “blonde dancer” has two independent units: a blonde person who is also a dancer.<p>This seems a rather odd "random" language example, especially coming from New Scientist. Being politically correct by then referring to the "blonde" as a "person" doesn't help much. May as well just use "brunette stripper" as an example - a brown haired person who takes their clothes off for money.