2

The Generative approach on syntax is very elegant, useful and very complete as far as I can see. I think that, as all theories have, there must have some critics on it. But I don't know where to find counter examples. What are the theories that refute (or try to do so) the theories of generative syntax?

Also, if someone know about the same thing but on generative phonology, I would like to learn about it too.

4
  • 1
    The main criticism I've heard is against universal grammar, because it's an unexplained source of innate knowledge whereas other aspects of human behaviour seem to be acquired as infants and children. – curiousdannii Apr 6 '20 at 1:09
  • Modern versions of Generative Grammar on the syntax end assume strict binarity of branching. The notion that all syntactic structure involves binary branching is not supported empirically. The issue is discussed here: linguistics.stackexchange.com/questions/6826/… – Tim Osborne Apr 6 '20 at 4:10
  • Another problem with modern version of Generative Grammar is the assumption that syntactic structures are built bottom up. There is no empirical evidence for this assumption. What CAN be shown, in contrast, is that syntactic structures are produced and processed in an online fashion, from earlier to later in the flow of speech (i.e. left to write in languages that are written from left to right). – Tim Osborne Apr 6 '20 at 4:15
  • 1
    There are two lines of criticism: 1) generative syntax isn't cognitively adequate, 2) generative syntax doesn't model the syntax of natural languages well. As for the former, none of the current mainstream theories is cognitively adequate. On the other hand, generative syntax models fairly well the structure of natural languages, but it's not complete — it doesn't account (at least not sensibly) for phenomena such as agreement, long-distance dependencies etc. I'd recommend reading about constraint-based theories, which literature addresses some of these shortcomings of GG. – Atamiri Apr 6 '20 at 9:06
4

The best argument I've encountered against generative syntax is that made in C.F. Hockett's State of the Art. Personally, I don't subscribe to it, but you may find it persuasive. Hockett compares the game of professional baseball with the similar pick-up game played on vacant lots or in parks by young people -- "sandlot baseball". The professional game is fully codified, with clear and explicit rules, while the rules for sandlot baseball are ad hoc and may be decided by the participants, for convenience, and depend on who happens to be playing.

Generative grammar is like the pro game, while the true nature of language is more like the casual and constantly varying sandlot ball.

13
  • 1
    surely the players still obey natural laws. Though they seem to defy gravity, they don't get the ball going by changing the rules – vectory Apr 6 '20 at 15:04
  • @vectory Yes, the players still obey natural laws. To make use of that, we need to know what those natural laws are. It would obviously be circular to assume that the natural laws were the rule book for pro baseball, then to conclude that the sandlot players are also controlled by the strict determinate rules for the pro game. Unfortunately, – Greg Lee Apr 6 '20 at 23:39
  • the players do not need to know the laws, is my point. PS: Circular reasoning is not necessarily a fallacy, only if the initial premisses is fallacious, that need not be part of the circle, which is why the circle might be vicious – vectory Apr 6 '20 at 23:41
  • ... in the comparable case in grammar, in the latter part of Aspects, this is exactly what Chomsky does, when he discounts speakers' rejection of some complicated constructions as "performance" errors. When judging the evidence, speakers are allowed to use "pencil and paper" analysis (presumably using the competence theory as a guide). – Greg Lee Apr 6 '20 at 23:41
  • 1
    You are wrong. Wikipedia (List of Fallacies) says it's not a fallacy, and it makes sense. There might be a fallacy in a given circular argument, sure, but, for example, any tautology is circular. An axiomatic rule may count as tautology, though we don't know if it is really true, and some people might reject them. To stick with the metaphor, being able to run quick is not prerequisite for having a game of ball. – vectory Apr 7 '20 at 15:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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