It very much depends on how you define "language". For Chomsky, human language has to contain a capacity for recursion - and not just any old capacity, it has to be a genetic capacity. That puts Chomsky's model into a series of quandaries, which he has never fully explored or reconciled.

First is the "what and how" quandary: what is it about recursion that makes it so game-changing for cognition? Recursion must have had a radical effect on human cognition to completely replace a non-recursive population; but there is no indication - either back then or now - of what that radical effect could have been. Even today, recursion remains a minor backwater of language use, neither vital for communication nor commonplace.

Second is the "where and whom" quandary: to be a genetic capacity that every human has, it has to have evolved and dominated the whole population quite quickly. Unfortunately, humans were already quite widely spread by 80kya (thousand years ago), so we would expect to see a centre of innovation with change rippling out from that centre - and we don't.

This leads on to the "when" quandary: 80kya to 40kya gives only about 1600 generations for all non-recursive humans to be replaced by recursive humans. We would expect to find isolated populations genetically incapable of recursion; but all we have is one group, the Piraha, with no evidence that their lack of recursion is genetically, rather than culturally, inspired.

And finally, there is the "why" quandary: why is recursion even being considered? This is where the argument becomes somewhat circular. Language is a special form of communication because only humans use it; so there must be a magic extra to human communication; and that extra is recursion, because only humans use it; making language a special form of communication.

Which brings us back to our definition of language. Is it just the way humans communicate, and not significantly different from other communication systems in nature? Or is it part of a myth of human specialness? Human language is special to humans in the same way that any species-defining functionality is special to members of that species. That does not make humans more special than other species, only different. Language is a tool we carry around in our heads, but a tool nonetheless. It is not language that makes humans special, but humans that make language special.

So, to answer your question, who is correct, they all are (and many others besides). Chomsky is correct in that language is special to humans, but wrong in that language is not special of itself. Everett is correct in that the Piraha pose a difficult case for universal recursion to address, and in that language is primarily a social tool, not a magical spell. And Kimura is correct about the physically proven change in the vocal tract; but how that physical change affected human communication remains contentious. For a start, most chimpanzee calls are collections of "vowels" without "consonants", and unstopped, voiced vocalisations are common throughout nature.

I hope this helps.

Thanks for your wonderful response, Martin.

The question of the date of the beginning of human verbal language has been a subject of debate among linguists, anthropologists, and psychologists. Noam Chomsky did not propose a specific date for the origin of verbal language, but rather argued for the existence of an innate and universal capacity for language in humans. His approach focuses on the structure and acquisition of language, not its historical chronology.

Chomsky's Approach to Language

Universal Grammar: Chomsky argued that all languages share a common underlying structure, called universal grammar, which is innate in humans and allows for rapid language acquisition in childhood (Chaudhary & Singh, 2024; Marler, 2007; Railianova, 2024).

Acquisition vs. Learning: He argued that language is not learned solely through imitation or reinforcement, but is acquired through specific and universal brain mechanisms (Marler, 2007; Ornat & Gallo, 2004; Chaudhary & Singh, 2024).

Criticism of previous theories: Chomsky challenged behaviorist explanations, which viewed language as a product of experience and learning, proposing instead a biological and universal basis (Marler, 2007; Ornat & Gallo, 2004; Chaudhary & Singh, 2024).

What does research say about the date of onset?

Chomsky and chronology: None of the cited works attribute a specific date to Chomsky for the onset of verbal language. His primary interest has been the "how" rather than the "when" of language (Marler, 2007; Railianova, 2024; Chaudhary & Singh, 2024).

... Historical debate: Other authors have discussed the history of linguistics and the origin of language, but there is no consensus on a precise date, and Chomsky's position focuses on universality and innateness, not on dating (Aarsleff, 1970; Railianova, 2024).

Summary of Positions on the Origin of Verbal Language

Position Main Focus Proposed Start Date Citations

Chomsky Innate ability, universal grammar Not specified (Marler, 2007; Railianova, 2024; Chaudhary & Singh, 2024)

Behaviorism (Skinner) Reinforcement/imitation learning Not specified (Ornat & Gallo, 2004)

History of Linguistics Debate on Philosophical Influences Not specified (Aarsleff, 1970; Railianova, 2024)

Conclusion

Chomsky did not propose a specific date for the origin of verbal language, but rather revolutionized linguistics by proposing that the capacity for language is innate and universal in humans. The exact date of the origin of verbal language remains debated, and there is no scientific consensus on the matter.

