Xref: utzoo comp.ai:1266 sci.lang:1782
Path: utzoo!utgpu!water!watmath!clyde!rutgers!sri-spam!ames!aurora!labrea!russell!zwicky
From: zwicky@russell.STANFORD.EDU (Arnold Zwicky)
Newsgroups: comp.ai,sci.lang
Subject: Re: words order in English and Japanese
Summary: Don't underestimate variability
Message-ID: <2019@russell.STANFORD.EDU>
Date: 28 Jan 88 17:36:25 GMT
References: <1671@russell.STANFORD.EDU> <236@pyuxf.UUCP> <3580@bcsaic.UUCP> <243@pyuxf.UUCP>
Reply-To: zwicky@russell.UUCP (Arnold Zwicky)
Organization: Center for the Study of Language and Information, Stanford U.
Lines: 43


There is no reason to think that everything in language is there "for a
purpose" any more than there is to think that everything in nature is.
There *is* an enormous amount of randomness in the world, both natural
and cultural.

In the linguistic case, there is a huge amount of variability in expression,
both across languages and within languages.  Within a language, there is
a purpose for much of this variability (though surely not all of it);
different word orders, constructions, intonations, pronunciations, etc.
are associated with different discourse functions and serve as 
sociolinguistic markers.  But these functions are largely arbitrarily
associated with linguistic form - the same form can serve different
functions in different languages, and different forms can serve the
same function in different languages.  (This is not to deny that
certain forms are particularly good for certain functions, like
rising intonation for asking questions.  It *is* to deny that this
form is locked onto that function; if the form is already taken for
some purpose, then any other form available in the language can be
pressed into service.)

None of this posits any universal scheme of association between language
(in particular, syntax) and other aspects of culture.  The bits of a
language have to fit together, elegantly or clunkily, as the particular
case might be, into a complex system with many purposes, but to ask
WHY these particular bits are assembled in particular languages is to
invite the answers:  Just because.  Whatever is, is right.

What I have just said will probably seem pretty unsatisfying to some
readers.  Lots of people would like to believe that there have to be
deep reasons for why things are the way they are, that historical
accident is never a satisfactory account of the panoply of nature
and culture.  If those are your beliefs, then I'm not likely to change
your mind.  But I'd like to try to change your mind, at least if you're
going to be thinking seriously about language.  Looking for the big WHY
answers is a bad research strategy here; it leads you to overlook much
of the essential complexity of the phenomena.  The question of WHAT there
is in language is a tougone, and insofar as we know the answers they
seem to require an intricate system oconpts and hypotheses, most
of them with no detectable connection to matters of culture or
geography or ethnicity.  If you attend only to phenomena that you can
plausibly interpret by reference to culture etc., then you're going
to miss most of the relevant data.