Yves here. It seems useful, when suitable opportunities present themselves, to range outside what these days is particularly stressful news-related fare for other informative topics. Animal communication, interests most pet owners as well as animal-lovers and animal professionals (such as farmers and zoo employees). But this article is separately instructive as an exercise in how to treat a natively difficult topic with rigor.
The author of the book profiled here sees animal communication as limited and as I infer, largely contained to alerts about dangers and identifying members of the immediate pack/tribe, and therefore by implication also fairly simple. Not having read it, I wonder how he explains crows being able to describe the faces of people they regard as threats to other crows, over time frames and participating crow numbers that would seem to make it impossible for all or even most of the crows to have seen the original “bad”‘ human (as in the number of scolding crows and time distance from the original human offending behavior can’t be readily explained by a crow simply issuing a call “That guy is nasty!”). So how do they convey what “evil person” looks like to crows that have not seen him?
You can find a detailed discussion of the crow study at issue here.
Having said that, considering how to convey information outside our structure of language is useful.
By Erica Goode, a science journalist, is a former reporter and editor at The New York Times and former managing editor of Inside Climate News. Originally published at Undark
Over the last decades, researchers who study animal behavior have succeeded in largely blurring the line between Homo sapiens and other animals. Like their human counterparts, animals feel emotions, they solve problems, they communicate and form complicated relationships, investigators have found.
Any number of books — think of Ed Yong’s “An Immense World” or Marc Bekoff’s “The Emotional Lives of Animals” — have been dedicated to exploring these relatively recently recognized abilities.
Yet few books on the ways animals communicate have been written through the eyes of a scientist as cautious and as thoughtful as zoologist Arik Kershenbaum, the author of “Why Animals Talk: The New Science of Animal Communication.”
Kershenbaum, a lecturer and fellow at the University of Cambridge, is distrustful of simplistic explanations, wary of assumptions, devoted to caveats — few statements come without qualification. In Socratic fashion, he asks a lot of questions, the answers to which, in many cases, neither he nor anyone else can yet provide.
That did not deter him from writing the book and it should not deter other people from reading it. But those who pick up “Why Animals Talk” expecting to find proof of animal telepathy or hoping for a dictionary of elephant-speak or a word-for-word translation of humpback whale songs, will be disappointed. (On Amazon, one disgruntled reviewer summarized the book: “Animals don’t really talk – The End.”)
If there is a message that Kershenbaum wants to get across, it’s that, as much as we’d like to be able to hold conversations with our pets or chat with chimpanzees at the zoo, it makes no sense to expect animals to communicate in the same way that humans do, “with the same equipment as we have, the same ears and eyes and brains.”
The idea of words, as we conceive them, has no meaning in the animal world; language is a human concept. Animal communication, Kershenbaum writes, is intimately bound up with evolutionary strategies for survival: Species develop forms of communication that allow them the best chance of successfully negotiating the environment and social structures they inhabit, whether it’s the underwater world of the dolphin or the highly social forest havens of the wolf.
Scientists, Kershenbaum tells us, are learning a great deal by analyzing animal sounds — the howls of wolves, the whistles of dolphins, the songs of the hyrax, a small, furry mammal related to the elephant and the manatee — and examining them for evidence of syntax and grammar, the building blocks of language. But the “why” in the title of the book is important, underscoring his view that knowing exactly what animals are “saying” is less important than trying to understand why they are communicating at all.
“Even if we discover that we will never talk to animals in the same way as we can talk to other humans, never hold a true conversation with a dolphin, still just by probing those possibilities we will find out why they live their lives the way they do,” he writes.
“Why Animals Talk” is organized around discussions of six different animals (seven if you count humans) extensively studied by researchers, with a chapter set aside for each. Kershenbaum has himself conducted research on some of these, including wolves, dolphins, gibbons, and parrots — but he also weaves in studies by other researchers: The parrot chapter, for example, centers largely on the work of Irene Pepperberg, who famously taught an African gray parrot named Alex to speak.
To understand animals, Kershenbaum tells us in the chapter on wolves, “is to understand the stories of animals as individuals, observed in the wild, but recognized as separate from their brothers and sisters, and from strangers of the same species — who might look the same to us, but can be perceived as mortal enemies by the animals themselves.”
He returns repeatedly throughout the book to these questions: What is language? How different from other animals are we really?
