118: Using tech to chat with bonobos, dogs, and whales

Introduction to Lingthusiasm

0:00Welcome to Lingthusiasm, a podcast that's enthusiastic about linguistics. I'm Gretchen McCulloch. And I'm Lauren Gawne. Today we're getting enthusiastic about using technology to communicate with animals, and how could we know if animals are using language? But

0:32first, we have a special treat for you on Patreon. A while back, I got asked to read the audiobook for Shakespeare's Sonnets, A Bridge Beyond the Point of Usefulness by Zach Wienersmith, which was a very fun day in a recording studio reading these classic sonnets distilled down into very silly rhyming couplets. It's impressive how Zach managed to write something that was both so elegant but also funny, and Gretchen's delivery is excellent. Zach has very kindly given us some free copies of this audiobook for patrons to download. This is available for all patrons, both free and paid. So even if you haven't

1:07checked us out on Patreon, we've been posting more and more fun stuff there for all levels of patronage. Including our very first bonus episode about swearing, which we refreshed with some new sweary information, and is also free for all patrons. Our most recent bonus episode over at Patreon includes some extra content from two excellent recent interviews. One is a discussion with Danny Bate about learning Czech and living in Czechia, and we also return to the mysterious Voynich manuscript with Claire Bowen, who has

1:37a very cool undergraduate university class about it. To listen to this bonus feature, as well as over 100 other bonus episodes, go to patreon.com slash linkthusiasm.

Patreon and Bonus Episodes

1:49Lauren, do you feel like people ask you a lot to adjudicate on specific examples of animal communication and whether or not they are truly language? Because this happens to me all the time. Oh, yes. I'm always so happy to have this conversation with people because it's always so interesting what animals and what theories about communication come up in these stories.

2:24I love hearing about all the different ways different animals communicate. I think that it's useful to be descriptive about what we mean and try to be very precise about what we mean here because I want to distinguish between two things. One is any type of communication. Yeah. The second is specifically what linguists mean when we talk about language because sometimes the word language gets used informally to refer to any type of communication writ large.

2:57Weirdly enough, linguists have a really specific idea of a specific bundle of features used in a specific way that is only found among humans across modalities. Right. The question of can animals communicate? Can they communicate with each other? Can they sometimes communicate with us as humans or with other species of animals? The answer to that is absolutely. They can communicate some ways in some things. Have we yet found an animal that uses a communication system in the same ways that human communication systems do, that all human languages

3:33have in common, whether they're spoken, whether they're signed, whether they're tactile, that we haven't. All of the human things, regardless of how related to each other they are, regardless of whether they have a writing system or not, they seem to have something in common that so far we haven't found any animals or aliens that do all of those things. And so, figuring out the specific thing that that communication is is interesting in itself. But also, what if we kind of reboot Dr. Doolittle in a really tech-savvy way? Can we bridge some of that gap? Is it

4:09just because they don't have the same face shape as us to make the same sounds or the same focus on making signs with hands and bodies? Can we use technology to see if this really is a level playing field or not? I saw the Dr. Doolittle movie back when it came out, and I confess to having that dream. What if we really could someday communicate with animals? I think that the quest for it has led in some interesting directions, even if – spoiler – we're gonna find out at the end that they're not quite the

4:44same so far as what humans are doing. Now, we are not going to, even though we are called to adjudicate on a wide range of animal communication stories today, Gretchen, we are not going to arbitrate every single animal communication story. We've already done an episode where we look at horses, ravens, dolphins, bees, dogs, other animals. Today, we're just gonna look at a few case studies

Bonobos and Language

5:08in particular. Gretchen So that we can go into them each in enough depth, and also so maybe we can preserve some material for future episodes. We're gonna talk about three animals – bonobos, dogs, and sperm whales. Gretchen Okay, so we've all had a chance, hopefully, to interact with at least one of these animals. Gretchen Yeah, I have definitely interacted with at least one of dogs, bonobos, and sperm whales. Gretchen I am here to tell you all about a very particular bonobo called Kanzi, who I – in a recently reading a textbook, he was described as a

5:40bonobo celebrity. Gretchen He's certainly the only bonobo I know by name. Gretchen He was a male bonobo. He lived from 1980 to 2025, so like a cuspy Gen X millennial. Gretchen I feel like he was really very famous when I was first getting into linguistics. I read a number of pop linguistic treatments of Kanzi, and I'm curious to know how well they stack up. Gretchen Yeah, so what makes him so compelling is that he wasn't intended to be one of the bonobos

