The scientists learning to speak whale - BBC

Marina: Welcome to Deep Voices — I'm so glad you're here. I'm Marina, a marine biologist and science communicator who has spent the better part of my life completely, hopelessly in love with whales and dolphins. This show is where we dive into their songs, their cultures, their family bonds, and the extraordinary science working to understand what they're actually saying to each other. Marina: Today's episode is a good one. We're going to meet the scientists using A I to try to decode whale language — and ask whether we might actually be on the brink of a genuine conversation with another species. We're going to talk about a tiny, heartbreaking porpoise that might vanish from the earth — and the digital lifeline scientists are throwing it. And we're going to celebrate some genuinely wonderful news about a whale species that has had so little to celebrate for so long. Marina: Before we dive in, a warm thank you to our sponsors who make this show possible. Ada helps clients integrate A I into their workflows — and given today's stories, that feels very fitting. Ago Consulting — pronounced "Ahh-Go" — offers silicon development consulting all the way from A I to SoC. And Zen Semiconductor is an umbrella company with a remarkable portfolio across the silicon industry. These are the kinds of companies building the technological foundation that makes science like this possible. We're grateful for their support. Marina: Alright. Take a breath. Let's go deep. Marina: So I want to start with something that stopped me in my tracks this week — a question that sounds almost too wild to ask seriously. Can we learn to speak whale? Marina: Not metaphorically. Not poetically. Actually learn to decode what whales are saying to each other — and maybe, eventually, say something back. Researchers are now pursuing this in earnest, and the tool they're reaching for is A I. Marina: Here's why this is so hard, and why it's taken this long to even try at this scale. Whale communication — and I'm thinking especially of sperm whales here, because they're at the center of a lot of this work — is not like birdsong, where you can sit with a spectrogram and start identifying patterns with your eyes and ears after a few months. Sperm whales communicate in clicks called codas. Short, rapid, rhythmic bursts of sound. And on the surface, some of them sound similar. But when you record thousands of them across different whale families, across different ocean basins, across years of interactions — the complexity starts to reveal itself. The combinations, the timing, the context. It becomes clear that there is structure here that we are only beginning to see. Marina: That's where Project CETI comes in. CETI stands for Cetacean Translation Initiative — and it is one of the most ambitious interspecies communication projects ever attempted. A genuinely multidisciplinary team: marine biologists, linguists, cryptographers, machine learning experts. They are deploying underwater listening arrays, non-invasive bio-logging tags on sperm whales, and underwater cameras in the waters off Dominica in the Caribbean, where a community of sperm whales lives year-round. Their goal is to collect billions — billions — of sperm whale clicks, and then feed that data into large-scale A I models to look for structure, meaning, and context. Marina: The approach draws on the same kinds of transformer models that power large language A I systems. The idea being: if you give an A I enough data, and enough context about when and how sounds are produced — who's present, what's happening socially, whether whales are diving or socializing or nursing — patterns that are invisible to the human eye might start to emerge. Marina: Now — and I want to be careful here — we are not at the point of translation. Nobody is claiming they've cracked the whale code. What researchers are saying is that the tools are now powerful enough, and the datasets are now large enough, that we can ask the question seriously for the first time. That is not nothing. That is extraordinary. Marina: And there's a philosophical dimension to this that I find almost dizzying. If we discovered that sperm whales have something like a language — with combinatorial structure, with the ability to convey different meanings depending on context and audience — what would that mean? What obligations would it place on us? How would it change the way we think about the ocean, about intelligence, about who gets a voice in decisions about the planet's future? I don't think we're ready for the answer. But I think we need to be asking the question. Marina: Okay. From the question of whether we can decode whale language to the question of how whales themselves pass on what they know. Because here's something that took the scientific community a while to fully accept, and that the public is still catching up to: whales have culture. Documented, demonstrable, transmissible culture. Marina: Let me tell you what I mean by that, because culture is a word that sometimes makes scientists nervous when applied to animals. But in the technical sense — behavior that is learned socially, passed from individual to individual or generation to generation, and that varies between groups in ways that can't be explained by genetics or environment alone — cetaceans qualify. Repeatedly, in multiple species. Marina: One of the clearest examples is humpback whale song. Male humpbacks sing — long, complex, evolving songs during breeding season. And the songs change. New phrases emerge, spread across populations, get adopted by other males. Researchers studying humpback populations in the South Pacific documented what they called a cultural revolution: a completely new song type emerged in Australian waters and spread eastward across the Pacific — through French Polynesia, the Cook Islands, Tonga — within just a couple of years. Thousands of males adopting a new song. That is cultural transmission on a breathtaking scale. Marina: Bottlenose dolphins in Shark Bay, Western Australia, use sponges as foraging tools — carrying them on their rostrums to protect their snouts as they probe the seafloor. This behavior is passed almost exclusively from mothers to daughters. It's a tradition. A skill with a lineage. Marina: And sperm whales — our friends from the first segment — live in matrilineal clans that span huge stretches of ocean, and the clans have distinct coda dialects. The pattern of clicks that identifies your clan, that you recognize as kin, that you share with the whales you grew up with. That is culture. That is identity carried in sound. Marina: Now — the A I work I mentioned in our first segment isn't just about translation. It's also about mapping this cultural complexity. Understanding which codas appear in which clans. How they vary. Whether they change over time. Because if you want to understand what sperm whales are saying, you first have to understand the cultural context in which they're saying it. You can't separate language from culture. We know this from human linguistics. It's almost certainly true in the ocean too. Marina: And the data being gathered by projects like CETI is going to take years to fully analyze. But early findings are already tantalizing — hints of combinatorial structure in sperm whale codas, patterns in how