No subject

[A. Katz]

I'm glad to see that my last posting generated some interesting responses.

By and large, I do think that one point may have been misunderstood. I was not
arguing that the length of a sentence in natural language is limited by the
mortality of its speakers or the amount of time in the day that they can
devote to speaking. While that point is valid, I would tend to agree with
Jean Hudson that it is not particularly interesting.

My focus was on the length limitation imposed on a sentence by the processing
abilities of speakers and hearers. Sergio Meira observes: "Sentences are
never infinite-- but they don't seem to be bounded either." My point is that
if we are talking about natural language processing in real time -- they ARE
bounded. (Although it is unclear precisely where the boundary lies, and it
would require a considerable quantity of experimental data to pinpoint.)

Using Sergio's analogy of baseball and volleyball matches, I would point out
that the unit more comparable to a sentence in these events is probably the
amount of time the ball can be kept in the air without touching the ground,
not the amount of time that the game can go on.

While our lives are finite and the amount of time we spend talking is limited,
that is not the real limitation on the length of a sentence. The President of
the United States could give a two hour speech, and people would take time out
from their schedules to listen to it, but what is the likelihood that the
whole speech would consist of a single sentence?

Or take the Mark Twain joke about a multi-volume opus in German in which all
the verbs appear in the last volume. That would never happen in real life,
(not even in German), because such a work, while it might be logically
meaningful, could not be processed by any human being.

We may occasionally listen to a long speech or read a very long book, but in
order for it to be comprehensible, it must be broken down into smaller,
self-contained parts. That is a limitation placed on language by human
cognition.

The reason this upper bound is very interesting to me is that it has profound
implications for language change and grammaticalization. I would be most
grateful for any of you who could provide citations to works on this subject
or who might have expertimental data that would shed light on the issue.



                         --Aya Katz

\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//
Dr. Aya Katz, 3918 Oak, Brookfield, Illinois 60513-2019 (708) 387-7596
//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\