Theiling Online    Sitemap    Conlang Mailing List HQ   

Computer syntax and VSO

From:Samuel Rivier <samuelriv@...>
Date:Monday, February 16, 2004, 21:14
Hey everyone. It's been a while since I've posted, but
I'm beginning some major-specific coursework in
computer science, and I've noticed something in data
structures that may be of interest.

When a program reads any type of semantic structure,
be it a string or code or math, it has to first
reorder it in a fashion that is easiest to read. In
the case of math, this is called postfix (VSO) or
prefix (SOV), and, by the way, it is for these
purposes only for functions of two arguments (add,
multiply, etc).
So, for instance,: (2+5)*6/(3+2)+5
in postfix is:     5 3 2 + 6 2 5 + * / +
which as you can see is kinda messy to look at, but is
quite legible with practice.

You guys may recall Bertrand Russell's interpretation
of grammar through symbolic logic (the basis of
lojban) in which verbs are seen as functions taking
both the subject and the object indiscriminantly as
arguments, which is admittedly flawed if one observes
that 95% of languages discriminate the subject and
object such that the subject always precedes the

However, what I am proposing here is that in some way
VSO or SOV sentence structure is "better", in terms of
mental processing, than SVO, by virtue of the
idealization that it is relatively unambiguous. Or is
it facetious of me to compare the human brain with
computer processing (though keep in mind that neurons
do transmit in terms of signal/no-signal)? What does
everyone here think about a mathematical or
computational representation of language, particularly
considering that even though Russell's analysis is
severely flawed, it is certainly quite flexible in
comparison to the subject, verb, adverb, adjective
concepts of the Western world which, by the way, do
not apply to, say, Mandarin.



Samuel Rivier

Case Western Reserve University
Undergraduate - Department of Fizzacks

Do you Yahoo!?
Yahoo! Finance: Get your refund fast by filing online.


Philippe Caquant <herodote92@...>
Christophe Grandsire <christophe.grandsire@...>