Talk:Colorless green ideas sleep furiously

From Wikipedia, the free encyclopedia

Contents

[edit] Three reasons?

Talk:Noam Chomsky has three different reasons why this sentence was created. Which is right?

  • "Colorless green ideas sleep furiously." This sentence was invented by Noam Chomsky as an example of a sentence whose grammar is correct but for which the semantics are nonsense.
  • Afair he invented it to shop sentence, which grammar was correct, but probablity of appearance of every word after its precedensor was almost zero, and it had nothing to do with semantics, but with criticizing some probablity-based theories of language.
  • Oh boy. I had remembered it as intending to show that the function of a word was dictated by its position in a sentence. For instance, that in English, adjectives come before nouns and adverbs come after verbs.

"Afair he invented it to shop sentence, which grammar was correct, but probablity of appearance of every word after its precedensor was almost zero, and it had nothing to do with semantics, but with criticizing some probablity-based theories of language." This is as good an example of gibberish as I have seen in a long time. I would prefer to see Chomsky discussed by people who have understood what he was saying, and I do not think most of you have. Dr Dark.

The second one I've not heard before, I think that's more of a side effect than Chomsky's main point. The other two seem to be saying the same thing, ie a sentence can be grammatically correct, but meaningless, or if you like "there's more to a sensible sentence than correct grammar." At http://www.cs.buffalo.edu/~rapaport/675w/colorless.html Chomsky is quoted as writing "Sentences (1) and (2) are equally nonsensical, but any speaker of English will recognize that only the former is grammatical.

  • 1. Colorless green ideas sleep furiously.
  • 2. Furiously sleep ideas green colorless."

from "Syntactic Structures" (1957), apparently, which should probably be added to the article if it's correct. --Camembert 12:22 Aug 1, 2002 (PDT)

On the other hand, (2) makes great beat poetry if you break off the last two words with pauses. --Brion VIBBER

A genius Chomsky he is, from my long gone school years of "Olympiads in Mathematics an Linguistics" I remember one problem: to define a plausible context in which the famous Chomsky phrase becomes semantically sensible. Redefinitions of the constituent words were forbidden. Mikkalai 07:12, 21 Jan 2004 (UTC)

Please, someone render this piece at the article's beginning into correct English.

The phrase offered proof that grammar was not the valid structure underlying language (as was thought at the time), rather that words are symbols with associated properties that will not function if they are not properly used. In demonstrating that "language" is not particular to any one localized language or culture, Chomsky's work established the theoretical basis for machine translation, and cognitive linguistics—the study of a "species-innate" cognitive structure, that provides for language development.

Mikkalai 21:13, 25 Feb 2004 (UTC)

The piece is good, although complicated, English. However I'll try to break it up into simpler sentences (without damaging the meaning, I hope). -- Derek Ross 22:43, 25 Feb 2004 (UTC)

That's what I was saying at your talk page; the English seems good. But the phrase "grammar was not the valid structure underlying language (as was thought at the time)" alarms me, to put it mildly. Something along the lines of your thought when you replaced the perfect English phrase
"that valid grammar (which dominated early 20th century linguistics) is not necessary for generating meaningful language".
by
"valid grammar (which dominated early 20th century linguistics) does not necessarily generate meaningful language".
at the Chomskybot. (BTW precisely because of that change I asked for your assistance, assuming you know what's this all about)

Mikkalai 23:44, 25 Feb 2004 (UTC)

Okay I hope that the new text is a little easier to understand. I have changed not the valid structure to not the fundamental structure because I think that good grammar normally conveys meaning more clearly than bad grammar, so in that respect it is a valid structure, just not the most important one. Likewise I had to think hard about what Steve meant by the second sentence. I hope that I'm not too far out with my paraphrase. -- Derek Ross 23:52, 25 Feb 2004 (UTC)

In both cases you hit the nail. If *you* had to think hard, what about us ignorant ones? :-) Mikkalai 00:16, 26 Feb 2004 (UTC)

The opening of the article still implies that before 1957, people believed that any grammatically correct sentence was meaningful. Now, I am no linguist, but I find that pretty hard to swallow. I can't really accept that anybody, even less so a person who devotes their life to the study of language, would ever believe something so easy to find counter-examples to. It seems to me the importance of this sentence lies elsewhere. Someone more well-versed in this really ought to clear this up. --130.232.31.109 17:14, 6 September 2005 (UTC)

Disappointed these related links were removed from the article, but I won't fight it. <>< tbc 00:01, 28 Apr 2005 (UTC)

John Hollander wrote a poem titled "Coiled Alizarine" in his book, The Night Mirror. It ends with Chomsky's sentence.

The results of a 1985 literary competition in which the contestants attempted to make this sentence meaningful using not more than 100 words of prose or 14 lines of verse has also been published.