6:15that they trained to use human-mediated communication to talk to humans. But he was just hanging out with his adoptive mom, Matata, when he was really young, where she would be being taught this communication system. And he just started using it. So, really, from an early age, giving the sense that he's kind of doing what human children do, which is just acquiring from the environment how to use language. Gretchen Because there was a whole fad, which has kind of faded now,

6:49maybe because it didn't turn out quite as well as people were hoping, of trying to teach various kinds of apes. You know, Coco the gorilla is famous among this series as well. Gretchen Another celebrity primate, absolutely. Gretchen Various celebrity primates to try to communicate in various systems. Since great apes don't have the same type of vocal tract as humans do, they can't shape human sounds. But their hands can do many of the things that human hands can do. So, there were attempts to teach them sign languages. There were attempts to teach them to communicate with symbolic boards as well.

7:20Gretchen Yeah. And Kanzi used a combo of these. I definitely agree. He's at the end of this trend that started in the mid-20th century. He's probably the peak and the end of this kind of research being driven where we force primates to communicate on our terms. He used some signs that were adapted from ASL, but he also used these symbols on a board called lexigrams. He would learn the symbol,

7:51and it would be sometimes attached to a synthetic voice so that he could press the button and it would vocalise the word that was attached to that symbol. I actually have a lexigram for you to look at here.

Yerkish and Lexigrams

8:07Yeah. And a screenshot of the board. I chose an aesthetic one. Gretchen Yeah. Let me see if I can figure out what it might mean or describe what it does. Gretchen Sure. You describe what it looks like and then have a go. Gretchen So, I'm seeing a symbol that's a circle with a straight vertical line going all the way through and spilling out of the edges. Then horizontally through the circle, there's a wave, maybe a sine wave or something similar to that that goes through horizontally along the middle. If I had to guess,

8:43I would say maybe they want this to indicate – well, it looks like a pokeball, but that's probably not what they wanted the bonobo to communicate. Maybe that's like a speech wave? Gretchen This is – Gretchen Or a thought wave? Gretchen This is the lexigram for juice. Gretchen Juice. Oh. It doesn't look like the juice emoji at all. Gretchen No. Like human language, it's pretty arbitrary in the relationship between form and meaning. There is structure to this set of symbols. Red is for things like food and drink. Blue is for activities. Purple is for animate beings

9:21like humans and bonobos. Gretchen And who even knows if bonobos have the same ontology that humans have? The colour system is helping the humans, but do we know that they distinguish between actions and foods or something in that sort of systematic way? Gretchen But this juice, like all the best lexigrams, looks like a really great album cover. Gretchen Oh, yeah. This could definitely be some sort of total eclipse of the heart situation, especially from the era. Gretchen Yeah. This

9:52particular set of visual communication that was designed to be used with primates like Kansi is known as Yerkish. Gretchen Yerkish. Gretchen I didn't realise it had a name until I did this research. It's named after Robert M. Yerkes, who founded the lab where Kansi spent his time. Gretchen Well, I guess it would make sense that Yerkish would be named after a guy named Yerkes. I was hoping for a more exciting etymology, but okay, yeah. Gretchen Yeah, it's a cool – like, Yerkish sounds way more metal than it looks as a set of images. Gretchen Yeah. And how did

10:28Kansi do at using these symbols? So, it was sort of a press a button and play a sound symbol, but the buttons were visually identifiable. Gretchen Yeah. I mean, I think the challenge with the research on Kansi, as with many of these primates, is A, there's only one of him. So, only a finite number of people can work with him, and that work is labour, and he needs to rest. So, like, there's a finite amount of research, there's a finite number of people who can do that research. Gretchen Mm-hmm . Gretchen And a lot of the debate has been around how much

10:58is it people scaffolding him into communicating the way they want him to, even if it's not consciously? How much of it is people who are communicating with him kind of projecting their own desired outcome on the communication? This has been a problem with a lot of this 20th century primates who sign, primates who use symbols type of communication. Gretchen It does seem like he had a pretty good understanding of these 350 Yerkish symbols and supposedly could understand many spoken English words.

