For anyone interested in topics like these, I recommend checking out Issac Arthur's "Cryptic Aliens" video, where he talks in depth about the assumptions that we can make about some of the forms of aliens just from natural laws, and about how much the aliens could understand of our language just from statistical analysis and a rough 1950s knowledge of science.
It's very interesting and it blew me a way a little.
Has anyone successfully decoded a language of another species with this or other efforts? Or does anyone know what's the furthest we think we've gotten when decoding language for another species?
Dolphins are understood well enough that we know there are regional dialects, and they broadcast individual name identifiers.
We know they're talking at least enough to have a sense of self, but AFAIK we haven't deciphered the culture/language enough to say anything meaningful to them.
I sleep outside most days year-round and the morning chatter of the birds is a delight. I understand none of it. Maybe they, like us, enjoy hearing themselves talk?
I always chuckle at birdsong, and how pretty it is. Most of it evolved to be some form of "GO AWAY" or "COME HAVE SEX PLEASE" that was loud and distinct enough to be heard over great distances and over the other noises in nature.
So, listening to sweet birdsong first thing in the morning is like going to a rowdy bar on a Saturday night and listening to the singles yell at each other over the music.
It genuinely never fails to make me chuckle when I'm on the porch listening to birds in the morning.
I always love that quote (can't find it) that Elephants have a special noise for "warning! there are bees nearby!", and someone pointing out that humans do too, it sounds like this: "warning! there are bees nearby!"
I’m in a tent, and would consider going without as there aren’t many insects around here that want to eat parts of me. There are coyotes and the occasional black bear and cougar, but I’m fairly close to the house. I grew up in rural USA and an comfortable in the woods. I try not to speak for others, and I haven’t yet met anyone who regularly sleeps outside when not on a camping trip.
"Decoding the language of plants and flowers has been a decades-long challenge. But that changes today. Thanks to great advancements in artificial intelligence, Google Home is now able to understand tulips, allowing translation between 'Tulipish' and dozens of human languages."
I think if anyone's had any success decoding animal language, it's been in a personal setting, with a whole lot of listening, observing, and paying attention.
Each individual animal has its own personality and language which they'll share with you, just like when humans who share a lot of time together tend to form their own language variants.
One of the most helpful books on the subject I've read is "Wild Animals I Have Known" (1898)
While that's personally appealing to me, I worry that we tend to understand what we want to understand - there's a reason we use the expression "if a were a horse" as a parody of empathy. Though of course that's even more true of trying to make sense of other human beings, which I don't think is impossible (just hard).
I think it is very similar to human beings, except relying even more on the non-verbal.
If you have doubts about interpretation vs "facts", then I can only guess/assume that you have not had this experience.
The best advice I can give for establishing a relationship is to think hard about what the animal wants most, and try to offer it in a way that it can understand is connected with you.
We know that cotton top tamarins have at least 38 calls, the calls have a grammatical structure, the calls are learned, and comprehension comes before usage mastery. That pretty much qualifies as a language.
Animals speak with scent, posture, gesture, and other means as well as with sound; and focusing solely on the audible portions of animal communication is never going to work well.
Raising my blind baby goat (as part of a dog pack), is showing me this anew: the dogs are having trouble talking to the baby because she can't see the postures and gestures. Scent and sound and feel are fine, but growly dogs' "respect my space and play nice" isn't being communicated at all.
Exactly, it seems to me that animals communicate within the context of the current situation (e.g. a dog would combining a specific bark, body orientation, neck fur standing up, to signify "there is something in this direction and I don't like it"). Decoding any one part of the expression wouldn't make us able to understand the whole meaning.
This is also what makes two-way communication so hard – we are so focused on our language that we often fail to develop or account for our other ways of expression, (and if we do rediscover our non-verbal expressiveness, it often come in a way of a hacky and very verbal ted-talk or book). So for two-way communication our animal counterparts first have to learn our repressed, underdeveloped body language, gesture, posture and facial expressions.
"How Animals Talk" first published in 1919. A newer edition was released in 2005. Would be worth reading if interested in working on this project.
