A very interesting read, which describes how for ~50 years the community was lead astray by a seemingly simple oversight: studying dead animals instead of living ones!
>Charles Darwin theorized that monkeys had the vocal anatomy needed for speech, but lacked the necessary neural mechanisms.
>This was the most popular hypothesis until 1969 [when] Lieberman studied the vocal anatomy of a monkey corpse and concluded that other primates could not produce as many vowels as humans because of the position of their larynxes [which] cemented the idea that a descended larynx is a prerequisite for speech.
>Towards the end of the 20th century, evolutionary biologist W. Tecumseh Fitch realized a crucial fact – all the existing evidence was based on the anatomy of dead primates. Surprised that this had not been done before, Fitch used x-ray imaging to study the vocal tracts of live animals while they were voicing sounds. He was amazed to observe that their larynxes at rest remained high and then descended during vocalization to a position very similar to the human larynx.
>All these studies indicate that apes have all the anatomical characteristics necessary for speech. The reason they don’t is purely neural. Humans have much better control of the larynx, not because of its position, but because of the neural connections that connect it to the brain. Parrots don’t even have a larynx, but they have wonderful control of their speech organ, which enables them to articulate intelligible words and phrases.
I'm pretty sure descended larynx is not prequisite for producing vowels.
Vowels are merely overtones which occurs because the waves generated in the vocal folds(with a charachtrized fundamental frequency) travel the vocal tract in 3d, which causes destructive interference for most frequencies except those with constructive interference (overtones / formants)
When humans move that tongue and lips we basically change the vocal tract shape which changes the overtones, the larynx is behind the lips and behind/in the back of the tongue anyway.
"require" is a fuzzy term. If you force your lips and tongue to stay in place, you can get partial intelligible vowel sounds from your larynx, and feel your larynx changing as vowels change.
Human exceptionalism has poisoned so many studies.
My pet-theory why primates do not vocalize is that the neural difference is not even that cemented, it has to be some difference at the language acquiring stage as babys and children, were one human pattern matching center for sound fires constantly, acquiring the language "keys" and theirs dont.
20th century hubris poisoned so many studies I would rather say.
I have this example but there are literally hundreds:
You know about the 5 μm particle size limit for it to become airborne? Washing hands for protecting against aerosols? It’s parroted bad research; and wrong.
The main infection path of upper respiratory diseases is through aerosols. The medical literature says it’s through hands that are touching the mouth and the nose but this is wrong. It was shown in the 1940s that viruses can spread through aerosols and that aerosols are particles under 100 μm. Research was done on tuberculosis. Tuberculosis is caused by bacteria that must infect the deep inside parts of the lung. For that it needs an aerosol size of 5 μm. Other diseases can do with aerosols of 5-100 μm.
Now, a meme was born that only 5 μm particles can get airborne.
And this was just accepted by the Western medical canon and the WHO.
One sort of R, sure. But native English speakers often have trouble with the "voiced alveolar trill" R which is present in many other European languages, unless they speak one of the less common English dialects where it is used.
Monkeys have a much easier time learning sign language than spoken, so there's probably something about vocal chords too. I suspect it's because humans and parrots independently developed music as a bonding mechanism and apes did not.
You're biased because you already know what it's supposed to be saying. The brain really likes to make us hear the sound it knows it's supposed to hear.
My chihuahua-pit mix has such a range I often describe it as “singing the songs of her people”.
Anyway, what qualifies as speech is very subjective and generally anthropomorphic. I understand what my pup is saying almost all of the time, which wouldn’t qualify it as talking by most standards but it’s plain as day to me.
Ya, humans are really good at pattern recognition. I can understand what my parrot (who can’t speak) is communicating.
But did I learn my parrot’s patterns or is he intending to communicate with me? Hard to say.
If he is intending to communicate, did he just learn what works through operant conditioning? Hard to say.
If it is just operant conditioning (for both of us) and a pinch of antropomorphizing from me, does that even matter when the end result is that we are successfully communicating? Hard to say.
Whatever it is, it works. We can’t have a philosophical discussion, but he can usually get what he needs from me and I can usually understand what he wants and get him to change course of action when needed or get him to perform tricks on command.
Haha, meanwhile my chihuahua-pit mix only communicates by trembling, booping, performative sneezing, huffing (as opposed to a woof), sometimes grumbling, and yelling at the mailman. Mostly trembles.
My first beloved Chihuahua was similarly very expressive, but almost never barked. Lots of trembling, sneezing, some cute vocalizations while sneezing, growling, grumbling, stamping her feet, pointing with her eyes, but no barking... unless other dogs she affiliated with were barking.
I was so amused, I always used to praise her for it. I'd say
> That's right, baby! Good job!! That was a good, family bark!!!
My gorl doesn’t get the chihuahua shivers, but she definitely does some big sneezes and huffs. She also does a big performative snorty sigh when she settles into a snuggle. And she has the loudest sing-songy yawns I have ever imagined, sometimes while dismounting the couch with her front paws and letting them slide across the floor before she can bother with even moving her hind paws.
Cats don't even meow or purr at each other much. They just stare each other down and, as my parents told me when I was a lil kid, they speak through mind reading. I still believe that.
They do communicate, usually in the form of yelling to indicate they want food, pettig, to be let outside or let back in. That's err... all the communication they need, the rest of the time they sleep.
They did develop fine control in their eyebrows to appeal / communicate to humans. And cats developed mewling similar to human babies to draw their attention.
That’s not the conclusion at all. The article says monkeys can understand and produce words (via signs). They can’t talk because they don’t have control of their larynx. They’re mute essentially.
Yes, we just put some AI via electrodes into the monkeys. If we can also add some blockchain, I think we'll have a winner. Let's circle back Monday on this idea, I'd like to pick your brain some more.
In any case, you bring up an important issue: can monkeys bitcoin?
Could we backport a blockchain experiment in the same way that demoscene Hungarians have backported math hacks to fuel unexpectedly sophisticated realtime graphics to the C64?
Yes, but Brain-Computer Interfaces are not currently an especially advanced field of study. We don't have the precision needed to do any of that as currently stands.
Chimpanzees are intelligent and willing to do something like communicating, and can maybe produce words/signs, but they don't produce sentences/syntax and they never ask questions. Supposedly parrots can do that one.
Related question: which animals understand humans pointing at things?
There are a number of interesting videos on social media sites ostensibly demonstrating dogs and birds using buttons and tablets to generate reasonably coherent communication like might be expected out of a two year old. Some of these have some backing scientific literature that provide some validation of the claims. If true it would indicate they have quite advanced mental faculties and are basically entirely limited by the lack of a appropriate vocal system for generating human speech.
Two year olds learn a lot more language in a year than chimps can. Can't have a sentence like "dog bites man" without word order even if it's made of logograms.
There's no reason to believe that LLMs are necessarily similar to humans. Just think of flight; both birds and planes can fly despite being very different. It's understandable that because humans are our main reference for intelligence that we see LLMs speaking language as a proof of similarity but there's no concrete reason to assume that.
There's something about watching other primates interact with screens that just feels so uncanny. It's an almost humbling experience, seeing how much they look like us tapping and scrolling. The only difference seems to be that we think we understand more of what we're looking at on the screen... but do we really?
Yes. We do. Significantly more. It’s our only power. We aren’t stronger or faster or more vicious, we cannot fly or breed dozens of humans at a time or puke honey. We can’t swim the ocean depths or swing through the trees.
But we sure can think, understand and create. For better or worse.
I’m all for not underestimating animals. They are sentient like us.
But not at the cost of dismissing the human animal.
We have very good color eyesight, are unusually flexible when it comes to weather, are pretty omnivorous and disease resistant, and we're absolutely unbeatable endurance runners.
IIRC we actually smell as well as dogs too, if we get down to floor level.
Planes and birds do operate on the same principles though, but instead of power through propellers birds are powered via wings.