The howl of the wolf is meant for long-range communication, very different from the variety of short-range sounds — growls, whimpers, yelps, and whines that are softer and contain more complex information. Howls can be heard 6 miles (or more) away and are marked by changes in pitch and tempo. Scientists are mostly in agreement, Kershenbaum notes, about three different functions of howls: They can be used to mark out territory, to keep in touch with other wolves in the pack, or just for the joy of howling.
But “do these three roles of howling mean that wolf howls have three different meanings?” he asks. Kershenbaum doesn’t think so.
Some vocalizations — a screech of fear, for example, or a soothing cooing sound — seem to convey emotions across species. But, he writes, “We understand ‘meaning’ as having a clear definition at all only because we have language.”
“If you don’t have a language — you don’t even have a concept of what a language might be — then you probably don’t have a clear concept of unique meaning,” he adds.
Instead, howls seem to convey ideas “without needing distinct, discrete meanings, in the sense that our language-infused brains expect.”
“Why Animals Talk” is full of interesting facts about animal communication: Dolphins identify themselves with signature whistles. The most dominant male hyrax is the one with the most complicated song. Greater honeyguides — small birds in sub-Saharan Africa — engage in a cooperative exchange of information with people hunting for honey, trading calls and whistles back and forth: The birds guide the humans to honey, and the humans break open the hive, giving the birds access to the beeswax and larvae.
The songs of the lar gibbon, one of the most sophisticated communicators of the animal world, continue for 300 or 400 notes, with the potential to be combined into a staggeringly large number of songs.
In every case, as Kershenbaum illustrates, from the basic “I am here” of a songbird to the far more sophisticated utterances of a gibbon or a chimpanzee, the way information is communicated is a product of the animal’s need to survive in specific circumstances, and is developed as far as it needs to be, no more and no less.
No book on animal communication would be complete without writing about dogs, and this one is no exception. Kershenbaum’s own dog, Darwin, makes guest appearances throughout the book. (Before it was finished, however, Darwin died at 16, and the book is dedicated to him.)
Dogs play an important role in any story of animal communication because so much information passes back and forth between dog and owner, in a relationship forged over tens of thousands of years, to the benefit of both. But readers will find no endorsement of their canine companions having telepathic abilities or other common notions that many people have about their dogs — and about animals more generally.
“Why Animals Talk” can be a frustrating book. Kershenbaum often writes in a circular manner — repeating what he has written earlier, in slightly different words, often with some element added — and is overly reliant on posing questions to convey ideas. One would like to think that a good editor could have straightened out these knots.
Yet there is no question that Kershenbaum’s book is also stimulating and that it challenges readers to think hard about what we share with the other animals on our planet, and what separates us from them.
Because there is, in fact, a line, as Kershenbaum eventually concludes. “Yes, there is a lot in common,” he writes. “Animals have some syntax and even some sounds that behave a little like words, and they sometimes communicate complex ideas, although more often simple ones. But animals don’t combine all these abilities with explosive effect as we do.”
What’s more, he adds: “We’ve moved so far away from what animals do, and moved away so quickly, that direct comparisons between ourselves and other species seem almost childish. It should be clear that ‘Do animals have language?’ was never a good question.”
23 comments
Comments are closed.
This post and the underlying book seem best at conveying that animals don’t have the language of academics. And these other species may be lucky for that reason.
It may be that the review places too much stress on words, yet it is more than obvious that animals learn words. There are zillions of amusing videos on Ytoob of crows stacking containers, crows giving gifts (lovely), parrots naming colors, dogs listening for words (cookie?), and the list goes on.
What Goode and Kershenbaum seem unable to appreciate is that there may be other ways of organizing language, although they admit to syntax. In spite of all the “socially constructed” baloney going around these days, human beings may be wired for words. Dolphins and whales may be wired for long songs (any vocalization that long has to have some meaning or the creature wouldn’t do it). Parrots and crows may be wired for all kinds of vocalizations and a kind of culture.
And dogs have a sense of humor, which anyone who has been around dogs will learn. Humor is situational and also involves language. Hmmm.
So I am unsatisfied. Further, the assertions that human beings will communicate with animals are unimportant. Heck, those of us living in Italy laugh (constantly) over Americans saying things like, “Italian, it’s just like Spanish.” Well, no. And if a human being can’t figure out that Italian depicts a world different from the world of an English speaker or a Castilian, image the problems of trying to learn what a crow is saying or what makes a catbird the expert on jazz stylings that it is.
So: Too much use of chattering-class humans as a measure, too little contemplation of what bees see and how, and what their dances mean. Or the wisdom of coyotes and wolves.