11:34Based on trials, apparently, he could identify symbols correctly, like, 89% to 95% of the time. Like, that's pretty cool. Gretchen Okay. So, that's like you show him an apple and he presses the symbol for apple. Gretchen Yeah. There did seem to be some, like, he could combine them to make slightly more complicated meanings, but there didn't appear to be any evidence of the kind of grammatical structure. Like, all those little small words that we have in English that do all the grammatical heavy lifting. Gretchen Right. So, the, you know,

12:05words like and and of and the or endings like ing and id and plural and stuff like that. All languages have something like this, whether they do it as short individual words or whether they do it as prefixes and suffixes. Languages have grammar. That's one of the things we've found in all the human languages we've encountered. It's something that we don't find in the bonobos. Gretchen And I really, really need to stress this point, just so we have a reason to read one of my favourite blue sky posts ever. Gretchen Okay.

12:39Gretchen So, this is an entirely fictional, not true account of Coco the gorilla and a researcher. Gretchen Right. This is Coco fan fiction. Gretchen It's not a real thing. Gretchen This is Coco fan fiction, but it captures what it feels like to read this research where you have these primates communicating without function words and researchers attempting to create a translatable structure out of it. Gretchen Right. Shall we each take a part?

13:09Gretchen I'm happy to take Coco because I've put this upon us. Gretchen Okay. Gretchen Coco Birken bag, practical Coco possession bag. Gretchen Then the researcher says, no, Coco, you can't have a Birken bag. Gretchen Good Birken, good Coco, give beautiful Coco deserve gorilla. Gretchen Coco, we simply can't afford a Birken bag. Gretchen It's an unjustifiable expense. Gretchen Jealousy professor. Gretchen This one gets me every time. Gretchen Every time. That's what it feels like, genuinely, to read this research and what it feels like to see this gap between

13:43chimp communication and human full grammatical function. Gretchen Even this is something that has been constructed to have what a human sees as a punchline, right? Gretchen Yeah. Gretchen Either the snippets that we get repeated for us have been excerpted from a larger context in which maybe the gorilla goes on or the bonobo or whatever goes on to say something else that makes it less punchline-y, or the human has put in some sort of narrative to again give it more actual

14:14structure than what the whole data set of what the great ape is trying to communicate. Gretchen But also, just to stress, that was an entirely fictional account. Gretchen What I think is really exciting is we've seen this dwindling of this particular kind of research. What we're seeing instead is this really exciting growth of research where we're seeing primates and great apes all researched on their own communicative terms rather than forcing them to communicate with humans or through human technology.

14:48Gretchen So, there's lots of work from great researchers like Kat Hobata where the researchers spend lots of time, they collect lots of video footage that they can then get independent verification to see how apes communicate with each other. That's showing some really interesting properties that seem to be this kind of core set of gesture that exist across gorillas and across apes and across potentially one of the substructures that then went on to contribute to human language.

15:20Gretchen That's really neat because I feel like I've also used gesture to communicate with humans and sometimes with dogs and other species when I don't share a language in common with them. There are some gestures like I'm not being threatening right now or I'm happy or this kind of thing that can cross communication barriers. Maybe that's something we can also share to some extent with our cousins, the great apes. Gretchen Yeah, there's some great projects they've done where they've collected

15:50kind of citizen science, like what do you think this gesture means type feedback from humans who are just watching these video interactions. As a human though, you're always gesturing with the knowledge that this other human has a language, so you might be stacking gestures together in a way that presumes language. What I find really interesting about chimp communication, there was this great paper where they observed some multi-gesture structures. Gretchen Ooh, that sounds so sophisticated.

16:21Gretchen These weren't the sophisticated chimps. These were younger chimps who hadn't learnt yet what was the most effective way to communicate. Gretchen Oh. Gretchen It's a good plot twist. Gretchen Plot twist – the chimp gestures share some features of human gestures and we can maybe guess what they mean at a rate higher than chance, but then they're not doing that human thing of doing sequences, which is what we really see in human language. Gretchen Yeah. Younger chimps are figuring out what's going to work, and the older chimps are like, here's how you do it. Gretchen Here's the one gesture you need. Gretchen Yeah. Gretchen Oh. Gretchen This is why I think it's really exciting to not

16:55force them into our technology and our attempts for what makes communication.

