From the Minimalist Program, Chapter 1 (20th Anniversary Edition, pg. 16):

It has sometimes been argued that linguistic theory must meet the empirical condition that it account for the ease and rapidity of parsing. But parsing does not, in fact, have these properties…

In general, it is not the case that language is readily usable or “designed for use.” The subparts that are used are usable, trivially.

There is no a priori reason to expect that the languages permitted by UG be attainable under normal circumstances.

If proposals within the P&P approach are close to the mark, then it will follow that languages are in fact learnable, but that is an empirical discovery, and a rather surprising one.

(edited and with my own omissions)

Why does Chomsky think UG would have evolved at all, if only to generate unlearnable languages?

What does he consider “attainable under normal circumstances” - environmental factors regarding availability of language “training” data, or the language itself being too complex for the human mind to grasp, or something?


1 Answer 1


Chomsky finds it surprising kids can learn languages for a couple reasons.

First, he believes all human languages follow some universal grammar rules built into our brains. But these rules could logically allow for tons of really complicated languages - even ones too wild for anyone to actually learn.

There's no reason to think the languages that evolved would just happen to be simple enough for kids to figure out. It could have been a mess! Second, children don't get perfect examples of language when they're learning. It's all mixed up, half-finished, with mistakes. It's amazing they can make sense of the disorder and get the grammar right.

It's like if I gave you a pile of jumbled lego pieces, only some assembly instructions, and you had to recreate a huge lego castle perfectly.

So Chomsky believes children have an innate universal grammar ability. But he finds it really surprising, almost unbelievable, that the complicated languages this allows are the same ones kids can actually learn based on the limited scraps of language they hear.

That's why he says language learning being possible at all is a lucky accident, not something he would have predicted from his theory. It just happened to work out, which is weird if you think about it. So in short, he finds it almost unbelievable that the grammar kids have built into their brains lines up so well with the messy complicated languages we speak.

  • 2
    I don't quite understand. Languages aren't subject to "natural selection" generally speaking, but in the particular example of a language that starts being used by adults but that children cannot learn, that language will die out as the adults who speak it die, and as such, it's extremely improbable that we'd be able to witness such a language, even if one or several had existed in the past. Their only fate can be death.
    – LjL
    Jul 22, 2023 at 19:48
  • 2
    It seems you're suggesting a kind of 'cultural selection' happening with languages. Languages that can't be learned by children would indeed likely disappear over time, effectively being 'selected out'. This is a compelling counterpoint to Chomsky's surprise about children's ability to learn complex languages. But I am not Chomsky and do not advocate his view. I only tried to explain WHY he might think so. Jul 22, 2023 at 19:52
  • 1
    I am not sure I believe in complicated languages. Also, he probably only speaks English. [still alive last time I looked].
    – Lambie
    Jul 23, 2023 at 19:21
  • I’m not sure about this answer, but I appreciate the attempt. I wonder if the key word is “learnable”, in that Chomsky is saying that it would be more expected for language to be fully innate with almost no external learning from the environment, but that there is some learning to a degree; possibly. Jul 26, 2023 at 4:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.