Yes. The book seems to be partly an exercise in subconscious insecurity. Rupert Sheldrake has a very different take on this subject, but he is generally dismissed by ‘real’ scientists, many of whom have a phobia about anything that can’t be pinned down and dissected under a microscope. Yet there are many such things. Scientists must learn to accept reality as it is – not as they want it to be.
Indeed. I think the author is very wrong when he says –
“But animals don’t combine all these abilities with explosive effect as we do.”
Oh yeah? Tell that to the little kicky dog that almost got ripped to shreds by some angry crows after going after one of their babies – https://harpers.org/archive/2021/04/the-crow-whisperer-animal-communicators/
Or to any number of large whales off the coast of Australia who were victims of orcas cooperating with humans – https://en.wikipedia.org/wiki/Killer_whales_of_Eden,_New_South_Wales
In this review Kershenbaum reminds me of Hogarth, the narrator in Lem’s His Master’s Voice, a fictional memoir of a mathematician asked to assist a Manhattan Project-like effort to understand a message from a very distant intelligence.
There is a great tradition in the academy of expounding on what makes humans superior. Sometimes it seems like a scientific version of Genesis 1:26-28. It doesn’t seem Kershenbaum is guilty of this. It’s more like, as in HMV, he looks askance at scientists trying to fit observations of animal behavior into their existing scientific languages (models) without asking what they are (might be) communicating for.
I like to observe and imagine the similarities between canine consciousness and ours and consider the limits to those similarities. I think the dogs lack three closely connected features we have: words, episodic memory, imagination. They do however have a sophisticated body language and some are oddly vocal. Some seem try to communicate beyond their capability and that reminds me of what it’s like with a human that lost language to a stroke.
I see little sign of music in the dogs. I’ve mentioned before that I think human music runs on the same hardware that supports our words. And whadya know, the animals that show the greatest musical communion with us often seem closest to talking like us: birds. I love those singing and dancing youtube birds.
Once I left the television on when I went to work, on Mezzo, a music channel that runs a lot of operas. When I got home, I could hear my dog singing all by himself with the TV. I opened the door and was surprised that my wife wasn’t there because I didn’t suspect that he sang just for himself. He’s old now and doesn’t sing anymore but when I put one of his favorite arias on (he adores O zittre nicht, mein lieber Sohn from Magic Flute) I can hear him quietly humming along.
The crows are not “describing” anything. All of them haven’t seen the original human. The mask in the original instance becomes a conditioned stimulus because of the threatening behavior of the people wearing it. Over time the offspring of the original crows who saw the threat (and other crows as well) learn that that mask is a threat by observing the “scolding” behavior of the original crows in the presence of the mask and scold as well. It’s just basic learning principles.
In a response to me on the earlier thread, you pointed to “fact that [Professor] Marzluff was surprised” as supporting (I think) something remarkable about how crows communicate but the only thing that I can see that he might have been surprised about (aside from the persistence of the memory of the crows) was that the scolding behavior propagated through the flock more extensively and for a longer period of time than Marzuff thought it would. All that means—all it has to mean—is that more crows observed the cawing behavior and followed suit (and remembered the threat for a longer period of time) than Marzuff expected. (Crows are really social so the fact that they’re attuned to observing others in the flock seems to be consistent with that.) That’s definitely an interesting finding but, again, it seems to me to be a finding based on basic learning principles—classical conditioning and observational learning—rather than communication per se. (The crows are communicating to the other crows that the mask constitutes a threat in a sense but they’re not scolding in order to communicate that, as far as I can tell. The scolding behavior seems to be directed at the perceived threat. But then again, I wouldn’t rule out that they could be communicating “threat” to the other crows as well.)
And I’m not disputing that other animals could “talk” in some sense as Arik Kershenbaum argues—just that we don’t have to resort to any special kind of communication on the part of crows to explain the Marzuff study.
Did you bother reading the original study?
There was a lapse of YEARS between one of the uses of the mask and a later test. Crows only live on average every 7-8 years. The researchers documented MORE, not less, scolding as time passed. You’d expect it to diminish over time if this was merely original crows who saw the masked person misbehaving then calling out and getting others to pile on.
And I’ve been in area with lots of crows and regularly heard scolding incidents. They are often not taken up by all the crows. Sometimes you’ll get a general din and other times only some crow whinging.