Dogs and Button Communication

17:02Gretchen The next animal that we want to talk about is dogs. I think dogs have a really interesting connection with the chimp research and the gorilla research and the great ape research because on the one hand, we can't teach dogs to sign because they don't have opposable thumbs for it. But they are already found in so many human homes and have a whole lot of evolutionary practice at understanding and being understood by humans, even with very different anatomy. But there have recently been

17:35a couple famous button dogs on social media. Gretchen Button dogs? Gretchen Ah! Their humans post these videos of the dogs pressing buttons with their paws or their noses that speak words out loud, and then the human ends up telling a story or attributing something to the dog's intentionality about pressing these buttons. Gretchen Right. Gretchen These buttons were first developed for humans who have language disorders like aphasia,

18:06and they can press these buttons in order to communicate. Gretchen Oh, yeah. Assisted communication is absolutely a valid communication tool that humans can use. Gretchen Exactly. When humans use it, we use it in ways that are structured, that are complex, that use things like function words and question words, and that use the full end of the spectrum. Often an assisted communication tool will have both some buttons for frequent words, but also a way to type in extra words that weren't in the system to give humans that open-endedness to communicate their thoughts. The dog systems are

18:41generally much simpler. They have just a few buttons, they may be larger for the dogs to be able to press them, and they record a voice of the human saying a word. A few of the famous dogs are Bunny, who is a sheep-a-doodle. Gretchen Great dog name. Gretchen Yeah. And Stella, who is another Gretchen Classic dog name. Gretchen There are a lot of these dogs at this point. Apologies if we don't cite your favourite. One such example is a female Labrador named Copper, who has a video that's titled Pool…mad, and presents this story about apparently the dog being upset about

19:15losing access to the pool. It opens with a shot of the dog calmly sitting there. The trainer's voice says, hey, cop, what do you want? The dog sluggishly presses the pool button. The trainer responds, I know, your pool is broken. I'm so sorry. The trainer asks, Copper, want to play? The dog presses the mad button and looks away from the camera. Gretchen This dog has impeccable comedic timing. Gretchen Yeah. There's a paper that analyzes some of this communication, which we're going to link to. There's a few things there that are a little bit suspicious if

19:48you know dogs, one of which is that we know what an angry dog looks like, right? Gretchen Yeah. Gretchen Like, an angry dog is sort of growling, their hackles are up, all of this sort of stuff. The dog presses this mad button, but otherwise, the dog has a steady posture, slow movements that don't convey a sense of anger or agitation.

20:13Pressing the mad button is great comedic timing from a human perspective, but doesn't necessarily follow that the dog is communicating – even something we know dogs can communicate like, I'm angry or I'm agitated or I'm upset. If you have a dog or you're familiar with dogs, you know that a dog can communicate like, back off, I'm annoyed. That's not quite what's happening here. Gretchen So, what is happening? Gretchen What seems to be happening is that humans love to tell a story. Gretchen Yes. Gretchen And that our impression of what

20:44the dog is communicating is being shaped by both the human who's behind the camera and then also the human who's choosing how to edit and post these videos. Not just specifically for this one video, but probably for the videos in general. We don't know how many videos get recorded that don't have a funny comedic thing happening at the end that make them worth posting. You can train pigeons to hit buttons. Gretchen Yeah. Tell me more. Gretchen Say you have a left button and a right button, right? You can train pigeons to do certain

21:22types of combinations of left, right, left, right. You have to only reward left and right patterns of pecs that had not occurred in some previous number of opportunities. Gretchen Right. Gretchen So, the pigeons might only be rewarded to get a little bit of food if the pattern of responses had not been seen in the last five pecs sequences. They're able to do that. They can reduce these variable patterns. Gretchen Okay. What you're saying is, I could get a pigeon to reply to my text messages for me. Gretchen Well, maybe, right? If you give a pigeon a bunch of buttons that say a bunch of English words and they press them in a whole bunch of

21:56random orders because you're encouraging them to do lots of different orders, one of those, which could be the only video you put on TikTok, might say, like, hey, Gretchen, I don't want to record today. I'm mad. Gretchen If you give the pigeon the right number of buttons and you get, like, it's the million monkeys, you know, typing on a million typewriters will eventually produce the works of Shakespeare sort of thing. Like, you can eventually get something that looks like a sentence. Gretchen But it's also because humans love to read meaning into human symbols. Like, we're not asking the dog to design this. It's all human words. Gretchen In some ways,

22:30I think of the buttons as less of a prosthetic device for the animals and more a prosthetic device for the humans because humans love to interpret language. It's pretty well known that you can teach a dog to ring a bell, you know, maybe for a certain cue. I've known dogs who had learned to ring a bell when they wanted to go outside to go to the potty kind of thing. Gretchen Yeah. Gretchen You can teach a dog to do that. Or, like, a dog can communicate with dog body language that they want to play or that they're tired or that they want some food or water. They can go over to their food bowl and, like,

23:02make some, you know, plaintive eyes at you. If you teach them to press a button to communicate the same thing, like, they can. Gretchen Yeah. Again, linguists are never disputing that dogs are communicating with humans or doing some kind of communication. We're just very fussy about when that crosses over into language. If ringing a bell and pressing a button that says, release me from my torment, does the same thing, that's probably not language.