There's no reason to think an LLM would not work for a monkey. Even if it is a different structure than how things work in humans a parallel structure could offer more or less the same benefits.
Anyway I buy the whole LLM thing, consider a baby, they can't talk at all and only become coherent when you talk to eachother just about non stop for 2 or 3 years.
You are right, there’s no concrete reason why an Internet comment should be taken to heart, yet somehow we still do react emotionally to online content all day long.
I get excited, happy, sad, jealous, proud and a bunch of other stuff just seeing stuff on a screen.
Anyway, what is life’s concrete reason to be? Does everything need one to be considered valid?
I've also been wondering this. Can we train an LLM on dolphin noises?
I suspect the problem is the training data. If we don't know what the dolphin squeaks mean, we can't really filter the training data to include only meaningful "language." But I do wonder how far we could get by just dumping a bunch of dolphin recordings into an LLM and getting it to make the same noises...
The other side of that same problem is that we would have no way of knowing whether it's working. Even if it could "communicate" perfectly, we would have no way of understanding it, aside from watching real dolphins interact with it. But maybe we could build an artificial dolphin, load an LLM into it, and drop it in a pool with real dolphins... just to see what happens...
One issue with this is that dolphins don't have a consistent "language". There are at the very least different dialects in different parts of the ocean. So one issue is ensuring your model is using sounds from a group of dolphins speaking the same "language" - otherwise you're at risk of trying to "translate human" by feeding a mix of English and Chinese.
They key issue with lack of training data is that we do have intelligent dolphins for communication experiments - for example, we can replay certain sounds that we have recorded, and we can "ground" that language with "nonverbal" communication - e.g. showing various things or pictures.
Yes that's the point of project CETI https://www.projectceti.org/ but for sperm whales instead of dolphins. I saw a video of it, it was so amazing but I forget which one of the videos it was.
This is an old tradition in popular science, replacing the humans are animals with souls analogy with the current fad. (Analogies to clockworks, printed circuits, et cetera.) They’re wrong because they’re oversimplifying for pithiness. Not explanatory value.
Hmmm. As I type this reply on my device powered by a semi conductor that wouldn’t exist without our knowledge of quantum physics I’m gonna say science.
As I type this from a free country that allows me unrestricted access to worldwide media, and affords me the right to express my opinion freely publicly, I would say politics.
We all have our own points of view. Yours is certainly an astute observation. But it is not the only one.
False dichotomy. This isn’t science versus religion, philosophy, the humanities or politics.
You repeated a meme: humans are animals plus some nebulous X. I’m calling that out as meaningless; it’s unfalsifiable. You switched to asking about the meaning of words, and are now trying to draw equivalence between your statements and the merits of politics as a category. This is an exercise in false equivalences, red herrings and wronger than wrongs [1].
Could you care to explain why you expected the comment to be about the conclusion of the article instead of an opinion about the comment it was replying to?
>Charles Darwin theorized that monkeys had the vocal anatomy needed for speech, but lacked the necessary neural mechanisms.
>This was the most popular hypothesis until 1969 [when] Lieberman studied the vocal anatomy of a monkey corpse and concluded that other primates could not produce as many vowels as humans because of the position of their larynxes [which] cemented the idea that a descended larynx is a prerequisite for speech.
>Towards the end of the 20th century, evolutionary biologist W. Tecumseh Fitch realized a crucial fact – all the existing evidence was based on the anatomy of dead primates. Surprised that this had not been done before, Fitch used x-ray imaging to study the vocal tracts of live animals while they were voicing sounds. He was amazed to observe that their larynxes at rest remained high and then descended during vocalization to a position very similar to the human larynx.
>All these studies indicate that apes have all the anatomical characteristics necessary for speech. The reason they don’t is purely neural. Humans have much better control of the larynx, not because of its position, but because of the neural connections that connect it to the brain. Parrots don’t even have a larynx, but they have wonderful control of their speech organ, which enables them to articulate intelligible words and phrases.
Fascinating