There’s gotta be a process like epigenetics that guides the development of language. Humans have virtual libraries. but I think it might be interesting to ask the question, How much can humans really communicate? It would be interesting to document and analyze some subject of deliberation at the UN and take all that commentary and trace it backwards, maybe all the way to vitamin deficiencies or ancient trauma. Because we discuss and analyze incessantly. If all we are doing is rationalizing then maybe the actual communication is as simple as fatigue and consensus. My guess is that the instinct to cooperate controls our communication, simply because it seems ineffable.
I’m assuming the study is this one and, if that’s the case, I’ve read it. (And, in fact, the abstract of the study says “We suggest that conditioned and observational learning of specific threats may allow local bird behaviours to include aversions to individual people,” which is identical to what I said.)
It’s not “merely original crows who saw the masked person misbehaving then calling out and getting others to pile on”—every new instance of crows seeing the threatening mask and responding is an opportunity for other crows to learn the behavior. They don’t have to have seen the original misbehaving episode—they’re learning the behavior from other crows with the mask as a conditioned stimulus. There could be a lapse of several years between one use of the mask and the next and, if it’s within the average lifetime of some of the crows—and it would be because it appears from the study that the longest gap of time between one exposure of the mask and another was 2.7 years—they could be responding in the way that they do based on the persistence of their memories and the strength of the conditioned response.
The amount of scolding could go up over time as more and more crows learn the behavior and then diminish over time as that behavior undergoes extinction (as fewer and fewer crows associate the mask with a threat). The latter, in fact, appears to be what is happening:
IMHO, animal communication ‘suffers’ mainly from its nearly total lack of negation. They can say “I don’t like this”, but they can’t say “X is not true” or “Y is not real”. No imagination and no lies.
It’s much easier to express a negative in words than in primary process but to a very limited extent it can be done. For example, dogs can communicate something like “I won’t fight you,” by showing what they’re talking about and then switching. But yes, it’s ridiculously hard to say there are no elephants in the garden without words. Much easier to say there’s elephants at the zoo or the garden looks safe.
I also find it interesting to consider the human mental processes that lack negation, such as dreams, visual imagination, desire separately from ability to describe these experiences in words that can use negatives.
There’s a lot of deceit in the animal kingdom. Does that not count as lies?
I think I’ve read of animals pilfering goods from some common ‘store’ and ‘cheating’ on mates, but I’m not inclined to count anti-communal or simple selfish acts as, what’s the word? — ‘misinformation’.
I think of dreams as negation in the sense that they often complement conscious thought, as in XOR. Optical illusions are often rooted in neural XOR. Much of Dada’s ‘imagination’ is just the mapping of linguistic negation (perhaps nihilism) onto canvas.
There seems to be a gap between thinking about human communication (linguistics) and animal communication.
The author mentions “true conversation.” What would that be? Humans and animals use their communication systems to entertain, soothe, establish social bonds and hierarchy, punish and reward, woo and retain mates, and yes, convey information–but that information may be lies or just mistaken.
In-person conversation (and even a lot of writing) does pretty much what animals do, in more complex ways. They may be talking about the past, or the future or hypotheticals, but the humans are really displaying tail feathers, grooming their partners’ fur, sniffing butts, alerting to predators, etc.
I feel uneasy about this article since the author is talking about animal languages. So you have wolves, dolphins, gibbons, and parrots as examples and I am seeing a mammal, a sea-borne mammal, a primate and an avian species. Reminds me of the guy who went to twenty countries and said that he had seen them all when in reality he had only seen the airport and hotel for each. Do we even know how each species sees the world? Do we really understand the context of where they use their languages? Maybe in a few centuries we will have the technology to start uplifting some of them and then we can ask them ourselves. Till then, we have not yet really started to understand them.
-Do we even know how each species sees the world? Do we really understand the context of where they use their languages?
Yes!