Skepticism of Dog Language

23:32Gretchen Right. I think that at this point, there is quite a bit of skepticism. I really enjoyed some of the comments on the dog training subreddit about this type of research because I think people who know dogs are pretty aware of what dogs can and can't communicate. People were saying, okay, it's useful for these types of things. Like, I want to go out, I want some food, this kind of thing. There's a lot of skepticism on this thread. There's a lot of skepticism, I think, in general, that the dog is really communicating, like, I'm mad about the pool being closed, or why is this thing

24:05happening? Or these more complicated questions that we don't necessarily think dogs can really put sequences together in complex sorts of ways. Gretchen But also, dogs have so many communication skills. Their ability to follow human gaze and use gaze to direct humans is something that is not found among other canines. It's actually a weirdly complicated skill to know that if someone extends their hand or extends a finger, you don't look at that. I'm so obsessed with pointing and this

24:40imaginary line that we project out into the world. Gretchen And the fact that we can do the fake-out fetch with the dogs where you pretend to throw the object and you don't throw it and the dog is trying to go there and then is confused when it's not there. That's very advanced that dogs can follow our movement to try to end up at a trajectory. That's something to really appreciate about dogs in their own terms. Gretchen Yeah. When you break down communication into all these different skills and you can be like, there's this thing – we really struggle to get gorillas and chimpanzees to engage with pointing. They are in our genetic lineage and

25:16they can't do this thing that dogs can do that is unique within their genetic lineage. Putting together this more complicated puzzle tells a more interesting story. Gretchen And the thing about the buttons is – again, some of the dog trainer comments were like, I tried to use the buttons but I found that it was better for me to just pay attention to the dog's actual body language because when I was paying too much attention to the buttons, I was ignoring the ways the dog was already communicating with me. The buttons tend to be quite large. They stay at home. If you're taking the

25:48dog to the park or all of the other many places you can take the dog, they don't necessarily have the buttons. At that point, you're still trying to communicate with their body language, their whimpers, their howls, their barks – all of the ways that dogs do communicate with us. Many dogs have learned how to understand certain human words like, hey, do you want to go for a walk? I'm sorry. Gretchen, that's very mean to people who are maybe listening with their buttons. Gretchen I'm sorry if I just trolled your dog. But one of the cool things about how ubiquitous dogs are and how these buttons have become really

26:22popular is that there's a citizen science dog communication project that has enrolled over 10,000 dogs and 700 cats from 47 countries across every continent except Antarctica. You can still sign up for it. It's still running. They have been trying to figure out how these buttons are being used, what processes people use to try to train their dogs to use them, and can you put the dogs

26:53through some tasks with the researchers on the camera to monitor how they're doing them. So far, they don't have a ton of results yet. They do try to find – do the dogs have any language strings? L – Okay, so that's putting together multiple buttons to potentially communicate something more complicated. Gretchen Right. One dog had apparently been pressing water and bone in a sequence to request ice. L – Oh, that's very cute. L – Up until the owner acquired an ice button and the dog abandoned this phrase.

27:25So that's kind of cute. Again, I wonder how much of this is prosthetic technology for the humans, right? Because if the human sees water, bone, even if the dog has just sort of done that arbitrarily, and rewards the dog with, oh, yeah, here's some ice. You like ice? And then the next time the dog is like, oh, maybe I press button one and button three to request ice, even if they don't have an actual understanding of the meaning of those buttons in a combinatorial way. L – I guess the idea of having the multiple cameras and having the researchers

27:57join for some of the interaction is to avoid one of my favourite phenomena in human-animal communication, which is the clever Hans effect. G – Oh, yeah.