I'm not familiar with the second link, but it seems very odd indeed not to include the first. I would have thought the poem in its entireity (three lines, after all) would have been appropriate, since it is a direct (albeit somewhat facetious) response to the claim being made by Chomsky, it has 'for Noam Chomsky' as a subtitle, and is an oft related part of the story of this sentence. Ncsaint 22:30, 20 October 2005 (UTC)

[edit] Vacuous truth

I do not agree with this:

 Notice that, instead, saying all colorless green ideas sleep furiously is not only correct, but even true (see vacuous truth) since, as there is no such a thing as a colorless green idea, there can't even be such a thing as a colorless green idea which does not sleep furiously, which would be needed to deny this statement.

If "sleep furiously" is meaningless, then the whole proposition in which it appears has to be meaningless, for there is no way to determine "sleep furiously"'s impact on the other parts. Consider this sentence:

 All colorless green ideas formed one after another the driving force in the Revolution.

Is is true? If it is, then the driving force in the Revolution is the set of colorless green ideas, that is, nothing. The important word here is "the". (Note: my sentence may not be perfect English, as I'm not a native speaker. I hope you can still see my point. Suggestions for a better sentence are welcome!)

I should add that I strongly disagree with the meaninglessness of "sleep furiously"; I'm just trying to point out what I think is a contradiction.

Palpalpalpal 13:29, 6 September 2005 (UTC)

As a mathematician, every sentence of the form all foo bar, where the class of objects clasified as foo is empty, is true. Your sentence is more complex than that; it could be restated as "The driving force in the revolution was foo", where the fact that the set of foo is empty is important and negates the sentence.--Prosfilaes 03:13, 1 October 2005 (UTC)


Obviously it is true that in logic, the sentences you describe are considered true. But there is nonetheless some very poor reasoning going on here. Logicians typically translate a sentence of the form "all colorless green ideas sleep furiously" as "for all x, if x is a CGI, x sleeps furiously". The rules of modern logic render that true when no x are CGI. But this does not mean that we can go back to the sentence in English and say that, in natural language, the sentence is always true. Rather, if actual English speakers tend to think of that sentence as false or meaningless, all we have demonstrated is that we have rendered it into logical notation incorrectly. One might argue that if people generally consider statements about empty categories to be false, a more faithful rendering would include the claim that there exists and x such that x is a CGI. -- Ncsaint 18:54, 4 December 2005 (UTC)
I think I did not make my point clear. Of course I agree that "(X is the empty set) implies (for all x, (x is in X) implies P(x))" is true, regardless of the interpretation of P. My point is that Chomsky's sentence cannot be translated to such formal language (if we admit that parts of that sentence are nonsensical), because the result (which includes both a sequence of symbols and an interpretation function for the symbols) would not be well-defined. "Sleep furiously" cannot be used as a value for P, just like the number 6 or the operator +. Actually, maybe we could find a point of view where 6 or + would be acceptable, but when I just wrote "sleep furiously", I meant _the meaning_ of "sleep furiously", which is what we admitted as non-existing. There is no function that returns something that doesn't exist, thus no interpretation function can be defined.
I thought my example sentence would be obvious to people familiar with translating natural language into formal language, which is the problem at hand: presuppositions usually affect the whole discourse, and are notoriously non-obvious to translate: "all the balls are in the hat", which presupposes the existence of the hat, cannot be said to be true when there is no hat, even if there are no balls. Palpalpalpal 23:00, 13 February 2006 (UTC)
I don't know if you just can't explain your example or what. I did a preliminary proof(only half assed, but I belive accurate) that your claim is incorrect. That is, provided I understood the previous paragraph correctly. Admittedly I am not that great at predicate logic but here we go. B is the state of being a ball, H is the state of being a hat, and I is the state of being inside of any hat (rather that being inside of a particular [the] hat). The fact that
\neg (\exists x)Hx \to \neg (\exists x)Ix (i.e. If hats don't exist, then objects inside of hats don't exist either.) and also the fact that
(\exists x)Bx \lor \neg (\exists x)Bx(i.e. Either balls exist or they do not. Obvious I know, but you said it so added for completeness.) both imply that
\neg (\exists x)Hx \to \neg (\forall x)(Bx \to Ix) (i.e. If hats don't exist, then it is not the case that all balls are inside of hats.)
That is what you were saying right? Or am I missing something subtle? If I mischaracterized your argument, please clarify, otherwise let me know and a proof to the contrary will be forthcoming for your perusal. Again, I am not a logician, and I am not great with wiki formatting, which is why I request clarification. It is my belief that, at least in this case, that the natural language meaning is unambiguous. I don't actually have time at the moment to write up the proof but consider this: If balls exist, and hats don't exist, then obviously there have to be balls that aren't in any hats. But let's say that balls don't exist now and hats still do not exist. Nothing can be inside of a hat; that is obvious. We can also say of the scenario though, there are no balls that are not in hats, or more formally \neg (\exists x)(Bx \land \neg Ix) if we switch to a universal quantifier, simplify by DeMorgan's rule, and remove the double negation, that is equivalent to (\forall x)(\neg Bx \lor Ix) which is itself tautologous to (\forall x)(Bx \to Ix). And so I have deduced, albeit informally that it is true that "all the balls are in the hat" whether or not "are in the hat" has any meaning. Similarly it can be shown that "all colorless green ideas sleep furiously" irrespective of whether or not it is the case that there is at least an object x, such that x sleeps furiously. If that wasn't a good enough explanation, I will include a formal proof(as formal as I can manage) for verification. Although I suppose technically any proof would be merely restating what a material implication is. 66.102.196.3 00:48, 12 May 2006 (UTC)
Just an after thought, but maybe you were trying to say that meaningless predicates have no meaningful truth value? 66.102.196.3 16:06, 12 May 2006 (UTC)
I don't completely agree. The fact that a predicate, made of legitimate words combined in legitimate ways, is self-contraddictory, doesn't make the predicate meaningless, it just makes it always false. For example, nobody would call "meaningless" the statement "1987 (or 2000, or 23, or any other number) is a composite prime number", despite the fact that there is no such a thing as a composite prime number. Despite the fact that there is no such a thing as a time when Immanuel Kant was playing an electric guitar, the sentence "when I was born (or on 1 January 1900 at noon, or any other time) Immanuel Kant was playing an electric guitar" is not meaningless. It is false. If there existed no such a thing as a hat, the statement "this ball is inside a hat" (supposing "this ball" exists) is not meaningless. It is false. The sentence "I'm currently on planet Namek" is not meaningless. It is false.

So the fact that it is impossible to sleep furiously (or is it?) doesn't make the predicate "x is sleeping furiously" meaningless. It makes it false. --Army1987 11:02, 15 December 2006 (UTC)

[edit] Incorrect / unlikely phrase structure diagram

The current structure given for this sentence is incorrect because it has an NP being headed by an adjective ('colorless'). I've included here an x-bar diagram that shows the structure using a more traditional analysis. I've kept the S --> NP VP (instead of IP --> NP I') to avoid making the diagram too theory specific. If no one has any objections, I would suggest that we change the current image to one that is similar to the image below. --lxowle

X-Bar representation of "Colorless green ideas sleep furiously"

Um, that digram is either anachronistic (X-bar theory didn't exist at the time of at which Chomsky invented this example [1965, in Aspects of the Theory of Syntax]) or insufficiently current ("NPs" are now thought to be headed by determiners, so more correctly, DPs). Not that it matters, given the advent of bare phrase structure.

[edit] Purpose of the sentence

The article focuses on a relatively insignificant aspect - the sentence's meaninglessness. But the main point was proving that probabilistic models of grammar are inadequate, and for that the meaninglessness doesn't matter, it was just enough for neither the sentence nor any part of it to had ever been used in English. Taw 00:06, 4 December 2005 (UTC)

[edit] umm

Wouldn't using alternative meanings for each of the words to generate a meaning for the sentence be irrelevant to Chomsky's point since, green (the color) and green (the property of being unexperienced) are distinct words, despite their morphemetic and phonetic identity? I mean, it seems besides the point that some of the words have multiple meanings and we can pretend that they assemble into some weird-ass metaphor about ideas. Also, reiterating what that other guy said about how the point had more to do with statistical modeling of language than about the meaninglessness of this sentence.

If the two senses of "green" you mention were completely unrelated, you'd have a point. But one can consider (and I'd bet etymology would confirm that this is the origin of the "alternative" sense) that "green" can be narrowed to a (generally) distinctive property of "not ripened" fruits and vegetables, and thus makes a good metaphor for "young/unexperienced". Why Chomsky didn't use a color which could not lead to such interpretations, I don't know — maybe there aren't any. However, in any case, if the phrase is meant to exemplify nonsensical grammaticality, it cannot afford any interpretation. For if the phrase has one interpretation, just why can't we find a phrase which has none? Palpalpalpal 23:37, 13 February 2006 (UTC)

[edit] Article title

Shouldn't this article be titled "Colorless green ideas sleep furiously." with a period? It is a sentence, after all. Unless there is some naming convention or technical restriction of which I am not aware, I propose that this page be moved to the aforementioned title. If there are no objections, I'll do it in a few days. Imaginaryoctopus(talk) 06:28, 30 December 2005 (UTC)

