Talk:Universal grammar

Questions

It is difficult to understand what Universal Grammar is on the basis of this and other articles on Wikipedia (and elsewhere). Sometimes it is explained in a way that makes it sound completely obvious, and at other times it is explained in a way that makes it completely incomprehensible. As an example of the first phenomenon, one hears things like 'Proponents of this view argue that the pace at which children learn languages is inexplicably rapid, unless children have an innate ability to learn languages' (in the Noam Chomsky article, which is probably edited more often than this one). It is completely obvious that children have an innate ability to learn languages, since they learn languages, whereas cats, for example, don't. So unless linguists are using the words 'innate' and 'ability' in some unexplained way that does not correspond with the ordinary use of these words, this doesn't make any sense. So far, James Hurford seems to be right.

As an example of incomprehensibility – in this article it says 'Linguist Noam Chomsky made the argument that the human brain contains a limited set of rules for organizing language. In turn, there is an assumption that all languages have a common structural basis.' First of all, what does it mean for the brain to contain 'a set of rules'? Then at the end of the article, it says, 'Universal Grammar is made up of a set of rules that apply to most or all natural human languages. Most of these rules come in the form of "if a language has a feature X, it will also have the feature Y."' So it would seem that UG claims that the human brain, presumably already at birth, contains 'rules' like 'If a language has a word for purple, it will have a word for red.' It is nice to learn that my brain has innate knowledge of that, because just a few minutes ago I myself did not know it.

In any case, how is this rule about red and purple a rule about language? It seems to be a statement concerning color. It is roughly comparable to something like, 'if a child can count to ten, she can count to six'. KBry (talk) 18:01, 17 October 2009 (UTC)

answer to first criticizm

mentioned in the article: Some linguists oppose the universal grammar theory. Geoffrey Sampson maintains that universal grammar theories are not falsifiable, arguing that the grammatical generalizations made are simply observations about existing languages and not predictions about what is possible in a language.

first: the criticizm needs reason

second: every experiment is "observations about existing languages" but the result is used to predict future depending on the amount of times thhat the experiment is repeated and confirmed again. the languages of the world are so much that if we study even one hundred of them, we can rely on the results of this theory for prediction.

third: this so called theory is still a movement toward a theory. it has not been finished. so it must be clearly mentioned in sentences before it is rejected

Saeed.Veradi 07:20, 26 June 2008 (UTC)

Initial discussion

First proposed by Noam Chomsky, Universal Grammar is that part of language that can be considered to be innate. As children develop they go through a stage of language acquisition which occurs much more rapidly than for most other mental abilities. This, along with many structural similarities between languages, suggests that there is a part of the human brain pre-wired through evolution allowing us to quickly assimilate and recognize linguistic patterns.

For example, the use of certain types of verbs implies a fixed set of possible nouns. "Come" for example implies one associated noun, a subject. In English we say "John is coming", in French "Jean vient". "Send" implies two associated nouns, a subject and an object. "John is sending a letter" in English. Or "Jan posla dopis" in Czech. "Give" on the other hand implies three nouns; a subject, direct object and indirect object, in this case the recipient. In English: "John is giving the letter to Mary". In French: "Jean donne la lettre à Marie". While what makes languages unique is the inability of non-speakers to understand phrases, phrases and sentences governed by verbs show similar patterns through all languages, as illustrated by the previous sentences. This similarity suggests that while sound and word patterns are specifically unique for any given language, there remains certain aspects of the underlying structure which is universal to all languages.

Universal Grammar covers the search for these common linguistic reflexes, and research into the mechanisms at work within the brain that make language acquisition possible.

You are wrong. These restrictions are not restrictions of some "universal grammar", these are logical restrictions of our world. and another mistake: Jan posla dopis. is not correct. (wrong inflection) Jan posílá dopis. is correct. However, subject is not required in Czech, so posílá dopis ( =He is sending a letter) is correct as well. Languages are not as similar as you think.--SuperElephant 07:19, 3 September 2007 (UTC)

I have to disagree with SuperElephant on this one. Although inclusion of the subject may not be required in Czech (or Spanish or French, for that matter) the subject is still implied by the conjugation of the verb. Languages my friend, are MORE similar than you are willing to believe. Lpjurca (talk) 18:48, 20 May 2008 (UTC)

Article cleanup

There was considerable redundancy and verbal fat in this article. Expressions like underlying principles, shared in common, that there exists and all human beings are all good examples of verbal excess.--NathanHawking 00:34, 2004 Oct 14 (UTC)

It seems to me that the parenthetical aside in last sentence in the main entry is confusing: "Proponents of UG argue that their theories make extremely strong predictions of this kind (often too strong, failing to allow for grammatical phenomena which are in fact observed)."