clicks are combined that suggest a much richer system than a simple set of fixed signals. The work continues. And every hydrophone in the water, every tag on a whale, every hour of acoustic data is another piece of a puzzle that may be one of the most profound in the history of science. Marina: I want to pivot now to three stories that all sit in the space where cetaceans and humans meet — and they run the emotional gamut from heartbreak to hope. Marina: Let's start with the vaquita. If you haven't heard of the vaquita, I want you to know its name, because it may not exist much longer for us to say. The vaquita is the world's smallest porpoise. It lives only in the northern Gulf of California — a tiny, specific corner of the Pacific Ocean. It has dark rings around its eyes that make it look a little startled, a little wide-eyed. It's extraordinarily shy. And it is on the absolute edge of extinction. Estimates suggest that as few as ten individuals may still be alive. Ten. In the entire world. Marina: The vaquita's story is one of accidental destruction. It isn't hunted directly. But it gets caught as bycatch in gillnets set for another critically endangered species — the totoaba fish, whose swim bladder is highly valued in certain black markets. Despite years of conservation efforts, despite fishing bans, the illegal gillnet operations continue. And the vaquita keeps disappearing. Marina: So scientists have done something remarkable and quietly heartbreaking: they have digitally preserved the vaquita. Using advanced imaging technology, researchers have created extraordinarily detailed three-dimensional models of vaquita skeletons. A virtual archive — a digital record of a species that may soon exist only in our memories and our databases. Marina: I don't want to be too bleak about this, because the scientists doing this work are also doing it with the explicit hope that it will inspire conservation efforts. That seeing a vaquita — really seeing one, in digital detail — might move people to act. And there's something in that impulse I find deeply human and deeply moving. We document what we love because we can't bear for it to vanish completely. Marina: But I also want to sit with the weight of it for a moment. That we have gotten to a place where one of the ocean's most remarkable creatures — ancient, evolved over millions of years, perfectly suited to its little corner of the sea — is so close to gone that we are making digital monuments to its bones. Marina: Now. Let me tell you something that made me genuinely smile this week, because we need it after the vaquita. Marina: Randy Wells is retiring. And if that name doesn't immediately mean something to you, let me tell you why it should. Randy Wells is the director of the Sarasota Dolphin Research Program — the longest-running study of a wild dolphin population anywhere in the world. He started this work in Sarasota Bay, Florida, in 1970. Nineteen seventy. He was a teenager. And he has spent more than fifty years coming to know — really know — the individual dolphins of Sarasota Bay. Marina: Because of Randy Wells and his team, we know things about bottlenose dolphins that we simply could not know any other way. We know that dolphin mothers and daughters maintain close social bonds for life. We know how calves learn to forage. We know how dolphins navigate illness, injury, old age. We have multi-generational data — grandmothers, mothers, daughters, granddaughters — all documented, all known by name. Well, by name to us. They have their own names for each other, their own signature whistles, their own identities. But Randy Wells knew them too. Marina: The kind of science Randy Wells built is rare. It requires patience, consistency, institutional commitment, and a very particular kind of love for your subjects. Long-term field studies like Sarasota are the bedrock on which everything else is built — the A I models, the cultural analyses, the conservation strategies. You can't understand a species from a snapshot. You have to show up, year after year, and listen. Marina: And finally — and I really have saved the best for last today — let me tell you about the North Atlantic right whale calving season. Marina: If you follow cetacean conservation, you know that North Atlantic right whales have had a devastating decade. Ship strikes. Entanglement in fishing gear. Warming ocean temperatures shifting their prey. Their population has been declining for years and sits at a heartbreakingly small number — around 360 individuals. Every calf born is precious. Every calf lost is a wound. Marina: So when the New England Aquarium, which runs the North Atlantic Right Whale Catalog and tracks this population with extraordinary care, reported that this past calving season produced the highest number of births since 2009 — I actually got a little emotional. The exact number of new calves confirmed was in the mid-teens, which for a population this size is genuinely significant. Marina: Right whales give birth in the warm, shallow waters off the southeastern United States — Georgia and Florida — in winter. The mothers nurse their calves for about a year, swimming north with them into feeding grounds in the Gulf of Maine and Canadian waters as spring arrives. Each new calf is catalogued, photographed, given a number, and often a name. And each one represents not just a life, but a thread of genetic diversity, a potential future mother or father, a voice added back to a diminishing chorus. Marina: This news doesn't mean the crisis is over. Right whales still face enormous threats. But it is a sign that when humans reduce the pressures we put on these animals — even incrementally, through things like vessel speed restrictions and modified fishing gear — nature responds. Whales are resilient, if we give them even a little room to be. Marina: So that's where I want to leave you today — with that image of a right whale mother in warm southern waters, her new calf surfacing beside her for the first time, filling its lungs with air. Both of them alive. Both of them adding their voices to the deep. Marina: Here's what strikes me, pulling all of today's stories together: we are living in an extraordinary moment for our relationship with these animals. We have the technology to listen more carefully than ever before, A I systems working to find meaning in the complexity of whale sound, digital tools preserving what we risk losing, and decades of patient human watching that has given us the foundation to understand what we're hearing. And in the middle of all of it — new calves, new life, new voices entering the ocean. Marina: The sea is not silent. It never was. We're just finally learning to listen. Marina: Deep Voices is brought to you by Ada, Ago Consulting, and Zen Semiconductor — thank you for believing that this kind of storytelling matters. Marina: If today's episode moved you, please share it with someone who loves the ocean, or a curious kid, or honestly anyone who needs reminding that the world is still full of wonder. Subscribe wherever you get your podcasts, and check the link in show notes for more on every story we talked about today — Project CETI, the vaquita, the right whale calving season, Randy Wells and the Sarasota dolphins. It's all there. Marina: Stay curious, stay connected to the sea, and I'll see you in the deep.