Wittgenstein:
Wenn ein Löwe sprechen könnte, wir könnten ihr nicht verstehen
If a lion could talk, we could not understand him
I understand the unhappiness of people at seeing animals, in their view, denigrated. I don’t doubt that various animals and humans overlap on all sorts of emotional and communicative matters. But I have to say that I think GramSci above is on the right track in pointing out that animals seem to lack a basic communicative function, negation. I would go much further. I think – and this is not just a matter of language but a matter of basic cognition – animals do not have much idea of possibility, epistemic or moral necessity, or much ability to think about possible futures, be moved by fiction as much as by real events yet remember that fiction is not real, and so on. Can you imagine an animal thinking about its child’s future career possibilities or social standing and taking present steps to try to affect that? The ability to think about future possibilities as intensely as though they were reality, yet hold on to the fact that they are just possibility? I seriously doubt animals have the ability to come up with very striking ideas based on observations, like noticing the resemblances between parents and offspring (including grandparents) probably giving rise to ideas like inherited “blood” or reincarnation? Or for a modern example, there is the legal invention that a corporation is a human individual, which has led to so many consequences (anti-worker organization in the nineteenth century using the 14th amendment, corporate political contributions freed up). Some of these inventions, like those of physics (‘gravity,’ ‘energy’) seem to approximate reality, some are just fantasy. And then there are the remarkable religious formulations. If any gifted chimpanzee has a great religious vision, they cannot communicate it to other chimpanzees. I am very sympathetic to the idea that in important basic emotional matters and beginning abilities, animals have a good deal that overlaps with us. But I think this idea that “we aren’t so different” has become a somewhat overextended, even silly fad. I think humans are way, way different from other animals in profound ways, and apparently this remains true even if discussing it becomes somewhat unpopular; people want to avoid elitist oppression. (As someone who spent a career studying language development, I also agree with the reported opinion about language of the author of the book reviewed here – some overlap, but profound and extensive non-overlap).
You would have to go a long way to find an animal stupid enough to entertain the ideas of its child having ‘future career possibilities’ or ‘social standing’ or of a ‘corporation’ or of such an entity being ‘a human individual’. These are mostly inventions of the last few hundred years. For most of its existence the human race has lived without them, and within the next seventy or so years will do so again, probably. Social standing is the only exception, and is shared by animals. The only advance we could make over animals out of all your examples is to do away with ‘social standing’.
“The idea of words, as we conceive them, has no meaning in the animal world; language is a human concept.”
I don’t agree with this. By this argument deaf people who use pure sign language aren’t human, which is obviously false. Or we can say that words and language can take many forms and include gesture, body, movement and facial expression, in which case animals obviously have language.
I’ve always thought animals communicate more with sign language, with body language and various modes of silent expression, than with anything resembling syntax, so if we’re looking for words similar to written or spoken within an array of animal sounds then I think we’re looking for the wrong things and in the wrong direction. To me the topic should begin with a survey of the many ways of communication other than utterance. Yves is right to bring up the crow and facial expression example.
Even my cat doing a foxtrot towards me with tail raised is “hi!” and “I’m happy to see you” in cat sign language. She thumps her tail in disapproval, the human equivalent of “will you stop doing that already”. She stares at the doorknob to summon the door opener. She’s also remarkably good at telling time using various means to disrupt what I’m doing at 11pm exactly. You can also find online plenty of vids of deaf people who’ve taught their pets sign language, which I think is meaningful here.
And, probably related, sign language is often recommended for use with human babies as a way to allow for expression and communication even before forming language ability, so sign as method of communication, would seem to form before spoken words as a language.
I also want to say that as someone in the tech field there is not a day goes by when I don’t see firsthand the limitations of spoken and written language, that we even communicate at all, using words, is a bit of a miracle in itself.
It’s not entirely clear but the sentence you quoted and disagree with may be just defining terms. Perhaps the author intends to limit “language” to refer to the words humans use. Everything else is therefore some other kind communication that we don’t call language. By this definition, all the important stuff that inflects the meaning of the sentence “I love you,” such as tone of voice, facial expression, body language, context, etc. are not language. Makes things simple: everything animals do isn’t language.
This is not at all how I use the word. I prefer a very broad definition. But it might serve the author if what Kershenbaum wants to do is encourage investigators to relate animals’ behavior to the animals’ purposes instead of to human use of words. Maybe his point is that we shouldn’t try to measure and model what animals do using the tools we have developed to measure and model human use of words.
Sign languages of the deaf are much more than elaborate miming. ‘Speakers’ of Sign Language and French Sign Language can’t really understand each other, despite Thomas Gallaudet having been profoundly influenced by the Institut National de Jeunes Sourds de Paris.
DSL’s gestures are symbols, notably including gestural symbols of abstract negation, per my comment above. Deaf people can pun in DSL, a simole attestation of its symbolic nature.
Whoops. Should read: ” ‘Speakers’ of American Sign Language and French Sign Language can’t really understand each other”
Almost any emotion dogs express can be seen in their eyes.
Most professors who study animal behavior and communication remind me of this article by the late, great David Graber on animals being seen as utility-maximizing machines.