Rupert Sheldrake controversial work would also be of interest. Starting with his book "Dogs who know when their owners are coming home". It can not always be attributed to sound, smell or time of day.
I imagine different local groups of animals may have their own linguistic culture. This has been seen with whales. People who have cats used to being fed at regular times will be familiar with whatever food calls they have decided on. Cats will sometimes use different vocalizations for different people when referring to the same thing. Cats rarely vocalize when communicating with each other. They clearly want to be understood by people, but i worry the fragmentation will make this kind of broad project very difficult.
They're hiring at the moment apparently - looked like some interesting roles. https://www.earthspecies.org/jobs
(No affiliation - just saw it via Twitter earlier today)
It will be interesting to see what they can come up with.
Still, given all the research so far, there is no such thing as animal language, not in the sense we understand human language. In particular, it's very clear that animal calls are not in any way equivalent to human words, they are at best equivalent to human facial expressions - quasi-automatic responses to certain stimuli, normally without any syntax or other signifiers.
There may be some exceptions, given that, for example, some dolphins have successfully been taught simple syntax (the difference between "Go there, do this" and "do this, go there", which no other mammal has successfully comprehended).
I know people whose communication appears to be quasi-automatic responses to certain stimuli.
I thought they'd worked out that dolphins have language and complex social structures? I was definitely told that when I saw the dolphins at Monkey Mia.
And dogs definitely communicate with humans, and understand human speech ("W.A.L.K" or you get zoomies for 10 minutes).
What's the research that says all of this is "not language"?
You're mixing the ideas of communication and language.
Communication is not language. Language is a type of communication. Language involves the use of symbols in a grammar that is sufficiently complex enough to facilitate recursive structures of arbitrary length.
I think it is likely that many animals like birds, cetaceans and elephants can use language and that it is possible that some dogs and cats can too but your example of a dog understanding that the word walk correlates to going for a walk is not an example of a dog being able to grasp human language.
My mom has Shelties which are a disconcertingly smart herding dog. I've witnessed one of her dogs overhearing relatively complex sentences, pausing to parse them, and then acting on the in away that indicates that they understand not just the correlation between subjects and objects but also the conditional logic in the sentence.
Here[0] is a paper going into some detail, comparing human language with animal communication systems. The mostly define the difference as related to composition: in animal communication, each "word" almost always carries individual meaning, while multiple "word" together carry all the meanings together (the paper also discusses cases where simple composition has in fact been identified). In human language, the meaning of a combination of words depends not just on the words themselves, but on the way they are combined as well.
To show an example: if you give me some bananas and I say "I like bananas but not these ones", it means I don't like the bananas you gave me. But a bonobo or dog "saying" some equivalent of "like" and "banana" would always mean that they like them (according to the paper, they could also "say" something like "not-like", "banana" showing some simple compositionality).
Another important observation (that is not found in the paper) is that animals have never been observed to "discuss" about abstract things or counter-factuals, which is what I mean when I say that their communication is quasi-automatic. For example, a monkey will only "say" "leopard" if it notices a leopard, or if it hears another monkey "say" "leopard" (or, most sophisticatedly, if it wants to elicit the leopard response from the other monkeys, e.g. to steal some food). But there is never a "discussion" about leopards that are not present and that does not cause the monkeys to have a leopard response.
Essentially animal communication is much more similar to our non-verbal queues, such as moaning or gasping or screaming, rather than to verbal communication. It can be more sophisticated than that, especially in primates, whales, and dolphins, encoding some information about social relationships, but 1 year old children already display much more complex language abilities than the most sophisticated animal.
30 comments
[ 0.14 ms ] story [ 1400 ms ] threadIt's very interesting and it blew me a way a little.
We know they're talking at least enough to have a sense of self, but AFAIK we haven't deciphered the culture/language enough to say anything meaningful to them.
So, listening to sweet birdsong first thing in the morning is like going to a rowdy bar on a Saturday night and listening to the singles yell at each other over the music.
It genuinely never fails to make me chuckle when I'm on the porch listening to birds in the morning.