Clever Hans Effect

28:08L – Clever Hans was a horse in Germany in the early 1900s. In fact, I've never thought about this before, but of course, it was German, so his name wasn't Clever Hans. It was de Kluger Hans. Hans drew massive crowds for solving mathematical puzzles by beating his hoof on the ground for the correct answer. He could do all kinds of operations. Very clever horse. Except this

28:39researcher, Oskar Fungst, figured out that what was happening is that his trainer and other people subconsciously were letting Hans know when he should stop tapping. They were all leaning in, and then when he got up to the number of hoofbeats, they'd all step back and subtly let him know to stop tapping. They didn't even realise they were doing this. L – So, not obviously leaning in and out, but they would just relax a little bit. L – Just whatever subtle cues. This horse

29:12wasn't a maths genius. It was a human body language genius. L – Which is itself kind of interesting, right? L – In itself, an amazing skill. L – Because horses have also co-evolved with humans to be able to interpret what we're doing with our eye gaze and our posture and the rest of our bodies. It's just not quite the same as being able to do long division. L – Yeah. This was such a scandal. Hans was such a celebrity that this has become known as the Clever Hans effect. When you are giving an

29:46animal or a research participant in general – human animal or non-human animal – you're giving them some kind of subtle indicator of whether they are performing the experiment correctly. L – The kind of thing you want to do to prevent that is ask the question from behind a screen maybe where the horse or the dog or whatever can't see you or ask a question that the asker and no one in the room knows the answer to. L – Yeah. L – Because otherwise, we know dogs can track our eye gaze. If we're looking at the button that would be the funniest punchline to what the dog could be quote-unquote saying,

30:21they could be following that eye gaze and being like, well, my human really likes it when I press this button. I'm going to do that because I like to please my human, you know? L – Yeah, which is why it's so great they're doing this big citizen science project. L – Yeah. What they've learned so far is that now is a favourite dog word whereas owners generally preferred later. L – Okay. L – It's not simply that whichever buttons the owner presses more frequently are also going to be pressed more frequently by the dogs. L – That's kind of nifty. L – But also – this is a real sick burn – dogs communicate love to their owners far less

30:55frequently than their owners do to them. L – Oh, poor dog owners. L – Which I don't necessarily think means your dog doesn't love you. I think it means that maybe the abstract concept of love as expressed by a button is less well communicated by a dog than the dog wanting to come sit by you and all these other ways that dogs express being excited to see you in these types of things. L – Again, I think your observation that it's probably better to communicate on the dog's terms is really going to pay off on that one. L – Yeah. They have been trying a few

31:26more complicated controlled experimental tests. A researcher wearing noise-canceling headphones presses a button on the dog's soundboard, which is covered by a plain coloured sticker so the researcher doesn't know what one they're pressing. That allows the dog a full minute to respond to the press. The researcher provides no feedback. The owners say this – a word or press a button on the dog's soundboard while wearing sunglasses so you can't see the eye gaze and do this type of stuff. L – Scientifically rigorous and very cool. L – Yeah. We should show things like dogs being

31:58more likely to move towards the door after the outside button is pressed than when the food button is pressed or something like that. They're still working on actually getting those results. There's another trial that they're doing where there's literally 24-7 cameras at multiple angles in the house facing the dog communication board and a bunch of other angles. I don't know who's willing to wire up their whole house to be live streaming data to researchers on the internet, but thanks for contributing to science. L – Thanks, folks. L – Then the next study they're trying to investigate is

32:30whether dogs can generalize an action label. If they have a button that says help or something like help that they've observed in situations like the owner retreating a ball that rolled under the couch, would they also produce that button when they're faced with a locked food container, which is a novel context that could still require help? L – Right. Again, this more sophisticated trying to see if they can do something closer to human language around generalization. L – Right. Which doesn't necessarily mean that they have a symbolic meaning attached to help,

33:01but at least that they can generalize it across certain types of contexts. It's certainly way more data than you can get of trying to do with chimpanzees. Maybe we'll learn a few things. L – Our final animal, there's no button boards. There's no symbols. We're going to look at sperm whales

Sperm Whales and Communication

33:19as our third example where instead of us giving them a technological medium to communicate with, we are using our technology to understand how sperm whales are communicating with each other, which we don't know a lot about. In fact, it was only in 1957 we learned that sperm whales even made sound. L – Oh, whoa, that's so recent. L – That's the first report in Nature. L – Less than 100 years ago that we even know that they make the whale songs. I thought we've always known about whale songs. Okay. L – We've known about whale songs, but sperm whales in particular, in 1957, there's this nature

