So why can't animals talk anyway?

A forum for discussing linguistics or just languages in general.
Locked
hogsaloft
rupestrian
rupestrian
Posts: 3
Joined: 13 Jun 2021 19:45

So why can't animals talk anyway?

Post by hogsaloft »

Does anyone have a better argument than this? [link removed -Aevas]
User avatar
eldin raigmore
korean
korean
Posts: 5989
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: So why can't animals talk anyway?

Post by eldin raigmore »

There’s a movement of pets (mostly dogs) “talking” by means of buttons.
One of them is Bunny.
She usually “speaks” in three-word utterances.
She can clearly do what Chomsky says dogs can’t do.
What she can’t do is pronounce those words with her mouth.

Her current button-set is about 50 words IIANM.
I believe dogs are capable of mastering a vocabulary of up to 400 words?

I’ve seen claims some dogs can “speak” several human languages.
Salmoneus
MVP
MVP
Posts: 2500
Joined: 19 Sep 2011 19:37

Re: So why can't animals talk anyway?

Post by Salmoneus »

As with a lot of articles by "look at me, I 'have studied' 50 languages and once met Frank Zappa! Buy my book about linguistics (no, I don't have a qualification in linguistics)!" people, it's a shallow and wildly misguided article.


It misunderstands human language - I cannot in fact 'put a thought into your head'. But more importantly, it completely misunderstands animal language. It's simply not true, for example, that only humans can give directions to Trafalgar Square - it's rare, but some animals can do that too. Ravens have been shown to give directions (to items that are out of sight of both parties), and he even mentions bee-dance. He's also completely wrong when he says that 'a bark is a bark' - varying in intensity and quality (and of course duration) IS a change! And let's not even get started on the astonishing inventory size of some bird song! He's also wrong in saying that without language, individuals must deal with problems alone from scratch - several animals display culturally-learnt and taught behaviours, despite having no language (that we can make sense of).

He lights on dobule articulation as 'the answer', but this is clearly ridiculous. There's no conceptual reason why double articulation (using small figure inventories combined into signs, rather than using more unique figures as signs) would be necessary for language - humans seem to do it simply because human speech isn't capable of many unique figures (phonemes). Notably, sign languages tend to have far larger phonemic inventories (because the hands are more agile than the tongue) and so much less use of composition. There have even been questions over whether some sign languages may actually lack systematic double articulation, particularly in 'new' sign languages. Double articulation appears to be a feature that develops as sign languages fossilise, not as they develop: essentially, sign languages begin by using huge inventories of unique signs, and only as the language evolves do speakers begin to make them easier to remember by reducing unique simple signs into complexes of non-unique sub-signs. Bedouin sign language, for instance, is nearly a century old, but is seemingly only now beginning to develop double articulation. So double articulation may be something that human minds instinctively prefer, but it's certainly not essential to language.

Nor is it (probably) unique to humans. Birdsongs, dolphin conversations and bat calls all appear to have double articulation (although of course it's hard to 'know' that for certain since we don't know what they're 'saying'...).

Finally, he's ended up exactly where he said he didn't want to be: saying "dogs can't speak because they don't have double articulation" is little better than "because their brains haven't evolved language" - in fact, if you think that double articulation IS essential to language, then these claims are essentially synonymous.

--------

What would a real answer be? Well, dogs just haven't evolved sophisticated language. Specifically, a sophisticated language probably requires a high intellect, and probably requires an extreme social orientation. Dogs can't talk because they're stupid, which isn't a great surprise because their brains are much smaller than ours.
User avatar
Creyeditor
MVP
MVP
Posts: 4438
Joined: 14 Aug 2012 19:32

Re: So why can't animals talk anyway?

Post by Creyeditor »

So, apart from double articulation and recursion as the "structuralist"/"generativist" answer to what makes human language unique, I think I agree that linguistics has been trying to come up with whatever properties fit. So animals don't have human language, because we have defined human language in a way that excludes animal communication systems.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
hogsaloft
rupestrian
rupestrian
Posts: 3
Joined: 13 Jun 2021 19:45

Re: So why can't animals talk anyway?