You missed "DANGER DANGER" and "FEED ME" and "I'M HERE WHERE ARE YOU" and "HEADS UP GUYS (let's mob that raptor)"
https://www.blog.google/products/home/google-tulip/
Each individual animal has its own personality and language which they'll share with you, just like when humans who share a lot of time together tend to form their own language variants.
One of the most helpful books on the subject I've read is "Wild Animals I Have Known" (1898)
If you have doubts about interpretation vs "facts", then I can only guess/assume that you have not had this experience.
The best advice I can give for establishing a relationship is to think hard about what the animal wants most, and try to offer it in a way that it can understand is connected with you.
https://en.wikipedia.org/wiki/Cotton-top_tamarin
Raising my blind baby goat (as part of a dog pack), is showing me this anew: the dogs are having trouble talking to the baby because she can't see the postures and gestures. Scent and sound and feel are fine, but growly dogs' "respect my space and play nice" isn't being communicated at all.
It’s seems that various version of sign language will be their first target of non audio communication.
This is also what makes two-way communication so hard – we are so focused on our language that we often fail to develop or account for our other ways of expression, (and if we do rediscover our non-verbal expressiveness, it often come in a way of a hacky and very verbal ted-talk or book). So for two-way communication our animal counterparts first have to learn our repressed, underdeveloped body language, gesture, posture and facial expressions.
Rupert Sheldrake controversial work would also be of interest. Starting with his book "Dogs who know when their owners are coming home". It can not always be attributed to sound, smell or time of day.
"They're saying.. 'Thanks for the fish'.
https://www.sciencedaily.com/releases/2011/05/110512104252.h...
Still, given all the research so far, there is no such thing as animal language, not in the sense we understand human language. In particular, it's very clear that animal calls are not in any way equivalent to human words, they are at best equivalent to human facial expressions - quasi-automatic responses to certain stimuli, normally without any syntax or other signifiers.
There may be some exceptions, given that, for example, some dolphins have successfully been taught simple syntax (the difference between "Go there, do this" and "do this, go there", which no other mammal has successfully comprehended).
I thought they'd worked out that dolphins have language and complex social structures? I was definitely told that when I saw the dolphins at Monkey Mia.
And dogs definitely communicate with humans, and understand human speech ("W.A.L.K" or you get zoomies for 10 minutes).
What's the research that says all of this is "not language"?
Communication is not language. Language is a type of communication. Language involves the use of symbols in a grammar that is sufficiently complex enough to facilitate recursive structures of arbitrary length.
I think it is likely that many animals like birds, cetaceans and elephants can use language and that it is possible that some dogs and cats can too but your example of a dog understanding that the word walk correlates to going for a walk is not an example of a dog being able to grasp human language.
My mom has Shelties which are a disconcertingly smart herding dog. I've witnessed one of her dogs overhearing relatively complex sentences, pausing to parse them, and then acting on the in away that indicates that they understand not just the correlation between subjects and objects but also the conditional logic in the sentence.
To show an example: if you give me some bananas and I say "I like bananas but not these ones", it means I don't like the bananas you gave me. But a bonobo or dog "saying" some equivalent of "like" and "banana" would always mean that they like them (according to the paper, they could also "say" something like "not-like", "banana" showing some simple compositionality).
Another important observation (that is not found in the paper) is that animals have never been observed to "discuss" about abstract things or counter-factuals, which is what I mean when I say that their communication is quasi-automatic. For example, a monkey will only "say" "leopard" if it notices a leopard, or if it hears another monkey "say" "leopard" (or, most sophisticatedly, if it wants to elicit the leopard response from the other monkeys, e.g. to steal some food). But there is never a "discussion" about leopards that are not present and that does not cause the monkeys to have a leopard response.
Essentially animal communication is much more similar to our non-verbal queues, such as moaning or gasping or screaming, rather than to verbal communication. It can be more sophisticated than that, especially in primates, whales, and dolphins, encoding some information about social relationships, but 1 year old children already display much more complex language abilities than the most sophisticated animal.
[0] https://royalsocietypublishing.org/doi/10.1098/rstb.2019.006...