Why would proponents argue that their theories are too strong? Isn't the parenthetical comment meant to say something like..."while opponents say that such predictions are often too strong...etc."? Or am I misreading the sentence? --User:Jeffmatt

I've removed it. This point cannot be dealt with in a sentence. It needs proper discussion, with examples and references. dab (ᛏ) 17:43, 18 February 2006 (UTC)

What does this mean?

The idea that universal grammar is supported by the creole languages is the fact that such languages all share certain features.

Should this be something like "...comes from the fact..."? Loganberry (Talk) 00:57, 22 June 2006 (UTC)

Eurocentric

The theory is also criticized because Chomsky makes too much of the similarity of a small group of languages (English, French, German, maybe some others - haven't read Chomsky lately, so I don't remember which), narrowing the subject matter so that one could hardly fail to find underlying forms. Truly universal grammar would include every actually existing language and any language that could exist (making it rather tough to falsify his theory, of course). 72.144.92.159 22:26, 14 February 2007 (UTC)

I'd love a citation for that for the article. CRGreathouse (t | c) 23:19, 14 February 2007 (UTC)

Erm...

I shan't edit the article because, frankly, this is a question. The poverty of stimulus argument seems rather out of whack. Sure, a child will not hear grammatical constructs unacceptable in the local language, but what of it? Having had experience with three year olds, they often produce themselves horribly incomprehensible (to anyone but their parents) grammatical nightmares. When they use them, adults tell them that what they have said makes no sense. That's not poverty of stimulus, that's an abundance of it.

In other words, while children may not hear unacceptable grammar very often, they can very easily come to understand what is and is not acceptable by experimentation.

As it currently reads, the argument in the article can be refuted by the very fact that children are not mute.

Either the article needs to do a better job of explaining the theory, or the theory is simply useless. I am more inclined towards the former. In either regard, I am not comptetent to fix the article. —The preceding unsigned comment was added by 69.113.219.44 (talk) 06:23, 11 March 2007 (UTC).

UG is falsifiable

"Some linguists oppose the universal grammar theory; it is outspokenly opposed by Geoffrey Sampson, who maintains that it is possible for children to learn a language without being born with grammatical rules. Sampson believes that universal grammar theories are not falsifiable, arguing that the grammatical generalizations made are simply observations about existing languages and not predictions about what is possible in a language."

Here's why I marked this with {{fact}}: If a language that violates UG is harder to learn, then there exists an experiment to judge whether language A is harder to learn than language B. Randomly selected parents fluent in A raise children. Randomly selected parents fluent in B raise children. If children in language A households take longer to speak "correctly", then language A is harder to learn.

Or by "falsifiable" do you mean "falsifiable without violating research ethics"? If so, please direct your comment to this other talk page, which I am also watching. --Damian Yerrick (talk | stalk) 16:42, 20 July 2007 (UTC)

that's nonsense. Your experiment just measures which language is harder, nothing to do with UG. Plus, there is no way you can {{fact}} a direct citation. The fact we report is that this is Geoffrey Sampson's opinion, not that UG is, in fact, not falsifiable. dab (𒁳) 17:18, 20 July 2007 (UTC)

If UG is defined as "patterns in language that the brain is wired to more easily accept", then "language A is hard" and "language A violates UG" are equivalent statements. And is a citation of an author's entire oeuvre considered a "direct citation", or can this paragraph cite a specific work by Sampson? --Damian Yerrick (talk | stalk) 17:37, 20 July 2007 (UTC)

You are saying "if you accept UG, you'll have to accept UG". That's precisely the argument of people who claim UG is not falsifiable: "violates UG" is just a fancy way of saying "is hard", then. Don't try to build a case here, ask for (and provide) citations. If your citation request is for a direct quote of Sampson, that's fair enough. dab (𒁳) 18:31, 20 July 2007 (UTC)

Even under a tautological definition of UG, a distinction must be made between the UG criticisms along the line "all languages are equally easy for a child to acquire" and those along the line "the learning tendencies that emerge from language acquisition do not rise to the level of 'grammatical rules'". And yes, I just want an example of a work so that other editors can interlibrary borrow a copy of one of his works and verify that he wrote what the article claims that he wrote. I hope that my revised position of the {{fact}} request within the paragraph makes this clearer. --Damian Yerrick (talk | stalk) 18:46, 20 July 2007 (UTC)

I don't think a "all languages are equally easy for a child to acquire" criticism even exists, since that's patently not the case. I agree we need a citation, and I suggest we strike Sampson, since he is, by all appearances, an idiot. Personally, I think UG is, in fact, falsifiable, and false at that. Unless, of course, you water it down enough so that it becomes just a tautology, in which case it is not false, but merely devoid of information. I'll provide some literature to that effect. dab (𒁳) 18:58, 20 July 2007 (UTC)

Idiot is who belives such pseudoscientific crap theories like universal grammar or intelligent design.--SuperElephant 07:47, 3 September 2007 (UTC)

Because many ( or even most? ) human languages have not been studied in any depth, the "rules" laid out in Universal Grammar are predictions of what will be discovered about languages whose structures have yet to be studied. Assuming they don't become extinct before we get around to it. By definition, this is falsifiable - just as Darwin himself predicted that, if his theory is correct, transitional fossils should be discovered. These aren't the types of predictions you would expect from particle physics being confirmed at CERN, but they're predictions nonetheless. —Preceding unsigned comment added by 207.138.32.33 (talk) 21:12, 30 December 2008 (UTC)

Examples

I was wondering if we could talk a bit about particular theories or aspects of particular theories of universal grammar, to illustrate the topic. I was also under the impression that some constructed languages violate the rules of some particular theories of universal grammar? -- Beland (talk) 17:54, 14 January 2008 (UTC)

Such an assertion about conlangs that fail UG used to be in the article until it got removed for lack of citation. --Damian Yerrick (talk | stalk) 18:27, 14 January 2008 (UTC)

Examples - rules laid out by UG - are sorely needed. I've only encountered a few specifics in my reading, and, honestly, they seem on weak footing to me. I'll add these, and hope that some ruleslawyering edit-warring Wikipedian doesn't remove them for personal taste reasons, but expect it to happen anyway. —Preceding unsigned comment added by 207.138.32.33 (talk) 21:15, 30 December 2008 (UTC)

Notes vs references

This is very confusing, and should be consolidated into only one list. —Preceding unsigned comment added by 207.138.32.33 (talk) 23:48, 30 December 2008 (UTC)

Criticism (CG, etc.)

The criticism section is still lacking the objection raised by Cognitive Linguistics (and largely confirmed by psycholinguistic data): that most or all "universals" we find in languages (e.g., categorical perception, physical and cognitive biases) are not specifically linguistic (and sometimes not even specifically human), and so the notion of a Universal Grammar is unnecessary and probably wrong (cf. Tomasello, Lakoff, etc.). Currently, the article makes it seem as though any confirmed universal must be evidence for "UG", which is obviously not true.

Incidentally, the article also fails to point out the fact that the overwhelming majority of psychologists, psycholinguists, biologists, educators and non-generative linguists don't buy the idea of UG as stated by Chomsky, precisely due to a lack of empirical evidence outside generative descriptive models.201.37.75.85 (talk) 22:32, 17 January 2010 (UTC)

Help needed for this poor article

This article is in sad shape, and the problems start from the very beginning. "Universal Grammar" was a term borrowed by Chomsky from French Enlightenment grammarians/philosophers -- and in Chomsky's usage was never intended or used as the name for a proposal or theory. So the very first sentence is wrong, and it's downhill from there on! "Universal Grammar" is the name for those aspects of linguistic structure whose properties arise from our common genetic heritage as humans, and more narrowly, for those properties that appear to be specific to language. The key point is that it's a name for those properties whatever they might turn out to be. That's how he introduced the term in his 1965 book Aspects of the Theory of Syntax (around page 6, if you want to look it up) and that's how he and others in the field have used it consistently ever since. Now Chomsky himself and many other linguists have made lots of specific proposals about what Universal Grammar does in fact consist of, and these are all controversial.

The task for a revision of this article needs to be first and foremost to make it clear how the term was used by Chomsky, how he adapted the earlier use by Enlightenment scholars, and then to give some NPOV characterization of how the wealth of commentary pitched as "criticism of UG" or "denial of UG" does or does not bear on actual proposals about UG.

In answer to the obvious critique of this comment - that "Wikipedia is written by us - why don't you start doing this yourself?", my worry is that the necessary revisions will inevitably amount to a rewriting of the entire article. That's a fearsome proposition, especially with so much more heat than light swirling around the concept "Universal Grammar" these days (see http://languagelog.ldc.upenn.edu/nll/?p=2507, for example). How can this be done in the real world in which we live?