References

Marler, P. (2007). A Dynamic Systems Approach to Development: Applications; The Language Instinct. Journal of Cognitive Neuroscience, 7, 101-109. https://doi.org/10.1162/jocn.1995.7.1.101

Aarsleff, H. (1970). The History of Linguistics and Professor Chomsky. Language, 46, 570. https://doi.org/10.2307/412308

Railianova, V. (2024). DEVELOPING IDEA OF UNIVERSAL GRAMMAR VIA NONINVASIVE IMAGING TECHNIQUES. DÉBATS SCIENTIFIQUES ET ORIENTATIONS PROSPECTIVES DU DÉVELOPPEMENT SCIENTIFIQUE. https://doi.org/10.36074/logos-01.03.2024.063

Ornat, S., & Gallo, P. (2004). Acquisition, Learning, or Development of Language? Skinner's “Verbal Behavior” Revisited. The Spanish Journal of Psychology, 7, 161 - 170. https://doi.org/10.1017/S1138741600004868

Chaudhary, A., & Singh, S. (2024). Noam Chomsky's Contribution to Linguistics. International Journal For Multidisciplinary Research. https://doi.org/10.36948/ijfmr.2024.v06i03.23721

It is interesting that your question asks about the start-date of “verbal” language. Chomsky, in fact, considers verbal language to be external to the core of language. He asserted many times that the essence of language is for expressing complex thoughts in the brain rather than for overt verbal communication. Therefore, to him, the start of language is the start of grammatical processes like recursion rather than the start of verbal language.

Also, Chomsky’s recursion is executed with words. But words are based on phonology. Without phonology, there cannot be words. But phonology is already very complex, and it is not just about vowels as concerned byKimura (1993), and it is not even just about vowels AND consonants, because they both need to be organized into syllables, rather than floating around on their own, to form phonological words.

Even syllables are already immensely complex, as beside human, only parrots and a few mammal species can imitate human syllables (They do not produce syllables in their own calls). The articulation of the syllable, as argued in Xu (2025), requires the synchronized onset of consonant, vowel, and tonal pitch, which likely requires dedicated neural circuits in the brain. So the origin of language should at least start from genetically enabled capacity to produce syllables. Complex grammar, including recursion, can only come later. This also means that the evolution of language into its homo sapiens form is likely a prolonged process across many, many generations.

Xu, Y. (2025). Syllable as a synchronization mechanism that makes human speech possible. Brain Sciences 15(33): doi.org/10.3390/brainsci15010033.

In response to Martin Edwardes:

If you check out the Old Testament of the Bible, King James, which is a translation, the text is highly non-recursive. I am checking with my Israeli colleagues to determine whether the Torah is also non-recursive. If so, this would suggest that ancient biblical texts were non-recursive and that recursion was introduced into written language post-300 BCE, the approximate date of the Old Testament. This would further highlight the contribution of Everett to language studies and that recursion is an invention much like AI--thereby suggesting that learning at the synapses and cultural transmission are central to recursion.

All responses on this point are welcome.

Edward,

If recursion is a linguistic trope to complicate language (a "shibboleth" device to identify strangers - which seems possible) then the fact that the Bible has been translated multiple times may be why the recursion has been ironed out of it. As the Torah has retained its single-language form, you may find less ironing out. I am not a Biblical scholar, so this is just a thought.

We also have to consider that linguistic recursion may be a transference from vocal recursion (which has been detected in the vocalisations of some nonhumans). What we think of as poetics do have recursive features in both rhythm and rhyme; they are not semantically recursive, but they do affect semanticity and semiosis (think of nursery rhymes, like "Jack Spratt could eat no fat, his wife could eat no lean; and so between the two of them they licked the platter clean"). Perhaps part of the problem is in the definition of recursion as a purely semantic phenomenon. Just another thought.

Thank you, Martin. My interest here is not recursion, per se, but rather to have an understanding of how the brain at the synaptic level processes different languages including the language of mathematics. Our book on this is currently being reviewed by Ox U Press. So far, one bad review (two more to go). If the British don’t bite then it’s off to Cambridge Mass at MIT Press and then to Stanford U Press, etc. You see where this is going.

Cheers,

Ed Tehovnik