▲ 113 points
back
91 comments
The article doesn't say what the headline says. At all. Never in the article does it say that "our capacity for language emerged 135,000 years ago".
What the article says is that they have genomically traced our species back to around 135,000 years ago to the point where we started splitting off in different groups, and as we all have the capacity for language we must have had it at the point of split as well (before that all could be said to be in the same local group. Possible, what with those population bottlenecks and everything).
There's nothing in there saying there wasn't language before, nor why it shouldn't have been present much earlier. Early enough that e.g. Neanderthals could equally well be covered.
Theoretically, it is possible language developed later. Some people hold the position that there is no specific language hardware in our brains, and that language is the result of a socio-cultural process (which then would spread).
Some animals have provably some sort of language, and most probably have them for tens of millions of years, or their ancestors did. No reason to believe there was no language before those 135k. Clearly you don't need something very special and unique to us.
Now where is the threshold when we consider something a proper language and something... less lets say I don't know. Science probably doesn't have a clue neither. The evolution was probably very gradual, ie every 1000 years the complexity grew up by a notch, on average. Compound it to a million years.
What about the speech centers? Broca and Wernicke areals is the closest thing humans have to language hardware.
What distinction are you trying to draw? To me the headline and the claim the article makes are effectively the same, just with the article claim dotting some i's in terms of assumptions being made (it seems reasonable to simify a headline).
Thank you, I read the same thing! Glad folks are commenting on this.
The above comment feels really spammy/AI generated but I promise I'm a real human and just genuinely appreciated someone mentioning the disparity between the headline and the actual research statement.
That's what "at least" means in the title.
If the title cannot be edited then the post should be removed, IMO.
It looks like they actually say at least 135K years ago, as that is the latest it could have happened.
So this finding is in the same vein as mathematicians who can't yet prove or disprove a conjecture, but can prove tighter restrictions than we knew before. Still interesting, and, more importantly, often productive of new paths or methods of analysis.
As for the article itself, I'm not sold. The analysis was about narrowing down to find the most recent regional split in our genetics, with the assumption that language formed before humans split across regions. That would work if we assumed the capability to develop language rested largely on genetics. If a lot of animals could develop a language, then the explosion of human language could come from imitation when one regional group without a language meets another with a language.
>"Human language is qualitatively different because there are two things, words and syntax, working together to create this very complex system," Miyagawa says. "No other animal has a parallel structure in their communication system. And that gives us the ability to generate very sophisticated thoughts and to communicate them to others."
I have seen a lot of ideas about human mental exceptionalism fall apart. It wouldn't surprise me to find out in a few years that dolphins developed syntax.
I think we already have decent evidence of animals having synytax. Presumably Miyagawa means something like recursive syntax.
Exactly. The article literally says
"I think we can say with a fair amount of certainty that the first split occurred about 135,000 years ago, so human language capacity must have been present by then, or before."
The title is wrong.
On that note, we know that interbreeding between Neanderthals and (the ancestors of) modern humans happened during at least two periods:
> Neanderthal-derived genes descend from at least 2 interbreeding episodes outside of Africa: one about 250,000 years ago, and another 40,000 to 54,000 years. Interbreeding also occurred in other populations which are not ancestral to any living person [0]
... which makes me think that it's more likely that both populations had the genomic capacity for language all the way back in the first "episode" 250k years ago as well (but to be clear, I'm saying it seems more likely to me, not that we can safely assume this must be true) .
Yes, I know that "humans are impossibly horny" is a bit of a meme, but we're not just talking about sex here, but about having offspring that is successful enough within then-existing society over many generations, to the point where the Neanderthal DNA was absorbed into the common gene pool and survived all the way to this day.
It seems very unlikely to me that that would happen unless both species were on a comparable level of mental capabilities. So we either both evolved language capabilities separately around similar times somewhere after 250k years ago, or already had the basis for it before our ancestral populations split and then later met again.