33:54report where they're like, oh, well, this makes sense of some urban legend-type stories. They were like, oh, we spent an hour hanging out near five sperm whales. At first, we thought there was some kind of problem on the ship. But once we figured that out, the whales were in plain sight and there was just this sound, like a grating groan or a hammering sound as well. They were like, unfortunately, phonographic equipment was not available. This is 1957, guys. What do you think we carry sound

34:26equipment all the time? L – Instead, they had this description of it reminded some of a rusty hinge creaking. L – Yeah. They had to go very poetic to capture it in the absence of a recording. L – But now we have recordings. L – We do have recordings of their sequences of clicks, which are called coders. A lot of very interesting work is currently being done by a team with a project that is called Project SETI, C-E-T-I. L – This reminds me of a different Project SETI, S-E-T-I, which is about extraterrestrial

35:02intelligence searching. L – Yes. SETI is the search for extraterrestrial intelligence. L – Right. L – These folk have made a great pun. They've called it SETI, as in cetacean, which is the family that sperm whales are in. They refer to themselves as doing intra-terrestrial research. L – What have we found? What sort of technology are we using? L – We've found so far that we need to invest a lot in waterproof robotics and drones and high-quality waterproof microphones. L – That sounds useful, yes.

35:32L – So, a lot of their early work has just been literally developing their field kit to capture video and audio recordings of sperm whale communication and interaction. They refer to it a lot as communication, which I think is correct. L – I really respect that choice. L – They do occasionally talk about translating, and I'm just like, oh, I think you've got to do what the funding dollars call you to do, but I don't think we're really going to translate. We might learn something about the structure of whale communication.

36:03L – I didn't realise before we started looking into it for this episode how very behind we are in terms of understanding whales compared to understanding more accessible animals like birds and even chimps and dogs where we have many more centuries of awareness of the kinds of communication that they're doing. L – Yeah, right. One of the big studies this team has done is that they recorded a birth within one of the sperm whale pods that they research. L – Yes, okay.

36:39L – There were 11 members of this pod of sperm whales who were present and participated in the birth. L – That's so cool, even if they aren't necessarily just the whale's mother or even related to the whale. L – It's the whole community that were there. As a bit of a midwifery geek, I love a family-assisted home birth. I think that's really beautiful. L – I don't think I would record my own birth to use in research, but I'm glad these whales were

37:13able to share this moment with us. L – We have tons of data of what human births look like at this point. As someone who's a fan of Call the Midwife, I really want to see the whale sequel that's maybe called Call of the Midwife or Call of the Mid-Whale. L – And so, this research has these – all the whales kind of line up along the same axis to support the birthing mother whale and support the baby when it comes out. Obviously, that's all very interesting

37:43from a social dynamic perspective. We don't really know a lot about the social dynamics of sperm whales, but what's also really interesting is they have three and a half hours of audio. They used machine learning and computational methods to help them extract 5,731 of those coder clicks so they could analyze them. L – So, that's like clusters of clicks or sequences of clicks. L – And so, you see these changes in the rate at which these coders are produced and the structure

38:22of them across the birth. They were longer in duration during the birth and then those sequences shortened after the birth was complete. L – So, it seems to be related to something. L – And there's this greater homogeneity and similarity in vocal style at any point except just before and during the birth. There's heaps of variability that comes in at that point. We have no idea why any of this is happening, but just that there is this variation and we now have both the recording equipment and then the computer technology to analyze it is so cool.

38:55L – And that we have the context of what's going on because there's video as well of what's going on at the same time so we can maybe map that onto something. This is the first time we've witnessed a sperm whale birth in recorded form. We have one point of data, which is a very rich point of data, but we don't have like a million births or even a hundred or a thousand births to compare it with from other sperm whales yet. Maybe we'll have some in the future. L – The team have also been doing this really cool work where they look at

Analyzing Whale Communication

39:24the acoustic properties of these clicks and they use spectrograms which we've met before in acoustic phonetics. L – Well, that's fun because they do have sounds. L – They do have sounds and when they look at these sounds – and these are very quick so you're slowing them down, but when you slow them down and you look at those formant bands within the spectrogram and you read them just like you read human spectrograms, they identify what they call them R and E codoclicks because they

39:55have properties more like an R vowel in English or an E vowel in English. L – This reminds me, people also use spectrograms to analyze birdsong. With birdsong, we know things like, okay, this is a particular bird's courtship call or this is a particular bird's warning call or we have certain types of categories for things. Do we know anything like this for the whales yet? L – No, we're literally at like, did you know these clicks aren't even all the same? L – Oh, boy. Okay. We've got a long way to go before we even understand as much about whale

40:27song as we do about birdsong. L – Yeah, and there's individual whales have their own unique timing. Some click faster, some click slower. It's not like just this one homogenous thing. There is lots of variation. There's lots of things that are clearly happening and we are just beginning to collect enough recordings to see what that variation is. Whether we will get towards the whale acoustic model or WHAM, which is their attempt to move towards something – so good.

