In an ambitious project, scientists are attempting to decode the communicative clicks of sperm whales using artificial intelligence, and then talk back to them.
An ambitious interdisciplinary group of scientists wants to use artificial intelligence (AI) to decipher the language of sperm whales, Hakai magazine reports. In what may be the largest interspecies communication effort in history, researchers at CETI (Cetacean Translation Initiative) plan to feed a large amount of information into an AI system to teach it whale language and even talk to animals. Should the initiative succeed, it would be the first time that we actually understand what animals are chatting about – and maybe even have a conversation with them.
But do animals have language at all? The question has been debated by scientists for a long time. Many see language as one of the last bastions of human exclusivity. According to Austrian biologist Konrad Lorenz, one of the pioneers of modern ethology, animals communicate with each other but “do not possess a language in the true sense of the word.” German marine biologist Karsten Brensing, on the other hand, argues that the communication between many animals meets the conditions that make a language a language.
One such condition, or feature rather, is that language has semantics, involving certain vocalizations that have a fixed meaning that does not change. Siberian jays, a type of bird, for example, are known to have a vocabulary of about 25 calls, some of which have a fixed meaning.
The second condition is grammar, that is, the rules of building sentences. For a long time, scientists were convinced that animal communication lacked any sentence structure. But in 2016, Japanese researchers published a study in Nature Communications on the vocalizations of great tits. It turned out that in certain situations, the birds combine two different calls to warn each other when a predator approaches. Furthermore, they reacted when the researchers played this sequence to them. When the call order was reversed, however, the birds reacted far less. “That’s grammar,” Brensing says.
And there is a third criterion: you wouldn’t call the vocalizations of an animal species a language if they are completely innate. Although Lorenz believed that animals were born with a repertoire of expressions and did not learn much in the course of their lives, several animal species have proved to be vocal learners – acquiring new vocabulary, developing dialects, and even identifying each other by name. And even imitating cellphone ringtones…
Sperm whales’ clicks are ideal for attempting to decode their meanings – unlike other whale species, they produce non-continuous sounds that are easy to translate into ones and zeros. Also, they dive down into the deepest ocean depths and communicate over great distances, making the use of body language and facial expressions impossible for them.
“It is realistic to assume that whale communication is primarily acoustic,” says Michael Bronstein, an Israeli computer scientist involved in the project. Since sperm whales have the largest brains in the animal kingdom, six times the size of ours, why wouldn’t it be realistic to wonder whether they have something to say to each other? Do they give each other tips on the best fishing grounds? Or do whale parents talk to each other about raising their offspring, like their human counterparts?
To find out, programmers trained GPT-3’s neural network, (a so-called language model developed by the company OpenAI), with about 175 billion words. Language models statistically process huge amounts of text pulled from the Internet, and not only know which words appear together frequently, but also learn the rules of composing sentences.
By comparison, Gero’s Dominica Sperm Whale Project has collected less than 100,000 sperm whale codas. So first of all, the new research project will have to vastly expand that collection, to about four billion words – although nobody knows yet what a “word” is in sperm whale language.
If the idea works, it is quite realistic to develop a system analogous to human language models that eventually generates grammatically correct whale utterances. As a next step, an interactive chatbot could be set up to engage in a dialogue with free-living whales. Of course, no one knows whether the animals would answer. “Maybe they would just reply, ‘Stop talking such garbage!’” says Bronstein.
But is it conceivable that we could overlay the maps of a human and an animal language? Former physicist, designer, and entrepreneur turned critic of technology Aza Raskin, who had a similar idea in 2013, is convinced that it’s all possible, at least in principle. According to him, “there is almost certainly some kind of shared set of experiences, especially with other mammals. They need to breathe, they need to eat, they grieve their young after they die.”
That said, Raskin believes that there will be a lot of areas where the maps don’t fit. “I don’t know what is going to be more fascinating—the parts where we can do direct translation, or the parts where there is nothing which is directly translatable to the human experience.” Once animals speak for themselves and we can listen, says Raskin, we could have “really transformational cultural moments.”
Other scientists are very skeptical about whether the project will shed light on anything interesting. Steven Pinker, the renowned linguist and author of the book The Language Instinct, doubts that we can find rich content and structure in the sperm whale codas. “I suspect it won’t be much beyond what we already know, namely that they are signature calls whose semantics is pretty much restricted to who they are, perhaps together with emotional calls. If whales could communicate complex messages, why don’t we see them using it to do complex things together, as we see in humans?”
Even the CETI researchers themselves admit that their search for meaning in whale codas might not turn up anything interesting. “We understand that one of our greatest risks is that the whales could be incredibly boring,” says program lead David Gruber, a marine biologist at City University of New York.”“But we don’t think this is the case. In my experience as a biologist, whenever I really looked at something closely, there has never been a time when I’ve been underwhelmed by animals.”
The brainchild of Shafi Goldwasser, a computer scientist and cryptography expert also from Israel, the name of the CETI project evokes SETI, the search for extraterrestrial intelligence, which has scanned the sky for radio signals of alien civilizations since the 1960s, so far without success. So why not try our decoding skills on signals that we can detect here on Earth? Instead of pointing our antennas toward space, we can eavesdrop on a culture in the ocean that is at least as alien to us.