What are the most convincing and most popular arguments against the Innateness Hypothesis of Universal Grammar or Universal Grammar as described by Chomsky?

  • What do you mean by Universal Grammar? Do you mean Chomsky's Innateness Hypothesis, or his current formulation of the nature language faculty? – Dan Milway Nov 6 '11 at 14:40
  • I was thinking of the Innateness Hypothesis, but has it been superseded? By Chomsky himself? – Lucas Nov 6 '11 at 14:56
  • No, It's still the basis for most of current linguistics but the field has split into various frameworks for explaining innateness. – Dan Milway Nov 6 '11 at 15:04
  • Here is an essay you can find interesting: timothyjpmason.com/WebPages/LangTeach/CounterChomsky.htm – Joe Nov 6 '11 at 16:48
  • 3
    After reading some of the arguments, here's a pretty humourous counterpoint from Elan Dresher of the University of Toronto – Dan Milway Nov 6 '11 at 17:12
up vote 8 down vote accepted

Obviously there is much written on this topic. A good place to start reading might be Evans & Levinson's (2009) article in Behavioral and Brain Sciences which is accompanied by responses.

See The myth of language universals: Language diversity and its importance for cognitive science by Nicholas Evans and Stephen C. Levinson.

  • 2
    Journals in linguistics are closed to outsiders--- it is hardly possible to even put a mathematical description of English grammar. The innateness of recursion is dead, it's wrong, and Evans defends it with the most pompous nonsense against the ideal counterexample of Piraha. There is nothing to say--- he is not a serious person, and I don't see any point in referencing him. – Ron Maimon Mar 12 '12 at 23:18
  • 1
    @RonMaimon are you talking about the same Evans (& Levinson) who write "Recursion ... is a capacity languages may exhibit, not a universally present feature." (443) ? (BBS is not specifically a linguistics journal, but I admit that it is difficult to get published there) – jlovegren Mar 13 '12 at 0:48
  • 1
    Please let's avoid personal attacks. If you disagree with someone ideas or even if you dislike them, then write "I dislike X's ideas" or "I strongly disagree with X ideas". – Alenanno Nov 4 '12 at 20:16
  • 2
    This isn't an answer, merely a pointer to an answer. – Joe Dec 10 '12 at 5:58

You may be interested in A Thousand Plateaus, a book Felix Guattari co-authored with Gilles Deleuze, in which they discuss Chomskyian linguistics in some depth and with a great deal of care.

Some caution may be warranted: they may seem to take Chomsky rather lightly, and their position may appear to be antithetically opposed in many ways -- you will find them, for instance, arguing against 'tree-based' sentence analysis, promoting in its stead a kind of 'an-hierarchical' or 'rhizomatic' language analysis emphasizing the pragmatic and collective aspects of discourse. However, a careful reading of this work will find a unique encounter with Chomsky's linguistics that is well worth the time to unpack.

They also offer a detailed reading of the Labov-Chomsky debate, which may be of some interest in this context.

The UG proposed by Chomsky has to be one that is consistent with transformational grammar, because Chomsky proposed TG. Chomsky does not like to be wrong.

But TG is wrong -- there are no transformations. Once we give up on transformations, we can see that the answer is at hand. There is a UG, which was originally proposed by Chomsky himself, and that is CFG (context free phrase structure grammar). The basic problem with Chomsky's latter day attempts to find a satisfactory UG theory is that he is looking for one that is consistent with some form of TG. Such a UG theory will never be found, not because the UG hypothesis is wrong, but because there are no transformations.

Well, then, if the UG problem was solved over 50 years ago, why are we still suffering over whether there is a UG theory? It's because there are some problems with CFG to get past. I think they're all soluble, but it takes some work. Here's a list:

  1. Describing constructions by appealing to movement of constituents is part of traditional grammar. How can there not be movement?
  2. It seems evident that sometimes word choice is not independent of context, so CFG is doomed from the outset, because it is context free.
  3. In his profoundly influential classic Syntactic Structures, Chomsky introduced CFG only to knock it down. He offered there several empirical arguments that natural languages cannot be described by CFGs. Other such evidence has been found since, notably the cross-serial constructions investigated by Stuart Shieber.
  4. Once Gerald Gazdar had shown how it was possible to get past the difficulty of describing apparent movement in a CFG, the descriptive grammar of English that emerged from applying the theory, GPSG (Generalized Phrase Structure Grammar), had too many rules and categories. It seemed obvious that mere humans could not cope with the tremendous complexity of detail that seemed necessary to describe a natural language within the confines of CFG.

protected by Alenanno Sep 27 '16 at 22:23

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

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