Post by hogsaloft »

Well, I note you don't have a better answer. "Probably" isn't much help either. I'm merely observing that the one undeniable fact about human language is that its core structure is digital. Just as with numbers (and protein structure), this allows us to create an almost infinite number of referents using a minimal number of phonemes. Even with the smallest phoneme inventory known (8 consonants and 3 vowels) it is possible to create as many words as the average speaker knows within the space of three syllables. And of course, everything involved in sentient communication is on a spectrum -- that's the whole point! -- it's just that once you go digital, you put yourself far ahead of the game. Since you clearly consider yourself a superior being (despite not having a better answer) you'll probably not read the book -- which is a pity, because it's laid out in much greater detail there.
User avatar
Pabappa
greek
greek
Posts: 548
Joined: 18 Nov 2017 02:41
Contact:

Re: So why can't animals talk anyway?

Post by Pabappa »

I sincerely believe that dolphins have their own fully formed languages and that the reason we don't notice is because their languages are beyond *our* comprehension, not ours beyond theirs.

I dont have much evidence for my theory beyond that they seem to be able to communicate with each other, and that they've been observed repeating human language snippets to each other, suggesting t hat their own languages must be at least as complex as ours, and are clearly of a different nature. I may never convince anyone of my theory and I don't plan to try .... I'll just leave it at saying that it *could* be true, and it fits perfectly well with our observations.
Fafa žayas šifap tanaba.
The wind combs the hair of the unkempt. (Play)
Salmoneus
MVP
MVP
Posts: 2500
Joined: 19 Sep 2011 19:37

Re: So why can't animals talk anyway?

Post by Salmoneus »

hogsaloft wrote: 13 Jun 2021 22:42 Well, I note you don't have a better answer. "Probably" isn't much help either. I'm merely observing that the one undeniable fact about human language is that its core structure is digital. Just as with numbers (and protein structure), this allows us to create an almost infinite number of referents using a minimal number of phonemes. Even with the smallest phoneme inventory known (8 consonants and 3 vowels) it is possible to create as many words as the average speaker knows within the space of three syllables. And of course, everything involved in sentient communication is on a spectrum -- that's the whole point! -- it's just that once you go digital, you put yourself far ahead of the game. Since you clearly consider yourself a superior being (despite not having a better answer) you'll probably not read the book -- which is a pity, because it's laid out in much greater detail there.
Ahh, so this is a spam marketing thread! I thought it might be.

[Talking about language being "digital" is also very misguided. Human vocal sounds are analogue; they are divided into discrete phonemes by analysis. But exactly the same would be true of any spoken language, or indeed sound - dog barks, bird song, and so forth. The supposed characteristic you're actually talking about - double articulation - is the fact that human morphemes are generally (though not always) composed of (mostly) sequential non-morphemic phonemes. This has nothing to do with the analogue/digital distinction. In any case, it's clearly NOT an undeniable fact about human languages, because many people have denied it. It's also far from clear that it's not also true of some animal language. Meanwhile, none of this is related to your 'infinite number of referents' business, since you can produce similar numbers of morphemes in a non-composable way. Which is done in sign language. But even if we had to stick to hearing: in theory, the human brain can hear and distinguish around 2,400 distinct fundamental pitches (although actual humans can't remember them well enough to identify them without cues). Even if you only took, say, the lowest 500 of them, when you then factor in the relative frequency range of the higher formants, you can easily produce hundreds of thousands of distinct tones. But more dramatically, if a word were indicated by a unique frequency contour, there would indeed be a virtually infinite number of possible morphemes without any double articulation!]

Anyway, it's a bit silly complaining that people don't have a better answer than yours, when yours isn't an answer. You've described one of the famous (alleged) characteristics of human language, as found on page one of any textbook, but that description of how humans speak does not actually furnish us with an explanation of why animals can't speak. Since, a) a description of something just isn't an explanation for it (you've just shifted "the universe gave humans the ability to speak, and no other animal" to "the universe gave humans the ability to use double articulation, and no other animal", which is no more a 'reason' than we started with), and b) even if it were, it wouldn't work in this case, because of course many animals are perfectly capable of learning a small phonemic inventory and then combining it to form longer units. [word formation and recognition is actually the really easy bit of language, which we have little trouble teaching to animals. It's the syntax of sentences that is difficult! Animals have been taught to recognise and repeat thousands of words, and have been taught to remember the meanings of over a thousand - enough to hold a basic conversation. So the mystery of language is not 'how can we have enough words to have basic conversations' - the conversational ability of animals is way, way, way below their capacity as measured by their abilities to learn and use vocabulary.]