Yeah, this is a common occurrence in that something is reported to have happened at year X, but in practice it will have emerged rouglhy around that time, and never at one fixed moment in time. Possibly re-emerged over and over again until it stuck.
Title was modified from what it originally said.
This appears to be an argument for terminus ante quem and useful in that sense but it ignores the possibility there is a far earlier terminus post quem when the actual language capacity emerges.
I think it true(ish) that in a model strongly aligned to a single root language the point of segmentation is the last point language seen in all post-fragmented states can exist. But I don't see why that is also the first point. It's just the one we can detect genetically. There will be some subsequent genetic evidence perhaps to a specific structural change in the brain, or vocal chords, or something else, indicative of language emergence.
If holographically defined families tools pre-date this time, then abstract concepts were being communicated, even if not vocally. Show-and-tell has limits and I would argue strongly suggest concepts inherent in language existed to communicate how to do the tool making.
Abstract conceptual thinking cannot be separated from our language ability, both of which are dependent on our access to mind. [Our brains are the tuners we use to select which realms of thought we are attempting to focus on. Some people choose to tune into universal compassion, some prefer pack-mentality behaviors such as racism or religious bigotry (including against those who haven't any), some just choose to make money or pursue greater pleasure, the list goes on and on.]
Our bodies are mammalian in the basic pattern, but we are clearly distinct in ability. It wasn't 6000 frickin years ago, but there was a creation event where we instantaneously diverged from pure animal life.
Mitochondrial Eve traces us to our root. Thanks again, science!
The word "instantaneously" seems a bit strong, even assuming you don't mean it literally. Many different kinds of animals apparently exhibit many of the prerequisites or precursor traits that seem necessary for language, even if they don't posses all of them in order to make the leap humans apparently did. Communication is not unique to humans. Vocalization is not unique to humans. Awareness of self is likely not unique to humans. We are clearly unique in some way, but it still seems plausible that it was a gradual (if relatively rapid) process that could have occurred from the confluence of many unrelated circumstances and abilities that has simply only happened once on this planet (as far as we're aware). Language, and everything derived from it, is also a positive feedback loop that only accelerates its own advancement once it occurs. And by looking at the world today, it also seems to preclude similar advancements happening in other species once it's occurred in one.
> Abstract conceptual thinking cannot be separated from our language ability
1. Clearly false (what words do you use when you plan to throw a ball to a target?)
2. There is no reason to believe other animals are incapable of abstract thinking (although we can clearly see they are far less intelligent)
3. I don't know what your long screed about bigotry and religion has to do with language ability
Just imagine how slow university-level maths would be if you had to do all the thinking in words in your head.
Sorry if I sound rude, but this is such a common misconception people believe in, and it's so blatantly not true so I'm baffled people keep on claiming it is 'scientific' to believe that language is identical to intelligence. No wonder some people insist LLMs are just weeks away from becoming AGIs.
The invention of writing in Ur has an uncanny alignment with the Jewish calendar.
I recommend the book “Why Only Us?” by Chomsky and Berwick on this topic. It gets quite technical in places but I still got quite a lot of out it.
If you’re interested in this topic then you should definitely know the Chomsky/Berwick position, but it’s not the only view and has never been the consensus. For an alternative argument, “The Origin of Human Communication” by Michael Tomasello is good.
Can someone recommend good reading on the "psychology of languages", i.e., how does a language develop in the brain; why is it easy for me (as someone who grew up in the East) to learn guttural sounds of Arabic, but it's nearly impossible for my kids raised in the West? Similarly, I could never hope to speak clicking sounds of Southern African languages. Does the writing system affect speech, and if yes, then how? Like for example, in some Central Asian countries the alphabet changed multiple times within a century - from Arabic-based script, to Latin, then Cyrillic, then Latin again. What are some known neuroscientific tricks (for adults, not kids) for acquiring specific languages?
Chomsky has that wrong theory about language innateness. In reality languages that don't fit child brains don't survive. It's the other way around!