41:00L – They want to move towards what they're calling a translative model of vocalizations. Whether that will actually happen or whether we're translating whales is just a bit of funding, advertising. Either way, the whales are doing interesting things and it's exciting that they are finally getting data. I'm so glad that we're starting this on the whale's terms and we're not going to make, hopefully, the same mistakes we made with primates of trying to get them to communicate on our terms.

41:34L – There's no plans to create giant, floatable button boards and try to get the whales to press them with their noses. That's not on the horizon because we've realised it's better to study animals in their own terms in terms of their communication systems. L – Thankfully, yes. L – I think that one of the things that trying to communicate with animals and use technology reminds us is all of the different kinds of relationships we can try to have and that we can really wish we were able to have when it comes to animals. Trying to find ways of doing that,

42:11that try to understand the animals on their own terms and treat them in ways that are ethical for what the animal actually is and how they have a relationship with humans or don't want a relationship with humans while avoiding trying to make them into circus shows like The Clever Hands Effect or these things where you train the animals to just perform that they don't really understand what

42:43they're doing. L – Social media reels of the new circus. L – Yeah. Creating some sort of catchy soundbite that can be shared for entertainment speaks, I think, to this desire of humans to feel like we're not alone and that we can communicate in interesting and meaningful ways with animals. L – Which sometimes things like recording technology can really help, but also using our own senses as fellow animals can help us avoid getting distracted by the flashy technology-driven ways of communicating.

43:25L – For more Lingthusiasm and links to all the things mentioned in this episode, go to Lingthusiasm.com. You can listen to us on all the podcast platforms or Lingthusiasm.com. You can get transcripts of every episode at Lingthusiasm.com slash transcripts, and you can follow at Lingthusiasm on all the social media sites. You can get scarves with lots of linguistics patterns on them, including the international phonetic alphabet, branching tree diagrams, booba and kiki, and our favourite esoteric Unicode symbols, plus other Lingthusiasm merch like our retro-style gaffa guy rabbit shirts, totes, and scarves at Lingthusiasm.com slash merch.

43:58L – I can be found at GretchenMcCulloch.com, and I'm on social media as GretchenMcCulloch.com on bluesky, at Gretchen.McCulloch on Instagram, my blog is allthingslinguistic.com, and my book about internet language is called Because Internet. – My social media and blog is Superlinguo. Lingthusiasm is able to keep existing thanks to the support of our patrons. If you want to get an extra Lingthusiasm episode to listen to every month, our entire archive of bonus episodes to listen to right now, or if you just want to help keep the show running ad-free, go to patreon.com slash lingthusiasm or follow the links from our

44:28website. Patrons can also get access to our Discord chatroom to talk with other linguistics fans and be the first to find out about new merch and other announcements. Recent bonus topics include the linguistics of idioms, where things really hit the fan, a chat with Helen Zaltzman about linguistics podcasting, and some bonus deleted scenes about Czech with Danny Bate and the mysterious Voynich manuscript with Claire Bowen. – Can't afford to pledge? That's okay too. We also really appreciate if you can recommend Lingthusiasm to anyone in your life who's curious about language, or leave a nice review. Like this one from Sandman Stories Presents who said,

45:00so enjoyable. All of the working parts of language that you didn't know had names are covered on this pod. One of the best podcasts to listen to. – Lingthusiasm is created and produced by Gretchen McCulloch and Lauren Gorn. Our senior producer is Claire Gorn. Our editorial producer is Sarah Doppiarella. Our production assistant is Martha Tsutsui-Billens. Our editorial assistant is John Crook. Our technical editor is Leah Vellman. Our music is Ancient City by The Triangles. – As the bonobos, dogs, and sperm whales have not yet managed to say, stay Lingthusiastic! – I think it is a great song! – It was a funny song!

45:34– And it's all great song! – Thank you so much! – And I'll see you next time! – You're sick! – Thank you! – I'll see you next time! – It's all fun. – Thank you!