FWIW, you're probably looking at the question the wrong way around. Humans don't have conversations about the offside rule and quantum physics because we have sophisticated language; humans have sophisticated language because we have conversations about the offside rule and quantum physics. The big problem with trying to teach a dog, say, to form complex sentences is that there are very few, if any times that a dog needs to convey the meaning of a complex sentence - their needs and intellectual capacities are too limited to give them the need for grammar.
hogsaloft
rupestrian
rupestrian
Posts: 3
Joined: 13 Jun 2021 19:45

Re: So why can't animals talk anyway?

Post by hogsaloft »

The fact that "human morphemes are generally (though not always) composed of (mostly) sequential non-morphemic phonemes" has everything to do with the analog/digital distinction. At the phonemic level, this is the key to unlocking the so-called "digital infinity" effect that gives language its flexibility. As Chomsky has observed, "Language is, at its core, a system that is both digital and infinite" (though he spoils the effect by asserting that "To my knowledge, there is no other biological system with these properties." He has evidently not studied cell biology: this is precisely how DNA works. And yes of course, you don't need to use a digital system in order to produce large numbers of morphemes -- it's just infinitely more efficient, as the parallel evolution of writing systems in independent traditions clearly reveals in the historical record. Despite our vaunted intelligence, like most animals humans have a limited capacity to retain unstructured information – hence the limitations of emoji. Your reference to sign language merely proves the point: the fact that analog morphemes eventually 'fossilize' into digital phonemes is precisely what I am pointing to as the mechanism for the evolution of language.

It is also somewhat disingenuous to suggest that because animals don't understand how to use grammar they must be stupid. At best, some animals are capable of grasping the meaning of a few hundred words, which they probably hear as holistic noises. Human language is a highly cultural tool, learnt in context by children who have the benefit of several hundred thousand, even millions of years of development behind them. Not only that: the best explanation for the otherwise puzzlingly rapid encephalisation of our species is that the enormous potential opened up by the emergence of a digital language would have created a positive feedback for the selection of bigger, smarter brains. But this would not have happened overnight, so to expect a chimp to burst into speech after a few months of fumbling 'tuition' is somewhat preposterous, to say the least. Yet as we start to look, we increasingly see the process of digitization in embryo in the communicative strategies of other species. This is usually cited as potential evidence of the emergence of syntactic structure, but it is equally plausible to see it as a staging post to the emergence of true phonocoding, a necessary first step to the isolation and digitization of analog sounds that is essential for the formation of words.

As to looking at it the wrong way round, just think about what you wrote for a moment: “Humans don't have conversations about the offside rule and quantum physics because we have sophisticated language; humans have sophisticated language because we have conversations about the offside rule and quantum physics.” That is patently absurd. It is precisely because the digital nature of words gives us the flexibility to define, develop and discuss abstract concepts that we are able to have sophisticated conversations. Otherwise, what would be the point of education? There is nothing to suggest that the ‘standard model’ has it the right way round. In fact if we take Chomsky at his word that linguistics is still in a ‘pre-Galilean state’ – by which he means that the inadequacy and complexity of our current explanations mirror the baroque convolutions of geocentric theory – there is every reason to suspect the presence of a fundamental conceptual flaw. As Chomsky himself puts in an attempt to justify the retrospective reduction of his monumental work on generative grammar to the single function ‘Merge’, “It has long been recognized that only simple theories can attain a rich explanatory depth.” It is my contention that the simple switch from analog to digital communication provides exactly that rich explanatory depth – it was a paradigm shift that changed the game completely.

Finally, I can only apologise if mentioning my book offends your sense of propriety, but those who dismiss my views due to a supposed lack of “linguistic credentials” (what's an MA from Oxford, anyway?) leave me with few alternative avenues for getting my ideas heard -- it's a book-length argument, after all. If you were to deign to read it, you would see that the Appendix contains a closely-argued and fully-referenced academic paper laying out my thinking in detail, which I have been unable to get published elsewhere due to absence of the ‘appropriate’ accreditation. But that doesn’t mean I’m wrong. Any more than ‘client’ means “a person I met once”.
User avatar
Dormouse559
moderator
moderator
Posts: 2903
Joined: 10 Nov 2012 20:52
Location: California

Re: So why can't animals talk anyway?

Post by Dormouse559 »

I have locked this thread since it was created mainly for the purpose of promoting OP's book.
Locked