Not sure why you think that would prove Chomsky wrong. First, languages that don't fit brains (children's or otherwise) aren't going to be spoken by humans in the first place. And second, languages that fit our brains--all 7000 of them spoken (or signed) today, as well as those that are now extinct--don't fit chimpanzees' brains. So there's something about our brains that makes it possible for us to learn language(s).
Are Chomsky's ideas on language still taken seriously these days?
Yes, although there are other theories. Some of those other theories actually retain some of Chomsky's ideas, such as our innate capability to learn language (which seems to be the idea most commonly disagreed with). Two examples are Lexical Functional Grammar (LFG) and Head-driven Phrase Structure Grammar (HPSG).
Why not? Do you think they are somehow refuted?
The paper https://www.frontiersin.org/journals/psychology/articles/10.... appears in Frontiers in Psychology.
Does beginning a submission title with "Genomic study:" provide an air of credibility? As in, before any reference to what is being studied, genomic study is mentioned. I'm actually curious.
I think it's less about credibility in general rather than the source of the information. It could have been some sort of archaeological evidence instead.
I wonder, if it is possible to detect what exactly changed in brain, as this could be a clue to create GAI.
Even strict list of changed genes could be extremely helpful for AI progress.
Hey, heads up. It's now 2025, machines have mastered language, displaying better linguistic abilities than those of a large percentage of the population. They also do so using a small fraction of the neurons and synapses found in the human brain, despite being able to express themselves in tens or hundreds of languages and across an extremely wide range of subjects (much beyond the average general knowledge of human beings), dispelling the myth that artificial neurons are much less powerful than the real ones.
LLMs are interest case, but they are nearly flat. Neural system of mammals is structured, as I understand, neocortex consists of about 100 millions structures each ~ 100x400 neurons, something like this. Anyway, just from calculation of human thinking delay appear human NN is just about 500 layers.
Second difference, natural I are feed-forward network, not back propagated as typical AI NN, because natural use some chemical method to "calculate" parameters.
I think, with right structure and some FF method, GAI is already technically achieved.
PS FF NN already exist, but problem that it now using classical method of differential equations which is prohibiting real use, because too much computation need.
It is impossible to figure this out and it's dumb to try to give a number. We don't think Neanderthals could speak? They also had languages.
We already know that genes regulates speech - for example, FOXP2 [0] - and have successfully sequenced the human genome, and have started similar initiatives on other archaic human and primate species.
Phylogenetic Analysis has been fairly successful already in analyzing our genetic history, so I'm not sure why you'd think it's impossible.
There are many cases, when animals (or birds) literally speak and even showing very human like behavior on talking.
But, as I could see, only humans have so significant language culture that lead to great number of manuscripts.
I will even accent - little number of humans write books, but near none other species do this.
I'm not an expert, but I guess they mean "complex enough language" to discard isolated words and consider only "languages" that have "sentences that may have 5 words" or some more accurate criteria.
How do we know Neanderthals had languages? I'm curious what the evidence is for this.
We do not know whether Neanderthals could speak.
I'd say those guys back then started fuddling around with the definition of genocide about that time too.
Existence of language requires 2 mutation trees, one for hearing, one for sound making. Could this be why it took so long?
Sign language doesn’t need hearing or sound making!
I suppose language requires a means of communication and someone to communicate to at a minimum. Although actually we think in a language so I guess we don’t need someone to communicate to.
Then maybe language creation just requires the right sort of thoughts?
Edit:
I should have read the article first LOL!
They say as much:
> "Language is both a cognitive system and a communication system," Miyagawa says. "My guess is prior to 135,000 years ago, it did start out as a private cognitive system, but relatively quickly that turned into a communications system."
There’s a lot of this in evolution. Look into how possibly unlikely the emergence of eukaryotes was.
The most likely explanation for the Fermi paradox is just that high intelligence is incredibly rare.
Why are we here? Because we are having this conversation over a global digital network. The universe is probably full of bubbling mats of bacteria.
But hearing and sound making both have many other uses. Jellyfish can "hear", crickets can make noises. I'm guessing we had both of these long before language.