Re: Comparing languages

From: LEE Sau Dan (danlee_at_informatik.uni-freiburg.de)
Date: 07/16/04 

Date: 16 Jul 2004 10:01:47 +0200

>>>>> "Nathan" == Nathan Sanders <nsanders.DIE.SPAM@wso.williams.edu> writes:

    Nathan> (Interesting side note: our most irregular forms tend to
    Nathan> be high up on the frequency list, and our least frequent
    Nathan> irregular forms tend to be regularized over time. I don't
    Nathan> know if this is true for other lannguages, but a cursory
    Nathan> examination seems to show that it might be the case.

I discussed this with someone in sci.lang a few years ago. Our
conclusion was: that's not a mere coincidence. It's a result of
evolution:

If the irregular forms do not occur frequently, then even the native
speakers will forget them. The language either [unlikely] dies
because the native (and non-native) speakers cannot communicate with
one another anymore, or [more likely] the language changes in a way so
that those infrequent words are regularized.

So, the only way for a language to survive through a prolonged period
of time is to have regular forms for the infrequent words.

i.e. infrequent word ==> regularly inflected.

The other direction "regularly inflected ==> infrequent word" is not
necessarily true. (e.g. the isolating languages.) But a reason for a
language to develop irregular forms for frequent words is economy!
Instead of having to use a regular but long long word, one can use an
irregular but very short word to save time and energy. In addition,
some highly irregular forms (like "to be" in English) usually carry
some redundancy to facilitate communication. That's another factor
favouring the development of irregular forms.

    Nathan> I'm just not convinced that such a measure can say
    Nathan> anything useful by about the learnability of a language.
    Nathan> "More tokens" isn't universally harder to learn than
    Nathan> "fewer tokens", since the structure of the tokens can
    Nathan> impede or facilitate learning. By analogy, learning (by
    Nathan> which I mean "memorizing") the long string 1357924680 is
    Nathan> significantly easier than learning the shorter string
    Nathan> 94706215, and if you didn't know what the symbol <2> meant
    Nathan> before reading these strings, you would have a much easier
    Nathan> time figuring it out from the longer string than from the
    Nathan> shorter string.

And I have memorized the string

14159265358979323846264338327950288419716939937510...
<100 more digits omitted>

But if you ask me to memorize an international phone number (i.e. with
all those country code + area prefix) of a new made friend, I'd find
it a bit challenging. If he lives in a city I'm familiar with, I'll
make use of my background knowledge of the country code and area code
of that city and memorize only the local part of his number together
with the city. Otherwise, I'd be lost. 

-- 
Lee Sau Dan                     +Z05biGVm-                          ~{@nJX6X~}
E-mail: danlee@informatik.uni-freiburg.de
Home page: http://www.informatik.uni-freiburg.de/~danlee

Relevant Pages

  • Re: Comparing languages
    ... > Nathan> irregular forms tend to be regularized over time. ... > of time is to have regular forms for the infrequent words. ... > language to develop irregular forms for frequent words is economy! ...
    (sci.lang)
  • Re: Ottawa unicyclists, READ THIS!
    ... I agree with nathan though that being there is a current group that has ... regular get togethers why start another... ... Victory's Profile: http://www.unicyclist.com/profile/10062 ...
    (rec.sport.unicycling)
  • Re: Beyond?
    ... Having read Issue #1, it doesn't give any explanation, but it is ... Seems to be consciously trying to follow the regular MU to me, ... Nathan P. Mahney - ...
    (rec.arts.comics.marvel.universe)