I'm not versed in either biology or linguistics so please forgive any naiveties I may commit.

I've learned that Noam Chomsky thinks that language is a result of a single genetic mutation in humans. It would be a mutation allowing the human brain to conceive of natural numbers as I understand. Have there been any (even speculative) attempts by biologists to locate that mutation in the human genome and/or in time? Is it even feasible?

  • $\begingroup$ It seems a very simplistic view. After all, spoken language relies on vocalization, which plenty of species have. On the other hand, sign language can be used for communication. There are also other types of non-verbal signaling in humans and other species. So what exactly is language? $\endgroup$ Aug 28, 2012 at 9:23
  • $\begingroup$ @Lubo As far as I know, most linguists agree that no animal apart from humans has been observed to use language. Animal communication lacks certain characteristics they require in a language. I understand that what Chomsky considers the constituent characteristic of language is what he calls recursion. $\endgroup$
    – ymar
    Aug 28, 2012 at 9:46
  • 1
    $\begingroup$ Please add a link to source your statement of Chomsky's theory. I very much doubt he claimed it was a single mutation. That is very unlikely. $\endgroup$
    – terdon
    Sep 20, 2012 at 16:04
  • $\begingroup$ @terdon en.wikipedia.org/wiki/Origin_of_language#Approaches That's all I've got. I understand that Chomsky's view is not shared by most scholars. $\endgroup$
    – ymar
    Sep 21, 2012 at 0:05
  • $\begingroup$ @ymar Chomsky's linguistic theories are rather mainstream, and have biological implications, as I discuss in my answer. The claim about a single mutation or gene is however a rather singular one, devoid of serious scientific support. $\endgroup$
    – Roger V.
    May 28, 2021 at 8:32

2 Answers 2


The anthropologist Svante Pääbo is more recently famous for trying to track down the 'language gene'. There isn't a lot of reference to Chomsky in his work as I've noticed, but it is to me the same intriguing idea.

That being that the chimpanzee and the bonobo have 99+% identical to human genome sequences and we also have data from human variations as well from a few thousand individuals. Using this data you could try to evaluate a small list of differences which might have led to the inborn trait human beings have to acquire language.

Pääbo's work focused on FOXP2, which leads to language disorders in humans when it is damaged in human beings. On the other hand its also known that primates can learn to speak rather well if they are trained their entire lives by the right people. They can even reflect on abstract ideas such as life after death. This is a typical result when looking for a single gene that causes something- it rarely leads to a conclusive result as a phenotype like verbality, or even height or body mass is the result of the action and fine tuning of many genes acting in concert.

That is, if you introduced human FoxP2 into transgenic chimps its unlikely that they would be as verbal as human beings. This current reference for instance talks about how important the slower development rate for human children is. This is one of the most unusual thing about human beings - it takes 1/3 to 1/5 of the human lifetime before we are fully mature. This phenotype is incredibly unlikely to be the result of FoxP2 alone, or any single gene.

There has been recent work published (2012) where the unique human FOXP2 gene was put into transgenic mice. They seem to have different sort of personality tendencies and are said to vocalize differently - brain activity profiles are also different. Kinda exciting, but not the Secret of NIMH yet....

  • $\begingroup$ Thanks a lot! As for "rather well", that strongly depends on the expectations. If you watch Koko on Youtube (and there are plenty of videos), you can indeed see many surprisingly human-like behaviors, but I was disappointed but her "language" personally. Her trainer was very clearly not objective. When Koko said the wrong thing, she'd say things like, "Why are you joking Koko?" and repeat that until Koko said what she was expected to say. Koko was largely unaware of syntax -- word order was random. $\endgroup$
    – ymar
    Sep 23, 2012 at 10:25
  • $\begingroup$ @ymar yes well - Koko and any other animal (including parrots) trained to cognate can't seem to get past 3-5 year old level. Another question that never seems to resolve. $\endgroup$
    – shigeta
    Sep 23, 2012 at 23:42

To my knowledge, Chomsky does not claim that there is a specific gene or mutation responsible for the human linguistic abilities - at least not in strictly biological sense. The claims of Chomsky and his followers, known as generative grammar, are more general: the study and the comparison of many human languages have shown that all the languages are structured according to the same principles (as opposed to the earlier Sapir-Whorf hypothesis that languages are products of purely random evolution). This similarity is the basis for the Chomsky's claim that this language structure is likely hard-coded in the human brain. This is further supported by mathematical complexity arguments, showing that the language rules are too complex, and the variety of the possible phrases that one can produce is too large to be learned within a reasonable time, without having a prior model of how the langauge is structured.

Thus, the linguistic ability, like many other traits, is likely a result of many mutations and action (and interaction) of many genes. Note that this involves not only the mental abilities, but also purely physical capacity to produce wide range of sounds that many animals do not possess.

Remark: Noam Chomsky is a renowned mathematical linguist, chose theories of generative/transformational grammar are pretty much the basis of the modern mainstream linguistics, and play an important role in the modern computer science. Thus, these are not to be dismissed, unlike his more controversial political views, which are anything but mainstream.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .