Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Infinitely many sentences & childrens' input v. output?

  1. Jul 23, 2005 #1

    honestrosewater

    User Avatar
    Gold Member

    I've just started a linguistics book, and I'd like to get others' opinions on two points that I have reservations about. I get the sense that they may be overstating things a bit in trying to make the case for Universal Grammar.

    I was going to let this go, but they bring it up elsewhere too. Perhaps theoretically there's no limit on sentence length, but I think there are real limits in practice, which shouldn't be ignored. Even when creating a sentence, where I imagine the theoretical lack of limits is most relevant, I am thinking about keeping the length within certain bounds. Perhaps 'acceptable' sentence length is something we learn through experience? Still,
    This seems just plain wrong. In what way is it even possible to use and understand an infinite set of sentences?

    The other point:
    Is the input really so different and the result really so similar? Are they are using different standards for each? I would think that children learning the same language are exposed to much the same set of utterances. No? If they're talking about across languages, then saying that they all arrive at 'pretty much the same grammar' seems wrong, especially while holding the input to not being 'exactly the same'.
     
  2. jcsd
  3. Jul 23, 2005 #2

    selfAdjoint

    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    I think the word infinite in the book might be better replaced with "open ended". The point is that children come to be able to understand sentences they have never heard before, and every natural language has the capability to freely generate novel sentences. In the author's perhaps unquantitative mind, infinite means just what its Latin roots suggest, unbounded. And I think he is emphasizing content and inner relationships of a sentence, not just length.
     
  4. Jul 23, 2005 #3
    I agree basically with selfAdjoint, however I do have a bit of a problem with the statement that goes:

    I find the grammar being used around me (children and adults) quite varied, and I'd say that the grammar from the children is at least as varied as that, and similar to, that of the adults, with whom they associate.

    By the way, I like the fact that you are writing on the subject. I hope that:
    1) It will be mostly about English. We are in danger of losing our grip
    on it (mostly through neglect).
    2) I hope that syntax will be featured, and the logic well explained. (Our
    traditional English books have left a lot to be desired.)
    3) you'll share some of it with us.

    KM
     
  5. Jul 23, 2005 #4
    By the way:
    I think that there should be a forum here on "Linguistics", probably under the heading of 'Philosophy'. I can think of nothing more important to all of us than the ways in which we communicate. In particular, I would like to see 'syntax' and 'semantics' discussed, but also things like derivation, change and some of the latter day movements to make our language less precise.

    KM
     
  6. Jul 23, 2005 #5
    By the way again:
    I do believe that 'Linguistics' is a science, or at least, it was the last time I checked; and language, even though informally derived contains within it some incredibly subtle and wonderful structures and rules. Also interesting, are the relative strengths of the different languages.


    KM
     
  7. Jul 23, 2005 #6

    selfAdjoint

    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    I think that threads on linguistics belong here on social sciences, not in philosophy. Scientists can check their theories against reality, which makes their study a science in my opinion.

    Your objection about variant grammars has a simple answer I think. Most children only learn one grammar, which is that of the dialect their parents and peers speak. In the rare cases where children learn two dialects from the beginning, that is just the same case as learning two languages from scratch. It takes longer, but toddlers can do it. The structures ("deep structures" in Chomsky's terminology) that support learning one language support learning all, but of course particular language skills once learned remain in long term memory, and the ability to access the deep grammar engine fades as the child matures.
     
  8. Jul 23, 2005 #7
    remember linguists aren't necessary mathematicians...they may use terms somewhate differently from math.

    first quote:
    remember sometimes they look at children for studies...and children can put strings of sentences together without realizing length...even babababab babbababa baba...can be a sentence its up to the second party to interpret whether it has meaning. I can even regurgitate a full run on sentence in this paragraph by removing all the periods and it would be considered a sentence it would be up to you to interpret it or break it oapart.

    second quote:
    againt its teh second party reference, I can say a whole lot of crap to you and its your interpretation that goes through your head. I can remove all the verbs and a child can still interpret the way they choose...and like selfadjoint said "open-ended" may be a better term though that can also mean infinite, perhaps a "larger set" term could be employed. Its really a "combinatorial mathematics" thing.

    third quote:
    this came across odd the first time i heard this theory...but i believe it is to mean
    most grammars have similar structures in that they handle a noun/verb/adjective in similar ways not necessarily the exact same order.
     
  9. Jul 24, 2005 #8

    honestrosewater

    User Avatar
    Gold Member

    Sorry, I've just started to read a linguistics book. I'm not writing one.

    I understand the point about the creative aspect of language. I'm really just interested in the narrower question of quantity and limits. Splitting it up into knowledge and use, I can see describing a person's linguistic knowledge as vast, extensive, or such, but what about it even has the potential to be limitless? And it seems that our use of language being boundless is at odds with it following rules and patterns.

    I can let the other point go - it's just the beginning of the book.
     
  10. Jul 24, 2005 #9

    selfAdjoint

    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed


    It is precisely because our language capacity exceeds any specific set of rules that Chomsky's generative grammar is so persuasive. Instead of having rules in our head, we have a "program" that can generate rules, and is never stuck for a new rule.
     
  11. Jul 25, 2005 #10

    honestrosewater

    User Avatar
    Gold Member

    Okay, maybe I was taking it the wrong way. They're speaking of the general potentials of the 'program' and I was thinking of the specific reality of individuals. Anyway, I'm satisfied and moving on. :smile:

    Heh, a few pages later:
    :rolleyes:
     
    Last edited: Jul 26, 2005
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Infinitely many sentences & childrens' input v. output?
  1. A sentence like many. (Replies: 7)

  2. Children's Cartoons (Replies: 5)

  3. Children fighting (Replies: 30)

  4. The Children of Hurin (Replies: 1)

Loading...