I've moved it back, as virtually all of the incoming links don't have the period. The period isn't essential in the article title, as the phrase could be grammatically used as part of a longer sentence, for example. — sjorford++ 10:07, 25 May 2006 (UTC)
Like "Colorless green ideas sleep furiously as another pair of socks drapes backwards"? ~user:orngjce223how am I typing? 02:30, 7 January 2007 (UTC)

[edit] Picture

There is a rather famous picture of Chomsky standing in front of a blackboard with the given sentence written on it. It would be great, I think, if that picture was added to this article. I also found an other suitable picture for this article on the following site: http://www.sccs.swarthmore.edu/users/01/bnewman/visions/cgisf.gif. I will it at that, though, since I am not sure about the legal issue pertaining to the use and enjoyment of these pictures. -PJ

[edit] Eduard Vilde

  • Eduard doesn't have his own page yet, so I'll have to ask this here... the dates in this sentance seem entirely wrong... he died in 1933, and then published a book in 1954 containing theories he'd written about in the late 1930s. I can understand him dying in 1933 and then a book of his works being published at a later date (1954), but the "late 1930s" doesn't seem to fit there at all. Can somebody familiar with this material check this for me? Thanks.
the Estonian writer Eduard Vilde (1865–1933) made the same point in his 1954 book (which was mostly written in the late 1930s), writing "The urchin sat on the roof and cried, as if rabbits were running along the tin-pipe towards the eternity, holding lycoperdons in their mouths", etc.
Maelwys 20:21, 2 March 2006 (UTC)

[edit] Alternative Meaning

Colorless could mean that ideas don't not have color because they are untanglable things. Green could mean that the idea of which the sentence refers to is about nature and respecting the enviroment. The reason for this is that many times people say "Think green". Sleeping furiously could mean that the ideas is left untouched for awhile, but only because nothing can be done about it, therefore it sleeps furiously.

[edit] "The full passage says..."

Is it possible we can have the quote unbolded? It makes it fantastically annoying to read when applied to such an amount of text. W guice 20:51, 14 April 2006 (UTC)

[edit] "Colorless green ideas sleep furiously" =? "meaningless"

Is it fair to assume that by now this phrase is now well known enough to be synonymous with the idea of meaninglessness? (Which would, of course, mean that it is no longer a meaningless phrase.) ----Andymussell 02:01, 30 September 2006 (UTC)

Question from a different user: Anybody try asking Chomsky what the purpose of the CGI sentence is? He's still alive (as of Jan 8 07).

The purpose was to provide a sentence that people had not encountered before, and yet is grammatical. "Colorless green ideas sleep furiously" would never be generated by a Markov Chain (well, now it would, but when he introduced it, it wouldn't have.) The sentence didn't need to be (and still doesn't need to be) meaningless, it just had to be something that a Markov Chain would never generate, and something a child would never have heard. If you can take something that it is guarenteed that no child has heard before, and who's individual words would not occur in succession, and despite this it is still grammatical to them, then you prove that the human mind doesn't work on a Markov Chain. The mere fact that Chomsky was not imaginative enough to create a statistical model that would allow the two to be distinguished doesn't stop it from breaking people's ideas that simple statistical models showed human usage. --Puellanivis 19:29, 8 January 2007 (UTC)
I agree with Puellanivis, completely. There's no ulterior motives behind his sentence and no hidden meanings like "green" meaning "novice". It's akin to Syntax teachers using the Jabberwocky as an example of how the structure, not the words themselves, can give meaning. JesseRafe 07:10, 9 January 2007 (UTC)

[edit] Lucien Tesnère's Quote

I've update the translation. "Le silence vertébral indispose la voile licite" is more accuratly reflected with "The vertebral silence indisposes the licit sail" than "The vertabral silence worries the legal sail". 156.34.25.187 03:26, 5 March 2007 (UTC)

[edit] Picture

Which picture is better?

The current one: Image:Syntax tree.png Or this one: Image:Cgisf-tgg.png

Replace with better one. --128.6.175.89 18:46, 5 March 2007 (UTC)

[edit] poetry contest

There was a contest in 1985 for poems and prose using the sentence: http://www.linguistlist.org/issues/2/2-457.html 64.160.39.153 03:28, 13 March 2007 (UTC)

[edit] NPOV?

The introduction includes the following line:

It can be argued that Chomsky simply was not imaginative enough to put the sentence into a context which would give it a sensible meaning.

This doesn't seem very NPOV to me, but I don't understand the argument well enough to be able to replace it. Can someone please clarify this statement? What context could it be placed in? What sensible meaning could it have? David 19:30, 17 March 2007 (UTC)

Retrieved from "http://en.wikipedia.org../../../c/o/l/Talk%7EColorless_green_ideas_sleep_furiously_5af7.html"